Postoperative Sore Throat: More Answers Than Questions

EDITORIAL

Phillip E. Scuderi, MD

ostoperative sore throat (POST) is a common adverse oral airway). The site or sites of mucosal injury would event after general anesthesia. Typically, the incidence obviously vary depending on the airway device. For in- Pof POST is highest in patients who are tracheally stance, endotracheal intubation can result in injury to any intubated; however, POST also occurs when a laryngeal mask portion of the pharynx as well as injury to the larynx and airway (LMA) is used.1 Even patients who are managed with trachea. Placement of an LMA can reasonably be expected a facemask are not immune.1 Most of the measures that have to cause injury to pharyngeal mucosa in the supraglottic been recommended for reducing this complication have been regions only, whereas the use of a facemask with an oral directed at limiting the physical trauma that might result from airway should result in injury to only the oropharynx, airway instrumentation and manipulation. Surprisingly few assuming that no other injuries occurred because of suc- investigations have evaluated pharmacologic interventions tioning or other airway maneuvers. It is therefore some- as a means of reducing POST. Furthermore, no single drug what surprising to note that the reported incidence of POST has achieved widespread acceptance in the clinical commu- after LMA insertion is, at least in some studies, remarkably nity. In this issue of Anesthesia & Analgesia, 4 articles similar to that seen with endotracheal intubation.6,7 Al- describe simple prophylactic measures that seem to signifi- though this might lead one to infer that the mechanism and 2–5 cantly reduce the incidence of POST. Two of these location of injury must also be similar, a number of facts articles evaluate the effectiveness of topical benzydamine seem to contradict this assumption. For instance, reducing hydrochloride applied to the cuff of the endotracheal tube, the size of endotracheal tubes results in a significant 2,3 directly to the pharyngeal mucosa, or both. A third article decrease in the incidence of POST.8 The design of tube cuffs 4 evaluates the efficacy of inhaled fluticasone propionate. has also been an area of intense research. The size, ௡ The fourth article evaluates Strepsils , a nonprescription pressure/volume characteristics, and shape of cuff have all lozenge that contains 2 active ingredients, amylmetacresol been implicated in tracheal mucosal injury and resultant 5 and 2,4-dichlorobenzyl alcohol. POST.9–12 Conversely, it has been suggested that cuff When considered in aggregate, these 4 articles raise a inflation pressure has less of a role in POST when an LMA number of interesting questions. The expression “sore is used.6 Both airway devices are clearly capable of induc- throat” is obviously common to the vernacular of many ing mucosal irritation and both can cause POST in patients different cultures, yet it provides at best a parsimonious at rates that are not strikingly different. Yet, anatomically, description of the actual phenomena. Consequently, the the site or sites of injury cannot be the same. expression “postoperative sore throat” likely represents a There are several interesting observations that arise broad constellation of signs and symptoms. For instance, in when one examines the data presented in the 4 articles its simplest form, sore throat is a lay description of phar- published in this issue of Anesthesia & Analgesia. The data yngitis, which in itself can have a variety of causes. 4 from Tazeh-kand et al. demonstrate that inhalation of However, sore throat may also include a variety of symp- fluticasone propionate before the induction of anesthesia toms including laryngitis, tracheitis, hoarseness, cough, or significantly reduces the incidence of POST at 1 hour and dysphagia. Postoperatively, it seems most plausible that the 24 hours after surgery compared with a placebo control. symptoms are the result of mucosal injury with resulting This is not necessarily an unexpected result. Topical13,14 inflammation caused by the process of airway instrumen- and systemic steroids15 have been demonstrated to reduce tation (i.e., laryngoscopy and suctioning) or the irritating the incidence of POST presumably because of their sys- effects of a foreign object (i.e., endotracheal tube, LMA, or temic antiinflammatory effects. More puzzling are the data presented by Ebneshahidi and Mohseni.5 Patients who From the Department of Anesthesiology, Wake Forest University School of Medicine, Winston-Salem, North Carolina. received a Strepsils lozenge before the induction of anes- Accepted for publication June 14, 2010. thesia had a significantly lower incidence of POST and Disclosure: The author reports no conflicts of interest. hoarseness both in the postanesthesia care unit and at 24 Address correspondence and reprint requests to Phillip E. Scuderi, MD, hours after surgery than did the placebo control group that Department of Anesthesiology, Wake Forest University School of Medicine, received a flavored lozenge without the active ingredient. Medical Center Blvd., Winston-Salem, NC 27157-1009. Address e-mail to [email protected]. Unless one postulates a systemic effect from the active Copyright © 2010 International Anesthesia Research Society ingredients in the Strepsils lozenge (i.e., amylmetacresol DOI: 10.1213/ANE.0b013e3181ee85c7 and 2,4-dichlorobenzyl alcohol), the effect site must be the

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 831 EDITORIAL

pharyngeal mucosa. Whereas it is plausible to postulate a eliminated. The 4 studies presented here may provide the reduction on pharyngeal irritation due to the lozenge, it is impetus for a more careful evaluation of POST resulting in harder to postulate a mechanism for the reduction in more precisely targeted therapies. “hoarseness” that was reported in this study. It is also difficult to understand how a preoperative lozenge could AUTHOR CONTRIBUTIONS reduce a reaction to injury to the larynx and trachea and the PES designed and conducted the study, analyzed the data, and resultant laryngotracheitis that must have a role in POST wrote the manuscript. This author approved the final manuscript. that occurs after intubation. 2,3 Conversely, 2 of the articles describe a reduction in REFERENCES POST with the application of benzydamine hydrochloride 1. Higgins PP, Chung F, Mezei G. Postoperative sore throat after to the endotracheal tube cuff alone compared with a ambulatory surgery. Br J Anaesth 2002;88:582–4 placebo control (normal saline and distilled water, respec- 2. Hung NK, Wu CT, Chan SM, Lu CH, Hang YS, Yeh CC, Lee MS, tively). It seems unlikely that the small dose of benzydam- Cherng CH. The effect on postoperative sore throat of spraying the endotracheal tube cuff with benzydamine hydrochloride, 10% ine hydrochloride (1.5 and 0.75 mg, respectively) used lidocaine, and 2% lidocaine. Anesth Analg 2010;111:882–6 would have resulted in a systemic effect. Therefore, the 3. Huang YS, Hung NK, Lee MS, Kuo CP, Yu JC, Huang GS, reduction in POST that was observed in both of these Cherng CH, Wong CS, Chu CH, Wu CT. The effectiveness of studies must be assumed to have resulted from a localized benzydamine hydrochloride spray on the endotracheal tube decrease in mucosal injury and/or inflammatory response. cuff or oral mucosa for postoperative sore throat. Anesthesia Analg 2010;111:887–91 The incidence of and reduction in POST is strikingly similar 4. Tazeh-kand NF, Eslami B, Mohammadian K. Inhaled flutica- when Strepsils lozenges were used when compared with the sone propionate reduces postoperative sore throat, cough and application of benzydamine hydrochloride to the endotra- hoarseness. Anesth Analg 2010;111:895–8 ௡ cheal tube cuff. There are unavoidable questions that must be 5. Ebneshahidi A, Mohseni M. Strepsils tablets reduce sore throat and hoarseness after tracheal intubation. Anesth Analg 2010;111:892–4 asked: How can a lozenge that is administered orally result in 6. Wakeling HG, Butler PJ, Baxter PJ. The laryngeal mask airway: a similar reduction in POST when compared with the topical a comparison between two insertion techniques. Anesth Analg application of benzydamine hydrochloride to the endotra- 1997;85:687–90 cheal tube cuff? In addition, How can either an oral lozenge or 7. Joshi GP, Inagaki Y, White PF, Taylor-Kennedy L, Wat LI, topical antiinflammatory agent applied to an endotracheal Gevirtz C, McCraney JM, McCulloch DA. Use of the laryngeal mask airway as an alternative to the tracheal tube during tube cuff yield reductions in POST that compare favorably ambulatory anesthesia. Anesth Analg 1997;85:573–7 with a more widespread application of topical steroids to both 8. Stout DM, Bishop MJ, Dwersteg JF, Cullen BF. Correlation of the pharyngeal and laryngotracheal mucosa? endotracheal tube size with sore throat and hoarseness follow- POST is unquestionably a common adverse event after ing general anesthesia. Anesthesiology 1987;67:419–21 9. Loeser EA, Kaminsky A, Diaz A, Stanley TH, Pace NL. The general anesthesia. A number of physical factors have been influence of endotracheal tube cuff design and cuff lubrication implicated as noted above. Most notable would seem to be on postoperative sore throat. Anesthesiology 1983;58:376–9 endotracheal tube and cuff design and the approach to airway 10. Loeser EA, Bennett GM, Orr DL, Stanley TH. Reduction of management (i.e., endotracheal tube, LMA, or mask anesthe- postoperative sore throat with new endotracheal tube cuffs. sia). In addition, female gender, younger patients, gynecologic Anesthesiology 1980;52:257–9 11. Loeser EA, Hodges M, Gliedman J, Stanley TH, Johansen RK, surgery, and the use of succinylcholine also seem to increase Yonetani D. Tracheal pathology following short-term intuba- 16 the incidence. Of particular note, the use of topical lidocaine tion with low- and high-pressure endotracheal tube cuffs. appears to confer no benefit and may in fact make POST Anesth Analg 1978;57:577–9 worse,17,18 a fact that seems to have been confirmed by Hung 12. Loeser EA, Orr DL II, Bennett GM, Stanley TH. Endotracheal 2 tube cuff design and postoperative sore throat. Anesthesiology et al. However, what actually causes POST remains some- 1976;45:684–7 thing of a mystery. Hoarseness is a physical sign and can 13. Ayoub CM, Ghobashy A, Koch ME, McGrimley L, Pascale V, certainly be evaluated objectively by a careful observer. Dys- Qadir S, Ferneini EM, Silverman DG. Widespread application phagia, although a symptom, is perhaps less likely to be of topical steroids to decrease sore throat, hoarseness, and influenced by intersubject variation in reporting, particularly cough after tracheal intubation. Anesth Analg 1998;87:714–6 14. Sumathi PA, Shenoy T, Ambareesha M, Krishna HM. Con- if questioning is done by a trained observer and is consistently trolled comparison between betamethasone gel and lidocaine applied. Sore throat is more problematic. As noted above, the jelly applied over tracheal tube to reduce postoperative sore location of mucosal injury can vary widely but still result in a throat, cough, and hoarseness of voice. Br J Anaesth 2008; subject complaining of a “sore throat.” The 4 articles pre- 100:215–8 15. Park SH, Han SH, Do SH, Kim JW, Rhee KY, Kim JH. sented in this issue of Anesthesia & Analgesia describe very Prophylactic dexamethasone decreases the incidence of sore different strategies yet all achieved positive results. However, throat and hoarseness after tracheal extubation with a double- we are no closer to understanding the actual etiology of POST lumen endobronchial tube. Anesth Analg 2008;107:1814–8 than we were before. A reasonable question to ask is “Does it 16. McHardy FE, Chung F. Postoperative sore throat: cause, pre- matter?” I believe that the answer is “yes.” If the precise vention and treatment. Anaesthesia 1999;54:444–53 17. Herlevsen P, Bredahl C, Hindsholm K, Kruhøffer PK. Prophy- etiology (or etiologies) of pain after airway management can lactic laryngo-tracheal aerosolized lidocaine against postopera- be determined, it increases the likelihood that a specific tive sore throat. Acta Anaesthesiol Scand 1992;36:505–7 therapy or therapies could be recommended that would 18. Loeser EA, Stanley TH, Jordan W, Machin R. Postoperative decrease symptoms and improve outcomes. POST is not the sore throat: influence of tracheal tube lubrication versus cuff design. Can Anaesth Soc J 1980;27:156–8 most important adverse event to avoid, at least from a 19 19. Macario A, Weinger M, Carney S, Kim A. Which clinical patient’s perspective. Nevertheless, it is an adverse event anesthesia outcomes are important to avoid? The perspective that could easily be significantly decreased or even potentially of patients. Anesth Analg 1999;89:652–8

832 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA EDITORIAL

Activated Clotting Times, Heparin Responses, and Antithrombin: Have We Been Wrong All These Years?

Jerrold H. Levy, MD, FAHA, and Roman M. Sniecinski, MD

t was recognized early in the development of modern well. In this issue of the journal, 3 investigations from the cardiopulmonary bypass (CPB) that the fluidity of same group add further information and pose new questions blood needed to be maintained for extracorporeal cir- about the interrelationship of antithrombin and heparin. I 9 culation, especially with the evolution of early oxygen- In the first study, Garvin et al. retrospectively evaluated ators.1 It was not until the anticoagulant properties of the Hepcon HMS PLUS System (Medtronic Inc., Minneapo- heparin were discovered that blood could circulate over lis, MN) for its accuracy in predicting heparin dose re- nonendothelial surfaces without activating the coagulation sponses for cardiac surgery with CPB. ACT-measured cascade. Today, almost a century after its isolation by heparin dose response and heparin concentrations were Howell and McLean in 1916, heparin remains the mainstay evaluated in 3880 patients after a heparin dose calculated to 2 agent for cardiac surgery because of several advantages. achieve a target ACT. The result was wide variability in Profound levels of anticoagulation can be quickly obtained measurements. A target ACT of 300 seconds was not (AntiXa levels of 3–5 U/mL) and an antidote, protamine, is obtained in 7% of patients, and a target ACT of 350 seconds readily available. In addition, heparin has a relatively short was not obtained in 17% of patients. The investigators also context-sensitive half-life, and can be used in patients with found that calculated and measured heparin dose re- renal dysfunction. However, an important limiting aspect sponses were not related at any heparin dose. of heparin is that it requires a cofactor, antithrombin (also In the second study, the authors examined whether referred to as antithrombin III or ATIII), for its anticoagu- there was a direct association among preoperative anti- lation effect. Thus, heparin is still used extensively as an thrombin activity, heparin dose responses, and the heparin anticoagulant despite the development of newer agents sensitivity index from 304 patients after CPB using Hepcon that have been the focus of significant clinical interest and 10 3 HMS PLUS Systems. The authors used multivariate linear consternation. regression to identify independent predictors of heparin The most common method for determining the adequacy dose response. The baseline antithrombin activity was of anticoagulation for cardiac surgery is the activated normal in this study and was not associated with either clotting time (ACT), a modified whole blood Lee-White baseline or postheparin ACT, heparin dose response, or clotting assay that can be affected by multiple factors.4 heparin sensitivity index. Of note, only 16% of patients (49 Although there is not universal agreement on the ideal of 304) in the study presented with low baseline antithrom- ACT for CPB, when heparin does not produce the bin levels as defined as Ͻ80%. desired ACT increase, “heparin resistance” is said to In the third study, the authors prospectively evaluated have occurred.5 The term is misleading, however, be- whether low levels of antithrombin were associated with cause it is really an alteration in the heparin dose postoperative major adverse cardiac events in 1403 patients response, possibly because of decreased antithrombin. undergoing coronary artery bypass graft.11 Major adverse “Heparin resistance” is perhaps more aptly termed “al- cardiac events were defined as postoperative death, reop- tered heparin responsiveness.”6 7 eration for coronary graft occlusion, myocardial infarction, Antithrombin levels markedly decrease after CPB. stroke, pulmonary embolism, or cardiac arrest until first These decreased levels may lead to a dangerous increase in thrombin activity.8 It would seem logical that this could hospital discharge. Antithrombin activity levels were mea- lead to postoperative thrombotic complications, suggesting sured preoperatively, post-CPB, and on postoperative days the need for antithrombin not only before CPB, but after it as 1 to 5. Major adverse cardiac events occurred in 146 patients (10.4%) and were independently associated with

From the Department of Anesthesiology, Emory University School of postoperative antithrombin but not preoperative anti- Medicine, Cardiothoracic Anesthesiology and Critical Care, Emory Health- thrombin levels. care, Atlanta, Georgia. What do these 3 studies tell us? For starters, they prove Accepted for publication June 22, 2010. that an altered heparin response cannot be predicted by Disclosure: The authors report no conflicts of interest. preoperative antithrombin alone. Heparin’s effect on co- Address correspondence and reprint requests to Jerrold H. Levy, MD, FAHA, agulation is rather unpredictable, at least as measured by Department of Anesthesiology, Emory University Hospital, 1364 Clifton Rd., NE, 12,13 Atlanta, GA 30322. Address e-mail to [email protected]. ACT, an important perspective but not a new finding. 14 Copyright © 2010 International Anesthesia Research Society For example, Metz and Keats gave 193 patients a single DOI: 10.1213/ANE.0b013e3181f08a80 dose of heparin (300 U/kg). In 51 patients (26.4%), ACT

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 833 EDITORIAL

Figure 1. The activated clotting time (ACT) is a whole blood point of care coagulation test used extensively in cardiac surgery and in the cardiac catheterization laboratory to monitor the anticoagulant effect of different agents including unfractionated heparin. In the ACT, blood is added to a cartridge or tube that contains an activator, usually celite or kaolin, to speed the process by increasing contact activation by the intrinsic coagulation cascade. Clot formation in the ACT represents the interaction of plasma coagulation components (e.g., factors and fibrinogen), platelets, and red blood cells as this is a whole blood clotting assay. However, clot formation in the ACT is influenced by multiple factors that include platelet count and platelet function, factor deficiencies, fibrinogen levels, pharmacologic agents (anticoagulants in platelet inhibitors), temperature(especially hypothermia), and contact activation inhibitors (e.g., aprotinin).

values were Ͻ400 seconds, including 4 patients Ͻ300 there are still only a few supporting studies demonstrating seconds. that antithrombin supplementation improves clinical out- The target ACT values of 300 or 350 seconds for the 3 comes, and these studies were not conducted in patients studies may seem relatively low for many clinical practices undergoing cardiac surgery.24,25 and the rate of failure to meet those targets may seem In summary, the current 3 studies provide additional unusually high. Previous reports suggest there is a 5% to data about the complex issues of anticoagulation, heparin 10% chance of a patient developing altered heparin respon- responsiveness, and outcomes, including the role of anti- Ͼ siveness. The current studies targeted ACT values 350, thrombin. Have we been wrong all these years regarding which may account for the authors finding a 17% incidence ACTs, heparin responsiveness, and the role of antithrom- 14,15 of altered heparin responsiveness. bin? More than 50 years after the development of CPB, we Realistically, we still do not know the ideal ACT to are still asking many of the questions that the early initiate CPB, which is reflected in the widespread variabil- pioneers confronted. We need better monitors and better ity in clinical practice.4,16 Several studies have demon- understanding of anticoagulation adequacy to treat alter- strated that thrombin is still activated at ACT values of 400 ations in heparin response and assess therapeutic efficacy. to 480 seconds during CPB.17,18 As shown by the response We hope it will not take another 50 years to find these of patients with factor XII deficiency, a prolonged ACT answers. does not guarantee anticoagulation.19 All of this calls into question the validity of using ACTs to assess adequate anticoagulation. Clinicians need to un- REFERENCES derstand the factors that influence both anticoagulation and 1. Galletti PM. Cardiopulmonary bypass: a historical perspective. Artif Organs 1993;17:675–86 its clinical measurement. The ACT is a relatively primordial 2. Wardrop D, Keeling D. The story of the discovery of heparin tool, consisting only of a tube, some dirt or glass for and warfarin. Br J Haematol 2008;141:757–63 activation, and some heat and agitation to speed the 3. Spyropoulos AC. Brave new world: the current and future use reaction. As shown in Figure 1, there are a host of other of novel anticoagulants. Thromb Res 2008;123:S29–35 factors besides heparin concentration and antithrombin 4. Despotis GJ, Gravlee G, Filos K, Levy J. Anticoagulation levels that affect ACT values. Unfortunately, there is no monitoring during cardiac surgery: a review of current and emerging techniques. Anesthesiology 1999;91:1122–51 “gold standard” measurement to validate ACTs, complicat- 5. Staples MH, Dunton RF, Karlson KJ, Leonardi HK, Berger RL. ing assessment of the relationship between ACT and al- Heparin resistance after preoperative heparin therapy or in- tered heparin responsiveness. traaortic balloon pumping. Ann Thorac Surg 1994;57:1211–6 Based on the current studies, is there a role for 6. Levy JH. Heparin resistance and antithrombin: should it still be antithrombin administration to improve heparin respon- called heparin resistance? J Cardiothorac Vasc Anesth 2004; siveness? Multiple studies report that antithrombin 18:129–30 7. Zaidan JR, Johnson S, Brynes R, Monroe S, Guffin AV. Rate of supplementation improves intraoperative anticoagulation, protamine administration: its effect on heparin reversal and increases ACT, and reduces biochemical markers of hemo- antithrombin recovery after coronary artery surgery. Anesth static activation.13,20–22 Although these 3 studies did not Analg 1986;65:377–80 investigate antithrombin administration, the authors did 8. Sniecinski R, Szlam F, Chen EP, Bader SO, Levy JH, Tanaka find a relationship between postoperative antithrombin KA. Antithrombin deficiency increases thrombin activity levels and major adverse cardiac events. This corroborates after prolonged cardiopulmonary bypass. Anesth Analg 23 2008;106:713–8 the observation by Ranucci et al., that low levels of 9. Garvin S. Heparin concentration-based anticoagulation for antithrombin activity in the intensive care unit are associ- cardiac surgery fails to reliably predict heparin bolus dose ated with a poor outcome in cardiac surgery. However, requirement. Anesth Analg 2010;111:849–55

834 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Activated Clotting Times, Heparin Responses, and Antithrombin

10. Garvin S. Heparin dose response is independent of perioper- 19. Salmenpera M, Rasi V, Mattila S. Cardiopulmonary bypass ative antithrombin activity in patients undergoing coronary in a patient with factor XII deficiency. Anesthesiology 1991; artery bypass graft surgery using low heparin concentrations. 75:539–41 Anesth Analg 2010;111:856–61 20. Avidan MS, Levy JH, van Aken H, Feneck RO, Latimer RD, Ott 11. Garvin S. Postoperative activity, but not preoperative activity, E, Martin E, Birnbaum DE, Bonfiglio LJ, Kajdasz DK, Despotis of antithrombin is associated with major adverse cardiac GJ. Recombinant human antithrombin III restores heparin events after coronary artery bypass graft surgery. Anesth responsiveness and decreases activation of coagulation in Analg 2010;111:862–9 heparin-resistant patients during cardiopulmonary bypass. 12. Gravlee GP, Case LD, Angert KC, Rogers AT, Miller GS. J Thorac Cardiovasc Surg 2005;130:107–13 Variability of the activated coagulation time. Anesth Analg 21. Levy JH, Despotis GJ, Szlam F, Olson P, Meeker D, Weis- 1988;67:469–72 13. Levy JH, Montes F, Szlam F, Hillyer CD. The in vitro effects of inger A. Recombinant human transgenic antithrombin in antithrombin III on the activated coagulation time in patients cardiac surgery: a dose-finding study. Anesthesiology 2002; on heparin therapy. Anesth Analg 2000;90:1076–9 96:1095–102 14. Metz S, Keats AS. Low activated coagulation time during 22. Williams MR, D’Ambra AB, Beck JR, Spanier TB, Morales DL, cardiopulmonary bypass does not increase postoperative Helman DN, Oz MC. A randomized trial of antithrombin bleeding. Ann Thorac Surg 1990;49:440–4 concentrate for treatment of heparin resistance. Ann Thorac 15. Ranucci M, Isgro G, Cazzaniga A, Ditta A, Boncilli A, Cotza M, Surg 2000;70:873–7 Carboni G, Brozzi S. Different patterns of heparin resistance: 23. Ranucci M, Frigiola A, Menicanti L, Ditta A, Boncilli A, Brozzi therapeutic implications. Perfusion 2002;17:199–204 S. Postoperative antithrombin levels and outcome in cardiac 16. Lobato RL, Despotis GJ, Levy JH, Shore-Lesserson LJ, Carlson operations. Crit Care Med 2005;33:355–60 MO, Bennett-Guerrero E. Anticoagulation management during 24. Fourrier F, Chopin C, Goudemand J, Hendrycx S, Caron C, cardiopulmonary bypass: a survey of 54 North American Rime A, Marey A, Lestavel P. Septic shock, multiple organ institutions. J Thorac Cardiovasc Surg 2010;139:1665–6 failure, and disseminated intravascular coagulation: compared 17. Brister SJ, Ofosu FA, Buchanan MR. Thrombin generation patterns of antithrombin III, protein C, and protein S deficien- during cardiac surgery: is heparin the ideal anticoagulant? cies. Chest 1992;101:816–23 Thromb Haemost 1993;70:259–62 25. Eid A, Wiedermann CJ, Kinasewitz GT. Early administration of 18. Slaughter TF, LeBleu TH, Douglas JM Jr, Leslie JB, Parker JK, high-dose antithrombin in severe sepsis: single center results Greenberg CS. Characterization of prothrombin activation from the KyberSept-trial. Anesth Analg 2008;107:1633–8 during cardiac surgery by hemostatic molecular markers. Anesthesiology 1994;80:520–6

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 835 EDITORIAL

Bridging the Gap to Reduce Surgical Site Infections

Frank J. Overdyk, MSEE, MD

urgical site infections (SSI) remain one of the most perfusion, reduces SSI, and shortens hospitalization.4 Mild persistent and costly preventable complications in intraoperative hypoventilation has been shown to increase health care. Although as anesthesiologists, we may tissue oxygen levels due to the vasodilating effect of carbon S 5 perceive our responsibility limited to the timely administra- dioxide in obese patients. More clarity may be brought to tion of prophylactic antibiotics and sterile placement of inva- the optimal intraoperative fluid, temperature, ventilation, sive catheters, we feel the pain of our surgical colleagues and oxygenation strategies that minimize SSI if we could when their service suffers a rash of SSI, and they resort to easily measure subcutaneous tissue oxygen partial pressure strategies ranging from ones with sound scientific basis to (Sto2). Arguably, Sto2 represents the common pathway of mere superstition. Over the years, I have seen operating cellular “contentment” during the “decisive period,” which rooms undergo lockdowns worthy of a maximum-security is defined as the hours surrounding surgical incision in prison, bathed in dance club–like ultraviolet light, or trans- which prophylactic antibiotics have been shown to have the formed to resemble a spacewalk rehearsal. Yet the problem greatest impact on reducing SSI.6 of SSI persists, second only to urinary tract infections In this issue of Anesthesia & Analgesia, Govinda et al. use among the most common types of nosocomial infections in a noninvasive infrared spectroscope (NIRS) to measure hospitals.1 Approximately 1 in 50 surgical procedures are Sto2 as a predictor of SSI that manifests as late as 30 days complicated by SSI, which translates to an estimated postoperatively.7 The authors used a receiver operator 290,485 cases annually, of which 13,088 result in death. It is curve to optimize the sensitivity and specificity of upper- no surprise that SSI have made the 2010 Joint Commission arm Sto2 measurements to differentiate patients who de- National Patient Safety Goals for the second consecutive velop an SSI from those who do not. They found Sto2 below year.* 66% measured 75 minutes after colonic surgery on the The importance of high tissue oxygen levels in im- upper arm to be predictive for SSI diagnosed, on average, 9 proved surgical outcomes and reduced surgical infection days after surgery. This NIRS measurement was a better rates has been understood for many years. High tissue predictor for SSI than were 2 clinically accepted SSI risk oxygen concentrations enhance the effects of leukocytes assessment metrics, the National Nosocomial Infections and antibiotics on microbes and are indicative of adequate Surveillance System and Study on the Efficacy of Nosoco- tissue perfusion. Oxygen delivery to the tissues, a function mial Infection Control. of both bloodflow and oxygen-carrying capacity, has been As an introductory study on this technology, its limitations the underpinning of many of the nonpharmacologic strat- are several and well delineated in the article. However, if this 2 egies to prevent surgical wound infection. Traditionally, result is reproducible upon further investigation, we may liberal intraoperative fluid management was thought to have an important “early warning system” for SSI and a tool improve outcomes by improving perfusion, yet more re- with which to dissect the problem of SSI into 2 components. cent data suggest that tissue edema from overly aggressive First, this noninvasive technique may allow us to investigate hydration may be detrimental. We remain without a con- the different intraoperative anesthetic strategies involving sensus on the optimal volume replacement strategy for fluid management, ventilation, oxygenation, hemodynamic optimal surgical outcomes.3 Similarly, although the sym- indices, and temperature that optimize Sto2 during the imme- pathectomy accompanying regional anesthesia improves diate postoperative period. Second, once we optimize and tissue perfusion and certain outcomes (such as preserved control for Sto , we can explore additional postoperative graft patency in vascular surgery and decreased incidence 2 strategies to minimize SSI. The current debate on the of deep vein thrombosis), its impact on SSI is unclear. efficacy of high intraoperative and postoperative Fio in Avoiding intraoperative hypothermia helps maintain tissue 2 reducing SSI demonstrates in which cases this Sto2 measurement may be helpful. A recent meta-analysis of 5 *From http://www.jointcommission.org/PatientSafety/NationalPatient SafetyGoals/. randomized, double-blind clinical trials found a relative risk reduction for SSI of 33% when hyperoxic gas mixtures From Medical University of South Carolina, Charleston, South Carolina. Ϸ (Fio2 80%) were used intraoperatively and immediately Accepted for publication June 14, 2010. postoperatively for 2 to 6 hours versus a normoxic (Fio2 Address correspondence to Frank J. Overdyk, MSEE, MD, Medical Univer- Ϸ30%) mixture.8 Yet, few anesthesiologists and surgeons sity of South Carolina, 167 Ashley Avenue, Suite 301, MSC 912 Charleston SC 29425-9120. Address e-mail to [email protected]. have adopted this approach, perhaps because a benefit Copyright © 2010 International Anesthesia Research Society realized only once in every 33rd patient (number needed to DOI: 10.1213/ANE.0b013e3181ee85b1 treat) may not justify the inconvenience and risks of

836 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Noninvasive Infrared Spectroscope and Surgical Site Infection patients wearing nonrebreather oxygen masks up to 6 better sensitivity, if most of the intra- and postoperative hours postoperatively. Atelectasis and increased pulmo- risk mitigation strategies can be implemented cost- nary complications in the hyperoxic group, particularly in effectively. NIRS should help answer these questions. obese patients, remain a concern.9 Furthermore, these re- As the above discussion illustrates, SSI is not merely a sults were contradicted by a randomized control trial surgical issue, but one that anesthesiologists influence powered to detect an identical 33% relative risk reduction through intraoperative decisions and postoperative pain between an Fio2 of 80% and one of 30%, continued for only management strategies. This may be reflected at some point 10 2 hours postoperatively. That study found no decreased by the inclusion of SSI in the National Anesthesia Clinical risk of SSI with the higher inspired oxygen concentration. Outcomes Registry.† Our involvement in investigating and Was the shorter duration of postoperative hyperoxia deci- reducing SSI will be appreciated by our administrators, sive? Or were intraoperative fluid and temperature man- surgeons, and our patients, and we will continue to be agement suboptimal for preventing SSI, as is suggested by recognized as critical and invaluable members of the peri- 11 Hunt and Hopf in an accompanying editorial? Perhaps operative team. immediate postoperative Sto2 measurements will give us a “halftime score,” which we can use to alter the game plan, REFERENCES and concentrate on postoperative interventions to reduce 1. Klevens RM, Edwards J, Richards C, Horan T, Gaynes R, SSI, such as high supplemental oxygen concentrations, Pollock D, Cardo D. Estimating health care–associated infec- effective pain management, postoperative warming, and tions and deaths in U.S. hospitals. Public Health Rep 2007; appropriate discontinuation of prophylactic antibiotics 122:160–66 within 24 hours of surgery end.12 However, if Sto mea- 2. Sessler D. Non-pharmacologic prevention of surgical wound 2 infection. Anesthesiol Clin 2006;24:279–97 surements in the recovery room suggest that the “SSI die is 3. Chappel D, Jacob M, Hofmann-Kiefer K, Conzen P, Rehm M. cast,” it is clear that our efforts to reduce SSI should be Rational approach to perioperative fluid management. Anes- focused on intraoperative strategies, including those not thesiology 2008;109:723–40 associated with oxygenation and perfusion, such as timely 4. Kurz A, Sessler D, Lenhardt R. Perioperative normothermia to antibiotic prophylaxis within 1 hour before surgical inci- reduce the incidence of surgical wound infection and shorten hospitalization. NEJM 1996;334:1209–15 sion and tight control of blood glucose levels. 5. Hager H, Reddy D, Mandadi G, Pulley D, Eagon JC, Sessler D, Additional work with Sto2 is warranted. Although dif- Kurz A. Hypercapnia improves tissue oxygenation in mor- ferences in dependent variables such as intraoperative bidly obese surgical patients. Anesth Analg 2006;103:677–81 temperature, volume replacement, hemodynamics, surgical 6. Classen DC, Evans RS, Pestotnik SL, Horn SD, Menlove RL, duration, and surgical technique between the patients with Burke JP. The timing of prophylactic administration of antibiotics and the risk of surgical-wound infection. N Engl J Med and without SSI were nonsignificant in this study, the 1992;326(5):281–6 patients with SSI weighed more and had more pain, and 7. Govinda R, Kasuya Y, Bala E, Mahboobi R, Devarajan J, Sessler Fio2 during NIRS measurements was not controlled. D, Akca O. Early postoperative subcutaneous tissue oxygen Clearly, the impact of these confounding variables on the predicts surgical site infection. Anesth Analg 2010;111:946–52 results must be clarified. This study also contradicted 8. Qadan M, Akca O, Mahid S, Hornung C, Polk H. Perioperative previous work in terms of the anatomic location of the Sto supplemental oxygen therapy and surgical site infection: a 2 meta-analysis of randomized controlled trials. Arch Surg measurement, suggesting that measurement location is not 2009;144(4):359–66 merely nuance, but may have important ramifications in 9. Zoremba M, Dette F, Hunecke T, Braunecker S, Wulf H. The terms of the reproducibility of these results. influence of perioperative oxygen concentration on postopera- Lastly, the trade-off between sensitivity and specificity tive lung function in moderately obese adults. Eur J Anaesthe- in determining a cutoff point (Sto Ͻ66%) needs to be siol 2010;27:501–7 2 10. Meyhoff C, Wetterslev J, Jorgensen L. For the PROXI random- considered in the context of the outcome. This threshold for ized clinical trial. JAMA 2009;302(14):1543–50 Sto2 was chosen by the authors from the receiver operator 11. Hunt T, Hopf H. High inspired oxygen fraction and surgical curve to optimize both sensitivity and specificity. In the site infection. JAMA 2009;302:1588–9 case of SSI, the annual cost to hospitals is approximately 12. Bratzler DW, Houck PM. Antimicrobial prophylaxis for sur- $7.5 billion, and the toll on patients and their families gery: an advisory statement from the National Surgical Infec- 13 tion Prevention Project. Amer J Surg 2005;189:395–404 incalculable. It may be wise to sacrifice specificity for 13. Stone PW, Braccia D, Larson E. Systematic review of economic analyses of health care–associated infections. Am J Infect †From http://www.aqihq.org/. Accessed May 25, 2010. Control 2005;33:501–9

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 837 EDITORIAL

Labor Pain, Analgesia, and Chronobiology: What Factor Matters?

Yvan Touitou, PhD,* Garance Dispersyn, PhD,† and Laure Pain, MD†

lleviation and control of pain are of major impor- investigated, considerably greater than that of previous tance in medicine. Depending on the causes and on studies of this nature. Athe clinical situations, the patterns of pain may The paper by Shafer et al.2 represents the final product vary during the day, with peak and trough times reported of a manuscript initially submitted to Anesthesia & Analge- at different times of day, although sometimes with sia, reporting the chronobiology of the duration of analgesia contradictory results.1 The therapeutic effects of local following intrathecal bupivacaine. After peer review and anesthetics and opioid analgesics also demonstrate cir- extensive reanalysis of the original data, the manuscript has cadian variations.1 evolved into a much more interesting exploration of the Two articles in this issue of Anesthesia & Analgesia report methods used in chronobiological analysis to detect the investigations in parturients into the relationship between influence of external factors and in particular how one can time of administration of intrathecal bupivacaine,2 or intra- use these methods to detect possible artifacts in the data. thecal fentanyl versus systemic hydromorphone,3 and the This is an important illustration of the value of a well- duration of analgesia. In both studies, the duration of conducted peer review process. 2 analgesia was defined as the time from the first adminis- Shafer et al. make the important point in their paper tration of analgesia during labor until the second request that very different conclusions can be drawn, depending on for analgesia. The 2 groups of investigators approached the the statistical methods used to analyze the data. They used analysis of the chronobiology of analgesic requirements 3 different smoothing functions to explore circadian during labor from somewhat different perspectives. Scav- rhythms graphically. These revealed a bimodal pattern in one et al.3 addressed the problem from a clinical perspec- the duration of analgesia, with 1 peak at around 0630 hours tive, with their main objective being to determine whether and a subsequent peak in the afternoon or evening. In time-related effects might have confounded the results of a contrast, an analysis of variance (ANOVA) did not show previous study by the group.4 Shafer et al.,2 on the other any significant difference in the duration of analgesia, hand, used their clinical data to examine the problems that irrespective of the timing of the intrathecal injection of arise when developing models to analyze the often com- bupivacaine. Fitting the data to a simple cosine function of plex periodic data obtained from clinical situations. analgesia duration versus time demonstrated a periodic Scavone et al.3 compared neuraxial versus systemic waveform with a period of 8 hours, but with peaks that opioid analgesia in 692 healthy parturients early in labor. corresponded with only 1 of the 3 peaks identified by the At the first request for analgesia, patients in the neuraxial smoother functions. The authors then used a bootstrap group were given intrathecal fentanyl 25 ␮g, whereas analysis to show that 2 individual points were responsible patients in the systemic analgesia group received 1 mg IV for the statistical significance in their cosine fit. Removing and 1 mg IM hydromorphone. Subjects given their initial these points from the dataset also removed the previously analgesia between 0701 hours and 2300 hours were consid- observed rhythmic effect of intrathecal bupivacaine on analgesia. The authors concluded that these 2 points were ered as the daytime group; those given their first analgesic likely to be artifacts, though they acknowledge this might between 2301 hours and 0700 hours, the nighttime group. be uncertain, corresponding to the change in nursing and The authors found no difference in the median duration of anesthesia shifts. either neuraxial or systemic analgesia, irrespective of the The literature contains conflicting reports on the time- time of administration. A major advantage claimed by the dependent effects of neuraxial local anesthetics and opi- authors for their study was the large number of patients oids.1 These differences may be a consequence of the wide variety of factors that need to be considered when inter- From *Unite´ de Chronobiologie, Fondation Ophtalmologique A. de Roths- child, Paris, France; and †INSERM U666 (GRERCA), CHU de Strasbourg, preting the results from chronobiology experiments. They Faculte´deme´decine, Strasbourg, France. also emphasize the importance of a very stringent experi- Accepted for publication June 6, 2010. mental protocol, including methodology and statistical Address correspondence to Yvan Touitou, Chronobiology Unit, Fondation methods used to analyze the potential rhythmic effects of a Ophtalmologique A. de Rothschild, 29 rue Manin, 75019 Paris, France. Address e-mail to [email protected]. drug. Many factors, both internal and external, can influ- Copyright © 2010 International Anesthesia Research Society ence the biological rhythms of a drug’s action. These DOI: 10.1213/ANE.0b013e3181ee85d9 include temporal variations in light–dark, rest–activity,

838 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Labor Pain, Analgesia, and Chronobiology fasting–eating cycles, and other environmental factors such study dealing with biological rhythms to avoid potential as stress, alcohol consumption, tobacco use, and caffeine biases. intake. All of these are able to alter the parameters charac- Many exogenous factors (masking agents) can influence terizing a biological rhythm and should be considered as circadian patterns by masking the endogenous rhythms of cofactors that can mask or unmask any circadian effects of the biological clock. A circadian rhythm can be masked by drugs.5 When studying labor pain, specific obstetrical fac- any environmental signal to which the organism is sensi- tors should be controlled for, because they may have tive.13 To unmask the endogenous circadian structure, an considerable influence on pain intensity, e.g., parity, spon- experimental protocol called “constant routine” can be taneous or pharmacologically induced labor, stage of cer- used. This involves subjects staying awake but lying down vical dilation, rupture of the membrane, labor duration, for 24 to 36 hours, in an environment of constant tempera- and pharmacological induction of uterine contractions ture, humidity, and lighting, with identical and regularly (oxytocin administration).6–8 spaced meals. Though this protocol is considered by many Ideally, although not always feasible, studies should con- as the gold standard, it obviously presents several limita- sider subpopulations of patients, considering all of these tions,14 and for research in fields such as obstetrics is different clinical situations. As an example, one may analyze seldom practicable. the duration of intrathecal analgesia in a population of partu- Subtracting the estimated exogenous or “masking” com- rients, nulliparous, with cervical dilation at 2 cm, with no ponent of a rhythm from the observed rhythm may reveal administration of oxytocin at all, and find circadian variations the underlying endogenous component. Regression models that could be masked in another study examining a different have been developed to identify masking effects, but sepa- subgroup of population. This might explain the discrepancies ration of the exogenous from the endogenous component is between the published studies. However, other factors can difficult.5 When the data can be approximated by a sinu- bias results. Although factors such as the timing of the shift soidal model, the cosinor function allows calculation of changes in the work team can bias results, one needs to be rhythm parameters such as the mesor (the mean level that cautious about this type of explanation. For example, the time is equal to the 24-hour average), amplitude (half of the when the morning nursing shift begins could coincide with peak-to-trough difference of the fitted cosine function), and the peak in cortisol, which is lowest in the evening and acrophase (the peak time of the rhythm given in degrees or reaches its highest concentration during the early waking hours and minutes). When the number of subjects is small hours. Which factor is then influencing the duration of anal- or the density of samplings low, the interindividual vari- gesia, the change of nursing team or the endogenous circadian ability should be smoothed. Nonparametric tests are useful rhythm of cortisol? because they ensure that the absolute values do not influ- Although biological rhythms are genetically deter- ence the results. Whatever methods are used, they should mined, they are continuously modulated (adjusted in time) be critically evaluated for their ability to test the hypotheses by periodic events in the environment called synchroniz- 5 under question and to assess the time series data of the ers. In mammals, the main synchronizer is light. Rhythm variable(s) being studied. synchronization of subjects is the “gold standard” of any Finally, why are the studies by Scavone et al.3and Shafer chronobiological study protocol. This means that, in addi- et al.2 important? Both studies found that time of day did tion to the factors discussed above, the subjects should keep not appear to influence the duration of analgesia produced as close as possible to their usual lifestyle, maintaining the by intrathecal local anesthetics or opioids. Knowledge of environmental factors (synchronizers) that impose their the time dependency of drugs is important because drug period on biological rhythms, such as sleep–wake time, effect can be optimized and toxicity minimized by basing rest–activity cycle, light–dark cycle, fasting–eating, etc.5 drug administration on the circadian patterns of drug Thus, it is advisable to verify the synchronization of the activity. Several studies have demonstrated circadian time– subjects’ circadian time organization by checking rhythm dependent changes in the toxicity and the pharmacokinetic markers (body temperature, cortisol, melatonin, etc.) as an disposition of local anesthetics.15 Shafer et al., in addition to initial step or as an integral aspect of the investigative demonstrating the importance of the peer review process, protocol.9 The importance of the synchronizers that influ- provide a set of clear guidelines that future investigators in ence endogenous rhythms varies according to the variable this field should find valuable, and hopefully help them or the function under investigation. Some synchronizers avoid many of the potential pitfalls in chronobiological are more predominant than others, e.g., the light–dark research. cycle and the sleep–wake activity in humans. In some experimental situations, however, synchronizers that otherwise might be considered minor can assume predomi- ACKNOWLEDGMENTS nance. Changes in any one of these synchronizers may lead This editorial was solicited and handled by Dr. James Bovill, to changes in the temporal relationship between biological Guest Editor-in-Chief for Anesthesia & Analgesia. rhythms.5 Circadian rhythms can also be subject to seasonal modu- REFERENCES lation, and different circadian rhythm patterns have been 1. Bruguerolle B, Labrecque G. Rhythmic pattern in pain and described in the same individuals but at different times of their chronotherapy. Adv Drug Del Rev 2007;59:883–95 10–12 2. Shafer SL, Lemmer B, Boselli E, Boiste F, Bouvet L, Allaouch- the year. Failure to consider this could explain some iche B, Chassard D. Pittfalls in chronobiology: a suggested of the apparently discrepant results in the literature. All of analysis using intrathecal bupivacaine analgesia as an ex- those factors need be considered and standardized in any ample. Anesth Analg 2010;111:980–5

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 839 EDITORIAL

3. Scavone BM, McCarthy RJ, Wong CA, Sullivan JT. The influ- 10. Touitou Y, Lagoguey M, Bogdan A, Reinberg A, Beck H. ence of time of day of administration on duration of opioid Seasonal rhythms of plasma gonadotrophins: their persistence labor analgesia. Anesth Analg 2010;111:986–91 in elderly men and women. J Endocrinol 1983;96:15–21 4. Wong CA, Scavone BM, Peaceman AM, McCarthy RJ, Sullivan 11. Levi F, Canon C, Touitou Y, Reinberg A, Mathe´ G. Seasonal JT, Diaz NT, Yaghmour E, Marcus RJ, Sherwani SS, Sproviero modulation of the circadian time structure of circulating T and MT, Yilmaz M, Patel R, Robles C, Grouper S. The risk of natural killer lymphocyte subsets from healthy subjects. J Clin cesarean delivery with neuraxial analgesia given early versus Invest 1988;81:407–13 late in labor. N Engl J Med 2005;352:655–65 12. Reinberg A, Touitou Y, Levi F, Nicolai A. Circadian and 5. Touitou Y, Haus E, eds. Biological Rhythms in Clinical and seasonal changes in ACTH-induced effects in healthy young Laboratory Medicine. Berlin: Springer, 1992 men. Eur J Clin Pharmacol 1983;25:657–65 6. Sheiner E, Sheiner EK, Shoham-Vardi I. The relationship between parity and labor pain. Int J Gynecol Obstet 1998;63:287–8 13. Mrosovsky N. Masking: history, definitions, and measure- 7. Aya AGM, Vialles N, Mangin R, Robert C, Ferrer JM, Ripart J, ment. Chronobiol Int 1999;16:415–29 de la Courssaye JE. Chronobiology of labour pain perception: 14. Waterhouse JM, Redfern P, Minors DS. An introduction to an observational study. Br J Anaesth 2004;93:451–3 circadian rhythms and their measurement in humans. In: 8. Veira WS, Hidalgo MPL, da Silva LT, Caumo W. Biological Redfern P, ed. Chronotherapeutics. London: Pharmaceutical rhythms of spinal-epidural labor analgesia. Chronobiol Int Press, 2003:1–30 2010;27:865–78 15. Chassard D, Bruguerolle B. Chronobiology and anesthesia. 9. Selmaoui B, Touitou Y. Reproducibility of the circadian Anesthesiology 2004;100:413–27 rhythms of serum cortisol and melatonin in healthy subjects. A study of three different 24-h cycles over six weeks. Life Sci 2003;73:3339–49

840 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA EDITORIAL

Can an Acute Pain Service Be Cost-Effective?

Eric Sun, MD, PhD,*† Franklin Dexter, MD, PhD,‡ and Alex Macario, MD, MBA§

n many countries around the world, the anesthesiolo- representative of 90% of all hospital discharges in the USA. gist is the primary physician responsible for pain con- For 2008, the year with the most recently available data, Itrol in the first 24 hours after surgery. However, in the there were approximately 9.1 million inpatients having United States of America (USA), postoperative analgesia is surgery in the USA. This constitutes approximately 23% of typically managed by the surgeon. This is because their all inpatient visits, and a 4% annual growth rate since 1998 professional fee includes this responsibility while the pa- (Table 1). These data exclude federal and prison hospitals. tient remains in the hospital and when the patient returns The data in Table 1 provide insight into the size and home. At the other end of the spectrum is a dedicated scope of the demand for pain management in the postop- Acute Pain Service team with expertise and authority for erative setting. From a national health policy perspective, managing a patient’s surgical pain. there are many different surgical procedures.2 However, The well-done randomized clinical trial conducted by cesarean deliveries, orthopedics, and general surgery pro- Lee et al.1 and described in this issue of the journal cedures should be a priority for optimizing clinical acute estimates the cost-effectiveness of an anesthesiologist-led, postoperative pain care because they account for a large nurse-based Acute Pain Service, mainly charged with man- fraction of all inpatient surgery. aging IV patient-controlled analgesia. The control group With public and private payers continuing to face tight- consisted of patients receiving IM or IV boluses of opioids ening budgets, and with single bundled episode of care as needed by nurses on the ward.1 It is not known how payments for hospital and physician care looming, it is an frequently this technique is used in other parts of the opportunity for anesthesiologists to consider additional world, including the USA. venues where the specialty can provide value. The dedi- Our objectives for this editorial were to: cated Acute Pain Service seems to be a logical choice, given • Estimate the total annual number of surgical inpa- that it is a straightforward extension of the care anesthesi- tients in the USA that might benefit from care pro- ologists already provide in the operating room and chronic vided by an Acute Pain Service. pain settings. However, given the expenses of such a • Define possible structures and functions of an Acute service, what are the incentives for hospitals and anesthe- Pain Service because these may vary among institutions, siologists to participate? On the clinical side, controlling and affect the economic viability of such a service. pain is of importance to patients, a goal made more difficult • Outline research questions that deserve further study by the increasing number of patients who are admitted to help hospitals make the needed investment in an with chronic pain syndromes including opioid prescrip- Acute Pain Service. tions superimposed on their acute surgical pain. The potential clinical benefits of a dedicated Acute Pain Service have resulted in support from a variety of organizations HOW MANY PATIENTS ARE AT RISK FOR PAIN such as the American Society of Anesthesiologists,3 the AFTER INPATIENT SURGERY IN THE USA? Royal College of Surgeons, and the College of Anesthetists. To obtain an estimate of the number of inpatients having In addition to these clinical benefits, an Acute Pain Service surgery, the Nationwide Inpatient Sample from the Health- provides important visibility to the anesthesia group within a care Cost and Utilization Project was used. It samples 20% hospital. Patient satisfaction with pain control comprises an of inpatient discharges from 1056 academic, community, important component of many measures of hospital quality, and acute care hospitals in 42 states, and is intended to be such as Press Ganey scores. Many facilities are providing resources to an Acute Pain Service to increase these scores. In From the *RAND Corporation, Santa Monica; †the Department of Health fact, a patient’s Press Ganey response to, “How well your pain Services, University of California, Los Angeles, Los Angeles, California; ‡University of Iowa, Iowa City, Iowa; and §Department of Anesthesia, was controlled” was the second most important variable Stanford University School of Medicine, Stanford, California. correlated to whether a patient recommended the hospital to Accepted for publication July 9, 2010. someone else. By the way, the number one item was “Staff Dr. Sun is funded by the Agency for Healthcare Research and Quality worked together to care for you.”4 through the UCLA/RAND training grant T32 HS 000046. Disclosure: The authors report no conflicts of interest. POSSIBLE STRUCTURES AND FUNCTIONS OF AN Address correspondence and reprint requests to Alex Macario, MD, MBA, ACUTE PAIN SERVICE Department of Anesthesia H3580, Stanford University School of Medicine, Stanford, CA 94305-5640. Address e-mail to [email protected]. Despite these benefits, the use of a dedicated Acute Pain Copyright © 2010 International Anesthesia Research Society Service to manage postoperative pain did not gain wide- 5–7 8,9 DOI: 10.1213/ANE.0b013e3181f33533 spread adoption in the 1990s in the USA or abroad.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 841 EDITORIAL

Table 1. Thirty Most Common Diagnosis Related Groups for Inpatients Having Surgery Requiring Anesthesia Discharges coded without Discharges coded with complications and complications and Total in the DRG name comorbidities comorbidities USA in 2008 Cesarean 904,000 472,000 1,376,000 Major joint replacement or reattachment of lower extremity 912,000 56,000 968,000 Uterine and adnexa procedure for nonmalignancy 430,000 97,000 527,000 Major small and large bowel procedures 95,000 275,000 370,000 Laparoscopic cholecystectomy without common duct exploration 191,000 137,000 328,000 Hip and femur procedures except major joint (e.g., hip fracture) 85,000 169,000 254,000 Appendectomy 196,000 26,000 222,000 Back and neck procedures except spinal fusion 157,000 51,000 208,000 Spinal fusion except cervical 198,000 9100 207,100 Lower extremity and humerus procedures except hip, foot, femur 149,000 54,000 203,000 (e.g., tibia fracture) Cervical spinal fusion 123,000 26,000 149,000 Major cardiovascular procedures 84,000 53,000 137,000 OR procedures for obesity 109,000 20,000 129,000 Craniotomy and endovascular intracranial procedures 58,000 66,000 124,000 Extensive OR procedure unrelated to principal diagnosis 24,000 99,000 123,000 Major chest procedures 37,000 86,000 123,000 Extracranial procedures 85,000 25,000 110,000 Coronary bypass with cardiac catheterization 68,000 38,000 106,000 Other respiratory system OR procedures 14,000 92,000 106,000 Vaginal delivery with sterilization and/or D&C 101,000 None 101,000 Major male pelvic procedures 79,000 17,000 96,000 Infectious and parasitic diseases with OR procedure 1000 91,000 92,000 Transurethral prostatectomy 44,000 45,000 89,000 Peritoneal adhesiolysis 35,000 53,000 88,000 Coronary bypass without cardiac catheterization 63,000 24,000 87,000 Revision of hip or knee replacement 41,000 43,000 84,000 Hernia procedures except inguinal and femoral 50,000 33,000 83,000 Appendectomy with complicated principal diagnosis 48,000 24,000 72,000 Cardiac valve and other major cardiothoracic procedures without 12,000 59,000 71,000 cardiac catheterization Total 4,393,000 2,240,100 6,633,100 Note: Because these are hospital discharges, a patient could have had more than 1 visit to the operating room for a surgical procedure during 1 hospitalization. Thus, the actual number of operations is likely higher than the numbers reported here. DRG ϭ diagnosis related group; USA ϭ United States of America; OR ϭ operating room; D&C ϭ dilatation and curettage.

Sometimes Acute Pain Services are services in name only. The heterogeneity in structure and function of pain For example, 36% of 446 hospitals in Germany operated services across facilities makes it difficult to state unequivo- an Acute Pain Service but half did not have specific cally whether an Acute Pain Service is or is not cost- personnel assigned to the service, policies for nights and effective. To define the exact nature of an Acute Pain weekends, written protocols for pain management, or Service is a crucial component of its economic study. regular assessment and documentation of pain scores at Anesthesia groups in the USA currently are likely to least once a day.10 The challenges are in part financial, provide an Acute Pain Service for a variety of reasons such as how coding/billing affects profitability, but also such as: include organizational change barriers such as compet- • ing managerial and clinical agendas, staff shortages, An anesthesia residency training program exists at the local politics, professional hierarchies, and workload hospital. As of 2008, a month-long acute pain experi- changes.11 ence for anesthesia trainees is required. • In Italy, although 42% of 163 public hospitals had an A hospital’s practice is to place a large number of organized Acute Pain Service, continuous IV analgesia peripheral nerve catheters, which then extends the using elastomeric infusion systems was the most frequently anesthesiologist’s responsibility longer and leads to an used analgesic technique, and IV patient-controlled analge- Acute Pain Service. sia and epidural techniques were used in fewer than 14% of • The anesthesia group is paid for its service by third patients.12 In Ireland, a 2007 study showed that 71% party payers such as the federal government or pri- of teaching hospitals had a formalized Acute Pain Service, vate insurance, and the marginal revenue exceeds the of which 85% were established after 1990.13 marginal costs. Moreover, where a dedicated Acute Pain Service exists, • The anesthesia group’s contract and professional there is variability in its structure and the types of services service agreement with the hospital stipulates its it provides. To illustrate the variability in Acute Pain existence. The hospital may subsidize the Acute Service staffing and services provided at several USA Pain Service budget if there is a perceived competi- hospitals, we created Table 2, which is based on informal tive advantage to the hospital to attract more communication with colleagues at those facilities. cases.

842 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Acute Pain Service Economics

Table 2. Stafﬁng and Care Provided by Acute Pain Services Hospital AB C DEFG Anesthesiologist (FTEs/d) 0.8 1 1 1 1 1 X Housestaff (FTEs/d) 1 1 1 No 1 1 No RN (FTEs/d) No 1 0 No 1 No No NP (FTEs/d) No No 1 No No No No Other staff Fellow Pharmacist No No No No No No. of patients/d 12 18 12 (Monday)–30 4 10–20 8–20 10–15 (Friday) Services 0 ϭ never provided; 10 ϭ most frequently provided service Manage routine IV PCA 0 0 0 2 1 0 10 Manage complex IV PCA (e.g., patients 108 3 7382 on chronic pain medications) Manage peripheral nerve catheter/ 10 10 10 10 10 3 0 infusions Manage intraoperative spinal opioids 10 10 0 2 2 1 3 Consult on nonsurgical acute pain (e.g., 103 5 2431 lumbar puncture, epidural catheters for rib fractures) Manage epidural catheter 10 10 10 0 9 10 1 Outpatient nerve block catheters No Yes No No Yes No No 24-h coverage Yes Yes Yes Yes Yes Yes Yes Note: Hospitals listed in this table are academic medical centers except hospitals D and G, which are community hospitals. Hospital D has no formal acute pain service in place but it is being proposed for funding. Hospital G has the post 2nd call person round on the patients in the morning, and the 1st call person answers the calls at all other times. FTE ϭ full time equivalent; RN ϭ registered nurse; NP ϭ nurse practitioner; PCA ϭ patient-controlled analgesia.

• The surgeons’ desire to deflect the night phone calls considered is some measure of pain scores. The Quality-of- and pain control questions to the anesthesia group Recovery instrument, for instance, has 9 questions, at least especially if the anesthesia group is being paid for 4 of which pertain to pain or side effects from analgesics.15 taking calls (depending on the contract) whereas the However, a challenge for the health economist is that these surgeons are not. measures need to be translated into units more easily understood by decision makers. Financial considerations are often largely responsible for For example, efforts could be made to translate these the slow adoption of a dedicated Acute Pain Service. pain scores to quality-adjusted life years, the measure of Although the study by Lee et al. in this issue of the journal, 16 14 efficacy used by many public payers. However, the time along with the studies that preceded it, represent important first steps, more research is needed to find ways to horizon in acute surgical care studies is typically measured overcome organizational constraints and deliver the spe- in days and weeks and not months or years as for treat- cialized care of an Acute Pain Service in an economically ments aimed at relieving chronic conditions. As a result, a viable manner. dedicated Acute Pain Service may have little effect on the conventional metric of quality-adjusted life years. There- RESEARCH QUESTIONS THAT DESERVE fore, additional efforts should be made to define what FURTHER STUDY measures payers should consider, and present economic The literature leaves unanswered the question of what ser- analyses in a way that practicing clinicians and payers can vices a dedicated Acute Pain Service should provide and the understand and translate to their practice. most cost-effective method in which to provide them. Studies Future studies could also report alternative outcomes of to date have compared a wide variety of staffing models and relevance to hospital administrators. For example, a group of services to more conventional methods of postoperative pain pain service interventions increased Press Ganey measure- management. However, no studies have explicitly compared ments of patient satisfaction from the 87th to the 99th percen- 17 various Acute Pain Service models with each other. In all tile. More understanding is needed regarding which of the likelihood, the most cost-effective way to implement an Acute several reported interventions had the largest impact. Pain Service will vary among hospitals because of local Efforts to date to estimate the efficacy of a dedicated Acute variations in culture and personnel, for example. In addition, Pain Service have included consideration of the average effect smaller hospitals that perform a range of surgical cases may in a study population. However, this approach may obscure face more variability in the demand for expert postoperative the utility of an Acute Pain Service for 2 reasons.18 The first pain management. This observation is based on the law of reason concerns differences in patient responses. If a dedicated large numbers, which states that there is more variability in Acute Pain Service provides a large benefit for a small portion smaller samples. Such low volume hospitals will need to find of the population, the average effect will be small, but clearly ways to manage this variability, such as pooling staff and the Acute Pain Service still has an important clinical role. resources across hospitals. The second reason concerns dependence in responses More work also needs to be done to quantify the efficacy across different alternatives to pain management. Put sim- of a dedicated Acute Pain Service. A typical outcome ply, the issue here is whether patients who fail to respond

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 843 EDITORIAL to basic methods of pain management are likely to see large REFERENCES improvement with the expertise of an Acute Pain Service. If 1. Lee A, Chang SKC, Chen PP, Gin T, Lau ASC, Chiu CH. The an Acute Pain Service is likely to provide large benefits for costs and benefits of extending the role of the acute pain service on clinical outcomes after major elective surgery. patients who are not responding to other modalities, the Anesth Analg 2010;111:1042–50 service also has an important clinical role, even if its 2. Macario A. Truth in scheduling: is it possible to accurately average effect (measured across a larger population) is predict how long a surgical case will last? Anesth Analg small. This latter possibility is particularly relevant in some 2009;108:681–5 hospitals where the anesthesiologist-staffed Acute Pain 3. American Society of Anesthesiologists Task Force on Acute Service typically focuses on particularly difficult cases that Pain Management. Practice guidelines for acute pain management in the perioperative setting: an updated report by the have been refractory to more conventional methods. Ulti- American Society of Anesthesiologists Task Force on Acute mately, these factors suggest that a dedicated Acute Pain Pain Management. Anesthesiology 2004;100:1573–81 Service can improve its cost-effectiveness by targeting its 4. HospitalPulseReport2009.Availableat:http://www.pressganey. efforts toward patients most likely to benefit. Unfortu- com/galleries/default-file/Hospital_Pulse_Report_2009.pdf. Ac- nately, our ability to identify these patients remains limited cessed June 15, 2010 5. Carr DB, Miaskowski C, Dedrick SC, Williams GR. Manage- and is another area for further study.19 ment of perioperative pain in hospitalized patients: a national Finally, in calculating the costs of an Acute Pain Service, survey. J Clin Anesth 1998;10:77–85 the present study focuses on the costs of the service itself, 6. Ready LB. How many acute pain services are there in the measured as staff time and the costs of administered United States, and who is managing patient-controlled analge- medications. Personnel costs will differ among countries sia? Anesthesiology 1995;82:322 both in the absolute cost of an anesthesiologist and nurse, 7. Warfield CA, Kahn CH. Acute pain management: programs in U.S. hospitals and experiences and attitudes among U.S. and in the relative cost of one to the other. This markedly adults. Anesthesiology 1995;83:1090–4 affects the optimal mix of providers for an Acute Pain 8. Rawal N, Allvin R. Acute pain services in Europe: a 17-nation Service.20 In addition, these measures of costs are incom- survey of 105 hospitals. The EuroPain Acute Pain Working plete, because they do not account for the net costs, taking Party. Eur J Anaesthesiol 1998;15:354–63 into account the possibility that improved pain manage- 9. Powell AE, Davies HT, Bannister J, Macrae WA. Rhetoric and ment may reduce costs by shortening hospital stay or reality on acute pain services in the UK: a national postal questionnaire survey. Br J Anaesth 2004;92:689–93 decreasing the probability of complications. As an example, 10. Stamer UM, Mpasios N, Stu¨ ber F, Maier C. A survey of acute total knee arthroplasty patients discharged home with a pain services in Germany and a discussion of international continuous femoral nerve block had reduced hospital survey data. Reg Anesth Pain Med 2002;27:125–31 length of stay and associated costs and charges, not includ- 11. Powell AE, Davies HT, Bannister J, Macrae WA. Challenge of ing postdischarge resource utilization.21 Because stake- improving postoperative pain management: case studies of three acute pain services in the UK National Health Service. holders are likely to be interested in the net costs of running Br J Anaesth 2009;102:824–31 an Acute Pain Service, future studies should investigate 12. Coluzzi F, Savoia G, Paoletti F, Costantini A, Mattia C. how pain management affects total hospitalization and Postoperative pain survey in Italy (POPSI): a snapshot of downstream costs. current national practices. Minerva Anestesiol 2009;75:622–31 Providing an Acute Pain Service is not solely an anes- 13. Hu P, Owens T, Harmon D. A survey of acute pain services in thesia issue. The hospital facility needs to invest in training, teaching hospitals in the Republic of Ireland. Ir J Med Sci 2007;176:225–8 supplies, and policies. In fact, a dedicated Acute Pain 14. Lee A, Chan S, Chen PP, Gin T, Lau AS. Economic evaluations Service provides the opportunity for anesthesiologists to of acute pain service programs: a systematic review. Clin J Pain extend their influence in perioperative medicine for the 2007;23:726–33 millions of inpatients having surgery. The tasks involved 15. Myles PS, Hunt JO, Nightingale CE, Fletcher H, Beh T, Tanil D, align well with the anesthesiologist’s existing skill set. Nagy A, Rubinstein A, Ponsford JL. Development and psychometric testing of a quality of recovery score after general However, the wide variety of practice models in the USA anesthesia and surgery in adults. Anesth Analg 1999;88:83–90 and abroad suggests there has not been agreement on how 16. Tan JM, Macario A. How to evaluate whether a new technol- to provide these services in a financially viable way. Given ogy in the operating room is cost-effective from society’s the large number of surgical inpatients, optimizing pain viewpoint. Anesthesiol Clin 2008;26:745–64 management may yield large savings. 17. Philips BD, Liu SS, Wukovits B, Boettner F, Waldman S, The truth is that whether a dedicated Acute Pain Service Liguori G, Goldberg S, Goldstein L, Melia J, Hare M, Jasphey L, Tondel S. Creation of a novel recuperative pain medicine can be a cost-effective way of managing postoperative pain service to optimize postoperative analgesia and enhance pa- depends on multiple conditions, such as work culture, tient satisfaction. HSS J 2010;6:61–5 surgical case mix, and reimbursement systems. 18. Basu A, Philipson TJ. The impact of comparative effectiveness research on health and health care spending. NBER Working AUTHOR CONTRIBUTIONS Paper No. 15633, 2010 All authors helped design and conduct the study, analyze the 19. Raja SN, Jensen TS. Predicting postoperative pain based on data, and write the manuscript. All authors approved the final preoperative pain perception: are we doing better than the manuscript. weatherman? Anesthesiology 2010;112:1311–2 20. Macario A. What does one minute of operating room time RECUSE NOTE cost? J Clin Anesth 2010;22:233–6 21. Ilfeld BM, Mariano ER, Williams BA, Woodard JN, Macario A. Franklin Dexter is the Section Editor of Economics, Education, Hospitalization costs of total knee arthroplasty with a continu- and Policy for the Journal. The manuscript was handled by ous femoral nerve block provided only in the hospital versus Spencer Liu, Section Editor for pain Medicine. Dr. Dexter was on an ambulatory basis: a retrospective, case-control, cost- not involved in any way with the editorial process or decision. minimization analysis. Reg Anesth Pain Med 2007;32:46–54

844 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Society of Cardiovascular Anesthesiologists

Cardiovascular Anesthesiology Section Editor: Charles W. Hogue, Jr. Perioperative Echocardiography and Cardiovascular Education Section Editor: Martin J. London Hemostasis and Transfusion Medicine Section Editor: Jerrold H. Levy

Heparin Concentration–Based Anticoagulation for Cardiac Surgery Fails to Reliably Predict Heparin Bolus Dose Requirements

Sean Garvin, MD,* Daniel C. FitzGerald, CCP,† George Despotis, MD,‡§ʈ Prem Shekar, MD,† and Simon C. Body, MBChB, MPH*

BACKGROUND: Hemostasis management has evolved to include sophisticated point-of-care systems that provide individualized dosing through heparin concentration–based anticoagulation. The Hepcon HMS Plus system (Medtronic, Minneapolis, MN) estimates heparin dose, activated clotting time (ACT), and heparin dose response (HDR). However, the accuracy of this test has not been systematically evaluated in large cohorts. METHODS: We examined institutional databases for all patients who underwent cardiac surgery with cardiopulmonary bypass (CPB) at our institution from February 2005 to July 2008. During this period, the Hepcon HMS Plus was used exclusively for assessment of heparin dosing and coagulation monitoring. Detailed demographic, surgical, laboratory, and heparin dosing data were recorded. ACT, calculated and measured HDR, and heparin concentrations were recorded. Performance of the Hepcon HMS Plus was assessed by comparison of actual and target ACT values and calculated and measured HDR. RESULTS: In 3880 patients undergoing cardiac surgery, heparin bolus dosing to a target ACT resulted in wide variation in the postheparin ACT (r2 ϭ 0.03). The postheparin ACT did not reach the target ACT threshold in 7.4% (i.e., when target ACT was 300 s) and 16.9% (i.e., when target ACT was 350 s) of patients. Similarly, the target heparin level calculated from the HDR did not correlate with the postbolus heparin level, with 18.5% of samples differing by more than 2 levels of the assay. Calculated and measured HDR were not linearly related at any heparin level. CONCLUSIONS: The Hepcon HMS Plus system poorly estimates heparin bolus requirements in the pre-CPB period. Further prospective studies are needed to elucidate what constitutes adequate anticoagulation for CPB and how clinicians can reliably and practically assess anticoagulation in the operating room. (Anesth Analg 2010;111:849–55)

dministration of heparin, and monitoring of whole although there is still a lack of consensus in this area.9,10 blood activated clotting time (ACT), is still the Previous studies of heparin-level monitoring devices, includ- Amainstay of anticoagulation for cardiopulmonary ing the Hepcon HMS Plus, had varied results, with some bypass (CPB).1,2 Hemostasis management has evolved to studies showing good agreement between whole blood hep- 11 include more sophisticated point-of-care systems that pro- arin concentration and either ACT before CPB or plasma 12 vide individualized dosing of heparin and protamine, anti-Xa levels before or during CPB, whereas others showed through heparin concentration–based estimation of hepa- a lack of significant relationship between ACT and heparin 13 rin dosage, and measurement of ACT.3,4 Individualized concentration or whole blood versus anti-Xa heparin con- 14 heparin concentration–based anticoagulation management centration. All of these studies were limited by their small has been reported to cause less activation of thrombin, sample sizes. Importantly, there is previously observed inter- decreased fibrinolysis, reduced blood loss, fewer transfusions, individual variation in measured heparin levels for a target and a decreased risk of reexploration for hemorrhage,5–8 ACT required for CPB. Therefore, we examined the relationship between individualized in vitro estimation of the heparin level for a From the *Department of Anesthesiology, Perioperative and Pain Medicine, specific target ACT, and the subsequently observed ACT and †Division of Cardiac Surgery, Brigham and Women’s Hospital, Harvard and heparin level measured after bolus administration of Medical School, Boston, Massachusetts; and ‡Departments of Pathology, §Immunology, and ʈAnesthesiology, Washington University School of the heparin dose response (HDR)-estimated heparin Medicine, St. Louis, Missouri. dose. We hypothesized that: 1) there would be strong Accepted for publication June 22, 2009. correlation between calculated and measured HDR for Supported by departmental funds. individual patients; 2) residual variation in the predicted Address correspondence and reprint requests to Simon C. Body, MBChB, HDR would be due to patient factors that are accounted for MPH, Brigham and Women’s Hospital, Harvard Medical School, 75 Francis St., Boston, MA 02215. Address e-mail to [email protected]. in the calculation of the HDR; and 3) the HDR-estimated Copyright © 2010 International Anesthesia Research Society heparin dose would achieve the desired ACT for CPB in all DOI: 10.1213/ANE.0b013e3181b79d09 patients.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 849 Anticoagulation for Cardiac Surgery

METHODS loss but were not supported by prior clinical evidence of The Hepcon HMS Plus (Medtronic, Minneapolis, MN) was efficacy. introduced in our institution in February 2004, in concert We also observed variability in anticoagulation practice with a change in heparin coagulation management. Use of that was not proscribed by a protocol. Some perfusionists this system was based on a belief that adequate anticoagula- administered additional heparin when the measured hep- tion for CPB using a heparin-coated circuit could be achieved arin level was Ͻ1.4 or 2.0 U/mL, according to personal with lower ACTs as long as reasonable, albeit lower heparin preference. These patients were excluded from analysis, as levels than had been historically used were maintained. previously described. Furthermore, there was a perception that reduced heparin administration may result in reduced postoperative bleeding. Data Collection and Statistical Analysis After IRB approval, we examined institutional databases Detailed demographic data, surgical indications, preopera- for all patients who underwent CPB for cardiac surgery at tive laboratory values, operative data, heparin bolus dose, our institution, from February 2005 to July 2008 (n ϭ 4667). ACT values, and measured heparin concentrations were Patients with incomplete perioperative anticoagulation routinely collected into a centralized database. The calcu- data were excluded from analysis (n ϭ 562). The majority of lated HDR was calculated as the target ACT Ϫ baseline patients were excluded for having a target ACT of Ͼ350 s ACT divided by the target heparin level and reported as ⅐ Ϫ1 ⅐ Ϫ1 or subsequent administration of aprotinin (n ϭ 157), no s U mL . The calculated HDR differed from the 15 record of postheparinization heparin level (n ϭ 144), no method used by the Hepcon HMS Plus, but prospective record of baseline ACT (n ϭ 91), or no record of height or comparison of the Hepcon HMS Plus and the above 2 ϭ weight (n ϭ 67). Furthermore, we excluded an additional calculation in 15 patients showed excellent correlation (r 225 patients who were not anticoagulated using the insti- 0.991). This calculation uses the target heparin concentra- tutional protocol described below, yielding 3880 analyzed tion identified from the HDR slope generated by the patients. Hepcon HMS Plus, not the protocol-driven heparin level. Thus, the calculation of the HDR slope is independent of the clinical protocol and unaffected by the heparin level Anticoagulation Management desired or obtained after heparin dosing. The Hepcon HMS Plus was used for anticoagulation man- The measured HDR was calculated as postheparin agement according to the manufacturer’s recommenda- ACT Ϫ baseline ACT divided by the measured heparin Ϫ Ϫ tions.15 The estimated blood volume for each patient was level and reported as s ⅐ U 1 ⅐ mL 1. By using the measured calculated using the manufacturer’s instructions,15 accord- postheparinization ACT and measured heparin level, the ing to the method described by Allen et al.14 After induc- measured HDR is independent of the clinical protocol tion of anesthesia, baseline kaolin ACT, predicted HDR, because the HDR slope has been demonstrated to be linear predicted heparin concentration, and heparin bolus calcu- over the clinical range of heparin concentrations.11 lations were performed according to the manufacturer’s The Hepcon HMS Plus heparin assay has limited fidel- instructions, using HDR cartridges encompassing whole ity, reporting only 6 categorical heparin levels, with inter- blood heparin concentration ranging from 0.4 to 3.4 U/mL. mediate values of heparin reported to the closest level. The Hepcon HMS Plus recommended CPB prime heparin Accordingly, we could not directly compare the calculated dose based on a 750- or 1000-mL prime volume was added and measured HDR. Therefore, we report the calculated to the calculated heparin bolus and administered via a HDR for each measured postbolus heparin level. Pearson central venous catheter. Heparin was not added to the CPB product-moment correlation was performed between cal- pump prime. Three minutes after United States Pharma- culated and measured values of HDR, and heparin concen- copeia porcine heparin administration and 10 min after tration and ACT data. ACT data at each heparin level were onset of CPB, heparin concentration and ACT were remea- compared with Wilcoxon’s ranked sum test. Statistical sured. All patients received an ⑀-aminocaproic acid load of analysis was performed with JMP version 7.02 (SAS, Cary, 7.5–10 g over 1 h, after the initial blood draw for baseline ACT, NC). All data are presented as mean Ϯ sd or median and but before heparin administration and blood sampling for interquartile range, as appropriate. A 2-sided P Ͻ 0.05 measurement of postheparin ACT. ⑀-Aminocaproic acid was was considered as showing statistical significance. subsequently infused at a rate of 1.25–1.5 g/h. Throughout the entire study period, an ACT of more RESULTS than 350 s and a minimum heparin concentration of 2 A total of 3880 patients were analyzed. Baseline demo- U/mL were used in patients undergoing non–coronary graphics and perioperative data are summarized in Table 1. artery bypass graft (CABG) procedures or CABG with the Target ACTs of 300 and 350 s were used in 23.4% and 76.6% use of cardiotomy suction. Patients undergoing primary of patients, respectively. Heparin dose, ACT, and heparin CABG surgery without the use of cardiotomy suction level after the heparin bolus for each target ACT are before March 2007 were anticoagulated using a protocol detailed in Table 2. The mean heparin dose calculated to that proscribed a minimum ACT of 300 s before the institu- achieve a target ACT of 300 s was 152 U/kg, and for a tion of CPB. After February 2007, to standardize anticoagula- target ACT of 350 s was 179 U/kg. After administration of tion management, the minimum ACT before CPB was in- the HDR-estimated heparin dose, the target ACT was not creased to 350 s with a minimum heparin concentration of 2.0 achieved in 7.4% of patients with a target ACT of 300 s, and U/mL for CABG surgery. These protocols were driven by a it was not reached in 16.9% of patients with a target ACT of desire to reduce heparin and protamine doses to reduce blood 350 s (Fig. 1). Additional heparin was administered either

850 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Demographics and Perioperative Data Study cohort (3880 ؍ N) Age 65.2 Ϯ 13.4 Male gender 2538 (65.4%) Height (cm) 171 Ϯ 10 Weight (kg) 82.2 Ϯ 19.0 BSA (m2) 1.94 Ϯ 0.24 Indication for surgery Coronary artery disease 2394 (63.3%) Valve disease 1932 (53.1%) Aortic dissection 41 (1.2%) Medical history Previous myocardial infarction 989 (26.2%) Hypertension 2550 (67.5%) Insulin-dependent diabetes 343 (9.1%) Renal insufﬁciency (creatinine 145 (4.0%) Ն2.0 mg/dL) Figure 1. Observed activated clotting time (ACT) at each target ACT Preoperative medications level. Box plot represents the 5th, 25th, median, 75th, and 95th Aspirin 2387 (63.1%) percentiles of data. Adenosine diphosphate inhibitors 182 (4.8%) Intravenous heparin 249 (6.6%) Warfarin 34 (0.9%) Preoperative laboratory values before or immediately after initiation of CPB to 91% of Hematocrit (%) 38.8 Ϯ 4.9 patients who failed to achieve their target ACT of 300 s and Platelet count (ϫ109/L) 247 Ϯ 81 to 77% of patients who failed to achieve their target ACT of INR 1.14 Ϯ 0.30 350 s. INR Ͼ1.4 280 (7.8%) Partial thromboplastin time (s) 36.2 Ϯ 15.9 There was wide variation among patients for the ACT Partial thromboplastin time Ͼ37 s 733 (20.9%) seen at each heparin level (Fig. 2). Measured HDR varied substantially among patients (mean Ϯ sd, 104.7 Ϯ 24.3 Data are presented as mean Ϯ 1SD for demographics. ⅐ Ϫ1 ⅐ Ϫ1 The upper limit of the laboratory reference range for partial thromboplastin s U mL ; median [10%–90% interquartile range], 102 time is 37 s. [77–137]). Correlation (r2) between the calculated and mea- BSA ϭ body surface area; CABG ϭ coronary artery bypass graft; INR ϭ sured HDR was poor at all heparin levels (r2 Ͻ 0.02; Fig. 3). International normalized ratio.

Table 2. Intraoperative Anticoagulation Management

(Mean ؎ SD Median (10th–90th percentile Baseline ACT (s) 138 Ϯ 17 137 (118–159) Target ACT of 300 s (N ϭ 908) Initial heparin dose (U) 12,994 Ϯ 3909 12,000 (9000–18,000) Heparin dose (U/kg) 152 Ϯ 38 149 (108–203) Postbolus ACT (s) 387 Ϯ 73 377 (309–466) Postbolus ACT Ͻ300 s 67 (7.4%) CPB ϩ 10 min ACT (s) 340 Ϯ 54 335 (274–409) Postheparin bolus CPB ؉ 10 min Heparin level (U/mL) Ͻ1.4 16 (1.8%) 250 (27.9%) 1.4 232 (25.6%) 498 (55.5%) 2.0 460 (50.7%) 118 (13.2%) 2.7 137 (15.1%) 21 (2.3%) 3.4 60 (6.6%) 10 (1.1%) Ͼ3.4 3 (0.3%) 0 (0%)

(Mean ؎ SD Median (10th–90th percentile Target ACT of 350 s (N ϭ 2972) Initial heparin dose (U) 14,313 Ϯ 4013 14,000 (10,000–20,000) Heparin dose (U/kg) 179 Ϯ 42 174 (132–232) Postbolus ACT (s) 418 Ϯ 83 406 (332–509) Postbolus ACT Ͻ350 s 501 (16.9%) CPB ϩ 10 min ACT (s) 386 Ϯ 73 377 (313–466) Postheparin bolus CPB ؉ 10 min Heparin level (U/mL) Ͻ1.4 20 (0.7%) 278 (9.4%) 1.4 438 (14.7%) 1716 (58.2%) 2.0 1449 (48.8%) 789 (26.8%) 2.7 731 (24.6%) 125 (4.2%) 3.4 315 (10.6%) 41 (1.4%) Ͼ3.4 19 (0.6%) 0 (0%) ACT ϭ activated clotting time; CPB ϭ cardiopulmonary bypass.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 851 Anticoagulation for Cardiac Surgery

Figure 2. Observed activated clotting time at each measured postbolus heparin level. Box plot represents the 5th, 25th, median, 75th, and 95th percentiles of data.

To exclude an outlier effect, the analysis was repeated in a subgroup of patients within 1 sd of the mean postheparin ACT, with similar results. The predicted heparin level expected after heparin bolus administration was compared with the heparin level measured after heparin bolus. Levels that differed by more than 0.3 U/mL differed by more than the sensitivity (1 channel) of the assay and were observed in 51.0% of samples (Fig. 4). Levels that differed by more than 0.7 U/mL differed by more than 2 channels in the assay and were observed in 18.5% of samples.

DISCUSSION We examined the performance of the Hepcon HMS Plus to guide anticoagulation in 3880 patients undergoing cardiac surgical procedures. Data provided by the Hepcon HMS Plus were used to calculate heparin dosing throughout CPB and to direct the administration of protamine after CPB. We observed wide variability in heparin dose requirements to achieve an adequate ACT for CPB, as others have previously described.1 We also observed poor correlation of the calculated HDR with the measured HDR. This led to ACT values that were less than the target ACT values when heparin dosing was guided by estimation of HDR, in 7.4%–16.9% of patients. The imprecision for the calculated HDR may explain the high frequency of failure to reach the Figure 3. Correlation between calculated and measured heparin target ACT. dose response. Each panel contains data for individuals with a Interpatient variability in HDR is well described, and speciﬁc measured heparin level after heparinization. Panel A repre- use of a dose-response plot recommended by Bull et al.3 sents a measured heparin level of 1.4 U/mL (r2 Ͻ 0.01). Panel B 2 provides the basis of Hepcon HMS Plus calculations. Our represents a measured heparin level of 2.0 U/mL (r ϭ 0.02). Panel C represents a measured heparin level of 2.7 U/mL (r2 ϭ 0.01). initial hypothesis that the HDR slope calculated by the Panel D represents a measured heparin level of 3.4 U/mL (r2 ϭ Hepcon HMS Plus would correlate with the measured 0.03). HDR slope is not supported by the data. The interindividual variability in heparin response could be attributable to a

852 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA heparin responsiveness may be underestimated by the initial HDR calculation performed by the Hepcon HMS Plus. This source of error in the calculated HDR would seemingly lead to increased heparin dosages and higher postheparin ACT, which is incongruous with our finding of 7.4%–16.9% of patients failing to achieve their target ACT, but congruous with the mean ACT being higher than the target ACT in our overall population. Although the exact contribution of TFPI to point-of-care measurements of heparin responsiveness has not been established, strong correlation between the TFPI-responsive Heptest (Ameri- can Diagnostica, Stamford, CT) measurements of heparin level and ACT in the pre-CPB period22 has been observed, and release of TFPI correlates with the Heptest measurement of heparin response in normal volunteers receiving IV heparin.23 These observations reinforce the importance of TFPI in heparin responsiveness and point to a likely impact on point-of-care assessments of HDR. Earlier studies of these heparin-level monitoring de- Figure 4. Target and measured heparin level observed after bolus vices, including the Hepcon HMS Plus, had varied results heparin administration. Box plot represents the 5th, 25th, median, with some studies showing good agreement between 75th, and 95th percentiles of data. whole blood heparin concentration and either ACT before CPB11 or plasma anti-Xa levels,12 whereas others showed a lack of significant relationship between ACT and heparin 24 number of in vivo factors, including error in the estimation of concentration or whole blood versus anti-Xa heparin 13,14 12 blood volume, effects of release of tissue factor pathway concentration during CPB. One study revealed sub- inhibitor (TFPI) by heparin in vivo but not ex vivo, extravascu- stantial interpatient variability in HDR; however, the linear lar sequestration of heparin, plasma protein binding, circulat- relationship between heparin concentration and kaolin ing antithrombin, and platelet activation.16 ACT within individual patients was generally exceptional. Estimation of blood volume is a potential source of error Our retrospective analysis demonstrated significant vari- in heparin dose calculation. The methodology used by ability in the responsiveness of the kaolin ACT and heparin several previous studies5,6,12 is predominantly based on the in vivo versus kaolin ACT and heparin ex vivo (i.e., after method described by Allen et al.14 in 1956. The magnitude IV administration of heparin), and calculated versus of error in estimation of blood volume in individual measured HDR, respectively (Fig. 3). This was paralleled patients undergoing cardiac surgery has not been system- by the finding that the Hepcon HMS Plus–guided hep- atically evaluated. It is probable that the severity and arin administration failed to result in adequate anti- nature of cardiac disease may be a substantial source of coagulation for CPB in 16% of patients in this study variation in blood volume estimates, thereby contributing when the target ACT was 350 s, even with the CPB prime to imprecision of point-of-care anticoagulation management heparin dose being administered as part of the initial IV instruments. Further investigation aimed at improving estima- bolus dose. tion of blood volume in patients undergoing cardiac surgery Early work performed by Bull et al.25 and Young may result in better performance of point-of-care heparin et al.26 continues to form the basis for the ACT target used concentration–based anticoagulation systems. to safely institute CPB, with many centers using an ACT of Potentially contributing to the imprecision of the Hep- more than 480 s as a target. In 1981, Jobes et al.27 deter- con HMS Plus are the effects of release of TFPI by heparin mined that a heparin level of 2 U/mL was associated with in vivo but not ex vivo. TFPI in vivo is bound to endothelial an ACT Ͼ300sinϾ95% of patients. These data were not cell surfaces with small amounts circulating in plasma and the basis for our use of a heparin level of 2 U/mL in the bound to platelets.17 TFPI is an endogenous serine protease latter portion of the study period, but are in line with our inhibitor that exerts its action primarily through neutraliz- observation of wide variation in ACT values at specific ing factor Xa by complexing with factor Xa and forming a heparin levels. The theoretical benefit of using heparin TFPI-FXa complex, while also providing feedback inhibi- concentration– based anticoagulation is that a stable hepa- tion of the factor VIIa–tissue factor complex.18 Plasma TFPI rin level may be achieved, thereby minimizing hemostatic is released from endothelial surface stores by administra- activation7,28 and keeping the patient fully anticoagulated tion of unfractionated and low-molecular-weight hepa- throughout the CPB period, which would lead to reduced rins,19 increasing plasma levels during CPB.20 Brodin et blood loss and transfusion.5,6 We observed wide variability al.21 recently demonstrated significant synergy between in ACT for any given heparin level (Fig. 2). Almost all antithrombin and TFPI in postheparin plasma with an studies using heparin concentration have used this histori- almost equal contribution of each to the prolongation of cal reference as the target for defining adequate anticoagu- anticoagulation in a laboratory setting. In the absence of an lation, but establishment of a “safe heparin level” has not endothelial source of TFPI, the contribution of TFPI to been systematically studied.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 853 Anticoagulation for Cardiac Surgery

This study is limited by its retrospective nature. We 7. Despotis GJ, Joist JH, Hogue CW Jr, Alsoufiev A, Joiner-Maier were not able to capture all of the perioperative factors that D, Santoro SA, Spitznagel E, Weitz JI, Goodnough LT. More effective suppression of hemostatic system activation in pa- may influence anticoagulation and an individual’s re- tients undergoing cardiac surgery by heparin dosing based on sponse to heparin administration. In addition, important heparin blood concentrations rather than ACT. Thromb Hae- clinical outcomes are not part of this data set, including most 1996;76:902–8 perioperative complications that may relate to anticoagula- 8. Bowie J, Kemma G. Automated management of heparin anticoagulation in cardiovascular surgery. Proc Am Acad Cardio- tion management, such as bleeding, transfusion, reexplora- vasc Perf 1985;6:1–10 tion, myocardial infarction, new or worsening renal insuf- 9. Shore-Lesserson L, Reich DL, DePerio M. Heparin and prota- ficiency, and cerebral vascular accident. Furthermore, our mine titration do not improve haemostasis in cardiac surgical comparison of the calculated and measured HDR is limited patients. Can J Anaesth 1998;45:10–8 10. Slight RD, Buell R, Nzewi OC, McClelland DB, Mankad PS. A by the imprecision of the automated protamine titration comparison of activated coagulation time-based techniques for method, which measures whole blood heparin concentration anticoagulation during cardiac surgery with cardiopulmonary using discrete cutoffs using only the 6 channels of the car- bypass. J Cardiothorac Vasc Anesth 2008;22:47–52 tridge. Thus, we are left with 4 likely sources of disparity 11. Despotis GJ, Alsoufiev AL, Spitznagel E, Goodnough LT, Lappas DG. Response of kaolin ACT to heparin: evaluation between calculated and measured HDR: 1) inaccurate esti- with an automated assay and higher heparin doses. Ann mate of the patient’s blood volume; 2) lack of fidelity of Thorac Surg 1996;61:795–9 measured heparin concentration because only 6 categories 12. Despotis GJ, Summerfield AL, Joist JH, Goodnough LT, San- of heparin level describe the linear range of heparin con- toro SA, Spitznagel E, Cox JL, Lappas DG. Comparison of activated coagulation time and whole blood heparin measure- centration; 3) inherent inaccuracy of the device; and 4) ments with laboratory plasma anti-Xa heparin concentration in differences between the anticoagulant activity of heparin ex patients having cardiac operations. J Thorac Cardiovasc Surg vivo compared with its several actions in vivo, notably upon 1994;108:1076–82 13. Hardy JF, Belisle S, Robitaille D, Perrault J, Roy M, Gagnon L. the release of TFPI. We were unable to subset these possible Measurement of heparin concentration in whole blood with causes further, but they seem to be significant. the Hepcon/HMS device does not agree with laboratory We conclude that the Hepcon HMS Plus fails to consis- determination of plasma heparin concentration using a chro- tently provide the therapeutic heparin bolus dose uniformly mogenic substrate for activated factor X. J Thorac Cardiovasc Surg 1996;112:154–61 in all patients based on the wide discrepancy in calculated 14. Allen TH, Peng MT, Chen KP, Huang TF, Chang C, Fang HS. versus measured HDR. This can lead to inadequate heparin Prediction of blood volume and adiposity in man from body doses to achieve a target ACT for CPB in as much as 16.9% of weight and cube of height. Metabolism 1956;5:328–45 patients. However, the Hepcon HMS Plus was able to identify 15. Medtronic Perfusion Systems. Hepcon HMS Plus operator’s manual. Minneapolis, MN, 2001 an adequate heparin dose for the majority of the patients. 16. Ranucci M, Isgro G, Cazzaniga A, Ditta A, Boncilli A, Cotza M, Because this study, did not compare the Hepcon HMS Plus Carboni G, Brozzi S. Different patterns of heparin resistance: with empiric dosing regimens, we are uncertain whether therapeutic implications. Perfusion 2002;17:199–204 empiric regimens can either under- or overperform when 17. Bajaj MS, Kuppuswamy MN, Saito H, Spitzer SG, Bajaj SP. Cultured normal human hepatocytes do not synthesize compared to this system. Further prospective studies are lipoprotein-associated coagulation inhibitor: evidence that en- needed to elucidate what constitutes adequate anticoagula- dothelium is the principal site of its synthesis. Proc Natl Acad tion for CPB and how clinicians can reliably and practically Sci USA 1990;87:8869–73 assess anticoagulation in the operating room. 18. Broze GJ Jr, Miletich JP. Characterization of the inhibition of tissue factor in serum. Blood 1987;69:150–5 19. Hoppensteadt DA, Walenga JM, Fasanella A, Jeske W, Fareed REFERENCES J. TFPI antigen levels in normal human volunteers after 1. Akl BF, Vargas GM, Neal J, Robillard J, Kelly P. Clinical intravenous and subcutaneous administration of unfraction- experience with the activated clotting time for the control of ated heparin and a low molecular weight heparin. Thromb Res heparin and protamine therapy during cardiopulmonary by- 1995;77:175–85 pass. J Thorac Cardiovasc Surg 1980;79:97–102 20. Adams MJ, Cardigan RA, Marchant WA, Grocott MP, Mythen 2. Hattersley PG. Activated coagulation time of whole blood. MG, Mutch M, Purdy G, Mackie IJ, Machin SJ. Tissue factor JAMA 1966;196:436–40 pathway inhibitor antigen and activity in 96 patients receiving 3. Bull BS, Huse WM, Brauer FS, Korpman RA. Heparin therapy heparin for cardiopulmonary bypass. J Cardiothorac Vasc during extracorporeal circulation. II. The use of a dose- Anesth 2002;16:59–63 response curve to individualize heparin and protamine dos- 21. Brodin E, Appelbom H, Osterud B, Hilden I, Petersen LC, Hansen JB. Regulation of thrombin generation by TFPI in age. J Thorac Cardiovasc Surg 1975;69:685–9 plasma without and with heparin. Transl Res 2009;153:124–31 4. Raymond PD, Ray MJ, Callen SN, Marsh NA. Heparin moni- 22. Hellstern P, Bach J, Simon M, Saggau W. Heparin monitoring toring during cardiac surgery. Part 1: Validation of whole- during cardiopulmonary bypass surgery using the one-step blood heparin concentration and activated clotting time. Per- point-of-care whole blood anti-factor-Xa clotting assay heptest- fusion 2003;18:269–76 POC-Hi. J Extra Corpor Technol 2007;39:81–6 5. Despotis GJ, Joist JH, Hogue CW Jr, Alsoufiev A, Kater K, 23. Hoffmann U, Harenberg J, Bauer K, Huhle G, Tolle AR, Goodnough LT, Santoro SA, Spitznagel E, Rosenblum M, Feuring M, Christ M. Bioequivalence of subcutaneous and Lappas DG. The impact of heparin concentration and activated intravenous body-weight-independent high-dose low- clotting time monitoring on blood conservation. A prospective, molecular-weight heparin Certoparin on anti-Xa, Heptest, and randomized evaluation in patients undergoing cardiac opera- tissue factor pathway inhibitor activity in volunteers. Blood tion. J Thorac Cardiovasc Surg 1995;110:46–54 Coagul Fibrinolysis 2002;13:289–96 6. Jobes DR, Aitken GL, Shaffer GW. Increased accuracy and 24. Culliford AT, Gitel SN, Starr N, Thomas ST, Baumann FG, precision of heparin and protamine dosing reduces blood loss Wessler S, Spencer FC. Lack of correlation between activated and transfusion in patients undergoing primary cardiac opera- clotting time and plasma heparin during cardiopulmonary tions. J Thorac Cardiovasc Surg 1995;110:36–45 bypass. Ann Surg 1981;193:105–11

854 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 25. Bull BS, Korpman RA, Huse WM, Briggs BD. Heparin therapy 27. Jobes DR, Schwartz AJ, Ellison N, Andrews R, Ruffini RA, during extracorporeal circulation. I. Problems inherent in Ruffini JJ. Monitoring heparin anticoagulation and its neutral- existing heparin protocols. J Thorac Cardiovasc Surg ization. Ann Thorac Surg 1981;31:161–6 1975;69:674–84 28. Koster A, Fischer T, Praus M, Haberzettl H, Kuebler WM, 26. Young JA, Kisker CT, Doty DB. Adequate anticoagulation Hetzer R, Kuppe H. Hemostatic activation and inflammatory during cardiopulmonary bypass determined by activated clot- response during cardiopulmonary bypass: impact of heparin ting time and the appearance of fibrin monomer. Ann Thorac management. Anesthesiology 2002;97:837–41 Surg 1978;26:231–40

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 855 Heparin Dose Response Is Independent of Preoperative Antithrombin Activity in Patients Undergoing Coronary Artery Bypass Graft Surgery Using Low Heparin Concentrations

Sean Garvin, MD,* Daniel Fitzgerald, CCP,† Jochen D. Muehlschlegel, MD,* Tjo¨rvi E. Perry, MD,* Amanda A. Fox, MD,* Stanton K. Shernan, MD,* Charles D. Collard, MD,‡ Sary Aranki, MD,† and Simon C. Body, MBChB, MPH*

BACKGROUND: Unfractionated heparin’s primary mechanism of action is to enhance the enzymatic activity of antithrombin (AT). We hypothesized that there would be a direct association between preoperative AT activity and both heparin dose response (HDR) and heparin sensitivity index (HSI) in patients undergoing coronary artery bypass graft surgery. METHODS: Demographic and perioperative data were collected from 304 patients undergoing primary coronary artery bypass graft surgery. AT activity was measured after induction of general anesthesia using a colorimetric method (Siemens Healthcare Diagnostics, Tarrytown, NY). Activated coagulation time (ACT), HDR, and HSI were measured using the Hepcon HMS Plus system (Medtronic, Minneapolis, MN). Heparin dose was calculated for a target ACT using measured HDR by the same system. Multivariate linear regression was performed to identify independent predictors of HDR. Subgroup analysis of patients with low AT activity (Ͻ80% normal; Ͻ0.813 U/mL) who may be at risk for heparin resistance was also performed. RESULTS: Mean baseline ACT was 135 Ϯ 18 seconds. Mean calculated HDR was 98 Ϯ 21 s/U/mL. Mean baseline AT activity was 0.93 Ϯ 0.13 U/mL. Baseline AT activity was not significantly associated with baseline or postheparin ACT, HDR, or HSI. Addition of AT activity to multivariable linear regression models of both HDR and HSI did not significantly improve model performance. Subgroup analysis of 49 patients with baseline AT Ͻ80% of normal levels did not reveal a relationship between low AT activity and HDR or HSI. Preoperative AT activity, HDR, and HSI were not associated with cardiac troponin I levels on the first postoperative day, intensive care unit duration, or hospital length of stay. CONCLUSION: Although enhancing AT activity is the primary mechanism by which heparin facilitates cardiopulmonary bypass anticoagulation, low preoperative AT activity is not associated with impaired response to heparin or to clinical outcomes when using target ACTs of 300 to 350 seconds. (Anesth Analg 2010;111:856–61)

ore than 50 years after heparin’s discovery,1 platelet count Ͼ300,000 cells/mm3, recent heparin expo- Rosenberg and Damus2,3 identified its primary sure, and AT deficiency.9 Preoperative AT activity of Ͻ80% Mmechanism of action as enhancement of the en- of normal has been associated with reduced heparin re- zymatic activity of antithrombin (AT). AT binds and inac- sponse in adults undergoing cardiac surgery.11 Postopera- tivates serine proteases’ contact activation and common tive AT deficiency has also been associated with worsened pathways, principally factors IIa, Xa, IXa, and VIIa, by clinical outcomes, including increased intensive care unit decreasing their binding efficiency for substrate. AT’s inhi- (ICU) length of stay, risk of reexploration for bleeding, and bition of serine protease activity is increased several orders thromboembolic events.12 of magnitude by heparin as a result of a conformational Because the primary mechanism of action of heparin is change induced by binding of a specific pentasaccharide to facilitate the enzymatic activity of AT, we surmised that unit of heparin with high affinity for AT.4 reduced AT activity would be associated with heparin Heparin resistance has been reported in 4% to 22% of response. Little evidence is available to support this hy- patients undergoing cardiopulmonary bypass (CPB).5–10 pothesis, with most data regarding AT activity and heparin Potential risk factors include age older than 65 years, dose response (HDR) reported from studies examining restoration of heparin responsiveness in cardiac surgical patients From the *Department of Anesthesiology, Perioperative and Pain Medicine, by administration of recombinant AT concentrate, usually and †Division of Cardiac Surgery, Brigham and Women’s Hospital, Harvard without measurement of the patient’s AT level.5–8,13,14 We Medical School, Boston, Massachusetts; and ‡Division of Cardiovascular Anesthesia at the Texas Heart Institute, Baylor College of Medicine, Saint aimed to assess the relationship between preoperative AT Luke’s Episcopal Hospital, Houston, Texas. activity and heparin sensitivity in a cohort of primary Accepted for publication November 18, 2009. coronary artery bypass graft (CABG) patients and addi- Address correspondence and reprint requests to Simon C. Body, MBChB, tionally assess this association in a subgroup of subjects MPH, Brigham and Women’s Hospital, Harvard Medical School, 75 Francis St., Boston, MA 02215. Address e-mail to [email protected]. with low preoperative AT activity. We further examined Copyright © 2010 International Anesthesia Research Society whether AT activity or measures of heparin sensitivity DOI: 10.1213/ANE.0b013e3181ce1ffa were associated with severity of myocardial injury or

856 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 duration of ICU and hospital stays, as surrogates of sever- For anticoagulation management, the Hepcon HMS Plus ity of illness. system (Medtronic, Minneapolis, MN) was used according to the manufacturer’s recommendations.15 The estimated METHODS blood volume for each patient was calculated using the 15 With IRB approval and individual patient consent, 346 manufacturer’s instructions, according to the method 16 patients undergoing primary CABG using CPB from Feb- described by Allen et al. After induction of anesthesia, ruary 2005 to December 2006 were enrolled into a parent baseline kaolin ACT, predicted HDR, predicted heparin study called the CABG Genomics Program with the aim of concentration, and heparin bolus calculations were per- determining genetic risks for adverse perioperative out- formed according to the manufacturer’s instructions, using comes. Patients were excluded from the parent study if heparin-protamine titration cartridges encompassing whole they were younger than 20 years of age; underwent repeat blood heparin concentration ranging from 0.7 to 3.4 U/mL or off-pump CABG, planned concomitant valve, or other and kaolin as the activator. The recommended Hepcon cardiac surgery; had a preoperative hematocrit Ͻ25%; or if HMS Plus CPB prime heparin dose based on a 750- or they had received leukocyte-rich blood products within 30 1000-mL prime volume was added to the calculated hepa- days before surgery. For this secondary analysis, detailed rin bolus and administered via a central venous catheter. demographic data, preoperative laboratory values, opera- Heparin was not added to the CPB pump prime. Through- tive data, heparin bolus dose, anticoagulation data includ- out the entire study period, an ACT of Ͼ350 seconds was ing activated coagulation time (ACT) values, and measured used in patients undergoing surgery where cardiotomy heparin concentrations were collected from hospital records. suction was to be used. Patients undergoing primary CABG Patients with missing laboratory or demographic data (n ϭ surgery without the use of cardiotomy suction were anti- 27) and those receiving warfarin (n ϭ 15) were further coagulated using a protocol that prescribed a minimal ACT excluded from analysis, yielding 304 analyzable patients. of 300 seconds before the institution of CPB. Three minutes ICU length of stay until first discharge was recorded in after USP porcine heparin (APP Pharmaceuticals, Schaum- hours. Hospital length of stay was recorded in days, with burg, IL) administration, heparin concentration and ACT the surgical day and last hospital day included as whole were remeasured. This heparin management protocol days. including use of heparin-coated circuits was adopted in Baseline blood samples for ACT, HDR, AT, and other a comprehensive institutional program to reduce the rate assays were obtained after induction of anesthesia but of reoperation for bleeding. All patients received an before surgical incision (described as preoperative hereaf- ⑀-aminocaproic acid initial loading dose of 7.5 to 10 g over ter). Free unbound AT activity was measured with a 1 hour, after the initial blood draw for AT level and baseline colorimetric method (Modular Analytics biochemistry ana- ACT, but before heparin administration and blood sam- lyzer, Siemens Healthcare Diagnostics, Tarrytown, NY) pling for measurement of postheparin ACT. performed by Charles River Laboratory (Montreal, Can- HDR was measured as the difference in ACT between ada). The assay limit of quantitation was 21.6%. To report target and baseline ACT measurements, divided by target human plasma AT activity results in IU/mL, the National heparin level estimated from the Hepcon HMS Plus system. Institute for Biological Standards and Control Second In- Because the Hepcon HMS Plus system has a limited fidelity ternational Reference Standard was used to determine a in reporting whole blood heparin concentrations, in that it conversion factor of activity in IU/mL ϭ AT activity in % ϫ provides values with discrete categories (i.e., 0.7, 1.4, 2.0, 0.0102. AT content was measured using an immunoneph- 2.7, and 3.4 U/mL) rather than a continuous variable, the elometric method using a BN-100 Prospec nephelometer heparin sensitivity index (HSI) was also calculated from (Dade Behring Diagnostics, Marburg, Germany). Coeffi- change in ACT between before and after heparin adminis- cient of variation for AT measurement was Ͻ5% within tration, divided by heparin dose, per kilogram of body assay and Ͻ10% across assays. The assay limit of quanti- weight.11 The same calculation was performed using hep- tation was 0.00672 mg/mL. To report human plasma AT arin dose per liter of estimated blood volume without content results in IU/mL, National Institute for Biological significantly changing the results; therefore, body weight Standards and Control Second International Reference was used. Standard was used to determine a conversion factor of We estimated the accessible effect size for HDR, using content in IU/mL ϭ content in g/L ϫ 3.64. Both assays the available sample size (n ϭ 319), a 20% rate of heparin measure free AT, rather than AT complexed with heparin. resistance based on prior studies, a mean HDR of 99 Plasminogen-activator inhibitor-1, tissue factor, d-dimer, s/U/mL and an sd of 22 s/U/mL, a type I error rate of 5% protein C, and cardiac troponin I (cTnI) were measured and a type II error rate of 20%. We estimated that we would using a sandwich immunoassay on a triage platform using be able to observe differences in HDR of 9 s/U/mL, which monoclonal and polyclonal antibodies (Biosite, San Diego, we thought would be more sensitive than a clinically CA). Hemoglobin, platelet, and white blood counts, along important difference. with prothrombin time, partial thromboplastin time, and Data are presented as mean Ϯ sd when normally international normalized ratio were measured by a central distributed, or median and 5th and 95th percentiles when hematology laboratory according to institutional protocols. not normally distributed. Data that were nonnormally Complete blood counts were performed using the Advia distributed were compared using the Wilcoxon ranked sum 2120 Hematology System (Siemens Healthcare Diagnostics, test. The Student t test was used to compare means of Deerfield, IL), and coagulation studies were performed on normally distributed data. Subgroup analysis of patients STA-R Evolution (Diagnostica Stago, Parsippany, NJ). with low AT activity (Ͻ80% of laboratory normal; AT

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 857 Antithrombin and Heparin Response

Table 1. Demographics and Perioperative Data of Patients with Low Antithrombin (AT) Activity (<80% of Normal; Deﬁned as AT Activity <0.813 U/mL) and Normal AT Activity Study cohort AT activity <80% of normala AT activity >80% of normala P (255 ؍ N) (49 ؍ N) (304 ؍ N) Age 65 Ϯ 970Ϯ 964Ϯ 9 Ͻ0.0001 Male gender 247 (81.3%) 42 (85.7%) 205 (80.4%) 0.43 Race (Caucasian) 272 (89.5%) 44 (89.8%) 228 (89.4%) 1.0 Height (cm) 173 Ϯ 9 174 Ϯ 9 173 Ϯ 9 0.56 Weight (kg) 88.3 Ϯ 18.6 87.0 Ϯ 16.6 88.4 Ϯ 19.0 0.62 Medical history Previous myocardial infarction 130 (42.7%) 27 (55.1%) 103 (40.4%) 0.06 Myocardial infarction within prior 2 wk 56 (18.4%) 16 (32.7%) 40 (15.7%) 0.008 Hypertension 224 (73.7%) 30 (61.2%) 194 (76.1%) 0.03 Hypercholesterolemia 236 (77.6%) 33 (67.4%) 203 (79.6%) 0.06 Insulin-dependent diabetes 31 (10.2%) 9 (18.4%) 22 (8.6%) 0.07 Renal insufﬁciency (creatinine Ն2.0 mg/dL) 8 (2.6%) 2 (4.1%) 6 (2.4%) 0.62 Liver failure 0 (0%) 0 (0%) 0 (0%) — Preoperative medications Aspirin 256 (84.2%) 38 (77.6%) 218 (85.4%) 0.20 Platelet inhibitor (excluding aspirin) 71 (23.3%) 14 (28.6%) 57 (22.4%) 0.36 Intravenous heparin in current admission 88 (28.9%) 33 (67.4%) 55 (21.6%) Ͻ0.0001 Any statin 252 (82.8%) 41 (83.7%) 211 (82.8%) 1.0 Preoperative laboratory values Hematocrit (%) 39.9 Ϯ 4.5 39.5 Ϯ 5.3 40.0 Ϯ 4.3 0.52 Platelet count (109/mL) 237 (169–322) 245 (170–339) 236 (167–317) 0.55 White cell count (103/mL) 8.0 (5.5–11.2) 8.6 (5.4–11.8) 7.8 (5.5–11.1) 0.27 INR 1.0 (1.0–1.1) 1.0 (1.0–1.2) 1.0 (1.0–1.1) 0.02 INR Ͼ1.4 0 (0%) 0 (0%) 0 (0%) — Partial thromboplastin time (s) 30.2 (26.5–63.2) 39 (28–94) 30 (26–40) Ͻ0.0001 Antithrombin content (U/mL) 0.862 Ϯ 0.123 0.703 (0.582–0.790) 0.881 (0.772–1.04) Antithrombin content (mg/mL) 0.23 (0.196–0.282) 0.193 (0.160–0.217) 0.242 (0.212–0.286) Antithrombin activity (U/mL) 0.933 Ϯ 0.126 0.761 (0.649–0.811) 0.961 (0.855–1.11) Antithrombin activity (%) 91.1 (76.3–107.2) 74.8 (63.6–79.5) 93.8 (83.7–108.7) Protein C (␮g/mL) 5.01 (2.27–6.98) 4.75 (3.32–6.63) 5.07 (3.53–6.96) 0.10 Plasminogen activator inhibitor-1 (ng/mL) 1.53 (0.0–20.2) 1.18 (0.23–3.88) 1.58 (0.06–6.92) 0.16 D-dimer (ng/mL) 183.3 (0.0–2569) 142.8 (6.8–1084) 186.6 (7.2–1381) 0.52 Tissue factor (pg/mL) 16.4 (0.03–495.6) 12.6 (0.04–73.3) 16.5 (0.04–118.8) 0.49 Heparin administration Baseline ACT (s) 135 Ϯ 18 138 (114–163) 136 (114–156) 0.12 Target ACT 300 s 257 (84.5%) 41 (16.0%) 216 (84.0%) 350 s 47 (15.5%) 8 (17.0%) 42 (83.0%) 0.86 Heparin dose (U/kg) 150 (86–246) 154 (112–198) 150 (111–211) 0.75 Heparin dose Ͼ200 U/kg 59 (19.4%) 8 (16.3%) 51 (20.0%) 0.69 ACT after heparin administration 300 s 366 (300–456) 357 (300–469) 366 (300–456) 0.53 350 s 389 (324–460) 395 (366–425) 389 (319–470) 0.81 Calculated HDR slope (s/U/mL) 98 Ϯ 21 96 Ϯ 20 98 Ϯ 21 0.50 Heparin sensitivity index (s/U/kg) 1.59 Ϯ 0.41 1.58 Ϯ 0.40 1.60 Ϯ 0.41 0.68 Clinical outcomes Cardiac troponin I on POD 1 (␮g/L) 0.99 (0.13–7.72) 1.17 (0.31–3.21) 0.97 (0.32–3.89) 0.48 ICU duration (h) 45 (21–94) 48 (24–116) 44 (21–90) 0.13 Postsurgical hospital stay (d) 6 (4–10) 6 (4–11) 6 (4–10) 0.17

Data presented as mean Ϯ SD or median (10th–90th percent conﬁdence intervals). INR ϭ International Normalized Ratio; ACT ϭ activated clotting time; POD 1 ϭ postoperative day 1; HDR ϭ heparin dose response; ICU ϭ intensive care unit. a Deﬁned as 100% of the laboratory normal for AT activity, not corrected for the National Institute for Biological Standards and Control Second International Reference Standard (see Methods).

activity Ͻ0.813 U/mL) who might be at risk for heparin RESULTS resistance was performed. Multivariable linear regres- Baseline demographics and perioperative data for 304 sion modeling was performed to examine for clinical and patients are summarized in Table 1. Thirty-two patients laboratory predictors of HDR and HSI; variables with (10.5%) required additional heparin administration before univariate P values Ͻ0.2 were entered into a combined the institution of CPB for failing to achieve their respective forward/backward stepwise linear regression model, with target ACT with administration of the heparin dose esti- an exit P value of 0.05. A 2-sided P Ͻ 0.05 was considered mated by the HepCon HMS Plus system. No patient as showing statistical significance. Statistical analyses were received Ͼ300 U/kg heparin. No patients were given fresh performed using SAS version 9.1.3 and JMP 7.03 (SAS frozen plasma or AT supplementation to reach target ACT Institute, Cary, NC). values.

858 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 1. Preoperative antithrombin (AT) activity versus heparin dose response (HDR) (r2 Ͻ 0.001; P ϭ 0.59). Conﬁ- dence interval of HDR for a single measurement of AT activity is shaded. HDR slope (s/U/mL) ϭ 94.3 ϩ 5.31 ϫ AT activity (U/mL).

Table 2. Multivariable Predictors of Heparin-Dose Response (HDR) Without AT activity With AT activity 0.140 ؍ r2 0.138 ؍ r2 Heparin dose response (s/U/mL) Estimate P Estimate P Female gender Ϫ1.82 Ϯ 1.66 0.27 Ϫ2.02 Ϯ 1.67 0.23 Age (y) 0.16 Ϯ 0.13 0.23 0.176 Ϯ 0.131 0.18 Caucasian race Ϫ3.81 Ϯ 1.87 0.042 Ϫ3.69 Ϯ 1.87 0.049 Hypercholesterolemia 5.61 Ϯ 1.39 Ͻ0.0001 5.52 Ϯ 1.40 Ͻ0.0001 Hematocrit 0.71 Ϯ 0.29 0.016 0.70 Ϯ 0.29 0.016 Prothrombin time (s) 5.57 Ϯ 1.82 0.002 5.83 Ϯ 1.84 0.002 White cell count (ϫ106/mL) Ϫ1.52 Ϯ 0.51 0.003 Ϫ1.50 Ϯ 0.51 0.004 AT activity (U/mL) 8.32 Ϯ 9.64 0.39 AT ϭ antithrombin.

AT activity and content were highly correlated (r2 ϭ testing or management (Table 1). No relationship between AT 0.801), therefore AT activity is reported. Higher AT levels activity and HDR was observed in these patients (r2 Ͻ 0.001). (P Ͻ 0.05) were observed in females, younger individuals, Furthermore, there was no evidence of diminished heparin current smokers, hypercholesterolemia, and individuals responsiveness in this group, because 94.1% (48 of 51) who had not had a myocardial infarction within the previ- achieved the target ACT after administration of the calculated ous 2 weeks. No association was observed between base- heparin bolus dose. Neither HDR nor HSI was significantly line AT activity and baseline ACT (r2 Ͻ 0.001; P ϭ 0.10) or related to AT activity in univariate relationship (Fig. 1), or with HDR (r2 Ͻ 0.001; P ϭ 0.59) (Fig. 1), HSI (r2 Ͻ 0.001; P ϭ after accounting for other covariates using multivariable lin- 0.73), or platelet count (r2 ϭ 0.004; P ϭ 0.24). Those patients ear regression (Tables 2 and 3). with recent preoperative heparin exposure had signifi- Preoperative AT activity, HDR, and HSI were not cantly lower AT activity (0.87 vs 0.96 U/mL; P Ͻ 0.001) but associated with cTnI levels on the first postoperative day did not show a significant difference in heparin require- (all r2 Ͻ 0.002; P Ͼ 0.5), ICU duration, or hospital length ments or heparin responsiveness by any measure including of stay. HDR or HSI. Individuals with AT activity Ͻ80% normal (Ͻ0.813 U/mL) DISCUSSION had higher partial thromboplastin time and were more likely We observed no relationship between preoperative AT to have received heparin during the current admission before activity and response to heparin, measured using either surgery, but otherwise showed no differences in coagulation HDR or HSI, in these 304 patients undergoing primary

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 859 Antithrombin and Heparin Response

Table 3. Multivariable Predictors of Heparin Sensitivity Index (HSI) Without AT activity With AT activity 0.137 ؍ r2 0.136 ؍ r2 Heparin sensitivity index (s/U/kg) Estimate P Estimate P Female gender 0.070 Ϯ 0.032 0.029 0.068 Ϯ 0.032 0.037 Age (y) 0.001 Ϯ 0.003 0.79 0.001 Ϯ 0.003 0.72 Caucasian race 0.019 Ϯ 0.037 0.61 0.021 Ϯ 0.037 0.58 Platelet count (ϫ109/mL) Ϫ0.0007 Ϯ 0.0003 0.013 Ϫ0.0007 Ϯ 0.0003 0.013 PT (s) 0.10 Ϯ 0.04 0.006 0.10 Ϯ 0.04 0.005 White cell count (ϫ106/mL) Ϫ0.022 Ϯ 0.011 0.048 Ϫ0.022 Ϯ 0.011 0.048 PAI-1 level (ng/mL) 0.032 Ϯ 0.009 0.0005 0.031 Ϯ 0.009 0.0007 Tissue factor level (pg/mL) Ϫ0.0008 Ϯ 0.0003 0.015 Ϫ0.0008 Ϯ 0.0003 0.016 AT activity (U/mL) 0.121 Ϯ 0.198 0.54 PAI-1 ϭ plasminogen activator inhibitor level-1; AT ϭ antithrombin; PT ϭ prothrombin time.

CABG surgery. Additional subgroup analysis in patients patients with preoperative AT activity Ͻ80% of normal, with baseline AT activity Ͻ80% failed to identify AT whereas no relationship was found in patients with normal activity as a predictor of HDR or HSI. However, no patient AT levels. Our data contrast with this, in that we observed exhibited a requirement for a heparin dose Ͼ300 U/kg to no relationship between AT activity and heparin respon- achieve target ACTs of either 300 or 350 seconds. siveness in either group. Acquired AT deficiency is common in patients with The ACT response to heparin is complex and affected by 17 previous heparin administration, critical illness, severe different factors as previously noted. Although AT en- hepatic dysfunction, and after major cardiovascular sur- hancement is a primary mechanism of heparin’s action, 18,19 gery. After cardiac surgery, lower levels of AT have other factors may influence HDR including tissue factor been independently associated with prolonged ICU stay pathway inhibitor levels in vivo but not in vitro, extravas- and a higher incidence of neurologic and thromboembolic cular sequestration of heparin, plasma protein binding, 12 events. In this cohort of patients, we observed associa- leukocyte lactoferrin, and activated platelets.7,14,22 Tissue tions between clinical markers of anticoagulation (pro- factor pathway inhibitor is an endothelium-derived endog- thrombin time and ACT) and factors that may relate to the enous serine protease inhibitor with enhanced expression severity of illness including white blood cell count and after heparin administration23 that may enhance anticoagu- hematocrit, and HDR. However, neither AT activity nor lation in vivo.24 Other endogenous mechanisms may heparin sensitivity was associated with severity of myocar- further modify the activity of heparin in vivo. Higher dial injury, ICU length of stay, or hospital length of stay, as molecular weight heparin (Ͼ13 kDa) is sequestered and surrogates of severity of illness. deactivated by endothelial endocytosis and depolymeriza- Administration of unfractionated heparin remains the tion.25 In addition, neutrophil-derived lactoferrin can neu- mainstay of anticoagulation management for patients un- tralize heparin by ionic binding.22 Heparin has shown to be dergoing cardiac surgery requiring CPB, with the goal of similarly neutralized by histidine-rich glycoproteins such maintaining therapeutic anticoagulation, thereby prevent- as vitronectin, fibronectin, and kininogen.7 Platelet factor 4 ing thromboembolic complications. Its unique properties of is a potent heparin binding agent released from activated rapidly providing systemic anticoagulation that can be platelets that reverses the ACT,26,27 potentially contributing maintained for the duration of CPB and rapid complete reversal with subsequent administration of protamine to heparin resistance. These endogenous mechanisms may make heparin the anticoagulant of choice for CPB. Interin- be important in modifying heparin’s anticoagulant activity dividual variability in heparin response is well described,20 in vivo, but difficult to quantify in vitro. and contributing to this variation may be previously ob- This study has important limitations. Our study did not served risk factors for diminished heparin responsiveness identify patients with severe heparin resistance, with only that include low AT levels, platelet count Ͼ300,000 10.5% of patients failing to reach the target ACT, and none cells/mm3, and heparin pretreatment.9 Although we ob- required fresh frozen plasma or AT to initiate CPB. Perhaps served substantial variability in heparin responsiveness, the relatively low ACT target contributed to the lack of our data failed to show any relationship between heparin identification of heparin-resistant patients. Historically, response and AT activity or heparin pretreatment. CPB has been initiated with higher ACTs to give a margin 28 AT is critical for maintenance of anticoagulation during of safety and in many studies, a large number of patients CPB and is consumed during the process.21 Despotis et diagnosed as heparin resistant had target ACTs Ͼ400 al.14 demonstrated a strong association between in vitro AT seconds.5–8,13,14 Definitions of heparin resistance based on and heparin responsiveness measured by ACT slope over a dose of heparin to achieve a specific target ACT such as the range of AT levels of 0.2 to 1 U/mL. At AT levels above requiring Ͼ500 IU/kg to achieve a target ACT of 480 1 U/mL, there was no further increase in heparin respon- seconds assume linearity of the HDR.10 Although Despotis siveness. However, there was much weaker association in et al.29 observed a strong linear relationship between ACT 31 patients undergoing cardiac surgery with CPB.14 Simi- and heparin concentration observed over a range of hepa- larly, Dietrich et al.11 observed a diminished HSI in adult rin concentrations, to assume that our results could be

860 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA extrapolated to higher target ACTs would not be appropri- 11. Dietrich W, Braun S, Spannagl M, Richter JA. Low preopera- ate. Furthermore, we did not measure circulating concen- tive antithrombin activity causes reduced response to heparin in adult but not in infant cardiac-surgical patients. Anesth trations of several important members of the coagulation Analg 2001;92:66–71 pathway that may have affected these results. 12. Ranucci M, Frigiola A, Menicanti L, Ditta A, Boncilli A, Brozzi In conclusion, we found that heparin responses were S. Postoperative antithrombin levels and outcome in cardiac independent of preoperative plasma AT activity in patients operations. Crit Care Med 2005;33:355–60 undergoing primary CABG using target ACTs of 300 to 350 13. Lemmer JH Jr, Despotis GJ. Antithrombin III concentrate to treat heparin resistance in patients undergoing cardiac sur- seconds. Preoperative heparin exposure was associated gery. J Thorac Cardiovasc Surg 2002;123:213–7 with diminished AT activity, but no change in HDR was 14. Despotis GJ, Levine V, Joist JH, Joiner-Maier D, Spitznagel E. observed. Perioperative AT activity, HDR, or HSI were not Antithrombin III during cardiac surgery: effect on response of associated with cTnI levels on the first postoperative day, activated clotting time to heparin and relationship to markers of hemostatic activation. Anesth Analg 1997;85:498–506 ICU duration, or hospital length of stay. 15. Medtronic Perfusion Systems. Hepcon HMS Plus Operator’s Manual. Minneapolis, MN: Medtronic Perfusion Systems, 2001 DISCLOSURE 16. Allen TH, Peng MT, Chen KP, Huang TF, Chang C, Fang HS. Dr. Garvin was the recipient of a research fellowship funded Prediction of blood volume and adiposity in man from body weight and cube of height. Metabolism 1956;5:328–45 by Talecris Biotherapeutics, which allowed time for generation 17. Dietrich W, Spannagl M, Schramm W, Vogt W, Barankay A, of this and other articles. Dr. Body has received consulting fees Richter JA. The influence of preoperative anticoagulation on from Talecris Biotherapeutics for another study. heparin response during cardiopulmonary bypass. J Thorac Talecris Biotherapeutics paid for the costs of antithrombin Cardiovasc Surg 1991;102:505–14 analyses performed by Charles River Laboratories in Ontario, 18. Maclean PS, Tait RC. Hereditary and acquired antithrombin Canada, but had no input into these analyses. deficiency: epidemiology, pathogenesis and treatment options. Drugs 2007;67:1429–40 19. Levy JH, Despotis GJ, Szlam F, Olson P, Meeker D, Weisinger A. REFERENCES Recombinant human transgenic antithrombin in cardiac surgery: 1. McLean J. The thromboplastin action of cephalin. Am J Physiol a dose-finding study. Anesthesiology 2002;96:1095–102 1916;41:250–7 20. Bull BS, Korpman RA, Huse WM, Briggs BD. Heparin therapy 2. Rosenberg RD, Damus PS. The purification and mechanism of during extracorporeal circulation. I. Problems inherent in existing action of human antithrombin-heparin cofactor. J Biol Chem heparin protocols. J Thorac Cardiovasc Surg 1975;69:674–84 1973;248:6490–505 21. Hashimoto K, Yamagishi M, Sasaki T, Nakano M, Kurosawa H. 3. Rosenberg RD. Actions and interactions of antithrombin and Heparin and antithrombin III levels during cardiopulmonary heparin. N Engl J Med 1975;292:146–51 bypass: correlation with subclinical plasma coagulation. Ann 4. Langdown J, Johnson DJ, Baglin TP, Huntington JA. Allosteric Thorac Surg 1994;58:799–804; discussion 804–5 activation of antithrombin critically depends upon hinge re- 22. Wu HF, Lundblad RL, Church FC. Neutralization of heparin gion extension. J Biol Chem 2004;279:47288–97 activity by neutrophil lactoferrin. Blood 1995;85:421–8 5. Avidan MS, Levy JH, Scholz J, Delphin E, Rosseel PM, Howie 23. Bajaj MS, Kuppuswamy MN, Saito H, Spitzer SG, Bajaj SP. MB, Gratz I, Bush CR, Skubas N, Aldea GS, Licina M, Bonfiglio Cultured normal human hepatocytes do not synthesize LJ, Kajdasz DK, Ott E, Despotis GJ. A phase III, double-blind, lipoprotein-associated coagulation inhibitor: evidence that en- placebo-controlled, multicenter study on the efficacy of recom- dothelium is the principal site of its synthesis. Proc Natl Acad binant human antithrombin in heparin-resistant patients Sci USA 1990;87:8869–73 scheduled to undergo cardiac surgery necessitating cardiopul- 24. Brodin E, Appelbom H, Osterud B, Hilden I, Petersen LC, monary bypass. Anesthesiology 2005;102:276–84 Hansen JB. Regulation of thrombin generation by TFPI in 6. Avidan MS, Levy JH, van Aken H, Feneck RO, Latimer RD, Ott plasma without and with heparin. Transl Res 2009;153:124–31 E, Martin E, Birnbaum DE, Bonfiglio LJ, Kajdasz DK, Despotis 25. Barzu T, van Rijn JL, Petitou M, Tobelem G, Caen JP. Heparin GJ. Recombinant human antithrombin III restores heparin degradation in the endothelial cells. Thromb Res 1987;47:601–9 responsiveness and decreases activation of coagulation in 26. Lane DA, Pejler G, Flynn AM, Thompson EA, Lindahl U. heparin-resistant patients during cardiopulmonary bypass. Neutralization of heparin-related saccharides by histidine-rich J Thorac Cardiovasc Surg 2005;130:107–13 glycoprotein and platelet factor 4. J Biol Chem 1986;261:3980–6 7. Ranucci M, Isgro G, Cazzaniga A, Ditta A, Boncilli A, Cotza M, 27. Levy JH, Cormack JG, Morales A. Heparin neutralization by Carboni G, Brozzi S. Different patterns of heparin resistance: recombinant platelet factor 4 and protamine. Anesth Analg therapeutic implications. Perfusion 2002;17:199–204 1995;81:35–7 8. Williams MR, D’Ambra AB, Beck JR, Spanier TB, Morales DL, 28. Young JA, Kisker CT, Doty DB. Adequate anticoagulation Helman DN, Oz MC. A randomized trial of antithrombin during cardiopulmonary bypass determined by activated clot- concentrate for treatment of heparin resistance. Ann Thorac ting time and the appearance of fibrin monomer. Ann Thorac Surg 2000;70:873–7 Surg 1978;26:231–40 9. Ranucci M, Isgro G, Cazzaniga A, Soro G, Menicanti L, Frigiola 29. Despotis GJ, Summerfield AL, Joist JH, Goodnough LT, San- A. Predictors for heparin resistance in patients undergoing toro SA, Spitznagel E, Cox JL, Lappas DG. Comparison of coronary artery bypass grafting. Perfusion 1999;14:437–42 activated coagulation time and whole blood heparin measure- 10. Staples M, Dunton D, Karlson K, Leonardi H, Berger R. ments with laboratory plasma anti-Xa heparin concentration in Heparin resistance after preoperative heparin therapy or inpatients having cardiac operations. J Thorac Cardiovasc Surg traaortic ballon pumping. Ann Thorac Surg 1994;57:1211–6 1994;108:1076–82

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 861 Postoperative Activity, but Not Preoperative Activity, of Antithrombin Is Associated with Major Adverse Cardiac Events After Coronary Artery Bypass Graft Surgery

Sean Garvin, MD,* Jochen D. Muehlschlegel, MD,* Tjo¨rvi E. Perry, MD,* Junliang Chen, PhD,† Kuang-Yu Liu, PhD,* Amanda A. Fox, MD,* Charles D. Collard, MD,‡ Sary F. Aranki, MD,§ Stanton K. Shernan, MD,* and Simon C. Body, MB, ChB, MPH*

BACKGROUND: Low levels of antithrombin (AT) have been independently associated with prolonged intensive care unit stay and an increased incidence of neurologic and thromboembolic events after cardiac surgery. We hypothesized that perioperative AT activity is independently associated with postoperative major adverse cardiac events (MACEs) in patients undergoing coronary artery bypass graft (CABG) surgery. METHODS: We prospectively studied 1403 patients undergoing primary CABG surgery with cardiopulmonary bypass (CPB) (http://clinicaltrials.gov/show/NCT00281164). The primary clinical end point was occurrence of MACE, defined as a composite outcome of any one or more of the following: postoperative death, reoperation for coronary graft occlusion, myocardial infarction, stroke, pulmonary embolism, or cardiac arrest until first hospital discharge. Plasma AT activity was measured before surgery, after post-CPB protamine, and on postoperative days (PODs) 1–5. Multivariate logistic regression modeling was performed to estimate the independent effect of perioperative AT activity upon MACE. RESULTS: MACE occurred in 146 patients (10.4%), consisting of postoperative mortality (n ϭ 12), myocardial infarction (n ϭ 108), stroke (n ϭ 17), pulmonary embolism (n ϭ 8), cardiac arrest (n ϭ 16), or a subsequent postoperative or catheter-based treatment for graft occlusion (n ϭ 6). AT activity at baseline did not differ between patients with (0.91 Ϯ 0.13 IU/mL; n ϭ 146) and without (0.92 Ϯ 0.13 IU/mL; n ϭ 1257) (P ϭ 0.18) MACE. AT activity in both groups was markedly reduced immediately after CPB and recovered to baseline values over the ensuing 5 PODs. Postoperative AT activity was significantly lower in patients with MACE than those without MACE. After adjustment for clinical predictors of MACE, AT activity on PODs 2 and 3 was associated with MACE. CONCLUSIONS: Preoperative AT activity is not associated with MACE after CABG surgery. MACE is independently associated with postoperative AT activity but only at time points occurring predominantly after the MACE. (Anesth Analg 2010;111:862–9)

ntithrombin (AT) is a serine protease inhibitor AT levels are decreased after administration of heparin (serpin) and the principal inhibitor of the final due to degradation of the ternary complex. Additionally, Acommon pathway of the coagulation system by acquired AT deficiency is common in patients with critical inactivation of circulating thrombin (factor IIa) and factor illness, severe hepatic dysfunction, and after major cardio- Xa, among other serine proteases. Heparin increases AT vascular surgery.1,2 The magnitude of reduction in AT after activity 2000- to 4000-fold due to a conformational change cardiac surgery is similar to that in patients with heterozy- in the quaternary structure of AT by heparin binding and gous AT deficiency, which is associated with increased risk 3,4 through formation of a ternary complex of thrombin, AT, of thromboembolic events. After cardiac surgery, lower and heparin. Heparin-augmented AT activity is still the levels of AT have been independently associated with principal mechanism of anticoagulation for cardiopulmo- prolonged intensive care unit (ICU) stay and a higher 5 nary bypass (CPB). incidence of neurologic and thromboembolic events. We therefore examined for independent association between From the *Department of Anesthesiology, Perioperative and Pain Medicine, perioperative AT activity and the frequency of postopera- Brigham and Women’s Hospital, Harvard Medical School, Boston, Massa- tive major adverse cardiac events (MACEs) in patients chusetts; †Talecris Biotherapeutics, Research Triangle Park, Durham, North undergoing coronary artery bypass graft (CABG) surgery. Carolina; ‡Baylor College of Medicine Division of Cardiovascular Anesthe- sia at the Texas Heart Institute, Saint Luke’s Episcopal Hospital, Houston, Texas; and §Division of Cardiac Surgery, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts. METHODS Accepted for publication June 22, 2009. The study cohort was obtained from a continuing prospective The authors had full access to the data and take responsibility for its longitudinal parent study of 1447 patients undergoing primary integrity. All authors have read and agree to the manuscript as written. CABG surgery with CPB between August 2001 and May 2006 at Address correspondence and reprint requests to Simon C. Body, MB, ChB, 2 United States academic medical centers (CABG Genomics MPH, Department of Anesthesiology, Perioperative and Pain Medicine, Program; http://clinicaltrials.gov/show/NCT00281164). With Brigham and Women’s Hospital, 75 Francis St., Boston, MA 02115. Address e-mail to [email protected]. IRB approval, written informed consent was obtained from each Copyright © 2010 International Anesthesia Research Society patient. Patients were excluded from the parent study if they DOI: 10.1213/ANE.0b013e3181b7908c were younger than 20 yr old, underwent repeat or off-pump

862 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 CABG, had a preoperative hematocrit Ͻ25%, or if they had plasma AT content results in IU/mL, the National Institute received leukocyte-rich blood products within 30 days before for Biological Standards and Control Second International surgery. All patients enrolled in the parent study were included. Reference Standard was used to determine a conversion Patients without postoperative cardiac troponin I (cTnI) levels factor of content in IU/mL ϭ content in g/L ϫ 3.64. Both (n ϭ 43) were excluded from further analysis. Demographic assays measure free AT rather than AT complexed with data, medical and surgical history, medications, and out- heparin. Assays were performed by Charles River Labora- comes were recorded by trained research staff using defined tory in Montreal, Canada by personnel blinded to outcome protocols in a purpose-built case report form. The examina- status. Subsequent comparison of paired AT activity and tion of the relationship between AT levels and MACE was not content data revealed high correlation (r2 ϭ 0.878), so only prespecified in the original parent study. the activity is reported. Perioperative anticoagulation protocols differed between institutions. At Brigham and Women’s Hospital, Statistical Methods patients received 300 U per kg body weight of porcine Statistical analyses were performed using SAS, version heparin to achieve an activated clotting time (ACT) of Ͼ400 9.1.3, and JMP 7.0 (SAS Institute, Cary, NC). AT activity s, until February 2004. From February 2004, patients re- was normally distributed at all time points, so was not ceived a Hepcon HMS Plus (Medtronic, Minneapolis, MN) transformed. Data are presented as mean (sd) and median calculated dose of porcine heparin to achieve an ACT of with 10%–90% interquantile range, unless otherwise stated. either 300 or 350 s. At Texas Heart Institute, 300 U per kg Continuous variables were compared using analysis of vari- body weight of either bovine or porcine heparin was given ance or Wilcoxon Mann-Whitney ranked sum test when to achieve an ACT of Ͼ400 s. appropriate. Categorical variables were compared with ␹2 or Fisher’s exact test. Clinical End Points for the Test Cohort Multivariate logistic regression modeling was per- The primary clinical end point was prespecified as the formed to identify and account for MACE risk factors that occurrence of a MACE, defined as a composite outcome of might confound any association between low AT activity any one or more of the following: postoperative mortality and MACE. The multivariate analysis used a forward (defined as all deaths occurring within 30 days of the stepwise technique to identify independent risk factors for operation or occurring during the primary hospitalization), MACE, whereby clinically relevant demographic variables reoperation for coronary graft occlusion, myocardial infarc- and variables with a two-tailed univariate P Յ 0.2 were tion (MI) (predefined as peak postoperative cTnI concen- entered into the model and P Յ 0.2 was necessary to remain tration Ͼ12 ng/mL, being the upper 8th percentile), cardiac in the model. Age, gender, race, body mass index, and institu- arrest (defined as a postoperative event requiring cardio- tion were forced into the model. Nagelkerke generalized r2 and pulmonary resuscitation) until first hospital discharge, likelihood ratio test were used to determine the additional thromboembolic event consisting of stroke (defined as a predictive value of AT upon MACE. F tests were used to clinical diagnosis of focal or global neurological deficit), or compare generalized r2. Odds ratios and 95% confidence pulmonary embolism (diagnosed by ventilation perfusion intervals for a 0.1 IU/mL decrease in AT activity were scan of moderate to high probability or by a positive estimated. A two-sided P Ͻ 0.05 was considered significant. pulmonary angiogram). RESULTS Cardiac Biomarker Assay The cohort comprised 1403 patients undergoing CABG Blood samples were obtained before surgery, 5 min after surgery whose characteristics are described in Table 1. administration of post-CPB protamine, and on the mornings MACE occurred in 146 patients (10.4%), consisting of of postoperative days (PODs) 1–5. Citrated plasma was stored postoperative mortality (n ϭ 12), MI (n ϭ 108), stroke (n ϭ in vapor-phase liquid nitrogen until analysis for cTnI with a 17), pulmonary embolism (n ϭ 8), cardiac arrest (n ϭ 16), or sandwich immunoassay on a Triage® platform using mono- a subsequent postoperative or catheter-based treatment for clonal and polyclonal antibodies (Biosite, San Diego, CA) at graft occlusion (n ϭ 6). Nineteen patients had 2 or 3 events, a single core facility. Patient caregivers were not aware of usually MI, with either subsequent death or stroke. Most the results of the assays as they were performed after adverse events occurred before or on POD 2. Of 12 patients patient discharge. with operative mortality, 2 patients died at POD 0 and 10 patients died on or after POD 5. Of 17 patients with stroke, Antithrombin Assays 6 patients had a stroke on or before POD 2. Of 108 patients AT activity was measured with a colorimetric method using a with MI, 89 patients had a diagnosis of MI first occurring Modular Analytics biochemistry analyzer (Siemens Health- on POD 1. MACE frequency did not differ between insti- care Diagnostics, Tarrytown, NY). The assay limit of quanti- tutions (Table 1). tation was 21.6%. To report human plasma AT activity results AT activity and content were measured at the 7 time in IU/mL, the National Institute for Biological Standards and points (Table 2). AT activity at baseline did not differ Control Second International Reference Standard was used to between patients with MACE (0.91 Ϯ 0.13 IU/mL; n ϭ 146) determine a conversion factor of activity in IU/mL ϭ activity and those without MACE (0.92 Ϯ 0.13 IU/mL; n ϭ 1257) in % ϫ 0.0102. AT content was measured using an immuno- (P ϭ 0.18). AT activity was significantly reduced at the ephelometric method using a BN-100 ProSpec nephelometer post-CPB measurement compared with the preoperative (Dade Behring Diagnostics, Marburg, Germany). The assay time point (P Ͻ 0.0001) and returned to baseline levels over limit of quantitation was 0.00672 mg/mL. To report human the ensuing 5-day period in patients with and without MACE.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 863 Antithrombin and Adverse Events

Table 1. Demographic and Clinical Characteristics of the Cohort as a Whole and Stratiﬁed by the Occurrence of Major Adverse Cardiac Events (MACE) Entire cohort Patients with MACE Patients without MACE P (1257 ؍ n) (146 ؍ n) (1403 ؍ n) Age Ͻ55 yr 240 (17.1) 22 (15.1) 218 (17.3) 55 to Ͻ65 yr 494 (35.2) 51 (34.9) 443 (35.2) 65 to Ͻ75 yr 422 (30.1) 42 (28.8) 380 (30.2) 75 to Ͻ85 yr 231 (16.5) 29 (19.9) 202 (16.1) At least 85 yr 16 (1.1) 2 (1.4) 14 (1.1) 0.75 Male 1136 (81.0) 109 (74.7) 1027 (81.7) 0.045 Caucasian race 1189 (84.6) 116 (79.5) 1073 (85.4) 0.068 Body mass index (kg/m2) 29.4 (5.5) 30.2 (5.5) 29.3 (5.4) 0.063 Institution BWH 1061 (75.6) 109 (10.3) 952 (89.7) 0.76 THI 342 (24.4) 37 (10.8) 305 (89.2) Medical history Diabetes (drug treated; %) 466 (33.2) 52 (35.6) 414 (32.9) 0.52 Pulmonary disease (%) 224 (16.0) 17 (11.6) 207 (16.5) 0.13 Creatinine (mg/dL) 1.10 (0.334) 1.13 (0.322) 1.10 (0.335) 0.27 Hematocrit (%) 40.11 (4.724) 39.48 (4.825) 40.2 (4.708) 0.099 Hypertension (%) 1051 (74.9) 117 (80.1) 934 (74.3) 0.12 Hypercholesterolemia (%) 1045 (74.5) 106 (72.6) 939 (74.7) 0.58 Previous MI 620 (44.2) 83 (56.9) 537 (42.7) 0.002 Time since last MI Ͻ2 wk 256 (18.3) 44 (30.1) 212 (16.9) 2–13 wk 47 (3.4) 6 (4.1) 41 (3.3) Ͼ13 wk 266 (19.0) 31 (21.2) 235 (18.7) Never 834 (59.4) 65 (44.5) 769 (61.2) 0.0002 Previous thrombolysis 71 (5.1) 13 (8.9) 58 (4.6) 0.025 IABP placed preoperatively 38 (2.7) 11 (7.5) 27 (2.2) 0.001 Arrhythmia requiring therapy 148 (10.6) 19 (13.0) 129 (10.3) 0.31 Peripheral vascular disease 130 (9.3) 20 (13.7) 110 (8.8) 0.051 Prior PVD procedure 39 (2.8) 10 (6.9) 29 (2.3) 0.0049 Prior stroke 63 (4.5) 10 (6.9) 53 (4.2) 0.15 LVEF preoperative Ͻ40% 182 (13.0) 29 (19.9) 153 (12.2) 0.013 Medications—preoperative ACE inhibitor 648 (46.2) 66 (45.2) 582 (46.3) 0.86 Beta-blocker 1094 (78.0) 116 (79.5) 978 (77.8) 0.75 Caϩϩ antagonist 195 (13.9) 24 (16.4) 171 (13.6) 0.35 Aspirin 1072 (76.4) 113 (77.4) 959 (76.3) 0.84 HMG CoA reductase inhibitor 1082 (77.1) 107 (73.3) 975 (77.6) 0.24 Heparin (intravenous) 351 (25.0) 47 (32.2) 304 (24.2) 0.035 Platelet inhibitor (not aspirin) 304 (21.7) 32 (21.9) 272 (21.6) 0.94 Preoperative laboratory data Hemoglobin (g/dL) 13.7 (1.7) 13.5 (1.7) 13.7 (1.6) 0.12 Creatinine (mg/dL) 1.10 (0.33) 1.13 (0.32) 1.10 (0.34) 0.27 Platelet count (109/mL) 240 (72) 234 (70) 241 (72) 0.32 cTnI (ng/mL) 0.4 (2.5) 1.9 (7.4) 0.2 (0.7) 0.006 Intraoperative management Number of grafts 1 28 (2.0) 2 (1.4) 26 (2.1) 2 188 (13.4) 27 (18.5) 161 (12.8) 3 627 (44.8) 63 (43.2) 564 (44.9) Ն4 558 (39.8) 54 (37.0) 504 (40.2) 0.28 CPB duration (min) 99.2 (42.21) 119.63 (57.821) 96.78 (39.353) Ͻ0.0001 Aortic cross-clamp duration (min) 71.5 (35.03) 81.92 (45.125) 70.27 (33.471) 0.003 IABP placed intraoperatively 63 (4.5) 23 (15.8) 40 (3.2) Ͻ0.0001 Heparin administration (mg) BWH before February 2004 206 (83.7) 208 (76.0) 206 (85.0) 0.8319 BWH after February 2004 194 (61.0) 187 (54.7) 195 (61.5) 0.4608 THI 306 (92.0) 300 (80.6) 306 (93.4) 0.6998 Other surgical procedure Concurrent mitral valve 35 (2.5) 10 (6.9) 25 (2.0) 0.0020 Concurrent aortic valve 22 (1.6) 3 (2.1) 19 (1.5) 0.49 Concurrent other valve 2 (0.1) 2 (1.4) 0 0.01 Other cardiac surgery 108 (7.7) 24 (16.4) 84 (6.7) Ͻ0.0001 Other noncardiac surgery 17 (1.2) 1 (0.7) 16 (1.3) 1.0 (Continued)

864 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Continued Entire cohort Patients with MACE Patients without MACE P (1257 ؍ n) (146 ؍ n) (1403 ؍ n) Intraoperative inotropes Epinephrine 345 (24.6) 59 (40.4) 286 (22.8) Ͻ0.0001 Norepinephrine 169 (12.1) 20 (13.7) 149 (11.9) 0.52 Phenylephrine 668 (47.6) 63 (43.2) 605 (48.1) 0.25 Dopamine Ͼ5␮g ⅐ kgϪ1 ⅐ minϪ1 38 (2.7) 3 (2.1) 35 (2.8) 0.79 Dobutamine 18 (1.3) 1 (0.7) 17 (1.4) 1.0 Milrinone 29 (2.1) 6 (4.1) 23 (1.8) 0.11 Vasopressin 17 (1.2) 3 (2.1) 14 (1.1) 0.41 Postoperative inotropes Epinephrine 230 (16.4) 51 (34.9) 179 (14.2) Ͻ0.0001 Norepinephrine 143 (10.2) 31 (21.2) 112 (8.9) Ͻ0.0001 Phenylephrine 113 (8.1) 11 (7.5) 102 (8.1) 0.81 Dopamine Ͼ5␮g ⅐ kgϪ1 ⅐ minϪ1 30 (2.1) 8 (5.5) 22 (1.8) 0.009 Dobutamine 12 (0.9) 3 (2.1) 9 (0.7) 0.12 Milrinone 32 (2.3) 9 (6.2) 23 (1.8) 0.004 Vasopressin 91 (6.5) 26 (17.8) 65 (5.2) Ͻ0.0001 Intraoperative and postoperative transfusiona (median, 10–90 percentile) Red blood cell transfusion (units) 1 (0–5) 2 (0–8) 1 (0–5) Ͻ0.0001 Coagulation factor transfusion (units) 0 (0–2) 0 (0–4) 0 (0–2) 0.0055 Postoperative events HLOS (d) Ͼ12 d 130 (9.3) 43 (29.5) 87 (6.9) Ͻ0.0001 ICU LOS (d) Ͼ4 d 136 (9.7) 44 (30.1) 92 (7.3) Ͻ0.0001 Peak postoperative cTnI Ͼ12 ng/mL 108 (8.0) 108 (76.6) 0 Ͻ0.0001 Peak postoperative cTnI 4.06 (8.6) 21.9 (17.8) 2.0 (2.1) Ͻ0.0001 a All blood products administered during the hospital stay. Red blood cell transfusion includes both packed red blood cells and whole blood. Coagulation factor transfusion includes fresh frozen plasma, cryoprecipitate, and platelet transfusion. BWH ϭ Brigham and Women’s Hospital; THI ϭ Texas Heart Institute; MI ϭ myocardial infarction; cTnI ϭ cardiac troponin I; IABP ϭ intraaortic balloon pump; LVEF ϭ left ventricular ejection fraction; ACE ϭ angiotensin converting enzyme; CPB ϭ cardiopulmonary bypass; HLOS ϭ hospital length of stay; ICU LOS ϭ intensive care unit length of stay; PVD ϭ peripheral vascular disease.

Table 2. Postoperative Changes in Antithrombin Activity in Patients with Major Adverse Cardiac Events Results of Univariate Analysis at Each Time Point :(1257 ؍ and Without MACE (n (146 ؍ MACE) (n) Postoperative day Baseline Post-CPB 12345 Antithrombin activity (IU/mL) Patients with MACE 0.91 Ϯ 0.13 0.59* Ϯ 0.12 0.66* Ϯ 0.12 0.72* Ϯ 0.12 0.78* Ϯ 0.13 0.85* Ϯ 0.15 0.90† Ϯ 0.16 Patients without 0.92 Ϯ 0.13 0.62 Ϯ 0.11 0.71 Ϯ 0.13 0.77 Ϯ 0.12 0.83 Ϯ 0.12 0.90 Ϯ 0.13 0.95 Ϯ 0.14 MACE Change in antithrombin activity from baseline (IU/mL) Patients with MACE — Ϫ0.32 Ϯ 0.13 Ϫ0.24† Ϯ 0.13 Ϫ0.19† Ϯ 0.13 Ϫ0.13† Ϯ 0.14 Ϫ0.06* Ϯ 0.15 Ϫ0.00† Ϯ 0.16 Patients without — Ϫ0.30 Ϯ 0.12 Ϫ0.22 Ϯ 0.13 Ϫ0.16 Ϯ 0.13 Ϫ0.09 Ϯ 0.13 Ϫ0.02 Ϯ 0.13 0.03 Ϯ 0.13 MACE Percentage change in antithrombin activity from baseline (%) Patients with MACE — Ϫ34.7† Ϯ 12.8 Ϫ26.2† Ϯ 12.3 Ϫ20.4* Ϯ 12.7 Ϫ13.5† Ϯ 15.0 Ϫ5.9† Ϯ 16.4 0.7† Ϯ 17.9 Patients without — Ϫ32.0 Ϯ 11.2 Ϫ22.7 Ϯ 13.5 Ϫ16.1 Ϯ 13.3 Ϫ9.2 Ϯ 14.7 Ϫ1.1 Ϯ 14.5 4.6 Ϯ 15.5 MACE Data are reported as mean Ϯ standard deviation. CPB ϭ cardiopulmonary bypass. Signiﬁcance is reported between MACE groups at each time point: *P Յ 0.001 and †P Ͻ 0.05 by Student’s t-test.

Postoperative AT activity was significantly lower in patients variables that may possibly be indicative of recent heparin with MACE than those without MACE (Table 2). use at prior recent hospitalization, such as recent MI, were Decreased preoperative AT activity was independently also independently predictive. Decreased preoperative AT predicted by older age, male gender, and prior heparin use activity was also independently associated with lower within the same hospitalization (Table 3). Other clinical platelet count and increased partial thromboplastin time,

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 865 Antithrombin and Adverse Events

Table 3. Multivariable Predictors of Preoperative Antithrombin Activity Predictors of preoperative AT activity (U/mL) (0.168 ؍ 1194a; r2 ؍ n) Univariate Multivariate Standard error of Predictor estimate estimate multivariate estimate P Age (1 yr increment) Ϫ0.0020 Ϫ0.0022 0.0004 Ͻ0.0001 Gender (female) 0.0288 0.0177 0.0059 0.0028 Race (Caucasian) Ϫ0.0078 Ϫ0.0026 0.0054 0.624 Weight (1 kg increment) Ϫ0.0003 Ϫ0.0003 0.0002 0.233 Height (1 cm increment) Ϫ0.0007 Ϫ0.0031 0.0005 0.568 Institution 0.0183 0.0081 0.0053 0.129 Previous MI (Y) Ϫ0.0273 Ϫ0.0104 0.0036 0.004 Preoperative platelet count (10 ϫ 0.0023 0.0013 0.0005 0.006 109/mL increment) Preoperative PTT (1 s increment) Ϫ0.0014 Ϫ0.0008 0.0002 Ͻ0.0001 Preoperative heparin use Ϫ0.0863 Ϫ0.0352 0.0042 Ͻ0.0001 Preoperative diuretic use 0.0275 0.0147 0.0042 0.0005 a 209 subjects were missing one or more of the model’s predictor variables and are not included in this analysis. AT ϭ antithrombin; MI ϭ myocardial infarction; PTT ϭ partial thromboplastin time.

Table 4. Multivariable Predictors of Post-CPB Antithrombin (AT) Activity Predictors of post-CPB AT activity (U/mL) (0.504 ؍ 1205a; r2 ؍ n) Univariate Multivariate Standard error of Predictor estimate estimate multivariate estimate P Age (1 yr increment) Ϫ0.0030 Ϫ0.0006 0.0002 0.0110 Gender (female) 0.0356 0.0023 0.0039 0.5581 Race (Caucasian) Ϫ0.0068 Ϫ0.0022 0.0033 0.4979 Height (1 cm increment) 0.0023 0.0010 0.0003 0.0045 Weight (1 kg increment) 0.0015 0.0006 0.0001 Ͻ0.0001 Institution Ϫ0.0492 Ϫ0.0236 0.0038 Ͻ0.0001 Preoperative AT activity (U/mL) 0.4870 0.4698 0.0174 Ͻ0.0001 CPB duration (10 min increment) Ϫ0.006 Ϫ0.004 0.001 Ͻ0.0001 Lowest venous temperature during CPB 0.0030 0.0044 0.0011 Ͻ0.0001 (1°C increment) Intraoperative packed red blood cell Ϫ0.0095 Ϫ0.0085 0.0021 Ͻ0.0001 transfusion (1 unit increment) Intraoperative cryoprecipitate transfusion 0.0497 0.0491 0.0150 0.0011 (1 pooled unit increment) Post-CPB hemoglobin (1 g/dL increment) 0.0177 0.0184 0.0018 Ͻ0.0001 a 198 subjects were missing one or more of the model’s predictor variables and are not included in this analysis. CPB ϭ cardiopulmonary bypass. independent of recent heparin use, perhaps indicating a PODs 2 and 3 was independently predictive of MACE. MACE dose effect of heparin administration upon decreased AT was independently associated with change in AT activity activity. In the 1205 patients who had complete data for all from baseline on PODs 2–5 (Table 5). The receiver operat- variables in the model, decreased post-CPB AT activity was ing characteristic of the model was significantly improved independently predicted by lower preoperative AT activity by the addition of AT activity on PODs 2–4 to the model and clinical variables that indicate greater hemodilution, (Fig. 1). such as lower body weight and height and increased transfusion incidence or a prolonged procedure (Table 4). DISCUSSION A clinical model predicting MACE was developed This prospective, observational, cohort study confirms earlier (Table 5; adjusted r2 ϭ 0.156) for 1403 patients who had findings that preoperative AT activity is reduced in patients complete data for all variables in the clinical model. Vari- with recent exposure to heparin or recent MI.5,6 Furthermore, ables that have been associated with MACE in prior AT activity is reduced after CPB, likely due to consumption studies, notably recent MI, longer perfusion time, a require- by heparin administration6 and dilution. AT activity remained ment for intraaortic counterpulsation, and red blood cell significantly reduced from baseline levels over the postoperative transfusion were also associated with MACE in this study. period, recovering to baseline levels within 5 days, on average. AT activities at baseline, post-CPB, PODs 1–5, and the In this population of patients, MACE was associated change from baseline at these time points for each patient with factors that have been previously associated with MI were added to the clinical model, one time point at a time. or mortality including recent MI, peripheral vascular dis- Preoperative, post-CPB, and POD 1 AT activity were not ease, longer perfusion time, use of intraaortic counterpul- independently predictive of MACE, whereas AT activity on sation, and red blood cell transfusion.7–10 The clinical

866 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 5. Multivariable Predictors of Major Adverse Cardiac Events (MACE) Without AT activity information in model (0.156 ؍ r2 ;1403 ؍ n) 95% conﬁdence Predictor Odds ratio interval P Age Ͻ55 yr 1.00 — 55–64 yr 1.26 0.72–2.21 65–74 yr 1.26 0.70–2.28 75–84 tears 1.66 0.86–3.20 Ն85 yr 0.70 0.11–4.36 0.58 Gender (female) 1.13 0.72–1.78 0.60 Race (Caucasian) 0.59 0.36–0.95 0.03 Body mass index Ͻ20 0.67 0.15–3.04 20–24.9 0.52 0.28–0.96 25–34.9 0.48 0.30–0.77 Ն35 1.00 — 0.03 Institution 0.73 0.44–1.21 0.22 Myocardial infarction Ͻ2 wk prior 1.70 1.11–2.59 0.014 Prior peripheral vascular procedure 3.38 1.53–7.46 0.003 Perfusion time (10 min increment) 1.09 1.05–1.14 Ͻ0.0001 Concurrent other cardiac procedure 1.47 0.82–2.67 0.20 Preoperative or intraoperative IABP 4.02 2.18–7.42 Ͻ0.0001 RBC transfusion during hospital stay (per unit) 1.10 1.04–1.16 0.0004

Additional Predictive Value of AT Activity at Each Time Point Added to the Clinical Model

AT activity (0.1 IU/mL Odds 95% Conﬁdence P value of AT P value of improvement decrease) Ratio interval variable in overall model Preoperative AT activity 0.98 0.85–1.13 0.7820 0.7822 Postoperative AT activity 1.10 0.91–1.31 0.3193 0.3179 POD 1 AT activity 1.13 0.96–1.33 0.1565 0.1543 POD 2 AT activity 1.26 1.07–1.48 0.0056 0.0053 POD 3 AT activity 1.19 1.01–1.41 0.0412 0.0399 POD 4 AT activity 1.17 1.00–1.37 0.0506 0.0493 POD 5 AT activity 1.17 1.00–1.37 0.0553 0.0546 Change in AT activity Post-CPBa 1.15 0.93–1.41 0.1888 0.4058 Change in AT activity Day 1a 1.20 1.00–1.45 0.0558 0.1394 Change in AT activity Day 2a 1.36 1.13–1.62 0.0009 0.0035 Change in AT activity Day 3a 1.26 1.04–1.52 0.0163 0.0525 Change in AT activity Day 4a 1.26 1.06–1.51 0.0104 0.0294 Change in AT activity Day 5a 1.20 1.01–1.44 0.0407 0.1201 a Refers to change in AT activity from the baseline level. AT ϭ antithrombin; RBC ϭ red blood cell; POD ϭ postoperative day. model of MACE generated from this cohort had modest shown survival benefit, a meta-analysis of 20 trials encom- predictive value, similar to prior clinical models.9 The passing 3458 patients failed to show a survival benefit of addition of AT activity on POD 2 and beyond to the model administration of AT to critically ill patients.18 modestly improved model performance. However, the There are limited data regarding AT activity and ad- clinical value of AT activity as a predictor of MACE is verse outcomes after cardiac surgery. The majority of limited by the majority of adverse events that comprised studies describe use of AT for “heparin resistance” during MACE occurring before POD 2. Specifically, the majority of CPB and lack postoperative clinical outcome data.19–25 A MACE events were MI, which was typically manifested as single well-conducted observational study of 647 patients the peak cTnI level occurring on POD 1. Rather, it may be evaluated the association between preoperative and imme- that lower postoperative AT activity may be a consequence diate postoperative AT levels and outcomes in cardiac of more extensive surgery and other factors that are asso- surgery patients. Low levels of AT activity upon ICU ciated with increased incidence of MACE, although such arrival were associated with prolonged ICU stay, higher assertion cannot be proven in this observational cohort. We rate of reexploration for bleeding, thromboembolism, and cannot exclude the possibility that lower AT levels may adverse neurologic sequelae.5 Our study examined a longer have enhanced postoperative MACE. time period that encompassed the period of MACE occur- Increased risk of adverse cardiovascular events has been rence and the recovery of AT levels, thus providing addi- associated with lower levels of AT in other critically ill tional insights. Importantly, we replicated the finding that populations, such as patients with severe sepsis.11–13 These AT level after CPB is associated with clinical factors indica- observations have prompted clinical trials of supplemental tive of longer, more extensive procedures, perhaps with AT administration.14–17 Although several trials have more hemodilution.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 867 Antithrombin and Adverse Events

we believe there is no conflict of interest in the conduct of this study.

REFERENCES 1. Maclean PS, Tait RC. Hereditary and acquired antithrombin deficiency: epidemiology, pathogenesis and treatment options. Drugs 2007;67:1429–40 2. Levy JH, Despotis GJ, Szlam F, Olson P, Meeker D, Weisinger A. Recombinant human transgenic antithrombin in cardiac surgery: a dose-finding study. Anesthesiology 2002;96:1095–102 3. Sanson BJ, Simioni P, Tormene D, Moia M, Friederich PW, Huisman MV, Prandoni P, Bura A, Rejto L, Wells P, Mannucci PM, Girolami A, Buller HR, Prins MH. The incidence of venous thromboembolism in asymptomatic carriers of a deficiency of antithrombin, protein C, or protein S: a prospective cohort study. Blood 1999;94:3702–6 4. Simioni P, Sanson BJ, Prandoni P, Tormene D, Friederich PW, Girolami B, Gavasso S, Huisman MV, Buller HR, Wouter ten Cate J, Girolami A, Prins MH. Incidence of venous thromboembolism in families with inherited thrombophilia. Thromb Haemost 1999;81:198–202 5. Ranucci M, Frigiola A, Menicanti L, Ditta A, Boncilli A, Brozzi S. Postoperative antithrombin levels and outcome in cardiac operations. Crit Care Med 2005;33:355–60 6. Ranucci M, Ditta A, Boncilli A, Cotza M, Carboni G, Brozzi S, Bonifazi C, Tiezzi A. Determinants of antithrombin consump- Figure 1. Receiver operating characteristics of the addition of tion in cardiac operations requiring cardiopulmonary bypass. antithrombin (AT) activity on postoperative days (PODs) 2–4 to the Perfusion 2004;19:47–52 clinical model. 7. Bradshaw PJ, Jamrozik K, Le M, Gilfillan I, Thompson PL. Mortality and recurrent cardiac events after coronary artery bypass graft: long term outcomes in a population study. Heart 2002;88:488–94 Although our study contributes to our understanding of 8. Charlesworth DC, Likosky DS, Marrin CA, Maloney CT, the role AT plays in the perioperative period, we are Quinton HB, Morton JR, Leavitt BJ, Clough RA, O’Connor GT. concerned by the absence of a temporal correlation between Development and validation of a prediction model for strokes lower AT activity and MACE in our cohort, likely indicat- after coronary artery bypass grafting. Ann Thorac Surg 2003;76:436–43 ing that AT level is not an important determinant of 9. Koch CG, Weng YS, Zhou SX, Savino JS, Mathew JP, Hsu PH, thrombotic complications such as MI, stroke, and graft Saidman LJ, Mangano DT. Prevalence of risk factors, and not occlusion in the cardiac surgical setting. Thus, our obser- gender per se, determines short- and long-term survival after vational cohort cannot directly address any putative bio- coronary artery bypass surgery. J Cardiothorac Vasc Anesth logical mechanism for the role of AT in MACE generation. 2003;17:585–93 10. van Brussel BL, Plokker HW, Voors AA, Ernst JM, Ernst NM, Knaepen PJ, Koomen EM, Tijssen JG, Vermeulen FE. Multivar- CONCLUSION iate risk factor analysis of clinical outcome 15 years after In a large cohort of patients undergoing CABG surgery, venous coronary artery bypass graft surgery. Eur Heart J 1995; postoperative AT activity was independently associated with 16:1200–6 MACE. Because this occurs at time points predominantly after 11. Martinez MA, Pena JM, Fernandez A, Jimenez M, Juarez S, Madero R, Vazquez JJ. Time course and prognostic significance the MACE event, the clinical utility of AT as a biomarker of of hemostatic changes in sepsis: relation to tumor necrosis risk remains unknown. factor-alpha. Crit Care Med 1999;27:1303–8 12. Sakr Y, Reinhart K, Hagel S, Kientopf M, Brunkhorst F. DISCLOSURE Antithrombin levels, morbidity, and mortality in a surgical There are four potential conflicts of interest. Dr. Garvin was the intensive care unit. Anesth Analg 2007;105:715–23 recipient of a Research Fellowship funded by Talecris Biothera- 13. Wilson RF, Mammen EF, Tyburski JG, Warsow KM, Kubinec peutics, which allowed time for generation of this and other SM. Antithrombin levels related to infections and outcome. J Trauma 1996;40:384–7 articles. Dr. Chen is an employee of Talecris Biotherapeutics 14. Baudo F, Caimi TM, de Cataldo F, Ravizza A, Arlati S, Casella and performed the majority of the analyses. To prevent the G, Carugo D, Palareti G, Legnani C, Ridolfi L, Rossi R, appearance or substance of conflict, Simon Body personally D’Angelo A, Crippa L, Giudici D, Gallioli G, Wolfler A, Calori directed the conduct of all analyses and confirmed the conduct G. Antithrombin III (ATIII) replacement therapy in patients of the analyses by reviewing the code and output of all with sepsis and/or postsurgical complications: a controlled analyses. In addition, Simon Body personally reran the impor- double-blind, randomized, multicenter study. Intensive Care tant components of the analyses to confirm the findings and Med 1998;24:336–42 can attest that there was no potential for conflict in the analysis 15. du Cheyron D, Bouchet B, Bruel C, Daubin C, Ramakers M, Charbonneau P. Antithrombin supplementation for anticoagu- phase. Dr. Body received a total of Ͻ$5000 to allow his time for lation during continuous hemofiltration in critically ill patients travel to Talecris in North Carolina to coordinate the analyses. with septic shock: a case-control study. Crit Care 2006;10:R45 Talecris also paid for the costs of antithrombin analyses 16. Eid A, Wiedermann CJ, Kinasewitz GT. Early administration of performed by Charles River Laboratories in Ontario, Canada, high-dose antithrombin in severe sepsis: single center results but had no opportunity to intervene in these analyses. In brief, from the KyberSept-trial. Anesth Analg 2008;107:1633–8

868 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 17. Kienast J, Juers M, Wiedermann CJ, Hoffmann JN, Ostermann 21. Despotis GJ, Levine V, Joist JH, Joiner-Maier D, Spitznagel E. H, Strauss R, Keinecke HO, Warren BL, Opal SM. Treatment Antithrombin III during cardiac surgery: effect on response of effects of high-dose antithrombin without concomitant heparin activated clotting time to heparin and relationship to markers in patients with severe sepsis with or without disseminated of hemostatic activation. Anesth Analg 1997;85:498–506 intravascular coagulation. J Thromb Haemost 2006;4:90–7 22. Kanbak M. The treatment of heparin resistance with anti- 18. Afshari A, Wetterslev J, Brok J, Moller A. Antithrombin III in thrombin III in cardiac surgery. Can J Anaesth 1999;46:581–5 critically ill patients: systematic review with meta-analysis and 23. Lemmer JH Jr, Despotis GJ. Antithrombin III concentrate to trial sequential analysis. BMJ 2007;335:1248–51 treat heparin resistance in patients undergoing cardiac sur- 19. Avidan MS, Levy JH, van Aken H, Feneck RO, Latimer RD, Ott gery. J Thorac Cardiovasc Surg 2002;123:213–7 E, Martin E, Birnbaum DE, Bonfiglio LJ, Kajdasz DK, Despotis 24. Levy JH, Montes F, Szlam F, Hillyer CD. The in vitro effects of GJ. Recombinant human antithrombin III restores heparin antithrombin III on the activated coagulation time in patients responsiveness and decreases activation of coagulation in on heparin therapy. Anesth Analg 2000;90:1076–9 heparin-resistant patients during cardiopulmonary bypass. 25. Williams MR, D’Ambra AB, Beck JR, Spanier TB, Morales DL, J Thorac Cardiovasc Surg 2005;130:107–13 Helman DN, Oz MC. A randomized trial of antithrombin 20. Conley JC, Plunkett PF. Antithrombin III in cardiac surgery: an concentrate for treatment of heparin resistance. Ann Thorac outcome study. J Extra Corpor Technol 1998;30:178–83 Surg 2000;70:873–7

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 869 ␤ 2-Adrenergic Receptor-Coupled Phosphoinositide 3-Kinase Constrains cAMP-Dependent Increases in Cardiac Inotropy Through Phosphodiesterase 4 Activation

Christopher J. Gregg,* Jochen Steppan, MD,* Daniel R. Gonzalez, PhD,† Hunter C. Champion, MD, PhD,‡ Alexander C. Phan,* Daniel Nyhan, MD,* Artin A. Shoukas, PhD,* Joshua M. Hare, MD,† Lili A. Barouch, MD,* and Dan E. Berkowitz, MD*

BACKGROUND: Emerging evidence suggests that phosphoinositide 3-kinase (PI3K) may modulate cardiac inotropy; however, the underlying mechanism remains elusive. We hypothesized that ␤ 2-adrenergic receptor (AR)-coupled PI3K constrains increases in cardiac inotropy through cyclic adenosine monophosphate (cAMP)-dependent phosphodiesterase (PDE) activation. METHODS: We tested the effects of PI3K and PDE4 inhibition on myocardial contractility by using isolated murine cardiac myocytes to study physiologic functions (sarcomere shortening [SS] and intracellular Caϩ transients), as well as cAMP and PDE activity. RESULTS: PI3K inhibition with the reversible inhibitor LY294002 (LY) resulted in a signiﬁcant increase in SS and Ca2ϩ handling, indicating enhanced contractility. This response depended on

Gi␣ protein activity, because incubation with pertussis toxin (an irreversible Gi␣ inhibitor) abolished the LY-induced hypercontractility. In addition, PI3K inhibition had no greater effect on SS than both a PDE3,4 inhibitor (milrinone) and LY combined. Furthermore, LY decreased PDE4 activity in a concentration-dependent manner (58.0% of PDE4 activity at LY concentrations of 10 ␮ ␤ M). Notably, PI3K␥ coimmunoprecipitated with PDE4D. The 2-AR inverse agonist, ICI 118,551 (ICI), abolished induced increases in contractility. CONCLUSIONS: PI3K modulates myocardial contractility by a cAMP-dependent mechanism through the regulation of the catalytic activity of PDE4. Furthermore, basal agonist-independent ␤ activity of the 2-AR and its resultant cAMP production and enhancement of the catalytic activity of PDE4 through PI3K represents an example of integrative cellular signaling, which controls cAMP dynamics and thereby contractility in the cardiac myocyte. These results help to explain the mechanism by which milrinone is able to increase myocardial contractility in the absence of direct ␤-adrenergic stimulation and why it can further augment contractility in the presence of maximal ␤-adrenergic stimulation. (Anesth Analg 2010;111:870–7)

␤ ␤ he -adrenergic receptor ( -AR) isoforms character- complex, as this receptor is coupled to both Gs and Gi, the ized in the heart have distinct but overlapping signal latter of which is recognized to attenuate inotropy through Ttransduction mechanisms that regulate many aspects the cGMP dependent effects of nitric oxide produced from ␤ of myocardial pump function. Augmentation of cardiac endothelial nitric oxide synthase 3. 2-AR signaling is likely inotropy through ␤-ARs is mediated through a well- compartmentalized in light of the dual and seemingly studied system that increases intracellular cyclic adenosine opposing effects of these signaling cascades.2 Those cas- monophosphate (cAMP) leading to increases in cAMP- cades are critically dependent on cAMP concentrations and 1 ␤ 3 dependent protein kinase A activity. The 1-AR is coupled spatial localization. Thus, changes in the balance between to the heterotrimeric Gs protein and is considered the cAMP production through adenylyl cyclase activation and primary means for catecholamine-induced increases in cAMP breakdown through cAMP-dependent phosphodies- cAMP concentrations and thus myocardial contractility. terases (PDE) will modulate protein kinase A activity and ␤ The role of the 2-AR in modulation of contractility is more myocardial contractility. The roles of phosphoinositide 3-kinase (PI3K), a fam- From the *Johns Hopkins Medical Institutions (current affiliation: University ily of lipid kinases whose downstream targets include of California, San Diego), †Johns Hopkins Medical Institutions (current affiliation: University of Miami), ‡Johns Hopkins Medical Institutions, bioactive lipids and proteins, in signaling has emerged Baltimore, Maryland (current affiliation: University of Pittsburgh). over the past few years.4 Specifically, it has been dem- Accepted for publication June 1, 2010. onstrated that the ␤-AR kinase–1 (␤-ARK1) and PI3K This work was supported in part by a grant from the National Space interact and that ␤-ARK1 recruitment of PI3K is crucial Biomedical Research Institute (CA00405) through National Aeronautics & ␤ Space Administration (AS) and a National Institutes of Health grant (R01 for mediating 2-AR internalization as a component of AG 021,523) (DEB). the receptor/internalization scaffold complex.1 In addi- Disclosure: The authors report no conflict of interest. tion to its role in receptor internalization, the direct role Address correspondence to Dan E. Berkowitz, MD, Johns Hopkins Medical of PI3K in modulating myocardial contractility has been Institutions, 600 N Wolfe Street, CCM Tower 711, Baltimore, MD 21287. Address e-mail to [email protected]. investigated in an elegant study observing the role of Copyright © 2010 International Anesthesia Research Society PI3K in progression of myocardial hypertrophy. Specifi- DOI: 10.1213/ANE.0b013e3181ee8312 cally, mice deficient in the catalytic subunit of the PI3K␥

870 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Figure 1. Phosphoinositide 3-kinase constrains contractility in cardiac myocytes. (A,B) Effects of the reversible phosphoinositide 3-kinase ␮ ϭ Ͻ 2ϩ ϭ Ͻ (PI3K) inhibitor LY (10 M) on sarcomere shortening (SS) (A, n 14, *** P 0.001 vs baseline) and systolic [Ca ]i (B, n 12, *** P ϭ 2ϩ ϭ 0.001 vs baseline). Washout confirms the reversibility for both SS (n 9, n.s. vs baseline) and [Ca ]i (n 8, n.s. vs baseline). Sample transients are shown above the graphic panels. (C) LY (0.1 to 100 ␮M) dose response showing a concentration-dependent increase in SS with ␮ ϭ Ͻ ␶ an EC50 of 2.2 M. (D) Effects of wortmannin (5 nM) on SS (n 4, ** P 0.01 vs baseline). (E) Monoexponential time constant, (index of lusitropy), fit to diastolic portion of averaged Ca2ϩ fluorescence transients (n ϭ 15, * P Ͻ 0.05 vs baseline fit) and washout. isoform (p110␥Ϫ/Ϫ) demonstrate increased basal myo- committee of the Johns Hopkins University School of cardial contractility in a cAMP-dependent manner, and Medicine, as well as with the National Institutes of Health inhibition of PI3K augments isoproterenol-induced con- and American Physiological Society guidelines. Three tractile responses.5 The cellular mechanism of this effect, 5-month-old C57/BL6J mice were purchased from Jackson however, remains incompletely understood. Laboratories (Bar Harbor, ME). Thus, we hypothesized that PI3K is a negative upstream regulator of myocardial contractility by modulating cAMP- dependent PDE activity. Cardiac Myocyte Isolation Myocytes were isolated by enzymatic digestion with colla- METHODS genase type 2 (1 g/L; Worthington Biochemical, Lakewood, Animals NJ) and protease type XIV (0.1 g/L), as previously de- Animal treatment and care was approved and provided in scribed.6,7 Cell suspension was obtained by mechanically accordance with the institutional animal care and use disrupting digested ventricles, filtering, centrifugation, and

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 871 PI3K Constrains Cardiac Contractility

ϩ resuspension in Tyrode solution (1.0 mM Ca2 ). The myocyte cell suspension was incubated with 5 ␮mol/L fura- 2/am (Molecular Probes, Eugene, OR), then transferred to an inverted microscope (TE 200; Nikon), continuously superfused with Tyrode solution and stimulated at 1 Hz. Change in average sarcomere length was determined by fast Fourier transform of the Z-line density trace to the frequency domain. Calcium concentrations were measured using a dual-excitation spectrofluorometer (IonOptix, Mil- ton, MA), as previously described.6,7 The experiments were conducted in the presence or absence of the reversible PI3K inhibitor LY294002 (LY, 10 ␮M), the irreversible PI3K inhibitor wortmanin (5 nM), the Gi␣ protein inhibitor ␮ ␤ ␤ pertussis toxin (PTX, 1.5 g/mL), the 1-and 2- adrenoreceptor agonist isoproterenol (50 nM), the PDE3,4 inhibitor milrinone (0.3, 3, 30, 300 ␮M), the PDE4 inhibitor rolipram (1, 10, 100 ␮M), and ICI 118,551 (ICI, 100 nM). Phosphodiesterase Activity Analysis Total low Km cAMP-dependent PDE activity was assayed by fluorescence polarization (Molecular Devices, Sunny- Figure 2. Enhancement of myocardial contractility by PI3K inhibition vale, CA) and read on a microplate reader (Perkin Elmer, is sensitive to Gi activity. Effects of the irreversible Gi␣ inhibitor Wellesley, MA), as described by the manufacturer in the pertussis toxin (PTX) on basal and phosphoinositide-3 kinase (PI3K) presence or absence of LY (1, 3, 10 ␮mol/L), isoproterenol inhibited myocardial contractility. LY significantly increased contrac- (100 nmol/L), milrinone (30 ␮mol/L), and ICI (100 tility in non-PTX treated myocytes (n ϭ 6, ** P Ͻ 0.01 vs baseline). In PTX-treated myocytes, LY response was significantly blunted, nmol/L). compared with LY only group (n ϭ 6, ** P Ͻ 0.01, LY vs LY ϩ PTX) and was not significantly different than baseline (n ϭ 6, P Ͼ 0.05 vs Cyclic Nucleotide Assay baseline). Myocytes were used in a cAMP enzyme immunoassay (Amersham Pharmacia Biotech, Piscataway, NJ), as described by the manufacturer, in the presence or absence of Ͻ isoproterenol (50 nM), LY (10 ␮M), milrinone (Mil) (30 ␮M), value 0.05 after t tests for pairs or nonparametric ranking and ICI (100 nM). and one-way ANOVA. Coimmunoprecipitation PI3K (p110␥) or control (bovine serum albumin) antibodies RESULTS (both: Cell Signaling, Danvers, MA) were cross-linked to PI3K Constrains Contractility in Isolated protein A/G beads (Pierce Biotechnology, Rockford, IL) Cardiac Myocytes and used to immunoprecipitate proteins from mouse heart Figure 1 illustrates that the dose-dependent inhibition of cell lysates. Western blots were then performed with a PI3K results in an increased sarcomere shortening (SS) and 2ϩ PDE4D antibody. systolic [Ca ]i. In detail, inhibition of PI3K with the specific reversible inhibitor LY (10 ␮M) resulted in a 2.21 Ϯ Western Blots 0.12-fold increase in SS (Fig. 1A) and a 1.68 Ϯ 0.10-fold 2ϩ The entire coimmunoprecipitation was denatured, re- increase in systolic [Ca ]i (Fig. 1B). This increase in duced (150 mmol/L dithiothreitol), and separated by contractility was reversible and concentration dependent Tris-Glycine/sodium dodecyl sulfate-polyacrylamide gel (Fig. 1C). Similar results were observed with the irrevers- electrophoresis (Invitrogen, Carlsbad, CA). The proteins ible PI3K inhibitor wortmannin, which produced a 2.31 Ϯ were transferred to polyvinylidene fluoride membranes 0.04-fold increase in SS (Fig. 1D). Furthermore, the time ϩ (Billerica, MA) with Towbin transfer buffer, and the constant of Ca2 fluorescence decay decreased from 0.16 Ϯ membranes were blocked with 5% nonfat dry milk/tris 0.01 to 0.11 Ϯ 0.01 s upon PI3K inhibition (Fig. 1E). buffered saline (NFDM/TBS). Primary antibodies (PI3- kinase, Cell Signaling; PDE4D, Fabgennix, Frisco, TX) PI3K Is Regulated Through G ␣ Signaling were diluted in 5% NFDM/TBS. The immunoblots were i Figure 2 shows the relation between PI3K and Gi␣ signal- washed and incubated with the appropriate secondary ing. The enhanced contractility mediated by PI3K inhibi- antibodies (Santa Cruz Biotechnology, Santa Cruz CA). tion was abolished by preincubation with the Gi␣ inhibitor, The membranes were washed again and developed with PTX. LY increased SS 2.52 Ϯ 0.75-fold, while in the presence SuperSignal Pico and Femto substrate (Pierce Biotechnol- of PTX, SS did not change significantly (Fig. 2). ogy, Rockford, IL). Data Analysis and Statistical Procedures PI3K Inhibition Enhances ␤-AR-Stimulated Data are presented as mean Ϯ SEM, with the exception of Contractility in Myocytes physiologic data that are presented as change from base- Figure 3 demonstrates the additive effect of PI3K inhibition line Ϯ SEM. Statistical significance was considered for P and ␤-AR stimulation. The combination of PI3K inhibition

872 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 3. Synergistic effect of isoproterenol and LY on myocyte contractility. Cumulative data showing signiﬁcant increases in sarcomere shortening (SS) (A) with adrenergic stimulation alone (isoproterenol: n ϭ 9, * P Ͻ 0.05 vs baseline) and the addition of phosphoinositide-3 kinase (PI3K) inhibition (isoproterenol ϩ LY: n ϭ 9, ** P Ͻ 0.01 vs baseline, # P Ͻ 0.01 isoproterenol versus isoproterenol ϩ LY). Systolic 2ϩ ϭ Ͻ [Ca ]i (B) also showed signiﬁcant enhancement with adrenergic stimulation alone (isoproterenol: n 9, * P 0.05 vs baseline) and the addition of PI3K inhibition (isoproterenol ϩ LY: n ϭ 9, ** P Ͻ 0.01 vs baseline, # P Ͻ 0.01 isoproterenol versus isoproterenol ϩ LY). Sample transients from these experiments are shown above the graphic panels.

decreased PDE4 activity in a concentration-dependent manner, which paralleled the results from the PDE4 specific inhibitor CB524717 (Fig. 5A). The PDE3,4 inhibitor Mil (30 ␮M) and LY increased cAMP levels from 4.70 Ϯ 0.77 to 14.59 Ϯ 2.12 and 19.85 Ϯ 3.42 fmol/␮g protein, respectively (Fig. 5B). There was no difference when incubated with LY and Mil alone or a combination of both (18.28 Ϯ 4.31 fmol/␮g protein). Mil and the PDE4 specific inhibitor rolipram resulted in a concentration-dependent increase in SS and systolic 2ϩ ␮ [Ca ]i (Fig. 6A, B). At EC50 concentrations, Mil (10 M) Ϯ Figure 4. PI3K inhibition increases myocyte cAMP production and produced a 1.35 0.08-fold increase in SS, and LY pro- augments adrenergically stimulated cAMP. Effect of isoproterenol duced a 2.27 Ϯ 0.25-fold increase. The combination of Mil (50 nM), LY (10 ␮M) and their combination on myocyte cAMP and LY yielded no synergistic effect (2.25 Ϯ 0.18-fold ␮ ϭ Ͻ levels (fmol/ g protein, n 6, * P 0.05). Both LY and increase), while the addition of isoproterenol (100 nM) isoproterenol increased cAMP levels independently, however, the Ϯ combination of isoproterenol and LY increased cAMP levels above yielded an additional enhancement in contractility (3.41 levels seen with isoproterenol and LY alone, (n ϭ 8, *** P Ͻ 0.32-fold change in SS, Fig. 6C). 0.001, baseline versus isoproterenol ϩ LY, # P Ͻ 0.01 isoproterenol versus isoproterenol ϩ LY). ␤ Agonist Independent 2-AR Coupling Through PI3K Maintains Contractile Tone with LY and ␤-AR stimulation with isoproterenol aug- ϩ The results depicted in Figure 7 illustrate that decreasing mented SS 1.97 Ϯ 0.27-fold and systolic [Ca2 ] 1.50 Ϯ i basal agonist-independent cAMP production (with ICI), the 0.13-fold over isoproterenol alone (Fig. 3). catalytic activity of PDE isoforms (regulated by PI3K) cease PI3K Inhibition Increases Myocyte cAMP Levels to have an effect on cardiac contractility. Myocytes incu- ␤ As shown in Figure 4, increasing cAMP levels are linked to bated with the 2-AR inverse agonist ICI demonstrated a PI3K inhibition and ␤-AR stimulation. Isoproterenol in- time and concentration-dependent decrease in SS (data not creased cAMP levels from 5.20 Ϯ 1.20 to 18.48 Ϯ 2.21 shown) and exhibited a decrease in SS from baseline (Fig. fmol/␮g protein, and LY increased cAMP levels to 24.32 Ϯ 7A). Furthermore, ICI completely reversed LY and Mil- 2.73 fmol/␮g protein. The combination of LY and isopro- induced contractile responses, cAMP levels, and systolic 2ϩ terenol had a synergistic effect and augmented cAMP levels [Ca ]i (Fig. 7, B–D). further to 37.44 Ϯ 3.78 fmol/␮g protein (Fig. 4). Interaction of PDE4 and PI3K␥ PI3K Is Coupled to cAMP-Dependent PDE4D was detected in HeLA and U937 cells (control) and PDE Activity in homogenates immunoprecipitated with PI3K␥ (p110␥) Figure 5 and 6 demonstrate the mechanism by which PI3K but not in homogenates immunoprecipitated with bovine regulates contractility in a cAMP-dependent manner. LY serum albumin. As expected, PI3K␥ (p110␥) was detected

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 873 PI3K Constrains Cardiac Contractility

Figure 5. PI3K inhibition decreases the activity of cAMP-dependent PDE4 to increase myocardial cAMP levels. Concentration-dependent depression of LY (A) and CB524717 (panel inlay) on phosphodiesterase (PDE) 4 activity (** P Ͻ 0.01, n ϭ 5 vs baseline activity). (B) Effect of LY (10␮61,517), milrinone (30 ␮M) alone, and in combination on myocyte cAMP levels (fmol/␮g protein). LY and milrinone increased myocyte cAMP levels above baseline (* P Ͻ 0.05, n ϭ 4 vs baseline). There is no signiﬁcant increase in cAMP levels if the drugs are combined (n ϭ 4, p ϭ n.s. LY versus milrinone, milrinone versus LY ϩ milrinone, LY versus LY ϩ milrinone).

Figure 6. PI3K regulates myocardial contractility by cAMP-dependent PDE activation. Concentration-dependent enhancement of myocyte sarcomere shortening (SS) by milrinone (A, n ϭ 4) and rolipram (B, n ϭ 10). (C) Effect of milrinone (30 ␮M), LY (5 ␮M), and isoproterenol (10 2ϩ nM) alone and in combination on myocyte SS and systolic [Ca ]i. Milrinone and LY alone produced signiﬁcant increases in SS and systolic 2ϩ ϭ ϭ [Ca ]i. However, administration of a combination of both LY and milrinone produced no added effect over LY alone (n 13, p n.s. LY versus LY, milrinone). Addition of isoproterenol to milrinone and LY produced a further increase in SS (n ϭ 13, ** P Ͻ 0.01, milrinone ϩ LY ϩ isoproterenol versus milrinone ϩ LY). in lysates immunoprecipitated with the same antibodies inhibition decreases PDE4 activity and increases cAMP (Fig. 8). levels. Finally, this mechanism is constitutively activated ␤ given that the inverse 2-AR agonist ICI completely re- DISCUSSION verses both PI3K and PDE-mediated increases in contrac- In the present study, we have demonstrated that PI3K tility and cAMP levels (Fig. 9). negatively modulates myocardial contractility by a Gi␣- We demonstrated that inhibition of PI3K enhances ad- dependent mechanism. Furthermore, PI3K negatively regu- renergic stimulation of cardiac myocytes, which is consis- lates adrenergic stimulation of myocardial contractility tent with the findings of Jo et al.8 and Crackower et al.5 since inhibition significantly enhances ␤-AR–mediated con- However, it does differ significantly from Jo et al.8 with ␤ tractility. 2-AR–coupled PI3K decreases cAMP by a regard to the effect of LY on basal contractility. While they mechanism that involves activation of PDE4, as PI3K demonstrated no effect of PI3K inhibition with LY on basal

874 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA ␤ Figure 7. Agonist independent coupling of the 2-AR to cAMP production and the role of PI3K. (A) Depression of myocyte sarcomere shortening ␤ ϭ Ͻ (SS) by the 2-AR inverse agonist ICI 118,551 (100 nM) (n 6, * P 0.05 vs baseline). (B) ICI abolishes LY and milrinone-induced increases ϭ Ͻ ϭ ϩ 2ϩ ϭ in SS (n 10, ** P 0.01 LY or milrinone versus baseline; p n.s. LY or milrinone ICI versus baseline) and systolic [Ca ]i (D, n 11, * P Ͻ 0.05 LY versus baseline; ** P Ͻ 0.01 milrinone versus baseline; p ϭ n.s. LY or milrinone ϩ ICI versus baseline). (C) Both LY (n ϭ 6, *** P Ͻ 0.001 vs baseline) and milrinone (n ϭ 6, ** P Ͻ 0.01 vs baseline) increased cAMP levels that were completely abolished by ICI coincubation. Sample transients are shown above their respective graphic panels.

Figure 8. Molecular interaction between PI3K and PDE4D. Phospho- diesterase (PDE) 4D was detected in lysates from HeLa and U937 cells (control). In mouse heart homogenates immunoprecipitated with phosphoinositide 3-kinase (PI3K) (p110␥), PDE4D was detected but not in homogenates immunoprecipitated with A/G-linked bovine serum albumin Ab. Figure 9. Schematic of the proposed PI3K interaction with the ␤-AR. ␤ ϭ ϭ ϭ 2-AR beta-2 adrenoreceptor; AC adenylate cyclase; Gs ϭ ϭ G-protein S; Gi G-protein i; PI3K phosphoinositide 3-kinase; 8 PDE ϭ phosphodiesterase; cAMP ϭ cyclic adenosine monophos- cellular contractility, we demonstrated a concentration- ϭ 2ϩ phate; PKA protein kinase A. dependent increase in both SS and systolic [Ca ]i. This is consistent with the findings of Crackower et al.5 who demonstrated significantly enhanced basal contractility in were performed in the presence of the nonspecific PDE isolated myocytes from mice in which the catalytic subunit inhibitor isobutylmethylxanthine. This difference in the of the ␥ isoform of PI3K (p110␥) has been knocked out. We assays performed could explain their different conclusion, agree with Crackower et al.5 that myocyte cAMP levels are while indirectly supporting our hypothesis that inhibition modulated by the activity of PI3K. This is in contrast to the of PI3K enhances contractility through a pathway involving 8 ␤ conclusion made by Jo et al. that the 2-AR–mediated PDE inhibition. enhancement in contractility is not mediated by an increase We show that PI3K is coupled to cAMP-dependent in cAMP, because they demonstrate no difference in myo- PDE4 activity, which is in contrast to Patrucco et al.9 who ␤ cyte cAMP levels in response to a 2-AR selective agonist in demonstrated that mice expressing a PI3Kg kinase dead KD the presence or absence of LY. However, these experiments mutant (PI3Kg ) have no overt alteration in myocardial

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 875 PI3K Constrains Cardiac Contractility contractility. They interpreted their results to indicate that of cAMP production and cAMP breakdown in the cardiac the kinase domain of the PI3K was not necessary for myocyte and sheds a light on Milrinone’s mechanism of modulating PDE and constraining cAMP-dependent alter- action (Fig. 9). In another way, cAMP is a convergence point ations in myocardial contractility. Therefore, Patrucco et of two opposing signaling pathways. More specifically, cAMP al.9 suggested that protein-protein interaction between levels are the integrated output of two temporally and spa-

PI3Kg and the PDE3B isoform is sufficient for PDE3B tially sensitive signaling inputs that serve to regulate myocar- activation. Our data, on the other hand, suggest that the dial inotropic responses. This signaling convergence of cAMP PI3K catalytic domain may indeed be important in activa- production and breakdown at cAMP levels indicates tighter tion of a downstream PDE, in this case PDE4D. LY294002, regulation of cAMP availability than has been previously a specific reversible inhibitor of PI3K, does so by competing identified. This could potentially prevent a destructive, run- with adenosine triphosphate for the active site of the away cAMP event on the second timescale or could control ␥ 10 ␤ catalytic subunit for PI3K␥, p110 . Our results with the gain of the 2-AR over the lifetime of the organism by LY294002 and wortmannin support the notion that kinase relatively regulating transcription of both Gs and Gi proteins. activity is critically important for the modulation of cAMP- Furthermore, by indirectly inhibiting the breakdown of dependent inotropic changes. We cannot refute the idea cAMP, Milrinone is able to increase myocardial contrac- that LY294002, in binding and inhibiting the kinase activity tility in the absence of direct ␤-adrenergic stimulation and of PI3K, may, in addition, alter the binding of a PDE to further augments contractility even in the presence of p110␥. An alternative explanation is that PI3K may act maximal ␤-adrenergic stimulation, as seen clinically. Thus, through an intermediate to result in regulation of a PDE. an understanding of the importance of this physiologic This is hypothetically supported by the observation that pathway in the regulation of myocyte contractility is im- PDE has regulatory phosphorylation sites for Akt, the portant in defining the molecular mechanism underlying kinase downstream of PI3K.11,12 Thus, PI3K may not only Milrinone’s action as a critical inotrope in our limited act as a scaffold for PDE9 but may also be phosphorylated armamentarium. by the enzyme or its downstream effector Akt. PDE3 and PDE4 are the major isoforms that degrade REFERENCES cAMP in cardiac myocytes.13 Recent data suggest that these 1. Rockman HA, Koch WJ, Lefkowitz RJ. Seven-transmembrane- isoforms localize to distinct compartments in the cell and spanning receptors and heart function. Nature 2002;415:206–12 3,13–15 2. Devic E, Xiang Y, Gould D, Kobilka B. Beta-adrenergic receptor have different functional roles. Specifically, PDE4 subtype-specific signaling in cardiac myocytes from beta(1) appears to be functionally coupled to the pool of cAMP and beta(2) adrenoceptor knockout mice. Mol Pharmacol stimulated by ␤-ARs as a result of spatial confinement by a 2001;60:577–83 localization signal, while PDE3 appears to regulate the 3. Baillie GS, Houslay MD. Arrestin times for compartmentalised general pool of cAMP stimulated by forskolin.13 This is cAMP signalling and phosphodiesterase-4 enzymes. Curr 16 Opin Cell Biol 2005;17:129–34 consistent with the observations of Xiang et al. who 4. Oudit GY, Sun H, Kerfant BG, Crackower MA, Penninger JM, demonstrated that PDE4 rather than PDE3 is required for Backx PH. The role of phosphoinositide-3 kinase and PTEN in ␤ 2-AR signaling in neonatal mouse myocytes. They demon- cardiovascular physiology and disease. J Mol Cell Cardiol strated an enhanced response to isoproterenol in the presence 2004;37:449–71 5. Crackower MA, Oudit GY, Kozieradzki I, Sarao R, Sun H, Sasaki of selective PDE4 inhibitors and not in cells treated with a T, Hirsch E, Suzuki A, Shioi T, Irie-Sasaki J, Sah R, Cheng HY, selective PDE3 inhibitor. Furthermore, PDE4 knockout mice Rybin VO, Lembo G, Fratta L, Oliveira-dos-Santos AJ, Benovic JL, have augmented responses to isoproterenol, while myocytes Kahn CR, Izumo S, Steinberg SF, Wymann MP, Backx PH, ␤ Penninger JM. Regulation of myocardial contractility and cell size from 2-AR knockout mice have no augmented response to PDE4 inhibition.16 These findings are consistent with our by distinct PI3K-PTEN signaling pathways. Cell 2002;110:737–49 ␤ 6. Barouch LA, Harrison RW, Skaf MW, Rosas GO, Cappola TP, findings and support the idea that PDE4 is essential for 2-AR Kobeissi ZA, Hobai IA, Lemmon CA, Burnett AL, O’Rourke B, signaling. Moreover, we have extended the observation that Rodriguez ER, Huang PL, Lima JA, Berkowitz DE, Hare JM. this phenomenon is dependent on PI3K. Nitric oxide regulates the heart by spatial confinement of nitric There are several shortcomings of the present study that oxide synthase isoforms. Nature 2002;416:337–9 7. Khan SA, Lee K, Minhas KM, Gonzalez DR, Raju SV, Tejani need to be considered. First, all physiologic experiments AD, Li D, Berkowitz DE, Hare JM. Neuronal nitric oxide were done in a rodent model of isolated myocyte and not synthase negatively regulates xanthine oxidoreductase inhibi- verified in vivo or in human tissues. There are considerable tion of cardiac excitation-contraction coupling. Proc Natl Acad differences between species, and it is not possible to SciUSA2004;101:15944–8 generalize all results and translate them to a different 8. Jo SH, Leblais V, Wang PH, Crow MT, Xiao RP. Phosphatidyl- inositol 3-kinase functionally compartmentalizes the concur- species. Also, there could be a direct biochemical effect of rent G(s) signaling during beta2-adrenergic stimulation. Circ PI3K inhibitors on PDE activity. None, however, have yet Res 2002;91:46–53 been described in the literature. Moreover, even though we 9. Patrucco E, Notte A, Barberis L, Selvetella G, Maffei A, showed that PI3K acts through PDE4, the cAMP-dependent Brancaccio M, Marengo S, Russo G, Azzolino O, Rybalkin SD, Silengo L, Altruda F, Wetzker R, Wymann MP, Lembo G, PDE isoform remains controversial. Hirsch E. PI3Kgamma modulates the cardiac response to Milrinone, a well-known PDE 3,4 inhibitor, is frequently chronic pressure overload by distinct kinase-dependent and used to treat advanced heart failure, especially in the context -independent effects. Cell 2004;118:375–87 of cardiac surgery. Our novel finding that inhibition of PI3K 10. Walker EH, Pacold ME, Perisic O, Stephens L, Hawkins PT, Wymann MP, Williams RL. Structural determinants of phos- through Gi␣ enhances myocardial contractility by a PDE4- ␤ phoinositide 3-kinase inhibition by wortmannin, LY294002, dependent mechanism that is independent of direct -AR quercetin, myricetin, and staurosporine. Mol Cell 2000;6: agonist activation further demonstrates the exquisite balance 909–19

876 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 11. Kenan Y, Murata T, Shakur Y, Degerman E, Manganiello 14. Baillie GS, Sood A, McPhee I, Gall I, Perry SJ, Lefkowitz RJ, VC. Functions of the N-terminal region of cyclic nucleotide Houslay MD. beta-Arrestin-mediated PDE4 cAMP phosphodi- phosphodiesterase 3 (PDE 3) isoforms. J Biol Chem 2000; esterase recruitment regulates beta-adrenoceptor switching 275:12331–8 from Gs to Gi. Proc Natl Acad SciUSA2003;100:940–5 12. Wijkander J, Landstrom TR, Manganiello V, Belfrage P, Deger- 15. Perry SJ, Baillie GS, Kohout TA, McPhee I, Magiera MM, Ang man E. Insulin-induced phosphorylation and activation of KL, Miller WE, McLean AJ, Conti M, Houslay MD, Lefkowitz phosphodiesterase 3B in rat adipocytes: possible role for RJ. Targeting of cyclic AMP degradation to beta 2-adrenergic protein kinase B but not mitogen-activated protein kinase or receptors by beta-arrestins. Science 2002;298:834–6 p70 S6 kinase. Endocrinology 1998;139:219–27 16. Xiang Y, Naro F, Zoudilova M, Jin SL, Conti M, Kobilka B. 13. Mongillo M, McSorley T, Evellin S, Sood A, Lissandron V, Phosphodiesterase 4D is required for beta2 adrenoceptor Terrin A, Huston E, Hannawacker A, Lohse MJ, Pozzan T, Houslay MD, Zaccolo M. Fluorescence resonance energy subtype-specific signaling in cardiac myocytes. Proc Natl Acad transfer-based analysis of cAMP dynamics in live neonatal rat SciUSA2005;102:909–14 cardiac myocytes reveals distinct functions of compartmentalized phosphodiesterases. Circ Res 2004;95:67–75

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 877 ECHO ROUNDS Intraoperative Transesophageal Echocardiography- Guided Patent Ductus Arteriosus Ligation in an Asymptomatic Nonbacterial Endocarditis Patient

Haibo Song, MD, Fei Liu, MD, Ke Dian, MD, and Jin Liu, MD

n routine examination an asymptomatic 16-year- calcified vegetations across the lumen and at the pulmo- old girl was found to have a continuous (machin- nary artery end. Vegetations were removed from the pul- Oery like) murmur in the second intercostal space. A monic and aortic valves. Finally, aortic valvuloplasty was transthoracic echocardiogram (TTE) demonstrated a patent performed. TEE confirmed the absence of flow through the ductus arteriosus (PDA) with continuous left-to-right PDA, and the absence of vegetations and residual regurgi- shunting and a mobile hyperechogenic vegetation (40 tation through the aortic valve when the operation was mm ϫ 4 mm) attached to the wall of the pulmonary artery, finished. Pathologic examination of the vegetations proved with normal-sized right heart and pulmonary artery pres- to be thrombi, and cultures were negative. sure. Three consecutive blood cultures were negative for TEE has been shown to be more sensitive than TTE for 1 bacterial growth. A chest radiograph was normal. The identifying small (Ͻ1 cm) valvular vegetations. In this case patient was diagnosed with PDA and noninfective endo- it provided confirmatory information as well as detecting carditis. Consent for publication of this case was obtained an aortic valve vegetation and valvular insufficiency not from the patient and her parents. appreciated on preoperative TTE, which prevents infective Intraoperative transesophageal echocardiography (TEE) endocarditis (IE) and other complications after surgery. (Philip IE33 ϫ 7 to 3) during PDA ligation showed a We believe that this is the first case report of an slightly enlarged left atrium and ventricle, mild mitral asymptomatic patient with nonbacterial thrombotic endo- valve regurgitation, and a normal triscuspid valve. Mod- carditis in a patient with PDA. The fetal ductus arteriosus erate aortic regurgitation and a 6-mm vegetation (Fig. 1 arises from the aorta opposite the origin of the left subcla- A) (Video 1; see Supplemental Digital Content 1, vian artery to the bifurcation of the MPA at the origin of the http://links.lww.com/AA/A177; see the Appendix for left pulmonary artery. PDA normally closes in the weeks video legends) was noted in the midesophageal aortic valve after birth, but persists in 10% of cases of adult congenital 2 long-axis view. Using live three-dimensional TEE imaging, heart disease. PDA-related IE is a rare but very dangerous event, which could cause septic shock or even pulmonary we noted the vegetation to be present on the right coronary 3,4 cusp (Figure 1B) (Video 2; see Supplemental Digital Con- thrombolization. Turbulent bloodflow produced by PDA tent 2, http://links.lww.com/AA/A178; see the Appendix is likely a causative factor in the development of nonbac- for video legends). Imaging in the midesophageal right terial endocarditis and IE. Continuous turbulent flow ventricular inflow–outflow view revealed a dilated main causes microscopic trauma to the endothelium of the pulmonary artery and aortic valve, which provides an pulmonary artery (MPA) without pulmonic valve regurgi- environment conducive to platelet and fibrin deposition tation. A PDA that measured 18 mm long and 8 mm wide and progression to IE if septicemia sets in. Before the was detected in the modified upper esophageal aortic arch introduction of antibiotic therapy and surgical closure, short-axis view. A mobile vegetation was seen in the PDA-related IE had a 45% mortality rate. With recent pulmonary artery on the downstream side of the PDA (Fig. advancement in diagnostic tools, especially echocardiogra- 2A). Color Doppler imaging demonstrated flow into the phy, preoperative antimicrobial therapies, and surgery, the MPA from descending aorta (Figure 2B). Continuous wave mortality rate of patients with PDA-related IE seems to Doppler showed a continuous spectrum; the peak velocity have declined.3 of the systolic shunt flow was 4.5 m/s, and diastolic flow PDA is usually demonstrated in the upper esophageal was 3 m/s. ascending aortic short-axis view with color Doppler flow After institution of cardiopulmonary bypass, TEE- mapping.2 However, we made a small change to enhance guided ligation of the PDA was first performed (Video 2, visualization of the PDA. We first get the upper esophageal clip 2). During the procedure, TEE monitoring of the aortic arch short-axis view. We advanced the probe 1 to 2 anterior wall of the pulmonary artery documented the cm from this view and rotated it from 0° to 60° to obtain a presence of the vegetation. Intraoperative surgical findings view of the descending aorta and MPA (Figure 2). Color- included a dilated and atherosclerotic PDA with small flow Doppler was used to detect flow through the PDA from 0° to 60°, from which we could visualize left-to-right From the West China Hospital, Sichuan University, Chengdu, Sichuan, China. continuous shunt flow originating from the descending Accepted for publication May 27, 2010. aorta. The systolic velocity was 4.5 m/s, yielding a peak Supplemental digital content is available for this article. Direct URL citations gradient between the descending aortic artery and pulmo- appear in the printed text and are provided in the HTML and PDF versions nary artery of 81 mm Hg, and a diastolic velocity (3 m/s) of this article on the journal’s Web site (www.anesthesia-analgesia.org). yielding the lowest gradient of 49 mm Hg. Address correspondence to Fei Liu, MD, West China Hospital, Sichuan TEE visualization of the PDA is potentially problematic University, Chengdu, Sichuan 610041, China. Address e-mail to [email protected]. because of interposition of the trachea between the esoph- Copyright © 2010 International Anesthesia Research Society agus and aortic arch (Figure 3 A). We used the technique 5 DOI: 10.1213/ANE.0b013e3181effffb described by Li et al., in which a saline-filled balloon is

878 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Comprehensive TEE for PDA Ligation

Figure 1. A, Two-dimensional transesophageal echocardiographic examination of the aortic valve long axis view. Arrow indicates a vegetation overlying the aortic valve. B, Live 3-dimensional view shows the vegetation overlying the right coronary artery cusp. DAO ϭ descending aorta; AO ϭ aorta; RV ϭ right ventricle; LV ϭ left ventricle; AVV ϭ aortic valve vegetation; RCS ϭ right coronary cusp; LCS ϭ left coronary cusp; NCS ϭ non-coronary cusp.

Figure 2. A, Two-dimensional transesophageal echocardiographic examination of the upper esophageal aortic arch short-axis sectional view at 30°. Arrows indicate the vegetation in the pulmonary artery. B, Color doppler image of shunted ﬂow through the patent ductus arteriosus in the upper esophageal aortic arch short-axis view. MPA ϭ main pulmonary artery; PV ϭ pulmonary vegetation; DAO ϭ descending aorta; PDA ϭ patent ductus arteriosus.

Figure 3. A, Desending aorta and a blurred image of the main pulmonary artery. B, Relationship between the esophagus, left main bronchus, and patent ductus arteriosus (PDA). We used a saline balloon in the left main bronchus to improve the image of the PDA. C, Clear pulmonary artery after we used a saline balloon. DAO ϭ descending aorta, MPA ϭ main pulmonary artery, AAO ϭ; AO arch ϭ aortic arch, LPA ϭ left pulmonary artery, LB ϭ left main bronchus; TEE ϭ transesophageal echocardiography; RPA ϭ right pulmonary artery; LPA ϭ left pulmonary artery.

advanced through the endotracheal tube after initiation of In summary, in this case, TEE monitoring played an cardiopulmonary bypass but before aortic cross-clamping important role in PDA ligation, facilitating assessment of to attenuate the ultrasound scattering by the tracheal air residual bloodflow. It also allowed monitoring of the status column (Fig. 3 B). This resulted in substantial enhancement of the vegetations in the pulmonary artery and on the aortic of the PDA in this case (Fig. 3C). valve.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 879 ECHO ROUNDS

APPENDIX: VIDEO LEGENDS 2. Neuman MD, Fox JD, Muehlschlegel JD. Incidental discovery of Video 1. Real-time monitoring of persistent patent ductus arteriosus a large patent ductus arteriosus in an adult during aortic (PDA) during surgical ligation. Initial bloodflow from the descending reconstruction: echocardiographic findings and diagnostic di- aorta to the pulmonary artery ceases by the end of the video lemmas. Anesth Analg 2007;105:1227–8 recording. White arrow shows colorful bloodflow ceases after PDA 3. Ozkokeli M, Ates M, Uslu N, Akcar M. Pulmonary and aortic ligation. MPA ϭ main pulmonary artery; DAO ϭ descending aorta. valve endocarditis in an adult patient with silent patent ductus Video 2. In the first part of the video is the live 2-dimensional arteriosus. Jpn Heart J 2004;45:1057–61 transesophageal echocardiography (TEE) view of aortic valve vegeta- 4. Kouris NT, Sifaki MD, Kontogianni DD, Zaharos I, Kalkandi tion in the aortic valve long-axis view. White arrow indicates a EM, Grassos HE, Babalis DK. Patent ductus arteriosus endarte- vegetation overlying the aortic valve. In the second part of the video ritis in a 40-year-old woman, diagnosed with transesophageal is the live 3-dimensional TEE view of aortic valve vegetation. White echocardiography. A case report and a brief review of the ϭ arrow indicates a vegetation overlying the aortic valve. LV left literature. Cardiovasc Ultrasound 2003;1:2 ϭ ϭ ventricle; LA left atrium; AO aorta. 5. Li YL, Wong DT, Wei W, Liu J. A new method for detecting the proximal aortic arch and innominate artery by transesophageal REFERENCES echocardiography. Anesthesiology 2006;105:226–7 1. Shapiro SM, Young E, De Guzman S, Ward J, Chiu CY, Ginzton LE, Bayer AS. Transesophageal echocardiography in diagnosis of infective endocarditis. Chest 1994;105:377–82

Clinician’s Key Teaching Points By Martin M. Stechert, MD, Roman M. Sniecinski, MD, and Martin J. London, MD • Patent ductus arteriosus (PDA) is the persistence of a normal fetal connection between the left pulmonary artery and the descending aorta. Clinical manifestations are determined by the size of the PDA and the degree of left-to-right shunting. Although large PDAs cause left ventricular overload and are usually diagnosed in the newborn, a small lesion may remain undetected into adulthood. Turbulent flow through the PDA can lead to endothelial injury, complicated by nonbacterial thrombotic endocarditis, bacterial seeding, and infective endocarditis. • Transesophageal echocardiography (TEE) can demonstrate a PDA with the upper esophageal aortic arch views, which show a dilated main pulmonary artery and, when color-flow Doppler is applied, turbulent flow into the left pulmonary artery from the descending aorta. However, because of the interposition of the air-filled trachea over the aortic arch, the left pulmonary artery is not always well visualized. Therefore, transthoracic echocardiography using the suprasternal and parasternal windows is the preferred method of diagnosis. • In this case, the authors initially obtained the upper esophageal aortic short-axis view, advanced the TEE probe 1 to 2 cm, then adjusted the omniplane angle to 60° to successfully visualize the PDA. The image was further improved by inserting a saline-filled balloon through the endotracheal tube while on bypass, which decreased scattering of the ultrasound by endotracheal air. • TEE may be useful for perioperative PDA assessment, provided an appropriate echo window can be found. In select patients, visualization might be improved by filling the endotracheal cuff with normal saline to provide such a window.

880 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Society for Ambulatory Anesthesiology

Section Editor: Peter S. A. Glass

Effect on Postoperative Sore Throat of Spraying the Endotracheal Tube Cuff with Benzydamine Hydrochloride, 10% Lidocaine, and 2% Lidocaine

Nan-Kai Hung, MD,* Ching-Tang Wu, MD,* Shun-Ming Chan, MD,* Chueng-He Lu, MD,* Yuan-Shiou Huang, MD,* Chun-Chang Yeh, MD,* Meei-Shyuan Lee, DPH,† and Chen-Hwan Cherng, MD, DMSc*

BACKGROUND: Postoperative sore throat (POST) is a common complication after endotracheal intubation. We compared the effectiveness on POST of spraying the endotracheal tube (ETT) cuff with benzydamine hydrochloride, 10% lidocaine, and 2% lidocaine. METHODS: Three hundred seventy-two patients were randomly allocated into 4 groups. The ETT cuffs in each group were sprayed with benzydamine hydrochloride, 10% lidocaine hydrochloride, 2% lidocaine hydrochloride, or normal saline before endotracheal intubation. After insertion, the

cuffs were inflated to an airway leak pressure of 20 cm H2O. Anesthesia was maintained with propofol. The patients were examined for sore throat (none, mild, moderate, or severe) at 1, 6, 12, and 24 hours after extubation. RESULTS: The highest incidence of POST occurred at 6 hours after extubation in all groups. There was a significantly lower incidence of POST in the benzydamine group than 10% lidocaine, 2% lidocaine, and normal saline groups (P Ͻ 0.05) at each observation time point. At 6 hours after extubation, the incidence of POST was significantly lower in the benzydamine group (17.0%) compared with 10% lidocaine (53.7%), 2% lidocaine (37.0%), and normal saline (40.8%) groups (P Ͻ 0.05). The benzydamine group had significantly decreased severity of POST compared with the 10% lidocaine, 2% lidocaine, and normal saline groups (P Ͻ 0.05) at each observation time point. Compared with the 2% lidocaine and normal saline groups, the 10% lidocaine group had significantly increased severity of POST at 1, 6, and 12 hours after extubation. There were no significant differences among groups in local or systemic side effects. CONCLUSIONS: Spraying benzydamine hydrochloride on the ETT cuff is a simple and effective method to reduce the incidence and severity of POST. (Anesth Analg 2010;111:882–6)

ostoperative sore throat (POST) after general anes- properties.13–16 It has been reported that moderate to thesia with an endotracheal tube (ETT) is an unde- severe sore throat may be resolved with gargling benzy- Psirable outcome1 with an incidence varying from damine hydrochloride.17 In addition, it is widely used in 40% to 100%.2–5 Although the symptoms resolve spontane- radiation-induced oral mucositis,18 for arthritis as a gel ously without any treatment, prophylactic management for ointment preparation applied to the skin,13 and for symp- decreasing its frequency and severity is still recommended tomatic treatment of acute sore throat.17,19 Preventive topical to improve the quality of postanesthesia care.6–8 benzydamine hydrochloride applied to the oropharyngeal Several pharmacological methods have been suggested cavity before endotracheal intubation or before endotracheal to reduce POST including inhaling beclomethasone; apply- intubation and continuously for 48 hours postoperatively has ing lidocaine spray or lidocaine to the ETT; administering been reported to decrease the incidence and severity of POST aspirin, ketamine, or benzydamine hydrochloride; or gar- after ETT insertion and laryngeal mask airway inser- 19–21 gling with azulene sulfonate.6,9–12 Benzydamine hydro- tion. However, patients may sustain numbness or the chloride is a topical nonsteroidal antiinflammatory drug sensation of tingling in the tissues in the oral cavity, dry that also has analgesic, antipyretic, and antimicrobial mouth, thirst, and nausea because of application of benzydamine hydrochloride by gargling or oropharyngeal spray.21 Combes et al.22 demonstrated that mucosal dam- From the *Department of Anesthesiology, Tri-Service General Hospital and age occurring at the cuff level is thought to be an important National Defense Medical Center; and †School of Public Health, National causative factor for tracheal morbidity. Therefore, applica- Defense Medical Center, Taipei, Taiwan, Republic of China. tion of benzydamine hydrochloride to the ETT rather than Accepted for publication January 11, 2010. gargling may provide an alternative and effective method Supported by grants from Tri-Service General Hospital (TSGH-C98-125) of Taiwan, Republic of China. to reduce the incidence and severity of POST. Address correspondence and reprint requests to Dr. Chen-Hwan Cherng, The hypothesis of this study was that simply spraying Department of Anesthesiology, Tri-Service General Hospital and National benzydamine hydrochloride on the endotracheal cuff will Defense Medical Center, #325, Section 2, Chenggung Rd., Neihu 114, Taipei, Taiwan, ROC. Address e-mail to [email protected]. provide better prevention of POST after extubation than Copyright © 2010 International Anesthesia Research Society 10% lidocaine hydrochloride, 2% lidocaine hydrochloride, DOI: 10.1213/ANE.0b013e3181d4854e or normal saline sprayed over the cuff of the ETT.

882 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 METHODS At 1, 6, 12, and 24 hours after extubation, the patients After obtaining hospital ethics committee approval and were asked about sore throat and hoarseness by a single written informed consent, 420 patients (n ϭ 105 per group), investigator who was blinded to the group allocation. POST ASA physical status I to III, aged 18 to 84 years, were was graded by a modified 4-point scale (1–4): 1, no sore included in this prospective, randomized, and double-blind throat; 2, mild sore throat (complains of sore throat only on study. All patients were scheduled for surgical procedures asking); 3, moderate sore throat (complains of sore throat under general anesthesia with orotracheal intubation in the on his or her own); and 4, severe sore throat (change of supine position. The exclusion criteria included patients voice or hoarseness, associated with throat pain).10,11 undergoing oral cavity surgery, cervical spine surgery, We recorded the patients’ age, sex, weight, height, thyroid surgery, patients with a difficult airway (after Ͼ1 duration of surgery, total fentanyl consumption, type of orotracheal intubation attempt), requiring a nasogastric surgery, and postoperative analgesia methods. Potential tube, or a history of perioperative sore throat. side effects associated with tracheal intubation or the Patients were randomly allocated into 4 groups by study drugs, such as nausea, vomiting, cough, throat choosing blinded envelopes. All study solutions and nor- numbness or stinging, dry mouth, and hoarseness, were mal saline were contained in similarly configured bottles. also recorded. The grouping and medication were as follows: Before initiation of the study, power analysis was performed. A minimum of 91 patients was required in each 1. The benzydamine (Comfflam, United Biomedical, Asia) group to detect a decrease of the incidence of POST from (containing ethanol, glycerin, and menthol as additives) 40% to 20% with a power of 80% and a significance level of group: 10 puffs of benzydamine hydrochloride were 95%. Statistical analysis was performed using the SPSS for sprayed on the ETT cuffs, which contain approximately 1.5 mg benzydamine hydrochloride. Windows version 15 (SPSS, Chicago, IL). Results are ex- 2. The 10% lidocaine (Xylocaine Spray 10%, AstraZeneca, pressed as mean (SD), median with range, or percentage. Sweden) (containing ethanol, polyethylene glycol 400, Patient age, height, weight, and duration of surgery were menthol, saccharin, and essence of banana as addi- compared among groups and tested statistically by analysis tives) group: 10 puffs of 10% lidocaine hydrochloride of variance. The incidence of POST and side effects among ␹2 were sprayed on the ETT cuffs, which contain ap- groups were tested by tests. To avoid a type I error, for ␹2 proximately 100 mg lidocaine hydrochloride. those significant variables in the test, we recalculated all ϫ ␹2 3. The 2% lidocaine (Xylocaine 2%, AstraZeneca) (con- possible six 2 2 tests by applying the Bonferroni ␣ ␹2 Ͼ ϭ ϭ taining sodium chloride as an additive) group: 10 inequality to adjust the level (i.e., P[ 6.97] 0.05/6 puffs of 2% lidocaine hydrochloride were sprayed on 0.0083 for 1 degree of freedom). Differences in the severity the ETT cuffs, which contain approximately 20 mg of symptoms among groups were evaluated by Kruskal- lidocaine hydrochloride. Wallis tests. Furthermore, the Dunn procedure was applied 4. The normal saline group: 10 puffs of normal saline to compare the difference among groups. The area under were sprayed on the ETT cuffs, which contain ap- the curve (AUC) by adding 3 trapezoid areas was gener- proximately 1 mL normal saline. ated from the level of POST and time (within 24 hours). Both differences in the severity of symptoms and AUC Sterile ETTs (ILM Endotracheal Tube, Euromedical, Ke- were tested by Kruskal-Wallis analysis of variance and Ͻ dah, Malaysia) with high-volume, low-pressure cuffs were followed by the Dunn procedure. Values of P 0.05 were used. Application of substances to the ETT cuff was per- considered statistically significant. formed 5 minutes before the induction of anesthesia. Male and female patients received 7.5- and 7.0-mm inner diameter ETTs, respectively. Before intubation, oxygen was RESULTS administered, adequate IV access was established, and Characteristics of the study groups are shown in Table 1. standard American Society of Anesthesiologists clinical There were 11, 12, 13, and 12 patients withdrawn from the monitoring was applied. After administration of 2 to 3 benzydamine, 10% lidocaine, 2% lidocaine, and normal ␮g/kg fentanyl, induction was accomplished with 2 to 2.5 saline groups, respectively, because of Ͼ1 attempt at oro- mg/kg propofol and 0.6 mg/kg rocuronium. Endotracheal tracheal intubation or nasogastric tube insertion during the intubation was performed by residents with at least 2 years operation. Therefore, there were 372 patients enrolled in of experience and attending physicians, who were blinded this study. to group allocation. The ETT cuffs were inflated with room In the 24-hour evaluation period, the highest incidences air to achieve a seal at 20 cm H2O of peak airway pressure. of POST occurred at the sixth hour time interval after Anesthesia was maintained by a target-controlled infusion extubation, with incidences (95% confidence interval) system (Fresenius Base Primea௡, Brezins, France) with of 17.0% (9.4%–24.6%), 53.7% (43.6%–63.9%), 37.0% propofol, intermittent fentanyl, and cisatracurium/rocuronium (27.1%–46.8%), and 40.8% (30.7%–50.9%) in the benzydam- administered as required. Nitrous oxide was not used. At ine, 10% lidocaine, 2% lidocaine, and normal saline groups, the end of surgery, neuromuscular blockade was antago- respectively. There was a significantly lower incidence of nized by neostigmine and atropine. After full recovery and POST in the benzydamine group than the 10% lidocaine, awakening, the ETT was removed after gentle suctioning of 2% lidocaine, and normal saline groups (P Ͻ 0.05) at each oral secretions. Patients were then transferred to the post- observation time point (Table 2). Similarly, the benzydam- anesthesia care unit. ine group had significantly decreased severity of POST

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 883 Benzydamine Hydrochloride Reduces Postoperative Sore Throat

Table 1. Demographic Data of the Patients and Data Related to the Surgery Benzydamine 10% lidocaine 2% lidocaine hydrochloride hydrochloride hydrochloride Normal saline (93 ؍ n) (92 ؍ n) (93 ؍ n) (94 ؍ Group (n Gender (M/F) 45/49 48/45 46/46 48/45 Age (y) 48.5 (16) 47.8 (15) 45.3 (17) 46.3 (17) Height (cm) 162.9 (9.2) 162.5 (8.8) 162.7 (9.3) 161.6 (9.5) Weight (kg) 62.2 (11.4) 63.8 (13.4) 64.3 (12.5) 62.3 (12.1) Duration of surgery (min) 183.7 (110) 181.6 (91) 183.8 (114) 176.8 (100) Total fentanyl consumption (␮g) 173 (73) 173 (59) 173 (61) 171 (61) Postoperative analgesia method (PCA/meperidine IM) 78/16 76/17 75/17 78/15 Types of surgery Colon rectal surgery 6 4 6 4 General surgery 24 18 20 16 Genitourinary surgery 3 5 5 4 Gynecologic surgery 20 16 18 24 Ophthalmologic surgery 12 17 16 15 Orthopedic surgery 23 18 16 21 Plastic surgery 6 15 11 9

Values are presented as mean (SD) or number of patients. PCA ϭ patient-controlled analgesia.

Table 2. Incidence (n, %) and Severity (Median, Range) of Postoperative Sore Throat Benzydamine 10% lidocaine 2% lidocaine Normal saline *P (93 ؍ n) (92 ؍ n) (93 ؍ n) (94 ؍ Evaluation time (n 1 h after extubation Incidence 10 (10.6%)* 30 (33.2%) 21 (22.8%) 22 (23.7%) 0.005 Severity 1 (1–4)* 1 (1–4)†‡ 1 (1–4) 1 (1–4) 0.004 6 h after extubation Incidence 16 (17.0%)* 50 (53.7%) 34 (37.0%) 38 (40.8%) Ͻ0.001 Severity 1 (1–4)* 2 (1–4)†‡ 1 (1–4) 1 (1–4) Ͻ0.001 12 h after extubation Incidence 5 (5.3%)* 37 (39.8%) 22 (23.9%) 30 (32.2%) Ͻ0.001 Severity 1 (1–4)* 1 (1–4)†‡ 1 (1–4) 1 (1–4) Ͻ0.001 24 h after extubation Incidence 2 (2.1%)* 25 (26.9%) 16 (17.4%) 19 (20.4%) Ͻ0.001 Severity 1 (1–4)* 1 (1–4)† 1 (1–4) 1 (1–4) Ͻ0.001 Values are presented as number of patients (%) or median (range). * P Ͻ 0.05, benzydamine group versus 10% lidocaine, 2% lidocaine, and normal saline groups. † P Ͻ 0.05, 10% lidocaine group versus 2% lidocaine group. ‡ P Ͻ 0.05, 10% lidocaine group versus normal saline.

compared with the 10% lidocaine, 2% lidocaine, and nor- was that applying 10% lidocaine on the ETT cuff increased mal saline groups (P Ͻ 0.05) at each observation time point the severity of POST. (Table 2). Compared with 2% lidocaine and normal saline POST is one of the common side effects associated groups, the 10% lidocaine group had significantly in- with endotracheal intubation.2–5 Tracheal mucosa lesions creased severity of POST at 1, 6, and 12 hours after after intubation and an overinflated ETT cuff have been extubation (Table 2). Of the mean AUCs generated from the proposed to be a possible cause of POST.22 These com- level of POST and time (within 24 hours), the benzydamine plications can occur after even a “smooth” intubation. group (25.6) had a significantly smaller area than the 10% Immediate POST may be primarily due to the action of Ͻ lidocaine (41.4) and normal saline groups (33.0) (P 0.05), extubation, and late POST may be related to tracheal but not the 2% lidocaine group (37.5). There were no mucosa trauma.23 According to the results of this study, significant differences in AUC among any other groups the highest incidence of POST occurred at the sixth hour (data not shown). after extubation, but not the first hour. Sore throat at the There were no significant differences among groups for first hour after extubation might be masked by residual the potential side effects relevant to tracheal intubation or analgesic effects after general anesthesia. study drugs. Benzydamine hydrochloride is indicated for the relief of painful conditions of the mouth and throat such as tonsil- DISCUSSION litis, sore throat, radiation mucositis, and postorosurgical This study demonstrated that spraying benzydamine hy- and periodontal procedures. Previous studies demon- drochloride on an ETT cuff may reduce the incidence and strated that topical application of benzydamine hydrochlo- severity of POST compared with applying 10% lidocaine, ride to the pharynx before laryngeal mask airway or ETT 2% lidocaine, and normal saline. An unexpected finding insertion decreased the incidence of POST.19–21 The side

884 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA effects of topical use of benzydamine hydrochloride in- In conclusion, application of benzydamine hydrochlo- clude local numbness or burning, stinging sensation, nau- ride on the ETT cuff effectively attenuates the incidence sea or vomiting, cough, dry mouth, throat discomfort, and severity of POST. Application of 10% lidocaine spray drowsiness, and headache,24 which may be evident before should be avoided because of worsening of POST, and induction of anesthesia. To avoid these adverse effects, spraying 2% lidocaine on the ETT cuff does not prevent we applied benzydamine hydrochloride on the ETT cuff POST. instead of perioperative topical application to the oral pharyngeal cavity. We found that this maneuver pro- REFERENCES vided excellent prevention of POST and reduced its 1. Macario A, Weinger M, Carney S, Kim A. Which clinical incidence from placebo or 2% lidocaine spray by Ͼ50%. anesthesia outcomes are important to avoid? The perspective of patients. Anesth Analg 1999;89:652–8 Therefore, the application of benzydamine hydrochlo- 2. Biro P, Seifert B, Pasch T. Complaints of sore throat after ride on the ETT cuffs may provide a simple and effective tracheal intubation: a prospective evaluation. Eur J Anaesthe- method to attenuate the incidence and severity of POST siol 2005;22:307–11 after tracheal intubation. 3. Al-Qahtani AS, Messahel FM. Quality improvement in anesthetic practice—incidence of sore throat after using small Application of lidocaine spray to the oral pharyngeal tracheal tube. Middle East J Anesthesiol 2005;18:179–83 cavity before intubation seems to increase the incidence of 4. Monroe MC, Gravenstein N, Saga-Rumley S. Postoperative sore throat.5,25,26 In this study, we also found that spraying sore throat: effect of oropharyngeal airway in orotracheally 10% lidocaine on the ETT cuff also increased the severity of intubated patients. Anesth Analg 1990;70:512–6 POST compared with 2% lidocaine or placebo. Ten percent 5. McHardy FE, Chung F. Postoperative sore throat: cause, prevention and treatment. Anaesthesia 1999;54:444–53 lidocaine solution contains ethanol, polyethylene glycol 6. el Hakim M. Beclomethasone prevents postoperative sore 400, menthol, saccharin, and macrogolum as additives in throat. Acta Anaesthesiol Scand 1993;37:250–2 the solvent, whereas the 2% lidocaine solution we used 7. Monem A, Kamal RS. Postoperative sore throat. J Coll Physi- contained sodium chloride as an additive. In fact, both cians Surg Pak 2007;17:509–14 8. Park SH, Han SH, Do SH, Kim JW, Rhee KY, Kim JH. menthol and ethanol can irritate tracheal mucosa, poten- Prophylactic dexamethasone decreases the incidence of sore tially causing tracheal mucosa damage, thus leading to throat and hoarseness after tracheal extubation with a double- increased severity of POST. However, Soltani and Agha- lumen endobronchial tube. Anesth Analg 2008;107:1814–8 davoudi27 reported that using intracuff lidocaine (ETT cuffs 9. Sumathi PA, Shenoy T, Ambareesha M, Krishna HM. Con- trolled comparison between betamethasone gel and lidocaine prefilled with 7 to 8 mL of 2% lidocaine for 90 minutes jelly applied over tracheal tube to reduce postoperative sore before intubation and refilled with enough 2% lidocaine throat, cough, and hoarseness of voice. Br J Anaesth 2008; after intubation) was superior to spraying topical 10% 100:215–8 lidocaine on laryngopharyngeal structures or on the distal 10. Canbay O, Celebi N, Sahin A, Celiker V, Ozgen S, Aypar U. end of the ETT for decreasing the incidence of POST. Ketamine gargle for attenuating postoperative sore throat. Br J Anaesth 2008;100:490–3 Theoretically, chemical irritation from the additives may be 11. Agarwal A, Nath SS, Goswami D, Gupta D, Dhiraaj S, Singh avoided by using intracuff lidocaine. We also found that 2% PK. An evaluation of the efficacy of aspirin and benzydamine lidocaine spray did not attenuate the incidence and severity hydrochloride gargle for attenuating postoperative sore throat: of POST compared with normal saline. The duration of the a prospective, randomized, single-blind study. Anesth Analg 2006;103:1001–3 analgesic effect of lidocaine spray applied to oral mucosa is 12. Ogata J, Minami K, Horishita T, Shiraishi M, Okamoto T, 28 Ͻ15 minutes. In this study, at the end of surgery (aver- Terada T, Sata T. Gargling with sodium azulene sulfonate aging 180 minutes after tracheal intubation), the analgesic reduces the postoperative sore throat after intubation of the effect of lidocaine spray might have already disappeared. trachea. Anesth Analg 2005;101:290–3 This probably explains why we found the incidence of 13. Quane PA, Graham GG, Ziegler JB. Pharmacology of benzydamine. Inflammopharmacology 1998;6:95–107 POST to be no different between the 2% lidocaine and 14. Baldock GA, Brodie RR, Chasseaud LF, Taylor T, Walmsley normal saline groups. LM, Catanese B. Pharmacokinetics of benzydamine after intra- One limitation of our study is that there was no record venous, oral, and topical doses to human subjects. Biopharm of coughing or bucking at the time of extubation. Although Drug Dispos 1991;12:481–92 15. Guglielmotti A, Aquilini L, Rosignoli MT, Landolfi C, Soldo L, the extubation protocol was the same in all groups, we did Coletta I, Pinza M. Benzydamine protection in a mouse model not evaluate the correlation between the frequency of of endotoxemia. Inflamm Res 1997;46:332–5 coughing or bucking at the time of extubation and the 16. Modeer T, Yucel-Lindberg T. Benzydamine reduces prosta- incidence of POST. The second limitation is that benzy- glandin production in human gingival fibroblasts challenged damine hydrochloride is available under different trade with interleukin-1 beta or tumor necrosis factor alpha. Acta Odontol Scand 1999;57:40–5 names in different countries, its formulations are quite 17. Turnbull RS. Benzydamine Hydrochloride (Tantum) in the different in each country, and the additives might also vary. management of oral inflammatory conditions. J Can Dent The drug is not available in the United States, Canada, and Assoc 1995;61:127–34 most of Western Europe (the United Kingdom being an 18. Epstein JB, Silverman S Jr, Paggiarino DA, Crockett S, Schubert MM, Senzer NN, Lockhart PB, Gallagher MJ, Peterson DE, exception). For this reason, the results of this study might Leveque FG. Benzydamine HCl for prophylaxis of radiation- not be widely applicable. Moreover, the safety and dosage induced oral mucositis: results from a multicenter, random- of benzydamine hydrochloride applied to the trachea need ized, double-blind, placebo-controlled clinical trial. Cancer further investigation, even though we did not find any 2001;92:875–85 19. Gulhas N, Canpolat H, Cicek M, Yologlu S, Togal T, Durmus adverse effects in our patients. The third limitation is that M, Ozcan Ersoy M. Dexpanthenol pastille and benzydamine the additives to 2% and 10% lidocaine solution are differ- hydrochloride spray for the prevention of post-operative sore ent, which may have influenced the result. throat. Acta Anaesthesiol Scand 2007;51:239–43

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 885 Benzydamine Hydrochloride Reduces Postoperative Sore Throat

20. Mazzarella B, Macarone Palmieri A, Mastronardi P, Spatola R, 25. Maruyama K, Sakai H, Miyazawa H, Iijima K, Toda N, Lamarca S, De Rosa G, Mastella A. Benzydamine for the Kawahara S, Hara K. Laryngotracheal application of lidocaine prevention of pharyngo-laryngeal pathology following tra- spray increases the incidence of postoperative sore throat after cheal intubation. Int J Tissue React 1987;9:121–9 total intravenous anesthesia. J Anesth 2004;18:237–40 21. Kati I, Tekin M, Silay E, Huseyinoglu UA, Yildiz H. Does 26. Hara K, Maruyama K. Effect of additives in lidocaine spray on benzydamine hydrochloride applied preemptively reduce sore postoperative sore throat, hoarseness and dysphagia after total throat due to laryngeal mask airway? Anesth Analg 2004; intravenous anaesthesia. Acta Anaesthesiol Scand 2005;49: 99:710–2 463–7 22. Combes X, Schauvliege F, Peyrouset O, Motamed C, Kirov K, 27. Soltani HA, Aghadavoudi O. The effect of different lidocaine Dhonneur G, Duvaldestin P. Intracuff pressure and tracheal application methods on postoperative cough and sore throat. morbidity: influence of filling with saline during nitrous oxide J Clin Anesth 2002;14:15–8 anesthesia. Anesthesiology 2001;95:1120–4 28. Schonemann NK, van der Burght M, Arendt-Nielsen L, Bjer- 23. Keane WM, Denneny JC, Rowe LD, Atkins JP Jr. Complications ring P. Onset and duration of hypoalgesia of lidocaine spray of intubation. Ann Otol Rhinol Laryngol 1982;91:584–7 applied to oral mucosa—a dose response study. Acta Anaes- 24. Passali D, Volonte M, Passali GC, Damiani V, Bellussi L. thesiol Scand 1992;36:733–5 Efficacy and safety of ketoprofen lysine salt mouthwash versus benzydamine hydrochloride mouthwash in acute pharyngeal inflammation: a randomized, single-blind study. Clin Ther 2001;23:1508–8

886 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA The Effectiveness of Benzydamine Hydrochloride Spraying on the Endotracheal Tube Cuff or Oral Mucosa for Postoperative Sore Throat

Yuan-Shiou Huang, MD,* Nan-Kai Hung, MD,* Meei-Shyuan Lee, MPH,† Chang-Po Kuo, MD,* Jyh-Cherng Yu, MD,‡ Go-Shine Huang, MD,* Chen-Hwan Cherng, MD, DMSC,* Chih-Shung Wong, MD, PhD,* Chi-Hong Chu, MD, PhD,‡ and Ching-Tang Wu, MD*

BACKGROUND: The etiology of postoperative sore throat (POST) is considered to be the result of laryngoscopy, intubation damage, or inflated cuff compression of the tracheal mucosa. In this study, we compared the effectiveness in alleviating POST using different approaches to benzydamine hydrochloride (BH) administration by spraying the endotracheal tube (ET) cuff or the oropharyngeal cavity, or both. METHODS: Three hundred eighty patients were included in this prospective and double-blind study, which was randomized into 4 groups: group A, oropharyngeal cavity spray of BH, and distilled water on the ET cuff; group B, both the oropharyngeal cavity and the ET cuff received BH spray; group C, the ET cuff received BH spray, and the oropharyngeal cavity received distilled water; and group D, distilled water sprayed on both the ET tube and into the oropharyngeal cavity. The patients were examined for sore throat (none, mild, moderate, severe) at 0, 2, 4, and 24 hours postextubation. RESULTS: The incidence of POST was 23.2%, 13.8%, 14.7%, and 40.4% in groups A, B, C, and D, respectively. POST occurred significantly less frequently in groups B and C compared with group D (odds ratio: 0.36; 95% confidence interval: 0.21–0.60; P Ͻ 0.05). However, there was no significant difference between groups A and D (odds ratio: 0.62; 95% confidence interval: 0.38–1.01). Moreover, there was no significant interaction between spraying BH over the oropharyngeal cavity and the ET cuff on the incidence of POST (P ϭ 0.088). The severity of POST was significantly more intense in group D compared with groups B and C (P Ͻ 0.001). Group B had a significantly higher incidence of local numbness, burning, and/or stinging sensation compared with patients in group D (P Ͻ 0.05). CONCLUSIONS: This study indicates that spraying BH on the ET cuff decreases the incidence and severity of POST without increased BH-related adverse effects. (Anesth Analg 2010; 111:887–91)

ostoperative sore throat (POST) after intubated gen- antiinflammatory drug and has analgesic, antipyretic, an- eral anesthesia (GA) is a troublesome complication timicrobial, and antiinflammatory effects.10 It has been and is recognized as one of the undesirable outcomes reported that moderate to severe oral inflammatory condi- P 1 11 in the postoperative period. The incidence of POST ranges tions may be resolved by gargling BH. In addition, it is from 21% to 66%2,3 in accordance with different surgical widely used in radiation-induced oral mucositis,12 for and anesthetic manipulations.4,5 Therapeutic management arthritis as a gel ointment preparation applied to the skin,10 for decreasing its frequency and severity is still advised to and for symptomatic treatment of acute sore throat.3,11 The improve the quality of postanesthesia care even though it aim of this study was to compare the effectiveness of 0.15% will resolve without treatment.4,6 Various methods have BH in alleviating POST using different approaches: spray- been reported to alleviate POST, such as using a smaller ing on the ET cuff, into the oropharyngeal cavity, or both. sized endotracheal tube (ET), lubricating the ET cuff, and avoiding excessive intracuff pressure.2 In addition, several METHODS analgesic and antiinflammatory drugs have also been used, This study was approved by our institutional ethics com- 4 including IV dexamethasone, inhaled beclomethasone, mittee, and written informed consent was obtained from all 6 7 and gargling with azulene sulfonate, aspirin, or benzy- participants. In this prospective, randomized, and double- damine hydrochloride (BH).8,9 BH is a topical nonsteroidal blind study, we enrolled 380 patients. All patients were scheduled for elective surgery under GA with ET intuba- From the *Department of Anesthesiology, †School of Public Health, and tion. Patients who had head and neck surgery, cervical ‡Division of General Surgery, Tri-Service General Hospital and National Defense Medical Center, Taipei, Taiwan, Republic of China. spine surgery, or thyroid surgery, and had a history of Accepted for publication April 25, 2010. preoperative sore throat were excluded. Patients with first Disclosure: The authors report no conflicts of interest. attempt failed laryngoscopy, postoperative endotracheal Address correspondence and reprint requests to Ching-Tang Wu, MD, intubation or reintubation, and nasogastric tube insertion Department of Anesthesiology, Tri-Service General Hospital and Na- were also excluded from further analysis. tional Defense Medical Center, #325, Section 2, Chenggung Rd., Neihu All patients were randomly assigned into 4 groups (95 114, Taipei, Taiwan. Address e-mail to [email protected] or [email protected]. patients in each group). The grouping and medication were Copyright © 2010 International Anesthesia Research Society as follows: group A, 5 puffs of 0.15% BH (total 0.75 mg, DOI: 10.1213/ANE.0b013e3181e6d82a Comfflam; United Biomedical Asia, Taiwan) were sprayed

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 887 Topical Benzydamine Hydrochloride Attenuates Postoperative Sore Throat into the oropharyngeal cavity and 5 puffs of distilled water hoarseness, nausea and vomiting, and the type of postop- were sprayed on the ET cuff (5 puffs containing approxi- erative analgesia. Potential side effects associated with the mately 0.5 mL distilled water); group B, the oropharyngeal study drugs, such as local numbness, burning, and/or cavity and the ET cuff were both sprayed with 5 puffs of stinging sensation, cough, and dry mouth, were also re- BH; group C, the ET cuff was sprayed with 5 puffs of BH corded during spraying and postoperatively by partici- and the oropharyngeal cavity was sprayed with 5 puffs of pants’ self report. distilled water; and group D, the oropharyngeal cavity and Before initiation of the study, a power calculation was the ET cuff were both sprayed with 5 puffs of distilled performed to determine the required sample size based on water. All medications were sprayed 5 minutes before our institute’s previous data. A minimum of 91 patients induction of anesthesia by a nurse anesthetist blinded to the was required in each group to detect a decreased incidence treatment. The treatments were prepared by a pharmacist of POST from 40% to 20% with a power of 80% and a in our pharmacy department blinded to the medication so significance level of 0.05. To compensate for potential that the different treatments had the same external appear- dropouts, we enrolled 95 patients in each group. Statistical ance. No patient received premedication, and standard analysis was performed using SPSS for Windows version monitors were applied in the operating room. GA was 14 (SPSS, Chicago, IL). Data are expressed as mean (SD), induced with 2 to 3 ␮g/kg fentanyl, 1 to 1.5 mg/kg and median with range or percentage. We used a 1-way lidocaine, and 2 to 2.5 mg/kg propofol. Laryngoscopy and analysis of variance to compare patients’ ages, heights, ET intubation were facilitated by administration of 0.6 weights, and durations of surgery among groups. The mg/kg rocuronium. Patients were tracheally intubated overall incidence of POST and side effects among groups ␹2 with a 7.0- and 6.5-mm inner diameter ET with a high- were tested using tests. For those significant variables in the ␹2 test, we recalculated all possible six 2 ϫ 2 ␹2 tests by volume and low-pressure cuff (ILM Endotracheal Tube; applying the Bonferroni inequality to adjust ␣ level [i.e., Euromedical, Kedah, Malaysia), respectively. Intubations P(␹2 Ͼ 6.97) ϭ 0.05/6 ϭ 0.0083 for 1 degree of freedom] to were performed by residents with at least 2 years of avoid type I error. A Kruskal-Wallis test followed by the experience or attending physicians who were blinded to the Dunn procedure was applied to compare the differences in treatment. The ET cuffs were inflated with room air to a the severity of POST among groups. We used univariable cuff pressure of 20 to 25 cm H O. We measured the cuff 2 and multivariable logistic regressions for evaluating the pressure immediately after ET intubation using a manom- interaction between 2 treatments (mucosa and cuff) and the eter (VBM, Sulz, Germany) that was connected to the pilot relative risk, as odds ratio (OR) and 95% confidence inter- balloon of the ET cuff via a 3-way stopcock, and the cuff val (CI) of BH on POST. Probability values Ͻ0.05 were pressure was measured once in each patient at 60 minutes considered statistically significant. after intubation. Anesthesia was maintained with 8% to 12% desflurane in a total flow of 300 mL/min oxygen RESULTS under a closed system without nitrous oxide. Fentanyl was The characteristics of the study groups are shown in Table administered depending on the surgical stimulus and the 1; there were no significant differences among the 4 groups hemodynamic response; the dosage was approximately 1 in age, sex, height, weight, body mass index, and the ␮g/kg/h after intubation. Neuromuscular transmission duration of anesthesia. Two patients in groups B and D was monitored using train-of-four supramaximal stimula- received unanticipated nasogastric tubes, and the statistical tions (2 Hz, 50 mA) (TOF Watch SX; Organon, Dublin, analysis was performed without those patients. Ireland). Intravenous boluses of cisatracurium 2 mg were Table 2 lists the incidence and severity of POST for the administered when more than 2 responses were detected in study groups at 0, 2, 4, and 24 hours postoperatively. train-of-four stimulation until closure of the peritoneum Within the 24-hour period of evaluation, the overall inci- was commenced. At the end of surgery, residual neuro- dence of POST (patients with any POST during 24-hour muscular blockade was antagonized by neostigmine and evaluation/patient numbers) in groups A, B, C, and D was Ն atropine. When the T4/T1 ratio reached 90% and patients 23.2%, 13.8%, 14.7%, and 40.4%, respectively, mostly at the could follow simple commands, the patients were trache- second hour after extubation. There were significantly ally extubated after gentle oropharyngeal suction and then fewer incidences in groups B and C, but not in group A transferred to the postanesthesia care unit. compared with group D, for the total incidence of POST. When arriving at the postanesthesia care unit (0 hour) The severity of POST was significantly higher in group D and thereafter at 2, 4, and 24 hours, patients were assessed compared with groups B and C (P Ͻ 0.001). for the incidence and severity of sore throat by another No statistically significant differences of POST incidence anesthesiologist blinded to the treatment. POST was were found among patients who received BH spray only on graded on a 4-point scale (0–3): 0, no sore throat; 1, mild the ET cuff, only in the oropharyngeal cavity, or both, sore throat (complaints of sore throat only when which indicated no additive or synergistic effects of appli- prompted); 2, moderate sore throat (complaints of sore cation of BH (P ϭ 0.088). Patients who received BH on the throat volunteered by the patient without prompting); and ET cuff had a significantly lower risk of POST (OR: 0.36; 3, severe sore throat (change of voice or hoarseness, asso- 95% CI: 0.21–0.60) compared with the placebo group. ciated with throat pain).11 Total POST cases were those Those who received BH in the oropharyngeal cavity also patients who reported any degree of sore throat over the had less risk of POST (OR: 0.62; 95% CI: 0.38–1.01) com- 24-hour evaluation period. We recorded the patient’s age, pared with the placebo group, although not significant. sex, smoking history, weight, height, duration of surgery, After adjusting for potential confounders (age, gender,

888 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Demographic Data and Duration of Surgery Group P value (94 ؍ D (n (95 ؍ C (n (94 ؍ B (n (95 ؍ A(n Gender (male/female) 49/46 44/50 45/50 44/50 0.919 Age (y) 49.2 (18) 47.1 (15) 48.7 (16) 45.5 (18) 0.446 Height (cm) 163.7 (8.0) 162.1 (8.8) 162.9 (9.4) 162.4 (9.6) 0.631 Weight (kg) 63.2 (12.1) 62.9 (10.8) 62.0 (11.5) 62.7 (11.7) 0.897 BMI (kg/m2) 23.5 (3.5) 23.9 (3.6) 23.3 (3.4) 23.7 (3.9) 0.655 Duration of anesthesia (min) 175 (99) 186 (91) 176 (84) 175 (83) 0.917 Intraoperative fentanyl use 318 (65) 312 (57) 310 (60) 306 (60) 0.706 Smoking habit 35 40 38 42 0.964 Postoperative analgesia (PCA with 80/15 78/16 77/18 75/19 0.862 fentanyl/demerol, prn) Type of surgery Colon rectal surgery 86560.850 General surgery 21 20 21 18 0.954 Genitourinary surgery 45750.815 Gynecologic surgery 22 16 18 19 0.758 Ophthalmologic surgery 12 16 16 18 0.673 Orthopedic surgery 22 20 20 18 0.928 Plastic surgery 6 11 8 10 0.585 BMI ϭ body mass index; PCA ϭ patient-controlled analgesia; prn ϭ pro re nata (when necessary). Values are presented as mean (SD) or number.

Table 2. The Incidence and Severity of Postoperative Sore Throat for the Study Groups P value (94 ؍ Group D (n (95 ؍ Group C (n (94 ؍ Group B (n (95 ؍ Evaluation time Group A (n 0 h 6 (6.3%)* 4 (4.2%)* 5 (5.2%)* 20 (20.6%) Ͻ0.001 Grading of discomfort 0.001 Mild 4 2 3 12 Moderate 2 2 2 8 Severe 0 0 0 0 2 h 22 (23.2%) 13 (13.8%)* 14 (14.7%)* 38 (40.4%) Ͻ0.001 Grading of discomfort Ͻ0.001 Mild 13 5 6 9 Moderate 5 4 5 15 Severe 4 4 3 14 4 h 12 (12.6%)* 8 (8.3%)* 10 (10.4%)* 30 (30.9%) Ͻ0.001 Grading of discomfort 0.001 Mild 6 3 5 8 Moderate 2 1 2 8 Severe 4 4 3 14 24 h 10 (10.5%) 7 (7.3%)* 8 (8.3%)* 22 (22.7%) 0.003 Grading of discomfort 0.039 Mild 4 2 4 5 Moderate 2 1 1 3 Severe 4 4 3 14 Total incidence of POST 22 (23.2%) 13 (13.8%)* 14 (14.7%)* 38 (40.4%) Ͻ0.001 Severity of POST 0 (0–3) 0 (0–3)* 0 (0–3)* 0 (0–3) Ͻ0.001 POST ϭ postoperative sore throat. Values are presented as number of subjects (%) or median (range). * P Ͻ 0.05, compared with group D. The incidence of POST was done by ␹2 test by applying the Bonferroni inequality to adjust ␣ level ͓i.e., P(␹2 Ͼ 6.97) ϭ 0.05/6 ϭ 0.0083 for 1 degree of freedom͔ to avoid type I error. The severity of POST was determined by Kruskal-Wallis test and pairwise posteriori comparisons were done by Dunn procedure.

smoking, and duration of anesthesia), the association be- potential side effects relevant to BH, such as nausea or tween treatments for the risk of developing POST did not vomiting, cough, and mouth dryness (Table 4). change substantially (Table 3). The incidence of local numbness, burning, or stinging DISCUSSION sensation in groups A, B, C, and D was 8.42%, 10.6%, 3.19%, Our major finding was that spraying BH on the ET cuff may and 1.06%, respectively. Only group B had a significantly reduce the incidence and severity of POST up to 24 hours higher incidence of local numbness, burning, or stinging postoperatively compared with the application of distilled sensation than group D (P Ͻ 0.05). There were 4, 4, 3, and water. Moreover, spraying BH in the oropharyngeal cavity 5 patients who had hoarseness in groups A, B, C, and D, did not reduce the incidence and severity of POST. Indeed, respectively. There was no significant difference in the there was no additional advantage from spraying BH on the

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 889 Topical Benzydamine Hydrochloride Attenuates Postoperative Sore Throat

into the oropharyngeal cavity before induction of anesthe- Table 3. Odds Ratios and 95% Conﬁdence Intervals of Benzydamine Hydrochloride on sia may ameliorate the local irritation of BH. Postoperative Sore Throat at 2 Hours The first limitation in our study is the pharmacokinetics After Surgery and safety of BH on the tracheal mucosa. BH is well absorbed 15 No. of sore through the oral mucosa, and its effects last for 1.84 hours. In throats/no. Univariable Multivariablea our study, BH may be well absorbed after application to the of patients OR (95% CI) OR (95% CI) trachea because of its thinner epithelial layer. However, BH on mucosa concern for damage to the tracheal structure should be Yes 36/189 0.63 (0.39–1.03) 0.65 (0.40–1.06) considered, even though we did not find any adverse event in No 51/189 1.00 (reference) 1.00 (reference) 16 BH on cuff our limited data. Silvestrini et al. had reported that gross Yes 27/189 0.36 (0.22–0.60)* 0.36 (0.22–0.60)* and histological structure of the major organs such as lung, No 60/189 1.00 (reference) 1.00 (reference) liver, kidney, and spleen were not altered by benzydamine in OR ϭ odds ratio; CI ϭ conﬁdence interval; BH ϭ benzydamine hydrochloride. an animal study. However, no relevant human study regard- a Adjusted for gender, age (in years), smoking, and anesthetic time. ing a histological examination of the trachea has been re- * P Ͻ 0.05 compared with the reference group. ported. A second limitation is that we did not ensure that each group did in fact have an equal amount of BH, although we Table 4. The Local and Systemic Side Effects of limited the dose to 5 sprays each on either the ET or into the Benzydamine Hydrochloride in the Study Groups oropharyngeal cavity, and an equal dosage was prescribed. A Group A Group B Group C Group D third limitation of our randomized study is that postoperative pain management can influence POST, although there was no (94 ؍ n) (95 ؍ n) (94 ؍ n) (95 ؍ n) Nausea 20 25 22 25 significant difference in postoperative pain management. At Vomiting 18 16 24 20 the same time, dry oxygen and IV lidocaine would affect Cough 19 22 20 18 Local stinging or 8 10* 3 1 POST; however, we used these routinely and standardized numbness of the the dosage for each patient. throat and mouth In conclusion, BH topical spray on the ET cuff before Dry mouth 42 46 45 48 intubation and GA can reduce the incidence and severity of Hoarseness 4 4 3 5 POST without increased BH-related adverse effects. Values are presented as number. * P Ͻ 0.05 compared with group D. ␹2 test and applying the Bonferroni inequality to adjust ␣ level ͓i.e., P(␹2 Ͼ 6.97) ϭ 0.05/6 ϭ 0.0083 for 1 REFERENCES degree of freedom͔ to avoid type I error. 1. Macario A, Weinger M, Carney S, Kim A. Which clinical anesthesia outcomes are important to avoid? The perspective of patients. Anesth Analg 1999;89:652–8 cuff and into the oropharyngeal cavity simultaneously for 2. Al-Qahtani AS, Messahel FM. Quality improvement in anes- POST prevention, even though there was a decreased trend of thetic practice: incidence of sore throat after using small POST incidence when spraying BH on the ET cuff compared tracheal tube. Middle East J Anesthesiol 2005;18:179–83 with into the oropharyngeal cavity. Our results indirectly 3. Gulhas N, Canpolat H, Cicek M, Yologlu S, Togal T, Durmus M, Ozcan Ersoy M. Dexpanthenol pastille and benzydamine demonstrated that mucosal irritation occurring at the level of hydrochloride spray for the prevention of post-operative sore the ET cuff is the most important causative factor for tracheal throat. Acta Anaesthesiol Scand 2007;51:239–43 morbidity, and this outcome was consistent with previous 4. Park SH, Han SH, Do SH, Kim JW, Rhee KY, Kim JH. studies.13,14 Prophylactic dexamethasone decreases the incidence of sore Local spraying of BH can resolve moderate to severe sore throat and hoarseness after tracheal extubation with a double- lumen endobronchial tube. Anesth Analg 2008;107:1814–8 throat after ET intubation.11,15 Preventive topical application 5. Sumathi PA, Shenoy T, Ambareesha M, Krishna HM. Controlled of BH into the oropharyngeal cavity before intubation and comparison between betamethasone gel and lidocaine jelly ap- continuous use for 48 hours after the operation effectively plied over tracheal tube to reduce postoperative sore throat, decreased the incidence and severity of POST after ET and cough, and hoarseness of voice. Br J Anaesth 2008;100:215–8 also laryngeal mask airway insertion.3,8 In this study, we 6. Ogata J, Minami K, Horishita T, Shiraishi M, Okamoto T, showed that spraying BH into the oropharyngeal cavity did Terada T, Sata T. Gargling with sodium azulene sulfonate reduces the postoperative sore throat after intubation of the not reduce the incidence and severity of POST. This might be trachea. Anesth Analg 2005;101:290–3 because of the small dosage of BH used in the oropharyngeal 7. Agarwal A, Nath SS, Goswami D, Gupta D, Dhiraaj S, Singh cavity (0.75 mg) in our study compared with previous studies PK. An evaluation of the efficacy of aspirin and benzydamine (which ranged from 2.16 to 22.5 mg).3,7 hydrochloride gargle for attenuating postoperative sore throat: There are many side effects (64%) of topical use of BH, a prospective, randomized, single-blind study. Anesth Analg 2006;103:1001–3 including local numbness, burning or stinging sensation, 8. Kati I, Tekin M, Silay E, Huseyinoglu UA, Yildiz H. Does benzy- nausea or vomiting, cough, dry mouth, thirst, throat dis- damine hydrochloride applied preemptively reduce sore throat 15 8 comfort, drowsiness, and headache. Kati et al. and due to laryngeal mask airway? Anesth Analg 2004;99:710–2 Agarwal et al.7 reported local numbness, dysgeusia, and 9. Hung NK, Wu CT, Chan SM, Lu CH, Huang YS, Yeh CC, Lee throat irritation in BH-pretreated patients. In our current MS, Cherng CH. Effect on postoperative sore throat of spray- study, we found that 8% to 10% of patients experienced ing the endotracheal tube cuff with benzydamine hydrochloride, 10% lidocaine, and 2% lidocaine. Anesth Analg 2010 Mar these side effects after topical application of BH before 19 [Epub ahead of print] anesthesia induction. To avoid these drawbacks, we sug- 10. Quane PA, Graham GG, Ziegler JB. Pharmacology of benzy- gest that the application of BH onto the ET cuff instead of damine. Inflammopharmacology 1998;6:95–107

890 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 11. Turnbull RS. Benzydamine hydrochloride (Tantum) in the 14. Peppard SB, Dickens JH. Laryngeal injury following management of oral inflammatory conditions. J Can Dent short-term intubation. Ann Otol Rhinol Laryngol 1983;92: Assoc 1995;61:127–34 327–30 12. Epstein JB, Silverman S Jr, Paggiarino DA, Crockett S, Schubert 15. Passali D, Volonte M, Passali GC, Damiani V, Bellussi L. MM, Senzer NN, Lockhart PB, Gallagher MJ, Peterson DE, Efficacy and safety of ketoprofen lysine salt mouthwash versus Leveque FG. Benzydamine HCl for prophylaxis of radiation- benzydamine hydrochloride mouthwash in acute pharyngeal induced oral mucositis: results from a multicenter, random- inflammation: a randomized, single-blind study. Clin Ther ized, double-blind, placebo-controlled clinical trial. Cancer 2001;23:1508–18 2001;92:875–85 16. Silvestrini B, Barcellona PS, Garau A, Catanese B. Toxicology of 13. Combes X, Schauvliege F, Peyrouset O, Motamed C, Kirov K, benzydamine. Toxicol Appl Pharmacol 1967;10:148–59 Dhonneur G, Duvaldestin P. Intracuff pressure and tracheal morbidity: influence of filling with saline during nitrous oxide anesthesia. Anesthesiology 2001;95:1120–4

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 891 Strepsilsா Tablets Reduce Sore Throat and Hoarseness After Tracheal Intubation

Amin Ebneshahidi, MD, and Masood Mohseni, MD

BACKGROUND: Amyl-m-cresol (Strepsilsா) has been successfully used in the prophylaxis and treatment of oral inflammations, but its effects on postintubation sore throat and hoarseness are unknown. We conducted this study to evaluate the effects of Strepsils in reducing postintubation sore throat and hoarseness. METHODS: One hundred fifty patients, ASA physical status I to II, scheduled to undergo general anesthesia and elective orthopedic or gynecologic surgery were enrolled. Participants were randomly allocated to receive either Strepsils or identical-looking placebo tablets immediately before arrival to the operating room. The incidence and severity of postoperative sore throat and hoarseness were evaluated immediately and 24 hours after surgery. RESULTS: The incidence of early postoperative sore throat was 13.7% and 33.3% and hoarseness was 12.3% and 26.4% in the Strepsils and placebo groups, respectively (P Ͻ 0.05). One day after surgery, the incidence of sore throat decreased to 6.8% and 18.1% in the Strepsils and control groups, respectively. The incidence of hoarseness 1 day after the operation decreased to 8.2% in the Strepsils group and 19.4% in the placebo group, but the difference remained statistically significant (P Ͻ 0.05). CONCLUSION: Perioperative use of Strepsils tablets reduces postoperative sore throat and hoarseness of voice. (Anesth Analg 2010;111:892–4)

ostoperative sore throat and hoarseness are minor but either Strepsils or identical-looking placebo tablets imme- frequent complications of endotracheal intubation that diately before arrival to the operating room (45 minutes Poccur in up to 50% of patients.1,2 These complications before induction of anesthesia on average). A blinded nurse negatively influence patient satisfaction and occasionally re- administered the tablets to the patients. quire treatment with supplementary analgesics. Several drugs Strepsils honey and lemon lozenge (Boots, Nottingham, UK) such as clonidine,3 betamethasone gel,4 and chamomile- contained the active ingredients 2,4-dichlorobenzyl alcohol 1.2 extract spray5 have been used to reduce postoperative sore mg, amylmetacresol 0.6 mg, sucrose, glucose syrup, honey, throat and hoarseness with varying efficacy. A simple, safe, tartaric acid, peppermint oil, terpeneless lemon oil, and quinoline and inexpensive perioperative intervention to prevent post- yellow (E104). The placebo tablets (Anata, Tabriz, Iran) contained operative sore throat and hoarseness would be useful. sugar, glucose, citric acid, lemon flavor, and color (E104). The Amyl-m-cresol (Strepsils௡) has been successfully used in the lozenges were not easily distinguishable from placebo tablets treatment of oral inflammation and for the prevention of pain with regard to color, taste, and smell. The randomization was and inflammation after oral surgery,6,7 but the effects on postin- performed by the hospital pharmacy using a table of random tubation sore throat and hoarseness are unknown. This study numbers, and the patients, nurses, and anesthetic technician who was conducted to evaluate whether the perioperative use of were involved in the patients’ care and data recording were Strepsils lozenges would reduce the incidence of postoperative blinded to the nature of the assignment. sore throat and hoarseness compared with placebo. Method of Anesthesia All patients were premedicated with oral oxazepam 10 mg METHODS and ranitidine 150 mg 2 hours before surgery. Fentanyl 3 to Ϫ Ϫ Patients 4 ␮g ⅐ kg 1 and IV lidocaine 1.5 mg ⅐ kg 1 were adminis- This prospective study was approved by the ethics com- tered 3 to 5 minutes before tracheal intubation. After the Ϫ mittee of Sadi Hospital, and written informed consent was administration of 100% oxygen at5Lmin 1 for several obtained from all patients. One hundred fifty ASA physical minutes, anesthesia was induced with propofol 1.5 to Ϫ Ϫ status I to II patients, aged 19 to 63 years, undergoing 2mg⅐ kg 1 and atracurium (0.5 mg ⅐ kg 1). Laryngoscopy elective orthopedic or gynecologic surgery under general was performed 3 to 5 minutes after atracurium injection anesthesia who were expected to remain in the hospital for when adequate neuromuscular blockade was confirmed Ͼ 24 hours were enrolled. Patients with significant sore using train-of-four nerve stimulation. A train-of-four Ͻ10% throat for any reason or obvious hoarseness were not was used as an indication of satisfactory neuromuscular included. Participants were randomly allocated to receive blockade for intubation. Difficulty in laryngoscopy was graded as 1, no difficulty; 2, only posterior extremity of From the Department of Anesthesiology, Sadi Hospital, Isfahan, Iran. glottis visible; 3, only epiglottis seen; and 4, no recognizable Accepted for publication November 29, 2009. structures. Endotracheal intubation was performed using a Supported by departmental funds. polyvinyl chloride endotracheal tube (Supa, Tehran, Iran) Address correspondence and reprint requests to Amin Ebneshahidi, MD, with 7- to 8-mm internal diameter for women and 7.5- to Department of Anesthesiology, Sadi Hospital, Khodaverdi Alley, Sadi Blv, Isfahan, Iran. Address e-mail to [email protected]. 8.5-mm internal diameter for men. The high-volume, low- Copyright © 2010 International Anesthesia Research Society pressure cuff was inflated until no air leak could be heard DOI: 10.1213/ANE.0b013e3181d00c60 with peak airway pressure at 20 cm H2O. Cricoid pressure

892 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 was not applied for intubation. Patients requiring nasogas- Table 1. Group Baseline Characteristics tric tube placement or those in whom laryngoscopy was and Covariates attempted more than once were withdrawn. Strepsils Placebo (72 ؍ n) (73 ؍ General anesthesia was maintained with propofol, fentanyl, Variables (n and atracurium. Controlled mechanical ventilation with an initial a Ϫ Age (y) 29 (9) 31 (10) tidal volume of 10 mL ⅐ kg 1 and respiratory frequency of 10 Male sexb 43 (58.9) 40 (55.5) Ϫ breaths ⅐ min 1 was adjusted to maintain normocapnia. Weight (kg)a 58 (9) 57 (11) Duration of laryngoscopy (s)a 13.7 (3.1) 14.3 (3.6) Neuromuscular blockade was reversed with neostig- Duration of intubation (min)a 149 (44) 158 (51) Ϫ1 Ϫ1 mine 0.05 mg ⅐ kg and atropine 0.02 mg ⅐ kg . Grade of laryngoscopic viewc 2 (1,2) 2 (1,2) When spontaneous ventilation was adequate and the Type of surgeryb patients were able to follow verbal commands, tracheal Myomectomy/hysterectomy 9 (12.3) 7 (9.7) Ovarectomy 4 (5.5) 5 (6.9) extubation was performed immediately after suctioning of Diagnostic laparoscopy 10 (13.7) 12 (16.6) the oropharynx at the discretion of the responsible anesthe- Bone tumor resection 7 (9.5) 8 (11.1) siologist. An oral airway was inserted before extubation Total hip replacement 4 (5.5) 3 (4.2) Traumatic fractures 31 (42.4) 29 (40.2) and emergence in all patients. Shoulder surgery/arthroscopy 4 (5.5) 3 (4.2) Bone grafting 4 (5.5) 5 (6.9) Outcome Measures Positionb Postoperative sore throat and hoarseness were evaluated Supine 55 (75.3) 52 (72.2) within 20 minutes in the recovery room and 24 hours after Lateral decubitus 14 (19.2) 17 (23.6) Sitting 4 (5.5) 3 (4.2) surgery. At the time of the first evaluation, patients with a a Ramsay Sedation Score8 of 2 (cooperative, oriented, and Data are presented as mean (SD). b Data are presented as n (%). tranquil) or 3 (responding to commands only) were in- c Data are presented as median (25%, 75%). cluded. The severity of sore throat was graded as follows: 0, no sore throat; 1, minimal; 2, moderate; 3 severe; and for hoarseness, the grading was as follows: 0, no hoarseness; 1, slight hoarseness; 2, severe hoarseness; 3, cannot speak because of hoarseness. Additional analgesics were not administered until completion of the first evaluation. All scales were completed by a nurse blinded to the study groups. Demographic data, perioperative steroid use, type of surgery, the time needed for laryngoscopy (time from opening the mouth to placement of the endotracheal tube), intubation period (time from intubation to extubation), and grades of laryngoscopic view as well as the incidence of bucking were recorded by a blinded anesthetic technician. Figure 1. Incidence of sore throat and hoarseness at recovery from Patients’ discomfort after lozenge administration including anesthesia and 24 hours postoperatively in the Strepsils (white) and subjective symptoms of allergy to lozenges, bad taste or placebo (gray) groups. Data shown as percentages; P Ͻ 0.05 with ␹2 test. smell, and nausea were also recorded by the same technician before induction of anesthesia. room. The distribution of surgical procedures and positions Statistical Analysis are shown in Table 1. Patient characteristics, grade of diffi- Results are presented as either mean (SD) or percentages, as culty of laryngoscopy, and the duration of laryngoscopy appropriate. The incidences of sore throat and hoarseness (13.7% vs 33.3%, P ϭ 0.81) and intubation (12.3% vs 26.4%, were compared between the 2 groups using the Fisher exact P ϭ 0.14) were comparable between the 2 groups (Table 1). test. The scores of sore throat and hoarseness as well as grade None of the patients complained of any kind of discomfort of laryngoscopic view were compared between the 2 groups after using either Strepsils or placebo. The incidence of early with the Mann-Whitney U test. The duration of laryngoscopy postoperative sore throat in patients who received Strepsils and intubation period were compared between groups by an was about one-third compared with the placebo group (P ϭ independent 2-sample t test. A nomogram based on estimated 0.003). Likewise, hoarseness was reported less frequently in standardized difference was used to calculate the required the Strepsils than placebo group (P ϭ 0.04). As shown in sample size. With a power of 80% and ␣ level of 0.05 for Figure 1, the incidence of both sore throat and hoarseness 2-tailed statistical analysis, and estimated incidence of sore decreased at the 24-hour assessment after surgery, but the throat of 0.351,9 and 0.10 in the 2 groups, a sample size of 70 differences between the 2 groups remained statistically sig- patients for each group was calculated as being appropriate. nificant (P ϭ 0.04). Further analysis showed that the mean Ϯ Correlation coefficient between grade of laryngoscopic view SD severity scores of early sore throat (0.52 Ϯ 0.85 vs 0.20 Ϯ and sore throat/hoarseness score in either group was assessed 0.57), late sore throat (0.22 Ϯ 0.50 vs 0.08 Ϯ 0.32), early with Spearman analysis. Statistical analysis was performed hoarseness (0.36 Ϯ 0.65 vs 0.16 Ϯ 0.47), and late hoarseness with SPSS version 11.0 software (SPSS, Chicago, IL). (0.22 Ϯ 0.45 vs 0.08 Ϯ 0.27) in the placebo group were significantly higher than in the Strepsils group (P Ͻ 0.05). RESULTS None of the patients complained of grade 3 hoarseness Of 150 included participants, 5 patients were excluded be- (unable to speak because of hoarseness). Only 1 patient in cause of unfavorable Ramsay Sedation Scores in the recovery the Strepsils group and 3 patients in the placebo group had

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 893 Strepsils௡ Reduces Sore Throat and Hoarseness severe sore throat. Pearson correlation coefficient analyses with inappropriate sedation scores, fentanyl or propofol still between grade of laryngoscopic view and early sore throat circulating in the bloodstream immediately after surgery could (r ϭ 0.13, P Ͼ 0.05), 24-hour sore throat (r ϭ 0.10, P Ͼ 0.05), have affected their ability to accurately perceive their throat early hoarseness (r ϭ 0.16, P Ͼ 0.05), and 24-hour hoarse- symptoms. However, standardized protocols for intubation and ness (r ϭ 0.10, P Ͼ 0.05) showed no statistically significant extubation of patients along with similar results in the later results. Reanalysis of the data after the exclusion of 6 (24-hour) evaluation of patients support our initial findings. patients with perioperative steroid use yielded similar One limitation of this study is that we did not evaluate patient results in all sets of analysis (data not shown). satisfaction, which could be an important indicator of the efficacy of our intervention. Finally, we did not evaluate the best timing DISCUSSION for the administration of Strepsils and the effects of repeated use, Sore throat and hoarseness are minor but common postop- especially during the day after surgery. erative complaints with an estimated incidence of 14.4% to In conclusion, the perioperative use of Strepsils lozenges 1,9 1,10 90% for sore throat and 10% to 50.1% for hoarseness. may help eliminate sore throat and hoarseness after inpa- In this study, the overall incidence of sore throat and tient surgery. The best timing of lozenge administration hoarseness in the immediate postoperative period was and the efficacy of repeated use in patients with remaining 20.8% and 18.6%, respectively. Both complications were symptoms should be evaluated in further studies. shown to be significantly reduced with the use of Strepsils. Likewise, 1-day follow-up of patients confirmed the effi- REFERENCES cacy of Strepsils tablets for the prevention of postoperative 1. Maruyama K, Sakai H, Miyazawa H, Toda N, Iinuma Y, sore throat and hoarseness. Mochizuki N, Hara K, Otagiri T. Sore throat and hoarseness Several causal factors for sore throat and hoarseness after total intravenous anaesthesia. Br J Anaesth 2004;92:541–3 after intubation have been reported, including sex, large 2. Higgins PP, Chung F, Mezei G. Postoperative sore throat after tracheal tube size, cuff design,11,12 increased intracuff pres- ambulatory surgery. Br J Anaesth 2002;58:582–4 sure by nitrous oxide,13 use of succinylcholine, and pro- 3. Maruyama K, Yamada T, Hara K. Effect of clonidine premedi- 14 cation on postoperative sore throat and hoarseness after total longed laryngoscopy. In this study, patient characteristics intravenous anesthesia. J Anesth 2006;20:327–30 and duration of laryngoscopy and surgery were compa- 4. Kazemi A, Amini A. The effect of betamethasone gel in rable between the 2 groups. We excluded the potential reducing sore throat, cough, and hoarseness after laryngo- effects of difficult intubation on postoperative throat com- tracheal intubation. Middle East J Anesthesiol 2007;19:197–204 plications in our series and had a standardized protocol for 5. Kyokong O, Charuluxananan S, Muangmingsuk V, Rodanant induction, tracheal intubation, and extubation of patients. O, Subornsug K, Punyasang W. Efficacy of chamomile-extract spray for prevention of post-operative sore throat. J Med Assoc Thus, it seems reasonable to conclude that the observed Thai 2002;85(suppl 1):S180–5 difference in the outcome measures between the 2 groups 6. Ga´spa´r L, Turi J, Toth BZ, Suri C, Vago P. [The use of benzyl alcohol can be safely attributed to the favorable effects of Strepsils. and amyl-m-cresol (Strepsils) in the oral cavity: review of the Prior studies have demonstrated that mucosal damage literature and first clinical experiences]. Fogorv Sz 1998;91:143–50 can occur even after uneventful intubation for routine 7. Ga´spa´r L, Szmrtyka A, Turi J, To´th BZ, Suri C, Va´go´ P, Sefer A, al 12 Haj C. [Clinical experience with the use of benzylalcohol and surgery. Because of the potential role of inflammation in amyl-m-cresol (Strepsils) in stomatological diseases]. Fogorv Sz the causation of tracheal morbidities, earlier investigators 2000;93:83–90 have suggested the use of inhaled and topical steroids.15–17 8. Ramsay MA, Savege TM, Simpson BR, Goodwin R. Controlled Ga´spa´r et al.6 used Strepsils for the treatment of 22 patients sedation with alphaxalone-alphadolone. Br Med J 1974;22:656–65 with oral inflammatory diseases and preoperatively in 20 9. Lev R, Rosen P. Prophylactic lidocaine use preintubation: a review. J Emerg Med 1994;4:499–506 oral surgery cases. In this preliminary study, they reported 10. Christensen AM, Willemoes-Larsen H, Lundby L, Jakobsen that Strepsils may be effective in the prophylaxis and KB. Postoperative throat complaints after tracheal intubation. treatment of oral inflammation. In a complementary survey of Br J Anaesth 1994;73:786–7 272 patients with either oral inflammatory diseases or treated 11. Jensen PJ, Hommelgaard P, Sondergaard P, Eriksen S. Sore prophylactically for possible oral inflammation or infection, throat after operation: influence of tracheal intubation, intra- Ga´spa´r et al.7 found that healing was shortened by 30% and cuff pressure and type of cuff. Br J Anaesth 1982;54:453–7 12. McHardy FE, Chung F. Postoperative sore throat: cause, pre- pain and functional assessment were improved by 30% in vention and treatment. Anaesthesia 1999;54:444–53 patients treated with Strepsils compared with control. Strep- 13. Combes X, Schauvliege F, Peyrouset O, Motamed C, Kirov K, sils tablets were well tolerated by all patients. Taken together, Dhonneur G, Duvaldestin P. Intracuff pressure and tracheal these data suggest that Strepsils is an effective antiinflamma- morbidity: influence of filling with saline during nitrous oxide tory choice for mucosal damage in the orotracheal cavity. In anesthesia. Anesthesiology 2001;95:1120–4 14. Higgins PP, Chung F, Mezei G. Postoperative sore throat after this study, we demonstrated that postoperative sore throat ambulatory surgery. Br J Anaesth 2002;88:582–4 and hoarseness were reduced by using Strepsils lozenges. 15. Sumathi PA, Shenoy T, Ambareesha M, Krishna HM. Controlled Although we did not evaluate the effect of Strepsils on the comparison between betamethasone gel and lidocaine jelly ap- potential role of the inflammatory process in the generation of plied over tracheal tube to reduce postoperative sore throat, these adverse effects, it is likely that this is its mechanism of cough, and hoarseness of voice. Br J Anaesth 2008;100:215–8 action based on the above literature. 16. Ayoub MC, Ghobashy A, McGrimley L, Koch ME, Qadir S, Silverman DG. Wide spread application of topical steroids to decrease sore throat, hoarseness and cough after tracheal Study Limitations intubation. Anesth Analg 1998;87:714–6 Responsiveness of patients in the immediate postoperative 17. el-Hakim M. Beclomethasone prevents postoperative sore period may be questioned. Although we excluded patients throat. Acta Anaesthesiol Scand 1993;37:250–2

894 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Inhaled Fluticasone Propionate Reduces Postoperative Sore Throat, Cough, and Hoarseness

Nasrin Faridi Tazeh-kand, MD, Bita Eslami, MPH, and Khadijeh Mohammadian, RN

BACKGROUND: Sore throat is a common complication after surgery. Postoperative cough and hoarseness can also be distressing to patients. We sought to determine the effect of an inhaler steroid on sore throat, cough, and hoarseness during the first 24 hours of the postoperative period. METHODS: We enrolled 120 women with ASA physical status I or II and term singleton pregnancy who were scheduled for elective cesarean delivery under general anesthesia. Patients were randomized into 2 groups: in the sitting position, group F patients received 500 ␮g inhaled fluticasone propionate via a spacer device during 2 deep inspirations, after arrival in the operating room, and group C had no treatment. The patients were interviewed by a blinded investigator for postoperative sore throat, cough, and hoarseness at 1 and 24 hours after surgery. RESULTS: There were no significant differences in age, height, weight, body mass index, duration of surgery, intubation, and grade of laryngeal exposure between the 2 groups. The incidence of sore throat, cough, and hoarseness was significantly lower in group F (3.33%, 3.33%, and 3.33%) compared with the control group (36.67%, 18.33%, and 35%) (P Ͻ 0.05 for all comparisons), not only in the first postoperative hour but also 24 hours after surgery (13.33%, 13.33%, and 25% in group F vs 40%, 41.67%, and 50% in the control group). The incidence of moderate and severe hoarseness in group F at the first hour was significantly less than the control group (P Ͻ 0.05). CONCLUSIONS: Inhaled fluticasone propionate decreases the incidence and severity of postoperative sore throat, cough, and hoarseness in patients undergoing cesarean delivery under general anesthesia. (Anesth Analg 2010;111:895–8)

ore throat is a common complication after surgery. Ͼ115 kg, diabetes mellitus, pregnancy-induced hyperten- Postoperative cough and hoarseness also can be dis- sion, Ͼ2 attempts at intubation, and patients receiving steroid Stressing to patients and affect patient satisfaction. therapy were excluded from the study. Patients were random- Many factors can contribute to postoperative sore throat, ized into the 2 groups with the help of a computer-generated and the incidence has been found to vary with the method table of random numbers. 1 of airway management. The incidence is the highest after In the sitting position, group F received 500 ␮g inhaled tracheal intubation (45.4%), whereas after laryngeal mask fluticasone propionate (Flixotide™ Evohaler™ 250 ␮g, Glaxo airway use the incidence is lower (17.5%) and much less Wellcome Production, Evreux, France) via a spacer device (3.3%) when a facemask is used for the maintenance of 1–3 during 2 deep inspirations, after their arrival in the oper- anesthesia. Female sex increases the incidence and sever- ating room, and group control (C) received no treatment. In ity of postoperative sore throat.4 group F, the inhaler was given by an anesthesia nurse. All We sought to determine the effect of an inhaler steroid patients were given the impression they would receive on sore throat, cough, and hoarseness during the first 24 medication to reduce the incidence of sore throat. They postoperative hours. were told that some of them would be randomized to receive the medication by inhalation (F group) and some by METHODS injection (C group). Those randomized to group C did not After receiving hospital ethics committee approval and receive any active drug. informed consent from subjects, we enrolled 120 women Standard monitoring included noninvasive arterial with ASA physical status I or II and term singleton preg- blood pressure, pulse oximetry, electrocardiogram, and nancies who were scheduled for elective cesarean delivery end-tidal carbon dioxide. Uterine displacement was under general anesthesia. The study was conducted in a achieved by tilting the operating table to the left. We prospective, randomized, and single-blinded manner, from inserted a wide-bore IV catheter into a forearm vein and August to December 2008. Patients with a history of started a slow infusion of Ringer solution. In all patients, perioperative sore throat and asthma, Mallampati grade after 4 minutes of administration of oxygen, rapid sequence Ͼ2, recent nonsteroidal antiinflammatory drug use, weight induction with cricoid pressure was achieved using thiopental 5 mg/kg and succinylcholine 1.5 mg/kg. The tra- From the Department of Anesthesiology, Roointan-Arash Hospital, Tehran chea was intubated with a soft seal cuffed sterile polyvinyl University of Medical Sciences, Tehran, Iran. chloride endotracheal tube with a standard cuff (Supa Accepted for publication October 26, 2009. Medical Devices, Tehran, Iran) and an internal diameter of Address correspondence and reprint requests to Nasrin Faridi Tazeh-kand, 7 mm. The tracheal tube cuff was inflated until no air MD, Roointan-Arash Hospital, Rashid Ave., Tehranpars, Tehran, Iran. Address e-mail to [email protected]. leakage could be heard with a peak airway pressure at 20 Copyright © 2010 International Anesthesia Research Society cm H2O. The degree of laryngeal exposure was ranked DOI: 10.1213/ANE.0b013e3181c8a5a2 from complete visualization of the vocal cords (grade I) to

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 895 Postoperative Sore Throat and Fluticasone

Table 1. Scoring System for Sore Throat, Cough, Table 2. Characteristics of Study Population and Hoarseness Fluticasone Score propionate Control Sore throat group group ؍ ؍ 0 No sore throat (n 60) (n 60) P 1 Mild (less than a common cold) Age (y) 26.32 Ϯ 5.18 26.70 Ϯ 5.26 NS 2 Moderate (similar to a common cold) Height (cm) 162.03 Ϯ 4.90 161.39 Ϯ 4.99 NS 3 Severe (more than a common cold) Weight (kg) 79.18 Ϯ 10.91 76.79 Ϯ 12.16 NS Cough Body mass index 30.09 Ϯ 3.41 29.42 Ϯ 4.13 NS 0 No cough (kg/m2) 1 Mild (less than a common cold) Duration of surgery 39.92 Ϯ 8.36 40.67 Ϯ 11.55 NS 2 Moderate (similar to a common cold) (min) 3 Severe (more than a common cold) Duration of tracheal 53.83 Ϯ 8.65 54 Ϯ 11.96 NS Hoarseness intubation (min) 0 No hoarseness Grade of laryngeal 1 Mild (no hoarseness at the time of interview but had exposure it previously) I 43 (71.67) 43 (71.67) 2 Moderate (is only perceived by the patient) II 15 (25) 17 (28.33) 3 Severe (recognizable at the time of interview) III 2 (3.33) 0 (0) V 0 (0) 0 (0) NS NS ϭ not signiﬁcant. a partial view of the vocal cords (grade II), epiglottis only (grade III), and the inability to view even the epiglottis in group F had moderate or severe (grades 2 and 3) sore (grade IV). throat, cough, or hoarseness. However, in group C, mod- Anesthesia was maintained with 50% nitrous oxide in erate and severe symptoms were reported in 3 patients for oxygen and 0.5 minimum alveolar concentration of halo- sore throat, 3 for cough, and 5 for hoarseness. The incidence thane. Atracurium was given as required for further muscle of combined moderate and severe hoarseness in group F in relaxation. After delivery and clamping of the umbilical the first hour was significantly lower than in group C (P Ͻ cord, we administered 2 ␮g/kg fentanyl and 0.02 mg/kg 0.05). Meanwhile, the evaluation of complications 24 hours midazolam and 10 IU oxytocin IV. At the end of surgery, after surgery showed that in group F, only 2 patients had oxygen 100% was administered, and residual neuromuscu- moderate or severe sore throat, and 2 patients had moder- lar block was antagonized using neostigmine and atropine. ate or severe hoarseness. None of the patients in group F Oral suctioning was done just before extubation only. The had moderate or severe cough after 24 hours. However, trachea was extubated after deflating the cuff when the moderate and severe complications were common in group patient was fully awake and was following commands. All C and when combined, moderate and severe complications patients received oxygen by a facemask after surgery. The were significantly more frequent than in group F (P Ͻ 0.05). anesthesiologist intubating and providing care did not know whether a patient had been allocated to either the F DISCUSSION or C group. The patients were interviewed by a blinded We found that the incidence of postoperative sore throat, investigator for postoperative sore throat, cough, and cough, and hoarseness was significantly less when inhaled hoarseness at 1 and 24 hours after surgery, using the fluticasone (500 ␮g) was administered compared with no questionnaire based on the scoring system in Table 1. treatment. Some studies found the incidence of postopera- Based on the results of a pilot study with 30 patients in tive sore throat, cough, and hoarseness to be as high as 6.6% each group that showed an incidence of problems (sore to 90%. The results of our study showed that the incidence throat, cough, and hoarseness) of 56% in group F and 83% of these problems in group C was higher than in group F in group C, we calculated that 60 patients would be (40% vs 13.3% for cough; 41.67% vs 13.3% for sore throat; required in each group to detect a difference in the inci- and 50% vs 25% for hoarseness) after 24 hours. Many dence with a power of 90% and ␣ ϭ 0.05 by using the Epi factors including airway management, female sex, younger Info Web site (www.cdc.gov/epiinfo/). Statistical analysis patients, gynecological procedure, and succinylcholine ad- was performed with JMP software (version 4, SAS Institute, ministration predict postoperative sore throat.3 All patients Cary, NC). Statistical significance for differences was tested in this trial were young females and were candidates for by Student t test and ␹2 test when appropriate. A P value cesarean delivery. Therefore, age and sex were eliminated Ͻ0.05 was considered statistically significant. as possible confounding factors. Researchers recognizing the potential role of inflammation RESULTS in these postoperative airway sequelae have described the use As shown in Table 2, there were no significant differences of inhaled and topical steroids.5–7 Stride8 concluded that 1% in age, height, weight, body mass index, duration of hydrocortisone water-soluble cream was ineffective for surgery, intubation, and grade of laryngeal exposure be- reducing the incidence of postoperative sore throat. tween the 2 groups. The total incidence of sore throat, Sumathi et al.9 showed that the widespread application of cough, and hoarseness was significantly lower in group F betamethasone gel on the tracheal tube decreased the compared with group C (P Ͻ 0.05) not only in the first incidence and severity of postoperative sore throat, cough, postoperative hour but also 24 hours after surgery (Table and hoarseness. The differences in the findings between 3). During the first postoperative hour, none of the patients these 2 studies may be attributable to the fact that Stride

896 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 3. Postoperative Complications by Grade After 1 and 24 h After 1 h After 24 h Fluticasone Control Fluticasone Control propionate group group P propionate group group P Sore throat 1 2 (3.33) 19 (31.67) 6 (10) 15 (25) 2 0 (0) 2 (3.33) 2 (3.33) 7 (11.67) 3 0 (0) 1 (1.67) 0 (0) 2 (3.33) Total 2 (3.33) 22 (36.67) Ͻ0.0001 8 (13.33) 24 (40) 0.001 Cough 1 2 (3.33) 8 (13.33) 8 (13.33) 14 (23.33) 2 0 (0) 3 (5) 0 (0) 9 (15) 3 0 (0) 0 (0) 0 (0) 2 (3.33) Total 2 (3.33) 11 (18.33) 0.008 8 (13.33) 25 (41.67) 0.005 Hoarseness 1 2 (3.33) 16 (26.67) 13 (21.67) 23 (38.33) 2 0 (0) 4 (6.67) 2 (3.33) 4 (6.67) 3 0 (0) 1 (1.67) 0 (0) 3 (5) Total 2 (3.33) 21(35) Ͻ0.0001 15 (25) 30 (50) 0.005

lubricated the tube only to the 5-cm mark, whereas Sumathi therefore unlikely that a single dose of fluticasone will et al. lubricated the tube to the 15-cm mark. Thus, because affect the fetus. In this study, we used fluticasone propi- of more widespread application of steroid gel to the tube, onate in healthy women and did not observe any side more gel came in contact with the posterior pharyngeal effects. wall, vocal cords, and trachea and was not just confined to A limitation of our study was the infrequency of data the tip and cuff of the tracheal tube. collection, which we could also have performed at 6 and 12 Our study showed that inhaled steroid could be simi- hours. Also, we did not use an inert inhaler in the control larly effective in decreasing postoperative sore throat, group, leading to potential bias. cough, and hoarseness, and is similar to the study by In conclusion, inhaled fluticasone propionate (500 ␮g) Sumathi et al.9 In the 2 studies, the patient populations and decreases the incidence and severity of postoperative sore the type of surgery were different. Their patients were throat, cough, and hoarseness in patients undergoing cesar- either sex, aged between 18 and 50 years, and undergoing ean delivery under general anesthesia. elective surgery. Inhaled fluticasone delivers the drug in smaller doses REFERENCES and in a shorter time to the patient’s airway compared with 1. McHardy FE, Chung F. Postoperative sore throat: cause, pre- widespread lubrication of the tube with betamethasone, vention and treatment. Anaesthesia 1999;54:444–53 2. Joshi GP, Inagaki Y, White PF, Taylor-Kennedy L, Wat LI, which may increase the dose of drug that comes in contact Gevirtz C, McCraney JM, McCulloch DA. Use of laryngeal with the mucosa of the oropharynx, larynx, and trachea, mask airway as an alternative to the tracheal tube during resulting in higher systemic absorption and a possible ambulatory anesthesia. Anesth Analg 1997;85:573–7 aggravation of local subtle infection, especially in pregnant 3. Higgins PP, Chung F, Mezei G. Postoperative sore throat after ambulatory surgery. Br J Anaesth 2002;88:582–4 patients. 4. Maruyama K, Sakai H, Miyazawa H, Tida N, Iinuma Y, Coughing, wheezing, and shortness of breath are symp- Mochizuki N, Hara K, Otagiri T. Sore throat and hoarseness toms of asthma. Asthma treatment includes inhaled bron- after total intravenous anaesthesia. Br J Anaesth 2004;92:541–3 chodilators, which reverse airflow obstruction, and inhaled 5. Ayoub MC, Ghobashy A, McGrimely L, Koch ME, Gadir S, corticosteroids to prevent asthma exacerbations by damp- Silverman DG. Wide spread application of topical steroids to decrease sore throat, hoarseness and cough after tracheal ing the inflammatory processes that underlie asthma at- intubation. Anesth Analg 1998;87:714–6 tacks. Inhaled fluticasone propionate is a relatively new 6. Selvaraj T, Dhanpal R. Evaluation of the application of topical inhaled corticosteroid for the treatment of asthma. steroids on the endotracheal tube in decreasing postoperative In many studies, fluticasone propionate was used for sore throat. J Anaesthesiol Clin Pharmacol 2002;18:167–70 7. El-Hakim M. Beclomethasone prevents postoperative sore improved outcomes in children or adults who were at risk throat. Acta Anaesthesiol Scand 1993;37:250–2 of asthma. Treatment with fluticasone propionate or a 8. Stride PC. Postoperative sore throat: topical hydrocortisone. combination of this drug with a long-acting ␤-2 agonist Anaesthesia 1990;45:968–71 such as salmeterol was significantly effective in asthmatic 9. Sumathi PA, Shenoy T, Ambareesha M, Krishna HM. Con- patients.10–12 A study of the safety of intranasal corticoste- trolled comparison between betamethasone gel and lidocaine jelly applied over tracheal tube to reduce postoperative sore roids such as fluticasone propionate did not identify any throat, cough, and hoarseness of voice. Br J Anaesth 2008; systemic adverse events, which suggests that this drug can 100:215–8 be safely administered.13 10. Bacharier LB, Guilbert TW, Zeiger RS, Strunk RC, Morgan WJ, Some studies evaluated the effect of treatment with Lemanske RF Jr, Moss M, Szefler SJ, Krawiec M, Boehmer S, Mauger D, Taussig LM, Martinez FD. Patient characteristics fluticasone propionate nasal spray in rhinitis and inhaled associated with improved outcomes with use of an inhaled fluticasone during pregnancy. The authors did not report corticosteroid in preschool children at risk for asthma. J any adverse effects on maternal and fetal health.14–16 It is Allergy Clin Immunol 2009;123:1077–82

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 897 Postoperative Sore Throat and Fluticasone

11. de Blic J, Ogorodova L, Klink R, Sidorenko I, Valiulis A, Hofman 14. Choi S, Han JY, Kim MY, Vela´zques-Armenta EY, Nava-Ocampo J, Bennedbaek O, Anderton S, Attali V, Desfougeres JL, Poterre AA. Pregnancy outcomes in women using inhaled fluticasone M. Salmeterol/fluticasone propionate vs. double dose flutica- during pregnancy: a case series. Allergol Immunopathol 2007; sone propionate on lung function and asthma control in 35:239–42 children. Pediatr Allergy Immunol. 2009;20:763–71 15. Rahimi R, Nikfar S, Abdollahi M. Meta-analysis finds use of 12. Markham A, Jarvis B. Inhaled salmeterol/fluticasone propi- inhaled corticosteroids during pregnancy safe: a systematic onate combination: a review of its use in persistent asthma. meta-analysis review. Hum Exp Toxicol 2006;25:447–52 Drugs 2000;60:1207–33 16. Ellegård EK, Hellgren M, Karlsson NG. Fluticasone propionate 13. Demoly P. Safety of intranasal corticosteroids in acute rhino- aqueous nasal spray in pregnancy rhinitis. Clin Otolaryngol sinusitis. Am J Otolaryngol 2008;29:403–13 Allied Sci 2001;26:394–400

898 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA International Society for Anaesthetic Pharmacology

Anesthetic Pharmacology and Preclinical Pharmacology Section Editor: Marcel E. Durieux Clinical Pharmacology Section Editor: Tony Gin

Repinotan, a Selective 5-HT1A-R-Agonist, Antagonizes Morphine-Induced Ventilatory Depression in Anesthetized Rats

U. Guenther, MD,* H. Wrigge, PhD,* N. Theuerkauf, MD,* M. F. Boettcher, MD,† G. Wensing, MD,† J. Zinserling, PhD,* C. Putensen, PhD,* and A. Hoeft, PhD*

BACKGROUND: Spontaneous breathing during mechanical ventilation improves arterial oxygenation and cardiovascular function, but is depressed by opioids during critical care. Opioid-induced ventilatory depression was shown to be counteracted in anesthetized rats by serotonin(1A)-

receptor (5-HT1A-R)-agonist 8-OH-DPAT, which cannot be applied to humans. Repinotan hydrochloride is a selective 5-HT1A-R-agonist already investigated in humans, but the effects on ventilation and nociception are unknown. In this study, we sought to establish (a) the effects of repinotan on spontaneous breathing and nociception, and (b) the interaction with the standard opiate morphine. METHODS: The dose-dependent effects of repinotan, given alone or in combination with morphine, on spontaneous minute ventilation (MV) and nociceptive tail-ﬂick reﬂex latencies (TFLs) were measured simultaneously in spontaneously breathing anesthetized rats. An addi-

tional series with NaCl 0.9% and the 5-HT1A-R-antagonist WAY 100 135 served as controls. RESULTS: (a) Repinotan dose-dependently activated spontaneous breathing (MV, mean [95% conﬁdence interval]; 53% [29%–77%]) of pretreatment level) and suppressed nociception (TLF, 91% maximum possible effect [68%–114%]) with higher doses of repinotan (2–200 ␮g/kg). On the contrary, nociception was enhanced with a small dose of repinotan (0.2 ␮g/kg; TFL, Ϫ47% Ϫ maximum possible effect [ 95% to 2%]). Effects were prevented by 5-HT1A-antagonist WAY 100 135. (B) Morphine-induced depression of ventilation (MV, Ϫ72% [Ϫ100% to Ϫ44%]) was reversed by repinotan (20 ␮g/kg), which returned spontaneous ventilation to pretreatment levels (MV, 18% [Ϫ40% to 77%]). The morphine-induced complete depression of nociception was sustained throughout repinotan and NaCl 0.9% administration. Despite a mild decrease in mean arterial blood pressure, there were no serious cardiovascular side effects from repinotan.

CONCLUSIONS: The 5-HT1A-R-agonist repinotan activates spontaneous breathing in anesthetized rats even in morphine-induced ventilatory depression. The potency of 5-HT1A-R-agonists to stimulate spontaneous breathing and their antinociceptive effects should be researched further. (Anesth Analg 2010;111:901–7)

pioids are potent analgesics, but their clinical admin- has been demonstrated to overcome opioid-induced ventila- istration is limited by the intrinsic risk of fatal apnea. tory depression in anesthetized rats.5

Hence, pain therapy involving opioids must be bal- Nociception, another target of 5-HT1A-R-agonists, was O 6–8 1 anced against respiratory depression. The serotonin(1A)- reported either to be depressed or enhanced by 5-HT1A- R-agonists.9,10 Later, 5-HT -R-agonist F13640 was found receptor (5-HT1A-R)-agonist buspirone has been shown to 1A stimulate spontaneous breathing in cats2 and to overcome to exert a dual effect, hyperalgesic and analgesic, depend- 11 neurogenic breathing disturbances in humans.3,4 Another ing on plasma and brain concentrations. Recently, enhancement of nociceptive reflexes by small doses of 5-HT1A-R-agonist, 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT), a substance not approved for use in humans, 8-OH-DPAT and suppression by higher doses were confirmed in 2 different experimental models.12 Repinotan (R-(Ϫ)-2-{4-[(chroman-2-ylmethyl)-amino]- butyl}-1,1-dioxobenzo[d]-isothiazolone hydrochloride) is a From the *University Hospital of Bonn, Clinic of Anaesthesiology and 13,14 highly effective, selective, full 5-HT1A-R-agonist. Unlike Intensive Care Medicine, Bonn; and †Department of Pharmacological Re- search, Bayer Schering Pharma AG, Wuppertal, Germany. other 5-HT1A-R-agonists, repinotan is approved for IV use in Accepted for publication May 19, 2010. humans and has already undergone a series of clinical inves- Supported by Bayer Schering Pharma AG, Germany, and departmental tigations into the effects of neuroprotection after traumatic 15–18 funding. brain injury and stroke. However, its effects on nocicep- Address correspondence and reprint requests to Ulf Guenther, MD, Univer- tion and ventilation are not yet established. The aim of this sity Hospital of Bonn, Clinic of Anaesthesiology and Intensive Care Medi- study was to verify the effects of repinotan on spontaneous cine, Sigmund-Freud-Strasse 25, 53105 Bonn, Germany. Address e-mail to [email protected]. breathing and nociception simultaneously, and to determine Copyright © 2010 International Anesthesia Research Society the interaction with the standard opiate morphine on sponta- DOI: 10.1213/ANE.0b013e3181eac011 neous breathing and nociception in anesthetized rats. Two

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 901 Repinotan Antagonizes Ventilatory Depression

hypotheses were tested: (1) repinotan at a dose to stimulate A Repinotan spontaneous breathing does not enhance a nociceptive reflex, and (2) repinotan antagonizes morphine-induced depression of spontaneous breathing. Repinotan 0.02µg/kg WAY 100135 1µg/kg NaCl 0.9%

METHODS Repinotan 0.2µg/kg Repinotan 0.2µg/kg NaCl 0.9% Animals This study was performed with approval from the local Repinotan 2µg/kg Repinotan 2µg/kg NaCl 0.9% Institutional Animal Review Board for animal research and in accordance with the “Guide for the Care and Use of Labora- Repinotan 20µg/kg Repinotan 20µg/kg NaCl 0.9% tory Animals.” Animals were housed in standard laboratory conditions with a 12-hour light/dark schedule and free access Repinotan 200µg/kg Repinotan 200µg/kg NaCl 0.9% to food and water. Thirty-nine male Sprague-Dawley rats weighing 260 g (244–277 g) (mean, 95% confidence interval n=8 n=4 n=8 [CI]) were deeply anesthetized with sodium-pentobarbitone B Repinotan / Morphine (60 mg/kg) intraperitoneally and placed supine on a heating pad to maintain rectal temperature constantly at 37°C Ϯ 0.5°C. The right inguinal vessels were cannulated via a Morphine Morphine Morphine 12[6-18]mg/kg small surgical incision for continuous monitoring of 11[7-15]mg/kg 13[7-18]mg/kg arterial blood pressure and systemic administration of Repinotan 0.2µg/kg Repinotan 0.02µg/kg NaCl 0.9% study drugs. Anesthesia was maintained with sevoflurane, leveled at an inspiratory concentration of 1.5 to 2.5 Repinotan vol% to ensure immobility, stable spontaneous breath- 2µg/kg Repinotan 6µg/kg NaCl 0.9% ing, and detectable tail-flick reflex (TFR). Repinotan 20µg/kg Repinotan 60µg/kg NaCl 0.9% Measurements Animals breathed spontaneously via a tracheotomy tube Repinotan 200µg/kg NaCl 0.9% (inner diameter, 1.2 mm). The expired air was led through the flowhead (order number, MLT1L; ADInstruments n=8 n=5 n=6

GmbH, Spechbach, Germany) of a spirometer (ML141) Figure 1. A, Serotonin(1A)-receptor (5-HT1A-R)-agonist repinotan was connected to an A/D-interface (PowerLab 4/25®; all de- injected every 15 minutes at increasing doses. In a second series, vices from ADInstruments GmbH) to record respiratory the selective 5-HT1A-R-antagonist WAY 100 135 was given before repinotan. A third series involving only NaCl 0.9% served as controls. rate (RR) and tidal volume (Vt) by integration of ventila- B, Morphine was given at increments of 5 mg/kg until a target tory airflow over time. Minute ventilation (MV) was calcu- depression of respiratory rate of Ͼ50% was established. The mean lated as MV [mL/min] ϭ RR [1/min] ϫ Vt [mL]. required morphine dosing is given as mean (95% conﬁdence inter- The TFR was evoked by a 100-W light beam source val). Repinotan was given in 2 distinct series to cover all doses mounted 15 mm over the base of the tail to reach required to delineate the top of the bell-shaped dose-response curve and to maintain experiments at comparable length. maximum temperature within a second. The latency of the reflex response (TFR latency, TFL) was recorded with a strain gauge attached to the tail distal to the heating Repinotan spot. A shortened TFL indicates enhanced nociceptive Repinotan was injected IV every 15 minutes with doses responsiveness; an elongated TFL indicates depressed ranging from 0.02 through 200 ␮g/kg (Fig. 1A). The doses nociception. Heating was stopped when the tail flicked or were chosen because it was concluded from preliminary ␮ after a maximum heating time of 15 seconds (TFLoffset)to dose-finding experiments that the 20 g/kg dose was the prevent damage to the tail. TFLs were calculated as change most efficient to counteract opioid-induced ventilatory in percent of the maximum possible effect [% MPE] accord- depression. The wide range of dosage was necessary to 8 ϭ ϫ Ϫ ing to the formula : % MPE 100 [TFLtreatment verify whether repinotan also possesses dose-dependent ϫ Ϫ Ϫ1 TFLpretreatment] [TFLoffset TFLpretreatment] ). A 100% pro- and antinociceptive effects similar to the standard MPE means complete suppression of nociception. Three 5-HT1A-R-agonist, 8-OH-DPAT. The number of experi- sweeps were recorded and averaged. A blood pressure ments involving repinotan (n ϭ 8) was chosen based on our transducer, temperature probe, and strain-gauge trans- previous experience with 8-OH-DPAT, in which the small- ducer were also connected to the same A/D-interface such est effective dose increased spontaneous MV by 46% with a as the spirometer (PowerLab 4/25®; ADInstruments standard deviation of 35%. With an ␣ set at 0.05, the power GmbH). was calculated as 0.93 with n ϭ 8 experiments in the double-sided power analysis. Drug Administration Protocols Before the first drug administration, a series of 3 TFL Two different sets of experiments were performed: (a) the sweeps was averaged and taken as the pretreatment level. first set was aimed at determining the effects of repinotan Subsequent TFLs were taken 10 minutes after each drug on spontaneous breathing, and (b) the second set assessed administration. For control experiments, NaCl 0.9% was interactions of repinotan with the opiate morphine (see Fig. injected every 15 minutes instead of study drugs (n ϭ 8). In

1 for schematic overview). another series of 4 experiments, the selective 5-HT1A-R-

902 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA antagonist WAY 100 135 (1 mg/kg) was injected before A administration of repinotan (n ϭ 4).

Repinotan/Morphine Coadministration Morphine was injected at increments of 5 mg/kg until respiratory frequency was depressed to at least 50% of the pretreatment level. Thereafter, repinotan was added cumu- latively at the same doses as in the first set (Fig. 1B). Control experiments were again performed by injection of 200 ␮Lof ϭ NaCl 0.9% (n 6). After completion of this series, the 0.02 0.2 2 20 200 ventilatory dose-response curve had a bell shape. To delin- Repinotan (µg/kg) eate the top of the bell shape more precisely, 5 additional Repinotan experiments were performed with repinotan concentra- NaCl 0.9% tions that were between the initially intended measuring B points (Fig. 1). To verify integrity of the TFR at the end of experiments, naloxone (1 mg/kg) was given (not shown).

Statistical Analysis All data were tested for normal distribution (Kolmogorov- Smirnov test). Pretreatment levels of matched groups were compared with the Student t test. All ventilatory variables were calculated as change in percent of pretreatment level (% change). Results of experiments without morphine were compared with the values before the first administration of 0.02 0.2 2 20 200 study drugs (pretreatment level). Results of experiments Repinotan (µg/kg) involving morphine/repinotan coadministration were com- Repinotan pared with the variables obtained after morphine administra- NaCl 0.9% tion. The results of the ventilatory experiments were analyzed Figure 2. Effects of the serotonin(1A)-receptor (5-HT1A-R)-agonist ϭ by 1-way repeated-measures analysis of variance. Each repi- repinotan (n 8) on spontaneous minute ventilation (MV) and tail-flick reflex (TFR) latencies (TFLs). Control experiments were notan concentration was compared with the pretreatment performed with NaCl 0.9% (n ϭ 8). Values of MV are given as percent level (i.e., 0% change) by Dunnett multiple comparison post change from pretreatment level (% change), results of TFLs are test.19 TFLs were calculated as change in % MPE as stated shown as percent of maximum possible effect (% MPE; mean [95% above and also analyzed by comparison of each drug concen- confidence interval]). *P Ͻ 0.05, ***P Ͻ 0.001, compared with tration to pretreatment levels by Dunnett multiple compari- pretreatment level, 1-way repeated-measures analysis of variance. A, Repinotan dose dependently increased MV to a maximum of 53% son test. Data were processed with the Chart 4.0 and Scope 4.0 (10%–29%) above the pretreatment level with the 200 ␮g/kg dose. software package (ADInstruments GmbH); statistical analyses B, Repinotan effects on nociception were dose dependent. Initial were performed using Prism4® software package for Macin- shortening of TFL (Ϫ47 [Ϫ95 to 2], indicating enhanced nociceptive ␮ tosh (GraphPad Software Inc., San Diego, CA). Power analy- responsiveness) with small doses (0.2 g/kg) was followed by elongation of TFL to 91% MPE (68%–114% MPE) with the highest ses were done with the Simple Interactive Statistical Analysis dose (200 ␮g/kg), meaning that TFR was profoundly suppressed. (SISA) online software package (http://www.quantitativeskills. com/sisa/calculations/power.htm).

Drugs Hg (100–120 mm Hg) in the control group. The mean TFL Repinotan was provided by the manufacturer, Bayer was 7 seconds (5–9 seconds) in both groups. All data were Healthcare AG (Wuppertal, Germany). Morphine-sulfate normally distributed, and pretreatment values did not was purchased from Merck KG (Darmstadt, Germany), differ statistically. Repinotan dose dependently increased with permission from the German Institute for Pharmacy spontaneous MV (Fig. 2A), reaching the maximum effect ␮ and Medical Products. WAY 100 135 was obtained from (MV; 53% [29%–77%]) with the 200 g/kg dose. Tocris (Bristol, UK). Naloxone-HCl was purchased from Tail-Flick Reﬂex Latency Ratiopharm (Ulm, Germany). All compounds were diluted The repinotan effects on nociception were dose dependent: in isotonic saline at the respective concentrations; injection a small dose of repinotan (0.2 ␮g/kg) shortened TFL, volumes were 200 ␮L. whereas a high dose of repinotan (200 ␮g/kg) elongated TFL (Fig. 2B), meaning that nociception was enhanced with RESULTS small doses and suppressed with higher doses. Effects of Repinotan Spontaneous Ventilation Controls ϭ The mean pretreatment RR was 58 breaths/min (95% CI, The 5-HT1A-antagonist WAY 100 135 (1 mg/kg, n 4) did 46–70 breaths/min) in the repinotan group versus 61 not significantly alter MV itself, but prevented repinotan breaths/min (54–69 breaths/min) in the control group, and from activating spontaneous MV (data not shown). Injec- the mean arterial blood pressure (MAP) was 111 mm Hg tions of NaCl 0.9% had neither detectable effects on venti- (100–122 mm Hg) in the repinotan group versus 110 mm lation nor nociception.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 903 Repinotan Antagonizes Ventilatory Depression

the sustained action of morphine. Morphine depressed spon- A Pre-treatment Ϫ Ϫ Ϫ 10 taneous MV to 72% ( 100% to 44%) and always abolished the TFR with the first dosing increment (Fig. 4). Morphine mL/s TFR consumption did not differ between groups (Fig. 1). 0 2 s Spontaneous Breathing 10 s heat on heat off Repinotan (after morphine) dose dependently activated spontaneous breathing (Fig. 4), resulting in a maximum MV of 18% B Morphine 10 mg/kg (Ϫ40% to 77%) above the pretreatment level with the 20 10 ␮g/kg dose (P Ͻ 0.01, compared with morphine level). mL/s Further increases to doses Ͼ20 ␮g/kg returned MV to lower 0 levels, giving the dose-response curve a bell shape. 2 s 10 s Nociception heat on TFR was abolished with the first bolus of morphine (TFL, 100% MPE), and remained completely suppressed through- C Morphine 10 mg/kg, Repinotan 20 µg/kg out administrations of repinotan (n ϭ 13) and control drugs 10 (n ϭ 6; P Ͻ 0.001, compared with pretreatment levels). mL/s Naloxone-HCl at the end of experiments verified the integ- 0 rity of the TFR, because it returned with a TFL of 8% MPE 2 s (Ϫ18% to 33% MPE) (not shown). 10 s heat on Cardiovascular Side Effects Figure 3. Left panel, Expiratory airflow (mL/s). Animal was breathing Repinotan depressed MAP with higher doses (Table 1, n ϭ spontaneously through a tracheostomy tube. Right panel, Tracings of 8), but this did not have deleterious effects on the experi- the tail-flick reflex (TFR). Arrows indicate onset and offset of heat to the tail, and the circle marks the artifact evoked by the tail flick. A, ments. Likewise, although morphine (12 mg/kg [8–16 Pretreatment level. Left panel, Spontaneous respiratory frequency, mg/kg], n ϭ 13) markedly depressed MAP, repinotan did 51 breaths/min. Right panel, Heat to the tail evoked a tail flick with not further aggravate arterial hypotension, and neither did a latency (TFL) of 3.9 seconds. B, Morphine (10 mg/kg) completely NaCl 0.9% (Table 1). There were no serious cardiovascular depressed spontaneous ventilation; the last 3 breaths before apnea complications. are shown (left panel). The TFR was abolished (right panel). C, Repinotan (20 ␮g/kg) reestablished spontaneous ventilation (left panel), whereas TFR remained suppressed (right panel). DISCUSSION

This study was performed to clarify whether the 5-HT1A- R-agonist repinotan also antagonizes opioid-induced ven-

tilatory depression similar to other 5-HT1A-R-agonists. It was verified that (a) higher doses (2–200 ␮g/kg) of repinotan stimulated spontaneous breathing, small doses (0.2 ␮g/kg) enhanced nociception, and the highest dose (200 ␮g/kg) depressed nociception; and (b) morphine-induced

MV (%change [95% CI]) depression of spontaneous breathing was antagonized by n = 135885 88 5 higher doses of repinotan, whereas depression of nocicep- MO 0.02 0.2 2 20 200 Repinotan (µg/kg) tion persisted. Despite a mild depression of MAP, repinotan did not produce serious cardiovascular complications. Morphine/ Repinotan Morphine/ NaCl 0.9% These findings confirm previous work in which the 5-HT -R-agonist 8-OH-DPAT was shown to antagonize Figure 4. Effects of morphine (MO) and subsequent serotonin(1A)- 1A receptor (5-HT -R)-agonist repinotan on spontaneous minute venti- an opioid-induced ventilatory depression without impair- 1A 12 lation (MV). NaCl 0.9% served as controls (n ϭ 6). Values are given ing antinociception. 8-OH-DPAT, however, is not ap- as mean (95% conﬁdence interval) and percent change from pre- proved for human use. Buspirone, the only commercially treatment level (% change). *P Ͻ 0.05, **P Ͻ 0.01, compared with available 5-HT1A-R-agonist for use in humans, has been morphine levels, 1-way repeated-measures analysis of variance. 3 Morphine (11 mg/kg [8–16 mg/kg]) depressed MV to Ϫ72% shown to stabilize apneustic breathing disturbances. Bu- (Ϫ100% to Ϫ44%) of pretreatment level. Repinotan dose depen- spirone, being only a partial 5-HT1A-R-agonist, failed to dently activated MV to a maximum of 18% (Ϫ40% to 77%). Repino- counteract a morphine-induced ventilatory depression in tan dosage Ͼ20 ␮g/kg re-decreased MV, giving the dose-response healthy volunteers,20 nor did it cause antinociceptive ef- curve a bell shape (see Discussion). fects in healthy volunteers.21 In a direct comparison with 8-OH-DPAT, only a weak ventilatory stimulation in coadministration with fentanyl was found for buspirone in an in Repinotan/Morphine Coadministration situ perfused brainstem–spinal cord preparation, whereas Figure 3 shows a representative experiment on the antago- 8-OH-DPAT proved to be an effective ventilatory stimulant.22 nization of morphine-induced ventilatory depression, dur- Both enhancement and depression of nociception by ing which nociceptive TFR remains completely depressed by 5-HT1A-R-agonists, given alone or in combination with

904 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Effects of Morphine and Repinotan on Mean Arterial Blood Pressure Repinotan NaCl 0.9% Mean (95% CI) P value Mean (95% CI) P value Repinotan (% change of pretreatment) Repinotan 0.02 ␮g/kg 10 (4–17) Ͻ0.05 0 (Ϫ13 to 13) NS Repinotan 0.2 ␮g/kg 4 (Ϫ6 to 14) NS Ϫ3(Ϫ21 to 16) NS Repinotan 2 ␮g/kg 1 (Ϫ7to8) NS Ϫ5(Ϫ25 to 15) NS Repinotan 20 ␮g/kg Ϫ17 (Ϫ29 to Ϫ4) Ͻ0.001 Ϫ10 (Ϫ25 to 10) NS Repinotan 200 ␮g/kg Ϫ19 (Ϫ29 to Ϫ9) Ͻ0.001 Ϫ7(Ϫ21 to 10) NS Morphine, repinotan (% change of pretreatment) Morphine Ϫ30 (Ϫ46 to Ϫ14) Ͻ0.001 Ϫ21 (Ϫ54 to Ϫ13) Ͻ0.05 Repinotan 0.02 ␮g/kg Ϫ16 (Ϫ31 to Ϫ2) NS Ϫ21 (Ϫ54 to Ϫ11) NS Repinotan 0.2 ␮g/kg Ϫ24 (Ϫ40 to Ϫ8) NS Ϫ22 (Ϫ50 to Ϫ6) NS Repinotan 2 ␮g/kg Ϫ40 (Ϫ55 to Ϫ25) NS Ϫ21 (Ϫ53 to Ϫ11) NS Repinotan 6 ␮g/kg Ϫ28 (Ϫ50 to Ϫ5) NS Repinotan 20 ␮g/kg Ϫ30 (Ϫ46 to Ϫ14) NS Ϫ19 (Ϫ37 to 0) NS Repinotan 60 ␮g/kg Ϫ21 (Ϫ60 to Ϫ17) NS Repinotan 200 ␮g/kg Ϫ25 (Ϫ64 to 15) NS Ϫ19 (Ϫ49 to Ϫ11) NS CI ϭ conﬁdence interval; NS ϭ not signiﬁcant; MAP ϭ mean arterial blood pressure. MAP is given as mean (95% CI). Repinotan depressed MAP with higher doses (20, 200 ␮g/kg, n ϭ 8, repeated-measures analysis of variance, P Ͻ 0.001). Morphine (12 mg/kg ͓8–16 mg/kg͔) markedly depressed MAP (n ϭ 13, P Ͻ 0.001, compared with pretreatment level). Subsequent repinotan did not further aggravate hypotension. There were no serious cardiovascular complications.

8,9,23–26 opioids, have been variously reported. More re- 5-HT1A-R, once the administered dose was high enough to cently, the highly selective 5-HT1A-R-agonist F13640 was establish sufficient brain tissue concentrations. The under- reported to induce both hyperalgesia and/or analgesia lying mechanisms of hormetic dose responses clearly de- depending on the blood and brain concentration time serve further research.30,35 course.11 Most notably, F13640 was also shown to alleviate The cardiovascular depression after repinotan adminis- opioid-induced hyperallodynia and neuropathic pain in tration was much less severe than that by 8-OH-DPAT. In rats.27,28 The dose-dependent pro- and antinociceptive ef- previous work, we reported severe, occasionally fatal, fects of repinotan found in this study contribute to recon- cardiocirculatory depression with the highest dose of ciling the past contradictory findings, which were at least in 8-OH-DPAT (100 ␮g/kg) in anesthetized rats.12 Others saw part attributable to different experimental models, drug that 8-OH-DPAT prevented arterial hypotension induced administration routes, and dosing ranges. by the short-acting opioid remifentanil in conscious, non- 36 13 The dose-response curve of 5-HT1A-R stimulation of spon- anesthetized rats. Unlike repinotan, 8-OH-DPAT also 37 taneous breathing after morphine-induced ventilatory depres- stimulates 5-HT7-R, which are critical for activation of sion is inversely U shaped or “bell shaped,” meaning that cardiac vagal input.38 For instance, blockade of central 29 stimulatory effects subsided with high concentrations. Also, 5-HT7-R attenuates the bradycardia and pressor response repinotan produced a combination of low-dose stimulation of to both chemoreflex activation (induced by intracisternal the TFR followed by high-dose inhibition. This dose-response injection of potassium cyanide) and baroreflex activation 30 39 characteristic is generally referred to as “hormesis.” More (induced by IV phenylephrine). Activation of 5-HT7-R in than 30 receptor systems, including opioid and adrenergic turn might add to the depression of MAP seen in this study receptors, were identified to have hormetic dose responses, after morphine administration (Table 1), which was likely and the serotonin (5-HT) receptor system is among them.31 induced by peripheral vasodilation.40 Furthermore, it has

The neuroprotective effects of 5-HT1A-R-agonists have been been shown in anesthetized animals that the 5-HT1A- shown to have bell-shaped dose responses.32 It is proposed R-agonist F13640 markedly reduced the intraoperative re- that the basic biological principle behind this is that a mild quirement of the volatile anesthetic.36 The concentration of stress may promote function or action, and extreme stress the anesthetic was maintained constant in this study ac- may promote depressive or toxic action.30 cording to our protocol, which certainly contributed to

Although hormesis can be observed in a wide range of arterial hypotension caused by increasing Pco2 as the receptor systems and agents, there is no one-for-all molecu- consequence of hypoventilation. lar mechanism. The 5-HT1A-R are variously located and Some limitations of this study warrant comment. First, involved at different levels in the modulation of opioider- repinotan, a 5-HT1A-R-agonist, was developed as an antide- gic effects on nociceptive pathways.9,33 Activation of cen- pressant, and was also found to exert neuroprotective effects 15 16 tral 5-HT1A-R, for instance, has been shown to enhance on in vivo rats. Despite promising clinical data in humans, opioidergic inhibition of spinal reflexes,33 whereas systemic multicenter studies failed to show favorable effects on neuro-

(intraperitoneal, IV) administration of 5-HT1A-R-agonists logical outcomes in patients with stroke and traumatic brain produced both pro- and antinociceptive effects.7,11 Directly injury.17,18 Specific serotonergic complications of repinotan, applied onto the spinal cord, activation of 5-HT1A-R inhib- such as headache, nausea and vomiting, flush, tachycardia, ited nociceptive neural responses only with the highest and agitation, in humans were reported.17,41 These symptoms studied dose of 8-OH-DPAT.8,34 We speculate that IV may even be aggravated in coadministration with mor- repinotan overpowered possible pronociceptive effects phine.20 Specific serotonergic side effects were not seen in this

(mediated by spinal 5-HT1A-R) by actions via central study because of the experimental setup, but they could,

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 905 Repinotan Antagonizes Ventilatory Depression however, limit the clinical applicability of repinotan at least in 6. Bardin L, Tarayre JP, Malfetes N, Koek W, Colpaert FC. conscious patients. Profound, non-opioid analgesia produced by the high-efficacy 5-HT(1A) agonist F 13640 in the formalin model of tonic Second, it should be highlighted that the morphine nociceptive pain. Pharmacology 2003;67:182–94 concentrations in this investigation were much higher than 7. Bardin L, Tarayre JP, Koek W, Colpaert FC. In the formalin in studies aiming solely at nociception.42,43 This happened model of tonic nociceptive pain, 8-OH-DPAT produces because morphine dosage was targeted to produce venti- 5-HT1A receptor-mediated, behaviorally specific analgesia. latory depression, requiring higher dosing. The TFR was Eur J Pharmacol 2001;421:109–14 8. Nadeson R, Goodchild CS. Antinociceptive role of 5-HT1A always abolished with the first bolus of morphine and receptors in rat spinal cord. Br J Anaesth 2002;88:679–84 always before ventilatory depression occurred. The pos- 9. Clarke RW, Ogilvie J, Houghton AK. Enhancement and depression sible attenuation of morphine-induced antinociception by of spinal reflexes by 8-hydroxy-2-(di-n-propylamino) tetralin in the small doses of repinotan was presumably overpowered by decerebrated and spinalized rabbit: involvement of 5-HT1A- the strong morphine effect. We did not investigate to and non-5-HT1A-receptors. Br J Pharmacol 1997;122:631–8 10. Zhang YQ, Gao X, Ji GC, Huang YL, Wu GC, Zhao ZQ. determine whether small pronociceptive doses of repinotan Expression of 5-HT1A receptor mRNA in rat lumbar spinal would interfere with more moderate doses of morphine. dorsal horn neurons after peripheral inflammation. Pain This should be considered for further research. 2002;98:287–95 Third, the TFR is an acute, polysynaptic nociceptive 11. Bardin L, Assie MB, Pelissou M, Royer-Urios I, Newman- spinal reflex.44 Although pronociceptive effects were seen Tancredi A, Ribet JP, Sautel F, Koek W, Colpaert FC. Dual, hyperalgesic, and analgesic effects of the high-efficacy only with very small doses of repinotan, but not within the 5-hydroxytryptamine 1A (5-HT1A) agonist F 13640 [(3-chloro- dosing range to stimulate breathing, it is conceivable that 4-fluoro-phenyl)-[4-fluoro-4-{[(5-methyl-pyridin-2-ylmethyl)- small pronociceptive doses of repinotan could alleviate mor- amino]-methyl}piperidin-1-yl]methanone, fumaric acid salt]: phine antinociception. It was shown by others that 5-HT -R relationship with 5-HT1A receptor occupancy and kinetic param- 1A eters. J Pharmacol Exp Ther 2005;312:1034–42 influence nociceptive processing differently, according to the 45 12. Guenther U, Manzke T, Wrigge H, Dutschmann M, Zinserling type of noxious stimulus. Thus, nociceptive modalities other J, Putensen C, Hoeft A. The counteraction of opioid-induced than the one investigated here may be activated by small ventilatory depression by the serotonin 1A-agonist 8-OH- DPAT does not antagonize antinociception in rats in situ and doses of 5-HT1A-R-agonists, which may not be treated with opioids. This will be clarified by further investigations. in vivo. Anesth Analg 2009;108:1169–76 13. De Vry J, Schohe-Loop R, Heine HG, Greuel JM, Mauler F, In conclusion, this work confirmed that the 5-HT1A-R- Schmidt B, Sommermeyer H, Glaser T. Characterization of the agonist repinotan activates spontaneous breathing and aminomethylchroman derivative BAY ϫ 3702 as a highly suppresses nociception with higher doses, and that it potent 5-hydroxytryptamine1A receptor agonist. J Pharmacol antagonizes morphine-induced ventilatory depression in Exp Ther 1998;284:1082–94 anesthetized rats. Selective 5-HT -R-agonists thus are 14. Schwarz T, Beckermann B, Buehner K, Mauler F, Schuhmacher 1A J, Seidel D, Steinke W, Weinz C, Zimmerd D. Pharmacokinetics promising candidates for research into the stabilization of of repinotan in healthy and brain injured animals. Biopharm spontaneous breathing and pain therapy. Drug Dispos 2005;26:259–68 15. Harkany T, Mulder J, Horvath KM, Keijser J, van der AUTHOR CONTRIBUTIONS Meeberg EK, Nyakas C, Luiten PG. Oral post-lesion admin- UG, MFB, GW, HW, CP, and AH helped with study design; istration of 5-HT(1A) receptor agonist repinotan hydrochlo- ϫ UG, NT, JZ, and GW helped with study conduction; UG, NT, ride (BAY 3702) attenuates NMDA-induced delayed neuronal death in rat magnocellular nucleus basalis. Neu- and JZ helped with data collection; UG, NT, and JZ helped roscience 2001;108:629–42 with data analysis; and UG, HW, MFB, NT, CP, and AH helped 16. Ohman J, Braakman R, Legout V. Repinotan (BAY ϫ 3702): a with manuscript preparation. All authors read and approved 5HT1A agonist in traumatically brain injured patients. J Neuro- the final manuscript. UG and MFB reviewed the original study trauma 2001;18:1313–21 data and data analysis. UG maintains the study records. 17. Teal P, Silver FL, Simard D. The BRAINS study: safety, tolerability, and dose-finding of repinotan in acute stroke. Can DISCLOSURE J Neurol Sci 2005;32:61–7 18. Teal P, Davis S, Hacke W, Kaste M, Lyden PD, Fierus M. A Bayer Schering Pharma AG, Germany, provided the study randomized, double-blind, placebo-controlled trial to eval- drug and funded part of this study. MFB and GW are uate the efficacy, safety, tolerability, and pharmacokinetic/ employees of Bayer Schering Pharma AG. pharmacodynamic effects of a targeted exposure of intravenous repinotan in patients with acute ischemic stroke: modi- REFERENCES fied Randomized Exposure Controlled Trial (mRECT). Stroke 1. Couzin J. Medicine: a sigh of relief for painkillers. Science 2009;40:3518–25 2003;301:150 19. Motulsky HJ. Analyzing Data with GraphPad Prism. San 2. Garner SJ, Eldridge FL, Wagner PG, Dowell RT. Buspirone, an Diego: GraphPad Software Inc., 1999 anxiolytic drug that stimulates respiration. Am Rev Respir Dis 20. Oertel BG, Schneider A, Rohrbacher M, Schmidt H, Tegeder I, 1989;139:946–50 Geisslinger G, Lotsch J. The partial 5-hydroxytryptamine1A 3. Wilken B, Lalley P, Bischoff AM, Christen HJ, Behnke J, receptor agonist buspirone does not antagonize morphine- Hanefeld F, Richter DW. Treatment of apneustic respiratory induced respiratory depression in humans. Clin Pharmacol disturbance with a serotonin-receptor agonist. J Pediatr Ther 2007;81:59–68 1997;130:89–94 21. Pavlakovic G, Tigges J, Crozier TA. Effect of buspirone on 4. El-Khatib MF, Kiwan RA, Jamaleddine GW. Buspirone treat- thermal sensory and pain thresholds in human volunteers. ment for apneustic breathing in brain stem infarct. Respir Care BMC Clin Pharmacol 2009;9:12 2003;48:956–8 22. Guenther U, Bischoff A, Kettler D, Richter DW. 5-HT1A- 5. Sahibzada N, Ferreira M, Wasserman AM, Taveira-DaSilva AM, agnoists protect against opioidergic depression of respiration. Gillis RA. Reversal of morphine-induced apnea in the anesthe- In: Urban BW, Barann M, eds. Molecular und basic mecha- tized rat by drugs that activate 5-hydroxytryptamine(1A) recep- nisms of anesthesia D-49525 Lengerich: Papst Science Publish- tors. J Pharmacol Exp Ther 2000;292:704–13 ers, 2001:400–3

906 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 23. Colpaert FC, Wu WP, Hao JX, Royer I, Sautel F, Wiesenfeld- 36. Dutschmann M, Waki H, Manzke T, Simms AE, Pickering AE, Hallin Z, Xu XJ. High-efficacy 5-HT1A receptor activation Richter DW, Paton JF. The potency of different serotonergic causes a curative-like action on allodynia in rats with spinal agonists in counteracting opioid evoked cardiorespiratory dis- cord injury. Eur J Pharmacol 2004;497:29–33 turbances. Philos Trans R Soc Lond B Biol Sci 2009;364:2611–23 24. Galeotti N, Ghelardini C, Bartolini A. 5-HT1A agonists induce 37. Eriksson TM, Golkar A, Ekstrom JC, Svenningsson P, Ogren central cholinergic antinociception. Pharmacol Biochem Behav SO. 5-HT7 receptor stimulation by 8-OH-DPAT counteracts the 1997;57:835–41 impairing effect of 5-HT(1A) receptor stimulation on contex- 25. Bardin L, Colpaert FC. Role of spinal 5-HT(1A) receptors in tual learning in mice. Eur J Pharmacol 2008;596:107–10 morphine analgesia and tolerance in rats. Eur J Pain 38. Kellett DO, Ramage AG, Jordan D. Central 5-HT7 receptors are 2004;8:253–61 critical for reflex activation of cardiac vagal drive in anaesthe- 26. Graeff FG. Serotonin, the periaqueductal gray and panic. tized rats. J Physiol 2005;563:319–31 Neurosci Biobehav Rev 2004;28:239–59 39. Damaso EL, Bonagamba LG, Kellett DO, Jordan D, Ramage 27. Deseure K, Breand S, Colpaert FC. Curative-like analgesia in a AG, Machado BH. Involvement of central 5-HT7 receptors in neuropathic pain model: parametric analysis of the dose and modulation of cardiovascular reflexes in awake rats. Brain Res the duration of treatment with a high-efficacy 5-HT(1A) recep- 2007;1144:82–90 tor agonist. Eur J Pharmacol 2007;568:134–41 40. White DA, Reitan JA, Kien ND, Thorup SJ. Decrease in 28. Colpaert FC, Deseure K, Stinus L, Adriaensen H. High-efficacy vascular resistance in the isolated canine hindlimb after graded 5-hydroxytryptamine 1A receptor activation counteracts opi- doses of alfentanil, fentanyl, and sufentanil. Anesth Analg oid hyperallodynia and affective conditioning. J Pharmacol 1990;71:29–34 Exp Ther 2006;316:892–9 41. Lutsep HL. Repinotan, a 5-HT1A agonist, in the treatment of 29. Calabrese EJ, Baldwin LA. U-shaped dose-responses in biol- acute ischemic stroke. Curr Drug Targets CNS Neurol Disord ogy, toxicology, and public health. Annu Rev Public Health 2005;4:119–20 2001;22:15–33 42. Hutto CW Jr, Crowder WF. Dosage choices of rats for mor- 30. Calabrese EJ. Hormesis: basic, generalizable, central to toxicol- phine, for heroin, and between morphine and heroin. Pharma- ogy and a method to improve the risk-assessment process. Int col Biochem Behav 1997;58:133–40 J Occup Environ Health 2004;10:466–7 43. Eisenach JC, Gebhart GF. Intrathecal amitriptyline: antino- 31. Calabrese EJ, Baldwin LA. Hormesis: the dose-response revo- ciceptive interactions with intravenous morphine and intra- lution. Annu Rev Pharmacol Toxicol 2003;43:175–97 thecal clonidine, neostigmine, and carbamylcholine in rats. 32. Mauler F, Horvath E. Neuroprotective efficacy of repinotan Anesthesiology 1995;83:1036–45 HCl, a 5-HT1A receptor agonist, in animal models of stroke 44. Sandkuhler J, Gebhart GF. Characterization of inhibition of a and traumatic brain injury. J Cereb Blood Flow Metab spinal nociceptive reflex by stimulation medially and laterally 2005;25:451–9 in the midbrain and medulla in the pentobarbital-anesthetized 33. Clarke RW, Ward RE. The role of 5-HT(1A)-receptors in rat. Brain Res 1984;305:67–76 fentanyl-induced bulbospinal inhibition of a spinal withdrawal 45. Bonnefont J, Chapuy E, Clottes E, Alloui A, Eschalier A. Spinal reflex in the rabbit. Pain 2000;85:239–45 5-HT1A receptors differentially influence nociceptive process- 34. Gjerstad J, Tjolsen A, Hole K. The effect of 5-HT1A receptor ing according to the nature of the noxious stimulus in rats: stimulation on nociceptive dorsal horn neurones in rats. Eur effect of WAY-100635 on the antinociceptive activities of J Pharmacol 1996;318:315–21 paracetamol, venlafaxine and 5-HT. Pain 2005;114:482–90 35. Manzke T, Dutschmann M, Schlaf G, Morschel M, Koch UR, Ponimaskin E, Bidon O, Lalley PM, Richter DW. Serotonin targets inhibitory synapses to induce modulation of network functions. Philos Trans R Soc Lond B Biol Sci 2009;364:2589–602

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 907 Society for Technology in Anesthesia

Section Editor: Dwayne Westenskow

Goal-Directed Fluid Management Based on the Pulse Oximeter–Derived Pleth Variability Index Reduces Lactate Levels and Improves Fluid Management

Patrice Forget, MD,* Fernande Lois, MD,* and Marc de Kock, MD, PhD*

BACKGROUND: Dynamic variables predict fluid responsiveness and may improve fluid management during surgery. We investigated whether displaying the variability in the pulse oximeter plethysmogram (pleth variability index; PVI) would guide intraoperative fluid management and improve circulation as assessed by lactate levels. METHODS: Eighty-two patients scheduled for major abdominal surgery were randomized into 2 groups to compare intraoperative PVI-directed fluid management (PVI group) versus standard care (control group). After the induction of general anesthesia, the PVI group received a 500-mL crystalloid bolus and a crystalloid infusion of 2 mL ⅐ kgϪ1 ⅐ hϪ1. Colloids of 250 mL were administered if the PVI was Ͼ13%. Vasoactive drug support was given to maintain the mean arterial blood pressure above 65 mm Hg. In the control group, an infusion of 500 mL of crystalloids was followed by fluid management on the basis of fluid challenges and their effects on mean arterial blood and central venous pressure. Perioperative lactate levels, hemodynamic data, and postoperative complications were recorded prospectively. RESULTS: Intraoperative crystalloids and total volume infused were significantly lower in the goal-directed PVI group. Lactate levels were significantly lower in the PVI group during surgery and 48 hours after surgery (P Ͻ 0.05). CONCLUSIONS: PVI-based goal-directed fluid management reduced the volume of intraoperative fluid infused and reduced intraoperative and postoperative lactate levels. (Anesth Analg 2010; 111:910–4)

ypovolemia occurs frequently in the operating as accurately as does stroke volume variation.9 Neverthe- room. Its diagnosis remains difficult, but assessment less, it remains unknown whether the optimization of the Hof the adequacy of intravascular volume is of prime plethysmogram variability that occurs intraoperatively im- importance to maintain cardiac output and thus avoid tissue proves fluid management and circulation. To investigate hypoxia. In 1 meta-analysis the authors observed that periop- this, we used a pulse oximeter to continuously monitor the 7 erative hemodynamic optimization reduced mortality.1 PVI. We measured the impact of PVI-based goal-directed For many years, cardiac filling pressures were used to fluid management on perioperative lactate levels. guide intravascular volume therapy. This, however, is not a 2 reliable predictor of fluid responsiveness. Dynamic vari- METHODS ables (indices evaluating the response to a cyclic preload After approval of the Ethics Committee of St.-Luc Hos- variation) provide a better prediction of fluid responsive- pital (Brussels, Belgium) (www.clinicaltrials.gov, no. 3 ness. Among these, the arterial pulse pressure variation NCT00816153), and after obtaining written informed induced by mechanical ventilation has been demonstrated consent, a pilot study including 20 patients (10 per group) 3 as one of the best tools to guide volume therapy. Lopes et was conducted for the power analysis. Results showed an al. showed an improvement in postoperative outcome after improvement of 20% of the primary outcome (whole blood high-risk surgery when the pulse pressure variation was lactate levels) with the use of the PVI. A sample size of 37 used to guide intraoperative fluid therapy.4 Natalini et al. patients per group was calculated for a 0.05 difference and Cannesson et al. demonstrated that respiratory varia- (2-sided) with a power of 80%. tions in the amplitude of the pulse oximeter plethysmo- Between May and September 2008, we obtained written graphic waveform and in the pulse pressure both predict informed consent from 86 patients who met the inclusion fluid responsiveness.5–8 Zimmerman et al. showed that criteria: older than 18 years and the absence of cardiac pleth variability index (PVI) predicts fluid responsiveness arrhythmias, ultrasonographic cardiac ejection fraction Ͻ30%, lung pathology prohibiting mechanical ventilation Ϫ1 From the *Department of Anesthesiology, Universite´ catholique de Louvain, with tidal volumes larger than 6 mL ⅐ kg , and kidney St.-Luc Hospital, Brussels, Belgium. dialysis. They were scheduled for esophagectomy, gastric Accepted for publication May 27, 2010. resection/suture, hepatectomy, pancreatectomy, or intesti- Address correspondence and reprint requests to Patrice Forget, Department nal and colorectal surgeries. Patients were randomized to of Anesthesiology, St.-Luc Hospital, av. Hippocrate 10–1821, 1200 Brussels, Belgium. Address e-mail to [email protected]. either the PVI group or the control group. Copyright © 2010 International Anesthesia Research Society Heart rate, arterial blood pressure, oxygen saturation, DOI: 10.1213/ANE.0b013e3181eb624f inhaled gas concentrations, and temperature were measured

910 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 continuously by a Datex S/5 monitor (Datex Ohmeda®,GE Healthcare). A Masimo Set version V7.1.1.5 pulse oximeter (Masimo Co., Irvine, California) was placed on the patient’s finger for the continuous monitoring of PVI. A 20-G radial arterial catheter and a central venous access catheter were inserted at the end of the induction phase. A thoracic epidural catheter was placed before the induction of the general anesthesia. Anesthesia was induced with propofol 2 to 4 mg ⅐ Ϫ Ϫ kg 1 and atracurium or rocuronium 0.4 to 0.6 mg ⅐ kg 1 and maintained with sevoflurane or desflurane. The lungs were ⅐ Ϫ1 ϭ ventilated with 6 to 8 mL kg of tidal volume, I:E 1:2. The Figure 1. Trial proﬁle. PVI group: pleth variability index–guided ﬂuid ϭ frequency was set to maintain normocapnia (Paco2 target management. 40 Ϯ 3 mm Hg). In the PVI group, 500 mL of crystalloids (NaCl 0.9% or P-Lyte®, Baxter) was infused during induction, followed by Table 1. Preoperative Characteristics, Incidence Ϫ Ϫ a2mL⅐ kg 1 ⅐ h 1 continuous infusion. If PVI was higher of Chronic Diseases, Type and Duration of Surgery than 13% for Ͼ5 minutes, we gave a 250-mL bolus of and Anesthesia, Use of Epidural Analgesia in the colloid (hydroxyethyl starch 6%, Voluven®, Fresenius Pleth Variability Index (PVI) Group (PVI-Guided Fluid Management) and Control Group Kabi). The dose was repeated every 5 minutes if PVI was still higher than 13%. Norepinephrine was given as needed PVI group Control group (41 ؍ N) (41 ؍ N) to maintain a mean arterial blood pressure Ͼ65 mm Hg. Age (years) 59 Ϯ 14 61 Ϯ 12 In the control group, 500 mL of crystalloids was infused Weight (kg) 71 Ϯ 15 68 Ϯ 16 during induction, followed by a continuous infusion of Height (cm) 169 Ϯ 9 170 Ϯ 9 Ϫ Ϫ crystalloids (4 to 8 mL ⅐ kg 1 ⅐ h 1). A bolus of colloids was Sex (female/male) 16/25 (39/61) 16/25 (39/61) given if acute blood loss of Ͼ50 mL occurred, if the mean ASA score 2 22 (54) 22 (54) arterial blood pressure decreased below 65 mm Hg, or if the 3 19 (46) 19 (46) central venous pressure decreased below 6 mm Hg. A Chronic diseases repeat bolus was given after waiting 5 minutes if any one of Cirrhosis 3 (7) 0 (0) the criteria was met. If the mean arterial blood pressure Chronic obstructive pulmonary 2 (5) 2 (5) decreased below 65 mm Hg and remained unresponsive to disease Hypertension 18 (44) 13 (32) fluids, norepinephrine was given to maintain the mean Peripheral vascular disease 7 (17) 7 (17) arterial blood pressure above 65 mm Hg. Coronary artery disease 5 (12) 2 (5) Arterial blood samples were taken at the time of skin Other cardiomyopathy 2 (5) 4 (10) incision, each hour during surgery and 6, 12, 18, 24, 36, and Diabetes mellitus 4 (10) 2 (5) Preoperative biological values 48 hours after the end of surgery. The lactate concentration Hemoglobin (g ⅐ dLϪ1) 12.5 Ϯ 2 12.7 Ϯ 2 was measured using an ABL 620 analyzer (Radiometer, Serum creatinine (mg ⅐ dLϪ1) 0.96 Ϯ 0.2 0.97 Ϯ 0.3 Copenhagen, Denmark). Serum creatinine concentrations Type of surgery were measured 24 and 48 hours after surgery. The anesthe- Upper gastrointestinal 7 (17) 5 (12) Hepato-biliary 11 (27) 15 (37) siologist identified and recorded instances of intraoperative Lower gastrointestinal 24 (59) 22 (54) hypotension (systolic blood pressure 20% below the value Laparoscopic approach 5 (12) 5 (12) measured the day before surgery, while the patient was Duration of surgery (minutes) 295 Ϯ 125 301 Ϯ 154 resting quietly for at least 15 minutes) and oliguria (urine Duration of anesthesia 346 Ϯ 125 356 Ϯ 158 Ϫ output Ͻ0.5 mL ⅐ kg 1 for Ͼ2 hours). (minutes) Epidural analgesia 33 (81) 29 (71) During the first 30 days after surgery, a blinded postoperative care team member identified, collected, and re- One patient per group had two types of surgery. Data are presented as mean Ϯ SD or number (%). P Ͼ 0.05 for all the data. corded instances of postoperative infection, pulmonary embolism, acute myocardial infarction, acute lung injury/acute respiratory distress syndrome, pulmonary patients completed the protocol and were analyzed. No edema, arrhythmia, stroke, cardiac arrest, coagulopathy patient met abandon criteria. Ϫ (platelets Ͻ100,000 ␮L 1, international normalized ratio Student’s t test was used to compare normally distrib- ␹2 Ͼ2), hepatic dysfunction, nausea or vomiting necessitating uted continuous variables and for categorical variables. treatment, upper digestive hemorrhage, leakage of anasto- Homogeneity of variances was verified by the Levene’s Ͻ mosis, and mortality. test. A P value 0.05 was considered statistically significant. Data are expressed as mean (Ϯsd), mean [95% confidence interval], or number (percentage). STATISTICA (data Statistical Analysis analysis software system) version 7 (Statsoft, Inc., 2004) was Data were analyzed by comparing the patients in the PVI used for all analyses. group with those in the control group using a modified intention-to-treat analysis (4 patients were excluded after RESULTS the randomization for intraoperative arrhythmia or cancel- Table 1 lists the patients’ history and surgery. There were lation of the surgery, 2 per group; Fig. 1). The remaining 82 no preoperative differences between the goal-directed fluid

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 911 Pleth Variability Index and Fluid Management

Table 2. Fluids Administered, Blood Loss, Hemodynamic Status, Physiologic Status, and Renal Function During and After Surgery in the Pleth Variability Index (PVI) Group (PVI-Guided Fluid Management) and in the Control Group PVI group Control group P value (41 ؍ N) (41 ؍ N) Intraoperative fluids (mL) Crystalloids 1363 ͓1185–1540͔ 1815 ͓1568–2064͔ 0.004 Colloids 890 ͓709–1072͔ 1003 ͓779–1227͔ 0.43 Blood products 141 ͓53–230͔ 99 ͓20–179͔ 0.48 Total of intraoperative fluids 2394 ͓2097–2692͔ 2918 ͓2478–3358͔ 0.049 Blood losses 349 ͓230–468͔ 440 ͓242–637͔ 0.43 Postoperative fluids (24 hours) Crystalloids 3107 ͓2760–3454͔ 3516 ͓3009–4024͔ 0.17 Colloids 268 ͓126–409͔ 358 ͓175–540͔ 0.43 Blood products 8 ͓Ϫ8–25͔ 44 ͓Ϫ45–133͔ 0.41 Lactate levels (mMol ⅐ LϪ1) Maximum intraoperative 1.2 ͓1–1.4͔ 1.6 ͓1.2–2͔ 0.04 At 24 hours 1.4 ͓1.3–1.5͔ 1.8 ͓1.5–2.1͔ 0.02 At 48 hours 1.2 ͓1–1.3͔ 1.4 ͓1.2–1.5͔ 0.03 Lactate levels Ͼ1.7 mMol ⅐ LϪ1 Intraoperatively 7 (17) 4 (10) 0.33 At 24 hours 2 (5) 28 (68) <0.0001 At 48 hours 0 8 (20) 0.003 Lactate levels Ͼ5 mMol ⅐ LϪ1 Intraoperatively 0 1 (2) 0.31 At 24 hours 0 1 (2) 0.31 At 48 hours 0 1 (2) 0.31 Intraoperative hypotension 22 (54) 28 (68) 0.17 Continuous infusion of norepinephrine Intraoperative 9 (22) 9 (22) 1.0 At 24 hours 3 (7) 1 (2) 0.31 Renal function diuresis Intraoperative oliguria 13 (32) 17 (42) 0.34 Postoperative oliguria (24 hours) 3 (8) 3 (8) 0.97 Serum creatinine (mg ⅐ dLϪ1) At 24 hours 1.01 ͓0.9–1.1͔ 1.12 ͓0.9–1.3͔ 0.32 At 48 hours 0.91 ͓0.8–1͔ 1.09 ͓0.9–1.3͔ 0.11 Initiation of dialysis 1 (2) 0 (0) 0.32 Lactate levels: normal value 0.9–1.7 mMol ⅐ LϪ1. Oliguria was defined as a urinary output Ͻ0.5 mL ⅐ kgϪ1 for more than 2 hours. Data are presented as mean ͓95% confidence interval͔ or number (%). P Ͻ 0.05 was considered as statistically significant (boldface numerical entries).

that lactate levels were lower in the PVI group during and after surgery. There were no statistically significant differences in the incidence of hypotension, cardiovascular rescue, or renal dysfunction. Two patients in the PVI group died from septic shock 20 days and 33 days after surgery because of a failed anastomosis (Table 3).

DISCUSSION We found that PVI-guided fluid management resulted in less crystalloid administered perioperatively and reduced lactate levels during and after major abdominal surgery. Lactate levels provide an indirect but sensitive measure of organ perfusion. Lactate is clearly correlated with the adequacy of intravascular volume, tissue hypoxia, and energy failure due to bloodflow redistribution.10 Lactate Figure 2. Lactate levels during and after surgery in the pleth levels can be improved by the optimization of the fluid 2,4 variability index (PVI)–guided group (PVI-guided ﬂuid management) status and cardiac preload. and in the control group. Intraoperative: maximum intraoperative Our results confirm the conclusion of Lopes et al. The value. Data are presented as mean Ϯ SEM.*P Ͻ 0.05. use of the noninvasive PVI, or the invasively obtained pulse pressure variation, improves perioperative fluid management.4 In Lopes et al.’s study, the average management group and the control group. Table 2 shows amount of fluids was larger in the group guided by pulse that during surgery, patients in the PVI-directed fluid pressure variation, in contrast with our results. The management group were given less total fluid and less difference in results may be explained by the presence of crystalloid intraoperatively than was the control group. hypovolemia in some patients and hypervolemia in There were no differences postoperatively. Figure 2 shows others. These results therefore argue the superiority of

912 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 3. Postoperative Complications and Intensive Care Unit/Hospital Stay in the Pleth Variability Index (PVI) Group (PVI-Guided Fluid Management) and in the Control Group PVI group Control group P Value (41 ؍ N) (41 ؍ N) Postoperative complications Infection of surgery site 8 (20) 8 (20) 1.0 Other infections (pulmonary, line-related, other abdominal) 6 (15) 7 (17) 0.77 Cardiovascular complications (acute myocardial infarction, 4 (10) 8 (20) 0.26 acute lung injury/acute respiratory distress syndrome, pulmonary edema, arrhythmia) Coagulopathy 5 (12) 6 (15) 0.75 Nausea and/or vomiting 0 (0) 4 (16) 0.08 Upper digestive hemorrhage 4 (10) 3 (7) 0.78 Leakage of anastomosis 5 (12) 5 (12) 1.0 Morbidity (event per patient) 1.2 Ϯ 1.8 1.5 Ϯ 2.2 0.46 Mortality 2 (5) 0 (0) 0.16 Length of stay Postoperative mechanical ventilation 1 (2) 3 (7) 0.31 Intensive care unit (days) 2.2 Ϯ 5.7 1.8 Ϯ 7.2 0.71 Hospital (days) 15.1 Ϯ 14.3 16.0 Ϯ 17.8 0.78

Data are presented as mean Ϯ SD or number (%).

goal-directed fluid management over simplistic restrictive shown) and consequently chose 14% as the threshold for or liberal approaches for fluid management, avoiding hy- fluid loading. povolemia and hypervolemia.11,12 Whereas the PVI may be useful in most patients, our Unlike Lopes et al., we did not find an improvement exclusion criteria limit the application of our results in in terms of the number of complications. The much some patients. To maintain homogeneity between the 2 higher incidence of hypovolemia reported by Lopes et al. study groups, we did not include patients with severe may account for this difference. The clinical significance of cardiac insufficiency (ejection fraction Ͻ30%) or chronic lower lactate levels in our relatively small study may be dialysis. Moreover, the dynamic variables must not be questioned. Additionally, fluid management in the control calculated in the presence of arrhythmia. One patient per group was different by design, favoring greater fluid group was excluded because of an intraoperative arrhyth- Ϫ Ϫ crystalloid administration (2 mL ⅐ kg 1 ⅐ h 1 in the PVI mia. Additionally, these results cannot be extrapolated to Ϫ Ϫ group vs. 4 to 8 mL ⅐ kg 1 ⅐ h 1 in the control group), and other devices that calculate the respiratory variation of the it was possibly influenced by the fact that the control group plethysmographic curve. The algorithm used to process had a greater blood loss (440 [242 to 637] mL vs. 349 [230 to the signal may explain the poor accuracy observed by 468] mL) (although this was not statistically significant). others.13 Moreover, we did not measure the possible When mean arterial blood pressure decreased to Ͻ65 impact of the use of epidural analgesia and thoracotomy mm Hg, the PVI group received norepinephrine, whereas in some patients. the control group received norepinephirne and a bolus of In conclusion, the use of PVI-guided fluid management crystalloid. In our study a “learning contamination bias” was associated with lower lactate levels during major may have blunted the differences between the groups. This abdominal surgery. Patients in the PVI-guided group were bias occurs when a team member gains experience with given less crystalloid. Reduced lactate levels in PVI-guided pulse pressure variation and begins, intuitively, to use patients suggests that PVI-guided fluid management may respiratory variations of the arterial pressure curve to treat lead to fluid administration that is tailored to each indi- patients in the control group. However, small variations are vidual patient’s needs. difficult to see without using a device that makes the calculations from the curve. The PVI was calculated by the new Masimo Set pulse ACKNOWLEDGMENTS oximeter (Masimo Co., Irvine, California) from the respira- Masimo Corporation graciously provided devices during the tory variations in the perfusion index (PI). The PI is the study protocol. percentage amplitude difference between the pulsatile infrared signal and the nonpulsatile infrared signal. The PVI REFERENCES is calculated by measuring changes in the PI during the ϭ Ϫ ϫ 1. Poeze M, Greve JWM, Ramsay G. Meta-analysis of hemody- respiratory cycle: PVI [(PImax PImin)/PImax] 100. namic optimisation: relationship to methodological quality. Cannesson et al. have demonstrated that the PVI predicts Crit Care 2005;9:R771–9 fluid responsiveness in the operating room. They showed 2. Cavallaro F, Sandroni C, Antonelli M. Functional hemody- that the cutoff value to distinguish responders from non- namic monitoring and dynamic indices of fluid responsiveness. Minerva Anestesiol 2008;74:123–35 responders to intravascular volume expansion (in terms of 3. Michard F, Teboul JL. Predicting fluid reponsiveness in ICU Ͼ 7 an increase of cardiac index) was a PVI 14%. We patients. A critical analysis of the evidence. Chest 2002; confirmed their results in a preliminary study (data not 121(6):2000–8

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 913 Pleth Variability Index and Fluid Management

4. Lopes MR, Oliveira MA, Pereira VO, Lemos IP, Auler JO Jr, 9. Zimmermann M, Feibicke T, Keyl C, Prasser C, Moritz S, Graf Michard F. Goal-directed fluid management based on pulse BM, Wiesenack C. Accuracy of stroke volume variation com- pressure variation monitoring during high-risk surgery: a pilot pared with pleth variability index to predict fluid responsive- randomized controlled trial. Crit Care 2007;11(5):R100 ness in mechanically ventilated patients undergoing major 5. Desebbe O, Cannesson M. Using ventilation-induced plethys- surgery. Eur J Anaesthesiol 2009; [Epub ahead of print] mographic variations to optimize patient fluid status. Curr 10. Valenza F, Aletti G, Fossali T, Chevallard G, Sacconi F, Irace M, Opin Anaesthesiol 2008;21:772–8 Gattinoni L. Lactate as a marker of energy failure in critically ill 6. Natalini M, Rosano A, Taranto M, Faggian B, Vittorielli E, patients: hypothesis. Crit Care 2005;9(6):588–93 Bernardini A. Arterial versus plethysmographic dynamic 11. Bundgaard-Nielsen M, Holte K, Secher NH, Kehlet H. indices to test responsiveness for testing fluid administra- Monitoring of peri-operative fluid administration by indi- tion in hypotensive patients: a clinical trial. Anesth Analg 2006;103(6):1478–84 vidualized goal-directed therapy. Acta Anaesthesiol Scand 7. Cannesson M, Desebbe O, Rosamel P, Delannoy B, Robin J, 2007;51(3):331–40 Bastien O, Lehot JJ. Pleth variability index to monitor the 12. Bundgaard-Nielsen M, Ruhnau B, Secher NH, Kehlet H. Flow- respiratory variations in the pulse oximeter plethysmographic related techniques for preoperative goal-directed fluid optimi- waveform amplitude and predict fluid responsiveness in the sation. Br J Anaesth 2007;98(1):38–44 operating theatre. Br J Anaesth 2008;101(2):200–6 13. Landsverk SA, Hoiseth LO, Kvandal P, Hisdal J, Skare O, 8. Cannesson M, Attof Y, Rosamel P, Desebbe O, Joseph P, Kirkeboen KA. Poor agreement between respiratory variations Metton O, Bastien O, Lehot JJ. Respiratory variations in in pulse oximetry photoplethysmographic waveform ampli- pulse oximetry plethysmographic waveform amplitude to tude and pulse pressure in intensive care unit. Anesthesiology predict fluid responsiveness in the operating room. Anes- 2008;109(5):849–55 thesiology 2007;106(6):1105–11

914 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA A Long-Term Clinical Evaluation of AutoFlow During Assist-Controlled Ventilation: A Randomized Controlled Trial

Sigismond Lasocki, MD, PhD, Franc¸oise Labat, MD, Gaetan Plantefeve, MD, Mathieu Desmard, MD, and Herve´ Mentec, MD

BACKGROUND: Many new mechanical ventilation modes are proposed without any clinical evaluation. “Dual-controlled” modes, such as AutoFlow™, are supposed to improve patient– ventilator interfacing and could lead to fewer alarms. We performed a long-term clinical evaluation of the efﬁcacy and safety of AutoFlow during assist-controlled ventilation, focusing on ventilator alarms. METHODS: Forty-two adult patients, receiving mechanical ventilation for more than 2 days with a Dra¨ger Evita 4 ventilator were randomized to conventional (n ϭ 21) or AutoFlow (n ϭ 21) assist-controlled ventilation. Sedation was given using a nurse-driven protocol. Ventilator- generated alarms were exhaustively recorded from the ventilator logbook with a computer. Daily blood gases and ventilation outcome were recorded. RESULTS: A total of 403 days of mechanical ventilation were studied and 45,022 alarms were

recorded over a period of 8074 hours. The course of respiratory rate, minute ventilation, FIO2, positive end-expiratory pressure, PaO2/FIO2,PaCO2, and pH and doses and duration of sedation did not differ between the 2 groups. Outcome (duration of mechanical ventilation, ventilator- associated pneumonia, course of Sequential Organ Failure Assessment score, or death) was not different between the 2 groups. The number of alarms per hour was lower with AutoFlow assist-controlled ventilation: 3.3 [1.5 to 17] versus 9.1 [5 to 19], P Ͻ 0.0001 (median [quartile range]). In multivariate analysis, a low alarm rate was associated with activation of AutoFlow and a higher midazolam dose. CONCLUSIONS: This ﬁrst long-term clinical evaluation of the AutoFlow mode demonstrated its safety with regard to gas exchange and patient outcome. AutoFlow also allowed a very marked reduction in the number of ventilator alarms. (Anesth Analg 2010;111:915–21)

“ ual-controlled” ventilation modes1,2 are reported number of alarms may therefore improve alarm efficiency.9 to combine the advantages of both volume and We hypothesize that activation of the AF would reduce the Dpressure-controlled ventilation modes. The only number of ventilation-generated alarms, without impairing studies comparing dual modes to conventional assist- the patient’s ventilation. Because no clinical evaluation of controlled ventilation (ACV) focused on short-term effects this ventilation mode is available, this study was also (several hours). They report a reduction of inspiratory designed to clinically evaluate AF activation. pressure.3,4 Dra¨ger’s AutoFlow™ (AF) is one of these The aim of this randomized controlled study was to dual-controlled ventilation modes that also allows sponta- clinically evaluate AF activation during ACV, with regard neous breathing throughout the respiratory cycle. Drager to ventilator-generated alarms (primary aim) and to gas claims that AF is expected to improve patient–ventilator exchange and patient outcome (secondary aims). interfacing and could decrease the number of times patients fight the ventilator. AF could consequently reduce the METHODS The study protocol was approved by the Comite´ Consul- number of ventilator alarms. Alarms are partly responsible tatif de Protection des Personnes dans la Recherche for the high noise level in intensive care units (ICU),5–7 and Biome´dicale (independent ethics committee) of Saint- a reduction of ventilator alarms would therefore be benefi- Germain-en-Laye, France (ClinicalTrial.gov identifier: cial to both patients and ICU staff. ICU staff also fail to NCT0092774). Written informed consent was obtained from correctly identify many of these alarms.8 Reducing the the patient or next of kin. From the Reánimation Polyvalente, Centre Hospitalier Victor Dupouy, Patients Argenteuil, France. Adult patients admitted to the ICU of Victor Dupouy Accepted for publication June 19, 2010. Hospital, Argenteuil, France, were eligible when they re- Sigismond Lasocki, MD, PhD, is currently affiliated with Reánimation Chirurgicale, APHP, CHU Bichat Claude Bernard, Paris, France. quired ACV with an Evita4 ventilator (Dra¨ger Medical, Ͼ The Hospital Victor Dupouy, Argenteuil, France, supported this study. Antony, France) for an expected duration of 2 days. Dra¨gger SA provided only technical assistance for the recording of a Patients were not included in the case of coma, ventilation ventilator logbook, but had no access to the data and was not involved in for Ͼ12 hours before inclusion, pregnancy, or inclusion in the preparation of this manuscript. another study. Address correspondence to Sigismond Lasocki, MD, PhD, Reánimation Chirurgicale, CHU Bichat, 46 rue Henri Huchard, 75018 Paris, France. Address e-mail to [email protected]. AutoFlow Copyright © 2010 International Anesthesia Research Society AF is a dual-control mechanical ventilation mode associ- 1,10 DOI: 10.1213/ANE.0b013e3181f00015 ated with ACV. All breaths are pressure-controlled, with

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 915 Evaluation of AutoFlow Controlled Ventilation

Figure 1. Sedation algorithm used by nurses to conduct patient sedation (VAS ϭ visual analgesia scale; Ramsay ϭ sedation score according to the Ramsay scale11). a delivered level of pressure support that varies from Ventilator-Generated Alarms breath to breath to deliver the set tidal volume (Vt). AF Ventilators were periodically connected to a personal uses a feedback loop that regulates inspiratory flow. Dy- computer to download the number of alarms, changes in namic compliance is measured breath by breath, and the ventilator settings, alarm limit settings, and the number required ␦ pressure for the next breath is calculated by of times alarms were manually silenced (silence knob dividing the desired Vt by dynamic compliance. Changes activation). of inspiratory pressure from breath to breath are limited to 3 mbar. When inspiratory pressure reaches the upper Outcome pressure limit minus 5 mbar, inspiratory time is increased Organ failure was assessed daily using the Sequential 12 within the limits defined by the set respiratory rate. Organ Failure Assessment (SOFA) score. Duration of mechanical ventilation, ICU and hospital survival, the Mechanical Ventilation incidence of pneumothorax, and the incidence of ventilator- Patients were randomly assigned to the control group associated pneumonia were recorded. (AFϪ) or the AF group (AFϩ) by opening a sealed envelope. Attending physicians chose all respiratory settings, Statistical Analysis Ϯ without intervention by the study investigators. The upper Results are expressed as median[Q1–Q3] or mean sd. limit of inspiratory pressure alarm was initially set at 50 cm Comparisons were performed with a Mann–Whitney test or Student’s t test, a ␹2 test with Yates’ correction, or an H2O. Other alarm limits were set at the manufacturer’s default values, which varied according to the patient’s analysis of variance (ANOVA) for repeated measures as body weight. Attending physicians were allowed to change appropriate. Analysis was performed by either intention to any alarm limit and ventilator mode when clinically indi- treat on the whole study period or per protocol (for the cated, except that AF was always used with ACV in the period during which patients were on ACV, with or AFϩ group, and AF was never used with ACV in the AFϪ without AF), depending on the variable considered. Thus group. Discontinuation of mechanical ventilation was per- we report the alarm rate for the different ventilation mode ϩ formed according to our ICU standard protocol, by the (ACV or non-ACV) in both groups. Obviously, in the AF gradual reduction of pressure support. Blood gases were group, AF was not activated during non-ACV ventilation obtained at least once daily during the first days. Morning (because it was not available). To estimate the sample size of the study, we assumed values of ventilator settings, highest Fio2, highest positive that the total alarm rate in an ICU would be 36.5 end-expiratory pressure, highest Paco2, lowest Pao2/Fio2 13 ratio, and lowest pH were recorded daily. alarms/hr, and that ventilator alarms would account for 38% of all alarms.14 The alarm rate during ACV was Sedation assumed to be 14 alarms/hr. A 50% reduction when AF Patients were sedated by continuous infusions of midazo- was added to ACV was considered to be clinically relevant. lam and fentanyl according to a nurse-driven protocol (Fig. Twenty-one patients in each group were needed to achieve 1). The sedation goal was a Ramsay score11 of2or3. a 90% power, with an ␣ risk of 5%. To assess factors

916 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Patient Characteristics Table 2. Ventilator Settings and Gas P Exchange Variables (21 ؍ AF؊ (n (21 ؍ AF؉ (n P (21 ؍ AF؊ (n (21 ؍ Age (years) 71 ͓25–90͔ 65 ͓32–88͔ 0.70 AF؉ (n Body weight (kg) 73 ͓53–101͔ 64 ͓38–150͔ 0.75 RR (cycles/min) 18 ͓14–22͔ 20 ͓15–22͔ 0.87 Height (cm) 168 ͓150–191͔ 167 ͓150–185͔ 0.49 VT (mL) 500 ͓400–650͔ 500 ͓350–600͔ 0.40 ͓ ͔ ͓ ͔ Gender M/F 14/7 14/7 0.99 PEEP (cm H2O) 5 0–15 5 0–13 0.97 ͓ ͔ ͓ ͔ COPD (%) 2 (10) 6 (29) 0.24 FIO2 1 0.35–1 0.8 0.40–1 0.76 Type of patient 0.99 pH 7.31 ͓6.98–7.60͔ 7.28 ͓7.11–7.57͔ 0.76 ͓ ͔ ͓ ͔ Medical 15 15 PaCO2 (mm Hg) 39 31–61 42 29–110 0.51 ͓ ͔ ͓ ͔ Urgent surgery 6 6 PaO2 (mm Hg) 141 62–492 152 49–286 0.87 ͓ ͔ ͓ ͔ Indication of 0.31 PaO2/FIO2 (mm Hg) 192 62–532 242 49–356 0.98 mechanical Values are expressed as median ͓min–max͔. ventilation AFϩϭassist-controlled ventilation with AutoFlow activation; AFϪϭassist- ARF 15 12 controlled ventilation without AutoFlow activation; RR ϭ respiratory rate; VT ϭ Ventilation 43 ϭ ϭ tidal volume; PEEP positive end-expiratory pressure; FIO2 inspiratory without oxygen fraction. ARFa ARF on COPD 1 3 Postoperative 13 Ventilator Alarms and Interventions ventilation Admission 60 ͓25–101͔ 51 ͓19–105͔ 0.33 A total of 403 days (8074 hours) of mechanical ventilation SAPS II were studied. ACV was used for 3997 hours. AF was used Admission SOFA 9 ͓2–17͔ 9 ͓4–19͔ 0.57 for 2133 hours. A total of 45,022 alarms were recorded; Values are expressed as median ͓min–max͔ or numbers (%). nearly 1 alarm every 10 minutes. During ACV, 7060 alarms AFϩϭassist-controlled ventilation with AutoFlow activation; AFϪϭ were recorded in the AFϩ group, and 16,817 alarms were assist-controlled ventilation without AutoFlow activation; COPD ϭ chronic recorded in the AFϪ group. Figure 3 shows that the obstructive pulmonary disease; ARF ϭ acute respiratory failure; SAPS II ϭ simpliﬁed acute physiologic score (34); SOFA ϭ Sequential Organ Failure ventilator alarm rate was lower when AF was used in Assessment (12). conjunction with ACV (3.3 [1.5 to 17] alarms/hr with AF vs. a Ͼ Shock and PaO2/FIO2 250 mm Hg. 9.1 [5.2 to 19] without AF [P Ͻ 0.0001]). In the AFϪ group, the alarm rate was lower for ventilation modes other than ACV (mainly pressure support) than they were for ACV influencing the alarm rate, we divided patients into 2 (without AF), but no difference was observed between groups—higher and lower than the median alarm these ventilation modes and ACV for the AFϩ group (Fig. rate—and we performed a logistic regression, including all 3). The number of alarm setting modifications per hour was the parameters that had a P value of Ͻ0.1 in the univariate not different between groups (0.07 [0.02 to 0.23] vs. 0.09 [0.02 analysis (i.e., indexed midazolam, indexed fentanyl and AF to 0.23] modifications per hour for AFϩ and AFϪ patients, group, with indexed midazolam and indexed fentanyl respectively; P ϭ 0.85). Setting the high-pressure limit above ϩ being the total dose of midazolam or fentanyl indexed to 50 cmH2O was less frequent in the AF group (0 [0 to 2] vs. the patient’s body weight and the duration of drug infu- 2 [0 to 8] times per patient; P ϭ 0.0007). sion). Statistical analysis was performed with SPSS 15.0. The type of alarm differed between the 2 groups. ACV P Ͻ 0.05 was considered statistically significant. plus AF generated fewer pressure alarms than did ACV alone (P Ͻ 0.0001) (Fig. 4). The silence knob activation rate ϩ RESULTS was lower in the AF group during ACV (0.26 [0.1 to 1.1] vs. 0.72 [0.26 to 2] activation per hour; P ϭ 0.0013). Patients The median alarm rate during ACV was 6.37 alarms/hr. Forty-two patients were included. No statistically signifi- Patients with alarm rates lower than 6.37 alarms/hr were cant differences in demographic data were observed be- more frequently randomized to the AFϩ group (P ϭ tween the 2 groups (Table 1). Indications for mechanical 0.0002) and had a higher dosage of sedative drugs (fenta- ventilation were not statistically different between groups nyl, P ϭ 0.02; midazolam, P ϭ 0.07) (Table 4). In multivar- (P ϭ 0.31). iate analysis, an alarm rate lower than 6.37 alarms/hr was associated with activation of AF (OR [95% CI], 90 [5 to Ventilation and Gas Exchange 1570], P ϭ 0.002) and a higher midazolam dosage (OR 1801 Baseline ventilator settings and blood gases were not [3 to 1.1 106] per mg/d/kg, P ϭ 0.02). different between the 2 groups (Table 2). An ANOVA for repeated measures from day 1 to day 5 showed no signifi- Outcome cant difference between the 2 groups for either ventilator No patient suffered from pneumothorax in the AFϩ group settings or blood gases (Fig. 2). in comparison with 2 patients in the AFϪ group (P ϭ 0.48). Four cases of ventilator-associated pneumonia were ob- Sedation served in the AFϩ group in comparison with 8 in the AFϪ Sedation was not different between the 2 groups (Table 3). group (P ϭ 0.16). The median duration of ventilation was 6 Total doses of midazolam (114 [0 to 163] vs. 150 [0 to 316] [2 to 25] versus 9 [2 to 36] days in AFϩ and AFϪ groups, mg, P ϭ 0.5) and fentanyl (6600 [0 to 55,480] vs. 8000 [0 to respectively (P ϭ 0.33), and the median number of days free 21,500] ␮g, P ϭ 0.8) were not different between the 2 groups of mechanical ventilation at day 28 were 15 [0 to 26] and 13 (AFϩ and AFϪ groups, respectively). [0 to 26], respectively (P ϭ 0.55). An ANOVA for repeated

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 917 Evaluation of AutoFlow Controlled Ventilation

Figure 2. Time course of ventilatory settings, gas exchange, and Sequential Organ Failure Assessment (SOFA) score12 during the ﬁrst 5 days (D1 to D5) of mechanical ventilation. Solid squares represent the Dra¨ger’s AutoFlow (AF)ϩ group, and open circles represent the control (AFϪ) group. Mean Ϯ SEM. An analysis of variance (ANOVA) for repeated measures was not signiﬁcant. PEEP ϭ positive end-expiratory pressure. measures from day 1 to day 5 showed no significant DISCUSSION difference between the 2 groups for SOFA score (Fig. 2). Six This study is the first long-term clinical evaluation of AF patients in the AFϩ group and 9 patients in the AFϪ group during ACV. Clinical outcome and blood gas variables died in the ICU (P ϭ 0.39), whereas 8 and 10 patients died were not different with and without AF. There were fewer in the hospital, respectively (P ϭ 0.62). ventilator alarms when AF was used.

918 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 3. Sedation Variables P (21 ؍ AF؊ (n (21 ؍ AF؉ (n Sedation duration (days) 3 ͓1–10͔ 5 ͓0–12͔ 0.44 Indexed midazolam 0.38 ͓0–1.13͔ 0.44 ͓0–0.78͔ 0.73 (mg/d/kg)a Indexed fentanyl (␮g/d/kg)b 28 ͓0–93͔ 25 ͓10–52͔ 0.46 Sedation modiﬁcations 1.5 ͓0–3.3͔ 1.4 ͓0–3.8͔ 0.66 (per day) Ramsay score of 2 or 3 2.9 ͓0.5–6.3͔ 2.2 ͓0–5.5͔ 0.29 (per day)c Values are expressed as median ͓min–max͔. AFϩϭassist-controlled ventilation with AutoFlow activation; AFϪϭassist- controlled ventilation without AutoFlow activation. a Indexed midazolam: total dose of midazolam indexed to the patient’s body weight and the duration of drug infusion. b Indexed fentanyl: total dose of fentanyl indexed to the patient’s body weight Figure 4. Distribution of alarm types in the 2 groups during assist- and the duration of drug infusion. controlled ventilation (ACV). AutoFlow ACV group (AFϩ) is repre- c Ramsay score (11) of 2 or 3, the mean number of Ramsay scoring not equal sented with dark gray bars, and conventional ACV group (AFϪ)is to 2 or 3 per day. represented with light gray bars. The left side of the chart shows the total number of all volume alarms generated during ACV (i.e., low or high tidal volume (VT) and low or high minute ventilation alarms) and the total “partially delivered VT” alarm (this alarm is generated when the pressure limit is reached, leading to a stopping of inspiration), and the right side shows the total number of high and low airway pressure alarms. ACV without AF generated many more pressure alarms than did ACV with AutoFlow. Mean Ϯ SEM.

Table 4. Univariate Analysis of Factors Inﬂuencing Alarm Rate Low alarm High alarm P (21 ؍ ratea (n (21 ؍ ratea (n Age (years) 66 ͓26–90͔ 66 ͓25–88͔ 0.89 Gender (M/F) 15/6 13/8 0.11 COPD (%) 2 (10) 5 (24) 0.41 SAPS II (points) 59 ͓23–101͔ 54 ͓29–105͔ 0.27 Tracheal aspirate 1.5 ͓0.7–2.6͔ 1.9 ͓1.0–2.8͔ 0.27 volume Figure 3. Hourly alarm rates. Rectangular boxes represent the (points) median hourly alarm rate during assist-controlled ventilation (ACV) Nebulizations 0 ͓0–36͔ 0 ͓0–22͔ 0.25 and diabolo boxes represent the median hourly alarm rate during the (total number) other mechanical ventilation modalities, mainly during pressure Indexed 0.5 ͓0.2–1.1͔ 0.3 ͓0–0.8͔ 0.07 support (non-ACV). Upper and lower edges represent the 75th and midazolam 25th percentiles. Dark gray shapes represent rates observed for (mg/d/kg)b patients in the AutoFlow group (AFϩ), and light gray shapes repre- Indexed fentanyl 30 ͓7–93͔ 21 ͓0–53͔ 0.02 sent rates observed in the conventional ACV group (AFϪ). The (␮g/d/kg)c dashed line represents the overall median hourly alarm rate, which Sedation 4 ͓2–10͔ 4 ͓1–12͔ 0.72 was almost 1 alarm every 10 minutes. The lowest alarm rate was duration observed during ACV using AutoFlow, and the highest alarm rate was (days) observed during ACV without AutoFlow. ACV duration 85 ͓15–284͔ 54 ͓11–316͔ 0.16 (hours) AF is based on an attractive principle: to guarantee a set Silence knob 0.26 ͓0.1–1.1͔ 0.80 ͓0.26–2͔ 0.001 Vt and minute ventilation while maintaining the advan- activation (per tages of pressure-controlled ventilation.1,2 Despite this hour) Study group 17/4 4/17 0.0002 potential advantage, clinical evaluations have not been (AFϩ/AFϪ) performed. Even clinical efficiency in comparison with ͓ ͔ conventional ACV has not been formally demonstrated. Values are expressed as median min–max or number. COPD ϭ chronic obstructive pulmonary disease; SAPS II ϭ Simpliﬁed Acute This is the first report of around-the-clock observation Physiology Score II (34); ACV ϭ assist-controlled ventilation; AFϩϭassist- and long-term bedside evaluation of this ventilation controlled ventilation with AutoFlow activation; AFϪϭassist-controlled modality in the context of standard care. We found no ventilation without AutoFlow activation. a The patients were divided into 2 groups according to whether their ventilator differences with or without AF for gas exchange (P/F alarm rate was lower or higher than was the median alarm rate of the overall population (6.37 alarms/hr). ratio or Paco2) during ACV. No other studies are avail- b able for comparison. Previous studies have only reported Indexed midazolam, total dose of midazolam indexed to the patient’s body weight and the duration of drug infusion. the physiological advantages associated with pressure- c Indexed fentanyl, total dose of fentanyl indexed to the patient’s body weight regulated ACV in comparison with volume-controlled and the duration of drug infusion.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 919 Evaluation of AutoFlow Controlled Ventilation ventilation, such as lower inspiratory pressures3,4,15 and should have reduced the alarms rate in that group. The trial 3,15 lower Paco2. could not be blinded because the ventilator screen displays One concern regarding AF is that the level of support ventilator settings. However, the end points (gas exchange, (i.e., the level of pressure delivered) could theoretically alarms, and outcome) were assessed objectively. Lastly, no decrease as patient demand increases. Indeed, the work of modification of clinical outcome such as decreased dura- breathing increases when pressure support decreases.16 tion of mechanical ventilation or mortality was observed. It This has been recently confirmed in a lung simulator.17 is not surprising because no new mechanical ventilation However, the absence of altered gas exchange in the AFϩ mode has improved clinical outcome.10 In particular, a group and the trend towards a shorter duration of ventila- large-scale study comparing pressure-controlled ventilation do not support this hypothesis, in vivo. In fact, for a set tion with volume-controlled ventilation did not find any Vt, pressure-controlled ventilation reduces the work of direct benefit.30 In the present study, sedation was con- breathing in comparison with volume-controlled ventila- trolled by a nurse-driven protocol, but it can be assumed tion.18 In the present study, 2133 hours of AF ACV were that if physicians had prescribed sedation, they would have recorded, and no patient experienced any signs of respira- increased sedation during conventional ACV to decrease tory distress. However, our study, with a sample size of 42 the alarm rate, which could have accentuated the trends patients, is not powered to definitively assess these clinical observed in this study. aspects. Indeed, it was powered to compare the ventilation- related alarm rate with and without AF during ACV. CONCLUSIONS An original tool was used to record all ventilator- ACV with AF appears to be safe in terms of gas exchange generated alarms. A higher alarm rate was observed than and clinical outcome in this first long-term around-the- was previously reported by Chambrin et al. (0.6 alarm/hr 14 clock clinical evaluation. AF was associated with a marked in ). Gabor et al. found a higher rate of sound increase decrease in ventilator-generated alarms. The beneficial ef- Ϯ Ϯ 13 (37 20 to 72 13 times per hour of sleep), but they did fect of such a reduction of alarm rate on the comfort of not identify the source of each sound. The high alarm rate patients and ICU staff (quality of sleep and stress) deserves observed in the present study is certainly due to the further evaluation. method used, but could also be due to the target sedation level (Ramsay score of 2 to 3). A patient with less sedation REFERENCES may fight the ventilator. However, this low target is widely 19–24 1. Branson RD, Davis K Jr. Dual control modes: combining recommended and validated. The lowest alarm rate volume and pressure breaths. Respir Care Clin N Am was observed during ACV with AF, and the highest rate 2001;7:397–408 was observed during conventional ACV. It is not surprising 2. Campbell RS, Davis BR. Pressure-controlled versus volume- that a pressure-controlled mode, such as AF, is associated controlled ventilation: does it matter? Respir Care 2002; 47:416–24 with a reduction in pressure alarms. This is in accordance 3. Alvarez A, Subirana M, Benito S. Decelerating flow ventilation with studies demonstrating a reduction of peak inspiratory effects in acute respiratory failure. J Crit Care 1998;13:21–5 pressure during pressure-controlled ventilation.3,4,15 Mul- 4. Guldager H, Nielsen SL, Carl P, Soerensen MB. A comparison tivariate analysis found only 2 factors associated with a of volume control and pressure-regulated volume control ventilation in acute respiratory failure. Crit Care (London) lower alarm rate: midazolam doses and AF activation. 1997;1:75–7 It may be beneficial to reduce the number of alarms in an 5. Freedman NS, Gazendam J, Levan L, Pack AI, Schwab RJ. ICU, because alarms are partly responsible for the high Abnormal sleep/wake cycles and the effect of environmental noise level in an ICU.5–7 The first consequence of alarm noise on sleep disruption in the intensive care unit. Am J noise could be sleep disruption.25,26 In addition, an excess Respir Crit Care Med 2001;163:451–7 6. Meyer TJ, Eveloff SE, Bauer MS, Schwartz WA, Hill NS, of false positive alarms may decrease alarm efficiency: Only Millman RP. Adverse environmental conditions in the respira- 26.8% of ventilator-generated alarms lead to an action.14 tory and medical ICU settings. Chest 1994;105:1211–6 Only one half of critical alarms are correctly identified by 7. Walder B, Francioli D, Meyer JJ, Lancon M, Romand JA. Effects ICU staff.8 Reducing the total number of alarms should of guidelines implementation in a surgical intensive care unit 9 to control nighttime light and noise levels. Crit Care Med reduce noise in the ICU and improve alarm efficiency. We 2000;28:2242–7 also showed that AF activation was associated with a 8. Cropp AJ, Woods LA, Raney D, Bredle DL. Name that tone. reduction in a surrogate marker of care interruption, acti- The proliferation of alarms in the intensive care unit. Chest vation of the silence knob, which could help reduce cross- 1994;105:1217–20 infections, because the ICU environment can be a reservoir 9. Siebig S, Sieben W, Kollmann F, Imhoff M, Bruennler T, 27,28 Rockmann F, Gather U, Wrede CE. Users’ opinions on inten- for pathogens. Further studies will be needed to con- sive care unit alarms—a survey of German intensive care units. firm the alarm reduction observed during AF ACV. Anaesth Intens Care 2009;37:112–6 The major limitations of our study are the absence of 10. Branson RD, Johannigman JA. What is the evidence base for fixed alarm limits and the unblinded design. The attending the newer ventilation modes? Respir Care 2004;49:742–60 11. Ramsay MA, Savege TM, Simpson BR, Goodwin R. Controlled physician was allowed to modify alarm limits as usual, sedation with alphaxalone-alphadolone. Br MedJ 1974;2:656–9 29 because no clear recommendations have been published. 12. Vincent JL, de Mendonca A, Cantraine F, Moreno R, Takala J, Physicians are unlikely to have set high alarm limits in Suter PM, Sprung CL, Colardyn F, Blecher S. Use of the SOFA view of the very high alarm rate observed in this study. score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Furthermore, no difference was observed for changes of Working group on “sepsis-related problems” of the European alarm limits, apart from the upper-pressure alarm limit set Society of Intensive Care Medicine. Crit Care Med 1998;26: Ϫ above 50 cmH2O (more frequent in the AF group), which 1793–800

920 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 13. Gabor JY, Cooper AB, Crombach SA, Lee B, Kadikar N, 22. Kollef MH, Levy NT, Ahrens TS, Schaiff R, Prentice D, Bettger HE, Hanly PJ. Contribution of the intensive care unit Sherman G. The use of continuous i.v. sedation is associated environment to sleep disruption in mechanically ventilated with prolongation of mechanical ventilation. Chest 1998;114: patients and healthy subjects. Am J Respir Crit Care Med 541–8 2003;167:708–15 23. Kress JP, Pohlman AS, Hall JB. Sedation and analgesia in the 14. Chambrin MC, Ravaux P, Calvelo-Aros D, Jaborska A, Chopin intensive care unit. Am J Respir Crit Care Med 2002;166: C, Boniface B. Multicentric study of monitoring alarms in the 1024–8 adult intensive care unit (ICU): a descriptive analysis. Intens 24. Kress JP, Pohlman AS, O’Connor MF, Hall JB. Daily interrup- Care Med 1999;25:1360–6 tion of sedative infusions in critically ill patients undergoing 15. Edibam C, Rutten AJ, Collins DV, Bersten AD. Effect of mechanical ventilation. New Engl J Med 2000;342:1471–7 inspiratory flow pattern and inspiratory to expiratory ratio on 25. Freedman NS, Kotzer N, Schwab RJ. Patient perception of nonlinear elastic behavior in patients with acute lung injury. sleep quality and etiology of sleep disruption in the intensive Am J Respir Crit Care Med 2003;167:702–7 care unit. Am J Respir Crit Care Med 1999;159:1155–62 16. Kreit JW, Capper MW, Eschenbacher WL. Patient work of 26. Parthasarathy S, Tobin MJ. Sleep in the intensive care unit. breathing during pressure support and volume-cycled me- Intens Care Med 2004;30:197–206 chanical ventilation. Am J Respir Crit Care Med 1994;149: 27. Bures S, Fishbain JT, Uyehara CF, Parker JM, Berg BW. 1085–91 Computer keyboards and faucet handles as reservoirs of 17. Mireles-Cabodevila E, Chatburn RL. Work of breathing in nosocomial pathogens in the intensive care unit. Am J Infect adaptive pressure control continuous mandatory ventilation. Control 2000;28:465–71 Respir Care 2009;54:1467–72 28. Eggimann P, Pittet D. Infection control in the ICU. Chest 18. Cinnella G, Conti G, Lofaso F, Lorino H, Harf A, Lemaire F, 2001;120:2059–93 Brochard L. Effects of assisted ventilation on the work of 29. Les recommandations des experts de la Socie´te´deReánimation breathing: volume-controlled versus pressure-controlled ven- de Langue Française: monitorage de la ventilation mećanique. tilation. Am J Respir Crit Care Med 1996;153:1025–33 Reánim Urgences 2000;9:407–12 19. Brook AD, Ahrens TS, Schaiff R, Prentice D, Sherman G, 30. Esteban A, Alia I, Gordo F, de Pablo R, Suarez J, Gonzalez G, Shannon W, Kollef MH. Effect of a nursing-implemented Blanco J. Prospective randomized trial comparing pressure- sedation protocol on the duration of mechanical ventilation. controlled ventilation and volume-controlled ventilation in Crit Care Med 1999;27:2609–15 ARDS. For the Spanish Lung Failure Collaborative Group. 20. Heffner JE. A wake-up call in the intensive care unit. New Engl Chest 2000;117:1690–6 J Med 2000;342:1520–2 21. Jacobi J, Fraser GL, Coursin DB, Riker RR, Fontaine D, Witt- brodt ET, Chalfin DB, Masica MF, Bjerke HS, Coplin WM, Crippen DW, Fuchs BD, Kelleher RM, Marik PE, Nasraway SA Jr., Murray MJ, Peruzzi WT, Lumb PD. Clinical practice guidelines for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:119–41

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 921 Oxygen Delivery During Transtracheal Oxygenation: A Comparison of Two Manual Devices

Franc¸ois Lenfant, MD, PhD,* Didier Pe´an, MD,† Laurent Brisard, MD,† Marc Freysz, MD, PhD,* and Corinne Lejus, MD, PhD†

BACKGROUND: The Manujet™ and the ENK Oxygen Flow Modulator™ (ENK) deliver oxygen during transtracheal oxygenation. We sought to describe the ventilation characteristics of these 2 devices. METHODS: The study was conducted in an artificial lung model consisting of a 15-cm ringed tube, simulating the trachea, connected via a flow analyzer and an artificial lung. A 15-gauge transtracheal wire reinforced catheter was used for transtracheal oxygenation. The ENK and Manujet were studied for 3 minutes at respiratory rates of 0, 4, and 12 breaths/min, with and without the artificial lung, in a totally and a partially occluded airway. Statistical analysis was performed using analysis of variance followed by a Fisher exact test; P Ͻ 0.05 was considered significant. RESULTS: Gas flow and tidal volume were 3 times greater with the Manujet than the ENK (approximately 37 vs 14 L ⅐ minϪ1 and 700 vs 250 mL, respectively) and were not dependent on the respiratory rate. In the absence of ventilation, the ENK delivered a 0.6 Ϯ 0.1 L ⅐ minϪ1

constant gas flow. In the totally occluded airway, lung pressures increased to 136 cm H2O after 3 insufflations with the Manujet, whereas the ENK, which has a pressure release vent, generated acceptable pressures at a low respiratory rate (4 breaths/min) (peak pressure at 27.7 Ϯ 0.7 and Ϯ end-expiratory pressure at 18.8 3.8 cm H2O). When used at a respiratory rate of 12 breaths/min, the ENK generated higher pressures (peak pressure at 95.9 Ϯ 21.2 and Ϯ end-expiratory pressure at 51.4 21.4 cm H2O). In the partially occluded airway, lung pressures were significantly greater with the Manujet compared with the ENK, and pressures increased with the respiratory rate with both devices. Finally, the gas flow and tidal volume generated by the Manujet varied proportionally with the driving pressure. DISCUSSION: This study confirms the absolute necessity of allowing gas exhalation between 2 insufflations and maintaining low respiratory rates during transtracheal oxygenation. In the case of total airway obstruction, the ENK may be less deleterious because it has a pressure release vent. Using a Manujet at lower driving pressures may decrease the risk of barotrauma and allow the safe use of higher respiratory rates. (Anesth Analg 2010;111:922–4)

ranstracheal oxygenation may be lifesaving in “can- characterize these 2 devices in terms of oxygen flow, tidal not intubate/cannot ventilate” patients. The French volumes, and airway pressures. TSociety of Anesthesiologists has recently published updated guidelines for difficult airway management and recommends the use of dedicated devices for transtracheal METHODS oxygen delivery.1 Two devices are available, the ENK The Manujet III™ jet ventilator (VBM௡; Vitrolles, France) Oxygen Flow Modulator™ (ENK) and the Manujet™. Both was connected to a 4-bar oxygen source and the driving have been studied in a pig model.2 Both maintain oxygen- pressure was set at 3 bar. Manual activation of the trigger ation, but the ENK seemed to achieve better ventilation delivered oxygen to a transtracheal catheter. The ENK ௡ because of a continuous flow that provides CO2 washout Oxygen Flow Modulator™ (COOK , Charenton, France) between insufflations. Very little is known concerning the was connected to an oxygen wall flow regulator set at 15 lung pressures generated with these 2 devices.3 Because it L/min. Oxygen was delivered when the 5 holes in the is technically impossible to assess these variables in clinical ENK’s perforated tube were manually occluded. situations, we conducted this study in an artificial lung to The 2 devices were evaluated by connecting them to a 15-gauge (2-mm internal diameter, 75-mm length) wire reinforced transtracheal catheter (COOK). The catheter was From the *Department of Anesthesiology and Intensive Care, CHU de Dijon, inserted into the simulated trachea, a 15-cm ringed tube General Hospital, Dijon; and †Department of Anesthesiology, CHU de (Intersurgical, Fontenay sous Bois, France) occluded at one Nantes, Hotel Dieu, Nantes, France. end by a plastic cork. The other end of the tube was Accepted for publication June 8, 2010. connected to an IMT Medical Flow Analyzer PF-300™ (IMT Supported by departmental sources except for the PF-300™ flow analyzer, ௡ which was provided free of charge by SEBAC (Gennevilliers, France). Medical , Buchs, Switzerland) to measure flow rate, tidal Disclosure: The authors report no conflicts of interest. volume, peak pressure, mean pressure, and positive end- Address correspondence and reprint requests to Dr. Franc¸ois Lenfant, expiratory pressure. An adult SmartLung™ (IMT Medical) Department of Anesthesiology, CHU de Dijon, General Hospital, 3 rue du was connected to the analyzer output. The SmartLung Faubourg Raines, 21033 Dijon Cedex, France. Address e-mail to ϭ [email protected]. settings were as follows: total lung volume 1000 mL; ϭ ϭ Copyright © 2010 International Anesthesia Research Society airway resistance 5 mbar/L/s; and lung compliance 30 DOI: 10.1213/ANE.0b013e3181ee81b0 mbar/mL (Fig. 1).

922 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Table 1. Airway Pressure Delivered by the ENK Oxygen Flow Modulator™ During Total Airway Obstruction Mean pressure Mean pressure Respiratory during during rate (breaths/ inspiration Peak pressure expiration

min) (cm H2O) (cm H2O) (cm H2O) 0 13.2 Ϯ 0.0 4 21.9 Ϯ 4.4 27.7 Ϯ 0.7 18.8 Ϯ 3.8 12 62.7 Ϯ 24.3* 95.9 Ϯ 21.2* 51.4 Ϯ 21.4* * P Ͻ 0.05, 4 vs 12 breaths/min.

Table 2 shows tidal volumes, minute volumes, and mean and peak airway pressures during partial upper airway obstruction. Values were significantly higher with the Manujet. Table 3 shows how the Manujet gas flows and tidal volumes increased with an increase in driving pressure. Figure 1. Photograph showing the ENK Oxygen Flow Modulator™ device connected to the simulated trachea, the PF-300™ ﬂow DISCUSSION analyzer, and the artiﬁcial lung. The trachea is occluded at one end The Manujet delivers oxygen at a higher peak flow rate by a plastic cork. than the ENK. As a result, 1-second insufflations with the Manujet result in larger tidal and minute volumes and higher airway pressures than with the ENK. Our results Two investigators manually delivered 1-second insuffla- confirm the results of Flint et al.3 The tidal volumes we tions for 3 minutes at the rate of 4 breaths/min and 12 measured for the Manujet (650 mL) are close to those breaths/min with the plastic cork in place, simulating total calculated from the minute volumes reported by Flint et al.3 upper airway obstruction, and after making a 2-mm hole in (627 mL). For the ENK, we measured a 240-mL tidal the plastic cork, simulating partial upper airway obstruc- volume, whereas Flint et al. measured a 156-mL tidal tion. A metronome was used to keep the rate and the volume. This discrepancy may be explained by the differ- duration of the insufflations constant. Data were recorded ences in both the model and the methodologies used in the for 3 minutes with an insufflation rate of 4 breaths/min 2 studies. with the Manujet driving pressure set at 0.5, 1.0, 1.5, 2.0, 2.5, Yildiz et al.4 found that the oxygen that flows from the and 3 bar. The SmartLung was removed and the gas flow ENK between manual insufflations enhances oxygenation, delivery rates were recorded without insufflations (0 especially at low respiratory rates. Preussler et al.2 found breaths/min). that the constant flow improves CO2 elimination. Because All of the variables were sampled and recorded every 10 the Manujet generates higher airway pressure, it may result milliseconds using Flowlab™ software and transferred to a in better alveolar recruitment and better lung ventilation. personal computer as Excel™ files. Statistical analysis was Both investigators found that decreasing the Manujet driv- ௡ performed using StatPlus software (AnalystSoft, StatPlus ing pressure allows it to deliver higher respiratory rates Mac, Version 2008; http://www.analystsoft.com/fr). Con- without excessive airway pressure.2,4 Ϯ tinuous data were expressed as the mean SD. Data were Because the main risk of transtracheal oxygenation is analyzed using analysis of variance followed by a Fisher barotrauma, the Manujet should not be used in case of Ͻ exact test. P 0.05 was considered significant. total upper airway obstruction. The ENK seems to be less problematic because it delivers gas at a lower pressure RESULTS and has a pressure release vent that allows gas to escape The mean duration of the manually delivered insufflations between insufflations.5 However, even with the ENK, the was 1.01 Ϯ 0.06 second at 4 breaths/min and 0.95 Ϯ 0.05 peak and mean airway pressures may reach dangerously second at 12 breaths/min for the ENK, and 0.97 Ϯ 0.08 high levels at high respiratory rates (12 breaths/min) second at 4 breaths/min and 0.95 Ϯ 0.08 second at 12 during total airway obstruction. Increasing the respira- breaths/min for the Manujet. During insufflations, peak tory rate results in short expiratory times and, as a Ϫ flow was 36.0 Ϯ 6.3 L ⅐ min 1 with the Manujet and 13.7 Ϯ consequence, an increase in the volume of the gas Ϫ 2.42 L ⅐ min 1 with the ENK. When holes were not occluded, trapped in the lung. Our results confirm the absolute Ϫ the ENK delivered oxygen at the rate of 0.7 Ϯ 0.1 L ⅐ min 1. necessity of ensuring that the insufflated gas is exhaled Table 1 shows the airway pressure delivered by the ENK during the expiratory period.6 Sufficient exhalation can during total airway occlusion. With the Manujet, airway easily be checked by watching the chest fall before

pressure was 34 cm H2O after the first insufflation and 136 delivering a second tidal volume. cm H2O after the second, therefore the experiment was When using the Manujet during partial upper airway stopped. During total airway occlusion, airway pressures obstruction, great attention should be given to delivering were significantly lower with the ENK. insufflations that last Ͻ1 second and using low respiratory

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 923 Comparison of Two Transtracheal Oxygenation Manual Devices

Table 2. Gas Flow Rate During Inspiration, Tidal Volume, Minute Volume, and Mean and Peak Airway Pressures Measured with the Manujet and the ENK Oxygen Flow Modulator™ (ENK) During Partial Airway Obstruction Respiratory rate Flow Tidal volume Minute volume Mean pressure ⅐ ؊1 ⅐ ؊1 (breaths/min) (L min ) (mL) (L min ) (cm H2O) Peak pressure (cm H2O) ENK 4 12.4 Ϯ 2.0 263.0 Ϯ 26.6 0.91 Ϯ 0.01 11.3 Ϯ 1.8 14.1 Ϯ 0.3 Manujet 4 39.9 Ϯ 4.9 676.6 Ϯ 62.3 2.67 Ϯ 2.04 16.7 Ϯ 6.8 27.7 Ϯ 5.0 ENK 12 12.0 Ϯ 1.6 223.9 Ϯ 27.3 2.69 Ϯ 0.07 12.1 Ϯ 2.0 18.2 Ϯ 1.2 Manujet 12 38.8 Ϯ 5.8 753.4 Ϯ 88.9 9.06 Ϯ 0.02 16.7 Ϯ 6.8 58.3 Ϯ 25.0 All differences between Manujet and ENK were signiﬁcant at P Ͻ 0.05.

2. Preussler NP, Schreiber T, Huter L, Gottschall R, Schubert H, Table 3. Gas Flow Rate During Inspiration and Rek H, Karzai W, Schwarzkopf K. Percutaneous transtracheal Tidal Volume Versus Driving Pressure for the ventilation: effects of a new oxygen flow modulator on oxygen- Manujet at a Respiratory Rate of 4 breaths/min ation and ventilation in pigs compared with a hand triggered Driving pressure Gas ﬂow Tidal volume emergency jet injector. Resuscitation 2003;56:329–33 bar) (L ⅐ min؊1) (mL) 3. Flint NJ, Russell WC, Thompson JP. Comparison of different) 0.5 12.6 Ϯ 1.6 246.0 Ϯ 15.9 methods of ventilation via cannula cricothyroidotomy in a 1.0 17.4 Ϯ 2.5 304.0 Ϯ 34.9 trachea-lung model. Br J Anaesth 2009;103:891–5 1.5 23.7 Ϯ 2.9 356.0 Ϯ 25.2 4. Yildiz Y, Preussler NP, Schreiber T, Hueter L, Gaser E, Schubert 2.0 28.2 Ϯ 4.2 485.3 Ϯ 18.0 H, Gottschalll R, Schwarzkopf K. Percutaneous transtracheal 2.5 33.9 Ϯ 3.5 626.3 Ϯ 5.5 emergency ventilation during respiratory arrest: comparison of 3.0 37.1 Ϯ 5.8 653.3 Ϯ 50.3 the oxygen flow modulator with a hand-triggered emergency jet injector in an animal model. Am J Emerg Med 2006;24:455–9 5. Hamaekers A, Borg P, Enk D. The importance of flow and rates, to avoid excessive airway pressure. The most impor- pressure release in emergency jet ventilation devices. Paediatr Anaesth 2009;19:452–7 tant goal is to deliver oxygen to the lung, as recommended 6. Bourgain JL, Desruennes E, Fischler M, Ravussin P. Transtra- 7 by Cook and Alexander. cheal high frequency jet ventilation for endoscopic airway surgery: a multicentre study. Br J Anaesth 2001;87:870–7 REFERENCES 7. Cook TM, Alexander R. Major complications during anaesthesia 1. Combes X, Pean D, Lenfant F, Francon D, Marciniak B, Legras for elective laryngeal surgery in the UK: a national survey of the A. Difficult airway management devices: establishment and use of high-pressure source ventilation. Br J Anaesth maintenance—question 4. Societe Francaise d’Anesthesie et de 2008;101:266–72 Reanimation. Ann Fr Anesth Reanim 2008;27:33–40

924 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Two Serial Check Valves Can Prevent Cross-Contamination Through Intravenous Tubing During Total Intravenous Anesthesia

Oliver C. Radke, MD, PhD,*† Katrin Werth, MD,‡ Margarete Borg-von-Zepelin, MD, PhD,§ Petra Saur, MD, PhD,ʈ and Christian C. Apfel, MD, PhD†

BACKGROUND: Nonsterile handling of propofol for anesthesia has been linked with severe sepsis and death. Placing a single check valve in the IV tubing does not prevent retrograde ascension of pathogens into propofol-filled syringes, so we designed an IV tubing set with multiple check valves. To estimate the efficacy of this design, we measured the concentration of pathogens detected upstream in the IV tubing in relation to the pathogen concentration in a model of a contaminated patient. METHODS: A glass container with a rubber sealed port was filled with a suspension of either bacteria or phagocytes and kept at 37°C (“contaminated patient” model). A bag of normal saline was connected to an IV cannula, punctured through the rubber sealed port of the patient model. Two additional sidestream infusion lines were connected to syringes in 2 standard infusion pumps. One of the syringes contained propofol and the other contained normal saline as a substitute for an opioid preparation. After 5 hours of infusion, we obtained samples from different parts of the infusion lines and syringes. The samples were streaked out on blood agar plates and incubated at 37°C for 24 hours. We repeated this experiment with 6 different pathogens. RESULTS: We incubated 825 agar plates. Whereas the concentration of bacteria and phagocytes in the “patient” had significantly increased during the 5-hour experiments (positive control), no bacterial growth could be detected in any of the incubated plates. CONCLUSION: The data from this experimental setting suggest that the design with multiple check valves in paired configuration prevents retrograde contamination. Of note, this does not permit the reuse of propofol syringes because reusing is against the manufacturer’s recommendations. (Anesth Analg 2010;111:925–8)

ropofol infusions have been identified as a medium valves in each pathway would work in conjunction, ensur- supporting the growth of microorganisms.1 Contami- ing that the column of fluid between them did not move Pnation can occur during the preparation of the drug2 backward even if one of the valves failed. and during its administration.3–9 Clinicians often put a single To estimate the efficacy of this design, we conducted a 1-way valve into the IV line to prevent backflow but, unfor- series of experiments with a model of a contaminated tunately, these valves may not prevent ascension of bacteria patient. Our main objective was to measure the concentra- from the patient into the tubing and the syringes.10 tion of pathogens detected upstream in the IV tubing in To overcome the shortcomings of single check valves, relation to the pathogen concentration in the patient model. we propose placing 2 check valves in series.11 We hypothesized that the serial placement of check valves working in METHODS pairs would improve their function as a barrier to blood- Tests for Bacterial Contamination borne pathogens. Because fluids are not compressible, the 2 As a model of a contaminated patient, a 3000-mL glass container with a rubber-sealed port was filled with 2500 mL of a liquid medium (Fig. 1). The medium consisted of 25 g From the *Department of Anesthesiology and Critical Care Medicine, University Hospital Carl Gustav Carus Dresden at the TU Dresden, peptone, 5 g NaCl, 10 g dry meat extract, and 10 g glucose Dresden, Germany; †Perioperative Clinical Research Core, Department of in distilled water. For each experimental run, we infected Anesthesia and Perioperative Care, University of California at San Francisco, the medium with a bacteria suspension to achieve a con- UCSF Medical Center at Mount Zion, San Francisco, California; ‡Depart- ⅐ 6 Ϫ1 ment of Pediatrics, Klinikum Aschaffenburg, Aschaffenburg, Germany; centration of at least 1 10 mL colony-forming units §Institute of Infectiology & Pathobiology, Hufeland Klinikum GmbH, Stan- (CFUs). The bacteria were strains isolated from patients of ʈ dort Mu¨ hlhausen, Mu¨ hlhausen, Germany; and Department of Anesthesia, our university hospital (Staphylococcus aureus, Staphylococ- Intensive Care Medicine and Pain Medicine, Sana Kliniken Lu¨ beck GmbH, Lu¨ beck, Germany. cus epidermidis, Escherichia coli, Proteus mirabilis, and Pseudo- Accepted for publication May 28, 2010. monas aeruginosa). Funded by departmental resources. Medex Medical, now Smith Medical We assembled the IV tubing set (“TIVA-Set”; Smith International, Watford, UK, donated the IV tubings. B.Braun, Melsungen, Medical Deutschland GmbH, Grasbrunn, Germany) shown Germany, donated the infusion pumps for the experiments. AstraZeneca 11 GmbH, Wedel, Germany, donated the propofol. None of these companies in Figure 2 with 4 strategically placed check valves. A provided financial support. 500-mL bag of 0.9% NaCl was connected via the TIVA-Set Address correspondence and reprint requests to Oliver C. Radke, MD, to an 18-gauge IV cannula. This cannula was punctured PhD, DEAA, Klinik und Poliklinik fu¨ r Ana¨sthesiologie und Intensiv- through the rubber-sealed port of the patient model (Fig. 2). therapie, Fetscherstr. 74, 01307 Dresden, Germany. Address e-mail to [email protected]. The 2 infusion lines of the TIVA-Set were connected to Copyright © 2010 International Anesthesia Research Society 50-mL syringes (B.Braun, Melsungen, Germany) in Perfu- DOI: 10.1213/ANE.0b013e3181eb7194 sor௡ fm and perfusor compact infusion pumps (B.Braun).

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 925 Can Serial Check Valves Prevent Cross-Contamination?

Table 1. Infusion Program Carrier Propofol Saline Duration drip rate infusion infusion (min) (mL/h) rate (mL/h) rate (mL/h) Before induction 30 100 0 0 Maintenance I 120 100 10 10 Fluids run dry 30 0 10 10 Maintenance II 120 100 10 10 Standardized infusion program, controlled by a microprocessor. The program consisted of 4 parts mimicking a total IV anesthesia in clinical practice with low drug infusion rates. Total duration of each experimental run was 5 hours.

Before and after the infusion program, we took a sample of 100 ␮L from the bacteria suspension in the patient model and prepared a dilution series to estimate the bacteria concentrations. Additional undiluted samples of 500 ␮L were obtained under sterile conditions from the following locations (Fig. 2): IV tubing close to the cannula; both microtubings leading from the drug pumps to the main IV line; and leftover drugs in both 50-mL syringes. Each sample was spread out on 3 blood agar plates (100 ␮L per plate), and the plates were incubated at 37°C for 24 hours. At the end of the incubation, we manually counted the Figure 1. Photograph of the “contaminated patient” model. The rubber disc that seals the connector at the bottom has been number of CFUs on the plates. Bacteria concentrations in the punctured with a standard 16-gauge IV cannula. The thermometer at patient model were compared using the Student t test for the top is connected to the heated stirrer to ensure a constant repeated measurements (SPSS 16.0.2; SPSS Inc., Chicago, IL). temperature of 37°C. Test for Nonbacterial Contamination One of the syringes contained propofol 1% (Disoprivan௡; Because nonbacterial pathogens such as viruses or prions are AstraZeneca, Wedel, Germany) and the other contained much smaller than bacteria, we used a bacteriophage (Bacte- normal saline as a substitute for an opioid preparation. The riophage T3) as a sample of such a small infectious unit. The infusion pumps were approximately 30 cm above the level Bacteriophage T3 infects E coli bacteria and can be detected by of the port in the patient model. their ability to lyse the bacteria. The experimental runs with The flow of the normal saline (Table 1) was regulated the phage were conducted almost identically to the runs with by a drip pump (Infusomat fmS; B.Braun). All pumps bacteria, but because a phage cannot replicate on its own, we were operated by a microprocessor-controlled fluid man- added E coli B14 to the patient model. The blood agar plates ager (fm controller; B.Braun) that was programmed to for detection of pathogens in the samples were also flooded run a 5-hour infusion program (Table 1). This program with E coli B14 to provide the hosts for the phages’ replication. was designed to simulate an actual anesthetic. For each After incubation, the concentration of phages in a sample was pathogen, the experimental run was repeated at least 8 calculated from the number of lytic holes in the bacterial times (Table 2). spread on the blood agar plate.

Figure 2. Placement of check valves and sample sites. Schematic of the IV tubing. Four check valves are placed in paired conﬁgurations to prevent retrograde ﬂow. The 5 sample sites are marked.

926 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 2. Bacteria Concentrations in the Patient Model Growth rate Before After P value (95% CI) Runs Plates 106 ·mLϪ1 CFUs 106 ·mLϪ1 CFUs Staphylococcus aureus 12.3 Ϯ 6.3 110.6 Ϯ 4.0 Ͻ0.001 10.4 (8.1–12.6) 12 180 Staphylococcus epidermidis 9.4 Ϯ 1.2 34.0 Ϯ 25.2 0.028 3.6 (1.5–5.8) 8 120 Escherichia coli 50.7 Ϯ 38.8 249.4 Ϯ 135.4 Ͻ0.001 5.9 (3.8–8.0) 9 135 Proteus mirabilis 1.0 Ϯ 0.0 67.3 Ϯ 14.2 Ͻ0.001 67.3 (56.5–78.2) 8 120 Pseudomonas aeruginosa 20.7 Ϯ 4.0 66.6 Ϯ 12.0 Ͻ0.001 3.4 (2.4–4.5) 9 135 106 ·mLϪ1 PFUs 106 ·mLϪ1 PFUs Bacteriophage T3 1.4 Ϯ 1.7 37.8 Ϯ 43.7 0.015 97.2 (21.4–173.0) 9 135 CI ϭ conﬁdence interval; CFU ϭ colony-forming unit; PFU ϭ plaque-forming unit. Ϯ 6 6 Concentrations are given as mean standard deviation in 10 CFUs or 10 PFUs per mL. Growth rate is the CFUafter divided by CFUbefore. Total number of plates: 825.

RESULTS Validity of the Patient Model The test runs were repeated 8 to 12 times with each of the The pathogens tested in this study are representative of 5 bacterial isolates and the phage. We performed a total of infections found in the patients of our hospital. S aureus was 55 runs. For each bacterial strain, we found a statistically used because of its ubiquitous presence, accounting for 12% significant increase in concentration in the patient model of all nosocomial infections.17,18 S aureus strains with resis- (Table 2). Proteus mirabilis had the highest increase in tance to a wide spectrum of common antibiotics (not only concentration (67-fold). S aureus had the second highest methicillin-resistant S aureus) are an increasing problem in increase (10-fold), whereas the others increased their start- modern health care.19,20 S epidermidis is an important cause ing concentration 3- to 6-fold. The phages’ replication factor of infection in patients with a compromised immune sys- was even higher, ranging from 10 to 333 depending on the tem, or who have indwelling catheters. This bacterium has starting concentration. a special ability to produce a matrix (“biofilm”) that allows In 55 experimental runs, we obtained 275 samples from them to adhere to artificial surfaces, such as those of different parts of the IV tubing and the drug syringes and medical prostheses.21,22 E coli is frequently found in noso- incubated 825 plates. Even with the significant increase of comial infections, and strains that are highly resistant to the bacteria and phage concentrations in the patient model, antibiotics are increasingly isolated.20 Proteus mirabilis has no plates grew colonies or had lytic holes. an exceptionally high intrinsic motility and could potentially swarm actively into IV lines. Pseudomonas aeruginosa has a high intrinsic resistance to antibiotics and disinfec- DISCUSSION tants,20 and some strains can also generate a biofilm.23 Soon after propofol was introduced into clinical practice, The Bacteriophage T3 is a well-known phage without any there were several cases of severe sepsis resulting from the pathogenic potency for humans. It has approximately the use of propofol that was contaminated with bacteria (e.g., same size as a virus and therefore the same potential for see Veber et al.12). Contamination by the anesthesia pro- contamination. Its lytic effect on E coli bacteria makes it vider can be found in 8% to 11% of syringes whenever easy to detect. propofol is drawn up for clinical use,13 even when the The bacterial load in our contaminated patient model was 6 Ϫ1 manufacturer’s recommendations are closely adhered to.3 at least 1 ⅐ 10 CFUs mL . Studies in patients with endocar- 24 Because bacterial growth in propofol is slow at room ditis measured a maximum of 173 CFUs per mL of blood, temperature and begins after a latency period of 5 to 6 and children with manifested infections (meningitis, peritoni- 25 hours,14,15 the manufacturer recommends discarding any tis) had bacterial loads of up to 104 CFUs per mL of blood. propofol within 6 hours of opening the vial. Serious infec- Even blood samples from infected indwelling catheters mea- 26 tious outbreaks of sepsis were always attributed to propo- sured only up to 300 CFUs per mL. Therefore, our contami- Ͼ fol that had been opened several hours ahead of use nated patient model was 300 times more infectious than a (usually the day before); therefore, bacterial contamination worst case patient in real-life conditions. from opening the vials could grow to achieve bacterial counts that were high enough to cause severe sepsis. Validity of the IV Setup The risk of bacterial contamination by the anesthesia With our experimental setup, we tried to model real-life provider can be addressed by hand washing, aseptic tech- conditions during TIVA. We expected the risk for contami- niques, and using propofol within 6 hours after opening. nation to increase with duration of infusion. Also, high Transfer of pathogens from one patient to another by infusion rates would flush the IV tubing and make it less means of propofol syringes has not yet been documented, likely for pathogens to ascend. To facilitate the migration of but in vitro studies have shown that retrograde contami- bacteria from our contaminated patient model into the nation occurs easily.16 tubing, we ran an infusion pattern with low infusion rates With our experimental setup, we wanted to measure the over a period of 5 hours. incidence and the magnitude, if any, of retrograde bacterial It is obvious that an IV line without any check valves contamination through a special IV tubing set for total IV would not protect against contamination. In a survey of IV anesthesia (TIVA). tubings without check valves in the operating room, the

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 927 Can Serial Check Valves Prevent Cross-Contamination? authors found traces of blood from the patient in 3.3% of the 3. Lessard MR, Trepanier CA, Gourdeau M, Denault PH. A cases.4 In an experiment with a contaminated patient model microbiological study of the contamination of the syringes used in anaesthesia practice. Can J Anaesth 1988;35:567–9 similar to our setup but where the IV tubing was filled with 4. Trepanier CA, Lessard MR, Brochu JG, Denault PH. Risk of liquid culture medium instead of saline and propofol, the cross-infection related to the multiple use of disposable sy- researchers found a bacterial contamination of the line close to ringes. Can J Anaesth 1990;37:156–9 the patient in 87% of the cases,27 even though the line was 5. Zacher AN, Zornow MH, Evans G. Drug contamination from “protected” against backflow with a 1-way valve. The re- opening glass ampules. Anesthesiology 1991;75:893–5 6. Bach A. Syringe—or lead change for TCI? [in German]. searchers even increased the “venous” pressure of their model Anaesthesist 1998;47:434–6 to 150 mm Hg to simulate occlusion caused by noninvasive 7. el Mikatti N, Dillon P, Healy TE. Hygienic practices of consul- arterial blood pressure measurements.16 This maneuver did tant anaesthetists: a survey in the north-west region of the UK. not cause increased ascension of the pathogens into the Anaesthesia 1999;54:13–8 8. Halkes MJ, Snow D. Re-use of equipment between patients infusion tubing farther away from the patient. In our model, receiving total intravenous anaesthesia: a postal survey of we did not increase the venous pressure to such high read- current practice. Anaesthesia 2003;58:582–7 ings. Because the authors of the aforementioned article did not 9. Vonberg RP, Gastmeier P. Infection control measures in anaes- find an increase in contamination rates even with only a single thesia [in German]. Anasthesiol Intensivmed Notfallmed check valve, we do not expect to see an increase in contami- Schmerzther 2005;40:453–8 10. Gretzinger DT, Cafazzo JA, Ratner J, Conly JM, Easty AC. nation rates with our paired valves configuration. Validating the integrity of one-way check valves for the The lipid preparation containing propofol supports bac- delivery of contrast solution to multiple patients. J Clin Engl terial growth, but only after a latency period of 5 to 6 1996;21:375–82 hours.14,15 Additionally, the propofol preparation we used 11. Radke O. Das TIVA-Set. Anasthesiol Intensivmed 2003;44:787–8 12. Veber B, Gachot B, Bedos JP, Wolff M. Severe sepsis after contained EDTA, an additive that suppresses bacterial 28 intravenous injection of contaminated propofol. Anesthesiol- growth. The bacteriostatic effect of the EDTA could have ogy 1994;80:712–3 inhibited the growth of small amounts of bacteria that in 13. Magee L, Godsiff L, Matthews I, Farrington M, Park GR. fact entered the IV tubing and thus prevented their detec- Anaesthetic drugs and bacterial contamination. Eur J Anaes- tion. However, running a propofol/saline mixture in the IV thesiol Suppl 1995;12:41–3 14. Langevin PB, Gravenstein N, Doyle TJ, Roberts SA, Skinner S, tubing is closer to real-life conditions than running liquid Langevin SO, Gulig PA. Growth of Staphylococcus aureus in medium. Any bacteria that could not grow on our plates Diprivan and intralipid: implications on the pathogenesis of with optimal conditions for the growth of human patho- infections. Anesthesiology 1999;91:1394–400 gens probably would not have been capable of growing in 15. Sosis MB, Braverman B. Growth of Staphylococcus aureus in four intravenous anesthetics. Anesth Analg 1993;77:766–8 a human organism either. 16. Sim J, Choi Y, Yoon M, Lee D, Leem J. The peripheral venous pressure changes during non-invasive blood pressure mea- CONCLUSION surement. Can J Anaesth 1999;46:711–2 It is highly unlikely that during 5 hours of propofol anesthesia 17. Bouza E. Intravascular catheter-related infections: a growing the syringe filled with propofol and EDTA will become problem, the search for better solutions. Clin Microbiol Infect 2002;8:255 contaminated by pathogens from the patient’s blood if 2 check 18. Emori TG, Gaynes RP. An overview of nosocomial infections, valves are placed in series in the IV line. We did not find a including the role of the microbiology laboratory. Clin Micro- single trace of contamination in any of our 825 samples. biol Rev 1993;6:428–42 19. Gastmeier P, Geffers C. Nosocomial infections in Germany: AUTHOR CONTRIBUTIONS what are the numbers, based on the estimates for 2006? [in German] Dtsch Med Wochenschr 2008;133:1111–5 OCR: Inventor of the TIVA-Set, initiator of the study, and the 20. Toniolo A, Endimiani A, Luzzaro F. Microbiology of postop- main author of the manuscript. KW: Performed all the micro- erative infections. Surg Infect (Larchmt) 2006;7:S13–6 biological testing and did additional literature research. 21. Ziebuhr W, Hennig S, Eckart M, Kranzler H, Batzilla C, MB-v-Z: Designed and supervised the microbiological testing. Kozitskaya S. Nosocomial infections by Staphylococcus epider- PS: Coauthor of the manuscript and the advisor of study midis: how a commensal bacterium turns into a pathogen. Int design. CCA: Senior author of the manuscript, statistical J Antimicrob Agents 2006;28:S14–20 22. Pascual A. Pathogenesis of catheter-related infections: lessons analysis and interpretation of data. All authors contributed to for new designs. Clin Microbiol Infect 2002;8:256–64 the drafting of the article and gave final approval of the version 23. O’Toole GA, Kolter R. Flagellar and twitching motility are to be published. necessary for Pseudomonas aeruginosa biofilm development. Mol Microbiol 1998;30:295–304 DISCLOSURE 24. Werner AS, Cobbs CG, Kaye D, Hook EW. Studies on the The University of Go¨ttingen, Germany, holds a patent on the bacteremia of bacterial endocarditis. JAMA 1967;202:199–203 25. Santosham M, Moxon ER. Detection and quantitation of bac- IV tubing set and is required by German law to share a fraction teremia in childhood. J Pediatr 1977;91:719–21 of all royalties with the inventor, Dr. Oliver C. Radke. To avoid 26. Flynn PM, Shenep JL, Barrett FF. Differential quantitation with a conflict of interest, Dr. Radke was not involved in the a commercial blood culture tube for diagnosis of catheter- microbiological experiments. related infection. J Clin Microbiol 1988;26:1045–6 27. Eichler W, Schumacher J, Ohgke H, Klotz KF. Reuse of a set for total intravenous anaesthesia: safe against bacterial contami- REFERENCES nation? Eur J Anaesthesiol 2004;21:501–3 1. Thomas DV. Propofol supports bacterial growth. Br J Anaesth 28. Hart B. ‘Diprivan’: a change of formulation. Eur J Anaesthesiol 1991;66:274 2000;17:71–3 2. Lorenz IH, Kolbitsch C, Lass-Florl C, Gritznig I, Vollert B, Lingnau W, Moser PL, Benzer A. Routine handling of propofol prevents contamination as effectively as does strict adherence to the manufacturer’s recommendations. Can J Anaesth 2002;49:347–52

928 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA TECHNICAL COMMUNICATION Robot-Assisted Airway Support: A Simulated Case

Patrick J. Tighe, MD, S. J. Badiyan, MD, I. Luria, BS, MS, S. Lampotang, PhD, and S. Parekattil, MD

Recent advances in telemedicine and robotically assisted telesurgery may offer advanced surgical care for the geographically remote patient. Similar advances in tele-anesthesia will be necessary to optimize perioperative care for these patients. Although many preliminary investigations into tele-anesthesia are underway, none involves remote performance of anesthesia-related procedures. Here we describe simulated robotically assisted fiberoptic intubations using an airway simulation mannequin. Both oral and nasal approaches to fiberoptic intubation were successful, but presented unique opportunities and challenges inherent to the robot’s design. Robotically assisted airway management is feasible using multipurpose surgical robotic systems. (Anesth Analg 2010;111:929–31)

elemedicine has improved access to consultant-level fiberoptic bronchoscope (Karl Storz Endoscopy, El Seg- medical care by minimizing geographic obstacles. undo, California). The trocars used for mounting standard TSimilar advances in telesurgery could offer expert robotic instruments were removed from this arm, allowing surgical care for the geographically remote patient. Such the bronchoscope handle to slide into the mounting surgical advances have been historically preceded by simi- bracket. The bronchoscope handle was oriented to keep the lar advances in anesthesia. True to this paradigm, early tip actuator lever facing away from the arm, with the investigations into tele-anesthesia are already underway.1 suction port positioned to the side (Fig. 3). An external However, none of these efforts has tackled one of the brace was required to keep the bronchoscope firmly seated central tenets of anesthesiology: airway management. within the arm during manipulation of the actuator. In this report we present what we believe to be the first We manually loaded the endotracheal tube onto the description of a simulated robotically assisted fiberoptic bronchoscope. A camera was attached to the bronchoscope intubation. Instead of a procedure-specific device, we used and connected to the DVS Tilepro multivideo input system the multipurpose DaVinci Surgical System type S (DVS) (Intuitive Surgical, Sunnyvale, California), allowing simul- (Intuitive Surgical, Sunnyvale, California). This system taneous viewing of both the robot camera and broncho- incorporates 4 separate robotic arms, 1 of which is mated to scope camera in a single three-dimensional view (Fig. 4). a high-definition stereoscopic camera. The workstation Before attempting intubation, the urologist spent ap- allows the person performing the procedure to view the proximately 2 hours performing robotic dexterity exercises robot’s camera output, control the limbs, and receive simul- with the robotic surgical instruments and operator console. taneous video input from third-party sources.2,3 The DVS is After training was complete, an anesthesiologist manually already in widespread clinical use for a variety of urologic, placed the bronchoscope tip within the oropharnyx, and a gynecologic, and cardiothoracic surgical procedures.4 In urologist (S. Parekattil) used robot arms 2 and 3 to adjust this study involving an airway mannequin, we successfully the bronchoscope actuator and steer the bronchosc- used the DVS for both oral and nasal fiberoptic intubation. ope tip into the hypopharynx and through the vocal cords (Video 1; see Supplemental Digital Content 1, http://links.lww.com/AA/A171; see the Appendix for METHODS video legend). Next the bronchoscope tip was placed An adult airway mannequin was placed at the head of a external to the nare, an anesthesiologist manually ad- standard operating room bed. A stereoscopic video camera vanced and rotated the bronchoscope, and an urologist (DaVinci Surgical System, Intuitive Surgical) was mated to used the robot to steer the tip into the hypopharynx and the first robotic arm. This arm was situated above the through the vocal cords. mannequin in a sagittal plane, angled to view the mannequin from a caudal to rostral fashion. The second and third arms were equipped with large and small graspers (Figs. 1 RESULTS and 2). The fourth arm was equipped with a standard Two intubation attempts were completed for this demonstration. During oral intubation, it took 75 seconds to From the University of Florida College of Medicine, Gainesville, Florida. advance the bronchoscope tip from the oropharynx to Accepted for publication June 8, 2010. carinal visualization. For nasal intubation, it took 67 Funding was provided by the University of Florida College of Medicine, seconds to advance the tip from nasal entry to carinal Department of Anesthesiology, as well as by National Institutes of Health visualization. grant UL1 RR029890 Clinical and Translational Science Award, NIH (NCRR). Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions DISCUSSION of this article on the journal’s Web site (www.anesthesia-analgesia.org). Fiberoptic intubation is feasible with robotic equipment. Address correspondence to Patrick J. Tighe, MD, University of Florida College of Medicine, Department of Anesthesiology, PO Box 100254, Gaines- We did not encounter significant differences between the ville, FL 32610-0254. Address e-mail to [email protected]. nasal and oral intubation routes. Even if optimized for Copyright © 2010 International Anesthesia Research Society anesthetic practice, robotic-assisted anesthetic procedures DOI: 10.1213/ANE.0b013e3181ef73ec are not likely to become a part of routine anesthetic

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 929 TECHNICAL COMMUNICATION

Figure 1. Schematic depicting the arrangement of the DaVinci Surgical System type S (DVS), bronchoscope, and airway mannequin.

Figure 4. Image from the DaVinci workstation during the robotically assisted ﬁberoptic intubation. The robot operator can simultaneously visualize video output from both the bronchoscope and the robot’s main camera without turning attention away from the workstation.

practice. Their optimal role may ultimately be for environments hazardous to routine anesthetic practice, such as the battlefield or space-based environments.5 Our initial attempts at robotic-assisted intubation focused on direct laryngoscopy. However, we had considerable difficulty using the DVS robotic graspers to lift and manipulate both Macintosh and Miller laryngoscope light handles. Direct endotracheal tube manipulation with the DVS was quite challenging. On the other hand, the focus of the DVS on small-scale manipulation, coupled with its multiaxis flexibility and unifying video input, suggested that fiberoptic approaches to airway management would capitalize on the capabilities of the DVS. The robot operator controls the DVS through a dedi- Figure 2. Schematic depicting the arrangement of the DaVinci cated workstation in a corner of the operating room. The Surgical System type S (DVS), bronchoscope, and airway workstation could easily be placed in another room, build- mannequin. ing, or continent.6,7 Traditionally, such distances have been limited by latencies between user input, robot action, and robot-user feedback.8 Progress in adapting data packet transmission along existing telecommunication systems has minimized such limitations,9 allowing long-distance robotic telesurgery with acceptable latencies. 10 Ideally, this exercise would have required zero nonro- botic interventions. However, because of the cost of the bronchoscope, we elected to accept a loss in simulation fidelity to minimize potential damage to the costly equipment. Further work will be necessary to explore how well the DVS can rotate and advance a bronchoscope tip. Our experience indicates that because of the support and preparation required for robotically assisted intubation, current systems are not ready for such deployment into complex operating environments. The bedside presence of the anesthe- Figure 3. The handle of the bronchoscope was positioned into a siologist was necessary even in this focused simulation to robotic arm. Two other robotic arms were ﬁtted with graspers and assist with important maneuvers necessary during robotically used to control the ﬂexion and extension of the bronchoscope tip. assisted fiberoptic intubation. Furthermore, this exercise did

930 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Robot-Assisted Airway Support not include important factors such as patient preparation, 3. Rogers CG, Laungani R, Bhandari A, Krane LS, Eun D, Patel positioning, topical and systemic medication administration, MN, Boris R, Shrivastava A, Menon M. Maximizing console surgeon independence during robot-assisted renal surgery by and patient monitoring. using the fourth arm and TilePro. J Endourol 2009;23:115–22 This study demonstrated that a multipurpose surgical 4. Palep JH. Robotic assisted minimally invasive surgery. J Minim robot could be adapted for use in airway management. Access Surg 2009;5:1–7 Although limited in its approach to direct laryngoscopy, 5. Harnett BM, Doarn CR, Rosen J, Hannaford B, Broderick TJ. the DVS was able to assist with both oral and nasal Evaluation of unmanned airborne vehicles and mobile robotic telesurgery in an extreme environment. Telemed e-Health fiberoptic intubation. Future studies will be necessary to 2008;14:539–44 optimize robotic interfaces with other airway management 6. Sterbis JR, Hanly EJ, Herman BC, Marohn MR, Broderick TJ, techniques. Shih SP, Harnett B, Doarn C, Schenkman NS. Transcontinental telesurgical nephrectomy using the da Vinci robot in a porcine APPENDIX: VIDEO CAPTIONS model. Urology 2008;71:971–3 7. Marescaux J, Leroy J, Gagner M, Rubino F, Mutter D, Vix M, Video 1. This video demonstrates how the robotic manipu- Butner SE, Smith MK. Transatlantic robot-assisted telesurgery. lation arms adjusted the flexion and extension of the Nature 2001;413:379–80 bronchoscope tip, and how the bronchoscope itself was 8. Anvari M, Broderick T, Stein H, Chapman T, Ghodoussi M, secured to a third arm. Both oral and nasal intubations are Birch DW, Mckinley C, Trudeau P, Dutta S, Goldsmith CH. The impact of latency on surgical precision and task comple- demonstrated. The bronchoscope advancement was per- tion during robotic-assisted remote telepresence surgery. formed manually to avoid damage to the bronchoscope and Comp Aided Surg 2005;10:93–9 required very minimal human input. Advancement was 9. Rayman R, Primak S, Patel R, Moallem M, Morady R, Tavakoli performed at the direction of the robot operator. M, Subotic V, Galbraith N, van Wynsberghe A, Croome K. Effects of latency on telesurgery: an experimental study. Medi- cal image computing and computer-assisted intervention. REFERENCES MICCAI 2005;8:57–64 1. Hemmerling TM. Automated anesthesia. Curr Opinion Anaes- 10. Rayman R, Croome K, Galbraith N, McClure R, Morady R, thesiol 2009;103:811–6 Peterson S, Smith S, Subotic V, Wynsberghe AV, Primak S. 2. Bhayani S, Snow D. Novel dynamic information integration Long-distance robotic telesurgery: a feasibility study for care in during da Vinci robotic partial nephrectomy and radical ne- remote environments. Intl Med Robotics Comp Assist Surg phrectomy. J Robotic Surg 2008;2:67–9 2006;2:216–24

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 931 Anesthesia Patient Safety Foundation

Section Editor: Sorin J. Brull SPECIAL ARTICLE

Personal Protective Equipment for Care of Pandemic Inﬂuenza Patients: A Training Workshop for the Powered Air Purifying Respirator

Bonnie M. Tompkins, MD,* and John P. Kerchberger, MD†

Virulent respiratory infectious diseases may present a life-threatening risk for health care professionals during aerosol-generating procedures, including endotracheal intubation. The 2009 Pandemic Influenza A (H1N1) brings this concern to the immediate forefront. The Centers for Disease Control and Prevention have stated that, when performing or participating in aerosol-generating procedures on patients with virulent contagious respiratory diseases, health care professionals must wear a minimum of the N95 respirator, and they may wish to consider using the powered air purifying respirator (PAPR). For influenza and other diseases transmitted by both respiratory and contact modes, protective respirators must be combined with contact precautions. The PAPR provides 2.5 to 100 times greater protection than the N95, when used within the context of an Occupational Safety and Health Administration–compliant respiratory protection program. The relative protective capability of a respirator is quantified using the assigned protection factor. The level of protection designated by the APF can only be achieved with appropriate training and correct use of the respirator. Face seal leakage limits the protective capability of the N95 respirator, and fit testing does not assure the ability to maintain a tight face seal. The protective capability of the PAPR will be defeated by improper handling of contaminated equipment, incorrect assembly and maintenance, and improper don (put on) and doff (take off) procedures. Stress, discomfort, and physical encumbrance may impair performance. Acclimatization through training will mitigate these effects. Training in the use of PAPRs in advance of their need is strongly advised. “Just in time” training is unlikely to provide adequate preparation for groups of practitioners requiring specialized personal protective equipment during a pandemic. Employee health departments in hospitals may not presently have a PAPR training program in place. Anesthesia and critical care providers would be well advised to take the lead in working with their hospitals’ employee health departments to establish a PAPR training program where none exists. User instructions state that the PAPR should not be used during surgery because it generates positive outward airflow, and may increase the risk of wound infection. Clarifica- tion of this prohibition and acceptable solutions are currently lacking and need to be addressed. The surgical hood system is not an acceptable alternative. We provide on line a PAPR training workshop. Supporting information is presented here. Anesthesia and critical care providers may use this workshop to supplement, but not substitute for, the manufacturers’ detailed use and maintenance instructions. (Anesth Analg 2010;111:933–45)

xposure to virulent respiratory pathogens during in- PPE (personal protective equipment) for Healthcare Workers vasive airway procedures may present a life- stated that, “There is an urgent need to address the lack of Ethreatening risk for health care providers. The Institute preparedness regarding effective PPE for use in an influenza of Medicine (IOM) of the National Academies Committee on pandemic.” (Table 1, Ref. 1). In failing to anticipate and address PPE issues, health care workers may threaten not only personal, colleague, and family safety, but patient safety From the *Department of Anesthesiology, University of Wisconsin School of as well.1 Safe and effective use of PPE can only be optimized Medicine and Public Health, Madison, Wisconsin; and †Department of Anesthesiology, Rush University Medical Center, Chicago, Illinois. by practicing correct procedures and by being aware of the Accepted for publication May 6, 2010. operational factors that defeat its protective capacity. Disclaimer: Neither author has a financial relationship with any device manufacturer. Disclosure: The authors report no conflicts of interest. VIRAL PANDEMICS AND OUTBREAKS—PAST Address correspondence and reprint requests to Bonnie M. Tompkins, MD, AND PRESENT Department of Anesthesiology, University of Wisconsin Hospitals, 600 High- land Ave., Madison, WI 53792-0001. Address e-mail to [email protected]. Influenza pandemics typically occur several times each Copyright © 2010 International Anesthesia Research Society century. The 1957 and 1968 pandemics (Pandemic Severity DOI: 10.1213/ANE.0b013e3181e780f8 Index [PSI] 2, case fatality ratio [CFR] 0.1ϪϽ0.5%) did not

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 933 SPECIAL ARTICLE

Table 1. Website References Reference no. Website 1 Institute of Medicine (IOM) 2008. Preparing for an Influenza Pandemic: Personal Protective Equipment for Healthcare Workers. Washington, DC: The National Academies Press. Available at: http://www.nap.edu/catalog.php?record_idϭ11980. Accessed February 21, 2010 2 Interim Pre-pandemic Planning Guidance: Community Strategy for Pandemic Influenza Mitigation in the United States. Available at: http://pandemicflu.gov/professional/community/commitigation.html#IV. Accessed February 21, 2010 3 Department of Heath and Human Services (U.S. DHHS) Pandemic Influenza Plan 2005. Available at: http://hhs.gov/ pandemicflu/plan/appendixb.html. Accessed February 22, 2010 4 Occupational Safety and Health Administration (OSHA). Pandemic Influenza Preparedness and Response Guide 2007. Available at: http://www.osha.gov/Publications/3328-05-2007-English.html. Accessed February 22, 2010 5 Centers for Disease Control and Prevention (CDC). Estimates of 2009 H1N1 Influenza Cases, Hospitalizations, and Deaths in the United States. Available at: http://www.cdc.gov/h1n1flu/estimates/results_2009_h1n1.htm. Accessed January 16, 2010 6 Cumulative Number of Confirmed Human Cases of Avian Influenza A/(H5N1) Reported to World Health Organization (WHO). Available at: http://www.who.int/csr/disease/avian_influenza/country/cases_table_2010_02_17/en/index.html. Accessed February 18, 2010 7 WHO/Pandemic Influenza Preparedness and Response. Available at: http://www.who.int/csr/disease/influenza/ PIPGuidance09.pdf. Accessed February 21, 2010 8 Center for Infectious Disease Research and Policy (CIDRAP). Avian Influenza (Bird Flu): Implications for Human Disease. Last updated April 2, 2010. Available at: http://www.cidrap.umn.edu/cidrap/content/influenza/avianflu/biofacts/ avflu_human.html. Accessed February 21, 2010 9 WHO Summary of Probable SARS Cases with Onset of Illness from 1 November 2002 to 31 July 2003. Available at: http:// www.who.int/csr/sars/country/table2004_04_21/en/index.html. Accessed February 22, 2010 10 The SARS Commission Executive Summary: Spring of Fear. Available at: http://www.health.gov.on.ca/english/public/pub/ ministry_reports/campbell06/online_rep/V1.html. Accessed September 28, 2009 11 Adalja AA. Will the D225G Mutation Herald More Severe Illness in Patients with 2009 H1N1 Influenza? Clinician’s Biosecurity Network Report. Available at: [email protected]. Accessed December 11, 2009 12 1918 Polymorphism in Ukraine H1N1? Available at: http://www.recombinomics.com/News/11090902/Ukraine_1918.html. Accessed December 11, 2009 13 Siegel JD, Rhinehart E, Jackson M, Chiarello L, and the Hospital Infection Control Practices Advisory Committee (HICPAC). 2007 Guideline for Isolation Precautions: Preventing Transmission of Infectious Agents in Healthcare Settings. Available at: http://www.cdc.gov/ncidod/dhqp/pdf/guidelines/Isolation2007.pdf. Accessed February 20, 2010 14 Interim Domestic Guidance on the Use of Respirators to Prevent Transmission of SARS May 3, 2005. Available at: http:// www.cdc.gov/ncidod/sars/respirators.htm. Accessed February 19, 2010 15 IOM 2009. Respiratory Protection for Healthcare Workers in the Workplace Against Novel H1N1 Influenza A: A Letter Report. Washington, DC: The National Academies Press. Available at: http://www.nap.edu/catalog/12748.html. Accessed February 22, 2010 16 IOM 2006. Reusability of Facemasks During an Influenza Pandemic. Board of Health Sciences Policy. Washington, DC: The National Academies Press. Available at: http://books.nap.edu/openbook.php?record_idϭ11637&pageϭR1. Accessed February 22, 2010 17 Interim Guidance on Planning for the Use of Surgical Masks and Respirators in Health Care Settings during an Influenza Pandemic. Oct. 2006 Available at: http://www.flu.gov/professional/hospital/maskguidancehc.html#airborne. Accessed Feb 20, 2010 18 Proposed Guidance on Workplace Stockpiling of Respirators and Facemasks for Pandemic Influenza. Available at: http:// www.osha.gov/dsg/guidance/proposedGuidanceStockpilingRespirator.pdf. Accessed February 21, 2010 19 Infection Prevention and Control in Health Care for Confirmed or Suspected Cases of Pandemic (H1N1) 2009 and Influenza- Like Illnesses 25 June 2009. Available at: http://www.who.int/csr/resources/publications/SwineInfluenza_infectioncontrol.pdf. Accessed February 21, 2010 20 Interim Guidance on Infection Control Measures of 2009 H1N1 Influenza in Healthcare Settings, Including Protection of Healthcare Personnel. October 14, 2009. Available at: http://www.cdc.gov/h1n1flu/guidelines_infection_control.htm. Accessed February 21, 2010 21 OSHA Instruction Directive #: CPL-02-02-075 Effective Nov. 11, 2009. Enforcement Procedures for High to Very High Occupational Exposure Risk to 2009 H1N1 Influenza. Available at: http://www.osha.gov/OshDoc/Directive_pdf/ CPL_02_02–075.pdf. Accessed February 18, 2010 22 OSHA, Appendix C, Pandemic Influenza Preparedness and Response Guide 2007 Implementation and Planning for Respiratory Protection Programs in Healthcare Settings. Available at: http://www.osha.gov/Publications/OSHA_pandemic_health.pdf. Accessed February 22, 2010 23 OSHA State Occupational Safety and Health Plans. Available at: http://www.osha.gov/dcsp/osp/index.html. Accessed February 22, 2010 24 Assigned Protection Factors for the Revised Respiratory Protection Standard OSHA 3352–02 2009. Available at: http:// www.osha.gov/Publications/3352-APF-respirators.html. Accessed February 22, 2010 25 King JH. Assessing Powered Air Purifying and Supplied Air Respirator Performance: An Employer’s Guide to OSHA’s Final Rule on Assigned Protection Factors for Respirators. An E.D. Bullard Company White Paper. September 21, 2006. Available at: http://www.bullard.com/V3/resources/downloads/respiratory_dwnlds.php#Technical_Publications. Accessed February 22, 2010 26 3M™ Technical Data Bulletin #160; Reusable Respirator Facepieces and Powered Air Purifying Respirator Systems (PAPRs) in the HealthCare Environment: Considerations for Use; and TDB #175 Assigned Protection Factors. 3M Occupational Health & Environmental Safety Respiratory Protection Media Library. Available at: http://www.3M.com/occsafety. Accessed February 21, 2010 (Continued)

934 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Personal Protective Equipment for Pandemic Influenza

Table 1. (Continued) Reference no. Website 27 Understanding Respiratory Protection Against SARS. Available at: http://www.cdc.gov/niosh/npptl/topics/respirators/ factsheets/respsars.html. Accessed February 26, 2010 28 Adalja AA. Update on Personal Protective Equipment. Available at: http://www.upmc-cbn.org/report_archive/2008/ 12_December_2008/cbnreport_12192008.html. Accessed February 21, 2010 29 General Procedures for Putting On and Taking Off a Disposable Respirator. Available at: http://www.cdc.gov/flu/ freeresources/2009-10/pdf/n95respirator_instructions.pdf. Accessed February 19, 2010 30 3M Air-Mate™. Available at: http://www.3m.com/occsafety. Accessed February 21, 2010 31 3M Resource CD. Available on request from 3M Occupational Health and Environmental Safety Division Technical Service at 800-243-4630 and from [email protected]. Available at: http://www.3M.com/occsafety. Accessed October 11, 2009 32 Powered Air Purifying System PA20™. Available at: http://www.bullard.com/Respiratory/papr/pa20page.shtml. Accessed October 11, 2009 33 Bullard EVA™ Competitive Comparison. Available at: http://www.bullard.com/V3/resources/downloads/respiratory_ dwnlds.php#EVA. Accessed February 5, 2010 34 PAPR Training Workshop. Available at: http://www.med.wisc.edu/papr-workshop. Accessed October 12, 2009 35 OSHA Best Practices for Hospital-Based First Receivers of Victims from Mass Casualty Incidents Involving the Release of Hazardous Substances. Available at: http://www.osha.gov/dts/osta/bestpractices/html/hospital_firstreceivers.html. Accessed October 11, 2009 36 Full Barrier Personal Protective Equipment (PPE) with Powered Air Purifying Respirator (PAPR). Available at: http:// www.health.state.mn.us/divs/idepc/dtopics/infectioncontrol/ppe/ppepapr.html. Accessed September 28, 2009 37 3M Technical Data Bulletin #178: Maintenance and Care of 3M Powered Air Purifying Respirator (PAPR) Batteries. Published March 2007, Revised November 2008. Available at: http://www.3M.com/occsafety. Accessed February 21, 2010 38 Maintenance of Battery Packs for Bullard Powered Air-Purifying Respirators (PAPRs). Available at: http://www.bullard.com. Accessed February 21, 2010 39 Bullard Technical Advisory: Release of Filtered Particles. Available at: http://www.bullard.com. Accessed February 21, 2010 40 Stryker T4 Surgical Helmet System Filtration Testing Summary Report. Available at: www.sars.medtau.org/strykerreport.doc. Accessed August 31, 2009 41 Using the Stryker T4 Personal Protection System for High-Risk Procedures During SARS Outbreaks. Available at: sars.medtau.org/strykertraining.htm. Accessed April 16, 2010 42 Questions and Answers About CDC’s Interim Guidance on Infection Control Measures for 2009 H1N1 Influenza in Healthcare Settings, Including Protection of Healthcare Personnel. December 1, 2009. Available at: http://www.cdc.gov/h1n1flu/ guidance/control_measures_qa.htm. Accessed February 21, 2010 43 Personal Protective Equipment in Pandemic/Avian Influenza/SARS: N95 or PAPR for Intubation? ASA Newsletter. Available at: http://www.asahq.org/Newsletters/2008/01-08/tompkins01-08.html. Accessed September 28, 2009

approach the catastrophic nature of the 1918 Great Influ- The severe acute respiratory syndrome coronavirus enza, “Spanish Flu” (PSI 5, CFR Ն2%), wherein one-third (SARS-CoV) epidemics of 2002 to 2003 incurred Ͼ8000 (approximately 500 million) of the world’s population cases and 700 deaths in 29 countries (CFR 9.6%) (Table 1, became ill and 50 to 100 million died (Table 1, Refs. 2–4).2,3 Ref. 9). Twenty percent of the cases were in health care The Centers for Disease Control (CDC) midrange esti- workers.4 In Ontario, Canada, Ͼ50% of the 438 SARS cases mates for the 2009 Pandemic Influenza A (H1N1) (hereafter and 3 of the 43 deaths were in health care workers (Table 1, 2009 H1N1) in the United States from April 2009 to January Ref. 10).5–7 The SARS experience is of particular import to 2010 are 57 million total cases, 257,000 (0.45%) hospitaliza- anesthesiology and critical care specialists.1 tions, and 11,700 deaths (overall CFR 0.02%, hospitalized CFR 4.5%) (Table 1, Ref. 5). In seasonal influenza, the greatest mortality is among the very young and the very VIRULENCE FACTORS IN SEVERE VIRAL old; in contrast, the 1918 and 2009 H1N1 pandemics RESPIRATORY ILLNESS predominantly affected children and younger adults (Table The history and pathology of influenza have been studied 1, Refs. 3 and 5).2 and presented effectively by Taubenberger and Morens.8 The World Health Organization (WHO) reported that 478 Both pandemic and seasonal influenza viruses may repli- human confirmed cases and 286 deaths (CFR 60%) occurred cate throughout the respiratory tract. Disease is limited to from 2003 through February 2010 from the H5N1 highly the upper respiratory tract and trachea in nonfatal cases, pathogenic avian influenza (HPAI) (“bird flu”). Many cases but fatal cases show lung involvement.8,9 The 1918 and were in children and young adults (Table 1, Ref. 6). This virus 2009 H1N1 pandemic and H5N1 HPAI viruses exhibit a has not currently developed a high affinity for human respi- greater propensity than seasonal influenza to bind to viral ratory tract receptors; therefore, human-to-human transmis- receptors in the lower respiratory tract.3,8–11 The histopa- sion is nonsustained and cases have occurred only in small thology from autopsy of influenza victims is that of a clusters (WHO pandemic phase 3). Should H5N1 HPAI and primary viral hemorrhagic bronchitis and pneumonia with 2009 H1N1 occur simultaneously in the same individual, a diffuse alveolar damage and destruction.8,9,12 Secondary highly pathogenic reassortant pandemic strain could emerge bacterial pneumonia contributes prominently to the mor- (Table 1, Refs. 7 and 8). tality in seasonal and pandemic influenza.8,9,12

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 935 SPECIAL ARTICLE

The genomes of the 1918 and present-day influenza The term “airborne transmission” traditionally refers to viruses have been reconstructed, studied, and compared, the remote transmission and inhalation of yet smaller presenting opportunity for identifying preventive and droplets and aerosol-sized (respirable) particles that may therapeutic approaches. This work has clarified how the access the alveoli as well as the upper airway and may virus binds to lower as well as upper respiratory tract remain suspended in the air for an indeterminate time and receptors and has identified gene components associated distance (Table 1, Ref. 13).20 Airborne transmission of any with increased virulence of the 1918 pandemic, H5N1 respiratory infectious disease is difficult to prove or to avian, and 2009 H1N1 pandemic influenza viruses (Table 1, definitively exclude.23 Isolated specific outbreaks of SARS Refs. 11 and 12).3,8,11,13 Identified genetic amino acid se- (aerosolized remote fecal source) and influenza (airplane quences associated with host specificity and high virulence outbreaks) are believed to be examples of unusual, or may provide a predictive monitoring tool.14 opportunistic, airborne transmission; aerosol research sup- In addition to diffuse lung damage, altered immune ports the possibility of remote airborne transmission of mechanisms and a viral-associated extreme proinflammatory influenza and SARS (Table 1, Refs. 1, 14, 15, and 17).20,22,23 cytokine response, sometimes with hemophagocytosis, have Transmission and subsequent infection is influenced by been observed in young, previously healthy influenza pa- additional factors that include viability of the agent, expi- tients who died or were critically ill.8,15 These pathologic ratory force, distance from and duration of exposure to the features also have been described in animal models inocu- source, and environmental conditions (humidity, tempera- lated with the reconstructed 1918 virus,11 in animals and ture, and wind) (Table 1, Ref. 1).26–28 humans with H5N1 avian influenza,13,15,16 in SARS-CoV,17–19 and in 2009 H1N1 victims (Table 1, Ref. 12).9 AEROSOL-GENERATING PROCEDURES IN ANESTHESIA AND CRITICAL CARE TRANSMISSION MODES Aerosol particles are generated during all invasive airway The predominant transmission modes for influenza and procedures, noninvasive and positive pressure ventilatory SARS are respiratory droplet and direct and indirect con- support modes, suction, sputum induction, high-flow oxy- tact (fomite). (Table 1, Refs. 13 and 14). Epidemiologic and gen delivery, aerosolized or nebulized medication delivery, investigational evidence is strongly suggestive of near- interventions that stimulate coughing, and autopsies (Table range airborne transmission (Table 1, Refs. 1, 9, 10, and 1, Refs. 1, 4, 13, 15, and 17).21,22 Infection is established with 13–15).20–22 Remote airborne transmission is supported by a smaller quantity of aerosol than nasal instillation (Table 1, laboratory research and theoretical modeling of aerosol Ref. 1).22,29 Health care professionals who perform and behavior and is suspected in specific outbreaks, but be- assist with aerosol-generating procedures and therapies are lieved to be unusual (Table 1, Refs. 1 and 15).21–25 included in the Occupational Safety and Health Adminis- tration (OSHA) “very high occupational exposure risk” Contact Mode category (Table 1, Refs. 3, 4, and 18), and are at a higher risk The sources of contact transmission are often overlooked of infection than others. (contaminated surfaces, clothing, equipment, personal protective equipment, exposed skin) (Table 1, Refs. 1 and 13). TRANSMISSION-BASED PRECAUTIONS The infectivity of virus on surfaces, skin, and hands decays, Close Patient Care and the time that the fomite remains infective may vary Confusion persists about whether a facemask or respirator (Table 1, Ref. 13).22 is indicated for close care of influenza patients and those with a flu-like illness. For 2009 H1N1 influenza, because of Droplet and Airborne Modes the limited access to N95s in many countries, the WHO has Newer investigations on aerosol transmission of influenza stated that the N95 (or European equivalent, EU FFP2) is have challenged the traditional belief that airborne trans- only indicated during aerosol-generating procedures, and mission does not occur. Droplet and airborne modes are that a surgical mask in addition to face shield and eye now viewed as a continuum, with particle sizes ranging protection, with other contact precautions, is satisfactory from large to fine droplet or aerosol (Table 1, Refs. 1, 14, for all other patient care activity (Table 1, Ref. 19). In and 15).20–22 The term “droplet” is consistent with trans- contrast to the WHO, the CDC, OSHA, and the IOM of the mission by larger disease-bearing particles that settle out of National Academies all state that the N95 respirator, or the air onto surfaces within shorter distances from the higher ([N100], or powered air purifying respirator source or are inhaled into the upper airway and trachea. [PAPR]), should be used during close contact with influ- Larger droplets may evaporate to small “droplet nuclei” enza or SARS patients. Contact precautions (gown, gloves, that behave as aerosol (Table 1, Refs. 1, 13, and 16). hat, close-fitting eye protection, shoe covers) are also rec- “Near-range airborne” transmission is through smaller ommended for influenza and other virulent diseases that droplet particles (inspirable) that remain suspended in air are transmitted by both the respiratory and contact modes. for relatively short but variable distances (1 m with breath- The CDC describes close contact as within 6 to 10 ft. of the ing, 2 m with coughing, 6 m with sneezing), and when patient or in the patient’s room (Table 1, Ref. 20). inhaled, reach only the trachea and bronchi (Table 1, Refs. 1, and 13–15).20–22 Near-range airborne transmission Aerosol-Generating Procedures through a spectrum of disease-bearing particle sizes is now The CDC, OSHA, and the IOM further suggest that health acknowledged by the CDC and others to be operational in care personnel consider using the higher level of protection influenza and SARS (Table 1, Refs. 1, 8, 13, and 14).22 provided by the PAPR when performing or assisting with

936 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Personal Protective Equipment for Pandemic Influenza

aerosol-generating procedures (Table 1, Refs. 1, 3, 4, 13, 15, The APF of the N95 respirator is 10, meaning that the and 20). Some state and hospital pandemic influenza, wearer will expect to inhale no more than one-tenth of the SARS, and tuberculosis protocols assert definitively that hazardous airborne particles present (Table 1, Ref. 24). the PAPR is to be used for endotracheal intubation and OSHA has set a minimum APF of 1000 that a PAPR must bronchoscopy. The CDC states that these procedures achieve to obtain NIOSH certification, but charges the should be done in an airborne infection isolation room prospective buyer with obtaining confirmatory testing evi- (Table 1, Ref. 20). dence of the APF from each PAPR manufacturer (Table 1, Refs. 24 and 25).30 OSHA STANDARDS APPLICABLE TO HEALTH CARE FACILITIES TERMINOLOGY AND CATEGORIZATION OF FACEMASKS AND RESPIRATORS The General Duty Clause of the Occupational Safety and Health Act requires that employers abate or address recog- Medical Masks nized workplace hazards. OSHA standards for employee Medical masks of several different types are worn, as safety were developed initially for the industrial work- intended, for protecting the patient and the environment place; subsequently, standards for the health care venue from the respiratory secretions of health care workers were developed as well (Table 1, Refs. 4 and 21). The OSHA (Table 3A). Masks also provide the wearer with droplet and Blood-borne Pathogens, Personal Protective Equipment, contact protection for the skin area covered by the mask. and Respiratory Protection standards, as listed in the Code The degree of protection corresponds to the masks’ fluid of Federal Regulations, are of particular importance in resistance; surgical masks are fluid resistant, procedure pandemic influenza preparedness and response (Table 1, masks are less so, and isolation masks generally are not. ␮ Ref. 4). Airborne pathogens are included in the hazardous The filtering fabric of some masks is tested to 0.1- m airborne contaminants covered by the standards. OSHA particle size but this is not required for Food and Drug has posted inspection, investigation, and enforcement pro- Administration certification. Medical masks do not permit cedures that address 2009 H1N1 Influenza in health care a tight face seal, do not provide a barrier to aerosol workers in the “high” and “very high occupational expo- particles, and therefore do not meet the criteria for a sure risk” categories (Table 1, Ref. 21). respirator. Gauze masks are an effective droplet barrier Under the Respiratory Protection standard (29 CFR only with 6 or more layers (Table 1, Ref. 16). 1910.134), all respirators must be deployed within the The infection rate was reduced in Hong Kong during the context of a written, worksite-specific respiratory protec- SARS outbreak with consistent use of medical masks plus contact barrier garb that consisted of gowns, gloves, and tion program. The requisite respiratory protection program 6,26 director oversees and documents respirator selection, eye protection. Public use of masks, with hand hygiene, maintenance and cleaning procedures, employee medical also reduced the infection rate (Table 1, Refs. 1 and 16). A clearance to wear a respirator, fit testing of tight-fitting comparison of the infection rate between nurses wearing respirators, instruction, and periodic program review surgical masks or N95s found the influenza rate to be no different, but still considerable in both groups (approxi- (Table 1, Ref. 22). 31 States may elect to develop their own standards for mately 23%). employee safety, which the employers must follow, but Discussion and study continue on evidence-based choice of respiratory personal protective equipment (PPE) for health these plans must be approved by OSHA (Table 1, Ref. 23). 32,33 When applied, the OSHA standards maximize safety by care workers exposed to influenza. In the absence of a supporting the correct and consistent use of respirators. respirator or when supplies are limited, the surgical mask, with contact transmission precautions, should be used for routine bedside care of SARS and influenza patients (Table 1, QUANTIFICATION OF RESPIRATORY PROTECTION: Ref. 17). Masks must be removed carefully and discarded, THE ASSIGNED PROTECTION FACTOR followed by hand washing (Table 1, Refs. 13 and 20). The assigned protection factor (APF), applied to industrial respirators initially, is a value given or assigned to each Respirators respirator by OSHA and the National Institute of Occupa- In the context of PPE, a “respirator” reduces exposure to tional Safety and Health (NIOSH) denoting the factor by inhaled environmental contaminants (Table 1, Ref. 16). which a respirator reduces the contaminating substance in the Respirator classifications consider the air supply, power, ambient air (Table 1, Ref. 24). The APF is derived from filter type and efficiency, facepiece description, and pres- laboratory simulated workplace (SWPF) and actual work- ence of an exhalation valve. Additional descriptive fea- place protection factors (WPF) computed from photometric tures, such as air flow direction, clarify the respirator’s

determinations of the test contaminant outside (Co) the respi- protective mechanism (Tables 2 and 3B) (Table 1, Refs. 1, 4, 30 34 rator relative to inside (Ci) the respirator facemask. For each 15, and 16). The particulate respirator filter removes respirator tested, the lowest fifth percentile of test Co/Ci droplet and aerosol-sized airborne particles and an absor- values are averaged and further divided by a safety factor of bent respirator filter removes chemical vapors and gases. 25. This final value is assigned to the respirator as the APF. The nonpowered category of air purifying respirators APFs range from 10 to 10,000. This number represents the includes the elastomeric and the filtering facepiece respira- minimum factor by which exposure to contaminants is re- tors. The “elastomeric” respirator has a full or half facepiece duced when wearing the respirator. Hence, a higher number that is nondisposable, tight fitting, and made of silicone or indicates greater protection (Table 1, Refs. 25 and 26). rubber with either a replaceable particulate or vapor/gas

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 937 SPECIAL ARTICLE

Table 2. Categorizing Respirators Table 3A. Medical Masks Type of respirators Types of Isolation; Surgical; Laser; a Air purifying medical masks Procedure Dental Nonpowered: Depend on the wearer drawing air in through filters Water resistance b Variable Yes or cartridges Transmission mode Powered air purifying respirators: A blower draws air through the protection c filter and delivers it to the wearer For wearer: Air supplying Contact Yes Yes Self-contained breathing apparatus Droplet Very little Yes Type of filters Aerosol (airborne) No No Particulate filters For wearer’s contacts: P (oil proof; can survive oil exposure for Ͼ1 work shift) Contact Yes Yes R (oil resistant; can be used for oil exposure in 1 shift) Droplet Very little Yes N (not oil resistant; used for oil-free environments) Aerosol (airborne) No No Gas-vapor respirator Chemical protection d None None Combination particulate and gas vapor a Medical masks include various types of face masks used in the healthcare Filtering efficiency venue; they provide no protection from airborne biologic aerosol and are not Certified for a range of efficiency classes (e.g., 95%, 99%, 100%) classified as respirators. Masks are intended to protect patients and other Type of facepiece close contacts from the wearer’s secretions. Filtering facepieces b Only water resistant masks and respirators provide the wearer (patient or Replaceable filter components: half-mask and full-mask healthcare provider) with droplet protection from others. Water resistance is elastomeric respirators not a requirement for certification of procedure and isolation masks and may Loose-fitting facepieces vary among different brands. Surgical masks are water resistant and protect the wearer and their contacts from droplets and splashes. Use or nonuse of an exhalation valve c Contact and droplet protection for the wearer is only over the area covered Adapted from: Institute of Medicine. Preparing for an Influenza Pandemic: by the mask or respirator. Personal Protective Equipment for Health Care Workers. Washington, DC: The d Medical masks and respirators are particulate filters only and do not provide National Academies Press, 2008. Reprinted with permission from the Na- protection from chemical agents. tional Academies Press.

Reuse of the N95 is strongly discouraged but permissible if absorbing filter, or a combination filter. Examples are the supplies are limited, provided it is not soiled, creased, dam- “gas mask,” “MOPP” respirator (Mission Oriented Protec- aged, moist, or wet (Table 1, Refs. 14 and 16). Reaerosolization tive Posture) (dual filter), the painter’s respirator (vapor), of disease particles from the N95 does not occur (Table 1, Ref. and the particulate elastomeric respirator for use in health 16).34 However, the used respirator is a fomite and presents a care if N95 stores are exhausted. contamination risk. Covering the N95 with a face shield will The filtering facepiece respirators (N95 and higher) are protect it from droplets and splashes. A fluid-resistant N95 is classified by the resistance to degradation by oil, and by the available and should be used in surgery (Table 1, Refs. 14 and ␮ percent filtration efficiency of 0.3- m saline test particles, 16). The used N95 should be stored in a paper, not plastic, bag the most penetrating size and substance (Table 1, Ref. 4). to avoid condensation (Table 1, Ref. 26). When donning a used Both the N95 and the elastomeric are negative pressure mask, extreme care should be taken not to contaminate respirators; that is, they are dependent on the maintenance oneself, and hand washing should follow. The N95 filtering of a tight face seal to prevent contaminated air from facepiece respirator provides a 10-fold factor of protection bypassing the filter (Table 1, Refs. 4, 14, and 27). relative to ambient air (APF 10) (Table 1, Refs. 24 and 26). The notable advantages of the N95 are simplicity and 35 accessibility. The predominant disadvantages are the THE PAPR FOR BIOLOGICAL HAZARDS 36 well-known increased resistance to breathing and the The PAPR provides a higher level of protection than the propensity for a gap to occur in the face seal (face seal N95 respirator because it supplies maximally filtered air, leakage) (Table 4). Prior fit testing does not assure success eliminates face seal leak, and provides contact protection in attaining and maintaining a tight face seal. Only those for the head (Table 3B). It is composed of a belt-worn case wearing the N95 in their daily work were found to achieve that houses a battery, fan, and filter. The fan, referred to as a face seal with some consistency, irrespective of when the “blower,” draws ambient air through a high-efficiency 37 prior fit testing had occurred (Table 1, Ref. 28). The N95 particulate air (HEPA) filter (99.97% efficient) and blows it at don (put on) and doff (take off) procedure posted by the Ͼ170 L/min (6 cu. ft./min) through a flexible tube and into a CDC includes a positive and negative pressure face seal Tyvek or Tychem hood with a plastic face shield (Table 1, Refs. check, which should be performed when using any tight- 4 and 30–33). The high flow of air exiting the hood prevents the fitting respirator (Table 1, Refs. 4, 14, and 29). wearer from entraining contaminated ambient air. Hospitals are urged to stockpile supplies. OSHA has The PAPR hoods are available in 2 styles: the double- predicted that every nurse, using 4 N95s per shift, would shrouded hood and the loose-fitting face covering. The former use 480 over a 12-week pandemic wave, at a cost of $240 completely covers the head and shoulders and does not per worker (Table 1, Ref. 18). All disposables will present a permit use of a conventional stethoscope. The inner shroud substantial expense and will be depleted rapidly.38 The tucks beneath the neck of a gown, and the filtered air exits CDC recommends prioritization of N95 use based on risk of from under the inner shroud and gown. The PAPR used with exposure, to be initiated when N95 supplies are limited a double-shrouded hood provides more than a 1000-fold (Table 1, Ref. 20). protection relative to ambient air (APF 1000) and 100 times the

938 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Personal Protective Equipment for Pandemic Influenza

Table 3B. Features of Masks and Respirators

a Self-contained breathing apparatus. b Remote supplied air respirator. c Available with particulate filter, gas/vapor absorbent canister, or combination. d Powered air purifying respirator. e Water resistant N95 respirators are available and should be used during surgery. f Mission Oriented Protective Posture: Uniformed Services elastomeric respirator and chem warfare agent resistant clothing. g Chemical absorbent filters are agent specific; painter’s respirator is not protective against chemical and nerve agents. h The loose fitting face covering helmet covers the top of the head, the face and the chin, leaving the ears and neck exposed. i The double shrouded hood covers the entire head and shoulders. j Assigned Protection Factor (factor by which the test contaminant is reduced by the respirator). k Contact protection is only for the skin area covered by the mask or respirator. l Only with particulate filter. m Only with absorbent filter. n Storage sites and containers for biologic PAPRs should be labeled prominently, “Not for chemical protection”. o High Efficiency Particulate Air filter. p APF 50 for full face elastomeric. protection of the N95. The double-shrouded hood offers the and 2.5 times the protection of the N95 respirator (APF best protection for aerosol-generating procedures. 10) (Table 1, Refs. 30–33). The loose-fitting face covering protects the face, chin, Because the PAPR is a loose-fitting respirator, fit testing and top of the head but leaves the neck and ears exposed, is not required. The full hood, but not the loose-fitting face allowing use of a stethoscope. Illustrations are available cover, may be used by those with a beard. Facial hair can online (Table 1, Refs. 30–34). The high-flowing air exits permit unfiltered air to be entrained under the elastic band through perforations beneath the chin. This hood is border of the loose-fitting face cover. appropriate for continuous bedside care in the absence of The PAPR does not increase the work of breathing and aerosol-generating conditions. The protection factor of is more comfortable for extended wear than an N95 (Table this hood is 25-fold greater than no protection (APF 25) 1, Refs. 1 and 4) (Table 5). The PAPR and the nondisposable

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 939 SPECIAL ARTICLE elastomeric air purifying respirator are the only realistic or vapor protection.” The hazardous-material PAPRs, e.g., alternatives for respirators in the hospital during a pan- 3M Breathe-Easy™, Bullard PA30™, and Bullard EVA™, demic if N95 stocks are exhausted. have vapor-absorbing and dual (combination) cartridges and chemical-resistant hoods; these must be worn with THE PAPR FOR CHEMICAL OR additional full-body chemical protective suits (Table 1, HAZARDOUS MATERIAL Refs. 31, 33, and 35) (Appendix 1). The particulate HEPA filter in PAPRs used for protection To be certified to wear chemical protective equipment, the against biological hazard, e.g., 3M™ Air-Mate™, Bullard OSHA standard 29 CFR 1910.120 on Hazardous Waste Op- PA20™, or Bullard EVA™, do not protect against hazard- erations and Emergency Response (HAZWOPER) requires a ous chemicals, gases, or vapors. PAPRs for biological combination of 8 hours of OSHA-compliant awareness-level protection should be prominently labeled “not for chemical (informational) and operations-level (practical) training (Table 1, Ref. 35). Hospital security services and emergency departments strictly adhere to, and enforce, this mandate. Table 4. N95 Respirator Advantages and Disadvantages PAPR ISSUES AND LIMITATIONS Advantages Attention to many details of PAPR use and maintenance is ● Filters 95% of aerosol test particles essential for the assurance of effective protection (Table 6). ● 10-fold protection (APF 10) ● Easily accessible The Donning and Doffing Sequence Is Important ● Disposable ● No setup required Exposed contaminated skin and PPE constitute indirect ● No interference to using a stethoscope sources of infection (fomite) for health care workers and ● Not powered potentially their contacts. Using fluorescein dye as a surrogate ● Noiseless marker of contamination, a comparison of the N95 and PAPR Disadvantages ● Face seal leak common ● Increases resistance to breathing (exhalation valve may reduce discomfort but must cover with surgical mask during surgery) Table 6. Reasons for Failure of Powered Air ● Requires fit testing (costly, labor intensive, does not ensure a Purifying Respirator (PAPR) Protection tight face seal) ● Neglect of battery maintenance procedures ● Facial features may preclude a satisfactory fit with any model ● Failure to check equipment and flow rate before use ● Headaches, dizziness, shortness of breath, possible CO 2 ● Decreased air flow during use (likely incomplete battery charge) retention ● Absence or incorrect placement of filter and gasket ● Supplies rapidly depleted when demand is high ● Incorrect don/doff sequence ● Reuse risks self-contamination (use face shield if reuse is likely) ● Neglect of hood care/inspection and unit cleaning/reprocessing (store in paper bag) ● Removal of PAPR during a procedure (failure to practice ● Facial hair inhibits a face seal, defeats protection procedures while wearing the equipment) ● Ineffective when moist, wet, creased, or damaged Caution: A biological PAPR does not provide chemical or vapor protection.

Table 5. Powered Air Purifying Respirator (PAPR) Advantages and Disadvantages PAPR advantages PAPR disadvantages ● HEPA filtered air ● Initial cost high (but may be cost effective compared with N95 stockpiling/ use) ● Positive inside to outside air flow (170 L/min, 6 ● Larger requirement for storage space cfm ͓cu. ft./min͔) ● Highest level of protection for aerosol-generating ● Self-contamination from used equipment possible procedures ● Contact protection of head and neck: double- ● Reprocessing of hoods required between work breaks shrouded hood better than the loose-fitting face ● Longer training time than N95 (required by Occupational Safety and Health covering hood; both more than N95 Administration, essential for safety: awareness of equipment; limitations; ● Protection factor, relative to no protection: assembly; preuse check; maintenance/reprocessing; don/doff sequence; 1000 ϫ for PAPR with double-shrouded hood acclimatization and procedure practice) 25 ϫ for PAPR with loose-fitting face covering hood ● Battery recharging schedule is crucial 10 ϫ for N95 respirator ● Biannual discharge of unused units needed so battery maintains its capacity ● No fit testing required to accept a charge ● No entrainment of contaminated air ● Some models do not have real-time air flow indicator ● Comfortable for extended wear (bedside care, ● Care needed in seating of HEPA filter gasket after changing battery to avoid reprocessing worker, invasive airway procedures) misplacement ● Hoods are disposable (reusable, by single user ● Equipment may impede performance only, after reprocessing) ● Hearing reduced because of fan noise; speech muffled because of hood ● Facial hair is acceptable, with double-shrouded ● PAPR use contraindicated during surgery (issue has not been addressed, hood only, not with loose-fitting face covering hood remains unresolved) ● The surgical hood system is not certified as a respiratory protection device HEPA ϭ high-efficiency particulate air.

940 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Personal Protective Equipment for Pandemic Influenza

PPE ensembles showed that skin contamination was greater of contact but not respiratory transmission, because with the N95 ensemble. Errors in don/doff steps were more disease-bearing particles do not reaerosolize from the filter frequent with the PAPR ensemble, thereby increasing the material (Table 1, Refs. 13, 16, and 39). chance for self-contamination or exposure.39 However, in The filter seldom needs to be changed unless it becomes following the CDC proscribed sequence for removal of the wet or moist, which destroys its filtering capability, or N95 and contact garb, viral contamination of hands and unless the flow decreases despite a fully charged battery. clothing occurred. In that study, a fluorescein dye surrogate Large industrial particles, such as wood or asbestos, may marker was not a reliable indicator of viral contamination. clog a filter quickly, but accumulated biological particles Double gloving and the customary sequence for removal of rarely cause the flow to decrease (Table 1, Ref. 31). A gowns and mask after surgery were recommended.40 The Min- decreased flow is likely to be the result of an incompletely nesota Health Department has posted an illustrated don/doff charged battery. sequence for the PAPR and contact garb (Table 1, Ref. 36).

Initial Cost Is High and Storage Space Is Challenged PAPR Training Is Essential for Safe Use The requisite PAPR training is more involved and less avail- Budget constraints and scarce storage space limit supplies of able than N95 fit testing and training. As with the N95, the PAPRs and hoods in most hospitals. Hospitals may purchase PAPR training must be in cooperation with a hospital OSHA- PPE using federal funds for disaster preparedness. compliant respiratory protection program director, usually Battery Maintenance Is Crucial the director of employee health. This individual may be The 3M Air-Mate has an interchangeable rechargeable overextended with required N95 fit testing and may not have nickel-cadmium battery. A fully charged and maintained the time or expertise to undertake PAPR training. battery will run for longer than 8 hours when properly Although instructions may be sought on a “just in time” charged. Close attention to battery-charging procedures basis, OSHA strongly encourages users to achieve compe- will avoid failure of the PAPR during use (Table 1, Refs. 37 tency with a respirator in advance. The PAPR competency and 38). can be included in the Department of Anesthesiology initial The Bullard PA20 does not have an interchangeable and periodic required demonstration of competencies with battery so it must be removed from service for recharging. various other equipment. A rapid charger is available (Table 1, Refs. 32 and 33). The battery runs for 9 to 10 hours. PPE Impedes Performance PPE impedes performance of manual clinical procedures, A Flow Indicator Is Desirable impairs hearing and communication, and may trigger The 3M Air-Mate has no real-time flow or battery charge claustrophobia in some users. Although this is a recognized indicator to signal when airflow or charge is reduced (Table issue with military and industrial protective gear, the same 1, Refs. 31 and 33). The air flow must be checked with all could apply with the PAPR used in health care (Table 1, PAPRs with the test float each time the unit is turned on. Ref. 1).41 Practicing airway procedures on a mannequin The 3M Breathe-Easy PAPR for chemical protection does while dressed in the PAPR is strongly advised so that have an external air-flow gauge. The Bullard PA20 PAPRs patient care is not compromised in the clinical setting. have a low battery charge and flow sound indicator. Training is effective in improving compliance with PPE use and in reducing distressing symptoms while wearing pro- Reprocessing Is Essential tective equipment (Table 1, Ref. 1).35,41 The PAPR hoods are disposable and intended for single use. Reuse of a hood is acceptable, but only by the same individual. The hood must be cleaned every time it is PAPR Use Is Contraindicated During Surgery removed, because reuse of a soiled hood presents a contact The 3M Air-Mate user instructions state that the PAPR transmission risk. Reprocessing procedures must follow should not be used during surgery (Table 1, Ref. 31). The manufacturers’ instructions, should be documented in presumed rationale is that positive (outward) air flow writing, and should be approved by the hospital infection could increase the risk of wound infection. The user in- control practitioner. Reprocessing personnel and users structions do not specify whether the prohibition applies to must understand the equipment and be thoroughly versed both the double-shrouded full hood and the loose-fitting in procedural details. face cover hood styles, or only to the latter style. The air exit holes of the loose-fitting face cover are located under the The HEPA Filter and Gasket Placement chin; thus, if the wearer were standing at the operating Require Care table, air would exit directly onto the surgical field. In The 3M Air-Mate filter and gasket should be checked for contrast, the air exits from the double-shrouded hood integrity and correct placement. This should be done in the under the accompanying surgical gown and close to the reprocessing room while wearing gloves and before the floor. unit is returned to service. The PAPR reprocessing steps OSHA has acknowledged the stated prohibition of PAPR require removal of the filter to change the battery and to use during surgery but has not addressed the potential impact confirm that the filter gasket is present and aligned cor- on the safety of operating room personnel, most notably rectly in the groove under the filter. The PAPR unit must anesthesia providers (Table 1, Ref. 4); neither has the CDC not be used without an intact and correctly placed filter expressed an opinion on PAPR use in surgery. Therefore, in gasket. The used filter, as does a used N95, presents a risk consort with their hospitals’ infection control practitioner and

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 941 SPECIAL ARTICLE respiratory protection program and employee health direc- or avoiding manual ventilation will reduce exposure risk. A tors, anesthesiology departments should address PAPR use in technique that prevents coughing and avoids nebulized medi- the operating suite and develop procedures that maximize cation delivery will minimize aerosolization. safety for both patients and health care workers. At one Noninvasive ventilation, an aerosol-generating proce- author’s hospital (Rush), a Department of Anesthesiology dure, is generally avoided in patients with a virulent policy for using the PAPR with the double-shrouded hood in transmissible respiratory disease; however, noninvasive the surgical suite was developed, approved and documented ventilation, in conjunction with infection control air han- in cooperation with the hospital’s infectious disease and dling systems, was used successfully during the SARS infection control sections. epidemic without an increase in health care worker infections.17 Early noninvasive ventilation may reduce the risk Outcome Data of health care worker exposure by avoiding tracheal intu- A comparison of infection rates in health care workers who bation and ventilatory support.17 used either a PAPR or an N95 for aerosol-generating procedures would be pertinent, as would data on PAPR use OPERATING ROOM POLICY AND PROCEDURES issues and problems. However, no such data are available FOR INFLUENZA at this time. Previously published recommendations for the manage- The Surgical Hood System Is Not an Acceptable ment of SARS and tuberculosis patients in the surgical suite Alternative to the PAPR may be adapted to avian/pandemic/H1N1 influenza plans. These include recommendations on scheduling of During SARS outbreaks, some health care workers used the operations, air handling controls, isolation procedures, surgical hood system, and added goggles and an N95 upon 39 protection of the anesthesia machine, disinfecting and recommendation from Stryker (Table 1, Refs. 40 and 41). cleaning procedures, and communication with the director Although the surgical hood system used in orthopedics and of plant engineering (Table 1, Refs. 14 and 34).17,45 During spine surgery would seem to be a satisfactory substitute for the course of a pandemic, operating room schedules could the PAPR, it is neither classified nor certified as a respirator, be profoundly disrupted by staff and equipment shortages nor has NIOSH evaluated it for that purpose. The Stryker and policies that defer elective operations. Anesthesia surgical hood system draws ambient air through the material machines might be enlisted as surge capacity ventilators of the hood and gown, not through a HEPA filter. An and anesthesia professionals as critical care providers. evaluation of 2 surgical hood systems found both to have a 42 Consideration of these possibilities is encouraged. lower filtration efficiency than either the N95 or the PAPR. There is further concern that the incoming surgical hood system airflow blows directly over the wearer’s eyes. A TRAINING WORKSHOP A PAPR workshop in PowerPoint format is available for CRITICAL POINTS IN PAPR MAINTENANCE AND USE training purposes (Table 1, Refs. 34 and 43). This article and The respiratory and contact protection provided by the PowerPoint presentation support the workshop’s practical PAPR will fail if critical factors are neglected (Table 6). component that incorporates PPE don/doff sequence and procedure practice. The workshop was developed to EQUAL PROTECTION FOR HEALTH CARE supplement, not replace, the manufacturers’ training mate- WORKERS AND PATIENTS rials, and must be a component of an OSHA-compliant The health and safety of health care workers and patients respiratory protection program. alike are supported by a “hierarchy of controls” that reduce exposure, use engineering solutions, enact administrative SUMMARY policies, and lastly, facilitate the consistent and effective use Pandemic influenza is highly contagious and potentially of personal protective equipment and transmission-based many times more lethal than seasonal influenza. Those precautions (Table 1, Refs. 1, 20, 42, and 43).4 A culture of who participate in aerosol-generating procedures, in- safety is created when all parties participate with mutual cluding endotracheal intubation, are in the OSHA “very respect and support in a program that uses effective high occupational exposure risk” category. OSHA stan- measures known to minimize nosocomial and workplace- dards require that employers provide safe working con- acquired infections (Table 1, Ref. 1). ditions in a hazardous environment. We must work in concert with our hospitals and our fellow health care PROTOCOL FOR TRACHEAL INTUBATION professionals to create a culture of safety. OF PATIENTS WITH TRANSMISSIBLE The N95 respirator is limited by face seal leakage. RESPIRATORY DISEASE Wearing an N95 in daily work improved the success of A practical guide provides a template for anesthesia depart- maintaining a face seal. Using an N95 without having been ment preparedness (Appendix 2). A written protocol for fit tested voids the assigned protective factor (APF 10). endotracheal intubation of patients with a virulent transmis- When used within the context of an OSHA-compliant sible respiratory disease should be in place.1,43,44 The protocol respiratory protection program, the PAPR offers a higher level should emphasize the use of appropriate PPE within the of respiratory and contact protection than the N95. The PAPR context of an OSHA-compliant respiratory protection pro- has an APF of Ͼ1000 when used with the double-shrouded gram. Rehearsing the procedure, avoiding emergent intuba- head- and shoulders-covering hood, and an APF of 25 when tion through early intervention, and minimizing a mask leak used with the loose-fitting face covering hood. That is, the

942 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Personal Protective Equipment for Pandemic Influenza wearer may expect a 1000-fold (or more) or a 25-fold reduc- protection. The surgical hood system is not a respiration in contaminant within the respirator, depending on the tor and is not a satisfactory alternative. hood choice. Anesthesia providers must be able to perform procedures comfortably while wearing a PAPR. The double- Recommendations shrouded hood is most appropriate for aerosol-generating Prepare for Safety procedures and the loose-fitting face covering hood for usual bedside care. • Get vaccinated and practice sound respiratory, The PAPR carries substantial use and maintenance hand, personal, and workplace hygiene. issues that must be addressed by health care workers and • Communicate with: institutions alike. The package user instructions and OSHA • Hospital director of OSHA-compliant respiratory state that the PAPR is not to be used during surgery protection program regarding respirator choice because it could increase the risk of wound infection. This and training dilemma is as yet unresolved and should be addressed • Infection control practitioner regarding proce- with the hospital infection control practitioner. dures for personal protective equipment (PPE) don/doff sequence, and discuss/receive clearance for use of PAPR with full hood in the APPENDIX 1. Powered Air Purifying Respirator operating room (PAPR) Brand Name and Type of Protective Filter • Plant engineer regarding air flow/filtration controls • Identify an interested anesthesia department coor- Name Protection Filter type dinator to lead pandemic (and other) emergency a “Medical” PAPRs preparedness/response. 1. 3M™ Air-Mate™ Biologic Particulate (HEPA) • Select a respirator for aerosol-generating procedures. 2. Bullard PA20TM Biologic Particulate 3. Bullard EVA™ Biologic Particulate • If N95, be fit tested; remember your brand and “Chemical” PAPRs size; wear an N95 daily to improve (but not 1. 3M Breathe-Easy™ Chemical (gas/vapor) Absorbent assure) your ability to maintain a tight face seal; Bbiologic and Combination address reuse protocol with hospital infection chemical control practitioner; use a face shield to protect 2. Bullard PA30™ Chemical Absorbent the N95 if reuse is intended. Biologic and Combination • If PAPR, become familiar with chosen hospital chemical brand, or recommend a brand; know manufac- 3. Bullard EVA Biologic and Combination turer’s use and maintenance procedures; practice chemical 4. MSA OptimAirb TL Chemical Absorbent don/doff sequence of PAPR and associated con- Biologic and Combination tact PPE; practice procedures while dressed in chemical PPE; secure same protection for assisting staff. 4. North Chemical Absorbent b CompactAir® Biologic and Combination Prepare for Invasive Airway Procedure chemical • Anticipate/avoid the need for emergent intuba- HEPA ϭ high-efﬁciency particulate air. tion: use noninvasive ventilatory support; intu- See Table 1, Refs. 30, 32, and 33. bate preemptively. a Industrial use also. b Industrial PAPR. • Use air infection isolation room where available or high-efficiency particulate air (HEPA) filtered room exhaust. APPENDIX 2 • Experienced individual should intubate. Excuse: pregnant; nonessential personnel. • Don PPE (PAPR plus contact, or N95 plus contact) PRACTICAL GUIDE: AIRWAY MANAGEMENT per protocol, training, and hospital respiratory pro- OF PATIENTS WITH CONTAGIOUS tection plan before entering the room. RESPIRATORY DISEASE • Confirm that equipment and medications are imme- Background diately available. • Anesthesiology and Critical Care providers are in • Use closed suction when possible. the Occupational Safety and Health Administration • Use disposable or dedicated monitoring equipment. Remove unnecessary equipment. (OSHA) “very high exposure risk” category. • • Use a HEPA filter between mask and resuscitation bag. Some influenza patients progress to fatal destructive • Plan a procedure that will avoid all of the following: viral pneumonia and extreme inflammatory response, patient coughing, nebulized/topical/transtracheal “cytokine storm.” lidocaine, forceful bag-mask ventilation. • Airborne (droplet and aerosol) and contact transmis- • Rehearse the procedure just before intubation. sion precautions are indicated. • The limiting features of the N95 respirator are an Conduct Urgent/Emergent Intubation increased work of breathing and leakage around the face seal. • Patients in respiratory failure/arrest: Above, and • The powered air purifying respirator (PAPR) has a intubate using the practitioner’s technique of great- higher protection factor than the N95; however, inat- est success. A laryngeal mask airway may be used tention to details of maintenance and use will void as a bridge or a conduit.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 943 SPECIAL ARTICLE

• Patients in impending respiratory failure: Above, 11. Watanabe T, Watanabe S, Shinya K, Kim JH, Hatta M, and perform focused history and physical examina- Kawaoka Y. Viral RNA polymerase complex promotes optimal tion; assess the airway; consider glycopyrrolate, growth of 1918 virus in the lower respiratory tract of ferrets. preoxygenate. Proc Natl Acad Sci USA 2009;106:588–92 • For normal airway: Administer a muscle relax- 12. Gupta RK, George R, Nguyen-Van-Tam JS. Bacterial pneumo- ant, ventilate gently if at all. nia and pandemic influenza planning. Emerg Infect Dis 2008;14:1187–92 • For difficult airway: Consider deep sedation with 13. Stevens J, Blixt O, Tumpey TM, Taubenberger JK, Paulson JC, titrated midazolam, fentanyl, or ketamine. Use Wilson IA. Structure and receptor specificity of the hemagglu- lidocaine 1.5 mg/kg IV 1 minute before intuba- tinin from an H5N1 influenza virus. Science 2006;312:404–10 tion. Administer muscle relaxant after intubation 14. Allen JE, Gardner SN, Vitalis EA, Slezak TR. Conserved amino is confirmed. acid markers from past influenza pandemic strains. BMC • For all cases: Remove PAPR per don/doff proto- Microbiol 2009;9:77–87 col outside the room, with a gloved assistant to 15. de Jong MD, Simmons CP, Thanh TT, Hien VM, Smith GJ, place reusable equipment, including laryngo- Chau TN, Hoang DM, Chau NV, Khanh TH, Dong VC, Qui PT, scope, in a marked biohazard container for repro- Cam BV, Ha do Q, Guan Y, Peiris JS, Chinh NT, Hien TT, cessing. Reprocess PAPR equipment according to Farrar J. Fatal outcome of human influenza A (H5N1) is manufacturer’s recommendation and hospital associated with high viral load and hypercytokinemia. Nat protocol. Med 2006;12:1203–7 16. Baskin CR, Bielefeldt-Ohmann H, Tumpey TM, Sabourin PJ, Adapt Above for Perioperative Management Long JP, Garcia-Sastre A, Tolnay AE, Albrecht R, Pyles JA, Olson PH, Aicher LD, Rosenzweig ER, Murali-Krishna K, • Defer elective surgery. Clark EA, Kotur MS, Fornek JL, Proll S, Palermo RE, Sabourin • Follow hospital infection control guidelines for pa- CL, Katze MG. Early and sustained innate immune response tient transport. defines pathology and death in nonhuman primates infected • Use HEPA filters on both limbs of anesthesia circuit; by highly pathogenic influenza virus. Proc Natl Acad Sci USA consider disposable circuit. 2009;106:3455–60 • Remove gloves and wash hands after each patient 17. Lau AC, Yam LY, So LK. Management of critically ill patients contact. with severe acute respiratory syndrome (SARS). Int J Med Sci • Limit contamination of the anesthesia equipment. 2004;1:1–10 18. Ng PC, Lam CW, Li AM, Wong CK, Cheng FW, Leung TF, Hon • Recover the patient in isolation. EK, Chan IH, Li CK, Fung KS, Fok TF. Inflammatory cytokine • Change PPE before transporting the patient; used profile in children with severe acute respiratory syndrome. PPE and exposed skin are fomites. Pediatrics 2004;113:e7–14 19. Pang BS, Wang Z, Zhang LM, Tong ZH, Xu LL, Huang XX, References: Gao WJ, Zhu M, Wang C. Dynamic changes in blood cytokine 1,43,44,45 (Table 1, Refs. 4, 13, 20, 22, 26, 29, 30–34, and 36–38) levels as clinical indicators in severe acute respiratory syndrome. Chin Med J (Engl) 2003;116:1283–7 ACKNOWLEDGMENTS 20. Nicas M, Jones RM. Relative contributions of four exposure The authors gratefully thank Chelsea Wanta and Traci pathways to influenza infection risk. Risk Anal 2009; 29:1292–303 Nathans-Kelly for editing assistance. 21. Tellier R. Review of aerosol transmission of influenza A virus. Emerg Infect Dis 2006;12:1657–62 REFERENCES 22. Tellier R. Aerosol transmission of influenza A virus: a review 1. Kamming D, Gardam M, Chung FI. Anesthesia and SARS. Br J of new studies. J R Soc Interface 2009;6(Suppl 6):S783–90 Anaesth 2003;90:715–8 23. Roy CJ, Milton DK. Airborne transmission of communicable 2. Barry JM. The Great Influenza: The Epic Story of the Deadliest infection: the elusive pathway. N Engl J Med 2004;350:1710–2 Plague in History. New York: Penguin, 2004 24. Tong TR, Liang C, Nicastri E, Petrosillo N, Puro V, Fowler RA, 3. Taubenberger JK, Morens DM. 1918 influenza: the mother of Scales DC, Ilan R, Yu ITS, Li Y. Evidence of airborne transmis- all pandemics. Emerg Infect Dis 2006;12:15–22 sion of SARS [correspondence]. N Engl J Med 2004;351:609–11 4. Lau AC, Yip IK, Li MC, Wan M, Sit AW, Lee RA, Yung RW, 25. Yu IT, Li Y, Wong TW, Tam W, Chan AT, Lee JH, Leung DY, Yam LY. Response to SARS as a prototype for bioterrorism: Ho T. Evidence of airborne transmission of the severe acute lessons in a regional hospital in Hong Kong. In: McIsaac JH, ed. respiratory syndrome virus. N Engl J Med 2004;350:1731–9 Hospital Preparation for Bioterror: A Medical and Biomedical 26. Gamage B, Moore D, Copes R, Yassi A, Bryce E. Protecting Systems Approach. Boston: Elsevier Academic Press, 2006: health care workers from SARS and other respiratory patho- 281–94 gens: a review of the infection control literature. Am J Infect 5. Caputo KM, Byrick R, Chapman MG, Orser BJ, Orser BA. Control 2005;33:114–21 Intubation of SARS patients: infection and perspectives of 27. Scales DC, Green K, Chan AK, Poutanen SM, Foster D, Nowak healthcare workers. Can J Anaesth 2006;53:122–9 K, Raboud JM, Saskin R, Lapinsky SE, Stewart TE. Illness in 6. Nicolle L. SARS safety and science. Can J Anaesth 2003;50:983–8 intensive care staff after brief exposure to severe acute respi- 7. Shaw K. The 2003 SARS outbreak and its impact on infection ratory syndrome. Emerg Infect Dis 2003;9:1205–10 control practices. Public Health 2006;120:8–14 28. Varia M, Wilson S, Sarwal S, McGeer A, Gournis E, Galanis E, 8. Taubenberger JK, Morens DM. The pathology of influenza Henry B. Investigation of a nosocomial outbreak of severe virus infections. Annu Rev Pathol 2008;3:499–522 acute respiratory syndrome (SARS) in Toronto, Canada. CMAJ 9. Gill JR, Sheng ZM, Ely SF, Guinee DG, Beasley MB, Suh J, 2003;169:285–92 Deshpande C, Mollura DJ, Morens DM, Bray M, Travis WD, 29. Alford RH, Kasel JA, Gerone PJ, Knight V. Human influenza Taubenberger JK. Pulmonary pathologic findings of fatal 2009 resulting from aerosol inhalation. Proc Soc Exp Biol Med pandemic influenza A/H1N1 viral infections. Arch Pathol Lab 1966;122:800–4 Med 2010;134:235–43 30. Cohen HJ, Hecker LH, Mattheis DK, Johnson JS, Biermann AH, 10. Yeh E, Luo RF, Dyner L, Hong DK, Banaei N, Baron EJ, Pinsky Foote KL. Simulated workplace protection factor study of BA. Preferential lower respiratory tract infection in swine- powered air-purifying and supplied air respirators. AIHAJ origin 2009 A(H1N1) influenza. Clin Infect Dis 2010;50:391–4 2001;62:595–604

944 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Personal Protective Equipment for Pandemic Influenza

31. Loeb M, Dafoe N, Mahony J, John M, Sarabia A, Glavin V, 38. Phin NF, Rylands AJ, Allan J, Edwards C, Enstone JE, Nguyen- Webby R, Smieja M, Earn DJ, Chong S, Webb A, Walter SD. Van-Tam JS. Personal protective equipment in an influenza Surgical mask vs N95 respirator for preventing influenza pandemic: a UK simulation exercise. J Hosp Infect 2009; among health care workers: a randomized trial. JAMA 71:15–21 2009;302:1865–71 39. Zamora JE, Murdoch J, Simchison B, Day AG. Contamination: 32. Radonovich LJ Jr, Perl TM, Davey V, Cohen H. Preventing the a comparison of two personal protective systems. CMAJ soldiers of health care from becoming victims on the pandemic 2006;175:249–54 battlefield: respirators or surgical masks as the armor of choice. 40. Casanova L, Alfano-Sobsey E, Rutala WA, Weber DJ, Sobsey Disaster Med Public Health Prep 2009;3(Suppl 2):S203–10 M. Virus transfer from personal protective equipment to 33. Srinivasan A, Perl TM. Respiratory protection against influ- healthcare employees’ skin and clothing. Emerg Infect Dis enza. JAMA 2009;302:1903–4 2008;14:1291–3 34. Fennelly KP. The role of masks in preventing nosocomial 41. Barach P, Rivkind A, Israeli A, Berdugo M, Richter ED. transmission of tuberculosis. Int J Tuberc Lung Dis Emergency preparedness and response in Israel during the 1998;2:S103–9 Gulf War. Ann Emerg Med 1998;32:224–33 35. Yassi A, Lockhart K, Copes R, Kerr M, Corbiere M, Bryce E, 42. Derrick JL, Gomersall CD. Surgical helmets and SARS infec- Danyluk Q, Keen D, Yu S, Kidd C, Fitzgerald M, Thiessen R, tion. Emerg Infect Dis 2004;10:277–9 Gamage B, Patrick D, Bigelow P, Saunders S. Determinants of 43. Cooper A, Joglekar A, Adhikari N. A practical approach to healthcare workers’ compliance with infection control proce- airway management in patients with SARS. CMAJ 2003; dures. Healthc Q 2007;10:44–52 169:785–7 36. Li Y, Tokura H, Guo YP, Wong AS, Wong T, Chung J, Newton 44. Peng PW, Wong DT, Bevan D, Gardam M. Infection control E. Effects of wearing N95 and surgical facemasks on heart rate, and anesthesia: lessons learned from the Toronto SARS out- thermal stress and subjective sensations. Int Arch Occup break. Can J Anaesth 2003;50:989–97 Environ Health 2005;78:501–9 45. Stackhouse RA. Severe acute respiratory syndrome and tuber- 37. Lee MC, Takaya S, Long R, Joffe AM. Respirator-fit testing: culosis. Anesthesiol Clin North America 2004;22:437–55 does it ensure the protection of healthcare workers against respirable particles carrying pathogens? Infect Control Hosp Epidemiol 2008;29:1149–56

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 945 CME Early Postoperative Subcutaneous Tissue Oxygen Predicts Surgical Site Infection

Raghavendra Govinda, MD,*† Yusuke Kasuya, MD,†‡ Endrit Bala, MD,§ Ramatia Mahboobi, MD,§ Jagan Devarajan, MD,§ Daniel I. Sessler, MD,§ and Ozan Akc¸a, MD†ʈ

BACKGROUND: Subcutaneous oxygen partial pressure is one of several determinants of surgical site infections (SSIs). However, tissue partial pressure is difﬁcult to measure and requires invasive techniques. We tested the hypothesis that early postoperative tissue oxygen saturation

(StO2) measured with near-infrared spectroscopy predicts SSI. METHODS: We evaluated StO2 in 116 patients undergoing elective colon resection. Saturation was measured near the surgical incision, at the upper arm, and at the thenar muscle with an InSpectra™ tissue spectrometer model 650 (Hutchinson Technology Inc., Hutchinson, MN) 75 minutes after the end of surgery and on the ﬁrst postoperative day. An investigator blinded to

StO2 assessed patients daily for wound infection. Receiver operating characteristic curves were used to analyze the performance of StO2 measurements as a predictor of SSI. RESULTS: In 23 patients (Ϸ20%), SSI was diagnosed 9 Ϯ 5 days (mean Ϯ SD) after surgery. Patients who did and did not develop an SSI had similar age (48 Ϯ 14 vs 48 Ϯ 15 years, respectively; P ϭ 0.97) and gender (female:male, 15:8 vs 46:47, respectively), but patients who Ϯ Ϯ 2 Ͻ developed SSI weighed more (body mass index 32 7vs27 6 kg/m ; P 0.01). StO2 at the upper arm was lower in patients who developed SSI than in those who did not develop SSI (52 Ϯ 22 vs 66 Ϯ 21; P ϭ 0.033), and these measurements had a sensitivity of 71% and speciﬁcity

of 60% for predicting SSI, using StO2 of 66% as the cutoff point. CONCLUSION: StO2 measured at the upper arm only 75 minutes after colorectal surgery predicted development of postoperative SSI, although the infections were typically diagnosed

more than a week later. Although further testing is required, StO2 measurements may be able to predict SSI and thus allow earlier preventive measures to be implemented. (Anesth Analg 2010; 111:946–52)

urgical site infections (SSIs) are perhaps the most entire physiological range of tissue values.9 Interventions to common serious complication of anesthesia and sur- improve tissue oxygenation during or immediately after gery. They cause considerable morbidity and are surgery are thus most likely to reduce the morbidity and S 10,11 expensive to treat.1,2 The transition from contamination to mortality associated with SSI. established infection occurs during a decisive period that Tissue oxygenation has traditionally been measured probably lasts only a few hours, even though infections are with Clark-type electrodes or similar systems. However, typically detected a week or longer after surgery.3,4 If these methods are invasive, expensive, and require exper- 12,13 antibiotics are administered during this decisive period, tise to use. Near-infrared spectroscopy (NIRS) is an 14–17 they are more effective in reducing infection risk than when alternative noninvasive technique. We tested the hy- given later.5 Tissue oxygen tension levels are one of the best pothesis that tissue oxygen saturation (Sto2) measured soon factors established for predicting SSI.6,7 Oxidative burst after surgery with NIRS predicts ultimate development of SSI. Confirming this hypothesis would allow early clinical function of neutrophils is one of the primary defenses 8 interventions, which might reduce the risk of infection. against SSI. Oxidative burst depends on the Po2 over the

From the *Department of Anesthesiology, Tufts Medical Center, Boston, METHODS Massachusetts; †Department of Anesthesiology and Perioperative Medicine, and ʈNeuroscience ICU, University of Louisville, Louisville, Kentucky; With approval by the Human Studies Committees at the ‡Tokyo Women’s Medical University, Tokyo, Japan; and §Department of University of Louisville and the Cleveland Clinic, and Outcomes Research, Cleveland Clinic, Cleveland, Ohio. written informed consent by the participants, we included Accepted for publication May 6, 2010. 116 colorectal surgery patients. Forty-three of these patients Supported in part by Hutchinson Technology Inc. (Hutchinson, MN) and the Joseph Drown Foundation (Los Angeles, CA). had laparoscopic-assisted colorectal procedures; the re- Data were presented in part at the American Society of Anesthesiologists mainder had laparotomies (Table 1). Annual Meeting in Orlando, FL (October 2008). In a preliminary study, we found that patients who

All authors are also affiliated with the Outcomes Research Consortium. developed SSI had lower Sto2 measurements at the thenar Disclosure: The authors report no conflicts of interest. eminence ϳ15% Ϯ20% (mean Ϯ SD) than those who Address correspondence and reprint requests to Ozan Akc¸a, MD, Depart- remained uninfected postoperatively; the analogous abso- ment of Anesthesiology and Perioperative Medicine, University of Louis- lute difference was Ϸ20% Ϯ30% at the upper arm.18 Using ville Hospital, 530 S. Jackson St., Louisville, KY 40202. Address e-mail to [email protected]. these estimates, a sample size of 80 measurements at the Copyright © 2010 International Anesthesia Research Society thenar eminence and 132 at the upper arm was calculated DOI: 10.1213/ANE.0b013e3181e80a94 to achieve 90% power with an ␣ of 0.05. We therefore

946 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Table 1. Demographic and Morphometric Characteristics, and Potential Confounding Factors Patients with Patients without P value (93 ؍ SSI (n (23 ؍ SSI (n Demographic and morphometric characteristics Age (y) 48 Ϯ 14 48 Ϯ 15 0.971 Sex (female/male) 15/8 46/47 0.175 Weight (kg) 92 Ϯ 21 78 Ϯ 20 0.006 BMI (kg/m2) 32 Ϯ 6.8 27 Ϯ 5.5 Ͻ0.001 ASA physical status I/II/III, n 2/15/6 10/57/26 0.426 Smokers, n (%) 4 (17%) 18 (19%) 0.830 Cardiac disease, n (%) 7 (30%) 25 (27%) 0.732 Diabetes mellitus, n (%) 3 (13%) 4 (4%) 0.115 Previous chemotherapy, n (%) 2 (9%) 4 (4%) 0.394 Preoperative Hct (%) 40 Ϯ 440Ϯ 5 0.612 SENIC I/II/III (%) 17/78/5 9/80/11 0.373 NNISS I/II/III (%) 43.5/43.5/13 39/48/13 0.824 Intraoperative confounders Crystalloids (mL) 3273 Ϯ 1414 3274 Ϯ 1500 0.996 Crystalloids (mL)a 3150 (2550–3875) 3000 (2400–4000) Colloids (mL) 620 Ϯ 532 439 Ϯ 452 0.101 Colloids (mL)a 500 (0–1000) 500 (0–813) Blood loss (mL) 549 Ϯ 511 373 Ϯ 378 0.078 Blood loss (mL)a 350 (200–800) 250 (150–463) PRBC transfusion (U) 0.1 Ϯ 0.4 0.2 Ϯ 0.5 0.541 Duration of surgery (h) 3.4 Ϯ 1.6 3.5 Ϯ 1.5 0.784 Intraoperative core temperature (°C) 36.1 Ϯ 0.5 36.1 Ϯ 0.6 0.911 Ϯ Ϯ Intraoperative mean FIO2 62 18 61 19 0.816 Mean arterial blood pressure (mm Hg) 85 Ϯ 886Ϯ 9 0.470 Laparoscopic-assisted procedures, n (%) 6 (26%) 37 (40%) 0.335 Epidural analgesia, n (%) 0 (0%) 5 (5%) 0.581 SSI ϭ surgical site infection; BMI ϭ body mass index; SENIC ϭ Study on the Effect of Nosocomial Infection Control; NNISS ϭ National Nosocomial Infection ϭ ϭ ϭ Surveillance System; Hct hematocrit; PRBC packed red blood cell; FIO2 fraction of inspired oxygen. Data are presented as mean Ϯ standard deviation, count (percentage), or a Median (25th–75th quartile). enrolled 116 patients to complete the study with sufficient analgesia was provided with IV intermittent boluses or via power for both outcomes. patient-controlled analgesia with morphine or hydromor- The study enrollment period was from July 2007 to May phone. If an epidural catheter was placed preoperatively, 2008. Adults between 18 and 80 years of age with an ASA epidural analgesia with fentanyl (ϳ25 to 50 ␮m/h) was physical status I to III were included in the study. Patients used intraoperatively and approximately for the first 2 with intestinal obstruction, those in whom the surgeon did hours of the recovery if it was the preference of the not anticipate a primary wound closure, and those with a attending anesthesiologist. A combination of local anes- diagnosed or suspected intraabdominal abscess were ex- thetic and opioid via the epidural was only administered cluded from the study. Patients were also excluded if they after the study was completed. had severe chronic obstructive pulmonary disease, recent Spontaneously breathing patients were given oxygen myocardial infarction, unstable angina, or required oxygen via nasal cannula or Venturi mask. The inspired oxygen preoperatively. concentration was adjusted to maintain pulse oximeter Ն Protocol saturation (Spo2) 95%. Sto2 was measured 15 minutes All patients received antibiotics before surgery according to after weaning from oxygen and thereafter patients received Ն the protocols of the 2 participating institutions, each of supplemental oxygen as required to maintain Spo2 92%. which required administration of appropriate prophylactic antibiotic within an hour before surgical incision. Patients Measurements were given midazolam for premedication, propofol or The duration of surgery, blood loss, intraoperative crystal- etomidate for induction of anesthesia, and succinylcholine loid and colloid administration, hemodynamic variables, or rocuronium for initiation of muscle relaxation. Muscle relaxation was maintained with rocuronium or vecuronium. blood transfusion requirements, and urine output were Anesthesia was maintained with sevoflurane, desflu- recorded. rane, or isoflurane in 30% to 80% oxygen, supplemented Two systems were used to evaluate the SSI risk: the with fentanyl or morphine. Intraoperative IV fluid manage- Study on the Efficacy of Nosocomial Infection Control ment was at the discretion of the attending anesthesiologist, (SENIC) and National Nosocomial Infection Surveillance and consisted of 8 to 10 mL/kg. Core temperature was System (NNISS). The SENIC scoring system assigns 1 point maintained near 36°C by using forced-air warming blan- for each of the following factors: Ն3 underlying diagnoses, kets and fluid warmers. Hair was clipped from the surgical surgery that lasts Ն2 hours, an abdominal site of surgery, site immediately preoperatively, and the skin was prepared and the presence of a contaminated or infected wound.19 with a chlorhexidine-based antiseptic kit. Postoperative The NNISS predicts risk on the basis of the contamination

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 947 Tissue Oxygen and Surgical Site Infection

the probe’s detection site, but this hemoglobin value does not necessarily represent the body’s total hemoglobin concentration. THI monitors the total amount of hemoglobin in the tissue site where the NIRS probe is applied. Therefore, it may be considered more of a perfusion variable. During each set of measurements, mean arterial blood

pressure, heart rate, and Spo2 were recorded simultaneously. We asked the patients to rate their pain using a 10-cm visual analog scale.

An independent investigator not aware of the Sto2 measurements evaluated patients’ wounds daily throughout their hospitalization. SSI was diagnosed according to the surgical wound infection definitions from the Centers for Disease Control and Prevention (CDC).19,25

Data Analysis

Primary outcomes were the Sto2 measured around the Figure 1. InSpectra™ StO2 tissue oxygenation monitor. surgical incision, at the upper arm, and at the thenar

muscle. Surgical site Sto2 measurements were summarized of surgery, the rating of physical status on a scale devel- as a mean value and presented as a post hoc analysis. oped by the American Society of Anesthesiologists, and the Specifically, subcutaneous Sto2 values at the upper one- duration of surgery.20 third, middle one-third, and lower one-third of the surgical incision were averaged into a single “incision” value for Sto2 and tissue hemoglobin index (THI) were measured by an InSpectra™ tissue spectrometer model 650 (Hutchinson each patient. Ϯ Technology Inc., Hutchinson, MN) 75 minutes after surgery Normally distributed data are presented as means SD. and on the first postoperative day (Fig. 1). A time point of Skewed data are presented as medians and interquartile 75 minutes postoperatively was chosen because patients are ranges. After descriptive analysis of all parameters, univar- ␹2 usually weaned from supplemental oxygen at the end of the iate analysis was performed using the test for categorical first postoperative hour, and waiting an additional 15 minutes variables and unpaired, 2-tailed t and Kruskal-Wallis tests Ͻ generally allows full washout of supplemental oxygen. for continuous variables. P values 0.05 were considered statistically significant. Additionally, a multivariate analy- The InSpectra tissue spectrometer measures Sto2 using wide-gap second-derivative NIRS.21,22 The InSpectra spec- sis was performed to assess the independent contribution trometer makes use of the characteristic absorption prop- of each potential variable (SPSS Inc., Chicago, IL). erties of hemoglobin in the near-infrared wavelength range Receiver operating characteristic (ROC) curves were between 680 and 800 nm. The absorption spectrum of light developed for Sto2 in the upper arm and Spo2 on the first remitted from the tissue sample varies mainly with oxyhe- postoperative day to predict surgical wound infections. We moglobin and deoxyhemoglobin concentration; other compared our predictions based only on early postoperative upper arm Sto , using an Sto of 66% as the cutoff chromophores have minimal effect. Sto2 is a measure of 2 2 hemoglobin oxygen saturation of the blood contained in the point, with SENIC predictions. The CDC in the SENIC volume of tissue illuminated by the near-infrared light. The developed a predictive model for the risks of SSI that has 19 maximum depth of the tissue sampled is the distance become the standard. between the probe’s send and receive fibers. Mean measurement depth is half the probe’s spacing. We used a RESULTS

15-mm probe that measures Sto2 of 5- to 8-mm tissue depth. Of the 116 patients enrolled, 23 developed SSI (20%). If the For the forearm and wound sites, this corresponds to diverticulitis cases were excluded, the SSI rate would subcutaneous tissue above the skeletal muscle. For the decrease to 14%. All of the 23 patients who developed SSI thenar muscle site, this depth corresponds to muscle. had superficial incisional site infections. Three of these

Sto2 near the wound was measured 2.5 cm lateral to the patients also developed deep incisional site infections, and incision at the upper, middle, and lower third of the incision. 4 developed peritoneal infections as defined by CDC crite- The upper lateral arm site was chosen for measurement ria. Infections were diagnosed an average of 9 Ϯ 5 days because this site reflects the Sto2 of operative wounds in the after surgery. chest and abdomen even though it is approximately 10 mm Age, ASA physical status, and SENIC and NNISS risk Hg higher than in the wound area.6,12 The thenar muscle site scores were similar in the patients who developed SSI and was chosen because it is one of the best established sites for those who remained uninfected (Table 1). Patients with SSI tissue oximetry measurement.23,24 At each of these sites, the had a greater body mass index. There were no statistically probe was placed for 15 to 30 seconds until a stable oxygen significant or clinically significant differences in the dura- saturation value was obtained. tion of surgery, IV fluid administration, intraoperative THI was measured simultaneously throughout the study temperature, or blood transfusion requirements. Surgical from the same probe as the Sto2 was monitored. THI is not yet technique and use of epidural analgesia were also similar in a well-established value. As with several others, this NIRS- the patients who developed SSI and those who did not based device can provide a hemoglobin value obtained from (Table 1).

948 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 2. Major Study Outcomes: Tissue Oxygen Saturation Given as Percentage Patients Patients with SSI without SSI P value (93 ؍ n) (23 ؍ Measurements (n 75 min after surgery (early postoperative period) Ϯ Ϯ StO2 at surgical incision (%) 44 16 51 20 0.184 THI at surgical incision 3.8 Ϯ 1.1 4.3 Ϯ 2.0 0.229 Ϯ Ϯ StO2 at upper arm (%) 52 22 66 21 0.033 THI at upper arm 3.8 Ϯ 1.3 5.3 Ϯ 3.2 0.041 Ϯ Ϯ StO2 at thenar eminence (%) 69 17 74 16 0.210 Ϯ Ϯ THI at thenar eminence 8.2 3.6 8.9 3.1 0.403 Figure 3. Surgical wound infection rates (%) stratiﬁed by early SpO (%) 97 Ϯ 398Ϯ 2 0.193 2 postoperative tissue oxygen saturation (StO ) (%) values measured Ϯ Ϯ 2 MAP (mm Hg) 94 15 93 14 0.831 at the upper arm. The bars represent the difference between the Pain (cm VAS) 7.6 Ϯ 1.5 6.3 Ϯ 2.6 0.048 a observed infection rates and those expected based on the Study on Pain (cm VAS) 8.0 (7.0–8.3) 6.0 (4.0–8.0) 0.043 the Effect of Nosocomial Infection Control (SENIC) multivariate risk No. of patients on room air 13 (57%) 48 (52%) 0.403 index. No. of patients receiving 8 (35%) 35 (38%) 0.746

nasal O2 at 2–3 L/min No. of patients with FIO 2 (8%) 10 (10%) 0.467 2 measurements were done under supplemental oxygen be- Ͼ0.60 First postoperative day cause of the oxygenation needs of the patients as specified Ϯ Ϯ above. StO2 at surgical incision (%) 44 25 50 24 0.300 THI at surgical incision 4.5 Ϯ 1.3 5.0 Ϯ 2.1 0.410 In the early postoperative period (Table 2), Sto2 at the Ϯ Ϯ StO2 at upper arm (%) 61 13 60 22 0.825 surgical incision did not differ significantly between pa- THI at upper arm 4.1 Ϯ 0.9 4.9 Ϯ 2.6 0.193 Ϯ Ϯ tients who eventually developed SSI and those who re- StO2 at thenar eminence (%) 77 12 79 13 0.538 THI at thenar eminence 11.0 Ϯ 2.7 10.9 Ϯ 2.8 0.824 mained uninfected. The THI near the surgical incision was Ϯ Ϯ SpO2 (%) 95 3972 0.001 25% lower in patients who developed SSI, but this differ- MAP (mm Hg) 89 Ϯ 11 84 Ϯ 13 0.068 ence did not reach statistical significance. Pain (cm VAS) 5.8 Ϯ 2.2 4.9 Ϯ 2.5 0.152 In contrast, Sto2 at the upper lateral arm was signifi- Pain (cm VAS)a 5.0 (4.6–7.8) 5.0 (3.0–6.5) Ϯ Ϯ Ϯ cantly lower in the patients who developed SSI (52 22 Hct (%) 35 5325 0.436 Ϯ ϭ No. of patients on room air 16 (70%) 69 (74%) mm Hg) than in those who did not (66 21 mm Hg; P No. of patients receiving 6 (26%) 23 (25%) 0.033). ROC curve for Sto2 at the upper lateral arm had a nasal O at 2–3 L/min 2 sensitivity of 71% and specificity of 60% using an Sto2 of No. of patients with FIO 1 (4%) 1 (1%) 2 66% as the cutoff point for predicting SSI (Fig. 2). The Ͼ0.60 positive predictive value of this cutoff value was 29%, with SSI ϭ surgical site infection; MAP ϭ mean arterial blood pressure; Hct ϭ a negative predictive value of 90%. The THI measured in ϭ ϭ hematocrit; VAS 10-cm-long visual analog scale for pain; StO2 tissue ϭ ϭ the early postoperative period at the upper arm was also oxygen saturation; THI tissue hemoglobin index; SpO2 pulse oximeter ϭ Ϯ oxygen saturation; FIO2 fraction of inspired oxygen. statistically lower in the patients who developed SSI (3.8 Data are presented as mean Ϯ SD, count (percentage), or a Median 1.3 vs 5.5 Ϯ 3.2 g/dL). ROC curves for THI at the upper (25th–75th quartile). lateral arm in the early postoperative period had a sensitivity of 88% and specificity of 55% for predicting SSI at a A significant proportion of the patients (35%–38%) THI of 4.3. Figure 3 shows the difference between observed could not be weaned from supplemental oxygen at the and expected (based on SENIC scores) infection risk as a

75th-minute measurement period (Table 2). Additionally, function of early postoperative Sto2 at the upper arm. from this group of patients, there were 14 patients (3 with Because intraoperative oxygen concentrations were not SSI and 11 with no SSI) who required oxygen rates of more controlled per protocol, we calculated whether there was than 2 to 3 L/min. During the first postoperative day any correlation between inspired intraoperative oxygen measurements, 70% to 75% of the patients were weaned concentrations and postoperative Sto2 at the upper lateral from supplemental oxygen. Therefore, some of the study arm. There was no correlation (r2 ϭ 0.007).

Figure 2. A, Upper arm tissue oxygen saturation

(StO2) measured 75 minutes after surgery in patients who did and did not develop surgical site infections (SSIs). B, The associated receiver operating characteristic curve.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 949 Tissue Oxygen and Surgical Site Infection

patients who eventually developed SSI according to the CDC Table 3. Independent Contributors to Surgical criteria had lower Sto at the upper arm only 75 minutes after Site Infections (Multivariate Analysis) 2 surgery, which was approximately 9 days before the diagno- Odds 95% Conﬁdence ratio interval P value sis of SSI was made clinically. Body mass index (kg/m2) 1.081 0.984–1.188 0.106 Upper arm Sto2 provided a “fair” area under the ROC Blood loss 1.001 0.998–1.002 0.157 curve and was a better predictor of infection than the

Upper arm StO2 0.975 0.945–1.006 0.108 established SENIC or NNISS risk scores. This point is (postoperative 75 min) important because the high-risk patients we identified Pain VAS (postoperative 1.292 0.944–1.767 0.109 would generally not have been detected using the SENIC 75 min) risk score or other routinely available clinical systems. SpO (postoperative day 1) 0.765 0.603–0.971 0.044 2 As we have shown previously, subcutaneous tissue in ϭ ϭ ϭ StO2 tissue oxygen saturation; VAS visual analog scale; SpO2 pulse oximeter oxygen saturation. the upper lateral arm is relatively well oxygenated and perfused under general anesthesia and in the awake state.1,28,29 An average of 48 Ϯ 19 perforator arterioles from Thenar muscle oxygenation and THI values were both 15 vascular territories supply the integument of the upper higher than those of the subcutaneous tissue values in both extremity. Septocutaneous arteries predominate in the groups of patients. Neither the Sto2 nor the THI measured shoulder, elbow, distal forearm, and hand regions. Muscu- at the thenar eminence 75 minutes after surgery differed locutaneous perforators are more numerous in the upper significantly in patients who did or did not develop SSI arm and proximal forearm. The average perforator size in (Table 2). Early postoperative visual analog scale pain the upper extremity is approximately 0.7 Ϯ 0.2 mm in scores were slightly, but statistically significantly, higher in diameter and supplies an average area of 35 cm2.30,31 patients who developed SSI (Table 2). Therefore, we can extrapolate that our upper arm Sto2

Patients who developed SSI had lower Spo2 values on measurement likely included a site perfused with at least 1 the first postoperative day (95% Ϯ 3% vs 97% Ϯ 2%; P ϭ perforator artery.

0.001). The ROC curve for the first postoperative day Spo2 Leukocyte-mediated oxidative burst and collagen for- had a sensitivity of 75% and specificity of 73% using an mation require oxygen partial pressures of at least 40 mm 32 Spo2 of 95% as the cutoff point for predicting SSI. Hg. Upper arm subcutaneous tissue oxygen tension is Multivariate logistic regression analyses indicated that typically Ͼ50 mm Hg under a fraction of inspired oxygen Ͻ the first postoperative day Spo2 value was the only statis- 0.4 even during sympathetic vasoconstriction. Our tissue tically significant independent factor contributing to SSI. oxygen tension values were lower than those reported Although body mass index, postoperative pain, intraopera- previously (27 to 35 mm Hg). This difference might result tive blood loss, and upper arm Sto2 data provided a good from differences in oxygen monitoring techniques. Sto2 clinical and statistical difference in univariate analysis, they measures hemoglobin’s oxygen saturation in a focal area. did not reach independent significance levels with multi- Sto2 systems calculate the hemoglobin’s oxygen saturation variate statistics (Table 3). in the volume of tissue illuminated by near-infrared light. Because in the majority of the peripheral tissues tested DISCUSSION (surgical wound area and upper arm subcutaneous tissue) The link between tissue oxygenation and surgical wound the THI values were low, we can extrapolate that there was infection is well established. This concept, developed by less perfusion in the site of interest. Therefore, Sto2 pro- Thomas Hunt in the 1960s and 1970s26,27 led to a landmark vides a different oxygenation value than tissue oxygen article by Hopf et al.6 more than a decade ago. Hopf et al. tension, which monitors the partial pressure of free oxygen used a subcutaneous needle-guided tonometric silicon in the tissue. Free oxygen was reported to be the main catheter system into which they inserted a polarographic source of oxygen available for the tissues.33 This interesting Clark-type electrode to monitor tissue oxygen partial pres- concept of tissue oxygenation will continue to be an in- sure. They obtained measurements at 3 different times: tensely debated topic until applicability of monitoring, within 6 hours of surgery, on the first postoperative day, reproducibility of oxygenation data, and until it is proven and on the second postoperative day. They observed that with clinical outcomes. tissue oxygenation was a strong predictor of SSI. We are not the first to use Sto2 to predict SSI. Ives et 16,34 A subcutaneous tissue oxygen monitoring system (LICOX; al. reported that Sto2 of 53% at the surgical incision site, Medical Systems Corp., Greenvale, NY) has been used by measured 12 hours postoperatively, provided 71% sensitiv- various investigators, including us, for decades and is consid- ity and 76% specificity as a test to predict SSI. Our study ered the “gold standard” for tissue oxygen monitoring. How- extends previous work by showing that infection can be ever, the LICOX system is invasive, expensive, and requires predicted shortly after surgery, during the decisive period. approximately 45 minutes of calibration and equilibration But interestingly, Ives et al. could not find any difference in with tissue as well as considerable operator experience to be the upper arm Sto2 measurements between the patients accurate. Therefore, a simple and noninvasive tissue oxygen who did and did not develop SSI. A possible explanation is monitoring system might be easier and more feasible for that Ives et al. measured tissue oxygenation 10 cm below perioperative use. The InSpectra Sto2 system is noninvasive the shoulder tip over the bulk of the biceps muscle. In and relatively easy to use. Our main study question was contrast, we used the lateral upper arm, which was ap- whether NIRS Sto2, measured immediately postoperatively, proximately 15 cm below the shoulder tip, the area between would be able to predict SSI. Our principal finding was that the biceps and triceps brachii’s lateral head. This is the site

950 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA used in most previous studies.6,28,35 The reason behind our oximeters calculate the mean value of oxyhemoglobin across failure to find a statistically significant difference at the all the vessels of the microcirculation of the skin. Therefore, surgical site can be explained (1) by the lack of power and the derived oximeter saturation is a mean blood oxygen small sample size, and (2) because of the stretch of the saturation across arterioles, capillaries, and venules.48 peri-incisional tissues during surgery, there might have Another important limitation is in the claims made by been tissue edema to blunt the extent of oxygenation diagnosing potential infections in a decisive period. This difference. If the difference between the groups continued decisive period was established for antibiotic prophy- consistently, 170 to 180 patients would have provided a laxis as the intervention. Therefore, the decisive period statistically significant difference. responding to increased oxygenation needs to be tested Another interesting finding of the current trial is that before considering the validity of the proposed interven- 7 Spo2 values of the first postoperative day apparently tions. Using an animal model, Knighton et al. showed predicted SSI. Before taken into consideration, some major that oxygen might have a decisive period of at least 6 limitations of these data need to be addressed: (1) Spo2 was hours. not planned as one of the a priori outcomes of this study; In summary, Sto2 measured at the upper arm only 75 (2) the data were gathered from only a few values from a minutes postoperatively predicts development of surgical snapshot period; and (3) although the difference was wound infection after colorectal surgery, even though statistically significant, clinical meaning may be of limited infections were typically diagnosed more than a week later. value. Because high-risk patients can potentially be identified Early detection of patients at special risk of wound noninvasively, it may also be possible to intervene to infection is important because there are well-established improve tissue oxygenation. interventions that improve tissue oxygenation and may thus reduce infection rate and possibly improve surgical ACKNOWLEDGMENTS outcomes. For example, sympathetic nervous system acti- Samual Chen, MD, Luke Reynolds, MSc, Adrian Alvarez, MD, vation triggers vasoconstriction and reduces tissue oxygen- 36 and Gena Harrison, BA (Department of Outcomes Research, ation. A major mediator of sympathetic activity, and one Cleveland Clinic) are acknowledged for their assistance with that can often be treated, is surgical pain. In fact, it is well data acquisition. We appreciate the efforts of Nancy Alsip, established that adequate analgesia improves tissue oxy- PhD, and Gilbert Haugh, MS (OCRSS, University of Louis- 37,38 genation, which is supported by the fact that patients in ville), in medical editing and statistical assistance; and Joseph our study who eventually developed SSI had higher pain Ortner and Hutchinson Technology Inc. (Hutchinson, MN) for scores in the early postoperative period. technical support and for providing tissue oximeters and their Thermoregulatory vasoconstriction is another factor probes. that decreases tissue oxygen tension and perfusion.39,40 As might thus be expected, maintaining perioperative REFERENCES normothermia41 and local warming42 decreases SSI rate. 1. Greif R, Akc¸a O, Horn EP, Kurz A, Sessler DI, Outcomes Supplemental fluid administration increases tissue oxy- Research™ Group. Supplemental perioperative oxygen to re- genation,43,44 but it does not necessarily improve the SSI duce the incidence of surgical wound infection. N Engl J Med 45 2000;342:161–7 rate. However, supplemental fluids may have short- 2. Olsen MA, Chu-Ongsakul S, Brandt KE, Dietz JR, Mayfield J, term tissue oxygenation benefits that should be consid- Fraser VJ. Hospital-associated costs due to surgical site infec- ered despite the ongoing perioperative fluid debates.46 tion after breast surgery. Arch Surg 2008;143:53–60 Finally, supplemental oxygen (i.e., 80% inspired oxygen) 3. Miles AA, Miles EM, Burke J. The value and duration of 1 defence reactions of the skin to the primary lodgement of approximately doubles tissue oxygen partial pressure bacteria. Br J Exp Pathol 1957;38:79–96 37,47 without causing atelectasis. 4. Polk HC Jr. The prophylaxis of infection following operative There are thus at least 4 established interventions that procedures. J Ky Med Assoc 1974;72:139–43 improve tissue oxygenation and could be implemented 5. Classen DC, Evans RS, Pestotnik SL, Horn SD, Menlove RL, relatively easily in patients found to have low Sto during Burke JP. The timing of prophylactic administration of antibi- 2 otics and the risk of surgical-wound infection. N Engl J Med the decisive period. We caution, however, that although the 1992;326:281–6 link between tissue oxygenation and SSI is clear, whether 6. Hopf HW, Hunt TK, West JM, Blomquist P, Goodson WH III, titrating interventions perioperatively to improve Sto2 ac- Jensen JA, Jonsson K, Paty PB, Rabkin JM, Upton RA, von tually reduces infection risk is yet to be proven. Smitten K, Whitney JD. Wound tissue oxygen tension predicts the risk of wound infection in surgical patients. Arch Surg Despite the promising data provided in this trial sup- 1997;132:997–1004 porting the link between oxygenation and SSI, there are 7. Knighton DR, Halliday B, Hunt TK. Oxygen as an antibiotic: some important limitations that need to be addressed. The the effect of inspired oxygen on infection. Arch Surg 1984;119:199–204 values of the upper arm Sto2 were relatively weak (low ROC area), which were also apparent with the nonsignifi- 8. Allen DB, Maguire JJ, Mahdavian M, Wicke C, Marcocci L, Scheuenstuhl H, Chang M, Le AX, Hopf HW, Hunt TK. cant odds ratio obtained from the multivariate analysis. A Wound hypoxia and acidosis limit neutrophil bacterial killing larger trial aiming to reconfirm the current results in a mechanisms. Arch Surg 1997;132:991–6 larger and wider surgical population will be needed. More 9. Babior BM. Oxygen-dependent microbial killing by phagocytes importantly, no study in human subjects has validated the (first of two parts). N Engl J Med 1978;298:659–68 10. Leaper D, Burman-Roy S, Palanca A, Cullen K, Worster D, accuracy of Sto2 by comparing it with the “gold standard” Gautam-Aitken E, Whittle M. Prevention and treatment of tissue oxygen tension monitoring. Additionally, it is impor- surgical site infection: summary of NICE guidance. BMJ tant to recognize that the reflectance spectroscopy-based 2008;337:a1924

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 951 Tissue Oxygen and Surgical Site Infection

11. Lee JT. A new surgical site infection (SSI) prevention guideline. 29. Akça O, Doufas A, Morioka N, Iscoe S, Fisher J, Sessler D. Surg Infect (Larchmt) 2000;1:127–31 Hypercapnia improves tissue oxygenation. Anesthesiology 12. Gottrup F, Firmin R, Chang N, Goodson WH III, Hunt TK. 2002;97:801–6 Continuous direct tissue oxygen tension measurement by a 30. Morris S, Tang M, Geddes C. Vascular anatomical basis of new method using an implantable silastic tonometer and perforator skin flaps. Cir Plast Iberlatinamer 2006;32:1–5 oxygen polarography. Am J Surg 1983;146:399–403 31. Offman SL, Geddes CR, Tang M, Morris SF. The vascular basis 13. Hopf HW, Hunt TK. Comparison of Clark electrode and of perforator flaps based on the source arteries of the lateral optode for measurement of tissue oxygen tension. Adv Exp lumbar region. Plast Reconstr Surg 2005;115:1651–9 Med Biol 1994;345:841–7 32. Hopf HW, Holm J. Hyperoxia and infection. Best Pract Res 14. Soller BR, Idwasi PO, Balaguer J, Levin S, Simsir SA, Vander Clin Anaesthesiol 2008;22:553–69 Salm TJ, Collette H, Heard SO. Noninvasive, near infrared 33. Bitterman H. Bench-to-bedside review: oxygen as a drug. Crit spectroscopic-measured muscle pH and PO2 indicate tissue Care 2009;13:1–8 perfusion for cardiac surgical patients undergoing cardiopul- 34. Ives CL, Harrison DK, Stansby GS. Prediction of surgical site monary bypass. Crit Care Med 2003;31:2324–31 infections after major surgery using visible and near-infrared 15. Soller BR, Ryan KL, Rickards CA, Cooke WH, Yang Y, Soyemi spectroscopy. Adv Exp Med Biol 2007;599:37–44 OO, Crookes BA, Heard SO, Convertino VA. Oxygen satura- 35. Hopf HW, Viele M, Watson JJ, Feiner J, Weiskopf R, Hunt TK, tion determined from deep muscle, not thenar tissue, is an Noorani M, Yeap H, Ho R, Toy P. Subcutaneous perfusion and early indicator of central hypovolemia in humans. Crit Care Med 2008;36:176–82 oxygen during acute severe isovolemic hemodilution in 16. Ives CL, Harrison DK, Stansby GS. Tissue oxygen saturation, healthy volunteers. Arch Surg 2000;135:1443–9 measured by near-infrared spectroscopy, and its relationship 36. Jensen JA, Jonsson K, Goodson WH III, Hunt TK, Roizen MF. to surgical-site infections. Br J Surg 2007;94:87–91 Epinephrine lowers subcutaneous wound oxygen tension. 17. Ives CL, Harrison DK, Stansby G. Prediction of surgical site Curr Surg 1985;42:472–4 infections using spectrophotometry: preliminary results. Adv 37. Akça O, Melischek M, Scheck T, Hellwagner K, Arkiliç C, Kurz Exp Med Biol 2006;578:149–54 A, Kapral S, Heinz T, Lackner FX, Sessler DI. Postoperative 18. Govinda R, Kasuya Y, Mahboobi R, Devarajan J, Akca O. pain and subcutaneous oxygen tension. Lancet 1999;354:41–2 Oxygen saturation at thenar eminence predicts surgical wound 38. Buggy DJ, Kerin MJ. Paravertebral analgesia with levobupiva- infections. American Society of Anesthesiologists Annual caine increases postoperative flap tissue oxygen tension after Meeting, Orlando, FL, 2008 immediate latissimus dorsi breast reconstruction compared 19. Haley RW, Culver DH, Morgan WM, White JW, Emori TG, with intravenous opioid analgesia. Anesthesiology 2004;100: Hooton TM. Identifying patients at high risk of surgical wound 375–80 infection: a simple multivariate index of patient susceptibility 39. Rabkin JM, Hunt TK. Local heat increases blood flow and and wound contamination. Am J Epidemiol 1985;121:206–15 oxygen tension in wounds. Arch Surg 1987;122:221–5 20. Culver DH, Horan TC, Gaynes RP, Martone WJ, Jarvis WR, 40. Sheffield CW, Sessler DI, Hopf HW, Schroeder M, Moayeri A, Emori TG, Banerjee SN, Edwards JR, Tolson JS, Henderson TS, Hunt TK, West JM. Centrally and locally mediated thermo- Hughes JM; National Nosocomial Infections Surveillance Sys- regulatory responses alter subcutaneous oxygen tension. tem. Surgical wound infection rates by wound class, operative Wound Repair Regen 1996;4:339–45 procedure, and patient risk index. Am J Med 1991;91:152S–7S 41. Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to 21. Myers DE, Anderson LD, Seifert RP, Ortner JP, Cooper CE, reduce the incidence of surgical-wound infection and shorten Beilman GJ, Mowlem JD. Noninvasive method for measuring hospitalization. Study of Wound Infection and Temperature local hemoglobin oxygen saturation in tissue using wide gap Group. N Engl J Med 1996;334:1209–15 second derivative near-infrared spectroscopy. J Biomed Opt 42. Melling AC, Ali B, Scott EM, Leaper DJ. Effects of preoperative 2005;10:034017 warming on the incidence of wound infection after clean 22. Myers DE, Cooper CE, Beilman GJ, Mowlem JD, Anderson LD, surgery: a randomised controlled trial. Lancet 2001;358:876–80 Seifert RP, Ortner JP. A wide gap second derivative NIR 43. Arkilic CF, Taguchi A, Sharma N, Ratnaraj J, Sessler DI, Read spectroscopic method for measuring tissue hemoglobin oxy- TE, Fleshman JW, Kurz A. Supplemental perioperative fluid gen saturation. Adv Exp Med Biol 2006;578:217–22 administration increases tissue oxygen pressure. Surgery 23. Creteur J. Muscle StO2 in critically ill patients. Curr Opin Crit 2003;133:49–55 Care 2008;14:361–6 44. Jonsson K, Jensen JA, Goodson WH III, West JM, Hunt TK. 24. Cohn SM, Nathens AB, Moore FA, Rhee P, Puyana JC, Moore EE, Beilman GJ. Tissue oxygen saturation predicts the devel- Assessment of perfusion in postoperative patients using tissue opment of organ dysfunction during traumatic shock resusci- oxygen measurements. Br J Surg 1987;74:263–7 tation. J Trauma 2007;62:44–54 45. Kabon B, Akca O, Taguchi A, Nagele A, Jebadurai R, Arkilic 25. Horan TC, Gaynes RP, Martone WJ, Jarvis WR, Emori TG. CF, Sharma N, Ahluwalia A, Galandiuk S, Fleshman J, Sessler CDC definitions of nosocomial surgical site infections, 1992: a DI, Kurz A. Supplemental intravenous crystalloid administra- modification of CDC definitions of surgical wound infections. tion does not reduce the risk of surgical wound infection. Infect Control Hosp Epidemiol 1992;13:606–8 Anesth Analg 2005;101:1546–53 26. Hunt TK, Dunphy JE. Effects of increasing oxygen supply to 46. Liu B, Finfer S. Intravenous fluids in adults undergoing healing wounds. Br J Surg 1969;56:705 surgery. BMJ 2009;338:b2418 27. Hunt TK, Linsey M, Sonne M, Jawetz E. Oxygen tension and 47. Edmark L, Kostova-Aherdan K, Enlund M, Hedenstierna G. wound infection. Surg Forum 1972;23:47–9 Optimal oxygen concentration during induction of general 28. Akca O, Sessler DI, Delong D, Keijner R, Ganzel B, Doufas AG. anesthesia. Anesthesiology 2003;98:28–33 Tissue oxygenation response to mild hypercapnia during car- 48. Thorn CE, Matcher SJ, Meglinski IV, Shore AC. Is mean blood diopulmonary bypass with constant pump output. Br J An- saturation a useful marker of tissue oxygenation? Am J Physiol aesth 2006;96:708–14 Heart Circ Physiol 2009;296:H1289–95

952 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA American Society of Critical Care Anesthesiologists

Section Editor: Michael J. Murray

High-Fidelity Simulation Demonstrates the Inﬂuence of Anesthesiologists’ Age and Years from Residency on Emergency Cricothyroidotomy Skills Lyndon W. Siu, MBBS, FANZCA,* Sylvain Boet, MD,* Bruno C. R. Borges, MD,* Heinz R. Bruppacher, MD, FMH,* Vicki LeBlanc, PhD,† Viren N. Naik, MD, MEd, FRCPC,* Nicole Riem, MD,* Deven B. Chandra, MD, MEd, FRCPC,* and Hwan S. Joo, MD, FRCPC*

BACKGROUND: Age-related deterioration in both cognitive function and the capacity to control fine motor movements has been demonstrated in numerous studies. However, this decline has not been described with respect to complex clinical anesthesia skills. Cricothyroidotomy is an example of a complex, lifesaving procedure that requires competency in the domains of both cognitive processing and fine motor control. Proficiency in this skill is vital to minimize time to reestablish oxygenation during a “cannot intubate, cannot ventilate” scenario. In this prospective, controlled, single-blinded study, we tested the hypothesis that age affects the learning and performance of emergency percutaneous cricothyroidotomy in a high-fidelity simulated cannot intubate/cannot ventilate scenario. METHODS: Thirty-six staff anesthesiologists (19 aged younger than 45 years and 17 older than 45 years) managed a high-fidelity cannot intubate/cannot ventilate scenario in a high-fidelity simulator before and after a 1-hour standardized training session. The group division cutoff age of 45 years was based on the median age of our sample subject population before enrollment. The scenarios required the insertion of an emergency percutaneous cricothyroidotomy. We compared cricothyroidotomy skills in the older group with those in the younger group using procedural time, 5-point task-specific checklist score, and global rating scale score. Correlation based on age, years from residency, weekly clinical hours worked, previous continuing medical education in airway management, and previous simulation experience was also performed. RESULTS: In both prestandardization and poststandardization, age and years from residency correlated with procedural time, checklist scores, and global rating scores. Baseline, prestandardization variables were all better for the younger group, with a mean age of 37 years, compared with the older group, with a mean age of 58 years. Procedural time was 100 (72–128) seconds versus 152 (120–261) seconds. Checklist scores were 7.0 (6.1–8.0) versus 6.0 (4.8–8.0). Global rating scale scores were 22.0 (17.8–29.8) versus 17.5 (10.4–20.6). After the 1-hour standardized training session, the younger group continued to perform better than the older group with procedural time of 75 (66–91) seconds versus 87 (78–123) seconds, checklist scores of 10.0 (9.1–10.0) versus 9.0 (8.0–10.0), and global rating scale scores of 35.0 (32.1–35.0) versus 32.0 (29.0–33.8). Regression analysis was performed on the poststandardization data. Both age and years from residency independently affected procedural time, checklist scores, and global rating scale scores (all P Ͻ 0.05). CONCLUSIONS: Baseline proficiency with simulated emergency cricothyroidotomy is associated with age and years from residency. Despite standardized training, operator age and years from residency were associated with decreased proficiency. Further research should explore the potential of using age and years from residency as factors for implementing periodic continuing medical education. (Anesth Analg 2010;111:955–60)

ge-related deterioration in cognitive functioning and Cricothyroidotomy is a complex lifesaving procedure fine motor skills has been demonstrated in numerous that requires competency in the domains of both cognitive Astudies and reviews.1,2 The theoretical clinical signifi- processing and fine motor control. Proficiency in this skill is cance of aging has been extensively addressed in the litera- vital because minimizing time to achieve oxygenation is 3–5 ture. However, the specific impact of age on a particular essential in a cannot intubate/cannot ventilate situation. anesthetic procedure has never been objectively assessed. Age-related impact on this skill, and potentially other important skills, should be recognized to facilitate the From the *Department of Anesthesia, St. Michael’s Hospital, University of development of appropriate educational strategies. Toronto; and †Wilson Centre, University Health Network, Department of 6 Medicine, University of Toronto, Toronto, Ontario, Canada. Using a low-fidelity static model, Wong et al. observed Accepted for publication May 7, 2010. that anesthesiologists aged younger than 45 years per- Funded by departmental funds. formed cricothyroidotomies faster than those older than Disclosure: The authors report no conflicts of interest. 45 years. In another study, John et al.7 demonstrated that Address correspondence and reprint requests to Hwan S. Joo, MD, FRCPC, under stressful conditions in simulated cannot intubate/ St. Michael’s Hospital, 30 Bond St., Toronto, ON, M5B1W8, Canada. Address e-mail to [email protected]. cannot ventilate scenarios, procedural times were longer Copyright © 2010 International Anesthesia Research Society compared with times achieved on static mannequin DOI: 10.1213/ANE.0b013e3181ee7f4f models. However, neither of the 2 studies controlled for

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 955 Age and Percutaneous Cricothyroidotomy Skills confounders such as previous procedural and/or simula- did not have prior knowledge of the content of any tion experience. scenario. The primary purpose of this prospective, controlled, All scenarios were completed in a simulated operating single-blinded study was to investigate whether age affects room environment containing a high-fidelity mannequin learning and performance of simulated cannot intubate/cannot (Sim Man; Laerdal, Kent, UK) equipped with an anatomi- ventilate emergency percutaneous cricothyroidotomies. cally accurate larynx, properly designed for performance of We hypothesize that age may affect the ability to learn cricothyroidotomies, and standard monitors (electrocardio- and perform emergency cricothyroidotomy in a high- gram, noninvasive arterial blood pressure, oxygen satura- fidelity simulated setting. tion as measured by pulse oximetry, and end-tidal CO2). Appropriate equipment and airway devices in the room METHODS included multiple-sized laryngoscopes, endotracheal tubes, After institutional ethics approval, 36 attending anesthesi- laryngeal mask airways, gum elastic bougie, and anesthesia ologists in a tertiary care teaching hospital (19 aged drug cart. Airway adjuncts kept outside the room included younger than 45 years and 17 older than 45 years) managed a fiberoptic bronchoscope, a videolaryngoscope (Glide- ® a high-fidelity cannot intubate/cannot ventilate scenario Scope ; Verathon, Inc., Bothell, WA), intubating laryngeal immediately before and after a standardization process that mask airways, and a cricothyroidotomy kit. Two simula- included cricothyroidotomy training. The age 45 years was tion assistants played the scripted roles of a nurse and a chosen to divide the anesthesiologists into equal groups, junior resident. based on the median age of 39 possible participants before study enrollment. To clarify for this article, we have named Sample Size Calculation the “older than 45 years” group the “older” group and the We hypothesized that there would be a difference in the “younger than 45 years” group the “younger” group. All 39 procedural time in favor of the younger group. One stan- attending anesthesiologists in the Department of Anesthe- dard deviation for procedural times between the 2 groups sia of our institution were approached for recruitment; 3 was defined to be a clinically important difference. We declined to participate. Informed consent and confidential- calculated that 17 participants per group would be required ity agreements were obtained from all participants. to achieve a difference of 1 SD between groups in the poststandardization procedural times based on a 2-tailed ␣ Standardization Process of 0.05 and a power of 0.8. All participants received a 1-hour introduction of the simulation center consisting of an orientation session to the Data Collection human patient simulator and participation in an introduc- Demographic data including age, the number of years after tory high-fidelity airway management scenario, which was graduation from anesthesia residency, the number of hours not part of the study. of clinical practice per week, previous simulation and/or The prestandardization scenario was a cannot intubate/ airway simulation experience, and previous cricothyroid- cannot ventilate scenario. In this videotaped session, an otomy experience on both patients and mannequins were actor playing a second-year resident calls for help after 2 collected. All cricothyroidotomy performances (2 per sub- unsuccessful intubation attempts and difficulty with face- ject) were video-recorded and later evaluated by 2 blinded mask ventilation. The simulated patient’s oxygen satura- evaluators. The evaluators assigned were blinded to the tion was 89% when the subject arrived and decreased by study outcome, each other’s scores, and whether the 10% every minute. All alternative methods of intubation video was from the pre- or posttest cannot intubate/ were intended to be unsuccessful. This was accomplished cannot ventilate session. Blinding to age was attempted by changing the mannequin’s airway anatomy. The cervical but may not have been possible as the approximate age spine was immobilized, the tongue was made macroglos- may have been guessed because participants may have sic, and the vocal cords were adducted. When requested by been recognized. the participant, an ear-nose-throat surgeon and second anesthesiologist were called but would not be available. Outcome Measurement The scenario was designed to necessitate an emergency The primary outcome was the comparison of cricothyroid- percutaneous cricothyroidotomy using a 4.0-mm Melker otomy performances between the younger and the older emergency cricothyroidotomy catheter set (C-TCCS-400; groups. Performance was assessed with 3 variables: proce- Cook Inc., Bloomington, IN). The scenario only ended with dural time and 2 previously validated tools for procedural successful cricothyroidotomy, defined by positive capnog- skill evaluation (a 3-point task-specific checklist [Appendix raphy on the monitor. 1] and global rating scale [GRS] [Appendix 2]).8 Procedural Immediately after the first cannot intubate/cannot ven- time was measured during video review and was defined tilate scenario, a teaching session including practical in- as the time between the first instances when the subject structions on percutaneous cricothyroidotomy insertion grasps any equipment from the cricothyroidotomy kit to and video-assisted debriefing was provided. The same the time of successful cricothyroidotomy. The cricothyroid- individual conducted all debriefing sessions (LWS). otomy checklist was based on observation of common but Immediately after the standardization session, all par- important mistakes made by novice operators based on the ticipants managed the poststandardization scenario, an study by Friedman et al.8 A score of 0, 1, or 2 was given identical cannot intubate/cannot ventilate scenario to that when a stage was not performed, poorly performed, or presented during the prestandardization session. Subjects performed well, respectively. The GRS uses more general

956 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Demographics P value (17 ؍ Age >45y(n (19 ؍ Characteristics Age <45y(n Age 36.7 (Ϯ3.4) 58.0 (Ϯ8.3) Ͻ0.001 Gender (male/female) 13 (68.4%)/6 (32.6%) 15 (88.2%)/2 (11.8%) NS Years from residency graduation 4 (0.5–7.0) 28 (14.8–33.3) Ͻ0.001 Hours of clinical work per week 43.4 (Ϯ3.8) 38.5 (Ϯ8.6) 0.031 Previous simulation experience 16 (84.2%) 5 (29.4%) 0.002 Previous simulation experience in airway 4 (21.1%) 2 (11.8%) NS management Attended airway lecture or continued medical 12 (63.1%) 11 (64.7%) NS education within 10 y Previous percutaneous cricothyroidotomy experience 15 (78.9%) 7 (41.2%) 0.039 Patient 3 (15.8%) 3 (17.7%) NS Mannequin or pig 12 (63.2%) 4 (23.5%) 0.023 NS ϭ not signiﬁcant. Values are mean (Ϯstandard deviation), median (interquartile range), or number (percentage).

Table 2. Cricothyroidotomy Performance Before and After Standardization Prestandardization Poststandardization Age <45 y Age >45 y P value Age <45 y Age >45 y P value Procedural time (s) 100 (72–128) 152 (120–261) 0.003 75 (66–91)* 87 (78–123)* 0.018 Checklist scorea 7.0 (6.1–8.0) 6.0 (4.8–8.0) 0.024 10.0 (9.1–10.0)* 9.0 (8.0–10.0)* 0.005 Global rating scale scoreb 22.0 (17.8–29.8) 17.5 (10.4–20.6) 0.004 35.0 (32.1–35.0)* 32.0 (29.0–33.8)* 0.012 Data are median (interquartile range). a Checklist score ranges from 0 to 10. b Global rating scale ranges from 0 to 35. * Signiﬁcant difference with corresponding age group prestandardization data (P Ͻ 0.01).

descriptors and focuses on the overall performance of the dependent variables at P Յ 0.1 would be taken in subject, not the specifics of the manual task. consideration in the final analysis of age as a predictor variable. Statistical Analysis Analysis was performed using SPSS 11.0 software (SPSS, RESULTS Inc., Chicago, IL). To determine the reliability of assess- Thirty-six attending anesthesiologists were recruited over 9 ments provided by the 2 evaluators, intraclass correlation months. General demographics of the younger (aged 37 Ϯ coefficients (ICCs) were calculated for checklist and GRS Ϯ 9 3 years) and older (aged 58 8 years) groups are shown in scores. We compared cricothyroidotomy performances Table 1. between younger and older groups both before and after Interrater reliability was strong for both checklist and standardization. Procedural times, checklist scores, and GRS scores (checklist: ICC ϭ 0.911; GRS: ICC ϭ 0.837) (both GRS scores were compared using the Mann-Whitney test. P Ͻ 0.05). Correlations between demographic variables and proce- The performance of both age groups significantly im- dural time, checklist scores, and GRS scores were per- proved after teaching for all 3 variables (Table 2). Prestan- formed using Spearman’s correlation. All P values were 2 dardization procedural time was longer and checklist and sided except for correlation, which was 1 sided. Signifi- GRS scores were lower in the older group (Table 2). Ͻ cance for correlation was set at a value of P 0.05. Poststandardization procedural time was longer for the Significance for comparison between younger and older, older group compared with the younger group. Both and between pre- and poststandardization variables was checklist and GRS scores were lower in the older group set at a value of P Ͻ 0.025 to account for Bonferroni compared with the younger group (Table 2). correction. Age and years from residency correlated with proce- We used multiple regression analysis to account for dural time, checklist scores, and GRS scores, both before differences between the 2 groups regarding variables that and after standardization (Table 3). Previous simulation could influence our primary outcomes. Initially, we entered experience also correlated with GRS scores, and with our primary outcomes measurements (procedural times, procedural time but only before standardization. Weekly GRS score, and checklist score) as dependent variables and clinical hours correlated with GRS scores, but only after the other factors such as age, the number of hours of clinical standardization (Table 3). practice per week, previous simulation and/or airway Multiple regression analysis was performed on post- simulation experience, and previous cricothyroidotomy standardization data. Both age and years from residency experience on both patients and mannequins as predictor independently affected procedural time, checklist scores, variables. Any of these factors that correlated with the and GRS scores (all P Ͻ 0.05).

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 957 Age and Percutaneous Cricothyroidotomy Skills

Table 3. Predictor Variables: Correlation and P Values Airway continued Previous simulation Weekly clinical Years from medical education experience hours residency Age Statistical analysis of prestandardization data Procedural time Ϫ0.257 Ϫ0.437 Ϫ0.103 0.370 0.362 P ϭ 0.065 P ϭ 0.004* P ϭ 0.275 P ϭ 0.013* P ϭ 0.015* Checklist scores Ϫ0.031 0.186 0.130 Ϫ0.411 Ϫ0.503 P ϭ 0.43 P ϭ 0.139 P ϭ 0.225 P ϭ 0.006* P ϭ 0.001* Global rating scale scores 0.257 0.437 0.007 Ϫ0.454 Ϫ0.497 P ϭ 0.065 P ϭ 0.004* P ϭ 0.483 P ϭ 0.003* P ϭ 0.001* Statistical analysis of poststandardization data Procedural time Ϫ0.007 Ϫ0.041 0.070 0.341 0.310 P ϭ 0.484 P ϭ 0.407 P ϭ 0.344 P ϭ 0.021* P ϭ 0.033* Checklist scores 0.096 0.177 0.187 Ϫ0.333 Ϫ0.333 P ϭ 0.289 P ϭ 0.151 P ϭ 0.138 P ϭ 0.024* P ϭ 0.024* Global rating scale scores 0.076 0.302 0.344 Ϫ0.564 Ϫ0.521 P ϭ 0.330 P ϭ 0.037* P ϭ 0.02* P Ͻ 0.001* P ϭ 0.001* Values represent Spearman correlation coefﬁcient and P values. * Statistically signiﬁcant (P Ͻ 0.05).

DISCUSSION include the ability to perform complex tasks rapidly, to Before standardization, the younger group significantly process incoming information and make complex deci- outperformed the older group as assessed by all 3 variables. sions, and to perform effectively in a stressful environ- This difference in baseline (prestandardization) perfor- ment.11 All of them are required for the practice of mances between the 2 groups may reflect differences in anesthesiology as well. previous simulation exposure and/or cricothyroidotomy Proficient performance of emergency cricothyroidotomy experience on mannequins (Table 1). Of note, however, requires both the cognitive ability to recall essential steps those who had previously performed cricothyroidotomies for the procedure and psychomotor skills to execute on mannequins had only done so once or twice and none in planned actions efficiently. Both aspects may be adversely the 6 months before the study. In addition, a previous study affected by increasing age as a result of the general slowing suggests that cricothyroidotomy performance in low- of central cognitive processes.12–14 One probable neuro- fidelity models declines to near baseline after 3 months.10 physiological mechanism is the loss of neural connectivity Both groups improved significantly in time and checklist or decreased levels of neurotransmitters in the aging and GRS scores after standardization. This trend suggests brain.13,14 The difference in poststandardization procedural that training is effective in improving both the cognitive times, however, supports an age-related decline in psy- and psychomotor aspects of emergency cricothyroidotomy chomotor skills required for this procedure. Another pos- skills. sibility is that years from residency or training is just as To investigate whether the difference in performance important. Time from formal residency or training may was attributable to age, we focused our comparative anal- affect technical skills, even after standardized training. ysis on the poststandardization data. By priming partici- Nonetheless, the clinical significance of the difference in pants with a 1-hour standardization session consisting of 2 procedural times, measured within a simulation setting, is airway management simulation scenarios, debriefing, and unknown and may be difficult to determine. a practical instructional session on cricothyroidotomy im- It should be clear that our study was not focused on mediately before the study scenario, we aimed to minimize finding a cutoff age beyond which more training on crico- bias caused by the variable simulation and cricothyroid- thyrotomies is necessary, rather, simply on showing that otomy experience of participants. The results of this study age may well be a factor that affects learning and perfor- demonstrated that after standardized exposure to high- mance of emergency cricothyroidotomies. The cutoff age of fidelity simulation and after standardized teaching, an- 45 years was used mainly as a median-age dividing point. esthesiologists who are older and further away from Further research should explore the potential of using residency training require more time to perform emer- age and years from residency as factors for implementing gency percutaneous cricothyroidotomies and have lower periodic continuing medical education. As shown in our checklist and GRS scores. study, the older anesthesiologists benefited more from We failed to find studies in the literature that addressed standardized cricothyroidotomy training than the younger the impact of age on the performance of a specific anes- anesthesiologists. However, this may also be attributable to thetic procedure, except for the study by Wong et al.,6 a ceiling effect because the younger anesthesiologists may which also used cricothyroidotomies but that study was have been closer to their maximal potential in their pres- not primarily powered for this specific objective. However, tandardization scenario. the literature is rich with articles that address patient safety This study has several limitations. Even with our stan- concerns and aging physicians, including surgeons and dardized training, we might not have equalized previous anesthesiologists.3,4 Psychomotor and perceptual processes knowledge base in performing cricothyroidotomies. Previ- that are required in aviation that deteriorate with age ous knowledge and recent airway training in many

958 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA younger participants may have contributed to better times performed rarely, if at all during routine clinical practice, and scores. All participants were attending anesthesiolo- lack of experience further contributes to failure in real gists recruited from a single tertiary academic center. A life.15–17 Further research should explore the potential of multicenter study would render the results more general- using age and years from residency as factors for imple- izable. Because our study was conducted using 4.0-mm menting periodic specific continuing medical education. Melker cricothyroidotomy kits (C-TCCS-400; Cook Inc.), the results might not reflect performance using other AUTHOR CONTRIBUTIONS commercial kits, especially those that do not use LWS helped with study design, conduct of study, and manuscript Seldinger’s technique. In addition, as for all simulation preparation; SB helped with conduct of study, data collection and research, there are conflicting views on whether results can analysis, and manuscript preparation; BCRB helped with conduct be extrapolated to real-life clinical settings. of study and manuscript preparation; HRB helped with data In conclusion, there may well be an age/number of collection; VL helped with data analysis; VNN helped with study years from residency–related decline in proficiency for design; NR and DBC helped with manuscript preparation; and emergency cricothyroidotomy skills in a high-fidelity simu- HSJ helped with study design, conduct of study, and manuscript lated setting. Because emergency cricothyroidotomies are preparation, and is responsible for archival.

Appendix 1. Task-Specific Checklist for Cricothyroidotomy Score Task 0 1 2 Aspiration to identify trachea Does not aspirate Performed inadequately Aspirates with air-filled or fluid-filled syringe Ventilation during Does not ventilate during the Ventilates for part of the Ventilates for the entire cricothyroidotomy cricothyroidotomy duration of the duration of the cricothyroidotomy cricothyroidotomy Correct caudal angling during Cephalad angle 90° to trachea 45° caudad guidewire insertion (not during needle insertion) Adequate skin and membrane Does not use the scalpel for Performed inadequately Cuts skin and incision incision cricothyroid membrane with the scalpel Correct use of dilator and Attempts to insert cricothyroidotomy Dilates separately and Railroad entire cricothyroidotomy cannula without dilator in place then railroad entire assembly (dilator and assembly cricothyroidotomy cannula)

Appendix 2. Global Rating Scale for Cricothyroidotomy Score 12345 Preparation for procedure Did not organize equipment well. Equipment generally organized. All equipment neatly organized, Had to stop procedure Occasionally had to stop and prepared, and ready for use frequently to prepare prepare items equipment Respect for tissue Frequently used unnecessary Careful handling of tissue but Consistently handled tissues force on tissue or caused occasionally caused appropriately with minimal damage inadvertent damage damage Time and motion Many unnecessary moves Efficient time/motion but some Clear economy of movement unnecessary moves and maximum efficiency Instrument handling Repeatedly made tentative or Competent use of instruments Fluid moves with instruments awkward moves with but occasionally appeared and no awkwardness instruments stiff or awkward Flow of procedure Frequently stopped procedure Demonstrated some forward Obviously planned course of and seemed unsure of next planning with reasonable procedure with effortless move progression of procedure flow from 1 move to the next Knowledge of procedure Deficient knowledge Knew all important steps of Demonstrated familiarity with procedure all aspects of procedure Overall performance Very poor Competent Clearly superior

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 959 Age and Percutaneous Cricothyroidotomy Skills

REFERENCES 10. Amrhein PC, Stelmach GE, Goggin NL. Age differences in the 1. Krampe RT. Aging, expertise and fine motor movement. Neu- maintenance and restructuring of movement preparation. Psy- rosci Biobehav Rev 2002;26:769–76 chol Aging 1991;6:451–66 2. Birren J, Schaie K. Handbook of the Psychology of Aging. 5th 11. Eyraud MY, Borowsky MS. Age and pilot performance. Aviat ed. San Diego: Academic Press, 2001:313–48 Space Environ Med 1985;56:553–8 3. Katz JD. Issues of concern for the aging anesthesiologist. Anesth 12. Birren J, Schaie K. Handbook of the Psychology of Aging. 5th Analg 2001;92:1487–92 ed. San Diego: Academic Press, 2001:288–312 4. Greenfield LJ, Proctor MC. When should a surgeon retire? Adv 13. Salthouse TA. The processing-speed theory of adult age differ- Surg 1999;32:385–93 ences in cognition. Psychol Rev 1996;103:403–28 5. Boom-Saad Z, Langenecker SA, Bieliauskas LA, Graver CJ, 14. Birren J, Schaie K. Handbook of the Psychology of Aging. 5th O’Neill JR, Caveney AF, Greenfield LJ, Minter RM. Surgeons ed. San Diego: Academic Press, 2001:135–60 outperform normative controls on neuropsychologic tests, but 15. Schaumann N, Lorenz V, Schellongowski P, Staudinger T, age-related decay of skills persists. Am J Surg 2008;195:205–9 6. Wong DT, Prabhu AJ, Coloma M, Imasogie N, Chung FF. What Locker GJ, Burgmann H, Pikula B, Hofbauer R, Schuster E, is the minimum training required for successful cricothyroid- Frass M. Evaluation of Seldinger technique emergency crico- otomy. Anesthesiology 2003;98:349–53 thyroidotomy versus standard surgical cricothyroidotomy in 7. John B, Suri I, Hillermann C, Mendonca C. Comparison of 200 cadavers. Anesthesiology 2005;102:7–11 cricothyroidotomy on manikin vs. simulator: a randomised 16. Ravlo O, Bach V, Lybecker H, Moller JT, Werner M, Nielsen cross-over study. Anaesthesia 2007;62:1029 HK. A comparison between two emergency cricothyroidotomy 8. Friedman Z, You-Ten KE, Bould MD, Naik V. Teaching lifesav- instruments. Acta Anaesthesiol Scand 1987;31:317–9 ing procedures: the impact of model fidelity on acquisition and 17. Chang RS, Hamilton RJ, Carter WA. Declining rate of crico- transfer of cricothyrotomy skills to performance on cadavers. thyrotomy in trauma patients with an emergency medicine Anesth Analg 2008;107:1663–9 residency: implications for skills training. Acad Emerg Med 9. Landis JR, Koch GG. An application of hierarchical kappa-type 1998;5:247–51 statistics in the assessment of majority agreement among multiple observers. Biometrics 1977;33:363–74

960 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Adaptive Support Ventilation with Protocolized De-Escalation and Escalation Does Not Accelerate Tracheal Extubation of Patients After Nonfast-Track Cardiothoracic Surgery

Dave A. Dongelmans, MD, MSc,* Denise P. Veelo, MD, PhD,*†‡ Jan M. Binnekade, PhD,* Bas A.J.M. de Mol, MD, PhD,§ Anna Kudoga, MS,* Frederique Paulus, MSc,* and Marcus J. Schultz, MD, PhD*‡࿣

BACKGROUND: It is uncertain whether adaptive support ventilation (ASV) accelerates weaning of nonfast-track cardiothoracic surgery patients. A lower operator set %-minute ventilation with ASV may allow for an earlier definite switch from controlled to assisted ventilation, potentially hastening tracheal extubation. We hypothesized that ASV using protocolized de-escalation and escalation of operator set %-minute ventilation (ASV-DE) reduces time until tracheal extubation compared with ASV using a fixed operator set %-minute ventilation (standard ASV) in uncomplicated patients after nonfast-track coronary artery bypass graft. METHODS: We performed a randomized controlled trial comparing ASV-DE with standard ASV. With ASV-DE, as soon as body temperature was Ͼ35.0°C with pH Ͼ7.25, operator set %-minute ventilation was decreased stepwise to a minimum of 70%. RESULTS: Sixty-three patients were randomized to ASV-DE, and 63 patients to standard ASV. The duration of mechanical ventilation was not different between groups (10.8 [6.5–16.1] vs 10.7 [6.6–13.9] hours, ASV-DE versus standard ASV; P ϭ 0.32). Time until the first assisted breathing period was shorter (3.1 [2.0–6.7] vs 3.9 [2.1–7.5] hours) and the number of assisted ventilation episodes was higher (78 [34–176] vs 57 [32–116] episodes), but differences did not reach statistical significance. The duration of assisted ventilation episodes that ended with tracheal extubation was different between groups (2.5 [0.9–4.6] vs 1.4 [0.3–3.5] hours, ASV-DE versus standard ASV; P Ͻ 0.05). CONCLUSION: Compared with standard ASV, weaning of patients after nonfast-track coronary artery bypass graft using ASV with protocolized de-escalation and escalation does not shorten time to tracheal extubation. (Anesth Analg 2010;111:961–7)

daptive support ventilation (ASV) is an advanced in another study of fast-track cardiothoracic surgery pa- closed-loop mode of mechanical ventilation (MV) tients in which ASV was compared with pressure-regulated Athat maintains an operator preset minute ventila- volume controlled ventilation.4 However, in a recent study tion. ASV adjusts respiratory rates and pressure levels of nonfast-track cardiothoracic surgery patients, ASV com- 1 according to measured lung mechanics at each breath. In pared with traditional pressure support ventilation did not addition, ASV automatically switches between controlled shorten the time to tracheal extubation.3 2 and assisted ventilation according to the patient’s status. A lower operator set %-minute ventilation with ASV Previous randomized controlled trials have tested the effi- may allow for earlier and more frequent switches from cacy of ASV in patients after cardiothoracic surgery.3–5 In a controlled to assisted ventilation. Indeed, patients whose study of fast-track cardiothoracic surgery patients, ASV lungs are ventilated with lower minute volumes could be compared with synchronized intermittent mandatory ven- forced to breathe spontaneously sooner because arterial tilation or traditional pressure support ventilation short- 6 ened the time to tracheal extubation.5 This was confirmed Pco2 thresholds for breathing are reached faster. Second, in the above-mentioned study of nonfast-track cardiotho- 3 From the Departments of *Intensive Care Medicine, †Anesthesiology, and racic surgery patients, patients were able to trigger the §Cardiothoracic Surgery; and ‡Laboratory of Experimental Intensive ventilator early in the weaning process, at least suggesting Care and Anesthesiology (L.E.I.C.A.), Academic Medical Center, Univer- sity of Amsterdam; and ࿣HERMES Critical Care Group, Amsterdam, The that a lower operator set %-minute could push patients Netherlands. toward longer periods of assisted ventilation and thereby Accepted for publication June 18, 2010. earlier tracheal extubation. Supported by the Department of Intensive Care Medicine, Academic Medi- In a randomized controlled trial of patients after cal Center. planned and uncomplicated nonfast-track coronary artery Presented, in part, at the ATS conference, San Diego, CA, May 18, 2009. Disclosure: The authors report no conflicts of interest. bypass graft (CABG), we compared ASV using protocol- Address correspondence and reprint requests to Dave A. Dongelmans, MD, ized de-escalation and escalation of operator set %-minute MSc, Department of Intensive Care Medicine, G3-212, Academic Medical ventilation (ASV-DE) with ASV using a fixed operator set Center, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. Address e-mail to [email protected]. %-minute ventilation (standard ASV). We hypothesized Copyright © 2010 International Anesthesia Research Society that ASV-DE reduces time to tracheal extubation compared DOI: 10.1213/ANE.0b013e3181efb316 with standard ASV.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 961 Adaptive Support Ventilation Protocolized with De-Escalation After Cardiothoracic Surgery

METHODS to achieve a cardiac index Ն2.5 L/min/m2 or a mixed Patients and Setting venous oxygenation Ͼ60%. Consecutive patients after elective and uncomplicated Sedation and analgesics were given according to local CABG admitted to the 28-bed intensive care unit (ICU) of protocol for postoperative cardiothoracic surgery patients. the Academic Medical Center, Amsterdam, The Nether- Propofol was given for sedation via continuous infusion. lands, were eligible for inclusion. The study protocol was Infusion of propofol was stopped instantly when core approved by the local IRB, and preoperative written and temperature reached 35°C. The confusion assessment signed informed consent was obtained from eligible pa- method for the ICU was used to screen for delirium, and if tients programmed for surgery. present, haloperidol was started. Acetaminophen (4 g/d) was started in all patients. The requirement for additional Study Design analgesia was assessed by attending ICU nurses. Morphine According to an open-label randomized controlled design, was given in boluses of 1 to 2 mg IV until patients were free patients were assigned to receive MV with either ASV-DE of pain. The boluses were repeated as needed. Postopera- or standard ASV. tive shivering, if present, was treated with meperidine (25 mg IV). Muscle relaxants were not given in the ICU. Inclusion Criteria Patients were included and randomized using sealed num- MV Protocols bered envelopes on ICU arrival after surgery. We created a All patients’ lungs were ventilated by a Hamilton Galileo homogeneous group of patients after elective and uncom- ventilator (software version GMP03.41f, GCP03.40a, plicated CABG, i.e., without a history of chronic obstructive GTP01.00; Hamilton Medical AG, Rha¨zu¨ ns, Switzerland). pulmonary disease or hemodynamic instability. Patients Passive humidification of the ventilatory circuit was ap- with a history of chronic obstructive pulmonary disease or plied by means of an HME filter (Medisize Hygrovent S; a history of pulmonary surgery, and patients with an Medisize, Hillegom, The Netherlands). intraaortic balloon pump or inotropics and/or vasopres- In both groups, the initial levels of fraction of inspired sors at a more than usual rate (in milligrams per hour: oxygen (Fio2) (50%) positive end-expiratory pressure dopamine 20, norepinephrine 0.5, dobutamine 25, or epi- (PEEP) (5 cm H2O), peak airway pressure (35 cm H2O), and nephrine [any rate]) on ICU arrival were excluded. %-minute ventilation (a theoretical value based on ideal body weight, 100%) were set by the attending ICU physi- Cardiothoracic Surgery/Anesthesia Procedures cian. Flow trigger sensitivity was set at 2 L/s; active All patients in both groups were anesthetized according to patients could trigger the ventilator (i.e., actual minute our standard institutional protocol, starting with 1 or 2 mg ventilation could exceed set %-minute ventilation). An lorazepam as premedication, followed by etomidate, sufen- arterial blood gas analysis was performed 30 minutes after tanil, and rocuronium for induction of anesthesia and connection to the ICU ventilator, and 30 minutes after each facilitation of intubation. During the surgical procedure, modification of ventilator settings (except for Fio2), it was sufentanil was used as analgesic, and sevoflurane plus advised to perform an additional arterial blood gas analy- propofol were used to maintain anesthesia. Muscle relax- sis. Fio2 could be adjusted to maintain arterial oxygen Ն ants were not given during the surgical procedure. Mor- saturation of 95%. phine and midazolam could be administered at the end of In both groups, patients were tracheally extubated after the procedure. achieving general tracheal extubation criteria (i.e., respon- Ͼ Cardiopulmonary bypass was performed under moder- sive and cooperative, urine output 0.5 mL/kg/h, chest Ͻ ate hypothermia (28°C–35°C), using a membrane oxygen- tube drainage 100 mL last hour, no uncontrolled arrhyth- Ͼ ator and nonpulsatile blood flow. At the end of anesthesia, mia, and having a core temperature 36.0°C and a respi- Ͼ all patients were transferred to the ICU with tracheal ratory frequency of 10 breaths per minute without intubation. Anesthesiologists and surgeons at the operating machine-controlled breaths for at least 30 minutes). T-piece room were blinded for inclusion or randomization of weaning was not used; patients were tracheally extubated patients. when they reached the above-described extubation criteria.

ICU Management ASV-DE Versus Standard ASV Unit policy comprised that a patient was cared for by a With ASV-DE, as soon as body temperature reached 35.0°C Ͼ dedicated ICU nurse, responsible for 1 or 2 patients. with pH 7.25, irrespective of arterial Pco2, %-minute Attending ICU nurses were constantly at the bedside, and ventilation was decreased stepwise by 10% (de-escalation) changes in treatment according to the postoperative ICU until 70% of the theoretical value based on ideal body protocol, based on observations by ICU nurses, were ex- weight, only if pH declined Ͻ7.25%-minute ventilation was ecuted immediately. increased again (escalation) (Fig. 1). Indeed, by neglecting

The postoperative ICU protocol was similar for both the Pco2 safety limits as defined for standard ASV, we study groups and involved fluid resuscitation with normal created a span for de-escalation with ASV-DE. saline and starch solutions, blood transfusion to maintain With standard ASV, %-minute ventilation was only Ն Ͻ Ͼ hemoglobin concentration ( 5.0 mmol/L), norepinephrine changed if Pco2 was 3.5 or 5.5 kPa. Indeed, settings in continuous infusion to achieve mean arterial blood with ASV were based on previously used safety limits to pressure Ն70 mm Hg, and dobutamine and/or enoximone guarantee sufficient minute ventilation at all times.3

962 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 1. Flow sheet as used by intensive care unit (ICU) nurses and ICU physicians [translated from Dutch]. ASV ϭ adaptive support Figure 2. Flow diagram showing the ﬂow of the patients through each ventilation; %-MV ϭ percentage minute ventilation; F-spont ϭ fre- stage of the clinical trial. CABG ϭ coronary artery bypass graft; quency of spontaneous breath; ABG ϭ arterial blood gas. COPD ϭ chronic obstructive pulmonary disease; ASV-DE ϭ adaptive support ventilation de-escalation escalation. Data Collection Collected data included the patient characteristics of gender, age, weight, and height, and the operation characteristics of 16.4 hours in the ASV group.3 A reduction of approxi- number of bypasses, cardiopulmonary bypass time (pump mately 10% was expected for the total duration of tracheal time), and aortic clamp time. Intraoperative and ICU sedative intubation. A sample size of 61 patients in each group was and analgesic prescriptions; time from admission to ICU until deemed to have 80% power to detect a difference in the reaching a central body temperature of 36.0°C; ventilation duration of MV of 10%, assuming a common standard characteristics, tidal volume size, PEEP, inspiratory pressure deviation of 3.3 hours, using a 2-sided t test with a 0.05 (defined as the maximum airway pressure minus the level of 2-sided significance level. PEEP) and respiratory rate, %-minute ventilation, and arterial blood gas data in time were collected. Respiratory data were Statistical Analysis collected by a data logger connected to the ventilator (Ham- Descriptive statistics were used to summarize patient char- ilton data logger, version 3.27.1; Hamilton Medical AG) and acteristics. Categorical variables were compared between from our patient Data Management System (IMDsoft, Sassen- groups by ␹2 tests. If normally distributed, continuous heim, The Netherlands). values were expressed as means Ϯ SD; otherwise, medians Outcome variables (see Definitions) were calculated for and interquartile ranges were used. All analyses were each patient. Primary end point was total duration of performed in SPSS version 16.0 (SPSS, Inc., Chicago, IL). tracheal intubation; secondary end points were summed single assisted ventilation episodes, time to first single RESULTS assisted ventilation episode, time to the assisted ventilation episode that was followed by tracheal extubation, and Patients length of stay in the ICU. We included 126 consecutive patients after elective and uncomplicated CABG: 63 patients were randomized to ASV-DE Deﬁnitions and 63 to standard ASV (Fig. 2). Of patients enrolled in the Total duration of tracheal intubation was defined as the study, 2 patients were lost for analysis of the secondary end period from ICU admission until tracheal extubation, and a points because of data logger failures: 1 patient randomized to single assisted ventilation episode was defined as an epi- ASV-DE and 1 randomized to standard ASV. sode of Ն20 minutes during which the patient was breathing at least 5 breaths per minute. Baseline, Perioperative, and ICU Characteristics Opiate doses were all recalculated as morphine equipo- Groups were well balanced (Table 1). Arterial blood gas tent doses with the following formula: 10 mg morphine ϭ analyses on ICU admission were not different (data not 0.1 mg fentanyl ϭ 0.01 mg sufentanil.7 Doses of benzodi- shown). Core temperature on ICU admission was not Ϯ Ϯ azepines were similarly converted to equipotent doses of different (35.5°C 1.1°C vs 35.7°C 0.6°C, ASV-DE versus ϭ diazepam using the following formula: 5 mg midazolam ϭ standard ASV; P 0.32). The number of patients with a 8 Ͼ 10 mg diazepam ϭ 50 mg oxazepam. temperature 36°C on ICU admission was also not different (27 [44%] vs 32 [51%], ASV-DE versus standard ASV; Sample Size P ϭ 0.24). There were no differences in time to rewarming The study was powered on total duration of tracheal to 36°C (2.1 Ϯ 3.0 vs 1.7 Ϯ 2.1 hours, ASV-DE versus intubation. Sample size assumptions were based on results standard ASV; P ϭ 0.64). ICU survival was 100% for the 2 of our previous study, i.e., a mean duration of ventilation of randomization groups.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 963 Adaptive Support Ventilation Protocolized with De-Escalation After Cardiothoracic Surgery

Table 1. Patient Characteristics, and Intraoperative and Intensive Care Characteristics ASV-DE Standard ASV Patient characteristics Patients, n 63 63 Male gender, n (%) 56 (89) 55 (87) Age (y), mean Ϯ SD 68 Ϯ 10 65 Ϯ 9 Actual body weight (kg), mean Ϯ SD 82 Ϯ 14 83 Ϯ 10 Ideal body weight (kg), mean Ϯ SD 69 Ϯ 870Ϯ 8 Set body weight (kg), mean Ϯ SD 69 Ϯ 870Ϯ 8 Height (cm), mean Ϯ SD 172 Ϯ 15 175 Ϯ 8 Intraoperative characteristics No. of bypasses, mean Ϯ SD 3 Ϯ 13Ϯ 1 ECC time (h), mean Ϯ SD 1.7 Ϯ 0.7 1.7 Ϯ 0.6 AOX time (h), mean Ϯ SD 1.1 Ϯ 0.5 1.1 Ϯ 0.5 Lowest core temperature (°C), mean Ϯ SD 34 Ϯ 1.2 34 Ϯ 1.3 Sufenta dose (␮g), n; median (IQR)a 63; 200 (135–250) 63; 200 (150–250) Midazolam dose (mg), n; median (IQR)a 50; 15 (5–25) 49; 15 (5–20) Morphine dose (mg), n; median (IQR)a 28; 20 (20–20) 26; 20 (20–20) Clonidine dose (␮g), n; median (IQR)a 6; 150 (150–225) 3; 150 (150–300) Intensive care unit characteristics APACHE II score 17 Ϯ 516Ϯ 5 Morphine dose (mg/kg) n; median (IQR)a 0.05 (0.03–0.11) 0.05 (0.02–0.07) Propofol dose (mg) ICU, n; median (IQR)a 63; 1022 (621–1963) 63; 1125 (500–2122) Sedation duration (h), median (IQR)a 3.6 (2.2–6.8) 4.6 (2.6–7.0) Length of stay ICU (h), median (IQR) 27 (21–49) 27 (22–40) ASV ϭ adaptive support ventilation; ASV-DE ϭ ASV de-escalation escalation; ECC time ϭ duration of extracorporeal circulation; AOX time ϭ duration of aortic cross-clamping; IQR ϭ interquartile range; APACHE ϭ Acute Physiology and Chronic Health Evaluation; ICU ϭ intensive care unit. a Only dose of patients who actually received medication is displayed. There were no signiﬁcant differences between the randomization groups.

Protocol Adherence In the ASV-DE group, mean %-minute ventilation at tracheal extubation was 92% Ϯ 13% (14% were tracheally extubated at %-minute ventilation level of 70%, 77% at a level between 70% and 100%, and 9% at a level Ͼ100%). In the standard ASV group, mean %-minute ventilation at the time of tracheal extubation was 103% Ϯ 10% (2% at a level Ͻ100%, 78% at a level of 100%, and 20% at a level Ͼ100%) (P Ͻ 0.05 versus ASV-DE). Sedation and analgesic use was not different between randomization groups (Table 1). There were no patients who fulfilled the criteria for delirium, and haloperidol was never started.

Duration of MV and Assisted Figure 3. Tracheally intubated patients (%) expressed as Kaplan- Ventilation Episodes Meier curve in the adaptive support ventilation (ASV) and in the ASV de-escalation escalation (ASV-DE) groups. Duration of tracheal intubation was not different between groups (10.8 [6.5–16.1] vs 10.7 [6.6–13.9] hours, ASV-DE ϭ versus standard ASV; P 0.32) (Fig. 3; Table 2). Neither was not clinically significant and with 70%-minute ventilation time from admission to the first assisted breathing period the number of patients would result in too few patients. (3.1 [2.0–6.7] vs 3.9 [2.1–7.5] hours; P ϭ 0.49) nor the number of assisted ventilation episodes (78 [34–176] vs 57 Ventilator and Ventilation Variables [32–116] episodes; P ϭ 0.20) was different. However, dura- Ventilator and ventilation variables are presented in Figure tion of assisted ventilation episodes that ended with tra- 4. There were no differences between groups regarding cheal extubation was longer with ASV-DE (2.5 [0.9–4.6] vs tidal volume, respiratory rate, arterial pH, Pco2, and Po2. ϭ 1.4 [0.3–3.5] hours; P 0.05). The highest levels of arterial Pco2 were similar in the 2 In a per-protocol analysis in which only patients who randomization groups (5.9 [range, 5.2–6.4] vs 5.8 [range, reached Յ80%-minute ventilation before tracheal extuba- 5.0–6.4] kPa, ASV-DE versus standard ASV; P ϭ 0.82). tion were compared with patients in the standard ASV group, there were also no significant differences in time DISCUSSION until extubation: 10.5 (8.0–16.0) vs 10.0 (6.0–13.0) hours In this study of postoperative weaning of patients after (n ϭ 16 vs n ϭ 63). The choice for 80%-minute ventilation was planned and uncomplicated nonfast-track CABG, we found arbitrary; however, we believed that 90%-minute ventilation that ASV with protocolized de-escalation and escalation

964 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 2. Respiratory Characteristics ASV-DE Standard ASV P value Duration of tracheal intubation (h), median (IQR) 10.8 (6.5–16.1) 10.7 (6.6–13.9) 0.32 Time (h) from admission to ﬁrst spontaneous breathing period, median (IQR) 3.1 (2.0–6.7) 3.9 (2.1–7.5) 0.49 No. of spontaneous breathing periods, median (IQR) 78 (34–176) 57 (32–116) 0.20 Spontaneous breathing/total time (%), median (IQR) 36 (21–61) 31 (21–50) 0.39 Controlled breathing/total time (%), median (IQR) 64 (39–79) 69 (50–79) 0.39 Last spontaneous breathing period before extubation (h), median (IQR) 2.5 (0.9–4.6) 1.5 (0.3–3.5) 0.045 Tidal volume (mL), mean Ϯ SD 573 Ϯ 105 580 Ϯ 105 0.62 Tidal volume (mL/kg ideal body weight), mean Ϯ SD 8.1 Ϯ 1.5 8.4 Ϯ 1.5 0.27 Respiratory rate during spontaneous breathing, mean Ϯ SD 12.4 Ϯ 1.4 12.7 Ϯ 1.3 0.22 Ϯ Ϯ Ϯ PEEP level (cm H2O), mean SD 5.3 1.2 5.3 1.0 0.60 ASV ϭ adaptive support ventilation; ASV-DE ϭ ASV de-escalation escalation; IQR ϭ interquartile range; PEEP ϭ positive end-expiratory pressure.

Figure 4. Graphical display of ventilator and ventilation variables over time. Minute ventilation in %, positive end- expiratory pressure, tidal volume (VT)in mL/kg ideal body weight (IBW), L/min, mean pH, and PCO2 and PO2 in kPa; black circles represent adaptive support ventilation (ASV) group; open circles represent ASV de-escalation escalation (ASV-DE) group. RR ϭ respiratory rate in breaths per minute; I ϭ after stabilization; II ϭ after 4 hours; III ϭ before extubation.

compared with standard ASV did not shorten duration of otherwise. Of note, improved patient-ventilator synchrony tracheal intubation. The time to the first assisted breathing with ASV could also lengthen duration of tracheal intubation. period was also not different between groups. There was, Indeed, this could be attributable to increased comfort, fewer however, a difference in the duration of the assisted alarms, and a gradual transmission from controlled to spon- breathing period ending with extubation. taneous ventilation thus not leading to apnea. Our study was MV could harm all patients, including those whose lungs not designed to test this hypothesis. are ventilated for only hours.9 Therefore, it is imperative to In contrast to our study, a significant reduction of time strive for shorter duration of MV and tracheal intubation at all until tracheal extubation with ASV as compared with times, including the weaning phase after surgery. In addition, synchronized intermittent mandatory ventilation/pressure controlled forms of MV could rapidly cause muscle atrophy of support was found in patients after fast-track cardiotho- 10 the diaphragm. To counteract this phenomenon, it is impor- racic surgery.5 In this trial by Sulzer et al., initially ASV was tant to allow patients to use their diaphragm as soon as set at 100%-minute ventilation (phase 1). When spontane- possible while still being mechanically ventilated. ASV allows ous breathing occurred, %-minute ventilation was reduced for automatic switches between controlled ventilation and by 50% (phase 2), and if necessary again by 50% (phase 3). assisted ventilation, depending on the patient’s activity. As This weaning approach can be described as much more such, ASV with protocolized de-escalation and escalation may aggressive with respect to de-escalation, compared with improve outcome because there was a trend to shorter time our study. We chose to de-escalate stepwise until %-minute until the first assisted breathing period, and more assisted ventilation was 70%, as suggested on the Web site of the ventilation episodes. Our study, however, was underpowered manufacturer.¶ Our stepwise approach and the minimum to show statistical difference regarding these secondary end level of %-minute ventilation may have been a flaw. points. Although we hypothesized that protocolized de-escalation ¶Hamilton Medical. Available at: http://www.hamilton-medical.com/. Ac- would prevent longer intubation times, our study showed cessed May 14, 2009.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 965 Adaptive Support Ventilation Protocolized with De-Escalation After Cardiothoracic Surgery

In a more recent study in patients after fast-track cardio- available in the unit. Notably, however, we noticed a signifi- thoracic surgery, time to extubation was also significantly cant difference between the study groups regarding %-minute shorter with ASV as compared with pressure-regulated vol- ventilation at the time of extubation and the period of assisted ume controlled with automode ventilation.4 In this trial by ventilation leading to tracheal extubation. Gruber et al., weaning consisted of 3 phases: controlled Apart from the fact that this was a single center study, ventilation (phase 1), assisted ventilation (phase 2), followed which limits the generalizability of our conclusions, there by a T-piece trial (phase 3) that ended with extubation. This are other limitations of the study. Sedation and analgesia trial did not use de-escalation. An important similarity be- requirements were not reported using specific scales. How- tween the trials by Sulzer et al. and Gruber et al., and in ever, we did not gradually wean patients off of sedation, contrast to our trial, is that the former trials were both but stopped infusion of sedation completely when the core performed in fast-track cardiothoracic surgery patients, temperature was Ͼ35°C. Of note, sedation and analgesic whereas we explicitly included nonfast-track patients. requirements were the same in the 2 study groups. Another There were no differences in arterial blood gas variables, limitation is that we excluded patients with chronic ob- including arterial pH and Pco2, which could be explained structive pulmonary disease, also limiting generalizability. by the fact that patients in the ASV-DE group could adjust Compared with a previous study of ASV by our group, their minute ventilation when operator set %-minute ven- overall weaning time in the present study was considerably tilation was decreased (resulting in higher actual minute shorter. Indeed, median time to tracheal extubation was ventilation than the operator set %-minute ventilation). 16.4 (12.5–20.8) hours in the previous study.3 The question Indeed, time until the first assisted breathing period was must be raised whether tracheal extubation of cardiothoracic shorter, and more assisted ventilation episodes were found surgery patients is dependent on the ventilatory strategy in the ASV-DE group. It is also important to note that the alone, or (also) on factors independent of the ventilation 4 highest levels of arterial Pco2 were not different between strategy. The above-mentioned studies by Gruber et al. and the 2 randomization groups, indicating that ASV-DE is at Sulzer et al.5 certainly show that the ventilator strategy least as safe as standard ASV. influences weaning time in these patients. One ventilatory There are multiple reasons why we were unable to show strategy factor that could have influenced weaning time in our a difference between standard ASV and ASV-DE, including studies was the use of different levels of PEEP. Specifically, in standard of care in our setting and type of patients studied. our previous study of ASV, patients received 10 cm H2O In a study comparing SmartCare® (a knowledge-based PEEP in the first 4 hours after arrival in the ICU, and weaning tool including an automatic gradual reduction of thereafter 5 cm H2O PEEP until tracheal extubation. In the pressure support, automatic performance of spontaneous present study, patients received 5 cm H2O PEEP throughout breathing trials, and generation of an incentive message the complete weaning phase. This extra step in the weaning when a breathing trial was successfully passed) with con- process could have accelerated weaning in the present study. ventional weaning, Rose et al.11 found no differences In addition, this change in practice could have resulted in a regarding duration of MV. This finding was in sharp change in use of sedatives because we continued sedatives for contrast to the results from a study by Lellouche et al.12 at least the first 4 hours, or as long as the higher level of PEEP showing SmartCare to significantly reduce duration of MV. was used, in the first study. This usually took longer than the One important difference, however, between the control time needed to reach a core temperature Ͼ35°C, which was groups of the 2 studies was that duration of MV was the time to stop sedatives in the present study. Factors shorter in the study by Rose et al. The shorter duration of independent of ventilation strategy could also have a role. MV could have masked any beneficial effect of the interven- Both the surgical/anesthesiological team and the ICU team tion in the first study, whereas it allowed for an important gained experience over time, whereas the local guidelines effect of the intervention in the second study. Differences in (apart from the advice on PEEP) of these teams as well as their duration of MV could have resulted from differences in composition did not change. Better understanding of the patient case mix and differences in standard care surrounding needs of patients after cardiothoracic surgery could have led the studied patient populations in these 2 studies. We may to the use of less sedatives both intra- and postoperatively have encountered a similar problem: in our study, duration of despite the fact that no formal protocol changes were tracheal intubation was rather long compared with other trials implemented.13 Also, more experience with ASV in this of ASV. This may very well relate to the fact that we included particular patient group could have led to more confidence nonfast-track patients instead of most other studies of wean- in earlier tracheal extubation in our department. Finally, ing of cardiothoracic surgery patients. The rather long dura- better awareness of long weaning times in our institution tion of MV may have precluded any effect of ASV-DE over could have led to a more proactive behavior with regard to standard ASV in our study. tracheal extubation.14 Although we performed a random- Because this was an open-label, i.e., not blinded, ran- ized controlled trial, and as such all these factors should not domized clinical trial, we could not exclude the possibility have affected the primary outcome differently in the 2 that patients randomized to the standard ASV group also study arms, one could certainly suggest that other factors benefited from early de-escalation. This could have mini- than the ventilatory strategy have key roles in time until mized contrast between the study groups. To promote tracheal extubation in these patients. protocol adherence, nurses and physicians were able to In conclusion, compared with standard ASV, weaning of assess only 1 of the 2 flowcharts (as presented in Fig. 1), for patients after nonfast-track CABG using ASV with proto- ASV-DE or standard ASV, depending on randomization colized de-escalation and escalation does not shorten time group. We could not always prevent both flowcharts being to tracheal extubation.

966 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA AUTHOR CONTRIBUTIONS 5. Sulzer CF, Chiolero R, Chassot PG, Mueller XM, Revelly JP. DAD helped design and conduct the study, analyze the data, Adaptive support ventilation for fast tracheal extubation after and write the manuscript. This author has seen the original cardiac surgery: a randomized controlled study. Anesthesiol- ogy 2001;95:1339–45 study data, reviewed the analysis of the data, approved the 6. Santiago TV, Edelman NH. Opioids and breathing. J Appl final manuscript, and is the author responsible for archiving Physiol 1985;59:1675–85 the study files. DPV helped conduct the study, analyze the 7. Reisine T, Pasternak GW. Opioid analgesics and antagonists. In data, and write the manuscript. This author has seen the Hardman JG, Limbird LE, eds. Goodman & Gilman’s The original study data, reviewed the analysis of the data, and Pharmacological Basis of Therapeutics. 9th ed. New York: approved the final manuscript. JMB helped design the study, McGraw-Hill, 1996:521–55 8. Wilson WC, Smedira NG, Fink C, McDowell JA, Luce JM. analyze the data, and write the manuscript. This author has Ordering and administration of sedatives and analgesics dur- seen the original study data, reviewed the analysis of the data, ing the withholding and withdrawal of life support from and approved the final manuscript. BAJMdM helped design critically ill patients. JAMA 1992;267:949–53 the study. This author has seen the original study data and 9. Schultz MJ, Haitsma JJ, Slutsky AS, Gajic O. What tidal approved the final manuscript. AK helped conduct the study volumes should be used in patients without acute lung injury? and analyze the data. This author has seen the original study Anesthesiology 2007;106:1226–3 data, reviewed the analysis of the data, and approved the final 10. Levine S, Nguyen T, Taylor N, Friscia ME, Budak MT, Rothen- berg P, Zhu J, Sachdeva R, Sonnad S, Kaiser LR, Rubinstein manuscript. FP helped conduct the study. This author has seen NA, Powers SK, Shrager JB. Rapid disuse atrophy of dia- the original study data and approved the final manuscript. MJS phragm fibers in mechanically ventilated humans. N Engl helped design the study, analyze the data, and write the J Med 2008;358:1327–35 manuscript. This author has seen the original study data, 11. Rose L, Presneill JJ, Johnston L, Cade JF. A randomised, reviewed the analysis of the data, and approved the final controlled trial of conventional versus automated weaning manuscript. from mechanical ventilation using SmartCare/PS. Intensive Care Med 2008;34:1788–95 12. Lellouche F, Mancebo J, Jolliet P, Roeseler J, Shortgen F, Dojat REFERENCES M, Cabello B, Bouadma L, Rodriguez P, Maggiore S, Reynaert 1. Brunner JX, Iotti GA. Adaptive support ventilation (ASV). M, Mersmann S, Brochard L. A multicenter randomized trial of Minerva Anestesiol 2002;68:365–8 computer-driven protocolized weaning from mechanical ven- 2. Arnal JM, Wysocki M, Nafati C, Donati S, Graniet I, Corno G, tilation. Am J Respir Crit Care Med 2006;174:894–900 Durrand-Gasselin J. Automatic selection of breathing pattern 13. De Hert SG, Van der Linden PJ, Cromheecke S, Meeus R, ten using adaptive support ventilation. Intensive Care Med Broecke PW, De Blier IG, Stockman BA, Rodrigus IE. Anesthe- 2008;34:75–81 siology 2004;101:9–20 3. Dongelmans DA, Veelo DP, Paulus F, de Mol BA, Korevaar JC, 14. Hawkes CA, Dhileepan S, Foxcroft D. Early extubation for Kudoga A, Middelhoek P, Binnekade JM, Schultz MJ. Weaning adult cardiac surgical patients. Cochrane Database Syst Rev automation with adaptive support ventilation: a randomized 2003;4:CD003587 controlled trial in cardiothoracic surgery patients. Anesth Analg 2009;108:565–71 4. Gruber PC, Gomersall CD, Leung P, Joynt GM, Ng SK, Ho KM, Underwood MJ. Randomized controlled trial comparing adaptive-support ventilation with pressure-regulated volume- controlled ventilation with automode in weaning patients after cardiac surgery. Anesthesiology 2008;109:81–7

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 967 Lung Recruitment and Positive End-Expiratory

Pressure Have Different Effects on CO2 Elimination in Healthy and Sick Lungs

Gerardo Tusman, MD,* Stephan H. Bohm, MD,† Fernando Suarez-Sipmann, PhD,‡ Adriana Scandurra, Eng,§ and Go¨ran Hedenstierna, PhDʈ

BACKGROUND: We studied the effects that the lung recruitment maneuver (RM) and positive

end-expiratory pressure (PEEP) have on the elimination of CO2 per breath (VTCO2,br). METHODS: In 7 healthy and 7 lung-lavaged pigs at constant ventilation, PEEP was increased

from0to18cmH2O and then decreased to 0 in steps of 6 cm H2O every 10 minutes. Cycling RMs with plateau pressure/PEEP of 40/20 (healthy) and 50/25 (lavaged) cm H2O were applied for 2 minutes between 18-PEEP steps. Volumetric capnography, respiratory mechanics, blood gas, and hemodynamic data were recorded.

RESULTS: In healthy lungs before the RM, VTCO2,br was inversely proportional to PEEP decreasing from 4.0 (3.6–4.4) mL (median and interquartile range) at 0-PEEP to 3.1 (2.8–3.4) mL at Ͻ 18-PEEP (P 0.05). After the RM, VTCO2,br increased from 3.3 (3–3.6) mL at 18-PEEP to 4.0 Ͻ (3.5–4.5) mL at 0-PEEP (P 0.05). In lavaged lungs before the RM, VTCO2,br increased initially from 2.0 (1.7–2.3) mL at 0-PEEP to 2.6 (2.2–3) mL at 12-PEEP (P Ͻ 0.05) but then decreased Ͻ to 2.4 (2–2.8) mL when PEEP was increased further to 18 cm H2O(P 0.05). After the RM, the highest VTCO2,br of 2.9 (2.1–3.7) mL was observed at 12-PEEP and then decreased to 2.5 Ͻ (1.9–3.1) mL at 0-PEEP (P 0.05). VTCO2,br was directly related to changes in lung perfusion, the area of gas exchange, and alveolar ventilation but inversely related to changes in dead space.

CONCLUSIONS: CO2 elimination by the lungs was dependent on PEEP and recruitment and showed major differences between healthy and lavaged lungs. (Anesth Analg 2010;111:968–77)

he effect of positive end-expiratory pressure (PEEP) We have observed in patients that after a lung recruitment

on CO2 kinetics has been described. PEEP, at con- maneuver (RM), a ventilatory intervention aimed at restoring stant ventilation and body metabolism, is related to a pulmonary aeration, CO2 elimination increased despite the T 18–20 decrement in the elimination of CO2 by the lungs because use of high PEEP levels and low Vt values. These results of several factors: (1) a decrement in CO2 transport to the contradict the classical understanding of the effects that PEEP 3–5 lungs by a decrease in venous return and thus cardiac seems to have on CO2 kinetics. To our knowledge, a 1–3 output (CO), (2) a transient decrease in expired tidal systematic analysis of the elimination of CO2 during changes volume (Vt) caused by the sequential accumulation of air in PEEP combined with a lung RM has not been performed. within the lungs right after the increase in PEEP,3,4 (3) a These ventilatory interventions are likely to show different gain in functional residual capacity (FRC) leading to a responses in healthy and sick lungs because of the patho- physiology of acute lung injury resulting from massive lung filling of the lungs with inspiratory gases free of CO2 3–5 collapse, lung edema, and tissue inflammation. thereby inducing a transient dilution of alveolar CO2, and (4) an increase in airway and alveolar dead space Therefore, the aim of this study was to describe such causing a decrease in alveolar ventilation (V˙ a).6,7 changes in the elimination of CO2 in animals with healthy and The collapse of healthy anesthetized and acutely injured surfactant-depleted lungs and to determine whether lung lungs of patients is well described and the main ventilatory recruitment and PEEP induced retention of CO2 within the treatment of such collapse conditions is built upon blood. PEEP.8–11 PEEP and low Vt ventilation are parts of a lung-protective ventilatory strategy that aims to minimize METHODS the injury caused by tidal recruitment and overdisten- After approval by the Animal Research Committee of Upp- sion.12,13 However, because PEEP has been related to the 14–17 sala University in Sweden, 14 Swedish mixed country breed retention of CO2 within the body, this protective pigs (body weight ϭ 24.5 Ϯ 3 kg) were anesthetized with IV ventilatory strategy could lead to hypercapnia, especially in ␮ 12 ketamine 25 to 50 mg/kg/h, midazolam 90 to 180 g/kg/h, the context of low Vt ventilation. fentanyl 3 to 6 ␮g/kg/h, and pancuronium 0.25 to 0.50 mg/kg/h. The trachea was intubated with a 7-mm inner Author affiliations are provided at the end of the article. diameter cuffed endotracheal tube and air leaks were identi- Accepted for publication June 22, 2010. fied by incomplete flow-volume loops or changes in the Disclosure: The authors report no conflicts of interest. capnograms. The lungs were ventilated with a SERVO-i Reprints will not be available from the author. (Maquet Critical Care, Wayne, NJ) using a volume control Address correspondence to Gerardo Tusman, MD, Department of Anesthe- siology, Hospital Privado de Comunidad, Mar del Plata, Argentina. Address mode with a Vt of 6 mL/kg, respiratory rate of 30 e-mail to [email protected]. breaths/min, I:E ratio of 1:2, a fraction of inspired oxygen of 1, Copyright © 2010 International Anesthesia Research Society and initially without PEEP. Intravenous saline solution was DOI: 10.1213/ANE.0b013e3181f0c2da maintained at a fixed rate of 4 mL/kg during the study. Body

968 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 20 temperature was maintained by a warm blanket to keep rectal exchange with a better diffusion of CO2 and vice versa. Ϯ temperature within a range of 37.5°C 0.5°C. Pv-aco2 is the gradient between venous and arterial CO2. Shunt was calculated using a standard formula.27 CO2 Data Volumetric capnography was recorded on-line using the Hemodynamic Data NICO capnograph (Respironics, Wallingford, CT). The air- Electrocardiogram and pulse oxymetry were recorded, and a way flow and CO2 mainstream sensor were placed at the “Y” catheter for mean arterial blood pressure measurement was piece of the ventilator circuit and delivered data into a placed in the femoral artery. A 7.5F pulmonary artery catheter custom-made MatLab program (MathWorks, Natick, MA), CCOmbo (Edwards Lifesciences, Irvine, CA) placed into the which constructed volumetric capnograms by a Levenberg- right jugular vein provided continuous CO and pulmonary Marquardt fitting method, and all capnography-derived pa- pressures. CO was then subdivided into (1) an ineffective rameters were calculated from this mathematical function.21 shunting part (COSHUNT), calculated as the product between The Vtco2,br is the amount of CO2 eliminated in 1 breath shunt and CO to get the absolute value in L/min, and (2) an obtained by integrating expired airway flow and Pco signals. 2 effective pulmonary perfusion part (COEPP) or the portion of Petco is the partial pressure of CO at the end of 2 2 the CO that participates in CO2 exchange calculated by expiration and Peco is the mixed partial pressure of CO 2 2 subtracting the COSHUNT value from CO. This last parameter in 1 breath. Airway dead space was calculated as the was used to represent the effect of pulmonary blood flow on inflection point of the capnogram, which defines the Vtco2,br during the protocol. airway-alveolar interface or the limit between Vdaw and 21 the alveolar Vt (Vtalv). Protocol Dead space to Vt ratio (Vd/Vt) is an index that deter- Animals were randomly assigned to 2 groups: healthy (n ϭ mines the global inefficiency of ventilation. It was calcu- 7) or surfactant-depleted lungs (n ϭ 7). Lung lavages with 22 lated using the Bohr-Enghoff formula : 35 mL/kg warm isotonic saline solution were performed in the lavaged lung group.28 Lavages were repeated every 5 ؊ ؍ Vd/Vt (Paco2 Peco2)/Paco2 minutes until Pao2 stabilized between 100 to 150 mm Hg at pure oxygen and PEEP of 8 cm H2O. The mean value of 10 slopes of phase III (SIII) was 21 The protocol consisted of 3 sequential parts: calculated as previously described. SIII is a qualitative and noninvasive marker of the global ventilation-perfusion 1. An increasing PEEP limb, where PEEP was increased ˙ ˙ ˙ ˙ (V/Q) ratio, where low values are related to normal V/Q in steps of 6 cm H2Ofrom0to18cmH2O using a ratios whereas high values are indicators of an increased volume control mode of ventilation. These data rep- 23–26 dispersion of V˙ /Q˙ . resent the isolated effect that PEEP would have on

the elimination of CO2. Respiratory Mechanics Data 2. An RM, which consisted of a 2-minute cycling RM in Respiratory mechanics data were recorded using the flow pressure control ventilation using 40/20 cm H2Oin and pressure sensors of the same NICO capnograph. V˙ a is 29 30 healthy lungs or 50/25 cm H2O in sick lungs for the effective portion of ventilation and is defined as the plateau pressure and PEEP, respectively. product of Vtalv and respiratory rate. 3. A decreasing PEEP limb, as a mere mirror image of part Respiratory system dynamic compliance (Crs) was cal- 1 of the protocol where PEEP was decreased from 18 to ⌬ ⌬ ⌬ culated as Vt/ Paw. Changes in FRC ( FRC) induced by 0 PEEP in steps of 6 cm H O. These data represent the 3 2 PEEP were calculated by the following formula : cumulative effect that PEEP in combination with a prior

iϭn RM would have on CO2 elimination. (͸Vt(0) ؊ Vt(i ؍ FRC⌬ iϭ1 Each level of PEEP in parts 1 and 3 was maintained for 10 3–5 where Vt(0) was the reference exhaled Vt and n was the minutes because previous publications and our own number of breaths with decreased Vt [Vt(i)] after a PEEP results from pilot studies showed that Ͼ90% of all changes increase or increased Vt [Vt(i)] after a PEEP decrease, in Vtco2,br caused by PEEP occurred within this timeframe. respectively. Data Analysis Gas Exchange Data Before starting the protocol, in vitro and in vivo calibrations of Arterial blood gases were monitored on-line using the devices were performed following the manufacturer’s guides. multiparameter intraarterial sensor TrendCare (Diametrics Hemodynamic and on-line blood gas data were stored in a Medical Ltd., High Newcombe, UK) inserted into the right laptop by an acquisition system programmed in LabView carotid artery. Independent arterial and mixed venous (National Instruments, Austin, TX) while CO2 and respiratory blood gas samples were extracted at each protocol step and data were recorded in another laptop using the dedicated analyzed within 5 minutes using an ABL 300 (Radiometer, software Aplus (Respironics, Wallingford, CT). Both laptops

Copenhagen, Denmark). were synchronized in time. CO2, respiratory mechanics, and Pao2,Paco2, and Pvco2 are the Po2 and CO2 in the hemodynamic data belonging to the last minute of each PEEP ϭ arterial and venous blood, respectively. The Pa-etco2 is the step (30 breaths 30 data points), including the blood gas difference between arterial and end-tidal partial pressure of samples taken at this time, were analyzed.

CO2 representing the area for eliminating CO2 through the A descriptive statistical analysis was performed using the lungs where a low difference corresponds to a large area for MatLab program. Lilliefors test determined a non-Gaussian

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 969 CO2 Elimination by Lung Recruitment and PEEP

Table 1. Ventilation-Related Data During the Protocol in Healthy Animals

PEEP (cm H2O) Increasing limb of PEEP 0 6 12 18 VT (mL) 196 (184–208) 197 (187–207) 192 (180–204) 188 (165–211) V˙ A (L) 3.5 (3.1–3.9) 3.3 (2.8–3.8) 3.0 (2.6–3.4) 2.7 (2.3–3.1)

Pplat (cm H2O) 12 (10–14) 18 (17–19) 26 (25–27) 33 (31–35) ⌬FRC (mL) 96 (55–137) 279 (213–342) 570 (497–643)

Crs (mL/cm H2O) 17 (14–20) 16 (14–18) 15 (14–16) 15 (14–16) VD/VT 0.43 (0.41–0.45) 0.49 (0.45–54) 0.48 (0.44–0.52) 0.52 (0.48–56)

SIII (mm Hg/mL) 0.022 (0.012–0.032) 0.018 (0.008–0.028) 0.023 (0.001–0.044) 0.036 (0.010–0.062) pH 7.44 (0.43–0.45) 7.43 (0.38–0.48) 7.43 (0.37–49) 7.40 (0.33–0.47)

PaO2 (mm Hg) 496 (424–568) 483 (421–544) 514 (483–545) 552 (536–568) Pa–ETCO2 (mm Hg) 3 (2–4) 2 (1–3) 1 (0–2) 1 (0–2) ៮ Pv–aCO2 (mm Hg) 9 (5–13) 9 (6–12) 12 (9–15) 13 (5–21) Values are presented in median and interquartile range. Comparison of the same level of positive end-expiratory pressure (PEEP) before (increasing limb) and after (decreasing limb) a lung recruitment maneuver. ϭ ˙ ϭ ϭ ⌬ ϭ ϭ VT expired tidal volume; VA alveolar ventilation; Pplat plateau pressure; FRC change in functional residual capacity; Crs dynamic respiratory compliance; ϭ ϭ ϭ ϭ VD/VT ratio between physiological dead space and tidal volume; SIII slope of phase III; PaO2 arterial partial pressure of oxygen; Pa-ETCO2 difference between ៮ ϭ arterial and end-tidal partial pressure of carbon dioxide; Pv-aCO2 difference between mixed venous and arterial partial pressure of carbon dioxide. * P Ͻ 0.05.

Table 2. Ventilation-Related Data During the Protocol in Surfactant-Depleted Animals

PEEP (cm H2O) Increasing limb of PEEP 0 6 12 18 VT (mL) 176 (152–200) 177 (161–183) 175 (150–190) 172 (158–186) V˙ A (L) 3.0 (2.4–3.6) 2.6 (2.1–3.1) 2.5 (2–3) 2.3 (1.8–2.8)

Pplat (cm H2O) 27 (25–29) 26 (23–29) 27 (26–28) 32 (30–34) ⌬FRC (mL) 92 (76–108) 246 (206–286) 457 (406–508)

Crs (mL/cm H2O) 7 (6–8) 11 (9–11) 16 (14–18) 15 (12–18) VD/VT 0.73 (0.62–0.84) 0.64 (0.54–0.74) 0.58 (0.44–0.72) 0.58 (0.53–0.63)

SIII (mm Hg/mL) 0.090 (0.070–0.110) 0.072 (0.047–0.097) 0.039 (0.013–0.065) 0.067 (0.027–0.107) pH 7.29 (0.26–0.32) 7.28 (0.24–0.32) 7.34 (0.30–0.38) 7.36 (0.31–0.41)

PaO2 (mm Hg) 40 (18–62) 93 (32–154) 240 (155–325) 494 (421–565) Pa-ETCO2 (mm Hg) 33 (25–41) 13 (7–19) 6 (2–10) 4 (2–6) ៮ Pv-aCO2 (mm Hg) 7 (4–10) 8 (6–10) 10 (7–13) 12 (10–14) Values are presented in median and interquartile range. Comparison of the same level of positive end-expiratory pressure (PEEP) before (increasing limb) and after (decreasing limb) a lung recruitment maneuver. ϭ ˙ ϭ ϭ ⌬ ϭ ϭ VT expired tidal volume; VA alveolar ventilation; Pplat plateau pressure; FRC change in functional residual capacity; Crs dynamic respiratory compliance; ϭ ϭ ϭ ϭ VD/VT ratio between physiological dead space and tidal volume; SIII slope of phase III; PaO2 arterial partial pressure of oxygen; Pa-ETCO2 difference between ៮ ϭ arterial and end-tidal partial pressure of carbon dioxide; Pv-aCO2 difference between mixed venous and arterial partial pressure of carbon dioxide. * P Ͻ 0.05.

distribution of the data. Friedman nonparametric test was that the relative changes in the elimination of CO2 with used to compare the results of the same level of PEEP before increasing PEEP levels were mainly related to a decrease in Ͻ ˙ with those after RM in a 2-way direction. The same test was the efficacy of ventilation ( Va) and the decrease in COEPP. used to compare differences between results of consecutive The area for eliminating CO2 (Pa-etco2) showed a small levels of PEEP. Values are expressed as median (interquartile increase with increasing PEEP levels but with little effect on Ͻ range) and P values 0.05 were considered significant. CO2 elimination. Dead space (Vd/Vt) and SIII increased proportionally to PEEP (Table 1).

RESULTS After recruitment, Vtco2,br increased from 3.3 (3–3.6) All pigs completed the protocol successfully. Absolute values mL (18-PEEP) to 4.0 (3.5–4.5) mL (0-PEEP, P Ͻ 0.05) as

of the main variables belonging to the last minute for each PEEP was reduced. This increased CO2 elimination was ˙ PEEP step are presented in Tables 1 to 3. In general, PEEP associated with an increase in Va and COEPP. Initially, applied after lung recruitment improved lung function when Pa-etco2 decreased when going from 18 to 12 cm H2Oof compared with PEEP alone in both healthy and lavaged PEEP, but progressively increased again when going fur- lungs. The recruitment effect was characterized by a gain in ther down to 0-PEEP (Table 1). ⌬ FRC, an increase in Crs, and decreases in Vd/Vt and shunt, Vtco2,br presented a different behavior in lavaged ani- paralleled by improvements in gas exchange. mals (Figs. 1 and 2). Before recruitment, Vtco2,br initially In healthy pigs before recruitment, Vtco2,br decreased increased from 2.0 (1.7–2.3) mL (0-PEEP) to 2.6 (2.2–3) mL Ͻ from 4.0 (3.6–4.4) mL (0-PEEP) to 3.1 (2.8–3.4) mL (18- (12-PEEP) (P 0.05). This increment in CO2 elimination Ͻ PEEP, P 0.05) as PEEP increased (Fig. 1). Figure 2 shows went along with an increased COEPP and a decreased

970 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. (Continued)

PEEP (cm H2O) Decreasing limb of PEEP 18 12 6 0 191 (169–213) 195 (175–215) 200 (178–222) 201 (179–223) 2.9 (2.5–3.3) 3.0 (2.5–3.5) 3.6 (3.2–4.0)* 3.9 (3.4–4.4)* 29 (27–31) 19 (18–20)* 13 (12–14)* 9.6 (7–12) Ϫ628 (Ϫ597 to 659)* Ϫ401 (Ϫ370 to 432)* Ϫ406 (Ϫ371 to 441)* Ϫ181 (Ϫ151 to 211) 18 (16–20) 30 (27–33)* 31 (28–34)* 21 (18–24)* 0.51 (0.48–0.54) 0.44 (0.39–49)* 0.41 (0.38–0.44)* 0.42 (0.39–45) 0.031 (0.006–0.056)* 0.033 (0.023–0.043)* 0.014 (0.009–0.019)* 0.023 (0.010–0.036) 7.39 (0.32–0.46) 7.42 (0.37–0.47) 7.43 (0.39–47) 7.42 (0.38–0.47) 591 (570–612)* 576 (553–599)* 556 (540–572)* 476 (424–528) 1 (0–2) 0.6 (0.4–1)* 1 (0.7–2.1)* 5 (3–7)* 13 (8–17) 15 (7–22)* 12 (8–16) 11 (8–14)*

Table 2. (Continued)

PEEP (cm H2O) Decreasing limb of PEEP 18 12 6 0 171 (158–184) 172 (149–185) 174 (150–198) 177 (153–201) 2.4 (1.4–3.4) 2.8 (2.2–3.4)* 2.8 (2.3–3.3)* 2.9 (2.3–3.5) 29 (27–31)* 22 (19–25)* 20 (18–22)* 22 (19–25)* Ϫ607 (Ϫ541 to 673)* Ϫ377 (Ϫ354 to 400)* Ϫ384 (Ϫ353 to 415)* Ϫ185 (Ϫ165 to 205) 19 (15–23)* 23 (17–29)* 15 (12–18)* 9 (8–10) 0.56 (0.50–0.62) 0.56 (0.48–0.64) 0.59 (0.56–0.62)* 0.63 (0.57–0.69)* 0.035 (0.010–0.069)* 0.024 (0.004–0.048)* 0.031 (0.011–0.051)* 0.050 (0.028–0.072)* 7.40 (0.32–0.48)* 7.40 (0.33–0.47)* 7.34 (0.23–45)* 7.25 (0.20–0.30) 527 (501–553)* 399 (354–444)* 179 (90–268)* 80 (50–110)* 2 (1–3)* 6 (3–9) 9 (3–15)* 17 (13–21)* 14 (12–16) 10 (9–11) 9 (8–10) 7 (3–11)

Pa-etco2.Vtco2,br then decreased from 2.6 (2.2–3) mL Figure 3B shows the effect of a PEEP change from 6- to (12-PEEP) to 2.4 (2–2.8) mL (18-PEEP) (P Ͻ 0.05). This time, 12-PEEP before recruitment on the main variables for 18 the impairment of CO2 elimination was associated with a consecutive respiratory cycles. The effects were qualita- ˙ reduced Va and COEPP despite Pa-etco2 showing the tively similar, but quantitatively different in healthy and lowest value of the increasing limb of PEEP. in lavaged lungs. Vtco2,br decreased almost to zero in After recruitment, the highest Vtco2,br was observed at the first breath but recovered within 5 to 6 consecutive Ͻ 12-PEEP (2.9 [2.1–3.7] mL, P 0.05) which decreased to 2.5 breaths in both healthy and lavaged lungs. This change Ͻ (1.9–3.1) mL at 0-PEEP (P 0.05). The progressive decre- in Vtco was related to a decrease in expired Vt right ments in the area for CO exchange and in CO were 2,br 2 EPP after the PEEP change resulting in an expansion of ⌬FRC associated with a lower elimination of CO at 0-PEEP. Vd/Vt 2 by approximately 183 mL in healthy and 154 mL in and SIII increased with reductions in PEEP (Table 2). lavaged lungs. As opposed to Vtco2,br, CO was little Figure 3A represents the elimination of CO2 over time for affected by the change in PEEP. a single step change of PEEP from 6 to 12 cm H2O before recruitment. Compared with 6-PEEP, the change in median Pvco2,Paco2, and Petco2 are presented in Figure 1 and Ϫ ϩ Tables 1 and 2. In healthy lungs, no clinically significant Vtco2,br was 6% in healthy lungs and 8% in lavaged lungs at 12-PEEP (both P Ͻ 0.05). In both healthy and surfactant- changes in those variables were observed. In surfactant- depleted animals, Vtco2,br decreased in the first breaths after depleted lungs, however, the difference between Paco2 and Ͼ the PEEP increase with 90% of the effect occurring within 5 Petco2 narrowed significantly but in an inverse relation to minutes at the higher PEEP. the applied PEEP before and after lung recruitment with

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 971 CO2 Elimination by Lung Recruitment and PEEP

Table 3. Hemodynamic Data in Healthy and Surfactant-Depleted Animals

PEEP (cm H2O) Increasing limb of PEEP 0 6 12 18 Healthy lungs HR (bpm) 95 (68–122) 86 (48–124) 94 (66–122) 90 (68–112) MAP (mm Hg) 74 (47–101) 82 (61–103) 69 (49–89) 65 (44–86) MPAP (mm Hg) 21 (15–27) 25 (17–33) 25 (19–31) 28 (21–35) CVP (mm Hg) 7 (5–9) 8 (6–10) 9 (5–13) 11 (7–15) CO (L/min) 2.60 (2–3.2) 2.60 (2.1–3.1) 2.10 (1.6–2.6) 2.10 (1.4–2.8)

COEPP (L/min) 2.03 (1.1–3) 2.42 (2–2.8) 1.93 (1.3–2.5) 1.93 (1.2–2.6) COSHUNT (L/min) 0.31 (0.1–0.5) 0.29 (0.1–0.4) 0.17 (0.1–0.3) 0.09 (0.05–0.15) Lavaged lungs HR (bpm) 106 (75–137) 104 (71–135) 101 (63–139) 114 (85–143) MAP (mm Hg) 84 (62–106) 95 (70–120) 85 (65–105) 80 (55–105) MPAP (mm Hg) 45 (24–66) 38 (29–47) 34 (26–42) 35 (28–42) CVP (mm Hg) 9 (6–12) 9 (6–12) 9 (5–13) 11 (8–14) CO (L/min) 5.50 (4.3–6.7) 5.40 (4.2–6.6) 4.31 (3.5–5.1) 4.30 (3.4–5.1)

COEPP (L/min) 3.03 (2–4) 3.08 (2.3–3.8) 3.77 (3.3–4.3) 3.50 (2.9–4.1) COSHUNT (L/min) 2.48 (2–3) 1.70 (1.2–2.2) 0.84 (0.5–1.1) 0.30 (0.1–0.5) Values are presented in median and interquartile range. Comparison of the same level of positive end-expiratory pressure (PEEP) before (increasing limb) and after (decreasing limb) a lung recruitment maneuver. HR ϭ heart rate; MAP ϭ mean systemic arterial pressure; MPAP ϭ mean pulmonary artery pressure; CVP ϭ central venous pressure; CO ϭ cardiac output; ϭ ϭ COEPP effective part of CO; COSHUNT ineffective part of CO. * P Ͻ 0.05.

corresponding changes in pH. Both Pvco2 and Paco2 decreased with lung recruitment and PEEP.

DISCUSSION The main findings of this study can be summarized as follows: 1. Lung recruitment and PEEP have different effects on

CO2 elimination in healthy and surfactant-depleted lungs. 2. At constant metabolism and ventilatory settings, any

change in the elimination of CO2 can be explained by a combination of changes in (a) the effectiveness of

lung perfusion, (b) the area for CO2 exchange, and (c) the amount of V˙ a and, thus, in the global V˙ /Q˙ relationship of the lungs. 3. In healthy and lavaged lungs, changes in PEEP

altered the elimination of CO2 because of immediate effects on both expired Vt values and ⌬FRC for 5 to

6 consecutive breaths. Stable new values for Vtco2,br, ˙ lung perfusion, area of CO2 exchange, and Va were reached within 5 minutes.

4. In lavaged lungs, the efficacy of CO2 elimination was directly related to the recruitment/derecruitment ef-

fect. Lung recruitment and PEEP did not retain CO2 in the blood during the study periods.

Effect of PEEP and Lung Recruitment on CO2 Elimination Breen and Mazumdar3 and Johnson and Breen5 were the first to analyze the effect of PEEP on the non–steady-state

CO2 kinetics in healthy lungs. Our main goal was to further their work and to describe the same effect in the context of Figure 1. The elimination of CO2 per breath and tension-based CO2 a lung RM. Thus, to our knowledge, the data presented in values during the protocol. PVCO (dots), PaCO (squares) and PETCO 2 2 2 the current study are the first to describe the effect of PEEP (triangles). White columns in the lower panel are the elimination of and lung recruitments on the elimination of CO2 in CO2 per breath (VTCO2,br) values. *Comparisons between values of the same level of positive end-expiratory pressure (PEEP) before and after a non–steady-state conditions of healthy and surfactant- lung recruitment maneuver (RM); P Ͻ 0.05. depleted lungs.

972 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 3. (Continued)

PEEP (cm H2O) Decreasing limb of PEEP 18 12 6 0

92 (67–117) 91 (67–115) 89 (68–102) 92 (64–120) 67 (45–89) 66 (38–94) 86 (69–103) 83 (65–101)* 20 (14–26)* 23 (15–31) 20 (12–28)* 26 (18–34)* 12 (6–18) 10 (7–13) 9 (6–12) 6 (4–8) 1.80 (1.3–2.3)* 2.00 (1–3) 3.10 (2.1–4.1)* 3.70 (2.5–5.2)* 1.63 (1.1–2.1)* 1.93 (0.9–2.9) 2.95 (2–4)* 3.29 (1.3–5.3)* 0.07 (0.05–0.1) 0.11 (0.1–0.2)* 0.19 (0.1–0.3)* 0.47 (0.1–1)*

113 (73–153) 119 (91–149)* 117 (92–142)* 120 (85–155)* 77 (53–104) 80 (59–99)* 87 (65–109)* 85 (65–105) 36 (25–47) 36 (26–46) 40 (32–48) 46 (38–54) 12 (8–16) 11 (9–13) 9 (7–11) 9 (7–11) 3.29 (2–4.4)* 4.00 (2.8–5.2)* 5.00 (3.8–6.2)* 5.80 (4.5–7.2) 2.99 (2.3–3.6)* 3.52 (2.6–4.4) 3.95 (3.1–4.8)* 2.96 (2.3–3.7) 0.22 (0.2–0.3)* 0.48 (0.2–0.8)* 1.07 (0.5–1.5)* 2.56 (2–3.2)

Figure 2. Relative changes in the CO2 elimination per breath (VTCO2,br) during sequential changes of positive end-expiratory pressure (PEEP). The determinants of VTCO2,br are represented by the effective portion of cardiac output (COEPP), the area for CO2 exchange ˙ (Pa-ETCO2), and the effective portion of ventilation (VA). The relative changes in each of the variables as they are induced by the PEEP change are expressed in percentage, from zero to a maximum cutoff of 50%. An improvement is deﬁned as a change toward more normal ˙ values. Such improvements in CO2 elimination are characterized by increases in VA and COEPP and decreases in Pa-ETCO2 and are represented as white bars above the zero line. Impairment is deﬁned as a change toward more abnormal values. Such impairments in ˙ CO2 elimination are characterized by decreases in VA and COEPP and increases in Pa-ETCO2 and are represented by black bars below the zero line. *P Ͻ 0.05.

The elimination of CO2 by the lungs, at constant venti- influence on CO2 elimination of 1 factor differed from other lation and body metabolism, depends on its transport by factors. the blood into the lungs, its diffusion through the alveolar- Figure 3 shows the change in the elimination of CO2 ˙ 3,31 capillary membrane, and its elimination by the Va. during an increase in PEEP from 6 to 12 cm H2O without RM. Therefore, the final Vtco2,br value measured at the airway The layout of this figure is similar to the ones presented opening in response to a PEEP challenge with or without a in the articles by Breen and Mazumdar3 and Johnson and lung RM is the result of a complex interaction of several Breen5 to facilitate the reader’s comparison of the data. factors. Figure 2 and Tables 1 to 3 show that these interactions According to the results shown in this figure, changes in ˙ differ between healthy and lavaged lungs and, sometimes, the Vtco2,br, lung perfusion, CO2 exchange, and Va

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 973 CO2 Elimination by Lung Recruitment and PEEP

Figure 3. Time course of CO2 elimination during 2 consecutive positive end-expiratory pressure (PEEP) steps. A, Data of each of the animals is ˙ displayed for two 10-minute periods at 6 and 12 cm H2O of PEEP, respectively. VTCO2,br is the elimination of CO2 per breath, VA the alveolar ventilation, Pa-ETCO2 the arterial to end-tidal difference in PCO2 (calculated by the on-line PaCO2 value from the arterial catheter minus PETCO2 from capnograms), CO the continuous cardiac output, and ⌬FRC the change in functional residual capacity. B, Baseline is the mean value of the last 10 breaths on 6 cm H2O of PEEP (here summarized in breath 0) followed by 18 consecutive breaths at 12-PEEP. after a PEEP challenge vary between healthy and lavaged decreasing V˙ a and CO (Fig. 3A). The effects of a partial lungs. alveolar recruitment induced by the increasing PEEP levels

In healthy lungs, 12-PEEP decreased Vtco2,br because of are supported by concomitant increases in Pao2 and compli- decreased V˙ a and CO despite a slight improvement in gas ance at reduced dead spaces (Table 2). Although CO was exchange (Fig. 3A). More than 90% of the effects on Vtco2,br reduced by 21%, its COEPP increased by 22% because of the induced by 12-PEEP occurred within 5 minutes. These results recruitment of shunt areas (Table 3). The final result was a 3 ˙ ˙ are similar to the findings of Breen and Mazumdar observed better elimination of CO2 because of an improved V/Q ratio 5 in healthy dogs and those of Johnson and Breen in healthy as indicated by a decrement in SIII. The different findings in anesthetized humans. They also found that a PEEP challenge lavaged and healthy lungs point toward differences in the ˙ decreased Vtco2,br because of a decrement in Va and CO, physiological effects that PEEP and recruitment have on with most of the recovery of Vtco2,br within 10 minutes. normally aerated and on collapsed lungs.

Differences in the recovery of Vtco2,br observed between Figure 3B provides a zoomed view of the first breaths after these protocols could be explained by differences in the a change in a PEEP step. The decrement in Vtco2,br after an species studied, differences in the levels of PEEP applied (11 increase in PEEP was caused by the trapped gas within lungs cm H2O in the study by Breen and Mazumdar and 10 cm H2O at the onset of the higher PEEP, which diluted alveolar CO2. in the study by Johnson and Breen), or by differences in The opposite results were observed in cases of PEEP reduc- hemodynamic status and treatments. tion. After a PEEP change, a fast recovery in Vtco2,br was In lavaged lungs, Vtco2,br increased until 12-PEEP caused found within 5 to 6 consecutive breaths. This finding is similar 5 by an increment in the area for CO2 exchange despite to the one described by Johnson and Breen in anesthetized

974 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 3. Continued.

patients in which the recovery of Vtco2,br after a change of In healthy lungs, Vtco2,br decreased in proportion to PEEP PEEP from 0 to 10 cm H2O took 8 breaths. before the recruitment but increased after it. At constant

ventilation and metabolism, this decrease in CO2 elimination should lead to retention of CO within the body when PEEP Does PEEP and Lung Recruitment Retain CO2 in 2 the Blood? is applied without a prior recruitment. However, Pvco2, ៮ A protective ventilatory management with low Vt and Paco2, and thus P v-aco2 were not affected much during the plateau pressures is currently mandatory when treating protocol (Table 1 and Figs. 1 and 2), which must be inter- 12,13 acutely injured lungs. Arterial hypercapnia is thus a preted that CO2 was not retained within the blood, at least clinically tolerated negative consequence of an intentional during the study period. After recruitment, the elimination of

decrease in V˙ a. This hypercapnic state could become even CO2 and global lung physiology improved at PEEP levels

worse if PEEP, as many authors have postulated, caused down to 12 cm H2O (Table 1 and Figs. 1 and 2) with chances 14–17 CO2 retention in the body. for CO2 retention even lower than before the recruitment.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 975 CO2 Elimination by Lung Recruitment and PEEP

Our results differ from those of Breen and Mazumdar3 and would not have been able to show reproducibly the sequen- 5 Johnson and Breen who found increased Pvco2 and Paco2 in tial nature and the time dependence of CO2 elimination. healthy lungs at 10 and 11 cm H2O of PEEP, respectively. For the same reason, we chose to study the CO2 kinetics Differences in the hemodynamic status, protocol time, and during a short lapse of time because we were interested in experimental models might explain these differences. the CO2 kinetics during the non–steady-state conditions In surfactant-depleted lungs, however, results were induced by RMs and a PEEP titration process. This is totally different. Pv-aco2 increased whereas Pa-etco2 crucial new information, which should contribute to a decreased with positive-pressure ventilation as long as better understanding of CO2 kinetics, and we hope that it lung collapse was low and overall lung function pre- will finally have clinical implications for the monitoring of served (Table 2 and Fig. 1). The increased area of gas patients during mechanical ventilation. exchange leading to an augmented diffusion of CO2 across the alveolar-capillary membrane after a lung Limitations recruitment can explain this effect and was confirmed by Surfactant-depleted lungs as used in our experimental parallel increments in Pao2 and Crs, 2 well-known mark- study do not adequately represent the complex nature of 32,33 ers of lung recruitment. COEPP increased with in- acute lung injury in real patients, and thus our results creasing airway pressures because of the recruitment of should be interpreted with caution. previously collapsed capillaries in the atelectatic areas It is well documented that healthy and sick human lungs 29,30 (Table 3). Because Vtco2,br is directly proportional to have different opening pressures and therefore we lung perfusion, this increment in COEPP facilitates the chose our recruitment pressures in health and disease transport of CO2 from the body stores toward the lungs assuming that this difference would be true for animals where it is eliminated. Note that the absolute values of also. To achieve a complete recruitment effect in our

Pvco2 and Paco2 decreased with both lung recruitment surfactant-depleted animals, we decided to use an arbitrary and PEEP, thereby confirming that CO2 was not retained and fixed opening pressure of 50 cm H2O for all animals within the blood but rather eliminated more efficiently. with sick lungs based on our own previous experience33 In lavaged lungs, the highest recruitment pressures because we did not have access to lung imaging by applied in pressure control ventilation resulted in a mean computed tomographic scan to determine an optimal open- Vt of 7.4 mL/kg (221 [202–251] mL). This small and ing pressure individually for each animal. Had we used the transient increase in minute ventilation could have influ- same pressure as we did for healthy animals, the risk of enced the elimination of CO2 beyond the actual lung having incompletely recruited diseased lungs and thus recruitment effect during the descending PEEP steps. inconclusive study results would have been high.

However, we believe that the decreased values for Pvco2 and Paco2 on the descending limb of the PEEP titration CONCLUSIONS were mainly attributable to the recruitment of alveoli The results of this study show that lung recruitment and because: (1) changes in the area of gas exchange, and not PEEP have different effects on the elimination of CO2 in ˙ in Va, had a dominating influence on Vtco2,br (Fig. 2), healthy and lavaged lungs. These differences can be ex- (2) variables representing the recruitment effect and plained by a complex interaction between the key factors of ˙ ⌬ which are independent of Va (Crs, SIII, FRC, Pao2,or lung perfusion, diffusion through the alveolar-capillary shunt) improved after lung recruitment, and (3) a tran- membrane, and V˙ a. Our results suggest that sufficiently sient and marginal increase in minute ventilation for 2 high levels of PEEP applied after lung recruitment may minutes only would not have had a lasting impact on help decrease hypercapnia in patients treated with lung- Vtco2,br after a lapse of 10 minutes, the point in time protective ventilation and low Vt strategies. when the blood samples were taken. AUTHOR AFFILIATIONS Clinical Implications From the *Department of Anesthesiology, Hospital Privado de In contrast to continuous positive airway pressure ma- Comunidad, Mar del Plata, Argentina; †CSEM Centre Suisse neuvers, cycling RMs have the following advantages: (a) d’Electronique et de Microtechnique SA, Research Centre for they are hemodynamically better tolerated,34 (b) the Nanomedicine, Landquart, Switzerland; ‡Department of Criti- step-wise and sequential increments in PEEP allow the cal Care Medicine, Fundacioń Jimeńez Díaz-UTE, Madrid, gained volumes of air to spread progressively instead of Spain; §Bioengineering Laboratory, Electronic Department, abruptly throughout the lung parenchyma,35,36 and (c) University of Mar del Plata, Mar del Plata, Argentina; and ʈDepartment of Medical Sciences, Clinical Physiology, Univer- they allow real-time monitoring of respiratory variables sity Hospital, Uppsala Sweden. on a breath-by-breath basis. Today, the way to conduct and optimize RM is a topic of much debate, where the REFERENCES advent of volume-based capnographic monitoring may 1. Qvist J, Pontoppidan H, Wilson RS, Lowenstein E, Laver MB. add important new arguments in favor of such an Hemodynamic response to mechanical ventilation with PEEP: approach because of the noninvasive and real-time na- the effect of hypervolemia. Anesthesiology 1975;42:45–55 ture of this methodology. 2. Berglund JE, Haldeń E, Jakonson S, Landelius J. Echocardio- Because CO kinetics are context sensitive and highly graphic analysis of cardiac function during high PEEP venti- 2 lation. Intensive Care Med 1994;20:174–80 dependent on the sequence of steps during a cycling RM, in 3. Breen PH, Mazumdar B. How does positive end-expiratory our protocol, the levels of PEEP were not assigned in pressure decrease CO2 elimination from the lung? Respir random order. Had we randomized the protocol steps, we Physiol 1996;103:233–42

976 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 4. Elliott WR, Harris AE, Philip JH. Positive end-expiratory 22. Englhoff H. Volumen inefficax. Bemerkungen zur Frage des pressure: implications for tidal volume changes in anesthesia scha¨dlichen Raumes. Uppsala La¨kareforen Forhandl 1938; machine ventilation. J Clin Monit 1989;5:100–4 44:191–218 5. Johnson JL, Breen PH. How does positive end-expiratory 23. Harris B, Bailey DL, Chicco P, Bailey EA, Roach PJ, King GG.

pressure decrease pulmonary CO2 elimination in anesthetized Objective analysis of whole lung and lobar ventilation/ patients? Respir Physiol 1999;118:227–36 perfusion relationships in pulmonary embolism. Clin Physiol 6. Hedenstierna G, Lundberg S. Airway compliance during arti- Funct Imaging 2008;28:14–26 ficial ventilation. Br J Anaesth 1975;47:1277–81 24. Blanch LL, Fernandez R, Saura P, Baigorri F, Artigas A. 7. Coffey RL, Albert RK, Robertson HT. Mechanisms of physio- Relationship between expired capnogram and respiratory sys- logical dead space response to PEEP after acute oleic acid lung tem resistance in critically ill patients during total ventilatory injury. J Appl Physiol 1983;55:1550–7 support. Eur Respir J 1999;13:1048–54 8. Lundquist H, Hedenstierna G, Strandberg A, Tokics L, Brismar 25. Hofbrand BI. The expiratory capnogram: a measure of B. CT-assessment of dependent lung densities in man during ventilation-perfusion inequalities. Thorax 1966;21:518–24 general anaesthesia. Acta Radiol 1995;36:626–32 26. Stro¨mberg NO, Gustafsson PM. Ventilatory inhomogeneity 9. Brismar B, Hedenstierna G, Lundquist H. Pulmonary densities assessed by nitrogen washout and ventilation-perfusion mis- during anesthesia with muscular relaxation: a proposal of match by capnography in stable and induced airway obstruc- atelectasis. Anesthesiology 1985;62:422–8 tion. Pediatr Pulmonol 2000;29:94–102 10. Gattinoni L, Caironi P, Valenza F, Carlesso E. The role of 27. Berggren SM. The oxygen deficit of arterial blood caused by CT-scan studies for the diagnosis and therapy of acute respi- non-ventilated parts of the lungs. Acta Physiol Scand ratory distress syndrome. Clin Chest Med 2006;27:559–70 1942;Suppl 4:4–9 11. Ware LB, Matthay MA. The acute respiratory distress syn- 28. Lachmann B, Jonson B, Lindroth M, Robertson B. Modes of drome. N Engl J Med 2000;342:1334–49 artificial ventilation in severe respiratory distress syndrome: 12. The National Heart, Lung, and Blood Institute ARDS Clinical lung function and morphology in rabbits after wash-out of Trials Network. Ventilation with lower tidal volumes as com- alveolar surfactant. Crit Care Med 1982;10:724–32 pared with traditional tidal volumes for acute lung injury and 29. Rothen HU, Sporre B, Englberg G, Wegenius G, Hedenstierna the acute respiratory distress syndrome. The Acute Respira- G. Reexpansion of atelectasis during general anaesthesia: a tory Distress Syndrome Network. N Engl J Med 2000; computed tomography study. Br J Anaesth 1993;71:788–95 342:1301–8 30. Borges JB, Okamoto VN, Matos GF, Caramez MPR, Arantes 13. Amato MBP, Barbas CSV, Medeiros DM, Magaldi RB, Schet- PR, Barros F, Souza CE, Victorino JA, Kacmarek RM, Barbas tino GP, Lorenzi-Filho G, Kairalla RA, Deheinzelin D, Mun˜oz CSV, Carvalho CRR, Amato MBP. Reversibility of lung col- C, Oliveira R, Takagaki TY, Carvalho CRR. Effect of a protec- lapse and hypoxemia in early acute respiratory distress syn- tive ventilation strategy on mortality in the acute respiratory drome. Am J Respir Crit Care Med 2006;174:268–78 distress syndrome. N Engl J Med 1998;338:347–54 31. Cherniack NS, Longobardo GS. Oxygen and carbon dioxide 14. Dueck R, Wagner PD, West JB. Effects of positive end- gas stores of the body. Physiol Rev 1970;50:196–243 expiratory pressure on gas exchange in dogs with normal and 32. Lachmann B. Open up the lung and keep the lung open. edematous lungs. Anesthesiology 1977;47:359–66 Intensive Care Med 1992;18:319–21 15. Pesenti A, Marcolin R, Prato P, Borelli M, Riboni A, Gattinoni 33. Suarez-Sipmann F, Bo¨hm SH, Tusman G, Pesch T, Thamm O, L. Mean airway pressure vs positive end-expiratory pressure Reissmann H, Reske A, Magnusson A, Hedenstierna G. Use of during mechanical ventilation. Crit Care Med 1985;13:34–7 dynamic compliance for open lung positive end-expiratory 16. Isserles SA, Breen PH. Can changes in end-tidal PCO2 measure pressure titration in an experimental study. Crit Care Med changes in cardiac output? Anesth Analg 1991;73:808–14 2007;35:214–21 17. Hedenstierna G, White FC, Mazzone R, Wagner PD. Redistri- 34. Celebi S, Ko¨ner O, Menda F, Korkut K, Suzer K, Cakar N. The bution of pulmonary blood flow in the dog with PEEP venti- pulmonary and hemodynamic effects of two different recruit- lation. J Appl Physiol 1979;46:278–87 ment maneuvers after cardiac surgery. Anesth Analg 18. Tusman G, Bo¨hm SH, Suarez-Sipmann F, Turchetto E. Alveolar 2007;104:384–90 recruitment improves ventilatory efficiency of the lungs dur- 35. Hickling KG. Best compliance during a decremental, but not ing anesthesia. Can J Anaesth 2004;51:723–7 incremental, positive end-expiratory pressure trial is related to 19. Tusman G, Bo¨hm SH, Suarez-Sipmann F, Maisch S. Lung open-lung PEEP: a mathematical model of acute respiratory recruitment improves the efficiency of ventilation and gas distress syndrome lungs. Am J Respir Crit Care Med exchange during one-lung ventilation anesthesia. Anesth 2001;163:69–78 Analg 2004;98:1604–9 36. Albaiceta GM, Taboada F, Parra D, Luyando LH, Calvo J, 20. Tusman G, Suarez-Sipmann F, Bo¨hm SH, Pech T, Reissmann Menendez R, Otero J. Tomographic study of the inflection H, Meschino G, Scandurra A, Hedenstierna G. Monitoring points of the pressure-volume curve in acute lung injury. Am J dead space during recruitment and PEEP titration in an Respir Crit Care Med 2004;170:1066–72 experimental model. Intensive Care Med 2006;32:1863–71 21. Tusman G, Scandurra A, Bo¨hm SH, Suarez-Sipmann F, Clara F. Model fitting of volumetric capnograms improves calculations of airway dead space and slope of phase III. J Clin Monit Comput 2009;23:197–206

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 977 Society for Obstetric Anesthesia and Perinatology

Section Editor: Cynthia A. Wong

Pitfalls in Chronobiology: A Suggested Analysis Using Intrathecal Bupivacaine Analgesia as an Example Steven L. Shafer, MD,* Bjoern Lemmer, MD, PhD,† Emmanuel Boselli, MD, PhD,‡ Fabienne Boiste, MD,‡ Lionel Bouvet, MD,‡ Bernard Allaouchiche, MD, PhD,§ and Dominique Chassard, MD, PhD§

BACKGROUND: The duration of analgesia from epidural administration of local anesthetics to parturients has been shown to follow a rhythmic pattern according to the time of drug administration. We studied whether there was a similar pattern after intrathecal administration of bupivacaine in parturients. In the course of the analysis, we came to believe that some data points coincident with provider shift changes were influenced by nonbiological, health care system factors, thus incorrectly suggesting a periodic signal in duration of labor analgesia. We developed graphical and analytical tools to help assess the influence of individual points on the chronobiological analysis. METHODS: Women with singleton term pregnancies in vertex presentation, cervical dilation 3 to 5 cm, pain score Ͼ50 mm (of 100 mm), and requesting labor analgesia were enrolled in this study. Patients received 2.5 mg of intrathecal bupivacaine in 2 mL using a combined spinal-epidural technique. Analgesia duration was the time from intrathecal injection until the first request for additional analgesia. The duration of analgesia was analyzed by visual inspection of the data, application of smoothing functions (Supersmoother; LOWESS and LOESS [locally weighted scatterplot smoothing functions]), analysis of variance, Cosinor (Chronos-Fit), Excel, and NONMEM (nonlinear mixed effect modeling). Confidence intervals (CIs) were determined by bootstrap analysis (1000 replications with replacement) using PLT Tools. RESULTS: Eighty-two women were included in the study. Examination of the raw data using 3 smoothing functions revealed a bimodal pattern, with a peak at approximately 0630 and a subsequent peak in the afternoon or evening, depending on the smoother. Analysis of variance did not identify any statistically significant difference between the duration of analgesia when intrathecal injection was given from midnight to 0600 compared with the duration of analgesia after intrathecal injection at other times. Chronos-Fit, Excel, and NONMEM produced identical results, with a mean duration of analgesia of 38.4 minutes (95% CI: 35.4–41.6 minutes), an 8-hour periodic waveform with an amplitude of 5.8 minutes (95% CI: 2.1–10.7 minutes), and a phase offset of 6.5 hours (95% CI: 5.4–8.0 hours) relative to midnight. The 8-hour periodic model did not reach statistical significance in 40% of bootstrap analyses, implying that statistical significance of the 8-hour periodic model was dependent on a subset of the data. Two data points before the change of shift at 0700 contributed most strongly to the statistical significance of the periodic waveform. Without these data points, there was no evidence of an 8-hour periodic waveform for intrathecal bupivacaine analgesia. CONCLUSION: Chronobiology includes the influence of external daily rhythms in the environment (e.g., nursing shifts) as well as human biological rhythms. We were able to distinguish the influence of an external rhythm by combining several novel analyses: (1) graphical presentation superimposing the raw data, external rhythms (e.g., nursing and anesthesia provider shifts), and smoothing functions; (2) graphical display of the contribution of each data point to the statistical significance; and (3) bootstrap analysis to identify whether the statistical significance was highly dependent on a data subset. These approaches suggested that 2 data points were likely artifacts of the change in nursing and anesthesia shifts. When these points were removed, there was no suggestion of biological rhythm in the duration of intrathecal bupivacaine analgesia. (Anesth Analg 2010;111:980–5)

he duration of epidural analgesia, and the hypnotic has been investigated in the morning or in the evening; effect of many drugs used for anesthesia or sedation, significantly higher plasma lidocaine levels were obtained Thave been shown to vary in a periodic pattern after evening dosing in children.2 It has also been shown through the day.1 The transcutaneous passage of lidocaine under conditions of daily dental practice that the duration of local anesthesia after subcutaneous mepivacaine or arti- Authors’ affiliations are listed at the end of the article. caine is not constant throughout the day.3 These studies Accepted for publication March 10, 2010. suggest a longer duration of anesthesia in the afternoon at Supported solely by institutional sources. approximately 1500 for a variety of local anesthetic drugs.4 Supplemental digital content is available for this article. Direct URL citations We have demonstrated a significant effect of the time of appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.anesthesia-analgesia.org). administration on the duration of epidural ropivacaine Address correspondence and reprint requests to Dominique Chassard, MD, analgesia in parturients, with a 30% longer duration of PhD, Department of Anesthesiology and Intensive Care, Hoˆpital Me`re Enfant, analgesia when the block is placed in the afternoon.5 Bron 69500, France. Address e-mail to dominique.chassard @ chu-lyon.fr. Copyright © 2010 International Anesthesia Research Society Similarly, spinal opioids such as fentanyl or sufentanil do DOI: 10.1213/ANE.0b013e3181dd22d4 not have a constant duration of analgesia throughout a

980 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 24-hour period.6,7 These studies indicate that epidurally or by subtracting 24 hours from all data points Ͼ12 hours after spinally injected anesthetic drugs exhibit a circadian pat- the point, and adding 24 hours for each data point Ͼ12 hours tern. Because there are no such data on intrathecal local before the data point. This required calculating a new anesthetics, we tested the hypothesis that the duration of smoother function for every data point, rather than just once action of intrathecal local anesthetics has a circadian for across all data. However, for this data set, the compu- rhythm in parturients during the first stage of labor. In tational time was negligible. The R code for Figure 2 exploring this hypothesis, we developed several novel appears in Appendix 1 (see Supplemental Digital Con- approaches to assess the influence of external rhythms on tent1, http://links.lww.com/AA/A123). chronobiological models. The influence of parity and time of day (day [0800–2000] versus night [2000–0800]) on the VAS score before intra- METHODS thecal injection was analyzed using Mann-Whitney tests. After obtaining agreement from the Ethical Committee of Analgesia durations were divided into 4 periods based on the University of Lyon (France) and written informed the time of intrathecal injection: midnight to 0600, 0600 to consent, we enrolled pregnant women, ASA physical status 1200, 1200 to 1800, and 1800 to midnight. Times exactly on I, parity 0 and 1, with a singleton vertex pregnancy of at the border of 2 periods (e.g., 0600) were assigned to the least 36 weeks of gestation. Each woman was in the first later period. Differences in analgesia duration among peri- stage of spontaneous labor with a cervical dilation 2 to 4 cm ods were analyzed using analysis of variance. when requesting labor analgesia. Only patients with un- Parametric analysis estimated the duration of analgesia complicated pregnancy and normal fetal heart rate tracings as a cosine function of time, based on the equation: requesting analgesia were enrolled. We excluded patients receiving antihypertensive drugs and patients who had i ϭ ϩ ͸ͩ already received opioids during labor. analgesia duration mesor amplitudei 1 Each patient received an intrathecal injection of 2.5 mg of plain bupivacaine in 2 mL (a mixture of 1 mL 0.25% bupiva- 2␲ caine and 1 mL normal saline) using a combined spinal- ϫ ͫ Ϫ ͬͪ cos (t acrophasei) (1) epidural technique with the Braun Escopan௡ combination periodi needle (Braun Medical SA, Boulogne, France). After insertion of a 27-gauge pencil-point spinal needle and intrathecal where t is the time of initiation of intrathecal analgesia, bupivacaine injection, a 20-gauge multiorifice catheter was mesor is the mean of the fitted curve (i.e., average duration inserted in the epidural space. The epidural catheter was left of anesthesia), amplitude is half of the magnitude of the in position, but no drug was delivered epidurally until the circadian variation, and acrophase is the phase offset end of the intrathecal bupivacaine effect. relative to midnight (time 0). These parameters were calcu- Pain was assessed using a visual analog scale (VAS) of lated for periods of 6, 8, 12, and 24 hours. pain intensity at 15, 30, and 45 minutes after the block, and Three fitting programs were used: Chronos-Fit,†10 Excel when secondary analgesia was requested. Each patient was 2007 (Microsoft Corp., Redmond, WA) (using the Solver presented with a 100-mm line, labeled as no pain (left end) function to minimize Ϫ2 log likelihood), and nonlinear mixed and worst pain imaginable (right end). Patients were asked effect modeling (NONMEM) (version 6.2, Icon Development to mark the line to indicate the intensity of their pain at the Solutions, Ellicott City, MD) (which also minimized Ϫ2 log peak of a contraction. Patients could request additional likelihood). Results were considered significant if P values analgesia if pain relief was unsatisfactory by 15 minutes were Ͻ0.05. No adjustment was made for multiple tests (i.e., after injection of the intrathecal bupivacaine. If a patient we considered 4 different periodic signals but did not adjust requested relief within 15 minutes after injection, this was our P value to reflect the multiple tests). considered a technical failure; these patients were excluded The contribution of each subject to the final model was from further data collection. When additional analgesia estimated using NONMEM, comparing the individual Ϫ2 was requested, the study protocol and data collection were log likelihood of a null model (analgesia duration ϭ mesor terminated, and analgesia duration was recorded. Analge- [e.g., mean]) with the final model (Eq. 1, period ϭ 8 hours). sia duration was the time from intrathecal bupivacaine Ninety-five percent confidence intervals for the estimated administration to the first request for additional analgesia. parameters were calculated using a NONMEM bootstrap Ϯ Data are presented as mean SD. Circadian rhythms analysis of 1000 random bootstrap samples with replace- were explored graphically using the 3 smoothing functions, ment. The dependency of the statistical significance on a 8 9 Supersmoother,* LOWESS, and LOESS, as implemented subset of the data was estimated by NONMEM using in the R statistical programming language (R Project for bootstrap analysis to calculate the difference in Ϫ2 log Statistical Computing: http://www.r-project.org). Super- likelihood between the null (no period) and final (8-hour smoother is a variable span scatterplot smoothing function, period) models in 1000 random samples with replacement. whereas LOWESS and LOESS are locally weighted scatterplot NONMEM was run using the PLT Tools platform (PLTsoft, smoothing functions. To prevent edge effects, the smoothed San Francisco, CA; www.pltsoft.com). value for each time point was calculated by rescaling the X No formal power analysis was done. The sample size data to place the point in the center of the data. This was done †Zuther P, Lemmer B. Chronos-Fit, version 1.05, 2004. Available at: *Friedman JH. A variable span scatterplot smoother. Laboratory for Com- http://www.ma.uni-heidelberg.de/inst/phar/forschungLemmer.html. Ac- putational Statistics, Stanford University Technical Report No. 5, 1984. cessed June 6, 2008.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 981 Pitfalls in Chronobiological Studies

Table 1. Demographic Data (Mean ؎ SD) Table 2. Analgesia Duration by Period

(Age (y) 29 Ϯ 4 Period Analgesia duration (min ؎ SD Height (cm) 164 Ϯ 5 1 (0:01 to 6:00) 33.1 Ϯ 10.4 Weight (kg) 72 Ϯ 9 2 (6:01 to 12:00) 41.2 Ϯ 17.8 Dilation (cm) 3 (2–4) 3 (12:01 to 18:00) 40.0 Ϯ 14.5 Parity 0/1 (n) 38/44 4 (18:01 to midnight) 40.0 Ϯ 14.2

Table 3. Periodic Model Parameter Typical value 95% CI Mean (min) 38.4 35.4–41.6 Period (h) 8 na Amplitude (min) 5.8 2.1–10.7 Offset (h) 6.5 5.4–8.0 CI ϭ conﬁdence interval; na ϭ not applicable.

2349 (Supersmoother), 0609 and 2318 (LOWESS smoother), and 0722 and 1812 (LOESS smoother). The orange horizontal bar shows the change in the certified registered nurse anesthetist (CRNA) and midwife shift, whereas the purple horizontal bar shows the change in the shift for the anesthesia fellow and anesthesia attending. Figure 1. Analgesia duration versus the time of intrathecal injection. The duration of analgesia was modestly decreased dur- ϭ ϭ Black horizontal line median analgesic duration; red line ing the period from midnight to 0600 compared with the Supersmoother; blue line ϭ LOWESS smoother; green line ϭ LOESS smoother. Arrows denote the peaks for each smoother of the same other periods (Table 2). This trend did not reach statistical color. Orange horizontal bar ϭ daytime certiﬁed registered nurse significance by analysis of variance (P ϭ 0.42). anesthetist and midwife shift; purple horizontal bar ϭ daytime Chronos-Fit, Excel, and NONMEM produced identical anesthesia fellow and anesthesia attending shift. results. Excel and NONMEM returned identical Ϫ2 log likelihood estimates. There was just 1 statistically signifi- was chosen based simply on the experience of the investi- cant (P ϭ 0.04) periodic waveform, with an 8-hour period, gators. Power analysis for chronobiological studies is re- amplitude of 5.8 minutes, acrophase (phase offset) of 6.5 viewed in more detail in the Discussion section. hours, and a mean duration of analgesia (mesor) of 38.4 The data files for this analysis, the R programming code, minutes (Table 3 and Fig. 2). The periodic waveform the Excel spreadsheet for the nonlinear regression, and captured the peak seen in the original data at approxi- NONMEM control and output files for chronobiology mately 0630, and approximately captured the peaks at 18 analysis using these data as an example are available as a hours (LOESS) and 24 hours (Supersmoother, LOWESS). Web supplement to this article. In Figure 2, the contribution of each data point can be inferred from the size and color of the dot. Black data RESULTS points represent observations better fit by the model with Ninety-one parturients were enrolled in this study from the 8-hour period than by the null model, which only June 2005 through October 2005. Nine were excluded for modeled the mesor (average analgesic duration). Red data spinal analgesia failure, protocol violation, or emergency points represent observations worse fit by the model with cesarean delivery before a request for analgesia. Patient the 8-hour period than by the null model. The size of the characteristics (n ϭ 82) are provided in Table 1. data point represents the increase (black) or decrease (red) The mean VAS score before intrathecal injection was in Ϫ2 log likelihood of the 8-hour periodic model over the 70 Ϯ 17 mm (median, 70 mm; range, 40–100 mm). VAS scores null model. before intrathecal injection were not significantly affected by As shown in Figure 2, the 2 points that contributed most to the period of day (74 Ϯ 13 mm from 0800 to 2000 vs 69 Ϯ 12 the improvement in Ϫ2 log likelihood were from the 2 mm from 2000 to 0800; P ϭ 0.08) or parity (72 Ϯ 16 mm for intrathecal injections performed shortly before the change in parity 0 vs 68 Ϯ 10 mm for parity 1; P ϭ 0.17). Visual nursing shift at 0700. One was placed at 0624 and lasted 81 inspection of the baseline VAS data over the 24-hour time minutes, and the other was placed at 0653 and lasted 90 course did not suggest any periodic signal. Fifteen minutes minutes. after intrathecal bupivacaine, the mean VAS score decreased Figure 3 shows the result of the bootstrap analysis based to a median of 20 mm (range, 0–70 mm). on 1000 samples of the data set with replacement. The Figure 1 shows analgesia duration versus the time of x-axis shows the difference in Ϫ2 log likelihood, and the intrathecal injection. The black horizontal line is the aver- y-axis shows the cumulative fraction. The vertical lines age analgesic duration. The smoothers (red ϭ Super- mark differences in Ϫ2 log likelihood associated with P ϭ smoother, blue ϭ LOWESS, and green ϭ LOESS) all reveal 0.05, 0.01, 0.001, and 0.0001. If a model was highly statisti- a bimodal pattern with peaks (colored arrows) at 0624 and cally significant, regardless of which subset of the data was

982 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 2. Same result as Figure 1, with the 8-hour periodic waveform Figure 4. The same analysis as in Figure 1, but with the 2 outlying shown in gold. The size of the points reflects the magnitude of the data points at 0624 and 0653 removed. Only the LOWESS smoother contribution of each point toward (black) or against (red) statistical (blue line) continues to show evidence of periodicity. The Super- significance of the final model. smoother and LOESS smoother are essentially flat.

obtained by Wong et al.,11 39 Ϯ 25 minutes. Stocks et al.,12 using a technique of up-down sequential allocation, demonstrated that the minimum local analgesic dose of intrathecal bupivacaine was 1.99 mg, and that the duration of analgesia with this dose was 43 Ϯ 19 minutes. However, using the same dosage, Lim et al.13 reported a longer duration of action, 76.3 Ϯ 5.9 minutes. The results could have been influenced by several factors other than the time of the day: for example, the method to calculate analgesia duration, VAS scores, cervical dilation at the time of inclusion, parity, and the use of oxytocin. These factors must be incorporated in future studies of chronobiology in parturients, as well as the observation that the perception of labor pain varies with time of day.14 Figure 3. The cumulative distribution of the improvement in Ϫ2 log This article was initially submitted to Anesthesia & likelihood comparing the null model with the final model in 1000 Analgesia as a demonstration of the chronobiology of anal- bootstraps. In 40% of the bootstrap analyses, the difference was not gesia after intrathecal bupivacaine. Over the course of peer statistically significant, suggesting that the statistical significance depended on the inclusion of a subset of the data points in the review and additional analysis, this article has metamor- analysis. phosed into a cautionary tale of the influence of external rhythms on chronobiological analysis, and an exploration considered, then the vast majority of replicates in a boot- of data analysis methods to facilitate detection of artifacts. strap analysis would show a P value Ͻ0.05. That is not the Our presentation of the methods and results in this article case. Statistical significance was not achieved in 40% of the reflects the way we think this type of analysis should replicates, suggesting that statistical significance is highly proceed. The actual analysis was much more convoluted. dependent on the subset of data included. Given that the 2 The most critical element in data analysis is visualizing data points that contribute most to the model occur shortly the data. Figure 1 is our proposal for visualization of the before the change in nurse and anesthesia provider, we raw data in chronobiological analyses. The figure shows investigated the performance of the model with those the raw data and adds 3 smoothing functions to draw the outliers excluded. eye toward the underlying signal. Because smoothers are Figure 4 shows the data with the 2 suspected outlying data prone to edge effects, the smoothing functions have been points removed. When those 2 data points were excluded, calculated by centering each observation in a 24-hour time there was no longer a statistically significant periodic signal window, permitting the smoothing function to properly according to our reanalysis with Chronos-Fit, Excel, and “wrap around” midnight. The smoothers show that there NONMEM. Only the LOWESS smoother (blue line) offered can be, at most, 2 peaks. Thus, the data themselves suggest any evidence of periodicity. The Supersmoother and LOESS that there is little reason to explore a periodic signal with a smoother functions were essentially flat. period shorter than 12 hours. In addition, Figure 1 shows the primary external DISCUSSION rhythms likely to influence the data: the start and end of the The mean duration of action of intrathecal bupivacaine daytime nursing and anesthesia shifts. It is immediately reported in our study, 39 Ϯ 15 minutes, is similar to that obvious that the spike in analgesia duration for intrathecal

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 983 Pitfalls in Chronobiological Studies injections before the change in CRNA shift might be critical observations). Because 40% of the bootstrap analy- explained because the CRNAs, and subsequently the anes- ses did not reach statistical significance (Fig. 3), a small thesiologists, are starting their daytime shift, delaying number of data points are driving the statistical result. assessment of patients’ pain. We suggest that future chro- Supported by the evidence in Figures 1, 2, and 3 that the nobiological analyses include a figure similar to Figure 1 to statistical significance was driven by an artifact, the obser- identify the basic patterns in the data, assess the likely vations at 0624 (81 minutes of analgesia) and 0653 (90 period of any chronobiological signal by using smoothing minutes of analgesia) were removed from the data. Figure functions, and assess the relationship, if any, to external 4 shows that there is still a peak in duration at the time of rhythms. Of course, it is entirely possible that rhythms the nursing shift change, but that the peak is significantly unknown to the investigators (e.g., lunch breaks, the timing attenuated by the loss of these 2 points. Indeed, Super- of the change in bed sheets, and the regular 4 am test of the smoother and the LOESS smoother were essentially flat. backup generators) might be present. If that was the case, Similarly, no periodic signal was statistically significant isolated spikes would appear at given times, but an under- once these points were removed. lying sine rhythm would not be evident. Our lives are governed by powerful external rhythms. Our second step was to divide the data into 4 time periods Most obviously, we are synchronized to the rotation of the that approximately divided the 24-hour day into “night” earth, and the resulting periods of light and darkness. (midnight to 0600), “morning” (0600 to 1200), “afternoon” Other rhythms infuse our lives. These can affect studies of (1200 to 1800), and “evening” (1800 to midnight). We did not chronobiology in 2 ways. First, they may induce real see any statistically significant differences. physiological changes. For example, arterial blood pressure Our third step was modeling the data. The “model” is a might increase twice daily with the commute to and from work. However, external rhythms might produce artifacts simple cosine function. We chose to model periods of 24, that have nothing to do with biology, such as delayed 12, 8, and 6 hours. However, the only period that makes a patient assessment during the changing of provider shifts. priori sense biologically is the 24-hour period. Of course, We suggest 6 steps to assess the influence of external there could be 2 peaks in some biological function, but why artifacts on chronobiological analyses: should they necessarily be 12 hours apart? If the day is divided into a rhythmic pattern with multiple peaks, why 1. Graph the raw data over 24 hours, along with 1 or should it be sinusoidal rather than (for example) 16 hours more smoothing functions. To avoid edge effects, the alternating with 8 hours, as in the awake/asleep cycle? It is function needs to wrap around midnight. We elimi- even more mysterious why the day should be divided into nated edge effects by calculating the value of the sinusoidal 6- or 8-hour periods. Interestingly, even though smoothing function for each observation with the we only saw 2 peaks with the smoothing functions, the time centered in a 24-hour window. If the smoothing 12-hour rhythm was not statistically significant. The only functions do not suggest a periodic waveform, then rhythm that was statistically significant was a cosine signal no further analysis should be undertaken. with a period of 8 hours, shown in Figure 2. Figure 2 also 2. If a periodic pattern is evident in the smoothing introduces another new graphic element: showing the functions, compare the smoothed functions with ex- extent to which each observation contributes for, or against, ternal rhythms, such as shift changes, that might the statistical significance of the model. affect the observations. Inspection of Figure 2 reveals a problem with the model. 3. If a periodic pattern is evident in the smoothing It captures the observed peak at 0630, but the other 2 peaks functions, fit the data with a regression program, in the sinusoidal model do not correspond with peaks in such as Chronos-Fit, Excel, or NONMEM. If there is the data identified by the smoothing functions. Examining only 1 point per subject, then it makes no difference the individual points, the 2 largest black dots are the what program is used. In this analysis, all 3 programs observations that contribute the most to the statistical returned results that were identical in all reported digits. If there are multiple observations per subject, significance of the 8-hour periodic waveform. Looking at only NONMEM can separate interindividual vari- the external rhythms (horizontal orange and purple bars), ability from intraindividual variability. these are the very observations that are most likely to 4. Represent the data points to show the contribution reflect the influence of the change in nursing and anesthesia for, or against, the final model. This will identify shifts. whether the model is being driven by a subset of the Before discarding the fit entirely, we turned to a boot- data. strap analysis to address whether the statistical significance 5. Undertake a bootstrap analysis to confirm whether of the model was highly dependent on the specific data statistical significance is driven by a small subset of sampled. If the data broadly supported the statistical the data points. significance of the more complex model, then statistical 6. If data artifacts are driving the result, remove them significance should be evident from any subset of the data and reassess the model. analyzed. However, if statistical significance depends on just a few points, then a significant fraction of the bootstrap Removing data that appear to be artifacts may bias study analyses will fail to be statistically significant (the boot- results in favor of the investigator’s expectation. For example, straps that do not include the critical observations), we do not know for certain that the observations at 0624 and whereas some bootstraps will have greatly increased sta- 0653 are artifacts. These data points could be completely valid tistical significance (bootstraps that have Ͼ1 copy of the observations. However, this analysis demonstrates that even

984 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA if these points are valid (which we doubt), the lack of ACKNOWLEDGMENTS correlation between the data and the other 2 peaks in the The authors thank Peter Tucker (American Translator, Univer- “statistically significant” 8-hour periodic signal preclude our sity of Aix-Marseille, France) for assistance in the preparation having any confidence in the model. of the manuscript. The authors also acknowledge the many We have only explored rhythms that are synchronized insightful suggestions of Dr. Dennis Fisher to this work. among all patients. If each patient had a strong individual rhythm, but a random phase relationship to every other REFERENCES patient, then the underlying chronobiology would be 1. Chassard D, Bruguerolle B. Chronobiology and anesthesia. missed in this type of analysis. Individual rhythms with Anesthesiology 2004;100:413–27 variable phase offset could only be detected by analyzing 2. Bruguerolle B, Giaufre E, Prat M. Temporal variations in multiple observations in each individual, and using a transcutaneous passage of drugs: the example of lidocaine in children and in rats. Chronobiol Int 1991;8:277–82 mixed effects model with intersubject variability in the 3. Lemmer B, Wiemers R. Circadian changes in stimulus thresh- phase offset (“acrophase”). old and in the effect of a local anaesthetic drug in human teeth: Conventional power analyses do not readily handle a studies with an electronic pulptester. Chronobiol Int 1989; data analysis plan that involves comparing 1 model to a 6:157–62 second model, as in this case in which a straight line is 4. Reinberg A, Reinberg MA. Circadian changes of the duration being compared with a sine wave. In our view, the best way of action of local anaesthetic agents. Naunyn Schmiedebergs Arch Pharmacol 1977;297:149–52 to approach power analysis is to perform a Monte Carlo 5. Debon R, Chassard D, Duflo F, Boselli E, Bryssine B, Allaouch- simulation, in which multiple synthetic data sets are gen- iche B. Chronobiology of epidural ropivacaine. Anesthesiology erated for a range of sample sizes, and the percent of 2002;96:542–5 successful trials is calculated. The study size is based on the 6. Debon R, Boselli E, Guyot R, Allaouchiche B, Lemmer B, sample size in the Monte Carlo simulation that produces Chassard D. Chronopharmacology of intrathecal sufentanil for the desired percent of models correctly identified. labor analgesia: daily variations in duration of action. Anes- thesiology 2004;101:978–82 In conclusion, circadian rhythms are an important as- 7. Pan PH, Lee S, Harris L. Chronobiology of subarachnoid pect of human biology. On the basis of these data, there is fentanyl for labor analgesia. Anesthesiology 2005;103:595–9 no important chronobiological rhythm in spinal bupiva- 8. Cleveland WS. Robust locally weighted regression and caine analgesia during labor. External daily rhythms may smoothing scatterplots. J Am Stat Assoc 1979;74:829–36 contaminate the data when assessing biological rhythms. 9. Cleveland WS, Grosse E, Shyu WM. Local regression models. We suggest that the approach used in this study may help In: Chambers JM, Hastie TJ, eds. Statistical Models in S. Pacific to separate true biological signals from external artifact in Grove, CA: Wadsworth & Brooks/Cole, 1992:309–76 10. Zuther P, Witte K, Lemmer B. ABPM-FIT and CV-SORT: an future studies. easy-to-use software package for detailed analysis of data from ambulatory blood pressure monitoring. Blood Press Monit AUTHORS’ AFFILIATIONS 1996;1:347–54 From the *Department of Anesthesiology, Columbia University, 11. Wong CA, Scavone BM, Loffredi M, Wang WY, Peaceman AM, New York, New York; †Institute of Pharmacology & Toxicology, Ganchiff JN. The dose-response of intrathecal sufentanil added University of Heidelberg, Heidelberg, Germany; ‡Department of to bupivacaine for labor analgesia. Anesthesiology 2000; Anesthesiology and Intensive Care, Hoˆtel-Dieu Hospital; and 92:1553–8 §University Claude Bernard-Lyon 1, Lyon, France. 12. Stocks GM, Hallworth SP, Fernando R, Engl AJ, Columb MO, Lyons G. Minimum local analgesic dose of intrathecal bupiv- RECUSE NOTE acaine in labor and the effect of intrathecal fentanyl. Anesthe- Steven L. Shafer is Editor-in-Chief of the Journal. The manu- siology 2001;94:593–8 script was handled by James G. Bovill, Guest Editor-in-Chief, 13. Lim Y, Ocampo CE, Sia AT. A comparison of duration of and Dr. Shafer was not involved in any way with the editorial analgesia of intrathecal 2.5 mg of bupivacaine, ropivacaine, and levobupivacaine in combined spinal epidural analgesia for process or decision. patients in labor. Anesth Analg 2004;98:235–9 DISCLOSURE 14. Aya AG, Vialles N, Mangin R, Robert C, Ferrer JM, Ripart J, de La Coussaye JE. Chronobiology of labour pain perception: an Dr. Shafer is codeveloper of PLT Tools and has a financial observational study. Br J Anaesth 2004;93:451–3 interest in the software.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 985 The Inﬂuence of Time of Day of Administration on Duration of Opioid Labor Analgesia

Barbara M. Scavone, MD, Robert J. McCarthy, PharmD, Cynthia A. Wong, MD, and John T. Sullivan, MD

BACKGROUND: Medications administered into the epidural or intrathecal space for labor analgesia may demonstrate variable effects dependent on time of day, and this may affect clinical research trials investigating the pharmacology of specific drugs. In this retrospective study, we evaluated the effect of time of day of administration of intrathecal fentanyl and systemic hydromorphone labor analgesia from data collected as part of a randomized clinical trial examining the influence of analgesia method on labor outcome. METHODS: Six hundred ninety-two healthy parturients were randomized early in labor to receive combined spinal-epidural (intrathecal fentanyl 25 ␮g followed by a lidocaine and epinephrine containing epidural test dose) versus systemic (hydromorphone 1 mg IV and 1 mg IM) labor analgesia at first analgesia request. No further analgesics were administered until the patient requested additional analgesia (second analgesia request). Subjects were assigned to the daytime group (DAY) if initial analgesia (neuraxial or systemic) was administered between the hours of 07:01 and 23:00 and to the nighttime group (NIGHT) if it was administered between 23:01 and 07:00. Within each mode of analgesia study arm (neuraxial or systemic), the DAY and NIGHT groups were compared. The primary outcome variable was analgesia duration, defined as the time interval from administration of labor analgesia until the second analgesia request. Cervical dilation at first and second analgesia requests, pain score at first analgesia request, and average amount of pain between analgesia administration and second analgesia request were also compared between DAY and NIGHT groups. Rhythm analyses for duration of analgesia, cervical dilation, and pain scores were performed. RESULTS: There was no difference in the median duration of either neuraxial or systemic analgesia in DAY versus NIGHT subjects, and no harmonic variation was observed for analgesia duration. Rhythm analysis demonstrated a 24-h harmonic cycle for cervical dilation at first analgesia request with maximum values occurring near 17:00 and minimum values near 05:00, but the amplitude of the difference was very small. Rhythm analysis demonstrated a 24-h harmonic cycle with maximum values occurring near 22:00 and minimum values near 10:00 for the average amount of pain between analgesia administration and second analgesia request in neuraxial group patients, but amplitude was small. CONCLUSIONS: Time of day of administration did not seem to influence combined spinal- epidural or systemic labor analgesia duration under these study conditions. (Anesth Analg 2010;111:986–91)

hronobiology is the study of the effect of time, space for labor analgesia; specifically, differences in analgesia especially biologic rhythms, on living organisms; duration after administration of epidural ropivacaine and Cchronopharmacology is the study of the relation intrathecal sufentanil and fentanyl depending on time of day between time of administration and the effects of drugs. of administration.2–4 This has prompted concern regarding Chronopharmacologic effects can be either chronopharmaco- the impact of chronopharmacology on clinical research trials dynamic (time-dependent differences in organism re- investigating the pharmacology of specific drugs.1 However, sponse to drugs) or chronopharmacokinetic (time- these trials involved small numbers of subjects, especially dependent differences in pharmacokinetic parameters, during the night hours, and did not always control for such as absorption, distribution, and clearance).1 external factors that may indirectly influence duration of Several authors have noted time-dependent effects of analgesia (e.g., differences in labor unit staffing, visiting medications administered into the epidural or intrathecal hours, and sleep deprivation). The purpose of this retrospective study was to evaluate possible effects time of administration may have on intra- From the Department of Anesthesiology, Northwestern University Feinberg thecal fentanyl and systemic hydromorphone labor analge- School of Medicine, Chicago, Illinois. sia from data collected as part of a randomized clinical trial Accepted for publication September 7, 2009. examining the influence of analgesia method on labor Supported by Departmental funds. outcome.5 In the aforementioned study, duration of labor Cynthia A. Wong is Section Editor of Obstetric Anesthesiology for the Journal. This manuscript was handled by Steve Shafer, Editor-in-Chief, and analgesia was a secondary outcome, but the influence of Dr. Wong was not involved in any way with the editorial process or time of day was not examined. This study is a secondary decision. analysis of the previously obtained data, examining the Reprints will not be available from the author. influence of time of administration of analgesia on duration Address correspondence to Barbara M. Scavone, MD, 251 E. Huron St., F5-704, Chicago, IL 60611. Address e-mail to [email protected]. of labor analgesia. The null hypothesis for this analysis was Copyright © 2010 International Anesthesia Research Society that duration of labor analgesia would not vary with time DOI: 10.1213/ANE.0b013e3181c29390 of day.

986 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 METHODS DAY and NIGHT groups within each mode of analgesia The study was approved by the Northwestern University arm (neuraxial or systemic) were compared using the Mann- Office for the Protection of Research Subjects IRB. Details of Whitney U-test (age, height, weight, gestational age, time to the study design and protocol have been published.5 delivery, analgesia duration, cervical dilation, and pain scores) or Briefly, written informed consent was obtained from 884 ␹2 test (oxytocin administration and mode of delivery). The time healthy, opioid naïve, nulliparous women. Additional in- data were plotted against the data for duration of analgesia, as clusion criteria included uncomplicated, term, singleton well as cervical dilation and pain scores at first and second pregnancy with vertex position, and spontaneous labor or request and average pain score between analgesia interven- spontaneous rupture of membranes. Parturients were re- tions and evaluated using weighted least squares regression cruited shortly after admission to the Labor and Delivery (lowess), weighted quadratic least squares regression (loess), Unit by anesthesia personnel 24 h per day, 7 days per wk and nonparametric smoothing (the “super smoother” of over a 38-mo period commencing October 2000. Subjects Friedman).*6 Curve fitting trend lines were determined using were enrolled (n ϭ 750) in the study if the cervix was the R functions “lowess,” “loess,” and “supsmu”). To im- dilated Ͻ4 cm at the first request for analgesia and were prove smoothness at the beginning and end of the 24-h cycle, randomized to receive neuraxial versus systemic opioid data were concatenated from 25 h before and after the 24-h analgesia. At the first analgesia request, subjects in the cycle. Lowess and loess curves were estimated at a smoother neuraxial group received a combined spinal-epidural via a span of 0.15. Data were analyzed using R version 2.9.1 needle through needle technique. They received intrathecal (http://www.r-project.org). fentanyl 25 ␮g and an epidural test dose of lidocaine 45 mg Rhythm analyses for duration of analgesia, cervical dila- and epinephrine 15 ␮g in 3 mL total volume. Subjects in the tion, and pain scores were performed using a nonlinear systemic analgesia group received 1 mg IV and 1 mg IM WIN-ABPM-FIT program.† This analysis calculates the mesor hydromorphone. No further analgesics were administered (rhythm-adjusted mean), amplitude (half of the peak to until the patient requested additional analgesia (second trough of the rhythm-adjusted harmonic), and acrophase analgesia request). At the second analgesia request, a (time of the occurrence of the rhythm-adjusted harmonic). cervical examination was performed and the subject re- Data were evaluated for 24-, 12-, 8-, and 6-h harmonics. The ceived additional medications per study protocol. solver add in for Microsoft Excel 2007 was used to calculate Subjects rated peak contraction pain using a verbal rating the r2, Ϫ2 log likelihood, and P value for the harmonics scale for pain: a 0–10 scale in which 0 ϭ no pain and 10 ϭ worst identified. P Ͻ 0.05 was required to reject the null hypothesis. imaginable pain. Pain scores were obtained at the first analgesia request (pain score1st request). In addition, at the second anal- RESULTS gesia request, when the effects of analgesia were diminish- There were 366 patients assigned to the neuraxial analgesia ing, the patient was asked to rate the “average amount of arm and 362 to the systemic analgesia arm of the original pain” experienced between receiving intrathecal fentanyl study. Data from 343 of the neuraxial group subjects and or systemic hydromorphone and the second analgesia 349 of the systemic group subjects are included in the request (pain scoreaverage). present analysis (the remainder of subjects had protocol In this secondary analysis, subjects were assigned to the violations or missing data). Comparison of demographic daytime group (DAY) if initial analgesia (neuraxial or data between the DAY versus NIGHT groups is presented systemic) was administered between 07:01 and 23:00 h, and in Table 1. A significantly larger proportion of DAY sub- to the nighttime group (NIGHT) if it was administered jects than NIGHT subjects were receiving oxytocin at the between 23:01 and 07:00 h.2 Within each mode of analgesia time of analgesia in both the neuraxial and systemic arms of study arm (neuraxial or systemic), the DAY and NIGHT the study. DAY subjects in both analgesic arms of the study groups were compared. Duration of analgesia was the had a shorter analgesia to delivery time interval than primary outcome variable and was defined as the time NIGHT subjects. interval between the administration of intrathecal fentanyl Outcome data are presented in Table 2, and scatter plots or systemic hydromorphone and the second analgesia and smoother curves for duration of analgesia, cervical request. Age, height, weight, gestational age, oxytocin dilation, and pain scores are shown in Figures 1–4. There administration, time interval from analgesia administration was no difference in the median duration of either to delivery, infant weight, mode of delivery, cervical dila- neuraxial or systemic analgesia in DAY versus NIGHT tion, and pain scores were also compared in DAY versus subjects (Table 2). No harmonic variation was apparent for NIGHT group subjects. either neuraxial or systemic analgesia duration (Fig. 1). Previous studies examining the chronopharmacologic Despite a similar median value, there was a difference in effect of intrathecal opioids have demonstrated a 30-min distribution of cervical dilation at first analgesia request difference between day and night administration in the between the DAY and NIGHT subjects who received duration of analgesia after administration of intrathecal neuraxial analgesia. Subjects who received systemic anal- fentanyl 20 ␮g in 77 subjects,4 and a 23-min difference after gesia did not demonstrate the same variation. Rhythm administration of intrathecal sufentanil 10 ␮g in 91 sub- analysis of all subjects demonstrated a 24-h harmonic cycle jects.3 A sample of 345 subjects would achieve 96% power for cervical dilation at first analgesia request (Figs. 2 and 5). to detect a difference of 7 min between day and night duration of analgesia assuming an average duration of 90 *Friedman JH. A variable span smoother: technical report LC55. Palo Alto, ␣ CA: Department of Statistics, Stanford University, 1984. min with an estimated standard deviation of 25 min at an †Available at: http://www.ma.uni-heidelberg.de/inst/phar/forschungLemmer. equal to 0.05 using a 2-sided Mann-Whitney test. html. 2004. Accessed February 8, 2008.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 987 Opioid Labor Analgesia and Time of Day

Table 1. Subject Demographic Characteristics Neuraxial analgesia Systemic analgesia Time of analgesia administration Time of analgesia administration 07:01–23:00 h 23:01–07:00 hP 07:01–23:00 h 23:01–07:00 h P Age (yr) 32 (29–35) 31 (28–35) 0.34 31 (28–35) 32 (30–35) 0.22 Height (cm) 165 (160–170) 163 (160–169) 0.58 165 (160–170) 163 (160–170) 0.61 Weight (kg) 77 (69–88) 77 (70–85) 0.78 79 (71–88) 76 (69–87) 0.09 Gestational age (wk) 39 (38–40) 40 (39–40) 0.37 40 (39–40) 39.5 (39–40) 0.63 Oxytocin administered at 180/227 (79) 66/110 (60) Ͻ0.001 196/240 (82) 57/99 (58) Ͻ0.001 time of ﬁrst analgesia, n (%) Time to delivery (min) 419 (295–562) 491 (314–651) 0.03 505 (345–653) 559 (276–575) 0.03 Infant weight (g) 3490 (3168–3768) 3415 (3136–3679) 0.42 3410 (3120–3665) 3450 (3185–3732) 0.17 Vaginal delivery (%) 83 82 0.89 80 76 0.40

Table 2. Outcome Variables Neuraxial analgesia Systemic analgesia Time of analgesia administration Time of analgesia administration 07:01–23:00 h 23:01–07:00 hP 07:01–23:00 h 23:01–07:00 h P Duration of analgesia (min) 95 (73–117) 92 (75–125) 0.36 108 (80–143) 105 (80–145) 0.85 Cervical dilation at 1st analgesia 2 (1.5–3) 2 (1–3) 0.001 2 (1–3) 2 (1–3) 0.25 request (cm) Distribution, n (%) Յ1.5 cm 65 (27) 48 (39) 97 (38) 55 (50) Ͼ1.5 to Ͻ3.0 cm 80 (33) 50 (49) 83 (33) 28 (26) Ն3.0 cm 97 (40) 26 (21) 73 (29) 26 (24)

Pain score1st request 8 (7–9) 8 (7–9) 0.02 8 (7–9) 8 (7–9) 0.36 Distribution, n (%) 0–3 3 (2) 2 (2) 7 (3) 5 (5) 4–7 93 (39) 32 (26) 92 (38) 34 (31) 8–10 140 (59) 88 (72) 142 (59) 69 (64) Cervical dilation at 2nd 4 (3–5) 4 (3–5) 0.05 4 (3–5) 4 (2–5) 0.36 analgesia request (cm)

Pain scoreaverage 2 (1–3) 2 (0–3) 0.33 6 (4–7) 5 (4–7) 0.11

There was no difference observed between DAY and minimal. We also observed harmonic variations in the NIGHT subjects regarding cervical dilation at second anal- average amount of pain experienced between intrathecal gesia request, and no harmonic variation was observed for fentanyl administration and second analgesia request, this variable (Fig. 3). again, with small amplitude and r2 value.

The distribution of pain score1st request was different Biologic rhythms are both genetically determined and during the night period compared with the day in subjects influenced by environmental factors known as synchroniz- who received neuraxial analgesia despite a similar median ers, such as the light-dark cycle, eating, and fasting periods. value. Subjects who received systemic analgesia did not In humans, the main circadian (referring to a 24-h cycle) demonstrate a difference. Pain score1st request did not exhibit pacemaker is located in the hypothalamus. It is modulated any harmonic variation (Fig. 2). Median pain scoreaverage did by the external environment, receiving inputs from the not differ between DAY and NIGHT subjects in either the retina, as well as nonphotic synchronizers, such as those neuraxial or systemic arms of the study; however, a 24-h involved with locomotor activity and eating. Certain drugs harmonic variation was observed for pain scoreaverage in the also influence the pacemaker. The output from this pace- neuraxial group (Figs. 4 and 5); no harmonic variation was maker in turn influences a myriad of biologic activities, observed among subjects in the systemic group (Fig. 4). such as melatonin production, cortisol secretion, adrenocorticotropin releasing hormone secretion, and sensitivity DISCUSSION of end organs to adrenocorticotropin releasing hormone. Time of day of administration of analgesia did not seem to Organs such as the heart and liver, and even isolated cells, influence combined spinal-epidural or systemic labor anal- demonstrate their own circadian rhythms independent of gesia duration under the conditions of this study. We did the central circadian pacemaker.1 observe a harmonic cycle in the cervical dilation at which It can be difficult to separate true time of day effects labor analgesia was first requested; however, the small from external factors that may influence circadian human amplitude associated with the harmonic variation and the behaviors. Our study subjects demonstrated a minimum small r2 value imply that any clinical implications would be cervical dilation at first analgesia request in the very early

988 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 2. Cervical dilation (top panel) and pain score at ﬁrst request for analgesia (bottom panel) versus time of administration of analgesia. Black horizontal line ϭ mean cervical dilation/pain score; blue Figure 1. Analgesia duration versus the time of administration of line ϭ lowess smoother; green line ϭ loess smoother; and red analgesia (top panel ϭ neuraxial group; bottom panel ϭ systemic line ϭ supersmoother. Diamonds denote the maximum and mini- group). Black horizontal line ϭ mean analgesic duration; blue line ϭ mum values for each smoother of the same color. lowess smoother; green line ϭ loess smoother; and red line ϭ supersmoother. Diamonds denote the maximum and minimum values for each smoother of the same color. design does not distinguish between true time of day explanations for this versus external factors that influence morning hours, perhaps indicating increased early labor human behavior. pain perception during that time interval. Also, patients in To our knowledge, there have been no previous studies the neuraxial arm of the study had higher pain scores at examining the effect of time of administration on systemic first analgesia request between the hours of 23:01 and 07:00 opioid labor analgesia. Several authors have demonstrated than between 07:01 and 23:00. These findings are consistent time-related effects for neuraxial labor analgesia. Epidural with a previous study in which nulliparous women in ropivacaine had a longer duration between 07:00 and 13:00 spontaneous labor with cervical dilation between 3 and 5 h compared with other times of the day.2 Interestingly, this cm and ruptured membranes reported lower pain scores is the same time interval associated with lowest labor pain between 07:00 and 13:00 h than at other times of the day.7 scores.7 Intrathecal opioids given for labor analgesia also This circadian rhythmicity to pain perception could be exhibited chronobiologic characteristics. The duration of either attributable to true time of day effects or influenced intrathecal fentanyl 20 ␮g was longer between 12:00 and by external factors. For instance, it could be explained by 18:00 h than between 20:00 and 02:00 h.4 The duration of the circadian rhythm associated with ␤-endorphin secre- intrathecal sufentanil 10 ␮g followed a 12-h cycle, with tion (highest in the morning).8 However, it also could be peaks at 12:00 and 00:00 h.3 explained by the negative effects that factors such as sleep In contrast to these previous studies, we did not observe deprivation may have on pain perception. Alternatively, any harmonic cycle in the duration of labor analgesia. One subjects who experience labor pain during the day may be possible explanation for the difference in our results from more likely to go to the hospital when they are in less previous studies is that our patients were recruited in early distress than patients who experience that pain at night. labor, with a median cervical dilation of 2 cm, whereas Although we did not observe any time-related differences those in other studies were in more advanced labor. It is in analgesia duration, we did find some harmonic variation possible that chronopharmacologic analgesic effects are in effectiveness of intrathecal opioid analgesia as measured more likely to be observed as labor progresses and pain by pain scoreaverage, which showed a peak in the late worsens. Another explanation may lie in the fact that the evening and a nadir in the morning. Once again, our study analgesic doses differed. Not only did subjects in this study

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 989 Opioid Labor Analgesia and Time of Day

Figure 4. Average pain score between ﬁrst and second analgesia Figure 3. Cervical dilation at second request for analgesia (top administration (top panel ϭ neuraxial group; bottom panel ϭ sys- panel ϭ neuraxial group; bottom panel ϭ systemic group). Black temic group). Black horizontal line ϭ mean pain score; blue line ϭ horizontal line ϭ mean cervical dilation; blue line ϭ lowess lowess smoother; green line ϭ loess smoother; and red line ϭ smoother; green line ϭ loess smoother; and red line ϭ super- supersmoother. Diamonds denote the maximum and minimum smoother. Diamonds denote the maximum and minimum values for values for each smoother of the same color. each smoother of the same color.

Finally, none of the studies, including ours, controlled for other factors that may influence pain and analgesia, receive a relatively high dose of opioid but they also including presence or absence of medical personnel, lay received a lidocaine/epinephrine epidural test dose. No support people or family members, room lighting, sleep one has studied the time of day effect of a neuraxial mixture deprivation before enrollment in the study, time since last of opioid and local anesthetic. It is possible that pharmaco- sleep, sleep during the study period, the subjects’ normal logic rhythms are less likely to be observed with higher activity and sleep cycles, and other unknown factors. We doses of drugs or mixtures of drugs. Our results are manage a high volume of patients (approximately 10,000 consistent with a study examining chronic neuropathic deliveries per year) on our delivery unit, and therefore, the pain, which demonstrated chronobiology to pain percep- light and noise levels at night may differ from those of 9 tion but not to analgesic efficacy. smaller units. It would be desirable to control variables The number of subjects in previous studies was small such as these in a study specifically investigating the ϭ (largest n 194) compared with this study. Additionally, in influence of chronobiology on labor pain and analgesia. previous studies, explaining the study objective to the There are other limitations to this study. Variables such subject may have influenced the subject’s perception of as the administration of oxytocin were not controlled analgesic efficacy. It is well established that a significant resulting in a larger fraction of subjects during the day placebo effect can be measured in studies involving anal- receiving oxytocin at the initiation of analgesia. It is pos- gesia administration.10 Pain scores in the previous studies sible that both nurses and obstetricians are less likely to were assessed at regular intervals after the initial analgesic augment labor at night because of inconvenience or other dose, and the presence of an investigator assessing pain factors; alternatively, patients who come to the hospital at may have influenced when the subject requested additional night may be a class of patients in more active labor (in analgesia. Subjects in previous studies were recruited over more pain even at low cervical dilations) and therefore in a relatively brief interval (months) compared with our less need of oxytocin augmentation. It is possible that the study (years), and it is possible that this may also have increased use of oxytocin during the day masked time- influenced results. related effects that would have been evident had this

990 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Our aim in this analysis was not to study the chronopharmacology of systemic hydromorphone and intrathecal fentanyl per se, but to determine whether time-related effects confounded our original study results, as some authors have suggested they might. Time-related variables do not seem to have played a major role in our previous study. There were no time-related differences in analgesia duration, and time-related differences that we did observe, such as differences in cervical dilation and pain scores, were minimal. Furthermore, our study design does not allow us to draw conclusions regarding the cause for these observed differences; they may be related to true chronobiology or to environmental factors that influence human behavior. Chronobiology as a possible confounding factor in labor analgesia trials requires further examination.

ACKNOWLEDGMENTS The authors thank an anonymous reviewer for his insight into the analysis and presentation of the data in this article. REFERENCES 1. Chassard D, Bruguerolle B. Chronobiology and anesthesia. Anesthesiology 2004;100:413–27 2. Debon R, Chassard D, Duflo F, Boselli E, Bryssine B, Allaouchiche B. Chronobiology of epidural ropivacaine: variations in the duration of action related to the hour of administration. Anesthesiology 2002;96:542–5 3. Debon R, Boselli E, Guyot R, Allaouchiche B, Lemmer B, Chassard D. Chronopharmacology of intrathecal sufentanil for labor analgesia: daily variations in duration of action. Anesthesiology 2004;101:978–82 4. Pan PH, Lee S, Harris L. Chronobiology of subarachnoid fentanyl for labor analgesia. Anesthesiology 2005;103:595–9 5. Wong CA, Scavone BM, Peaceman AM, McCarthy RJ, Sullivan Figure 5. Harmonic cycles (a) identified by fitting the equation (b). JT, Diaz NT, Yaghmour E, Marcus RJ, Sherwani SS, Sproviero ϭ Upper panel: pain scoreaverage, neuraxial group. Orange line MT, Yilmaz M, Patel R, Robles C, Grouper S. The risk of ϭ ϭ predicted values from function with periodi 24 h, mesor 2, cesarean delivery with neuraxial analgesia given early versus ϭ ϭ 2 ϭ ϭ amplitudei 0.6, acrophasei 21.3 h (r 0.012, P 0.005), late in labor. N Engl J Med 2005;352:655–65 with maximum values occurring near 22:00 h and minimum values 6. Cleveland WS, Devlin SJ. Locally weighted regression: an ap- near 10:00 h. Lower panel: cervical dilation at first request for proach to regression analysis by local fitting. JASA 1988;83:596–610 ϭ analgesia. Orange line predicted values from function with pe- 7. Aya AG, Vialles N, Mangin R, Robert C, Ferrer JM, Ripart J, de ϭ ϭ ϭ ϭ riodi 24 h, mesor 2 cm, amplitudei 0.1 cm, acrophasei 16.4 h La Coussaye JE. Chronobiology of labour pain perception: an 2 ϭ ϭ (r 0.011, P 0.02), with maximum values occurring near 17:00 h, observational study. Br J Anaesth 2004;93:451–3 and minimum values near 05:00 h. 8. Lindow SW, Newham A, Hendricks MS, Thompson JW, van der Spuy ZM. The 24-hour rhythm of oxytocin and beta-endorphin secretion in human pregnancy. Clin Endocrinol (Oxf) 1996;45:443–6 variable been controlled. In addition, more subjects were 9. Odrcich M, Bailey JM, Cahill CM, Gilron I. Chronobiological recruited during the day than at night, and this may have characteristics of painful diabetic neuropathy and postherpetic confounded results. Patients at night may be different than neuralgia: diurnal pain variation and effects of analgesic therapy. Pain 2006;120:207–12 patients during the day (for instance, a group in more pain 10. Zubieta JK, Yau WY, Scott DJ, Stohler CS. Belief or need? and therefore more “motivated” to travel to the hospital at Accounting for individual variations in the neurochemistry of night). the placebo effect. Brain Behav Immun 2006;20:15–26

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 991 CASE REPORT Epidural Hematoma Nine Days After Removal of a Labor Epidural Catheter

Patrick J. Guffey, MD, Warren R. McKay, MD, and Rachel Eshima McKay, MD

Timely recognition and surgical decompression are crucial to minimize risk of permanent neurologic deficit from epidural hematoma. We present the case of a patient who developed acute back pain, sensory deficit, and ascending weakness 9 days after removal of a labor epidural catheter. Magnetic resonance imaging revealed a heterogeneous fluid collection extending from C6-7 through the lumbar region, with cord deformity at T9-11. Decompression laminectomy was performed within 4 hours of symptom onset. Twelve hours later, her motor function had fully recovered. Subsequent anatomic and hematologic workup was inconclusive. This presentation is atypical given the delayed presentation of symptoms after epidural placement. (Anesth Analg 2010;111:992–5)

pidural hematoma related to neuraxial anesthesia is was effective, and the remainder of labor and delivery (5 a rare but potentially devastating complication; pub- hours) progressed without complication. Shortly after de- Elished reports in patients undergoing epidural cath- livery, the patient developed profuse vaginal bleeding and eter placement estimate an incidence of about 1 case in was taken to the operating room for cervical dilation and 150,000 anesthetics.1–3 Most often, symptoms arise within a curettage; surgical anesthesia was successfully adminis- few hours after placement or removal of the catheter.3 tered via the in situ epidural catheter. The cause of the Cases of spontaneous epidural hematoma occurring in bleeding was not clearly attributed to retained placental pregnancy are exceedingly rare; only 6 have been re- fragments or cervical tear. Total estimated blood loss was ported in the English-language literature, only 1 of which 750 mL, and 2.5 L lactated Ringer solution was adminis- occurred after delivery.3–7 We found no previous case tered. After the procedure, the epidural catheter was re- reports of epidural hematoma presenting more than 1 moved, and the patient was discharged home 2 days later week after catheter removal. We report a case of an in excellent condition without complaints. epidural hematoma in a previously healthy woman who Nine days later, she presented to the emergency depart- presented 9 days after labor and vaginal delivery with ment with severe low back pain and acute, ascending lower epidural analgesia. Written informed consent was ob- extremity paralysis that began 2 hours before arrival. tained from the patient before preparation and submis- Immediate evaluation by obstetric and anesthesia physi- sion of the manuscript. cians revealed arterial blood pressure of 190/100 mm Hg and lower extremity paralysis and sensory deficit ascending to the umbilicus. Laboratory values included hemato- ϫ 9 ␮ CASE DESCRIPTION crit 30%, platelet count 358 10 / L, prothrombin time A 32-year-old, G1P0 woman with body mass index of 27 (PT) 14.0 seconds (normal, 12.5–16.0 seconds; international kg/m2 underwent epidural catheter placement during la- normalized ratio, 1.0), and fibrinogen 236 mg/dL (normal, bor. Localization of the epidural space required 2 passes in 202–430 mg/dL). Dexamethasone 12 mg was administered the midline of the L2-3 interspace with an 18-gauge Tuohy IV and urgent neurosurgical consultation was requested. needle. After the epidural space was identified by loss of Magnetic resonance imaging revealed a large, cylindrical resistance to saline at 7 cm from the skin, a flexible, heterogeneous fluid collection around the spine extending wire-embedded catheter (FlexTip Plus®; Arrow Interna- from C6-7 through the lumbar region, with significant mass tional, Reading, PA) was passed without resistance, pares- effect, cord deformity, and T2 signal prolongation at T9-11; thesia, or discomfort to a distance 6 cm beyond the tip of the majority of the blood had collected between T6 and T12 the needle. The catheter was secured at 13 cm at the skin. (Fig. 1). Within 2 hours of her arrival at the hospital, she Aspiration through the catheter with a 3-mL syringe failed was taken to the operating room to undergo decompression to yield blood or fluid. Lidocaine 45 mg and epinephrine 15 surgery. Baseline laboratory values revealed a hematocrit ␮g were administered without change in heart rate or of 28%, and platelet count, PT, and partial thromboplastin initiation of motor or sensory loss. Subsequent analgesia time within normal limits. Neither thromboelastography nor platelet function assay was available in our institution.

From the Department of Anesthesia and Perioperative Care, University of Because of ongoing blood loss and oozing in the field, 1 California San Francisco, San Francisco, California. pheresis pack of platelets and 4 U fresh frozen plasma were Accepted for publication June 3, 2010. administered, resulting in satisfactory hemostasis. Three Disclosure: The authors report no conflicts of interest. units of packed red blood cells were administered during Reprints will not be available from the author. surgery, and at the conclusion of the 4-hour case, blood loss Address correspondence to Rachel Eshima McKay, MD, Department of was estimated to be 1.5 L. The paralysis resolved com- Anesthesia and Perioperative Care, University of California San Francisco, pletely over the next 12 hours, and the sensory examination 521 Parnassus Ave., C450, San Francisco, CA 94143-0648. Address e-mail to [email protected]. returned to normal over the next week with the exception Copyright © 2010 International Anesthesia Research Society of an area of paresthesia along the medial aspect of her DOI: 10.1213/ANE.0b013e3181effd8f lower right medial leg.

992 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Hematoma After Labor Epidural Removal

Figure 1. A, Sagittal T1-weighted magnetic resonance image showing presence of extensive hematoma around the spinal cord, visible here from C7 to T12 and below. Blood here is visible anterior to the cord, and compression is visible most prominently between T9 to T12. B, Axial T2-weighted magnetic resonance image showing hematoma surrounding the spinal cord at the level of T12. C, Sagittal T2-weighted magnetic resonance image demonstrating extension of blood throughout the thoracic and lumbar region (T12 to sacrum shown).

Neuroangiographic evaluation for spinal arteriovascular malformation was negative. A subsequent detailed history, performed in conjunction with the hematology service, did not reveal personal or family history of abnormal bleeding. The patient was noted to have used ibuprofen 800 mg every 8 hours for 9 days after delivery for low back and generalized musculoskeletal pain. Further laboratory workup conducted 2 weeks after hematoma evacuation included a ristocetin cofactor assay of 88% (reference range, 42%–200%), fibrinogen of 301 mg/dL (normal range, 202–430 mg/dL), and factor VIII activity of 162% (normal, 43%–168%). These values were all within 2 standard deviations of mean values at our institution for a patient who is not pregnant. At 2-month follow-up, the patient had made a complete neurologic recovery.

DISCUSSION Spinal epidural hematoma is a rare but potentially devastating event, classified as spontaneous (occurring without Figure 2. Graph showing hours between initiation of neuraxial apparent cause or with delayed onset after minor injury), or procedure (lumbar puncture or spinal/epidural anesthesia) and secondary to identifiable cause, including spinal or epi- symptoms of epidural hematoma. Of the 63 cases of epidural dural anesthesia. Patients at higher risk for hematoma after hematoma associated with spinal or epidural anesthesia among 613 reported, 49 had documented evidence conﬁrming or refuting epidural anesthesia include those with advanced age, spi- presence of antiplatelet or anticoagulant medication use, coagulopa- nal stenosis, coagulopathy (longstanding or acute), platelet thy, or platelet abnormality. (The graph was constructed from data inhibition, arteriovenous malformation, or multiple punc- from Kreppel et al.3) tures.8 Overall, hematomas related to neuraxial anesthesia seem to be less frequent in the obstetrical compared with the elderly surgical population.2 There are also rare reports 49 had confirmed coagulopathy or were receiving concur- of spontaneous epidural hematoma in parturients that rent anticoagulation therapy; and 41 of these 49 patients occur in the absence of neuraxial block, and delay in developed symptoms within 72 hours of the neuraxial recognition and management has resulted in devastating procedure (Fig. 2).3 Five pregnant patients were among this consequences.2,9 This particular case was unusual because group of 63; 1 had an elevated PT from hepatic disease it occurred more than 1 week after epidural catheter related to the pregnancy, another was classified as hyper- removal, beyond the timeframe previously described after tensive, and the remaining 3 had no identifiable risk an inciting event.10 factors. A systematic review of 613 reported cases of spinal and What may have put our patient at risk? The presence of epidural hematoma demonstrates that trauma, neuraxial postpartum hemorrhage may have been related to numer- anesthesia, coagulopathy, arteriovenous malformation, or a ous factors other than coagulation disorder, but could also combination of these factors, are frequently associated with have been related to a subtle disorder of coagulation that hematoma; 30% were found to have no identifiable cause.3 was not detected during her workup after surgery.11,12 In all, 63 cases were associated with neuraxial anesthesia. Although the values of her PT, fibrinogen, and subsequent Of these 63 cases, coagulation status was known in 49; 34 of ristocetin cofactor assay and factor VIII activity were within

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 993 CASE REPORT the normal range for a nonpregnant patient, these may and exists within a low-pressure environment, any pres- have been falsely normal given the typical changes in sure exerted proximally because of a Valsalva maneuver, coagulation state that accompany pregnancy and the early vomiting, or mechanical compression of the vena cava can postpartum period, when circulating levels of clotting dramatically increase the transmural venous pressure, leav- factors may be 20% to 200% above nonpregnant levels and ing the veins vulnerable to injury. It has been postulated do not fully return to nonpregnancy baseline values until 8 that spontaneous epidural hematoma may occur from weeks postpartum.13 preexisting faults in the venous wall exacerbated by well- Additionally, we questioned whether the hematoma described mechanical and hyperdynamic conditions exist- could have been related to the patient’s nonsteroidal anti- ing in the peripartum period.4–7,18 inflammatory drug (NSAID) use, or to NSAIDs in combi- Numerous studies have demonstrated the most favor- nation with a subtle impairment in coagulation function. able outcomes when a symptomatic hematoma is decom- Antiplatelet drugs, including nonselective NSAIDs and pressed within 36 hours, and some authors suggest even aspirin, taken within 1 week of surgery, have been associ- faster intervention, within 6 hours.19,20 In this case, inter- ated with increased risk of hematoma after craniotomy and vention occurred within 4 hours of the onset of symptoms. hip arthroplasty.14,15 However, there is consensus among Through emergent, definitive treatment, the patient made a experts that NSAID use does not increase risk of epidural full recovery, illustrating the importance of prompt recog- hematoma. This consensus is based on prospective studies nition and treatment of epidural hematoma to maximize of hundreds of NSAID users undergoing neuraxial anes- the patient’s chance of a favorable outcome. thesia8 and epidural steroid injection.11 Use of thromboelastography or other platelet function assays at the time AUTHOR CONTRIBUTIONS of presentation might have established whether NSAID use PJG helped with patient consent, data collection, and manu- contributed to the development of this epidural hematoma. script preparation; WRM helped with manuscript preparation; and REM helped with data collection, manuscript preparation, A more thorough hematologic evaluation, performed fur- and construction of figures. ther out from the peripartum period, would be needed to exclude an underlying bleeding disorder; to date, the REFERENCES patient has declined further workup. 1. Loo CC, Dahlgren G, Irestedt L. Neurological complications in In either case, it is possible that inadequate hemostasis obstetric regional anaesthesia. Int Obstet Anesth 2000;9:99–124 2. Kopp SL, Horlocker TT. Anticoagulation in pregnancy and caused gradual expansion of a small collection of blood that neuraxial blocks. Anesthesiol Clin 2008;26:1–22 was initiated by epidural placement or removal. Over days, 3. Kreppel D, Antoniadis G, Seeling W. Spinal hematoma: a this small asymptomatic hematoma may have slowly con- literature survey with meta-analysis of 613 patients. Neurosurg tinued to bleed and expand, eventually reaching a critical Rev 2003;26:1–49 size sufficient to cause dramatic symptoms. The predomi- 4. Bidzinski J. Spontaneous spinal epidural hematoma during pregnancy: case report. J Neurosurg 1966;24:1017 nantly thoracic location of the hemorrhage, higher than the 5. Yonekawa Y, Mehdorn HM, Nishikawa M. Spontaneous spinal site of epidural puncture, does not argue against the epidural hematoma during pregnancy. Surg Neurol 1975;3: epidural catheter as the primary cause. First, the catheter 327–8 tip may have extended a considerable distance in a cepha- 6. Carroll SG. Spontaneous spinal extradural hematoma during pregnancy. J Matern Fetal Med 1997;6:218–9 lad direction, and catheter removal may have initiated 7. Bose S, Ali Z, Rath P, Prabhakar H. Spontaneous spinal trauma at that location. Second, blood from a hematoma haematoma: a rare cause of quadriplegia in the post-partum secondary to needle trauma in the lumbar region may have period. Br J Anaesth 2007;99:855–7 collected preferentially in the thoracic epidural space be- 8. Horlocker TT, Wedel DJ, Schroeder DR, Rose SH, Elliot BA, cause of anatomic factors such as the relatively narrower McGregor DG, Wong GY. Preoperative anti-platelet therapy does not increase the risk of spinal hematoma associated with dimension of the spinal cord, convexity of the thoracic regional anesthesia. Anesth Analg 1995;80:303–9 spine, the relatively lower intrathoracic pressure, and 9. Doblar DD, Schumacher SD. Spontaneous acute thoracic epi- greater capacity of the thoracic epidural space.16,17 dural hematoma causing paraplegia in a patient with severe Given the prolonged period of time between catheter preeclampsia in early labor. Int J Obstet Anesth 2005;14:256–60 10. Pear BL. Spinal epidural hematoma. Am J Roentgenol Radium manipulation and presentation of symptoms, it is possible Ther Nucl Med 1972;155:155–64 that this patient’s condition resulted from factors other than 11. Horlocker TT, Bajwa ZH, Ashraf Z, Khan S, Wilson JL, Sami N, epidural catheter placement. However, we think it more Peeters-Asdourian C, Powers CA, Schroeder DR, Decker PA, likely that the event was related to epidural catheter Warfield CA. Risk assessment of hemorrhagic complications placement/removal, with possible contributing factors in- associated with nonsteroidal anti-inflammatory medications in ambulatory pain clinic patients undergoing epidural steroid cluding elevated venous pressure related to the pregnancy, injection. Anesth Analg 2002;95:1691–7 platelet inhibition from persistent NSAID use, and hyper- 12. Kadir RA, Kingmam CEC, Chi C, Lee CA, Economides DL. Is tension. The relative contribution of these various factors, if primary postpartum haemorrhage a good predictor of inher- any, however, remains undetermined. ited bleeding disorders? Haemophilia 2007;13:178–81 13. Bremme KA. Haemostatic changes in pregnancy. Best Pract Pregnancy is characterized by a relatively hypercoagu- Res Clin Haematol 2003;16:153–68 lable state, effectively reducing the risk of epidural hema- 14. Palmer JD, Sparrow OC, Ianotti F. Postoperative hematoma: a toma. However, the epidural space houses an extensive 5-year survey and identification of risk factors. Neurosurgery venous system containing tributaries from the spinal cord 1994;35:1061–5 15. Robinson CM, Christie J, Malcolm-Smith N. Nonsteroidal and vertebral bodies that drain into the external vertebral anti-inflammatory drugs, perioperative blood loss, and trans- venous plexus and become more engorged during preg- fusion requirements in elective hip arthroplasty. J Arthroplasty nancy. Because this venous system does not contain valves, 1993;8:607–10

994 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Hematoma After Labor Epidural Removal

16. Visser WA, Liem TH, van Egmond J, Gielen MJ. Extension of 19. Groen RJ, van Alphen HA. Operative treatment of sponta- sensory blockade after thoracic epidural administration of a neous spinal epidural hematomas: a study of the factors test dose of lidocaine at three different levels. Anesth Analg determining postoperative outcome. Neurosurgery 1996;39: 1998;86:332–5 494–508 17. Igarashi T, Hirabayashi Y, Shimizu R, Saitoh K, Fukuda H. 20. Lawton MT, Porter RW, Heiserman JE, Jacobowitz R, Sonntag Thoracic and lumbar extradural structure examined by extra- VK, Dickman CA. Surgical management of spinal epidural duroscope. Br J Anaesth 1998;81:121–5 hematoma: relationship between surgical timing and neuro- 18. Beatty RM, Winston KR. Spontaneous cervical epidural logical outcome. J Neurosurg 1995;83:1–7 haematoma: a consideration of etiology. J Neurosurg 1984;61: 143–8

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 995 Society for Pediatric Anesthesia

Section Editor: Peter J. Davis

Ipsilateral Transversus Abdominis Plane Block Provides Effective Analgesia After Appendectomy in Children: A Randomized Controlled Trial

John Carney, MB,*† Olivia Finnerty, MB, FCARCSI,*† Jassim Rauf, MB,† Gerard Curley, MB, FCARCSI,*† John G. McDonnell, MB, FCARCSI,*† and John G. Laffey, MD, MA, BSc, FCARCSI*†‡

BACKGROUND: The transversus abdominis plane (TAP) block provides effective postoperative analgesia in adults undergoing major abdominal surgery. Its efficacy in children remains unclear, with no randomized clinical trials in this population. In this study, we evaluated its analgesic efficacy over the first 48 postoperative hours after appendectomy performed through an open abdominal incision, in a randomized, controlled, double-blind clinical trial. METHODS: Forty children undergoing appendectomy were randomized to undergo unilateral TAP block with ropivacaine (n ϭ 19) versus placebo (n ϭ 21) in addition to standard postoperative analgesia comprising IV morphine analgesia and regular diclofenac and acetaminophen. All patients received a standard general anesthetic, and after induction of anesthesia, a TAP block was performed using the landmark technique with 2.5 mg ⅐ kgϪ1 ropivacaine 0.75% or an equal volume (0.3 mL ⅐ kgϪ1) of saline on the ipsilateral side to the incision. RESULTS: The TAP block with ropivacaine reduced mean (ϮSD) morphine requirements in the first 48 postoperative hours (10.3 Ϯ 12.7 vs 22.3 Ϯ 14.7 mg; P Ͻ 0.01) compared with placebo block. The TAP block also reduced postoperative visual analog scale pain scores at rest and on movement compared with placebo. Interval morphine consumption was reduced over the first 24 postoperative hours. There were no between-group differences in the incidence of sedation or nausea and vomiting. There were no complications attributable to the TAP block. CONCLUSIONS: Unilateral TAP block, as a component of a multimodal analgesic regimen, provided superior analgesia compared with placebo in the first 48 postoperative hours after appendectomy in children. (Anesth Analg 2010;111:998–1003)

ppendectomy is one of the most frequently per- plane (TAP) between the internal oblique and the transver- formed surgical procedures in children and is asso- sus abdominis muscles.5 Using anatomic studies, our group ciated with significant postoperative discomfort and identified the lumbar triangle of Petit as an access point for A 6 pain.1 Multimodal approaches to the provision of postopera- introducing local anesthetic drugs into the TAP. This tive analgesia often incorporate blockade of the abdominal triangle is bounded posteriorly by the latissimus dorsi wall, such as ilioinguinal blockade2 or wound infiltration.3 How- muscle, anteriorly by the external oblique, and inferiorly by 7 ever, the efficacy of these approaches is unclear.3,4 The optimal the iliac crest. The floor of the triangle, from superficial to approach to the blockade of the abdominal wall in children deep, is composed of subcutaneous tissue, and the fascial undergoing appendectomy remains to be determined. borders of the external oblique, the internal oblique, and the transversus abdominis muscles, respectively. An important component of the pain experienced by The efficacy of the TAP block in providing postoperative children after appendectomy derives from the abdominal analgesia has been shown in adults undergoing bowel wall incision. The nerves that supply the abdominal wall surgery,8 cesarean delivery,9 and total abdominal hysterec- course through the neurofascial transversus abdominis tomy.9 Most recently, an ultrasound-guided approach to the TAP block has been described in a series of 8 children From the *Department of Anaesthesia, Clinical Sciences Institute, National undergoing inguinal hernia repair10 and has shown anal- University of Ireland, Galway; and †Department of Anaesthesia and Inten- sive Care Medicine, and ‡Clinical Research Facility, Galway University gesic efficacy in a randomized controlled trial for appen- Hospitals, Galway, Ireland. dicectomy in adults.11 However, there are no randomized Accepted for publication May 26, 2010. controlled clinical trials demonstrating the efficacy of the Supplemental digital content is available for this article. Direct URL citations TAP block in children. We hypothesized that, compared appear in the printed text and are provided in the HTML and PDF versions with a placebo block, the TAP block, as part of a multimo- of this article on the journal’s Web site (www.anesthesia-analgesia.org). dal analgesic regimen, would result in decreased opioid Funded by departmental resources. consumption and improved analgesia in the first 48 hours Disclosure: The authors report no conflicts of interest. for children undergoing appendectomy. Reprints will not be available from the author. Address correspondence to John G. Laffey, MD, MA, BSc, FCARCSI, Department of Anaesthesia, Clinical Sciences Institute, National University of Ireland, Galway, Ireland. Address e-mail to [email protected]. METHODS Copyright © 2010 International Anesthesia Research Society After obtaining approval by the Hospital Ethics Committee, DOI: 10.1213/ANE.0b013e3181ee7bba and written informed parental consent, we studied children

998 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 younger than 16 years, ASA grade I to III, scheduled for After completion of the surgical procedure, patients emergency open appendectomy, in a randomized, double- were transferred to the postanesthesia care unit. All pa- Ϫ blind, controlled clinical trial. Patients were excluded if tients received oral acetaminophen 20 mg ⅐ kg 1 every 6 Ϫ there was a history of relevant drug allergy, if they were hours and rectal diclofenac 1 mg ⅐ kg 1 every 12 hours receiving medical therapies considered to result in toler- postoperatively. Children older than 8 years were given IV ance to opioids, or if they were deemed to be unable to morphine patient-controlled analgesia (PCA) (bolus dose ␮ ⅐ Ϫ1 independently assess their pain. 20 g kg ; lockout 6 minutes). Children younger than 8 years were given nurse-administered IV morphine (20 Patients were randomly allocated to undergo TAP block Ϫ Ϫ ␮ ⅐ 1 with 2.5 mg ⅐ kg 1 ropivacaine 0.75% (to a maximal dose of g kg bolus) on demand. The presence and severity of 150 mg) on the right side or TAP block with an equal pain, nausea, and sedation were assessed systematically by Ϫ volume (0.3 mL ⅐ kg 1) of 0.9% saline. The allocation sean investigator blinded to group allocation. These assess- quence was generated by a random number table, and ments were performed in the postanesthesia care unit and at 2, 4, 6, 12, 24, 36, and 48 hours after TAP blockade. All group allocation was concealed in sealed, opaque enve- patients were asked to give scores for their pain at rest and lopes, which were not opened until patient consent had on movement (knee flexion) and for the degree of nausea at been obtained. The patients, their anesthetists, and staff each time point. Pain severity was measured using a visual providing postoperative care were blinded to group assign- analog scale (VAS) with superimposed faces pain severity ment. All patients received a standardized rapid sequence scale (10-cm line in which 0 cm ϭ no pain and10 cm ϭ induction of anesthesia. After application of standard worst pain imaginable) and a categorical pain scoring monitoring and oxygen administration for 3 minutes, an- system (none ϭ 0, mild ϭ 1, moderate ϭ 2, and severe ϭ 3). ⅐ Ϫ1 esthesia was induced with propofol (2–3 mg kg ), cricoid Nausea was measured using a categorical scoring system pressure was applied and muscle relaxation was produced (none ϭ 0, mild ϭ 1, moderate ϭ 2, and severe ϭ 3). The ⅐ Ϫ1 with succinylcholine (1–1.5 mg kg ), and the trachea was patient was deemed to have been nauseated if they had a intubated. Anesthesia was maintained using 1 to 1.5 mini- nausea score Ͼ0 at any postoperative time point. Sedation mum alveolar concentration sevoflurane in oxygen and air. scores were assigned by the investigator using a sedation Ϫ1 All patients also received IV morphine 0.15 mg ⅐ kg , scale (awake and alert ϭ 0, quietly awake ϭ 1, asleep but Ϫ rectal diclofenac 1 mg ⅐ kg 1, and rectal acetaminophen 20 easily roused ϭ 2, and deep sleep ϭ 3). The patient was Ϫ mg ⅐ kg 1 immediately before surgical incision. Prophylac- deemed to have been sedated if they had a sedation score tic antiemetics were not administered. Ͼ0 at any postoperative time point. Rescue antiemetics The TAP block was performed after induction of anes- were offered to any patient who complained of nausea or thesia but before surgical incision by 1 of 2 investigators vomiting. The study ended 48 hours after TAP blockade. (JC, OF) using the following technique.8 A 22-gauge 25-mm The primary outcome measure in this study was 48-hour or 50-mm blunted regional anesthesia needle (Plexufix௡;B. morphine consumption. Secondary outcome measures in- Braun, Melsungen AG, Germany) was attached with flex- cluded time to first request for morphine, VAS scores, and ible tubing to a syringe filled with the study solution. With side effects associated with morphine consumption. A pilot the patient in a supine position and the investigator stand- study of children undergoing open appendectomy found a ing on the contralateral side, the iliac crest was palpated mean 48-hour morphine requirement of 24 mg, with a from anterior to posterior until the latissimus dorsi muscle standard deviation of 8 mg in the control group. We insertion was appreciated. The triangle of Petit was pal- intended to be able to detect a minimum 33% reduction in pated between the anterior border of latissimus dorsi morphine requirement in the patients receiving TAP block- muscle, the posterior border of the external oblique muscle, ade. Based on these projections, we calculated that at least and the iliac crest. The skin over the triangle of Petit was 17 patients would be required per group for an experimental design incorporating 2 groups, with ␣ ϭ 0.05 and ␤ ϭ pierced with the needle held at right angles to the coronal 0.2. We therefore planned to recruit 40 patients into the plane. The needle was stabilized and advanced at right study. angles to the skin in a coronal plane until resistance was Statistical analyses were performed using a standard encountered. This first resistance indicated that the needle statistical program (SigmaStat 3.5; Systat Software, San tip was abutting the fascial extension of the external Jose, CA). Demographic data were analyzed using Student oblique muscle. Further gentle advancement of the needle 2 t, ␹ , or Fisher exact tests as appropriate. The data were resulted in a loss of resistance, or “pop” sensation, as the tested for normality using the Kolmogorov-Smirnov nor- needle entered the plane between the external and internal mality test. Repeated measurements (pain scores, nausea oblique fascial layers. Further gentle advancement resulted scores) were analyzed by repeated-measures analysis of in the appreciation of a second increased resistance and its variance where normally distributed, with further paired loss indicated entry into the TAP. After careful aspiration to comparisons at each time interval performed using the t exclude vascular puncture, a test dose of 1 mL was injected. test. For non-normally distributed data, between-group The presence of substantial resistance to this injection comparisons at each time point were made using Wilcoxon indicates the needle is not between fascial planes and the rank sum test. Categorical data were analyzed using the ␹2 needle should be repositioned. The study solution (0.3 analysis or Fisher exact test. The time to first request for Ϫ Ϫ mL ⅐ kg 1 ropivacaine 7.5 mg ⅐ mL 1 or saline 0.9%) was morphine was analyzed using the log rank test. Normally injected through the needle in increments while observing distributed data are presented as mean Ϯ SD, non-normally closely for signs of toxicity. distributed data are presented as median (interquartile

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 999 Transversus Abdominis Plane Block in Children

undergo TAP blockade with normal saline. Groups were Table 1. Baseline Patient Characteristics comparable in terms of age, which ranged from 4 to 16 Group years in the TAP group and 5 to 16 years in the control Control TAP block group. There were no differences between the groups regarding weight and height, history of prior abdominal (19 ؍ n) (21 ؍ Characteristic (n Weight (kg) 47.5 Ϯ 18.6 37.7 Ϯ 15.2 surgery, method of postoperative morphine administra- Ϯ Ϯ Height (m) 1.5 0.2 1.4 0.2 tion, or surgical technique (Tables 1 and 2). A similar Body mass index (kg/m2) 20.2 Ϯ 4.9 18.7 Ϯ 3.2 Prior abdominal surgery ͓n (%)͔ 0 (0) 0 (0) number of patients in both groups had appendicitis, with Histologic diagnosis 14 (74%) in the TAP group versus 14 (67%) in the control Appendicitis 14 14 group having a histologic diagnosis of appendicitis with or Appendicitis with perforation 54without perforation (Table 1). Five patients in the TAP of appendix group and 4 patients in the control group had histologic TAP ϭ transversus abdominis plane. evidence of perforation with peritonitis. In all patients, the Continuous data are presented as mean Ϯ SD. Categorical variables are presented as number and proportion. triangle of Petit was located easily on palpation, the TAP There were no signiﬁcant differences between groups. was localized after 1 or 2 attempts, and the block performed without complication. Children undergoing unilateral TAP block with ropiva- Table 2. Postoperative Analgesia, Nausea, caine had reduced 48-hour morphine requirements (Fig. 1) and Sedation and increased time to first requirement for morphine (Fig. Group 2). The total amount of morphine required in the first 48 Ϯ Control TAP block postoperative hours was 10.3 12.7 mg in the TAP group compared with 22.3 Ϯ 14.7 mg (P Ͻ 0.01) in the control (19 ؍ n) (21 ؍ n) Patient-controlled morphine 17 14 group. The median (interquartile range) time to first re- analgesia quirement for morphine was 55 (30–300) minutes in chil- Nurse-controlled morphine 45dren who received a TAP block compared with 16 (7–30) analgesia minutes in the control group (P Ͻ 0.001). The TAP block Interval morphine requirement (␮g/kg) with ropivacaine reduced cumulative postoperative mor- 0–6 h 117 (60–203) 55 (10–150)* phine consumption compared with placebo block at all 6–12 h 40 (0–115) 0 (0–17)* time points (Fig. 1). Interval morphine consumption was 12–24 h 60 (19–175) 0 (0–159)* also significantly lower at 6, 12, and 24 hours but not at the 24–36 h 0 (0–40) 0 (0–0) 36–48 h 0 (0–0) 0 (0–0) later time points in the patients who had TAP blockade Postoperative nausea scores (Table 2). There were no differences within either the TAP 0–6 h 0 (0–0) 0 (0–0) or control group regarding the amounts of morphine 6–12 h 0 (0–0) 0 (0–0) required between patients in whom the morphine was 12–24 h 0 (0–0) 0 (0–0) nurse administered versus those who received morphine 24–36 h 0 (0–0) 0 (0–0) 36–48 h 0 (0–0) 0 (0–0) via PCA. In addition, morphine consumption within each Postoperative sedation scores group did not differ significantly depending on whether or 2 h 1 (1–1) 1 (0.5–1) not the patient had histologic evidence of appendicitis. 4 h 1 (0.25–1) 1 (0–1) Postoperative VAS pain scores at rest and on movement 6 h 1 (0–1) 1 (0–1) 12 h 1 (0–2) 1 (0–1) were reduced after TAP block at all time points assessed 24 h 1 (0–1) 0 (0–0) (Figs. 3 and 4). Categorical pain scores were significantly 36 h 0 (0–0) 0 (0–0) lower in patients who received the TAP block at some but 48 h 0 (0–0) 0 (0–0) not all postoperative time points (data not shown). Pain TAP ϭ transversus abdominis plane. severity was low in both groups after the first 24 hours, and The data are presented as medians (interquartile range). several patients were discharged in both groups between *P Յ 0.05 (control versus TAP block). 24 and 48 hours. There was no significant difference in the incidence of range), and categorical data are presented as raw data and nausea or distribution of nausea scores between groups at frequencies. The ␣ level for all analyses was set at P Ͻ 0.05, any time point (Table 2). The limited nausea experienced and the Bonferroni correction for multiple comparisons resulted in low scores in both groups; the median and was used where appropriate. interquartile scores are shown in Table 2. Six patients (27%) in the control group and 4 patients (16%) in the TAP RESULTS blockade group developed postoperative nausea. There Forty-two children participated in this study. No patients was no significant difference in the incidence of sedation or were excluded on the basis of the exclusion criteria (Ap- distribution of sedation scores between groups at any time pendix Figure; see Supplemental Digital Content 1, point (Table 2). http://links.lww.com/AA/A161). Two patients, 1 from each group, were excluded after enrollment because of postoperative analgesic protocol violations. Of the remain- DISCUSSION ing 40 patients, 19 were randomized to undergo TAP Open appendectomy is one of the most frequently per- blockade with ropivacaine and 21 were randomized to formed surgical procedures in the pediatric population

1000 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 1. A box plot of postoperative cumulative morphine consumption in each group in the ﬁrst 48 postoperative hours. The middle line in each box represents the median value, the outer margins of the box represent the interquartile range, and the whiskers represent the 10th and 90th centile for each time point. *Signiﬁcantly (P Ͻ 0.05, Wilcoxon rank sum test) higher morphine consumption compared with the transversus abdominis plane (TAP) block group.

Figure 2. A Kaplan-Meier graph of the proportion of patients in each group over time that did not require supplemental morphine (P ϭ 0.004, log rank test).

worldwide and is a cause of significant pain in the postop- to 24 hours after surgery. Thereafter, morphine require- erative period.1 The optimal analgesic regimen should ments were very low in both groups, with no morphine provide safe, effective analgesia, with minimal side effects consumed over the second 24 postoperative hours in any for the child. A multimodal analgesic regimen is most likely patient studied. The TAP block delayed the time to first to achieve these goals; however, the optimal components request for supplemental opioid analgesia and reduced remain to be determined. The TAP block provides blockade pain scores at rest and on movement. Because the surgery of nociception from the abdominal wall; however, there is was 1 sided, the dose of local anesthetic used was limited to Ϫ also nociceptive input from the abdominal organs and the 2.5 mg ⅐ kg 1 on the ipsilateral side, which is within the onset of the block is not immediate. Therefore, the block is recommended safe dose range. Alternative abdominal tech- used as part of a multimodal approach. This trial demon- niques, such as the ilioinguinal iliohypogastric nerve block, Ϫ strated that an ipsilateral TAP block provides effective have been performed in children at a dose of 5 mg ⅐ kg 1 analgesia in children undergoing open appendectomy. ropivacaine without central nervous system or systemic A TAP block on the side of the surgical incision reduced toxicity.12 The reduction in opioid use, coupled with the overall postoperative morphine requirements by approxi- reduction in postoperative pain, highlights the potential of mately 50% and the interval morphine requirements for up the TAP block in children undergoing appendectomy.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1001 Transversus Abdominis Plane Block in Children

Figure 3. Box plots of postoperative visual analog scale (VAS) pain scores at rest in each group over the ﬁrst 48 postoperative hours. The middle line in each box represents the median value, the outer margins of the box represent the interquartile range, and the whiskers represent the 10th and 90th centile for each time point. *Signiﬁcantly (P Ͻ 0.05, Wilcoxon rank sum test) higher VAS pain scores compared with the transversus abdominis plane (TAP) block group.

Figure 4. Box plots of postoperative visual analog scale (VAS) pain scores on movement in each group over the ﬁrst 48 postoperative hours. The middle line in each box represents the median value, the outer margins of the box represent the interquartile range, and the whiskers represent the 10th and 90th centile for each time point. *Signiﬁcantly (P Ͻ 0.05, Wilcoxon rank sum test) higher VAS pain scores compared with the transversus abdominis plane (TAP) block group.

These findings, when taken in context with the demon- has been shown to be relatively unreliable, producing strated efficacy of the TAP block in adults,6,8–10 attest that deposition of local anesthetic solution in close proximity to the TAP block may provide effective postoperative analge- the nerves in as few as 14% of children.4 This may be sia for a wide variety of abdominal procedures in both improved by the use of ultrasound to guide needle posi- children and adults. tion.4 Potentially serious complications, such as bowel We used the landmark-based technique for the TAP wall hematoma,13 have been reported after the use of block, which was performed without difficulty in the ilioinguinal/iliohypogastric blocks in children. In our ex- children in this study. Alternative approaches to the TAP perience, performing the TAP block in a pediatric popula- block using ultrasound guidance have recently been de- tion has been technically easier than in adults because the scribed in a case series of children undergoing inguinal degree of obesity is usually less than that of an adult hernia repair,10 in adults for appendicectomy,12 and an population, there is a lesser degree of muscle laxity, and the adult study for laparoscopic cholecystectomy.11 The opti- 1-in. Plexufix needle that is available allows easier percep- mal approach remains to be demonstrated. Other regional tion of the loss of resistance as described. Local infiltration anesthesia–based approaches for analgesia after appendec- of the appendectomy wound with local anesthetic drugs is tomy include ilioinguinal/iliohypogastric nerve blockade also widely practiced. However, the efficacy of this ap- and local wound infiltration. However, in the case of proach is also unclear, with studies reporting no demon- ilioinguinal/iliohypogastric block, the landmark technique strable reduction in morphine consumption compared with

1002 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA standard care.3 Indeed, the technique has demonstrated to REFERENCES be inconsistent with a lack of evidence after surgery on the 1. Hale DA, Molloy M, Pearl RH, Schutt DC, Jaques DP. Appen- abdominal wall apart from herniorrhaphy.14 dectomy: a contemporary appraisal. Ann Surg 1997;225:252–61 There are now a variety of techniques for the TAP block 2. Willschke H, Marhofer P, Bo¨senberg A, Johnston S, Wanzel O, Cox SG, Sitzwohl C, Kapral S. Ultrasonography for and the analgesic merit of each is being elucidated in ilioinguinal/iliohypogastric nerve blocks in children. Br J ongoing studies. Although it is possible to ultrasonically Anaesth 2005;95:226–30 visualize the 3 muscle layers of the abdominal wall, there is 3. Jensen SI, Andersen M, Nielsen J, Qvist N. Incisional local variation in these muscle layers that can restrict the use of anaesthesia versus placebo for pain relief after appendectomy ultrasound over the triangle of Petit.15 As a result, the in children: a double-blinded controlled randomised trial. Eur needle insertion point as described in the ultrasound stud- J Pediatr Surg 2004;14:410–3 4. Weintraud M, Marhofer P, Bo¨senberg A, Kapral S, Willschke ies, which is dependent on the adequate identification of H, Felfernig M, Kettner S. Ilioinguinal/iliohypogastric blocks the 3 muscle layers, can vary. This will alter the location of in children: where do we administer the local anesthetic the injectate as will the angle of the needle insertion to skin, without direct visualization? Anesth Analg 2008;106:89–93 which contrasts to the landmark approach’s description. 5. Netter FH. Back and spinal cord. In: Netter FH, ed. Atlas of Anteriorly placed injectate may not spread throughout the Human Anatomy. Summit, NJ: Ciba-Geigy Corporation, TAP and does not according to ongoing work at this 1989:145–55 6. McDonnell JG, O’Donnell BD, Farrell T, Gough N, Tuite D, institution. Although there is an ever-increasing access to Power C, Laffey JG. Transversus abdominis plane block: a ultrasound, it is far from universal and there is a continuing cadaveric and radiological evaluation. Reg Anesth Pain Med interest in landmark techniques. 2007;32:399–404 There are a number of limitations to this study. First, 7. Netter FH. Abdomen posterolateral abdominal wall. In: Netter there are difficulties in adequately blinding these types of FH, ed. Atlas of Human Anatomy. Summit, NJ: Ciba-Geigy studies because of the loss of sensation of the abdominal Corporation, 1989:230–40 wall. Although patients and the investigator conducting the 8. McDonnell JG, O’Donnell B, Curley G, Heffernan A, Power C, Laffey JG. The analgesic efficacy of transversus abdominis postoperative assessments were technically blinded to plane block after abdominal surgery: a prospective random- group allocation, true blinding may not have been possible. ized controlled trial. Anesth Analg 2007;104:193–7 Second, 2 methods of postoperative administration of mor- 9. Carney J, McDonnell JG, Ochana A, Bhinder R, Laffey JG. The phine were used: patients younger than 8 years received transversus abdominis plane block provides effective postop- nurse-administered morphine whereas children older than erative analgesia in patients undergoing total abdominal hys- 8 years received PCA morphine. However, there were no terectomy. Anesth Analg 2008;107:2056–60 10. Fredrickson M, Seal P, Houghton J. Early experience with the differences in the proportions of each method between the transversus abdominis plane block in children. Paediatr An- groups, so these differences do not seem to have contrib- aesth 2008;18:891–2 uted to the reported findings. Third, rates of appendicitis in 11. Niraj G, Searle A, Mathews M, Misra V, Baban M, Kiani S, our study (66%–74%) are somewhat lower than those ob- Wong M. Analgesic efficacy of ultrasound-guided transversus tained internationally.16 This likely reflects a lower threshold abdominis plane block in patients undergoing open appen- for operative intervention in this pediatric population, as dicectomy. Br J Anaesth 2009;103:601–5 12. Dalens B, Ecoffey C, Joly A, Giaufre E, Gustafsson U, Huledal evidenced by lower rates of appendiceal perforation in our 16 G, Larsson LE. Pharmacokinetics and analgesic effect of ropi- patients compared with those reported internationally. vacaine following ilioinguinal/iliohypogastric nerve block in Fourth, the study was not large enough to independently children. Paediatr Anaesth 2001;11:415–20 assess safety. There is a risk of inadvertent peritoneal 13. Frigon C, Mai R, Valois-Gomez T, Desparmet J. Bowel hema- puncture with this block as with other abdominal toma following an iliohypogastric-ilioinguinal nerve block. blocks.17,18 Although the incidence is not known, if the Paediatr Anaesth 2006;16:993–6 block is performed as described, the risk of peritoneal 14. Moiniche S, Mikkelsen S, Wetterslev J, Dahl JB. A qualitative systematic review of incisional local anaesthesia for postopera- puncture is likely to be low. This institution has not tive pain relief after abdominal operations. Br J Anaesth encountered complications relating to peritoneal puncture. 1998;81:377–83 Finally, a dose-response study has not yet been performed 15. Loukas M, Tubbs RS, El-Sedfy A, Jester A, Polepalli S, Kinsela to determine whether effective postoperative analgesia C, Wu S. The clinical anatomy of the triangle of Petit. Hernia could be produced with a lower dose of ropivacaine. 2007;11:441–4 In conclusion, the TAP block holds considerable promise 16. Newman K, Ponsky T, Kittle K, Dyk L, Throop C, Gieseker K, Sills M, Gilbert J. Appendicitis 2000: variability in practice, as part of a multimodal analgesic regimen for postappen- outcomes, and resource utilization at thirty pediatric hospitals. dectomy analgesia. With experience, the TAP block is easy J Pediatr Surg 2003;38:372–9 to perform, has provided reliable and effective analgesia in 17. Farooq M, Carey M. A case of liver trauma with a blunt this study, and no complications from the TAP block were regional anesthesia needle while performing transversus ab- detected. dominis plane block. Reg Anesth Pain Med 2008;33:274–5 18. Frigon C, Mai R, Valois-Gomez T, Desparmet J. Bowel hema- ACKNOWLEDGMENTS toma following an iliohypogastric-ilioinguinal nerve block. Paediatr Anaesth 2006;16:993–6 The authors thank the nursing staff on the pediatric ward for their assistance with this study.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1003 Dexmedetomidine Infusion for Analgesia and Prevention of Emergence Agitation in Children with Obstructive Sleep Apnea Syndrome Undergoing Tonsillectomy and Adenoidectomy

Anuradha Patel, MD, FRCA,* Melissa Davidson, MD,* Minh C. J. Tran, MD, MPH,* Huma Quraishi, MD,† Catherine Schoenberg, BSN,* Manasee Sant, MD,* Albert Lin, MD,* and Xiuru Sun, MS*

␣ BACKGROUND: Dexmedetomidine, a specific 2 agonist, has an analgesic-sparing effect and reduces emergence agitation. We compared an intraoperative dexmedetomidine infusion with bolus fentanyl to reduce perioperative opioid use and decrease emergence agitation in children with obstructive sleep apnea syndrome undergoing adenotonsillectomy (T&A). METHODS: One hundred twenty-two patients with obstructive sleep apnea syndrome undergoing T&A, ages 2 to 10 years, completed this prospective, randomized, U.S. Food and Drug Administration–approved study. After mask induction with sevoflurane, group D received IV dexmedetomidine 2 ␮g ⅐ kg–1 over 10 minutes, followed by 0.7 ␮g ⅐ kg–1 ⅐ h–1, and group F received IV fentanyl bolus 1 ␮g ⅐ kg–1. Anesthesia was maintained with sevoflurane, oxygen, and nitrous oxide. Fentanyl 0.5 to 1 ␮g ⅐ kg–1 was given to subjects in both groups for an increase in heart rate or systolic blood pressure 30% above preincision values that continued for 5 minutes. Observers in the postanesthesia care unit (PACU) were blinded to treatment groups. Pain was evaluated using the objective pain score in the PACU on arrival, at 5 minutes, at 15 minutes, then every 15 minutes for 120 minutes. Emergence agitation was evaluated at the same intervals by 2 scales: the Pediatric Anesthesia Emergence Delirium scale and a 5-point scale described by Cole. Morphine (0.05 to 0.1 mg ⅐ kg–1) was given for pain (score Ͼ4) or severe agitation (score 4 or 5) lasting more than 5 minutes. RESULTS: In group D, 9.8% patients needed intraoperative rescue fentanyl in comparison with 36% in group F (P ϭ 0.001). Mean systolic blood pressure and heart rate were significantly lower in group D (P Ͻ 0.05). Minimum alveolar concentration values were significantly different between the 2 groups (P ϭ 0.015). The median objective pain score was 3 for group D and 5 for group F (P ϭ 0.001). In group D, 10 (16.3%) patients required rescue morphine, in comparison with 29 (47.5%) in group F (P ϭ 0.002). The frequency of severe emergence agitation on arrival in the PACU was 18% in group D and 45.9% in group F (P ϭ 0.004); at 5 minutes and at 15 minutes, it was lower in group D (P ϭ 0.028). The duration of agitation on the Cole scale was statistically lower in group D (P ϭ 0.004). In group D, 18% of patients and 40.9% in group F had ϭ an episode of SPO2 below 95% (P 0.01). CONCLUSIONS: An intraoperative infusion of dexmedetomidine combined with inhalation anesthetics provided satisfactory intraoperative conditions for T&A without adverse hemodynamic effects. Postoperative opioid requirements were significantly reduced, and the incidence and duration of severe emergence agitation was lower with fewer patients having desaturation episodes. (Anesth Analg 2010;111:1004–10)

denotonsillectomy (T&A) is one of the most com- estimated to be 1%–3%.1 Postoperative pain can be severe mon surgical procedures performed in children. after T&A, and providing effective and safe perioperative AThe presence of obstructive symptoms is replacing analgesia in this group of patients is challenging. Not only recurrent tonsillitis as the primary indication for T&A. The are children with OSAS undergoing T&A at significant risk prevalence of obstructive sleep apnea syndrome (OSAS) in of respiratory and cardiovascular complications,2 they also children, also referred to as sleep disordered breathing,is have enhanced analgesic sensitivity to opiates and reduced morphine requirements after T&A.3 A high incidence of From the *Department of Anesthesiology and Perioperative Medicine and emergence agitation (EA) in patients having otolarygologic the †Department of Otolaryngology, University of Medicine and Dentistry procedures adds another challenge.4 Dexmedetomidine of New Jersey, New Jersey Medical School, Newark, New Jersey. (Dex) (Precedex, Hospira Worldwide, Lake Forest, Illinois), Accepted for publication June 10, 2010. a specific ␣ 2-adrenergic receptor agonist, has sedative, Financial support was provided by an Institutional Cost of Drug support 5 grant from Hospira Worldwide, Lake Forest, Illinois. anxiolytic, and analgesic properties and is very effective in 6,7 Address correspondence and reprint requests to Dr. Anuradha Patel, prevention of EA in children. An intraoperative infusion Associate Professor, Department of Anesthesiology and Perioperative Medi- of Dex used as a substitute for fentanyl has been shown to cine, New Jersey Medical School, University of Medicine and Dentistry of reduce opiate use in the postoperative period in adult New Jersey, Medical Science Building E-581, 185 South Orange Avenue, 8 Newark, NJ 07101. Address e-mail to [email protected]. patients undergoing bariatric surgery, but clinical data on Copyright © 2010 International Anesthesia Research Society the analgesic-sparing effect of Dex in children are conflict- DOI: 10.1213/ANE.0b013e3181ee82fa ing. The present study was performed to evaluate whether

1004 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 an intraoperative infusion of Dex combined with general holding area for HR and systolic blood pressure were used anesthesia would be a safe and effective substitute to as baseline. Both groups received fentanyl 0.5 to 1 ␮g ⅐ kg–1 opiates intraoperatively, reduce opiate requirements post- for an increase in HR or systolic NIBP 30% above the value operatively, and also be effective in reducing the incidence before start of surgery and sustained for 5 minutes. Lac- and severity of EA in children with OSAS undergoing tated Ringer’s solution 15 mL/kg was administered as a T&A. fluid bolus for a 30% decrease of systolic blood pressure from baseline, which continued for 2 readings and glyco- METHODS pyrrolate 0.01 mg ⅐ kg–1 for a 30% decrease in HR. Sevoflu- An Investigational New Drug number (76,041) was ob- rane was discontinued once hemostasis was achieved and ⅐ tained from the U.S. Food and Drug Administration. The muscle relaxation was reversed with neostigmine 0.05 mg –1 ⅐ –1 study was registered at www.clinicaltrials.gov (registration kg and glycopyrrolate 0.01 mg kg . The time to awak- number NCT00468052) and approved by the IRB of the ening (TA), defined as spontaneous eye opening or on University of Medicine and Dentistry of New Jersey. One command from end of surgery, and the time to extubation hundred thirty-seven children ages 2 to 10 years, ASA (TE), defined as time from end of surgery to tracheal physical status II–III, undergoing elective T&A, were en- extubation, were recorded. All patients were observed rolled in this investigator-initiated, prospective, random- continuously in the PACU for 2 hours by observers blinded ized, blinded, controlled study. Informed, written consent to study group. Pain was evaluated using the objective pain 9 to participate in the study was obtained from the parent or score (OPS) in the PACU on arrival and at 5 minutes, at 15 legal guardian and assent from children older than 7 years minutes, and then every 15 minutes for 120 minutes. EA of age. All patients had OSAS on the basis of clinical was evaluated at the same intervals by 2 scales: the 10 symptoms or diagnostic polysomnography. Clinical grad- Pediatric Anesthesia Emergence Delirium (PAED) scale 11 ing of OSAS was done by the surgeon on the basis of and a 5-point agitation scale described by Cole. Duration severity of symptoms such as restless sleep, severe snoring, of severe EA was noted on the Cole scale. Morphine (0.05 to ⅐ –1 Ͼ apnea witnessed by the parents, nocturnal enuresis, stertor, 0.1 mg kg ) was given for pain (score 4) or severe hyperactivity, or failure to thrive. Exclusion criteria were agitation (score 4 or 5) lasting more than 5 minutes. HR, ␣ systolic and diastolic NIBP, respiratory rate (RR), and Spo known allergy to 2 agonists, developmental delay, cardiac 2 and craniofacial abnormalities, anxiety disorder, chronic were recorded in the PACU every 5 minutes for the first 15 disabilities or pain syndrome, and use of psychotherapeutic minutes, then at 15-minute intervals for the next 2 hours. ␤ ␣ Any desaturation episode with Spo below 95% was noted. medications, blockers, digoxin, cimetidine, 2 agonists, 2 anticonvulsants, or psychotropic medications. A random number table was used to assign subjects into 1 of 2 Statistical Methods treatment groups: Dex infusion (group D) or IV fentanyl A power analysis indicated that 60 subjects were required (group F). The anesthesiologists and data collectors in the per group to show that the number of patients needing operating room (OR) were not blinded; the subjects, their intraoperative rescue fentanyl and rescue morphine in the parents, and observers in the postanesthesia care unit PACU would be 50% lower in the subjects receiving Dex. (PACU) were blinded to treatment group. Sixty subjects were also required per group to determine No premedication was given. Monitoring included that treatment with Dex would decrease the incidence of ␣ ϭ pulse oximetry, electrocardiogram, noninvasive arterial severe EA after surgery by 50% with 80% power ( 0.05) in comparison with the control group. blood pressure (NIBP), end-tidal CO2 (etco2), and a depth of anesthesia monitor, the Bispectral Index (BIS; Aspect Data were analyzed using SPSS software (version 16, Medical Systems, Natick, Massachusetts). Anesthesia was Chicago, Illinois), and are presented as number (n)or Ϯ induced with 8% inspired sevoflurane and 60% nitrous percentage (%), mean sd, or median as appropriate. Student t test was used to compare the mean value of oxide (N2O) in oxygen by facemask. Group D received IV Dex (2 ␮g ⅐ kg–1 over 10 minutes, followed by 0.7 ␮g ⅐ kg–1 ⅐ h–1 quantitative data between the 2 groups. Two-way until 5 minutes before the end of the surgery), and repeated-measures analysis of variance (ANOVA) was –1 group F received IV fentanyl (1 ␮g ⅐ kg ) as a bolus, as soon used for NIBP, HR, Spo2, MAC, and BIS. Student t test was as IV access was obtained. A balanced salt solution was used for the comparisons of intragroup values of intraop- administered according to standard fluid administration erative and postoperative systolic blood pressure and HR. guidelines. Rocuronium 0.6 mg ⅐ kg–1 was used to facilitate Nonparametric data such as pain score, PAED score, and tracheal intubation. End-tidal sevoflurane concentration EA score on the Cole scale were compared between groups was maintained at 1 minimum alveolar concentration with Mann–Whitney U test. Fischer exact test was used for comparision of gender; percentage of patients in each (MAC) with 60% N2O as long as the BIS remained below 60 during surgery. If the BIS reached 60 or more, the sevoflu- group with a preoperative diagnosis of mild, moderate, or rane concentration was increased to reduce the BIS below severe OSAS; and number of patients rescued with fentanyl 60. All patients received IV dexamethasone 0.5 mg ⅐ kg–1 or morphine and those with episodes of severe EA. P value (maximum 10 mg) and rectal acetaminophen 30 to 40 mg ⅐ of 0.05 or less was considered statistically significant. kg–1 up to a maximum of 1000 mg before the start of surgery. The data collector recorded the heart rate (HR), RESULTS systolic and diastolic blood pressures (NIBP), hemoglobin Results are presented for 122 patients. One hundred thirty- oxygen saturation (Spo2), etco2 tension, MAC, and BIS seven subjects were enrolled in this study; 15 subjects were every 5 minutes during the anesthetic. The values in the eliminated from data analysis for the following reasons:

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1005 Dexmedetomidine Infusion for Adenotonsillectomy

Table 1. Demographic Data Group F Group D P value (61 ؍ n) (61 ؍ n) Age (years) 3.8 Ϯ 1.5 4.2 Ϯ 2.1 0.16 2–3 years old (%) 26 (42.6) 26 (42.6) 1 Gender (M/F) 35/26 35/26 1 Weight (kg) 18.3 Ϯ 5.7 20.4 Ϯ 8.6 0.12 Baseline HR (beats/ 105 Ϯ 18 104 Ϯ 15 0.77 minute) Baseline systolic 101 Ϯ 13.7 104 Ϯ 12.6 0.29 NIBP (mm Hg) OSAS (% patients) Mild 30 26 Moderate 50 60 Severe 20 14 0.55 Data are expressed as n (%) and mean Ϯ SD. Group D ϭ dexmedetomidine group; group F ϭ fentanyl group; HR ϭ heart rate; NIBP ϭ noninvasive arterial blood pressure; OSAS ϭ obstructive sleep apnea syndrome.

Table 2. Intraoperative Data Group F Group D P value (61 ؍ n) (61 ؍ n) Rescue by fentanyl, 22 (36.1) 6 (9.8) 0.001* n (%) Fentanyl rescue 1.04 Ϯ 0.67 0.73 Ϯ 0.25 0.312 dosage (␮g/kg) Figure 1. A, Heart rate and B, systolic blood pressure during the ﬁrst Time of rescue 10.82 Ϯ 12.5 17.6 Ϯ 6.77 0.256 60 minutes of the procedure. Both variables were statistically lower (minutes) in the dexmedetomidine group (P Ͻ 0.05). Acetaminophen 31.51 Ϯ 4.96 28.30 Ϯ 6.59 0.02* dosage (mg/kg) Ϯ Ϯ Dexamethasone 0.30 0.12 0.30 0.14 0.847 ϭ dosage (mg/kg) different between the 2 groups (P 0.015); MAC was lower Duration of surgery 43.33 Ϯ 17.36 37.54 Ϯ 13.33 0.041* in group D, ranging from 5.7% to 41.6%. There was a (minutes) statistical difference in TA and TE, both lower in group D Duration of 75.08 Ϯ 24.73 69.80 Ϯ 16.82 0.175 than in group F (P Ͻ 0.05). Duration of surgery was anesthesia statistically lower in group D (P ϭ 0.041). There was no (minutes) Time to awake 8.75 Ϯ 4.06 7.18 Ϯ 4.05 0.03* difference in the average dose of intraoperative fentanyl (minutes) and dexamethasone between the 2 groups. The dose of Time to extubate 10.44 Ϯ 4.15 8.59 Ϯ 4.51 0.02* acetaminophen was lower in group D. None of the subjects (minutes) needed glycopyrrolate for bradycardia or fluid bolus for Data are expressed as n, mean Ϯ SD, and percentage. hypotension in the OR. Group D ϭ dexmedetomidine group; group F ϭ fentanyl group; HR ϭ heart The variables measured in the PACU are shown in Table ϭ rate; NIBP noninvasive arterial blood pressure. 3. In group D 10 (16.3%) patients required rescue morphine, * P Ͻ 0.05. in comparison with 29 (47.5%) in group F (P ϭ 0.002). The median of the maximum OPS was 3 for group D and 5 for surgery was cancelled for 2 patients, 1 refused to participate group F (P ϭ 0.001). The percentage of patients with an after enrolling, and 1 patient had an intraoperative complica- OPS score of 4 and above (Fig. 2A) from arrival (P ϭ 0.001) tion. Eleven subjects who completed the study had deviation and at 5 and 15 minutes was statistically lower in group D from this strictly controlled protocol or incomplete data and (P Ͻ 0.05). On the Cole scale (5-point scale), severe EA was were also removed before data analysis. defined as a score of 4 to 5. The frequency of severe EA is The 2 groups were comparable in age, gender, baseline shown in Figure 2B. On arrival in the PACU it was HR, systolic NIBP, and diagnosis of OSAS (Table 1). The statistically lower, 18% in group D and 45.9% in group F age range of patients in the study was 2 to 10 years, 90% of (P ϭ 0.004). At 5 and 15 minutes it was statistically lower in patients were 6 years or younger, and 26 patients (46.2%) in group D (P ϭ 0.028). At 30 minutes none of the patients had each group were 2 to 3 years old. severe EA in group D, and 1.6% of patients in group F had Intraoperative data are presented in Table 2. In group D, severe EA. The duration of agitation on the Cole scale 6 patients (9.8%) needed rescue fentanyl in comparison showed statistical significance; it was 6.59 Ϯ 7.4 minutes with 22 (36%) in group F (P ϭ 0.001). Mean HR (P ϭ 0.001) (mean Ϯ sd) for group D and 11.85 Ϯ 12.0 minutes (mean Ϯ (Fig. 1A) and mean systolic NIBP (Fig. 1B) were signifi- sd) for group F (P ϭ 0.004). There was a statistical differ- cantly lower in group D during the first 60 minutes (P ϭ ence in the median of the highest score on the Cole scale, 3 0.019). Mean diastolic NIBP was not statistically different in for group D and 4 for group F (P ϭ 0.001). The percentage the 2 groups (P ϭ 0.29). During the first 60 minutes of the of patients with a score of 10 and above for the PAED (Fig. anesthetic, MAC values of sevoflurane were significantly 2C) was statistically lower in group D at arrival (P ϭ 0.004)

1006 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 3. Postanesthesia Recovery Unit Data Group F Group D P value (61 ؍ n) (61 ؍ n) OPS maximum 5 (0–10) 3 (0–10) 0.001* (range) EA score 4 (1–5) 3 (1–5) 0.001* maximum (range) Duration of 11.85 Ϯ 12.02 6.59 Ϯ 7.42 0.004* severe EA (minutes) PAED score 14 (0–20) 10 (0–20) 0.051 maximum (range) Rescue by 29 (48) 10 (17) 0.0003* morphine, n (%) Morphine dosage 0.073 Ϯ 0.033 0.074 Ϯ 0.033 0.928 (mg/kg) SpO2 below 25 (41) 11 (18) 0.01* 95%, n (%) OPS, PAED, and EA (Cole scale) are expressed as median values of the maximum score. Other data are expressed as n (%) and mean Ϯ SD. Group D ϭ dexmedetomidine group; group F ϭ fentanyl group; OPS ϭ objective pain score; EA ϭ emergence agitation; PAED ϭ pediatric anesthesia emergence delirium. * P Ͻ 0.05. and at 5 and 15 minutes (P Ͻ 0.05). The median of the highest score on the PAED scale did not show a statistical difference, 10 for group D and 14 for group F (P ϭ 0.051). On arrival in the PACU until 90 minutes, HR was statistically lower in group D (P ϭ 0.001) than in group F. There was no statistical difference in mean systolic NIBP, respiratory rate, and Spo2 in the PACU. There was a statistically significant difference in the number of patients Figure 2. A, Percentage of patients with an objective pain score with Spo2 below 95% between the 2 groups, 11 (18%) in (OPS) of 4 and above. Score 4 and above lasting more than 5 group D and 25 (40.9%) in group F (P ϭ 0.01). minutes was treated. Statistically lower in group D (dexmedetomidine) at arrival (P ϭ 0.001), at 5 minutes (P ϭ 0.028), and at 15 DISCUSSION minutes (P ϭ 0.011). B, Percentage of patients with severe emergence agitation (EA), deﬁned as a score of 4 or 5 on the 5-point In this study of children undergoing T&A, an intraopera- scale. Lower in group D at arrival (P ϭ 0.001), at 5 minutes (P ϭ –1 tive initial loading dose of 2 ␮g ⅐ kg Dex followed by an 0.028), and at 15 minutes (P ϭ 0.028). C, Percentage of patients infusion at 0.7 ␮g ⅐ kg–1 ⅐ h–1 decreased intraoperative with Pediatric Anesthesia Emergence Delirium (PAED) score of 10 opiate and anesthetic requirements and decreased opiate and above. Statistically lower in group D at arrival (P ϭ 0.001), at 5 ϭ ϭ requirements in the PACU, in comparison with a control minutes (P 0.028), and at 15 minutes (P 0.011). group receiving intraoperative IV fentanyl. Additionally, there was a significantly lower incidence and duration of severe EA in children who received Dex. Children in group D had statistically lower systolic blood We are reporting on the use of a high initial loading dose pressure and HR, almost the entire duration of the anes- of Dex (2 ␮g ⅐ kg–1) followed by a relatively high dose thetic (Fig. 1), but none of the patients needed intervention continuous intraoperative infusion, because T&A is a pro- for bradycardia or hypotension on the basis of study cedure with an intense surgical stimulus, and surgery starts criteria. Hemodynamic data are consistent with reports by immediately with no preparation time. An analgesic- other investigators. Mason et al.13 used higher doses of Dex sparing effect of Dex has been shown,5 and when combined (2 to 3 ␮g ⅐ kg–1 loading dose and infusion of 1.5 to 2 ␮g ⅐ –1 ⅐ –1 with N2O there is an additive interaction to enhance kg h ) as the sole drug for sedation in children and analgesia.12 We aimed to use the analgesic-sparing effect of observed a decrease in HR and blood pressure, which were Dex and evaluated whether a continuous infusion could be within 20% of awake normal range. In children anesthe- used as a substitute for bolus fentanyl. In our study, 90% of tized with one MAC sevoflurane or desflurane and given a patients in group D did not require any other intraopera- single, lower dose of Dex (0.5 ␮g ⅐ kg–1), Deutsch et al.14 tive analgesics. Because there are no studies using a high- found a significant decrease in HR, but neither the systolic dose continuous infusion of Dex combined with inhalation nor diastolic blood pressure was statistically lower. The anesthetics, the U.S. Food and Drug Administration man- biphasic response usually seen in adults, with an initial dated reporting hemodynamic changes as a safety concern. increase in systolic blood pressure and a reflex decrease in

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1007 Dexmedetomidine Infusion for Adenotonsillectomy

HR followed by stabilization of these variables below analgesia in the PACU in comparison with 53% in the baseline, is not observed in children.15 placebo group. Erdil et al.19 compared a single dose of 0.5 On the basis of routine clinical practice, fentanyl was ␮g ⅐ kg–1 Dex with 2.5 ␮g ⅐ kg–1 fentanyl in patients given in a dose of 1 ␮g ⅐ kg–1 as a bolus to the control group. undergoing adenoidectomy and concluded that Dex pro- This lower dose is based on the enhanced analgesic sensi- vided residual analgesia similar to that of fentanyl. tivity to opiates in children with OSAS.3 It is noteworthy Pain can be severe after T&A, and it is commonly treated that in the control group only 36% of patients needed with opioids, despite a known sensitivity of patients with rescue fentanyl, indicating that our technique of low-dose OSAS and recurrent hypoxemia to opiates. Brown et al.3 fentanyl is effective in almost two thirds of patients. HR reported enhanced analgesic morphine sensitivity in chil- and systolic blood pressure increase was used as the trigger dren with OSAS during T&A and reduced morphine for rescue fentanyl in both groups in response to surgical requirements after T&A. Therefore, several nonopioid an- stimulation. The BIS monitor was used to ensure that algesics such as ketorolac, ketamine, and tramadol have patients in group D had an adequate depth of anesthesia been evaluated for pain management after T&A,20–22 but because they may not display hemodynamic changes due none have gained widespread use or acceptance because of to the inherent sympatholytic properties of Dex. In an concerns with side effects or inadequate analgesia. A attempt to maintain equivalent depth of anesthesia in both morphine-sparing effect of acetaminophen has been dem- groups, the sevoflurane concentration was titrated to main- onstrated in pediatric day-case surgery,23 and dexametha- tain a BIS value below 60. Consistent with studies in adult sone also reduces post-tonsillectomy pain.24 In the present Ϫ patients, the concentration of sevoflurane required to main- study, all patients were given 30 to 40 mg ⅐ kg 1 of acetamin- tain the BIS below 60 was smaller in patients receiving Dex ophen rectally before start of surgery and intraoperative IV (MAC in group D was 5.7% to 41.6% lower). Tufanoguallari dexamethasone. A multimodal, opioid-sparing, analgesic ap- et al.8 found reductions in the average end-tidal desflurane proach including Dex, such as the one used in our study, is concentration of 19%–22%, depending on the rate of Dex worth considering in this patient population with a high infusion, which ranged from 0.2 to 0.8 ␮g ⅐ kg–1 ⅐ h–1. The potential for adverse respiratory events. The incidence of anesthetic-sparing effect of Dex appears to have an added nausea or vomiting was extremely low in this study. Only 1 advantage in facilitating earlier awakening and tracheal patient needed an antiemetic in the PACU, probably because extubation. In the present study, TA and TE were statisti- of the antiemetic effect of dexamethasone. cally lower in group D, despite the high dose used. Most EA is a complex phenomenon, the etiology of which is investigators using Dex as a low-dose intraoperative infu- multifactorial. The wide variability in the incidence of sion or as a single bolus reported no difference in TA and agitation in the different studies on EA may be due to the TE in comparison with placebo.6,16 Only 1 study reported criteria used to define this phenomenon and the time in the that a single dose of 0.5 ␮g ⅐ kg–1 Dex, 5 minutes before the PACU when EA was measured.17 We did repeated mea- end of surgery significantly prolonged TA and TE in surements at frequent time intervals, because a single comparison with placebo in patients having T&A.7 measurement may not reflect the true incidence of EA.11 Evaluation of postoperative pain is complicated by the Group D had a statistically lower frequency of severe EA difficulty in assessing pain in younger children and by the than did group F until 30 minutes (Fig. 2B). At 30 minutes occurrence of EA. It is often difficult to distinguish between there was no incidence of severe EA in group D, and in pain and EA because of the overlapping clinical picture, group F it was 1.6%. Severe EA lasting more than 5 minutes and pain itself can be the source of agitation.17 Most was treated. The incidence of severe EA on arrival in the investigators have used different assessment tools to try PACU in group D (18%) was similar to that reported by and separate the two, but there is generally overlap in the Guler et al.7 (17%), who used a single dose of Dex 5 minutes scales, because a child who is restless or thrashing will before the end of the procedure in children undergoing score high on both scales. We did find a positive correlation T&A. The occurrence of EA in younger patients and between agitation and pain; group F had higher pain and otolaryngologic procedures is reported to be high, although EA scores than did group D. Results on the OPS, Cole scale, the exact reason for this is not known.4 Ninety percent of and PAED showed a very similar trend in both groups; patients in our study were 6 years old or younger, and 26 scores were highest on arrival in the PACU and decreased patients (46.2%) in each group were 2 to 3 years old. over time (Fig. 2, A–C). A significantly smaller number of Hyperactivity and attention deficit disorder are frequently patients needed rescue morphine in group D, 18% in seen in children with OSAS, possibly explaining or contrib- comparison with 44% in group F. Because it is difficult to uting to a high incidence of EA in our T&A patients. separate pain and EA, and the fact that the rescue drug for Dexmedetomidine has been used successfully as an infu- both agitation and pain in our study was morphine, it is not sion (0.2 ␮g ⅐ kg–1 ⅐ h–1) continued into the postoperative possible to determine whether the morphine was given for period for 15 minutes or single dose at the end of surgery pain or for agitation. On the basis of the effectiveness of (0.5 ␮g ⅐ kg–1) to prevent or reduce emergence delirium in smaller doses of intraoperative Dex in adult patients for children.6,7,16 It must be noted that these studies compared reducing postoperative morphine consumption for 24 Dex with placebo, whereas our control group received hours,8,18 we could assume that an analgesic effect would fentanyl 1 ␮g ⅐ kg–1, which also reduces EA. However, a be present in our study patients in the immediate postop- higher dose is reported to be effective in patients having erative period. In children, Guler et al.7 found that 23% painful procedures.25 From our study and others, it re- patients who received a single dose of 0.5 ␮g/kg Dex mains difficult to discern whether the analgesic or sedative ␣ before the end of the procedure (T&A) required opoiods for effects of 2 agonists are responsible for reducing EA in

1008 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA children; regardless of the mechanism, Dex appears to be validated scale, but is easy to use, and defining the catego- effective in a wide range of doses. The half-life of Dex is ries of mild or severe is clear. reported to be 1.8 hours in children,15 but there are no data The OPS is not a validated scale, but this scale or some on duration of sedative or analgesic effects after discontinu- modification of it has been used in several studies in ation of Dex infusion. The HRs were significantly slower in children. Although 2 other studies on EA6,19 have used the group D until 90 minutes in the PACU. The residual effects OPS, it is perhaps not the best scale to use in a study on EA on HR of an intraoperative Dex infusion and the potential because of considerable overlap on the items being scored. for an attenuated response to postoperative bleeding in We did not follow patients once they were discharged T&A patients may be a concern and a disadvantage of from the PACU. A future study with overnight pulse using a Dex infusion. oximetry data and use of postoperative analgesics would The risk of respiratory morbidity after T&A in children be worthwhile to perform. with OSAS is reported to be about 20%.2 Sanders et al.26 reported that although the patients with OSAS were more CONCLUSION likely to require supplemental oxygen, oral airway use, or In children undergoing T&A, the goal is to minimize assisted ventilation on emergence, severe complications respiratory and airway compromise and have an awake, such as laryngospasm and bronchospasm were uncom- settled, comfortable child after the surgery. An opioid- mon. In the present study, there were no instances of sparing technique is particularly appealing in children with laryngospasm or bronchospasm after extubation. One pa- OSAS, when airway obstruction is known to preexist and tient developed intraoperative pulmonary edema and was may persist on the night after surgery. An intraoperative excluded from the study because she remained intubated infusion of Dex combined with sevoflurane and N2O provided satisfactory intraoperative conditions for T&A overnight. Although not a study variable, we noted that without adverse hemodynamic effects. TA and TE were extubation was much smoother with less coughing and shorter than they were for the patients receiving fentanyl. breath-holding in patients given Dex. All patients were Postoperative opoiod requirements were significantly re- observed continuously in the PACU for 2 hours, and the duced, and the incidence and duration of severe EA was observers were asked to record the lowest Spo during this 2 lower, resulting in a smooth recovery. We have described a period. There was a statistically significant difference in the practical, effective, and safe technique for using Dex infu- number of patients with Spo below 95% in the PACU 2 sion. A multimodal, opioid-sparing, analgesic approach between the 2 groups, 11 in group D and 25 in group F. This including Dex, such as the one used in our study, can be could be related to the smaller requirement for opiates in useful in children with OSAS undergoing other surgical the PACU in group D or to the lower incidence and procedures besides T&A, wherein the advantages of de- duration of severe EA in group D. The goal of having a creased perioperative opioid requirements and a reduced child who was settled, comfortable, and less restless, with occurrence of EA will be beneficial. application of monitors and administration of supplemental oxygen in the PACU, was easier to achieve in patients REFERENCES who received Dex. 1. Brown KA. What we don’t know about childhood obstructive A few methological considerations of this study need to sleep apnea. Paediatr Anaesth 2001;11:385–9 be mentioned. The anesthesiologist and the data recorder in 2. McColley SA, April MM, Carroll JL, Naclerio RM, Loughlin GM. Respiratory compromise after adenotonsillectomy in chil- the OR were not blinded to the study group. We believe dren with obstructive sleep apnea. Arch Otolaryngol Head that knowledge of study group assignment did not bias the Neck Surg 1992;118(9):940–3 conduct of the anesthetic, because the study protocol was 3. Brown KA, Laferriere A, Lakheeram I, Moss IR. Recurrent tightly controlled, with specific criteria regarding intraop- hypoxemia in children is associated with increased analgesic sensitivity to opiates. Anesthesiology 2006;105(4):645–7 erative rescue fentanyl, sevoflurane concentration, the time 4. Voepel-Lewis T, Malviya S, Tait AR. A prospective cohort to discontinue sevoflurane, extubation criteria, and use of study of emergence agitation in the pediatric postanesthesia rescue morphine in the PACU. care unit. Anesth Analg 2003;96(6):1625–30 The PAED is the only validated rating scale for emer- 5. Hall JE, Uhrich TD, Barney JA, Arain SR, Ebert TJ. Sedative, 10 amnestic, and analgesic properties of small-dose dexmedeto- gence delirium. The investigators who developed the midine infusions. Anesth Analg 2000;90:699–705 PAED scale rated emergence behavior 10 minutes after the 6. Shukry M, Clyde MC, Kalarickal PL, Ramadhyani U. Does child awakened and remained awake (did not fall back to dexmedetomidine prevent emergence delirium in children sleep). Early in the present study, we found this to be a after sevoflurane-based general anesthesia? Paediatr Anaesth 2005;15:1098–104 potential problem because children who were asleep were 7. Guler G, Akin A, Tosun Z, Ors S, Esmaoglu A, Boyaci A. receiving ratings of 4 on the first 3 items of the scale Single-dose dexmedetomidine reduces agitation and provides because they could not make eye contact, their actions were smooth extubation after pediatric adenotonsillectomy. Paediatr not purposeful, and they were not aware of their surround- Anaesth 2005;15(9):762–6 ings. Therefore we had to modify the scoring on the scale 8. Tufanoguallari B, White PF, Peixoto MP, Kianpour D, Lacour T, Griffin J, Skrivanek G, Macaluso A, Shah M, Provost DA. and rate these items as zero. Clearly, the children were not Dexmedetomidine infusion during laparoscopic bariatric sur- agitated if they were sleeping. Because we used a modified gery: the effect on recovery outcome variables. Anesth Analg version of the PAED, we used a second scale (Cole) to run 2008;106(6):1741–8 concomitantly to support the findings with the modified 9. Hannallah RS, Broadman LM, Belman AB, Abramowitz MD, Epstein BS. Comparison of caudal and ilioinguinal/ version of the PAED. The 1 to 5 scale described by Cole et iliohypogastric nerve blocks for control of post-orchiopexy pain al.11 has been used in several studies of EA. It is not a in pediatric ambulatory surgery. Anesthesiology 1987;66(6):832–4

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1009 Dexmedetomidine Infusion for Adenotonsillectomy

10. Sikich N, Lerman J. Development and psychometric evaluation 19. Erdil F, Demirbilek S, Begec Z, Ozturk E, Ulger MH, Ersoy MO. of the pediatric anesthesia emergence delirium scale. Anesthe- The effects of dexmedetomidine and fentanyl on emergence siology 2004;100(5):1138–45 characteristics after adenoidectomy in children. Anaesth Inten- 11. Cole JW, Murray DJ, McAllister JD, Hirshberg GE. Emergence sive Care 2009;37(4):571–6 behaviour in children: defining the incidence of excitement 20. Marret E, Flahault A, Samama CM, Bonnet F. Effects of and agitation following anaesthesia. Paediatr Anaesth 2002; postoperative, nonsteroidal, antiinflammatory drugs on bleed- 12:442–7 ing risk after tonsillectomy: meta-analysis of randomized, 12. Dawson C, Ma D, Chow A, Maze M. Dexmedetomidine controlled trials. Anesthesiology 2003;98(6):1497–502 enhances analgesic action of nitrous oxide: mechanisms of 21. Erhan OL, Go¨ksu H, Alpay C, Beçstaçs A. Ketamine in action. Anesthesiology 2004;100(4):894–904 post-tonsillectomy pain. Int J Pediatr Otorhinolaryngol 2007; 13. Mason KP, Zurakowski D, Zgleszewski SE, Robson CD, Car- 71(5):735–9 rier M, Hickey PR, Dinardo JA. High dose dexmedetomidine 22. Hullett BJ, Chambers NA, Pascoe EM, Johnson C. Tramadol vs as the sole sedative for pediatric MRI. Paediatr Anaesth morphine during adenotonsillectomy for obstructive sleep 2008;18(5):403–11 apnea in children. Paediatr Anaesth 2006;16(6):648–53 14. Deutsch E, Tobias JD. Hemodynamic and respiratory changes 23. Korpela R, Korvenoja P, Meretoja OA. Morphine-sparing effect following dexmedetomidine administration during general of acetaminophen in pediatric day-case surgery. Anesthesiol- anesthesia: sevoflurane vs desflurane. Paediatr Anaesth ogy 1999;91:442–7 2007;17:438–44 24. Afman CE, Welge JA, Steward DL. Steroids for posttonsillec- 15. Petroz GC, Sikich N, James M, van Dyk H, Shafer SL, Schily M, tomy pain reduction: meta-analysis of randomized controlled Lerman J. A phase I, two center study of the pharmacokinetics trials. Otolaryngol Head Neck Surg 2006;134:181–6 and pharmacodynamics of dexmedtomidine in children. 25. Cohen IT, Finkel JC, Hannallah RS, Hummer KA, Patel KM. Anesthesiology 2006;105:1098–110 The effect of fentanyl on the emergence characteristics after 16. Ibacache ME, Munoz HR, Brandes V, Morales AL. Single-dose desflurane or sevoflurane anaesthesia. Paediatr Anaesth dexmedetomidine reduces agitation after sevoflurane anesthe- 2002;12:442–7 sia in children. Anesth Analg 2004;98(1):60–3 26. Sanders JC, King MA, Mitchell RB, Kelly JP. Perioperative 17. Vlajkovic GP, Sindjelic RP. Emergence delirium in children: complications of adenotonsillectomy in children with obstruc- many questions, few answers. Anesth Analg 2007;104:84–91 tive sleep apnea syndrome. Anesth Analg 2006;103(5):1115–21 18. Gurbet A, Basagan-Mogol E, Turker G, Ugun F, Kaya FN, Ozcan B. Intraoperative infusion of dexmedetomidine reduces perioperative analgesic requirements. Can J Anaesth 2006; 53(7):646–52

1010 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Epidemiology of Ambulatory Anesthesia for Children in the United States: 2006 and 1996

Jennifer A. Rabbitts, MB, ChB,* Cornelius B. Groenewald, MB, ChB,* James P. Moriarty, MSc,† and Randall Flick, MD, MPH*

BACKGROUND: There are few data that describe the frequency, anesthetic type, provider, or disposition of children requiring outpatient anesthesia in the United States (US). Since the early 1980s, the frequency of ambulatory surgery has increased dramatically because of advances in medical technology and changes in payment arrangements. Our primary aim in this study was to quantify the number of ambulatory anesthetics for children that occur annually and to study the change in utilization of pediatric anesthetic care over a decade. METHODS: The US National Center for Health Statistics performed the National Survey of Ambulatory Surgery in 1994 through 1996 and again in 2006. The survey is based on data abstracted from a national sample of ambulatory surgery centers and provides data on visits for surgical and nonsurgical procedures for patients of all ages. We abstracted data for children who had general anesthesia, regional anesthesia, or monitored anesthesia care during the ambulatory visit. We obtained the information from the 2006 and 1996 databases and used population census data to estimate the annual utilization of ambulatory anesthesia per 1000 children in the US. RESULTS: In 2006, an estimated 2.3 million ambulatory anesthesia episodes of care were provided in the US to children younger than 15 years (38 of 1000 children). This amount compares with 26 per 1000 children of the same age group in 1996. In most cases, an anesthesiologist was involved in both time periods (74% in 2006 and 85% in 1996). Of the children, 14,200 were admitted to the hospital postoperatively, a rate of 6 per 1000 ambulatory anesthesia episodes. CONCLUSION: The number and rate of ambulatory anesthesia episodes for US children increased dramatically over a decade. This study provides an example of how databases can provide useful information to health care policy makers and educators on the utilization of ambulatory surgical centers by children. (Anesth Analg 2010;111:1011–5)

he introduction of the first freestanding ambulatory ASCs are less than for corresponding services in hospital- surgery centers (ASCs) in the 1970s resulted in a based outpatient departments. In addition, copayments Trapid increase in the proportion of operations per- and charges to patients are generally less at ASCs than at formed on an outpatient basis, from Ͻ10% in 1979 to 50% in hospitals. Almost 90% of all US freestanding ASCs are 1990.1 The number of ASCs continues to increase, with a wholly or partially owned by physicians and 96% are 4 150% increase per 100,000 population reported in metro- for-profit institutions. politan areas from 1993 to 2001.2 The number of Medicare- The purpose of this study was to describe, for the first certified ASCs increased 64% between 2000 and 2007, from time, the utilization of freestanding and hospital-based 3028 to 4964.3 Improvements in surgical and anesthetic ASCs in regard to their care of children. We quantified the number of ambulatory anesthesia episodes occurring an- techniques have increased the proportion of procedures nually for children in accordance with age group, anes- performed on an outpatient basis to Ͼ70% of the total thetic type, and anesthesia provider and described the surgical interventions currently performed in the United change in utilization over a decade. Secondary analyses States (US).1 examined the distribution of perioperative time and dispo- No quantification has been made of the pediatric proce- sition and used unplanned admission as an end point. dures occurring on an outpatient basis in the US. As the country enters an era of health care reform, epidemiologic data on the utilization of medical resources may be helpful METHODS The National Survey of Ambulatory Surgery (NSAS) is the to policy makers as health care expenditures are analyzed. only US national study of ambulatory surgery in hospital- For example, current Medicare payments to freestanding based and freestanding ASCs.5 We abstracted the data for ambulatory anesthesia of children from this public data- From the *Department of Anesthesiology, and the †Division of Health Care base for 1996 and 2006. National census data were used to Policy & Research, Mayo Clinic, Rochester, Minnesota. estimate utilization rates. Accepted for publication June 10, 2010. Supported by the Department of Anesthesiology, Mayo Clinic, Rochester, MN. The NSAS Database Disclosure: The authors report no conflicts of interest. The NSAS was performed by the National Center for Address correspondence and reprint requests to Randall Flick, MD, MPH, Health Statistics on a nationally representative sample of Department of Anesthesiology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905. Address e-mail to [email protected]. surgery centers that perform ambulatory procedures. The Copyright © 2010 International Anesthesia Research Society complete sampling and survey methods have been de- 5 DOI: 10.1213/ANE.0b013e3181ee8479 scribed and select data have been published for patients of

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1011 Ambulatory Anesthesia in Children all ages who had both surgical and nonsurgical procedures.6,7 In summary, eligible hospital-based facilities were identified from the SMG Marketing Group, Inc., Hospital Market Database5 and included all short-stay or general (medical, surgical, or children’s) noninstitutional, nonfed- eral hospitals in the 50 states and the District of Columbia with 6 or more beds staffed for patient use. Eligible freestanding facilities were identified from the SMG Free- Standing Outpatient Surgery Center Database and the Health Care Financing Administration Provider of Services Public Use File. Hospital-based and freestanding ASCs consisted of hospitals that were state regulated or certified for Medicare that performed at least 50 ambulatory procedures in the Figure 1. Rate of ambulatory anesthesia for children in 1996 and previous year and excluded dental, podiatry, pain, abor- 2006. Rate increased from 26 per 1000 children younger than 15 years in 1996 to 38 per 1000 children of this age group in 2006. tion, family planning, and birthing centers. The included procedures were both surgical and nonsurgical (e.g., lumbar puncture, computed tomographic scanning) procedures Data Abstraction from the National Hospital performed on an ambulatory basis in general operating Discharge Survey rooms, dedicated ambulatory surgery rooms, and other To help interpret the trends observed in the utilization of specialized rooms, including endoscopy units and cardiac ambulatory surgery facilities, we abstracted a limited catheterization laboratories. amount of information from the National Hospital Dis- A multistage probability design was used, in which charge Survey. The survey is a national database of inpa- independent samples of hospitals and freestanding ASCs tient medical and surgical care that is similar to the NSAS were selected at the first or second stages and visits to these database.8,9 It does not include information on the admin- 5 facilities were selected at the final sampling stages. An istration of anesthesia during procedures performed at NSAS medical abstract form (Appendix) was used to inpatient facilities, and therefore, we could not differentiate Ͼ collect data for each sampled visit during which 1 proce- the procedures performed with anesthesia from the nonin- dure may have been performed. Data were abstracted from vasive procedures (including imaging studies) or proce- the medical record by facility staff in 30% of cases and by dures performed without anesthesia. To help interpret the US Census Bureau personnel in 70% of cases. Data ab- change in rate of utilization of ASCs for surgical proce- stracted for the NSAS database included patient character- dures, we abstracted the number of inpatient visits in both istics, payment information, surgical and nonsurgical 1996 and 2006 for which tonsillectomy or adenoidectomy, procedures, surgical visit information (e.g., perioperative or both, was listed as the first procedure. We combined the times, anesthesia provider, type of anesthesia), and patient data on inpatients younger than 15 years with population disposition. census data to estimate the rate of these procedures. In 2006, data were collected for approximately 52,000 ASC visits at 437 centers (142 hospital-based and 295 RESULTS freestanding centers), with an overall response rate of 74% 6 Utilization of ASCs for Children, 2006 of sampled centers. Survey responses were received from In 2006, 2,300,651 (standard error [SE], 315,651) ambulatory 75% of sampled hospital-based ASCs and 74% of sampled anesthesia episodes of care were performed for patients freestanding ASCs. In 1996, data were collected for 125,000 younger than 15 years in the US, which is a rate of 38 ASC visits to 488 centers, with an overall response rate of 7 ambulatory anesthetic procedures per 1000 children (Fig. 81% of sampled centers. Survey responses were received 1). Among these cases, anesthetics were given to 1,329,976 from 91% of sampled hospital-based centers and 70% of (SE, 160,647) boys and 1,071,650 (SE, 168,697) girls, or rates sampled freestanding ASCs. of 43 (SE, 5.2) per 1000 boys and 36 (SE, 5.7) per 1000 girls. Data by age group and type of anesthesia are provided in Data Abstraction from the NSAS Database Table 1. We abstracted data pertaining to type of anesthetic admin- The 3 most frequently performed procedures were ton- istered, anesthesia provider present, procedure time vari- sillectomy, adenoidectomy, and myringotomy with ear ables, primary procedure, gender, source of payment, and tube.6 Data regarding the provider of anesthesia are dis- discharge status. We combined the data of patients younger played in Table 2. than 15 years with data from the population census to estimate the rate of visits to an ASC for ambulatory Perioperative Data procedures with anesthesia for US pediatric patients. Age The breakdown of perioperative times is displayed in categories were Ͻ1 year, 1 to 4 years, and 5 to 14 years Figure 2. Of the children who received anesthetics, 12,030 based on available census data. All statistical analyses were were admitted postoperatively to an inpatient facility (data conducted with Stata/SE 10.1 software (StataCorp LP, on those patients readmitted after discharge were not College Station, TX). Where data were missing, we catego- available), for a rate of 6 (SE, 1.3) inpatient admissions per rized the result as “not specified” (e.g., for the anesthesia 1000 ambulatory anesthetics. An estimated 2,193,686 (SE, provider category in 1996 data). 311,507) of the 2,401,626 children receiving ambulatory

1012 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Ambulatory Anesthesia Sessions for Monitored Anesthesia Care or Regional or General Anesthetics Only by Age Group, 2006 and 1996 Overall rate Age, y MAC (SE) Regional (SE) General (SE) Total (SE) per 1000 children 2006 Ͻ15 44,462 (10,149) 26,484 (7036) 2,241,985 (313,649) 2,300,651 (315,651) 38 a a Ͻ1 — — 196,991 (36,173) 202,412 (36,363) 49 a a 1–4 — — 963,733 (141,654) 974,915 (141,977) 60 5–14 38,215 (9823) 11,156 (3741) 1,081,261 (145,287) 1,123,295 (147,728) 28 1996b Ͻ15 53,943 14,776 1,490,686 1,522,883 26 a a Ͻ1 — — 138,661 140,639 37 1–4 12,283 3177 633,454 640,424 41 5–14 39,351 9338 718,571 741,820 19 MAC ϭ monitored anesthesia care; SE ϭ standard error. a Sample size was too small or SE was too large. b 1996 Data did not contain some of the survey sampling variables needed to accurately estimate the SEs and thus the SEs are not reported.

Table 2. Anesthesia Provider Involved During Admission to Ambulatory Center When Anesthesia Was Provided by an Anesthesiologist or CRNA Only, 2006 and 1996 Both anesthesiologist Age, y Anesthesiologist only (SE) CRNA only (SE) and CRNA (SE) 2006 Ͻ15 1,389,393 (209,784) 603,695 (158,713) 292,630 (52,055) Ͻ1 130,681 (23,159) 52,145a 18,375a 1–4 577,712 (89,577) 256,924a 135,772 (25,799) 5–14 681,000 (104,418) 294,626 (69,382) 138,483 (26,847) 1996b Ͻ15 936,944 219,716 314,919 Ͻ1 95,883 17,738 24,910 1–4 387,108 93,878 137,596 5–14 453,953 108,100 152,413 CRNA ϭ certiﬁed registered nurse anesthetist; SE ϭ standard error. a Sample size too small or SE too large. b Data of 1996 did not contain some of the survey sampling variables needed to accurately estimate the SEs and thus the SEs are not reported.

insurance or through self-pay. For the other visits, the cost of 816,185 visits was paid through public forms of funding (e.g., Medicaid, TRICARE). Of the visits for which funding was known, the cost for 65% of visits was paid from a private or commercial source and for 35% of visits from a government source.

Utilization of ASCs for Children, 1996 In 1996, an estimated 1,522,883 ASC visits included anesthesia administration, which is a rate of 26 ambulatory anesthetic procedures per 1000 children younger than 15 years. Data by age group and type of anesthetic are Figure 2. Mean perioperative times for children younger than 15 provided in Table 1. Data regarding the provider of anes- years. Postoperative time accounted for the largest portion of the thesia are displayed in Table 2. perioperative period during pediatric visits to ambulatory surgery centers for surgical procedures. Room time ϭ the difference between total operating room time (from entrance into until exit out of Payment Information, 1996 the operating room, or 45 [2] minutes) and surgical time (from the In 1996, most (1,142,481) of the ASC visits for children were operation’s start to its ﬁnish, or 26 [1] minutes); postoperative funded through private or commercial insurance or self- time ϭ from entrance into until exit from the recovery room, or 71 (3) pay; 494,665 (30%) were funded through public sources minutes; perioperative time ϭ from entrance into the operating room until exit from the recovery room. (including Medicaid and TRICARE). Of the visits for which funding was known, 70% of visits were paid from a private or commercial source and 30% from a government source. anesthesia were recorded as having routine discharge (913 of 1000 ambulatory anesthetics; SE, 138). Rate of Inpatient and Ambulatory Tonsillectomy and Adenoidectomy, 1996 and 2006 Payment Information, 2006 The rate of inpatient tonsillectomy or adenoidectomy, or In 2006, the cost of 1,547,744 visits to ASCs for children both, in 1996 was 0.39 (SE, 0.08) per 1000 children younger younger than 15 years was paid by private or commercial than 15 years. In 2006, it was 0.18 (SE, 0.04) per 1000

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1013 Ambulatory Anesthesia in Children

Table 3. Rates of Tonsillectomy or Adenoidectomy, or Both, per 1000 Children Performed on an Ambulatory Basisa and an Inpatient Basis,b 2006 and 1996 Rate per 1000 children (SE) Age, y Ambulatoryc Inpatient 2006 Ͻ15 9.7 (2.0) 0.18 (0.04) d d Ͻ1 — — d 1–4 13.2 (2.8) — d 5–14 9.2 (2.0) — 1996 Ͻ15 5.3 0.39 (0.08) d d Ͻ1 — — Figure 3. Provider of ambulatory anesthesia for children in 2006 and d 1–4 6.2 — 1996. An anesthesiologist was involved in most anesthesia epi- d 5–14 5.5 — sodes for ambulatory surgery in both time periods (61% in 2006 and 64% in 1996). CRNA ϭ certiﬁed registered nurse anesthetist. SE ϭ standard error. a From National Survey of Ambulatory Surgery data. b From National Hospital Discharge Survey data. c Ambulatory data from 1996 did not contain some of the survey sampling variables needed to accurately estimate SEs and thus SEs are not reported. American Hospital Association Annual Surveys of Hospi- d Sample size was too small or SE too large. tals, which showed a 28% increase in hospital-based outpatient surgery and a 4.5% decrease in inpatient surgery from 1993 to 2001.2 However, these data must be inter- children of that age. By comparison, the rate of ambulatory preted with caution because there may be a different tonsillectomy or adenoidectomy, or both, in 1996 was 5.3 explanation for this change. For example, surgeons may per 1000 children younger than 15 years; in 2006, it was 9.7 schedule tonsillectomies as outpatient procedures in chil- (SE, 2.0) per 1000 children of that age. Information by age is dren who stay overnight for payment reasons. provided in Table 3. During both 1996 and 2006, the highest rate of ASC visits with general anesthesia administration was in the 1 to 4 years age group and the lowest rate was in the 5 to 14 years DISCUSSION age group. Most of the ambulatory pediatric anesthesia was Over the 10 years between 1996 and 2006, pediatric visits to delivered by an anesthesiologist in both time periods (74% ASCs during which anesthesia was administered increased in 2006 and 85% in 1996). However, with the increased use almost 50%, from approximately 1.6 million in 1996 to 2.3 of ambulatory anesthesia, the proportion of anesthetics million in 2006. During that period, the population of provided by a certified registered nurse anesthetist alone pediatric patients increased only 5.3%, suggesting that the increased whereas the proportion of anesthetics provided increase in ASC visits requiring anesthesia was the result of by a certified registered nurse anesthetist working with an a change in overall utilization or a shift in practice from anesthesiologist decreased (Fig. 3). Nongovernmental inpatient to outpatient, or both. Overall utilization in- groups (private and commercial insurance and self-pay) creased from 26 to 38 ASC visits per 1000 children, repre- were the funding source for most visits in both 1996 and senting an almost 40% increase. 2006. Whether this increase in rate of ambulatory anesthesia is attributable to an increase in surgical procedures or a shift Economic and Educational Implications of procedures from inpatient to outpatient settings has This study is an example of how a database can be used to important implications for health care spending. No data abstract data useful to health care policy makers, adminis- are available that permit a direct comparison of inpatient trators, and educators and to provide important informa- and outpatient utilization rates for procedures requiring tion when changes have to be made in health care systems. anesthesia. The increase in ambulatory anesthesia itself may be Therefore, we abstracted the rate of either tonsillectomy interpreted as an increase in health care spending. How- or adenoidectomy and of both procedures from the NSAS ever, it may be associated with a decrease in inpatient database and the National Hospital Discharge Survey da- anesthesia, which could decrease health care expenditures.4 tabase, because tonsillectomy and adenoidectomy are com- If this trend continues, further savings may occur. mon pediatric procedures that may be performed in an The dramatic increase in pediatric ambulatory surgery inpatient or an outpatient setting and always require anes- has direct implications for residency and fellowship train- thesia. The rate of these procedures as an inpatient opera- ing, and this effect may be the most important impact of tion decreased approximately 54% from 1996 to 2006 this trend. Currently, programs are based at inpatient whereas the rate for the ambulatory setting increased 82%. medical centers, and training at ambulatory anesthesia This change suggests that there may have been a shift of centers may be limited. As pediatric anesthesia shifts to procedures from the inpatient, short-stay hospitals to the outpatient and ambulatory centers, education for residents hospital-based and freestanding ASCs during these 10 and fellows may need to be adapted to adequately prepare years. This shift is consistent with data from the Medicare anesthesiologists to manage the unique challenges of am- Online Survey Certification and Reporting System and the bulatory anesthesia in children.10,11

1014 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Limitations educators, and administrators, as well as other parties The main limitations of this study are those inherent to the involved in health care organization and provision. This NSAS database and the medical charts that were reviewed type of information is currently of particular importance in for it, because our study was reliant on data collected by the this era of health care reform when, to make decisions National Center for Health Statistics for the NSAS database. regarding health care spending and reform, data on utili- There was an average response rate of 74% by sampled zation of all aspects of health care are needed from all hospitals in 2006 and 81% in 1996. Data were extracted from groups. the medical records of sampled patients by nonmedical personnel after training,5 and it is possible that the medical APPENDIX abstract form (Appendix) was not uniformly interpreted. Medical Abstract Form of the National Survey of Ambula- This process was also limited by the data that were tory Surgery, NSAS-5 (2-1-2006). (Adapted from US Census available and retrievable from the medical records. Infor- Bureau and US Department of Commerce. Available at: mation was missing for some cases; specifically, the source http://www.cdc.gov/nchs/data/hdasd/nsas_participant/ of funding was unknown for a large portion of the pediatric nsas5.pdf.) ambulatory visits in 2006. The statistical software we used could abstract data only REFERENCES for specific visits and the primary procedure during the 1. Pregler JL, Kapur PA. The development of ambulatory anes- visit. These visits potentially could have included multiple thesia and future challenges. Anesthesiol Clin North Am procedures and anesthetics that were counted as 1 visit. 2003;21:207–28 Sample size was limited in the pediatric population and, 2. Bian J, Morrisey MA. Free-standing ambulatory surgery centers and hospital surgery volume. Inquiry 2007;44:200–10 therefore, further data could not be reported because of 3. Medicare Payment Advisory Commission (MedPAC). June 2008 unacceptable standard errors. Also, the 1996 and 2006 Healthcare Spending and the Medicare Program: A Data Book. NSAS medical abstracts were not identical. For example, Available at: http://www.medpac.gov/documents/Jun08Data the “not-specified” field used in 2006 was not used in 1996, Book_Entire_report.pdf. Accessed February 16, 2010 and thus “not specified” in 1996 was defined as no other 4. Medicare Payment Advisory Commission (MedPAC). Report to the Congress: Medicare Payment Policy. Available at: field filled. Options for payment source were slightly http://www.medpac.gov/documents/Mar09_EntireReport. different in the 2 time periods, and therefore, comparisons pdf. Accessed February 16, 2010 cannot be made for this category. 5. McLemore T, Lawrence L. Plan and operation of the National In addition, sampling variables were not available for Survey of Ambulatory Surgery. Vital Health Stat 1 1997;37:I–IV, 1–124 the 1996 NSAS database and thus accurate standard errors 6. Cullen KA, Hall MJ, Golosinskiy A. Ambulatory surgery in the could not be calculated for 1996 data. This lack of sampling United States, 2006. Natl Health Stat Report 2009;11:1–25 variables limited the comparisons that we could make 7. Hall MJ, Lawrence L. Ambulatory surgery in the United States, between the 2 time periods. Percentages do not add up to 1995. Adv Data 1997;296:1–15 100% because all data represent estimates based on sam- 8. DeFrances CJ, Lucas CA, Buie VC, Golosinskiy A. 2006 Na- tional Hospital Discharge Survey. Natl Health Stat Report pling rates and population size. 2008;5:1–20 9. Graves EJ, Owings MF. 1996 summary: National Hospital CONCLUSIONS Discharge Survey. Adv Data 1998;301:1–12 The rate of ambulatory anesthesia for children in the US 10. Emhardt JD, Saysana C, Sirichotvithyakorn P. Anesthetic con- Ͼ siderations for pediatric outpatient surgery. Semin Pediatr increased by 40% over a decade, partly because of a shift Surg 2004;13:210–21 in procedures from an inpatient to an outpatient setting. 11. Twersky RS. Educational protocols in ambulatory anesthesia. These databases are useful to health care policy makers, Ambul Surg 1997;5:117–9

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1015 REVIEW ARTICLE

CME The Anesthetic Considerations of Tracheobronchial Foreign Bodies in Children: A Literature Review of 12,979 Cases

Christina W. Fidkowski, MD,* Hui Zheng, PhD,† and Paul G. Firth, MBChB*‡

Asphyxiation by an inhaled foreign body is a leading cause of accidental death among children younger than 4 years. We analyzed the recent epidemiology of foreign body aspiration and reviewed the current trends in diagnosis and management. In this article, we discuss anesthetic management of bronchoscopy to remove objects. The reviewed articles total 12,979 pediatric bronchoscopies. Most aspirated foreign bodies are organic materials (81%, confidence interval [CI] ϭ 77%–86%), nuts and seeds being the most common. The majority of foreign bodies (88%, CI ϭ 85%–91%) lodge in the bronchial tree, with the remainder catching in the larynx or trachea. The incidence of right-sided foreign bodies (52%, CI ϭ 48%–55%) is higher than that of left-sided foreign bodies (33%, CI ϭ 30%–37%). A small number of objects fragment and lodge in different parts of the airways. Only 11% (CI ϭ 8%–16%) of the foreign bodies were radio-opaque on radiograph, with chest radiographs being normal in 17% of children (CI ϭ 13%–22%). Although rigid bronchoscopy is the traditional diagnostic “gold standard,” the use of computerized tomography, virtual bronchoscopy, and flexible bronchoscopy is increasing. Reported mortality during bronchoscopy is 0.42%. Although asphyxia at presentation or initial emergency bronchoscopy causes some deaths, hypoxic cardiac arrest during retrieval of the object, bronchial rupture, and unspecified intraoperative complications in previously stable patients constitute the majority of in-hospital fatalities. Major complications include severe laryngeal edema or bronchospasm requiring tracheotomy or reintubation, pneumothorax, pneumomediastinum, cardiac arrest, tracheal or bronchial laceration, and hypoxic brain damage (0.96%). Aspiration of gastric contents is not reported. Preoperative assessment should determine where the aspirated foreign body has lodged, what was aspirated, and when the aspiration occurred (“what, where, when”). The choices of inhaled or IV induction, spontaneous or controlled ventilation, and inhaled or IV maintenance may be individualized to the circumstances. Although several anesthetic techniques are effective for managing children with foreign body aspiration, there is no consensus from the literature as to which technique is optimal. An induction that maintains spontaneous ventilation is commonly practiced to minimize the risk of converting a partial proximal obstruction to a complete obstruction. Controlled ventilation combined with IV drugs and paralysis allows for suitable rigid bronchoscopy conditions and a consistent level of anesthesia. Close communication between the anesthesiologist, bronchoscopist, and assistants is essential. (Anesth Analg 2010;111:1016–25)

spiration of foreign bodies by children is a common purges, bleeding, and emetics were largely ineffective. problem around the world. Asphyxiation from Mortality was estimated at 23%. This rate plummeted with Ainhaled foreign bodies is a leading cause of acci- the development of bronchoscopic techniques for the re- dental death among children younger than 4 years. During moval of these foreign bodies.1 In 1897, Gustav Killian, a the 19th century, treatment of foreign body aspiration by German otolaryngologist, performed the first bronchoscopy using a rigid esophagoscope to successfully remove a pig bone from a farmer’s right main bronchus.1,2 Algernon From the *Department of Anesthesia, Critical Care and Pain Medicine, and the †Biostatistics Center, Massachusetts General Hospital, Boston; and the Coolidge performed the first successful removal of a tracheal ‡Department of Anesthesia, Massachusetts Eye and Ear Infirmary, Boston. foreign body in the United States at the Massachusetts Gen- Christina W. Fidkowski is now affiliated with the Department of Anesthe- eral Hospital in 1898.1 Shortly thereafter, Chevalier Jackson siology, Henry Ford Hospital, Detroit, Michigan. developed the lighted bronchoscope and several special- Accepted for publication June 10, 2010. ized instruments for the removal of foreign bodies.3 He Statistical and administrative support funded by the Department of Anes- thesia, Critical Care and Pain Medicine, Massachusetts General Hospital, pioneered developments in the field and saved the lives of Boston, and the Department of Anesthesia, Massachusetts Eye and Ear hundreds of children who had aspirated objects.4 While Infirmary, Boston. early clinicians used topical anesthesia, general anesthesia Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions became commonplace for the removal of aspirated objects of this article on the journal’s Web site (www.anesthesia-analgesia.org). with increased experience with the rigid bronchoscope and Address correspondence and reprint requests to Paul G. Firth, Department advances in anesthetic delivery. The flexible bronchoscope of Anesthesia, Critical Care and Pain Medicine, Massachusetts General 5 Hospital, Boston, MA 02114. Address e-mail to [email protected]. was introduced by Shigeto Ikeda in 1966, and the removal Copyright © 2010 International Anesthesia Research Society of an airway foreign body using this instrument was 6 DOI: 10.1213/ANE.0b013e3181ef3e9c reported in the 1970s.

1016 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Anesthetic Considerations of Tracheobronchial Foreign Bodies

Over the years, the steady accumulation of clinical Table 1. Sensitivity, Speciﬁcity, Positive reports has provided greater insight into the management Predictive Value (PPV), and Negative Predictive of foreign body aspiration in children. In part as a result, Value (NPV) of a Witnessed Aspiration Effect for morbidity and mortality from foreign body aspiration have Foreign Body Aspiration drastically diminished. This article analyzes the recent Sensitivity Speciﬁcity PPV NPV epidemiology of foreign body aspiration, reviews the cur- Aydogan et al. 93.2 45.1 86.5 63.6 rent trends in diagnosis and management, and discusses (1887, 1493)7 anesthetic management for bronchoscopy. Ciftci et al. 91.1 46.0 90.5 47.9 (663, 563)11 Tomaske et al. 74.7 53.7 70.5 58.8 35 CURRENT TRENDS IN FOREIGN BODY ASPIRATION (370, 221) Ayed et al. 81.6 37.9 90.3 22.4 A preliminary Medline search of the literature from 1950 to (235, 206)8 2009 yielded nearly 20,000 cases of foreign body aspiration Tokar et al. 84.9 87.1 94.2 70.1 in children. Before 2000, however, most case series were (214, 152)34 small, included both aspirated and ingested objects, Skoulakis et al. 91.5 56.3 77.3 80.4 31 mingled adult and pediatric patients, or were reported in (210, 130) Kiyan et al. 37.3 96.3 96.6 35.1 various styles in differing specialist journals that precluded (207, 153)25 meta-analysis. In addition, anesthetic and surgical tech- Erikci et al. 58.3 87.1 90.2 50.5 niques have altered considerably in recent years, making a (189, 127)16 detailed review of older series less relevant. However, Heyer et al. 75.4 92.1 96.8 53.8 numerous large case series have been published recently (160, 122)20 that collectively allow for a clearer representation of the Cohen et al. 83.6 32.1 48.1 72.2 (142, 61)12 problem of pediatric aspiration, as well as current trends in management. A Medline search using the keywords foreign Values are percentages. Data were available from 10 of the 30 studies that were reviewed. Study size body aspiration with limits of the year 2000 to present was is denoted (n, n) to represent the total number of patients and the number of performed on October 1, 2009. Analysis was limited to patients with an aspirated foreign body, respectively. studies (1) containing only patients with suspected or proven foreign body aspiration, (2) with Ͼ100 patients, (3) Most (81%, CI ϭ 77%–86%) of the aspirated foreign containing only children ages 18 years and younger, and (4) bodies are organic materials.7,9–11,19–21,23,24,26–36 Nuts (es- written in English. Of the 698 articles obtained, 33 met the pecially peanuts) and seeds (mainly sunflower and water- inclusion criteria, of which 3 were excluded for containing melon) are the most commonly aspirated foreign bodies duplicate patient data. The 30 articles reviewed report reported in almost all studies. The exception to this finding 12,979 children with suspected foreign body aspiration, of is an Italian series that found teeth to be the most com- whom 11,145 were found to have aspirated an object.7–36 monly aspirated objects (33/121).13 In adolescent Turkish Twenty-seven of these studies are retrospective analyses, females, headscarf pins are commonly aspirated.24,34 As and 3 studies are prospective analyses of all children with was reported in 24 studies, the majority of foreign bodies suspected or actual foreign body aspiration.12,13,19 The (88%, CI ϭ 85%–91%) lodge in the bronchial tree, with the cases in these series occurred within the last 20 years in 21 7,9,10,12–19,22,25–36 8,9,11,12,14–18,20,21,23–26,28–31,33,34 remainder catching in the larynx or trachea. A studies, 8 covered a larger higher incidence of right-sided foreign bodies (52%, CI ϭ time span dating back to the 1980s,10,13,19,22,27,32,35,36 and 1 7 48%–55%) in comparison with left-sided foreign bodies (33%, study covered a 30-year time frame starting in 1973. CI ϭ 30%–37%) was reported in all of these studies, with the To obtain a robust estimate of the various rates (true exception being a small series in Israel.12 A small number of positive, gender, foreign body type and location, and objects fragment and lodge in different parts of the airways. radiographic outcome distributions), we applied a meta- A history of a witnessed choking event is highly sug- analysis to the published data to account for the number of gestive of an acute aspiration. Data were available to patients, the number of foreign body cases, and the number determine the sensitivity, specificity, positive predictive 7–36 of outcomes reported in these 30 articles. These meta- value, and negative predictive value of a witnessed event in analyses use a Bayesian model to consider variations in 10 studies (Table 1).7,8,11,12,16,20,25,31,34,35 Children with as- study design, inclusion and exclusion criteria, and the pirated foreign bodies typically present with the symptoms study population among the different reported studies. of coughing, dyspnea, wheezing, cyanosis, or stridor. Data Most patients with aspirated foreign bodies are children were available in 10 studies to determine the sensitivity and younger than 3 years. Four series reported the median age, the specificity of these presenting signs and symptoms and 7 series reported the mean age of children with (Table 2).8,12,16,20,23,25,29,31,34,35 A history of cough is highly aspirated foreign bodies. The median and the mean age sensitive for foreign body aspiration but is not very specific. ranged from 1 to 2 years12,18,21,33 and from 2.1 to 3.8 On the other hand, a history of cyanosis or stridor is very years,11,12,14,26,28,29,32 respectively. With the exception of 2 specific for foreign body aspiration but is not very sensitive. Turkish studies with a high incidence of adolescent girls Signs and symptoms typical in delayed presentations in- aspirating headscarf pins,24,34 boys account for 61% (confi- clude unilateral decreased breath sounds and rhonchi, dence interval [CI] ϭ 59%–63%) of children with foreign persistent cough or wheezing, recurrent or nonresolving body aspiration.7,9,17,21,22,26–29,32,33 pneumonia, or rarely, pneumothorax.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1017 REVIEW ARTICLE

Table 2. Sensitivity (Sens) and Speciﬁcity (Spec) of Symptoms for Foreign Body Aspiration Cough Dyspnea Wheeze Cyanosis Stridor Sens Spec Sens Spec Sens Spec Sens Spec Sens Spec Tomaske et al. (370, 221)35 87.8 45.0 57.9 73.2 39.4 74.5 Ayed et al. (235, 206)8 80.1 34.5 30.1 65.5 16.5 65.5 Tokar et al. (214, 152)34 94.1 32.3 27.6 66.1 Skoulakis et al. (210, 130)31 82.3 53.8 24.6 85.0 5.4 100 11.5 98.8 Kiyan et al. (207, 153)25 67.3 20.4 16.3 74.1 79.1 27.8 7.2 98.1 Erikci et al. (189, 127)16 51.2 83.9 4.7 93.5 18.9 93.5 Shivakumar et al. (165, 105)29 92.4 8.3 61.9 66.7 64.8 0 12.4 100 4.8 100 Heyer et al. (160, 122)20 41.0 55.3 33.6 68.4 Kadmon et al. (150, 80)23 51.3 12.9 18.8 72.9 Cohen et al. (142, 61)67 93.4 28.4 14.8 92.6 Values are percentages. Data were available from 10 of the 30 studies that were reviewed to determine the sensitivity (Sens) and speciﬁcity (Spec) of the symptoms of cough, dyspnea, wheeze, cyanosis, and stridor for foreign body aspiration. Study size is denoted (n, n) to represent the total number of patients and the number of patients with an aspirated foreign body, respectively.

Table 3. Sensitivity (Sens) and Specificity (Spec) of Radiographic Findings for Foreign Body Aspiration Air trapping Atelectasis Mediastinal shift Infiltrate Sens Spec Sens Spec Sens Spec Sens Spec Tokar et al. (214, 152)34 41.7 91.9 12.6 71.0 11.9 74.2 Skoulakis et al. (210, 130)31 39.2 91.6 9.2 88.8 0 76.3 Kiyan et al. (207, 153)25 63.8 79.6 8.0 94.4 4.4 94.4 Shivakumar et al. (165, 105)29 49.5 80.0 22.9 83.3 3.8 41.7 Heyer et al. (160, 122)20 62.3 97.4 8.2 97.4 20.5 97.4 18.9 84.2 Kadmon et al. (150, 80)23 50.0 90.0 38.8 97.1 Cohen et al. (142, 61)67 49.2 86.4 6.6 96.3 13.1 100 14.8 79.0 Values are percentages. Data were available from 8 of the 30 studies that were reviewed to determine the sensitivity (Sens) and specificity (Spec) of the radiographic findings of localized air trapping, atelectasis, mediastinal shift, and infiltrate for foreign body aspiration. Study size is denoted (n, n) to represent the total the number of patients and the number of patients with an aspirated foreign body, respectively.

Most stable patients had chest radiographs. As was re- Hasdiraz et al.19 used paralysis as needed during the ported in 20 studies, only 11% (CI ϭ 8%–16%) of the foreign procedure and attempted to maintain spontaneous ventila- bodies were radio-opaque.7,9,12–17,19,20,23–25,27–29,32,34–36 Chest tion when possible. On the other hand, Divisi et al.13 radiographs were normal in 17% (CI ϭ 13%–22%) of children commented that spontaneous ventilation is not suitable for with aspirated objects as were reported in 14 studies. rigid bronchoscopy because of resultant oxygen desatura- 9,11,12,17,22,24,25,27–29,32–35 The common radiographic abnor- tion and used a balance anesthetic with sevoflurane and malities included localized emphysema and air trapping, remifentanil. Shivakumar et al.29 used jet ventilation to atelectasis, infiltrate, and mediastinal shift. Data were avail- prevent oxygen desaturation. None of the authors com- able in 8 studies to calculate the sensitivity and specificity of mented on using drying drugs such as glycopyrolate before these radiographic findings (Table 3).11,12,20,23,25,29,31,34 Pneu- bronchoscopy. Seven studies commented on using steroids mothorax and pneumomediastinum are less common find- for laryngeal edema, with the majority of those authors ings on chest radiograph (range: 0.1%–3.7%), as was reported favoring steroid use only as needed,7,13,19,22,30 as opposed in 7 studies.14,15,22,28,32,33 to routine administration.25,36 In 4 studies, antibiotics were While rigid bronchoscopy was used solely for the re- given routinely preoperatively,19,25 postoperatively,36 or as moval of foreign bodies in most studies, both flexible and a 5-day course,30 whereas authors in 3 studies favored rigid bronchoscopies were used in 4 series.12,13,20,33 A antibiotic administration only as needed for infection.7,13,24 minority of foreign bodies were removed by flexible bron- Major iatrogenic complications were specified in 21 choscopy in 3 of these studies (range: 4.1%–10.7%),12,13,20 studies with 9437 children with aspirated foreign bodies. whereas Tang et al33. reported successful removal by The other 9 studies did not provide details or rates of flexible bronchoscopy in 91.3% of children with foreign complications. These complications included severe laryn- body aspiration. For this study, local anesthesia with seda- geal edema or bronchospasm requiring tracheotomy or tion was used during bronchoscopy. For children undergo- reintubation, pneumothorax, pneumomediastinum, cardiac ing rigid bronchoscopy, general anesthesia was used in all arrest, tracheal or bronchial laceration, and hypoxic brain studies, and details regarding the anesthetic technique damage. These major complications occurred in 91 of these were provided in 12 studies. Both inhaled7,15,31,36 and 9437 children (0.96%) (Table 4). Of the 11 cardiac arrests IV13,19 inductions were reported. Similarly, anesthesia was that were reported, 1 occurred after induction of anesthesia maintained with either inhaled15,19,31,36 or IV20,22,25 drugs in a child who was hypoxic on admission, 5 occurred or a balanced anesthetic.13 Five studies reported the use of during bronchoscopy because of hypoxia (3) or bleeding neuromuscular blockers.7,9,15,19,22 Bittencourt et al.9 and (2), and the remaining 5 were not specified. Other reported

1018 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Anesthetic Considerations of Tracheobronchial Foreign Bodies

were in children presenting with acute cyanosis and respi- Table 4. Morbidity Associated with Bronchoscopy 19 for the Removal of Tracheobronchial ratory distress. Hasdiraz et al. reported 8 deaths in 911 Foreign Bodies Turkish children (0.88%) with foreign body aspiration undergoing rigid bronchoscopy. Three children developed Complication Total n cardiac arrest from total tracheal obstruction, 2 had heart ϭ Major nonfatal complications (n 91) failure and bronchopneumonia at the time of bronchoscopy Severe laryngeal edema or bronchospasm 43 requiring tracheotomy or and developed cardiac arrest postoperatively, 1 developed reintubation15,19,22,25,27,30,32 respiratory arrest due to inhalation of cement powder, 1 Pneumothorax or 27 developed sepsis and respiratory failure after explosive pneumomediastinum7,11,15,18,19,22,30,32,33 release of purulent discharge from behind the foreign body, 11,15,25,27 Cardiac arrest 11 and 1 developed a respiratory arrest after a negative Hypoxic brain damage20,21 5 bronchoscopy and was found to have a tracheal foreign Tracheal or bronchial laceration requiring 5 22 repair11,15,27 body at autopsy. In 2008, Hui et al. reported 3 deaths Other serious complications (n ϭ 136) among 1428 children (0.21%) undergoing rigid bronchos- Infection13,19,21,26,32 58 copy over a 22-year period in China. Two died after foreign Failed bronchoscopy requiring thoracotomy 37 body displacement during bronchoscopy, and 1 died of (27)7,8,11,13,15,19,36 or tracheotomy (10)7,15 15,19,27 asphyxia during a delay before bronchoscopy. In 2009, Bleeding 29 33 Thoracotomy (5)8,15,19 or tracheotomy 12 Tang et al. reported no deaths among 1027 children in (7)7,32—not specified China undergoing bronchoscopy for foreign body removal. In that series, 938 foreign bodies were removed by flexible Major iatrogenic complications, as were specified in 21 studies, occurred in 91 of the 9437 children with aspirated foreign bodies. Other serious bronchoscopy, and 89 foreign bodies were removed by complications occurred in 136 of these 9437 children. rigid bronchoscopy. Of the remaining 18 deaths in the 21 other reports, 10 were due to irreversible cardiac arrest on admission.9,11,27,32,36 Table 5. Mortality Associated with Bronchoscopy for the Removal of Tracheobronchial Foreign Bodies DIAGNOSIS AND MANAGEMENT A suggestive history is important in diagnosing an aspi- n (43 ؍ Cause of death (n rated object, because it is often difficult to make a definitive Cardiac/respiratory arrest 37 Hypoxic arrest at presentation7,9,11,22,27,32,36 15 diagnosis on the basis of an abnormal physical examination Arrest due to tracheal foreign body19,26 5 or radiological studies alone. The work-up of the stable Cardiac arrest during bronchoscopy, not 3 patient should include a chest radiograph to assess for specified10,11 other potential causes of symptoms, to identify a radio- 19,29 Postoperative arrest 3 opaque foreign body, or to detect the position of a foreign Hypoxic arrest due to shifting foreign body22 2 Rupture of puss under pressure behind 1 body on the basis of localized emphysema and air-trapping, 37 foreign body19 atelectasis, infiltrate, or mediastinal shift. The common ab- Respiratory arrest due to inhaled cement 1 normality of unilateral hyperinflation seen on the chest radio- powder19 graph due to air trapping behind the foreign body is best 15 Not specified 7 viewed at end expiration (Fig. 1). (Video 1; see Supplemental Bronchial rupture15 2 Severe bronchospasm15 1 Digital Content 1, http://links.lww.com/AA/A169; see the Postoperative infection11 1 Appendix for video legends). Although a decubitous view has Multiorgan failure21 1 been suggested to look for air trapping in the dependent lung 18 Not specified 1 of small children who cannot cooperate with expiratory films, Deaths, as were specified in 26 studies, occurred in 43 of the 10,236 one study found this to be an unreliable technique.38 children with aspirated foreign bodies. Neck radiographs may be helpful in managing upper aerodigestive tract foreign bodies. The alignment of flat serious complications included infection, bleeding, and objects, such as coins, may suggest the location of an object failed bronchoscopic removal that necessitated thora- (Fig. 2A, 2B). Tracheal objects tend to align in the sagittal cotomy or tracheotomy to remove the object (Table 4). plane, whereas esophageal objects tend to align in the Mortality data were obtained from 26 articles with 43 anterior plane. An object that overlaps the boundaries of deaths among 10,236 children (0.42%) with aspirated for- the airway on an anterior–posterior view is unlikely to be eign bodies (Table 5). The remaining 4 articles did not inside the airway. Lateral radiographs may show soft tissue provide details of death rates. Twenty-five deaths occurred swelling, loss of cervical lordosis, or an object posterior to in the 5 largest series with 5927 children (0.42%).7,15,19,22,33 the trachea. Proximal esophageal objects can be removed In 2003, Eren et al.15 reported 10 deaths in 1068 children with a forceps under direct vision, with the laryngoscope (0.94%) undergoing rigid bronchoscopy for foreign body blade inserted into the esophagus to visualize the body and removal under general anesthesia in Turkey. Seven died of protect the airway during removal of the object. hypoxic arrest during bronchoscopy, 2 of bronchial rup- Thoracic computed tomography (CT) and virtual ture, and 1 of intractable bronchospasm. Shortly thereafter, bronchoscopy—a reformatted 3-dimensional CT image that their countrymen Aydogen et al. reported 4 deaths in 1493 generates intraluminal views of the airway to the sixth and children (0.27%) with foreign body aspiration undergoing seventh generation bronchi—are emerging as new modali- rigid bronchoscopy over a 31-year period.7 All 4 fatalities ties to diagnose tracheobronchial foreign bodies in children

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1019 REVIEW ARTICLE

Figure 1. A, Chest radiograph on end inspiration of a patient with a delayed presentation of an aspirated foreign body aspiration. B, Chest radiograph on end expiration. Delayed emptying of the left lung suggests local air trapping. The foreign body was in the left bronchus. C, The offending object seen on rigid bronchoscopy. The airway edema (white-gray) can be seen around the black foreign body, with bubbles reﬂecting delayed air release during expiration. A Fogarty catheter is passed beyond the object in preparation for dislodgement. Images courtesy of Dr. Dan Doody, Department of Surgery, Massachusetts General Hospital, Boston, Massachusetts.

Figure 2. A, The sagittal orientation of a proximal aerodigestive foreign body suggests an esophageal location. B, A lateral view can demonstrate a position posterior to the esophagus. Images courtesy of Dr. Allan Goldstein, Department of Surgery, Massachusetts General Hospi- tal, Boston, Massachusetts.

Figure 3. A, B, Computerized tomography scan of an aspirated soda can top, using a low-resolution pediatric protocol to minimize radiation exposure. The object was not seen on initial chest radiograph. A small aluminum object, although metal, has insufﬁcient radiopacity for a plain chest radiograph, and the object did not produce major obstruction leading to overt pulmonary changes. A computed tomography (CT) scan has a greater range of sensitivity. C, The offending object in the bronchus intermedius. (Images courtesy of Dr. Pallavi Sagar, Department of Radiology, and Dr. David Lawlor, Department of Surgery, Massachusetts General Hospital, Boston, MA.)

(Fig. 3).39,40 CT and virtual bronchoscopy are more sensi- false-positive findings. In a retrospective analysis, spiral CT tive diagnostic modalities for foreign body aspiration in correctly identified all 42 children with aspirated foreign comparison with conventional chest radiography.41,42 Se- bodies.41 In that study, 3 children had false-positive CT cretions, tumors, or other obstructive lesions can produce images due to excess bronchial secretions, and 6 children

1020 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Anesthetic Considerations of Tracheobronchial Foreign Bodies had true negative scans. In 2 retrospective studies, virtual a diagnostic flexible bronchoscopy. When this algorithm bronchoscopy correctly identified 11of 11 and 15 of 23 was applied retrospectively, the negative finding rate of children, respectively, with aspirated foreign bodies.42,43 rigid bronchoscopy decreased from 18% to 4% and from No false-positive virtual bronchoscopies were reported in 16% to 6%, respectively.47,48 No adverse events were re- those studies. The diagnostic utility of virtual bronchos- ported with flexible bronchoscopy.47,48 Therefore, diagnos- copy has also been shown prospectively.44,45 Haliloglu et tic flexible bronchoscopy in selected children minimizes the al.45 demonstrated that virtual bronchoscopy findings cor- potential complications of rigid bronchoscopies. More re- related with those of conventional bronchoscopy in 23 cently, Kadmon et al.23 proposed a computer model based children, of whom 7 had foreign body aspiration and 16 did on history, physical examination, and radiographic find- not have foreign body aspiration. In a prospective study of ings to calculate a score that predicts the likelihood of 37 children with suspected foreign body aspiration, 16 had foreign body aspiration in children. They further suggested a positive virtual bronchoscopy, of whom 13 had a foreign an algorithm to observe a child, perform diagnostic flexible body found with conventional bronchoscopy, and 3 had bronchoscopy, or perform therapeutic rigid bronchoscopy either mucous plugs or a schwannoma found with conven- on the basis of the calculated score. A prospective study is tional bronchoscopy.44 The remaining 21 patients had a warranted to determine the utility of this model. negative bronchoscopy and were observed with improve- In addition to aiding in the diagnosis of aspirated foreign ment in their symptoms.44 These studies demonstrate that bodies, flexible bronchoscopy is becoming more popular for CT and virtual bronchoscopy correctly identified all cases the removal of foreign bodies.33,49–52 In a large retrospective of foreign body aspiration. Therefore, some authors sug- study, a foreign body was successfully removed by flexible gested that children with a negative CT and virtual bron- bronchoscopy in 938 (91.3%) children.33 Flexible bronchos- choscopy may not require conventional bronchoscopy as a copy is better suited for removing foreign bodies from distal definitive work-up.44 airways and upper lobe bronchi, because of the smaller A drawback of CT and virtual bronchoscopy is the diameter and greater flexibility in comparison with the rigid potential for excessive radiation exposure. A chest radio- bronchoscope. Fewer instruments, however, are available for graph exposes the child to 0.1 mSv of radiation, equivalent use with the flexible bronchoscope to remove the foreign to several days of background environmental radiation. bodies. Rigid bronchoscopy continues to be used to remove Although a high-resolution pediatric chest CT can involve aspirated foreign bodies because multiple extraction instru- up to 7 mSv of radiation, a lower-resolution scan protocol ments are available and because it provides good visualiza- using 1.5 mSv is usually sufficient to diagnose a foreign tion, controls the airway, and allows ventilation. body. Adequate 3-dimensional views can be subsequently formatted from this level of detail. ANESTHETIC MANAGEMENT FOR Further limitations include the cost and limited avail- BRONCHOSCOPY ability of equipment and radiologists. In addition, CT Anesthetic considerations encompass preoperative assess- examination is limited to stable and cooperative children, ment, management techniques for flexible or rigid bron- because anesthesia in a remote location for a child with an choscopy, and postbronchoscopic disposition. unstable object that can potentially acutely obstruct the airway poses significant risks. Preoperative Assessment Although rigid bronchoscopy has traditionally been the The preoperative assessment should determine where the definitive method to diagnose and remove tracheobron- aspirated foreign body has lodged, what was aspirated, chial foreign bodies, a diagnostic flexible bronchoscopy and when the aspiration occurred. If the foreign body is under local anesthesia may be indicated for patients with- located in the trachea, the child is at risk for complete out a clear history or findings of aspiration.23,46–48 In a airway obstruction and should be taken urgently to the prospective study, children with convincing evidence of operating room. Conversely, the risk of complete airway foreign body aspiration were examined with rigid bron- obstruction is less if the object is firmly lodged beyond the choscopy under general anesthesia, whereas others with carina. It is important to determine the type of foreign less-suggestive findings underwent flexible bronchoscopy body: Organic materials can absorb fluid and swell, oils with local anesthesia.47 Of the 28 children who underwent from nuts cause localized inflammation, and sharp objects rigid bronchoscopy, 23 (82%) had a foreign body aspiration. can pierce the airway. The time since the aspiration should Of the 55 children who underwent flexible bronchoscopy, be established because airway edema, granulation tissue, only 17 (34%) had a foreign body aspiration. Another and infection may make retrieval more difficult with de- prospective study found that 43 (84%) of 51 children who layed presentations. A recently aspirated object may move underwent rigid bronchoscopy and only 7 (37%) of 19 to a different position with coughing. children undergoing flexible bronchoscopy had positive The time of the last meal should be established to assess studies for foreign body aspiration.48 Both studies found a the risk of aspiration. There are no reports of aspiration of significant association of aspirated foreign bodies with gastric contents in the literature surveyed, although fatal unilateral decreased breath sounds, localized wheezing, progression of obstruction has been reported.7,9,11,22,27,32,36 and obstructive emphysema on chest radiograph.47,48 In acute cases, therefore, the dangers of delayed removal These authors recommended that children undergo rigid appear to outweigh the risk of a full stomach in a well- bronchoscopy only if they have acute asphyxiation, a conducted anesthetic. In urgent cases, the stomach can be radio-opaque foreign body, unilateral pulmonary signs, or suctioned through a large-bore gastric tube after induction but obstructive emphysema. All other children should undergo before the bronchoscope is inserted to minimize the risk of

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1021 REVIEW ARTICLE gastric aspiration. In delayed presentations in which bron- bronchial tree. Movement can be prevented with neuromus- choscopy is not urgent, a preanesthetic fast is appropriate. cular blocking drugs9,54,56,57 or with a deep level of anesthesia. The airway patency should be assessed. If the patient is One study suggests that topicalization of the tracheobronchial in severe distress, urgent bronchoscopy should be per- mucosal using a rigid bronchoscope coated with local anes- formed. If the patient is stable, however, some authors thetic gel improves surgical conditions and more effectively suggest that bronchoscopy may be performed during nor- maintains spontaneous ventilation while decreasing the doses mal daytime operating hours to ensure optimal conditions of anesthetics.58 Although the risk of positive pressure venti- with an experienced bronchoscopist and anesthesiologist.53 lation causing distal air trapping by a ball-valve effect has These authors found no increase in morbidity in stable been suggested,59,60 there is no clear clinical evidence in the patients by delaying bronchoscopy for a suspected foreign literature surveyed to support this as a practical concern. body until the next available elective daytime slot.53 A retrospective review of 94 children with aspirated foreign bodies detected no difference in adverse events on the basis of the type of ventilation.61 However, 5 of 18 Anesthetic Considerations for Rigid children who were maintained on assisted ventilation and Bronchoscopy 11 of 26 who were maintained on spontaneous ventilation Because surgeon and anesthesiologist share management of a were switched to controlled ventilation. A prospective potentially obstructed airway, clear communication and good study of 36 children with aspirated foreign bodies found cooperation are essential. Before induction, a detailed anes- that controlled ventilation is more effective than is sponta- thetic and operative plan should be discussed. The 3 main neous ventilation.54 All children in the spontaneous venti- anesthetic issues involve the methods of induction, ventilation lation group were switched to either assisted or controlled during bronchoscopy, and maintenance of anesthesia. ventilation because of coughing and bucking. It is possible, The choice of induction is dominated by the consider- however, that the necessity of switching from spontaneous ation of converting a proximal partial obstruction into a to either assisted or controlled ventilation was due to an complete obstruction. The conversion from spontaneous inadequate depth of anesthesia with inhaled drugs rather negative pressure breathing to positive pressure ventilation than an inherent problem with spontaneous ventilation.62 theoretically risks dislodging an unstable proximal body, Larger prospective studies, with both inhaled and IV 54 causing complete obstruction. Although hypoxic arrest maintenance techniques, are necessary to further evaluate during the initial stages of bronchoscopy is a recognized whether spontaneous or controlled ventilation is more 10,11,19 cause of death, the relative contributions of obstruction advantageous. In a nonrandomized observational study, on initial presentation, during the induction of anesthesia, and manual jet ventilation was shown to decrease the incidence from dislodgement during bronchoscopy, are unclear from of intraoperative hypoxemia in comparison with manual published accounts. A survey of 838 pediatric anesthesiolo- controlled ventilation and spontaneous ventilation.63 gists found that the majority preferred an inhaled induction Manual jet ventilation may better allow oxygenation and when foreign bodies were present in the tracheobronchial ventilation of the unobstructed lung during manipulation 55 tree. A cautious IV induction that maintains spontaneous of the foreign body because the jet ventilation catheter was ventilation is also possible, although this was not an option in inserted separately from the bronchoscope.63 that particular survey study. While the optimal method of Halothane and sevoflurane are 2 volatile anesthetics that induction is not definitively established, maintaining sponta- are widely used in pediatric practice. Meretoja et al.64 neous ventilation during the induction of a patient with a compared sevoflurane with halothane in 120 children proximal foreign body is commonly practiced. undergoing bronchoscopy, gastroscopy, or combined After induction of general anesthesia, the rigid broncho- procedures. They reported a higher incidence of cardiac scope is inserted through the glottic opening. The anesthesia arrhythmias (nodal rhythm, bigeminy or ventricular circuit is connected to the sideport of the bronchoscope to ectopy) in the halothane group (18/60 vs. 4/60). Batra et allow ventilation. Both spontaneous ventilation and con- al.65 compared the 2 drugs in 44 children undergoing trolled ventilation are feasible for removal of foreign bodies. bronchoscopy specifically for foreign body removal and Spontaneous ventilation around the bronchoscope may be found a higher incidence of cardiac arrhythmias in the more suitable for removal of proximal bodies, during which halothane group (7/22 vs. 2/22). When comparing halo- leakage around the scope may make effective positive pres- thane and sevoflurane for 62 pediatric bronchoscopies, sure ventilation difficult. Manually closing the mouth and Davidson66 found no differences in cord closure, desatura- nose can diminish a large leak around the scope and improve tions, breath holding, or coughing. ventilation. Positive pressure ventilation down the broncho- Although inhaled drugs have traditionally been used for scope, with intermittent apnea while manipulating the object, the maintenance of anesthesia,54,59,61,67 total IV techniques may be more suitable for distal retrieval. The use of optical are becoming more popular in the pediatric popula- forceps allows for positive pressure ventilation to be main- tion.56,62,68,69 A total IV anesthetic with propofol (200 to 400 Ϫ Ϫ Ϫ tained while the foreign body is being manipulated so that ␮g ⅐ kg 1 ⅐ min 1) and remifentanil (0.05 to 0.2 ␮g ⅐ kg 1 ⅐ Ϫ periods of apnea can be minimized (Video 2; see Supplemen- min 1) infusions in combination with vocal cord topical- Ϫ tal Digital Content 2, http://links.lww.com/AA/A170; see ization with lidocaine (1 mg ⅐ kg 1) allows for spontaneous the Appendix for video legends). Because airway trauma and ventilation.62 Children younger than 3 years of age can rupture are significant and potentially fatal complications, it is tolerate higher doses of remifentanil and still maintain essential to avoid coughing and bucking secondary to the spontaneous ventilation in comparison with older chil- intense stimulation from a rigid bronchoscope deep in the dren.70 An advantage of an IV anesthetic is that it provides

1022 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Anesthetic Considerations of Tracheobronchial Foreign Bodies a constant level of anesthesia irrespective of ventilation. By mg/kg) or midazolam (0.1 to 0.15 mg/kg) was used for contrast, hypoventilation and leaks around the rigid bron- sedation in 938 young children who had a foreign body choscope may produce an inadequate depth of inhaled removed by flexible bronchoscopy.33 In that series, the anesthesia. Pollution of the operating room, due to the flexible bronchoscope was inserted intranasally unless na- combination of leaks around the rigid bronchoscope and sal stenosis was present.33 high gas flows needed for ventilation, are additional draw- In smaller children who are unable to cooperate, several backs of inhalation anesthetics. Chen et al.63 showed that a techniques of general anesthesia have been reported. A total IV technique with spontaneous ventilation was asso- balanced anesthetic using IV propofol and sevoflurane with ciated with a higher incidence of body movement, breath topical lidocaine and oxymetazoline was used for 23 chil- holding, and laryngospasm in comparison with an inhaled dren ages 9 months to 16 years.50 The fiberoptic broncho- technique. However, the doses of IV propofol (100 to 150 scope was then inserted through a T-piece on the child’s Ϫ Ϫ Ϫ Ϫ ␮g ⅐ kg 1 ⅐ min 1) and remifentanil (0.1 ␮g ⅐ kg 1 ⅐ min 1) facemask and advanced transnasally. In a series of 6 were less than those previously described to provide anes- children ages 1.2 to 5 years spontaneously breathing under thesia and maintain spontaneous ventilation. sevoflurane anesthesia, the bronchoscope was inserted Dropping the foreign body during retrieval is a poten- through a swivel adapter on a laryngeal mask airway.52 tially life-threatening complication.71,72 The vocal cords The foreign bodies were removed en bloc with the laryn- should be well relaxed, either by residual topicalization, geal mask with no adverse events. Flexible bronchoscopy paralysis, or an adequate depth of anesthesia, before re- through endotracheal tubes under general anesthesia is moval of the foreign body through the larynx. Dropping also described in which the foreign body, bronchoscope, the foreign body has a higher correlation with the experi- and endotracheal tube are removed en bloc.51 A standard ence level of the bronchoscopist than with the mode of pediatric bronchoscope (3.6 mm outer diameter) can be ventilation.72 If the object is dropped in the proximal used with a size 4.5 or larger endotracheal tube, whereas airway and cannot immediately be removed, pushing it standard adult bronchoscopes (4.9 mm diameter) will fit back into a bronchus can eliminate an obstruction. If a into size 2 or larger laryngeal mask. bronchial body falls into the other bronchus, there is potential for complete airway obstruction due to edema Postoperative Considerations and inflammation at the original site.71 In the setting of a Early discharge after uncomplicated bronchoscopy is rea- marginal airway, optimization of other components of sonable. In one study, 187 (65%) children were discharged ventilation is essential. Ventilation may be impaired not home within 4 hours after rigid bronchoscopy.78 In another only by the object, but also proximally by upper-airway soft study, 82 (60.7%) children had a hospital stay Ͻ1 day.32 tissue or cord closure, and distally by atelectasis after Prolonged pulmonary recovery may prevent early dis- prolonged intraoperative hypoventilation. Optimal head charge. Predictive factors of prolonged recovery included position, open cords, reinflation of atelectatic segments, evidence of inflammation on preoperative radiographs, and slow prolonged breaths with adequate pressure can aggravation of pulmonary lesions on postoperative films, provide ventilation past a partial obstruction. If ventilation and a prolonged duration of bronchoscopy.28,79 Ciftci et is impossible, emergent efforts must be made to extract or al.11 found bronchoscopy time (57 Ϯ 2.9 minutes vs. 23 Ϯ move the object. In severe cases of cardiopulmonary failure 1.2 minutes) to be prolonged in children with postoperative due to foreign body obstruction, extracorporeal membrane complications in comparison with those without complica- oxygenation may facilitate foreign body removal and car- tions. Chen et al.63 found that postoperative hypoxemia diopulmonary recovery.73 was associated with prolonged emergence from anesthesia After the extraction of the foreign body and the removal and with foreign bodies that were plant seeds. of the rigid bronchoscope, the choice of ventilation during emergence is influenced by pulmonary gas exchange and CONCLUSIONS the degree of airway edema. For uncomplicated cases, Aspiration of a foreign body is a potentially lethal event. spontaneous ventilation assisted by mask ventilation as Although many deaths occur before arrival at the hospital, needed may be adequate. Intubation during emergence anesthesia and bronchoscopy to remove the offending item may be indicated for a marginal airway, pulmonary com- are associated with considerable mortality and morbidity. promise, or residual neuromuscular blockade. Outcomes have improved over the years because of advances in anesthesia and bronchoscopy. Although several Anesthetic Considerations for anesthetic techniques are effective for managing children Flexible Bronchoscopy with foreign body aspiration, there is no consensus from Flexible bronchoscopy can be performed with local anes- the literature as to which technique is optimal. An induc- thetic topicalization and sedation in both children and tion that maintains spontaneous ventilation is commonly adults.47–49,74–77 IM meperidine and oral diazepam,75 IV practiced to minimize the risk of converting a partial midazolam or fentanyl,75 and atropine and diazepam49,76 proximal obstruction to a complete obstruction. Controlled or sublingual codeine74 have been successfully used to ventilation combined with IV drugs and paralysis allows sedate adolescents and adults. Topical lidocaine to the for suitable rigid bronchoscopy conditions and a consistent nasopharynx and larynx was combined with 0.1 to 0.3 level of anesthesia. The use of CT and virtual bronchoscopy mg/kg rectal midazolam for 19 younger children.48 Aero- to diagnose foreign body aspiration and the use of flexible solized lidocaine in combination with an IM dose of either bronchoscopy for the diagnosis and removal of foreign atropine (0.01 to 0.02 mg/kg) and diazepam (0.1 to 0.2 bodies may decrease the necessity for rigid bronchoscopy

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1023 REVIEW ARTICLE under general anesthesia in patients with suspected foreign 16. Erikçi V, Karaçay S, Arikan A. Foreign body aspiration: a body aspiration. As a result, morbidity and mortality in four-years experience. Ulus Travma Acil Cerrahi Derg these children may further decrease. Regardless of the 2003;9:45–9 17. Girardi G, Contador AM, Castro-Rodríguez JA. Two new management strategy, close cooperation within a skilled radiological findings to improve the diagnosis of bronchial surgical and anesthetic team is essential to avoid the foreign-body aspiration in children. Pediatr Pulmonol 2004;38: potential hazards of foreign body aspiration. 261–4 18. Gregori D, Salerni L, Scarinzi C, Morra B, Berchialla P, Snidero ACKNOWLEDGMENTS S, Corradetti R, Passali D, Klaus A, Isidor H, Gernot S, Jan B, We wish to thank the following for assistance in preparation of Bernard B, Karchev T, Tzolov T, Ranko M, Lana K, Ivo S, Mirko T, Caye-Thomasen P, Anne P, Volker J, Onder G, Simasko N, this manuscript: Daniel Doody, MD, Gennadiy Fuzaylov, MD, Matilda C, Christopoulos I, PassA˜ li GC, PassA˜ li F, Damiani V, Allan Goldstein, MD, Kenan Haver, MD, David Lawlor, MD, MieczysÅ,aw C, Dorin S, Gheorghe DC, Janka J, Miha Z, Ales and Pallavi Sagar, MD, all of the Massachusetts General G, Ales M, Lorenzo R, Javier C, Pontus S, Philippe P, Ahmed C, Hospital; Cory Collins, DO, and Christopher Hartnick, MD, Metin OT, Ozden CA, Riza D, John G, Peter R, Rupert O. both of the Massachusetts Eye and Ear Infirmary; and all of Foreign bodies in the upper airways causing complications Harvard Medical School, Boston, Massachusetts. and requiring hospitalization in children aged 0–14 years: results from the ESFBI study. Eur Arch Otorhinolaryngol APPENDIX: VIDEO CAPTIONS 2008;265:971–8 19. Hasdiraz L, Oguzkaya F, Bilgin M, Bicer C. Complications of Video 1. Rigid bronchoscopy down the left mainstem bronchus. bronchoscopy for foreign body removal: experience in 1,035 Bubbles formed by release of trapped air can be seen during cases. Ann Saudi Med 2006;26:283–7 spontaneous breathing. 20. Heyer CM, Bollmeier ME, Rossler L, Nuesslein TG, Stephan V, Video 2. An optical forceps is used to grasp and remove the object Bauer TT, Rieger CH. Evaluation of clinical, radiologic, and via rigid bronchoscopy. laboratory prebronchoscopy findings in children with suspected foreign body aspiration. J Pediatr Surg 2006;41:1882–8 REFERENCES 21. Higuchi O, Adachi Y, Ichimaru T, Asai M, Kawasaki K. Foreign 1. Clerf LH. Historical aspects of foreign bodies in the air and body aspiration in children: a nationwide survey in Japan. Int food passages. South Med J 1975;68:1449–54 J Pediatr Otorhinolaryngol 2009;73:659–61 2. Zaytoun GM, Rouadi PW, Baki DH. Endoscopic manage- 22. Hui H, Na L, Zhijun CJ, Fugao ZG, Yan S, Niankai ZK, Jingjing ment of foreign bodies in the tracheobronchial tree: predic- CJ. Therapeutic experience from 1428 patients with pediatric tive factors for complications. Otolaryngol Head Neck Surg tracheobronchial foreign body. J Pediatr Surg 2008;43:718–21 2000;123:311–6 23. Kadmon G, Stern Y, Bron-Harlev E, Nahum E, Battat E, 3. Jackson C. The Life of Chevalier Jackson: An Autobiography. Schonfeld T. Computerized scoring system for the diagnosis of New York: Macmillan, 1938 foreign body aspiration in children. Ann Otol Rhinol Laryngol 4. Jackson C. Foreign Bodies in the Air and Food Passages. 2008;117:839–43 Charted Experiences in Cases from no. 631 to no. 1155 at the 24. Karakoc F, Karadag B, Akbenlioglu C, Ersu R, Yildizeli B, Bronchoscopic Clinic. New Bedford: Reynolds the Printer, 1923 Yuksel M, Dagli E. Foreign body aspiration: what is the 5. Miyazawa T, Miyazu Y, Kanoh K, Iwamoto Y. Flexible bron- outcome? Pediatr Pulmonol 2002;34:30–6 choscopy historical perspective. In: edited by Beamis J, Mathur 25. Kiyan G, Gocmen B, Tugtepe H, Karakoc F, Dagli E, Dagli TE. P, Mehta A, eds. Interventional Pulmonary Medicine (1st ed., Foreign body aspiration in children: the value of diagnostic chap. 3). New York: M. Dekker, 2004:33–48 criteria. Int J Pediatr Otorhinolaryngol 2009;73:963–7 6. Dikensoy O, Usalan C, Filiz A. Foreign body aspiration: clinical 26. Mahafza T, Khader Y. Aspirated tracheobronchial foreign bodies: utility of flexible bronchoscopy. Postgrad Med J 2002;78:399–403 a Jordanian experience. Ear Nose Throat J 2007;86:107–10 7. Aydogan LB, Tuncer U, Soylu L, Kiroglu M, Ozsahinoglu C. 27. Roda J, Nobre S, Pires J, Esteˆvaõ MH, Fe´lix M. Foreign bodies Rigid bronchoscopy for the suspicion of foreign body in the in the airway: a quarter of a century’s experience. Rev Port airway. Int J Pediatr Otorhinolaryngol 2006;70:823–8 Pneumol 2008;14:787–802 8. Ayed AK, Jafar AM, Owayed A. Foreign body aspiration in 28. Roh JL, Hong SJ. Lung recovery after rigid bronchoscopic children: diagnosis and treatment. Pediatr Surg Int 2003;19:485–8 removal of tracheobronchial foreign bodies in children. Int 9. Bittencourt PF, Camargos PA, Scheinmann P, de Blic J. Foreign J Pediatr Otorhinolaryngol 2008;72:635–41 body aspiration: clinical, radiological findings and factors 29. Shivakumar AM, Naik AS, Prashanth KB, Shetty KD, Praveen associated with its late removal. Int J Pediatr Otorhinolaryngol DS. Tracheobronchial foreign bodies. Indian J Pediatr 2003;70: 2006;70:879–84 793–7 10. Brkic F, Delibegovic-Dedic S, Hajdarovic D. Bronchoscopic removal of foreign bodies from children in Bosnia and Herze- 30. Shubha AM, Das K. Tracheobronchial foreign bodies in infants. govina: experience with 230 patients. Int J Pediatr Otorhino- Int J Pediatr Otorhinolaryngol 2009;73:1385–9 laryngol 2001;60:193–6 31. Skoulakis CE, Doxas PG, Papadakis CE, Proimos E, 11. Ciftci AO, Bingol-Kologlu M, Senocak ME, Tanyel FC, Buyuk- Christodoulou P, Bizakis JG, Velegrakis GA, Mamoulakis D, pamukcu N. Bronchoscopy for evaluation of foreign body Helidonis ES. Bronchoscopy for foreign body removal in aspiration in children. J Pediatr Surg 2003;38:1170–6 children. A review and analysis of 210 cases. Int J Pediatr 12. Cohen S, Avital A, Godfrey S, Gross M, Kerem E, Springer C. Otorhinolaryngol 2000;53:143–8 Suspected foreign body inhalation in children: what are the 32. Tan HK, Brown K, McGill T, Kenna MA, Lund DP, Healy GB. indications for bronchoscopy? J Pediatr 2009;155:276–80 Airway foreign bodies (FB): a 10-year review. Int J Pediatr 13. Divisi D, Di Tommaso S, Garramone M, Di Francescantonio W, Otorhinolaryngol 2000;56:91–9 Crisci RM, Costa AM, Gravina GL, Crisci R. Foreign bodies 33. Tang LF, Xu YC, Wang YS, Wang CF, Zhu GH, Bao XE, Lu MP, aspirated in children: role of bronchoscopy. Thorac Cardiovasc Chen LX, Chen ZM. Airway foreign body removal by flexible Surg 2007;55:249–52 bronchoscopy: experience with 1027 children during 2000–2008. 14. Emir H, Tekant G, Besik C, Elicevik M, Senyuz OF, Buyukunal World J Pediatr 2009;5:191–5 C, Sarimurat N, Yeker D. Bronchoscopic removal of tracheo- 34. Tokar B, Ozkan R, Ilhan H. Tracheobronchial foreign bodies in broncheal foreign bodies: value of patient history and timing. children: importance of accurate history and plain chest radi- Pediatr Surg Int 2001;17:85–7 ography in delayed presentation. Clin Radiol 2004;59:609–15 15. Eren S, Balci AE, Dikici B, Doblan M, Eren MN. Foreign body 35. Tomaske M, Gerber AC, Stocker S, Weiss M. Tracheobronchial aspiration in children: experience of 1160 cases. Ann Trop foreign body aspiration in children—diagnostic value of Paediatr 2003;23:31–7 symptoms and signs. Swiss Med Wkly 2006;136:533–8

1024 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Anesthetic Considerations of Tracheobronchial Foreign Bodies

36. Yadav SP, Singh J, Aggarwal N, Goel A. Airway foreign bodies 57. Soysal O, Kuzucu A, Ulutas H. Tracheobronchial foreign body in children: experience of 132 cases. Singapore Med J 2007;48: aspiration: a continuing challenge. Otolaryngol Head Neck 850–3 Surg 2006;135:223–6 37. Pinto A, Scaglione M, Pinto F, Guidi G, Pepe M, Del Prato B, 58. Yu H, Yang XY, Liu B. EMLA cream coated on the rigid Grassi R, Romano L. Tracheobronchial aspiration of foreign bronchoscope for tracheobronchial foreign body removal in bodies: current indications for emergency plain chest radiog- children. Laryngoscope 2009;119:158–61 raphy. Radiol Med 2006;1112:497–506 59. Baraka A. Bronchoscopic removal of inhaled foreign bodies in 38. Assefa D, Amin N, Stringel G, Dozor AJ. Use of decubitus children. Br J Anaesth 1974;46:124–6 radiographs in the diagnosis of foreign body aspiration in 60. Farrell PT. Rigid bronchoscopy for foreign body removal: young children. Pediatr Emerg Care 2007;23:154–7 anaesthesia and ventilation. Paediatr Anaesth 2004;14:84–9 39. Sodhi KS, Saxena AK, Singh M, Rao KL, Khandelwal N. CT 61. Litman RS, Ponnuri J, Trogan I. Anesthesia for tracheal or virtual bronchoscopy: new non invasive tool in pediatric bronchial foreign body removal in children: an analysis of patients with foreign body aspiration. Indian J Pediatr ninety-four cases. Anesth Analg 2000;91:1389–91 2008;75:511–3 62. Buu NT, Ansermino M. Anesthesia for removal of inhaled 40. Veras TN, Hornburg G, Schner AM, Pinto LA. Use of virtual foreign bodies in children. Paediatr Anaesth 2005;15:533 bronchoscopy in children with suspected foreign body aspira- 63. Chen LH, Zhang X, Li SQ, Liu YQ, Zhang TY, Wu JZ. The risk tion. J Bras Pneumol 2009;35:937–41 factors for hypoxemia in children younger than 5 years old 41. Hong SJ, Goo HW, Roh JL. Utility of spiral and cine CT scans undergoing rigid bronchoscopy for foreign body removal. in pediatric patients suspected of aspirating radiolucent for- Anesth Analg 2009;109:1079–84 eign bodies. Otolaryngol Head Neck Surg 2008;138:576–80 64. Meretoja OA, Taivainen T, Raiha L, Korpela R, Wirtavuori 42. Kosucu P, Ahmetoglu A, Koramaz I, Orhan F, Ozdemir O, K. Sevoflurane-nitrous oxide or halothane-nitrous oxide for Dinc H, Okten A, Gumele HR. Low-dose MDCT and virtual paediatric bronchoscopy and gastroscopy. Br J Anaesth bronchoscopy in pediatric patients with foreign body aspira- 1996;76:767–71 tion. AJR Am J Roentgenol 2004;183:1771–7 65. Batra YK, Mahajan R, Bangalia SK, Chari P, Rao KL. 43. Huang HJ, Fang HY, Chen HC, Wu CY, Cheng CY, Chang CL. A comparison of halothane and sevoflurane for broncho- Three-dimensional computed tomography for detection of scopic removal of foreign bodies in children. Ann Card tracheobronchial foreign body aspiration in children. Pediatr Anaesth 2004;7:137–43 Surg Int 2008;24:157–60 66. Davidson A. The correlation between bispectral index and 44. Adaletli I, Kurugoglu S, Ulus S, Ozer H, Elicevik M, Kantarci F, airway reflexes with sevoflurane and halothane anaesthesia. Mihmanli I, Akman C. Utilization of low-dose multidetector Paediatr Anaesth 2004;14:241–6 CT and virtual bronchoscopy in children with suspected 67. Cohen SR, Geller KA. Anesthesia and pediatric endoscopy: the foreign body aspiration. Pediatr Radiol 2007;37:33–40 surgeon’s view. Otolaryngol Clin North Am 1981;14:705–13 45. Haliloglu M, Ciftci AO, Oto A, Gumus B, Tanyel FC, Senocak 68. Maddali MM, Badur RS, Fernando MS, Alsajwani MJ. Total ME, Buyukpamukcu N, Besim A. CT virtual bronchoscopy in contralateral atelectasis following rigid bronchoscopy in a the evaluation of children with suspected foreign body aspi- child with scarf pin aspiration. Paediatr Anaesth 2006;16:1095–7 ration. Eur J Radiol 2003;48:188–92 69. Perrin G, Colt HG, Martin C, Mak MA, Dumon JF, Gouin F. 46. Even L, Heno N, Talmon Y, Samet E, Zonis Z, Kugelman A. Safety of interventional rigid bronchoscopy using intravenous Diagnostic evaluation of foreign body aspiration in children: a anesthesia and spontaneous assisted ventilation. a prospective prospective study. J Pediatr Surg 2005;40:1122–7 study. Chest 1992;102:1526–30 47. Martinot A, Closset M, Marquette CH, Hue V, Deschildre A, 70. Barker N, Lim J, Amari E, Malherbe S, Ansermino JM. Rela- Ramon P, Remy J, Leclerc F. Indications for flexible versus tionship between age and spontaneous ventilation during rigid bronchoscopy in children with suspected foreign-body intravenous anesthesia in children. Paediatr Anaesth 2007;17: aspiration. Am J Respir Crit Care Med 1997;155:1676–9 948–55 48. Righini CA, Morel N, Karkas A, Reyt E, Ferretti K, Pin I, 71. Kumar S, Saxena AK, Kumar M, Rautela RS, Gupta N, Goyal Schmerber S. What is the diagnostic value of flexible bronchos- A. Anesthetic management during bronchoscopic removal of a copy in the initial investigation of children with suspected unique, friable foreign body. Anesth Analg 2006;103:1596–7 foreign body aspiration? Int J Pediatr Otorhinolaryngol 72. Pawar DK. Dislodgement of bronchial foreign body during 2007;71:1383–90 retrieval in children. Paediatr Anaesth 2000;10:333–5 73. Brown KL, Shefler A, Cohen G, DeMunter C, Pigott N, 49. Gencer M, Ceylan E, Koksal N. Extraction of pins from the Goldman AP. Near-fatal grape aspiration with complicating airway with flexible bronchoscopy. Respiration 2007;74:674–9 acute lung injury successfully treated with extracorporeal 50. Ramirez-Figueroa JL, Gochicoa-Rangel LG, Ramirez-San Juan membrane oxygenation. Pediatr Crit Care Med 2003;4:243–5 DH, Vargas MH. Foreign body removal by flexible fiberoptic 74. Mise K, Jurcev Savicevic A, Pavlov N, Jankovic S. Removal of bronchoscopy in infants and children. Pediatr Pulmonol tracheobronchial foreign bodies in adults using flexible bron- 2005;40:392–7 choscopy: experience 1995–2006. Surg Endosc 2009;23:1360–4 51. Swanson KL, Prakash UB, Midthun DE, Edell ES, Utz JP, McDou- 75. Tariq SM, George J, Srinivasan S. Inhaled foreign bodies in gall JC, Brutinel WM. Flexible bronchoscopic management of adolescents and adults. Monaldi Arch Chest Dis 2005;63:193–8 airway foreign bodies in children. Chest 2002;121:1695–1700 76. Uskul TB, Turker H, Arslan S, Slevi A, Kant A. Use of 52. Yazbeck-Karam VG, Aouad MT, Baraka AS. Laryngeal mask fiberoptic bronchoscopy in endobronchial foreign body re- airway for ventilation during diagnostic and interventional fibre- moval in adults. Turkish Resp J 2007;8:39–42 optic bronchoscopy in children. Paediatr Anaesth 2003;13:691–4 77. Wong KS, Lai SH, Lien R, Hsia SH. Retrieval of bronchial 53. Mani N, Soma M, Massey S, Albert D, Bailey CM. Removal of foreign body with central lumen using a flexible bronchoscope. inhaled foreign bodies—middle of the night or the next morn- Int J Pediatr Otorhinolaryngol 2002;62:253–6 ing? Int J Pediatr Otorhinolaryngol 2009;73:1085–9 78. Tomaske M, Gerber AC, Weiss M. Anesthesia and periinter- 54. Soodan A, Pawar D, Subramanium R. Anesthesia for removal ventional morbidity of rigid bronchoscopy for tracheobron- of inhaled foreign bodies in children. Paediatr Anaesth chial foreign body diagnosis and removal. Paediatr Anaesth 2004;14:947–52 2006;16:123–9 55. Kain ZN, O’Connor TZ, Berde CB. Management of tracheo- 79. Chung MK, Jeong HS, Ahn KM, Park SH, Cho JK, Son YI, Baek bronchial and esophageal foreign bodies in children: a survey CH. Pulmonary recovery after rigid bronchoscopic retrieval of study. J Clin Anesth 1994;6:28–32 airway foreign body. Laryngoscope 2007;117:303–7 56. Sersar SI, Rizk WH, Bilal M, El Diasty MM, Eltantawy TA, Abdelhakam BB, Elgamal AM, Bieh AA. Inhaled foreign bodies: presentation, management and value of history and plain chest radiography in delayed presentation. Otolaryngol Head Neck Surg 2006;134:92–9

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1025 Economics, Education, and Policy

Section Editor: Franklin Dexter SPECIAL ARTICLE

An Optimistic Prognosis for the Clinical Utility of Laboratory Test Data

Ming Zheng, PhD,* Palanikumar Ravindran, PhD,† Jianmei Wang, PhD,† Richard H. Epstein, MD,‡ David P. Chen, PhD,* Atul J. Butte, MD, PhD,* and Gary Peltz, MD, PhD*

It is hoped that anesthesiologists and other clinicians will be able to increasingly rely upon laboratory test data to improve the perioperative care of patients. However, it has been suggested that in order for a laboratory test to have clinically useful diagnostic performance characteristics (sensitivity and specificity), its performance must be considerably better than those that have been evaluated in most etiologic or epidemiologic studies. This pessimism about the clinical utility of laboratory tests is based upon the untested assumption that laboratory data are normally distributed within case and control populations. We evaluated the data distribution for 700 commonly ordered laboratory tests, and found that the vast majority (99%) do not have a normal distribution. The deviation from normal was most pronounced at extreme values, which had a large quantitative effect on laboratory test performance. At the sensitivity and specificity values required for diagnostic utility, the minimum required odds ratios for laboratory tests with a nonnormal data distribution were significantly smaller (by orders of magnitude) than for tests with a normal distribution. By evaluating the effect that the data distribution has on laboratory test performance, we have arrived at the more optimistic outlook that it is feasible to produce laboratory tests with diagnostically useful performance characteristics. We also show that moderate errors in the classification of outcome variables (e.g., death vs. survival at a specified end point) have a small impact on test performance, which is of importance for outcomes research that uses anesthesia information management systems. Because these analyses typically seek to identify factors associated with an undesirable outcome, the data distributions of the independent variables need to be considered when interpreting the odds ratios obtained from such investigations. (Anesth Analg 2010;111:1026–35)

n contemporary clinical practice, a large number of datasets from multiple institutions’ anesthesia information laboratory tests are performed to facilitate diagnosis, to management systems (AIMS) through data-mining tech- Iassess disease progression or the effect of a therapy, and niques could identify threshold parameters associated with for prognostication. With the advent of genomic science, an undesirable outcomes or to assess risk. even larger number of variables (mRNAs, proteins, and There is reason to believe that additional data and metabolites) can be measured in blood or tissues in a genetic risk factor measurements will improve our ability cost-effective way. For anesthesiologists, it is likely that to diagnose, stratify, and optimize the perioperative care of many different metabolites, proteins, or neural variables our patients. For example, a quantitative simulation dem- will be measured in the perioperative setting that could onstrated how the use of pharmacogenetic information to provide data useful for patient assessment and treatment. individualize drug dosage has the potential to significantly 1 Genetic risk factors can be measured to identify individuals improve treatment outcome. at increased risk for disease or perioperative complications, Nevertheless, several publications have suggested that or for predicting surgical outcome. Also, analyzing joined newly discovered laboratory tests would have limited diagnostic utility for individual patients. The argument advanced was that in order for a laboratory measurement From *Stanford University, Stanford, California; †Roche Palo Alto LLC, Palo Alto, CA; and ‡Jefferson Medical College, Philadelphia, Pennsylvania. to have clinically useful performance characteristics (sensi- Accepted for publication June 4, 2010. tivity and specificity), the magnitude of the odds ratio (OR) Gary Peltz was partially supported by funding (7R56 GMO68885 to 05) from for the test (Text Box) must be considerably higher than the NIGMS. Funding was provided by NLM T15 LM007033 (to David P. those seen in most etiologic or epidemiologic studies.2–4 Chen) and by the Lucile Packard Foundation for Children’s Health (to Atul For example, a hypothetical analysis by Ware2 indicated J. Butte). that an OR of 228 would be required for a test result to have Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions sufficient diagnostic or predictive utility (80% specificity of this article on the journal’s Web site (www.anesthesia-analgesia.org). and 80% sensitivity). Address correspondence to Gary Peltz, MD, PhD, Stanford University, 800 If this level of certainty is truly required, it would be Welch Road, Room 213, Palo Alto, CA 94304. Address e-mail to [email protected]. virtually impossible to identify a genetic test with accept- Copyright © 2010 International Anesthesia Research Society able performance for the vast majority of clinical situations. DOI: 10.1213/ANE.0b013e3181efff0c This is because ORs for most of the genetic risk factors for

1026 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 An Optimistic Prognosis for Clinical Utility of Lab Test quantitative traits or susceptibility to common diseases that adjusted P value was smaller than 0.05, the null hypoth- have been identified are usually much lower than the esis was rejected at the 5% significance level (i.e., the data proposed threshold. For example, analysis of the 1967 were not normally distributed). identified human single nucleotide polymorphisms with a a reported OR for a studied trait indicated that 67% have Biomarker Data Analysis ORs Ͻ5, and 95% have ORs Ͻ30. 4 For the 3 types of laboratory data studied in detail, deiden- In a similar negative vein, Pepe et al. investigated the tified data were obtained from the STRIDE. The Interna- relationship between the OR and classification accuracy. tional Classification of Diseases, Clinical Modification They analyzed hypothetical data with a normal distribution (ICD-9), codes for each individual with an available labo- and concluded that an OR of 74 was required to obtain ratory value were evaluated to identify the control and clinically useful performance characteristics (79% sensitiv- disease populations by using the STRIDE Anonymous ity and 79% specificity). Taken at face value, these 2 Patient Cohort Discovery Tool that was developed by the analyses provide a pessimistic outlook for the utility of Stanford Center for Clinical Informatics.10 The hemoglobin ϩ laboratory tests, because they imply that the performance A1C (immunoassay), CD19 cell counts (flow immunocy- of most laboratory tests will not be sufficient to have much tometry), and protein S activity (automated latex immuno- diagnostic or predictive utility. 2–4 assay) were measured using standard protocols in the However, these analyses are based on a fundamental clinical laboratory at Stanford University. assumption that laboratory data are normally distributed The following ICD-9 codes were used to classify indi- within case and control populations. They also assume viduals with available laboratory data: diabetes (250), lym- that currently used disease definitions and classification phoma (200,201), and coagulopathy (286). There were schema, developed before the discovery of genetic risk 33,958 A1C measurements from 20,590 nondiabetic indi- factors or novel laboratory tests, will continue to be viduals and 41,541 A1C measurements from 10,677 diabetic applied. individuals from the database. CD19ϩ cell counts included In this article, we demonstrate that the clinical utility of 17,706 measurements from 4498 individuals not diagnosed laboratory data has a much better prognosis than that as having a lymphoma, and 1861 measurements from 541 suggested by Ware and by Pepe et al. The vast majority of individuals with a lymphoma. Protein S activities included laboratory results do not have a normal distribution within 3701 measurements from 3385 individuals without a diag- either control or disease populations. As a consequence, the nosed coagulopathy and 519 measurements from 420 indi- performance characteristics (i.e., sensitivity and specificity viduals with a diagnosed coagulopathy. When a test was for disease diagnosis) of laboratory tests are substantially performed more than once on an individual, the value improved over Gaussian assumptions. Thus, laboratory obtained on the initial visit was used. (However, results data whose performance characteristics are within the were insensitive to the substitution of lab values from usual range observed in epidemiologic or etiologic studies subsequent visits.) can have substantial diagnostic utility. These consider- The A1C lab values were extensively right skewed; ations lead to a more optimistic assessment of the utility of therefore, a log-transformation was applied to make the laboratory tests in improving patient care. data distribution more symmetric. Because values of 0 were obtained for some CD19 and the protein S values, 1 was added to the original value before log transformation, METHODS because log(0) is undefined. All subsequent analyses were Analysis of Clinical Laboratory Test Data performed on the transformed data. All available data for 700 clinical laboratory tests per- Additional information about the OR determinations for formed between 2000 and 2006 were retrieved from the different data distributions and for A1C and other lab test Stanford Translational Research Integrated Database data, and for the other simulation studies, is provided in (STRIDE) according to a protocol that was approved by the the online Supplementary Methods (see Supplemental IRB. These tests were routinely performed at the Stanford Digital Content 1, http://links.lww.com/AA/A172) for and Lucile Packard Children’s Hospitals, and included this journal. patients between 0 and 108 years of age. At least 1000 measurements were available for each test (minimum 1001, maximum 2,090,227, median 3466). The Jarque–Bera test7 RESULTS was used to evaluate the normality of the data distribu- Clinical Laboratory Data Usually Are not tion in each test. The test was performed on the original Normally Distributed data and on the logarithm-transformed data, which is a Even if most of the population data appear to follow the transformation commonly used to analyze data with a shape of a normal distribution, the values at the tails significant rightward skewing. The larger of the 2 P (within the population extremes) may diverge markedly values obtained was reported. The Benjamini–Hochberg from that predicted by the normal distribution. For in- adjustment method was then used to adjust for multiple stance, the measured serum concentration of a protein testing.9 Because the null hypothesis was that the distri- could have a nonnormal distribution within the extremes bution was indeed normal, the smaller the adjusted P of a population because of a limitation in synthetic value, the greater the deviation from normal. When the capacity, a biological feedback mechanism that limits its maximum concentration, or a threshold level of stimula- ahttp://www.genome.gov/gwastudies, accessed on April 10, 2010. tion that may be required to initiate the synthesis of a

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1027 SPECIAL ARTICLE

Text Box: OR Deﬁnitions and Data Distribution. The OR compares the measured value for a risk factor within a population with a disease to that in a control population without the disease. Traditionally, the relationship between a binary test result and a binary outcome is represented in a two-way table. For example, the Bispectral Index (BIS) is used to titrate the administration of drugs during general anesthesia to minimize the risk of intraoperative awareness.5 This measure can be used to illustrate how a quantitative measure can be converted into a binary parameter (“BIS status”). Suppose analysis of an AIMS database that includes adverse outcome data leads to a putative test to predict the occurrence of intraoperative awareness if the BIS were more than 70 for more than 5 min during the interval from intubation to the end of surgery. Such a test might be useful to guide the rapidity with which BIS increases should be treated as well as to identify patients who need postoperative follow-up to assess and possibly treat the consequences of intraoperative awareness. Cases where the BIS was elevated as such are characterized as BISϩ and those where there were no such epochs as BIS-. Then, we can construct the following two-way table: Awareness Amnesia

BIS- n11 n12 ؉ BIS n21 n22 ϭ ϩ ϩ ϩ We have a total of n individuals (n n11 n12 n21 n22), where n11 of them are BIS- and have awareness during the surgery and n12,n21 and n22 are similarly defined according to the row and column labels for the corresponding cells in the table. Then, the odd of having awareness versus amnesia during the surgery in the BIS negative group is n11/n12. The odd of having awareness versus amnesia in the BIS positive group is n21/n22. The OR is then calculated as the ratio ϭ ϭ Ͼ of these two odds: OR (n11/n12)/(n21/n22) n11n22/n21n12. If the criterion of the BIS being more than 70 for 5 min were a good indicator of outcome, then n11 and n22 would be large, and n12 and n21 small, relative to n11 and n22.As a consequence, a large OR is obtained. In contrast, a small OR (close to 1) indicates that the test result has a small effect on outcome probability, indicating that it has a poor performance. Several other important measurements are frequently used to assess test performance. If we denote the individuals

having amnesia during the surgery as cases, and those with awareness as controls, then n11,n12,n21 and n22 are known as true negative (TN), false negative (FN), false positive (FP) and true positive (TP), respectively. Specificity is the probability that a control individual is correctly classified; and sensitivity is the probability that a case individual is correctly classified. Additionally, the positive predictive value and negative predictive value are the chances that an individual predicted as case (or control) is actually a case or control, respectively. The following table illustrates how these values are calculated:

Awareness Amnesia ؉ ؍ BIS- n11 (TN) n12 (FN) NPV TN/(TN FN) ؉ ؍ ؉ BIS n21 (FP) n22 (TP) PPV TP/(TP FP) ؍ Sensitivity ؍ Specificity (TN/(TN ؉ FP) TP/(TP؉FN However, by focusing on different case and control populations, different methods can be used to calculate the OR, which can markedly affect its value. In the analysis by Ware,2 the OR is defined as the ratio of the frequency of an event occurring within the group with the disease (cases) whose risk factor value places them at the 90th percentile (Figure T1, left panel, arrow labeled Group 2) relative to the frequency of events within the control group whose risk factor value is at the 10th percentile (Figure T1, left panel, arrow labeled Group 1). The difference in the mean frequencies between the case and control groups is used to calculate the OR for a risk factor. However, more conventionally, the OR is defined as the ratio of the odds of an event occurring in one group relative to the odds of it occurring in another group where all values are at or above a given value (90th percentile) in one group (Figure T1, right panel, orange area), and the values that are at or below a given value (10th percentile) in the second group (Figure T1, right panel, blue area). The conventional method for OR calculation can be more easily and accurately determined. The data distribution within a population also affects laboratory data performance. The data distribution for populations with a normal (Blue), a Double-Exponential (green) or a Cauchy (Red) distribution are shown in Figure T2. Note the differences at the extremes of the population distribution, where the Cauchy curve and the Double- Exponential curve do not tail-off as rapidly as does the normal curve. The maximum likelihood method is used to fit the actual data to different types of distributions. The fitted curve can be overlaid to the histogram. Visual inspection provides an estimate of the goodness-of-fit of the fitted distribution and statistical tests (Kolmogorov-Smirnov test6 for general distributions, and the Jarque-Bera7 or Shapiro-Wilk8 tests for normal distribution) can be used to rigorously assess the fit.

1028 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA An Optimistic Prognosis for Clinical Utility of Lab Test

Figure T1: Definition of odds ratio using the Ware (Left) and Conventional Definitions (Right). According to the Ware Definition, (2) the odds ratio is defined as the ratio of the frequency of an event occurring within the group with the disease (cases) whose risk factor value places them at the 90th percentile (arrow labeled in Group 2) relative to the frequency of events within the control group whose risk factor value is at the 10th percentile (arrow labeled in Group 1). The difference in the mean frequencies between the case and control groups is used to calculate the odds ratio for a risk factor. According to the Conventional Definition, the odds ratio is defined as the ratio of the odds of an event occurring in one group relative to the odds of it occurring in another group where all values are at or above a given value (90th percentile) in one group (orange area), and the values that are at or below a given value (10th percentile) in the second group (blue area).

Figure T2: The data distribution for populations with normal (Blue), Double- Exponential (green) or Cauchy (Red) distributions.

disease-associated protein. Any of these effects would Histograms of A1C lab values for 20,590 control (nondia- flatten the data distribution curve at the extremes of betic) individuals and 10,677 diabetic individuals reveal a control or diseased populations. This is important be- deviation of these data from a normal distribution in either cause calculation of ORs involves assessments at the tails population (Fig. 1). Although a superficial visual inspection of the distribution, whether one follows the calculation of the shape of the data distribution might seem to indicate method described by Ware or the more conventional a normal distribution, this is not a rigorous method for such approach (Text Box). determinations. Therefore, we used the Jarque–Bera test to To assess the possibility that the prior estimates of assess the normality of this data. The resulting P values are ORs required for adequate test performance are overly nearly zero for the case and control populations, indicating conservative because of deviations from the normal that these data have a highly nonnormal distribution, distribution at the tails, we evaluated the data distribu- which is consistent with the shape of the histograms. The tion of all 700 clinical laboratory tests. A detailed ex- same data are also graphed as quantile–quantile plots, ample follows. where the A1C lab values are plotted in relation to the The A1C test (HbA1c, glycated hemoglobin, or glycosy- percentile for the theoretical normal distribution and de- lated hemoglobin) is a commonly used laboratory test that partures from linearity indicate where the data do not have reflects the effectiveness of blood glucose regulation.11 a normal distribution. These plots demonstrate that the

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1029 SPECIAL ARTICLE

Figure 1. A1C lab values are not normally distributed in control (nondiabetic) or diabetic populations. Top panel, The histogram (left) and the quantile–quantile (QQ) plot (right) of the (log-transformed) A1C lab values in control populations. Bottom panels, The histogram (left) and the QQ plot (right) of the (log-transformed) A1C lab values in diabetic populations. In the histograms, the red curve represents the ﬁtted normal curve. In the QQ plots, the red lines indicate data with a normal distribution. The P values for the null hypothesis that the data are normally distributed are 0 for both the control and diabetic populations, indicating that the data have a highly nonnormal distribution.

distribution of A1C lab values at the extremes deviates distribution (Supplemental Figs. 1 [see Supplemental Digital significantly from the normal distribution, especially in the Content 2, http://links.lww.com/AA/A173 ] and 2 [see Supple- control population (Fig. 1). mental Digital Content 2, http://links.lww.com/AA/A174 ]; see We analyzed the raw data for all 700 laboratory Supplementary Methods section for figure legends, tests with the same methodology, and found that 699 http://links.lww.com/AA/A172). The dramatically different (99.9%) tests were not normally distributed. After log- shapes of these curves demonstrate that laboratory test per- transformation, the data in 694 (99.1%) tests remained non- formance (sensitivity and specificity) at a specified OR is normally distributed. The serum transferrin level was the only markedly altered if the data are not normally distributed. test whose data had a normal distribution; only 5 other tests Furthermore, the effect of a nonnormal data distribution (B-type natriuretic peptide, 1-hour glucose tolerance test, is especially pronounced under conditions in which high glucostatin, hematocrit, and total protein) were normally sensitivity and specificity are required. To illustrate this, we distributed after log-transformation. Thus, the data for nearly prepared a table showing the values of ORs calculated for all laboratory tests do not have a normal distribution. clinically useful levels of sensitivity and specificity for data A Nonnormal Distribution Signiﬁcantly Alters with a normal, a double-exponential or a Cauchy distribu- Laboratory Data Performance tion (Tables 1 and 2). The OR definition applied clearly Because the ORs and laboratory data performance are impacts laboratory test performance. In general, use of the assessed with data obtained from the extremes of a popu- Ware definition increased the ORs that were required for a lation (Text Box), the distribution of the data within the laboratory test to have clinically useful performance Ͼ population extremes has a large effect on its utility for (80%–90% sensitivity and 80%–90% specificity) by 10-fold disease diagnosis. For example, if laboratory data were in relation to the conventional OR definition. more flatly distributed at the extremes, the data distribu- However, independent of which OR definition was tion would resemble a double-exponential12 or a Cauchy used, the data distribution within the extremes of a popu- distribution13 (Text Box). These two types of data distribu- lation had a very significant effect on test utility and tion may better fit the rate of decay of the probability performance characteristics. A laboratory test with 80% density for the tails. Consequently, these distributions sensitivity and 90% specificity require an OR (Ware defini- better fit the actual data distribution at the extremes than tion) of 231 for normally distributed data. However, using does the normal distribution. the same assumptions as Ware (the data distribution in the To investigate the potential implications of this effect, case and control populations has the same shape), an OR of we plotted the OR as a function of the sensitivity and 15 or 25 can provide the same performance for data with a specificity for lab test data with a normal or a Cauchy Cauchy distribution or a double-exponential distribution,

1030 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA An Optimistic Prognosis for Clinical Utility of Lab Test

Table 1. Comparison of Odds Ratio Values Table 3. Performance Characteristics (Specificity Required for Laboratory Data with a Normal, and Sensitivity) of A1C Laboratory Test Data Double-Exponential, or Cauchy Distribution to No. correctly Achieve the Indicated Performance Characteristics identified Specificity (Sensitivity and Specificity) Control population Odds ratio Actual 17,319 84.1% (Ware definition) Normal distribution 13,880 67.4% Improvement ϩ3439 13.7% Double No. correctly Sensitivity Specificity Normal exponential Cauchy identified Sensitivity 80% 80% 75 25 11 Diabetic population 80% 90% 231 25 7 Actual 7532 70.5% 90% 80% 231 25 5 Normal distribution 7198 67.4% 90% 90% 713 25 4 Improvement ϩ334 3.1% The odds ratios were calculated as described in the METHODS section using Using 2.63 as the diagnostic cutoff, which corresponds to an odds ratio of the Ware (2) definition. Of note, regardless of the odds ratio definition 24.3, the number and percentage of individuals who were correctly classified used, the data distribution has a large impact on the required odds ratio for among the 20,590 control (nondiabetic) and 10,677 diabetic individuals are certain performance characteristics: the odds ratio required for normally shown. This analysis was repeated using data with a normal distribution that distributed data may be 1 or 2 magnitudes higher than that required for data had the same number of individuals and odds ratios. The number of with the double-exponential or Cauchy distribution. individuals who were correctly classified using the actual data was compared with that in the normally distributed data (improvement).

Table 2. Comparison of Odds Ratio Values Required for Laboratory Data with a Normal, To evaluate the impact that the data distribution had on Double-Exponential, or Cauchy Distribution to the required OR, we artificially shifted the distribution Achieve the Indicated Performance Characteristics curve for the A1C data in the case population to achieve the (Sensitivity and Specificity) desired performance characteristics. This enabled us to Odds ratio directly compare the calculated ORs from the actual A1C (conventional definition) data (after the hypothetical right shift) with those for data Double with a normal distribution. By varying the extent of the Sensitivity Specificity Normal exponential Cauchy rightward shift in the diabetic population, any desired level 80% 80% 433 38 8 of performance could be achieved, which enabled the 80% 90% 2415 100 11 90% 80% 2415 100 15 impact of the shape of the distribution curve (especially at 90% 90% 14,910 225 18 the 2 tails) on the OR (conventional definition) to be The odds ratios were calculated as described in the METHODS section using characterized (Table 3). the conventional definition. Of note, regardless of the odds ratio definition The required OR for the A1C test to achieve clinically used, the data distribution has a large impact on the required odds ratio for useful performance characteristics is significantly smaller certain performance characteristics: the odds ratio required for normally distributed data may be 1 or 2 magnitudes higher than that required for data than if the data had a normal distribution. This analysis with the double-exponential or Cauchy distribution. also indicated that a double-exponential distribution could reasonably approximate the properties of the A1C data respectively. The same trend was noted when the conven- distribution at the 2 tails. Thus, the nonnormal data distri- tional OR definition was applied. More dramatically, an OR bution at the extremes had a very significant effect on the of 14,910 is required to achieve 90% sensitivity and 90% performance of a laboratory test; the ORs required to specificity for data with a normal distribution, and an OR of achieve clinically useful performance characteristics were 225 or 6 is required for data that has a double-exponential significantly less than expected. or Cauchy distribution, respectively (Table 2). At the sensitivity and specificity values required for Nonnormal Data Distribution Improves clinical utility, the minimum required OR values for data A1C Performance with a Cauchy distribution are substantially smaller than We next evaluated the effect that the data distribution had those for data with a normal distribution. The same results on the performance of A1C laboratory data. To do this, we were consistently obtained when the shape of the data determined the number of individuals who would be distribution in the case and control groups were different correctly classified as control (nondiabetic) or diabetic on (Supplementary Table 1; see Supplementary Methods, the basis of actual A1C laboratory data, and compared that http://links.lww.com/AA/A172). with the expected results if the A1C data had a normal We further investigated whether the distribution of distribution (Table 3). Using the indicated cutoff of 2.63 actual laboratory data affects the OR that is required for a (which produced an OR of 24.3) for the A1C data, we laboratory test to achieve a certain level of diagnostic correctly identified 3439 more control (nondiabetic) indi- performance. Using a cutoff of 2.63 for the A1C lab values, viduals by this result than if the data had a normal this test would achieve 84% specificity and 71% sensitivity distribution. In addition, 334 more diabetic individuals for the diagnosis of diabetes in the populations evaluated. were correctly identified than if the A1C data were nor- Using the actual A1C data, the calculated OR was 24. If mally distributed (Table 3). In other words, the nonnormal AIC data were normally distributed, the OR required to distribution of the A1C laboratory data resulted in 13.7% achieve this performance would have been Ͼ10-fold fewer FP and 3.1% fewer FN classifications than if the data (271) higher. had followed a normal distribution.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1031 SPECIAL ARTICLE

Figure 2. Receiver operating characteristic (ROC) curves for (A) A1C, (B) CD19, (C) protein S actual, and (D) shifted protein S laboratory data. The ROC curve for the data distribution is shown in red, and the ROC curve for corresponding data with a normal distribution with the same odds ratio is shown in green. A, The areas under the ROC curve for the A1C and for the normal distribution data are 0.84 and 0.74, respectively. The specificity and sensitivity for the 2 points (triangle) are shown in Table 2. B, For CD-19 data, the actual odds ratio is 12.6, and the area under the curve (AUC) is 0.77. The corresponding AUC for normally distributed data with the same odds ratio is only 0.69. C, For protein S data, the actual odds ratio is 2.5, and the AUC is 0.6. The corresponding AUC for normally distributed data with the same odds ratio is 0.57. D, For protein S data, the data for the case group were artificially shifted to the right to achieve the performance of 80% specificity and 80% sensitivity. The odds ratio for the transformed data is 38, and the AUC is 0.86. The corresponding AUC for normally distributed data with the same odds ratio is 0.77.

Generation of a receiver operating characteristics (ROC) of 100% specificity and 100% sensitivity indicates nice curve is a useful procedure for assessing the overall per- performance, and a curve close to the diagonal line from 0% formance of a test. By varying the cutoff for A1C data, we specificity and 100% sensitivity to 100% specificity and 0% obtain different pairs of specificity and sensitivity. Instead sensitivity indicates a poor performance. The total area of focusing on a specific pair, we can plot all such pairs in under the ROC curve is also an indicator of the perfor- a scatter plot. The resulting curve is an ROC curve, showing mance, with a value of 1 indicating perfect classification the change of sensitivity with respect to the change in power and a value of 0.5 demonstrating that the variable specificity (Fig. 2A). An ROC curve close to the ideal point used for classification is irrelevant to the actual outcome.

1032 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA An Optimistic Prognosis for Clinical Utility of Lab Test

Table 4. Odds Ratios Required to Achieve the Table 5. Results of a Simulation Examining the Indicated Performance Characteristics (Sensitivity Impact of Misclassified Samples (% Mislabeled or and Specificity) Intercenter Differences in Data Distribution ⌬ Odds ratio Parameters [Intercenter ] on Biomarker (conventional definition) Sensitivity [at 80% Specificity]) Sensitivity Specificity A1C CD-19 Protein S Sensitivity 80% 80% 50 44 38 Actual Normal 80% 90% 165 139 87 % Mislabeled 90% 80% 165 71 61 0 74.5% 74.5% 90% 90% 332 235 129 1% 73.7%Ϯ0.1% 73.6%Ϯ0.5% The odds ratios were calculated using the conventional definition of the odds 3% 72.0%Ϯ0.2% 71.9%Ϯ0.5% ratio and under the assumption that the data distribution had the same shape 5% 70.4%Ϯ0.2% 70.1%Ϯ0.5% as that of A1C, CD-19ϩ cell count, or protein S activity. These distributions 7.5% 68.4%Ϯ0.3% 67.9%Ϯ0.5% were shifted in order to obtain the desired performance characteristics, as is 10% 66.3%Ϯ0.3% 65.6%Ϯ0.5% described in the METHODS section. Intercenter ⌬ 0 74.5% 74.5% The improved performance of the actual A1C data in 10% 74.1%Ϯ0.2% 74.3%Ϯ0.2% Ϯ Ϯ relation to that of the corresponding normal distribution 30% 71.9% 1.1% 72.6% 0.9% 50% 68.8%Ϯ2.1% 69.7%Ϯ2.0% is demonstrated by comparison of their ROC curves (Fig. 75% 64.3%Ϯ3.3% 64.8%Ϯ3.6% 2A); the area under the ROC curve for the A1C data 100% 60.4%Ϯ4.0% 59.9%Ϯ4.7% Ͼ (0.84) is 10% more than that of the normal distribution The results are shown for simulations using the actual A1C data or when it (0.74). was adjusted to have a normal distribution. The performances of 2 other laboratory tests were simi- ϩ larly evaluated. The number of CD19 cells is a potential biomarker for the diagnosis of lymphoma, and protein S distribution had on laboratory test performance when the activity is a potential biomarker for coagulopathy. The case and control populations were better separated by histograms and the quantile–quantile plots indicate that the test. In this case, there was a 10% gain in the AUC in the the data for both of these biomarkers have a nonnormal shifted data in relation to that of the normally distributed distribution (Supplementary Figs. 3 [see Supplementary data (Fig. 2D). These results further demonstrate that the Digital Content 4, http://links.lww.com/AA/A175] data distribution has a large impact on the performance of and 4 [see Supplementary Digital Content 5, laboratory test data. http://links.lww.com/AA/A176]; see Supplementary Methods for figure legends, http://links.lww.com/AA/A172). The Additional Factors Affecting the Laboratory ORs required for these 2 tests to achieve a useful level of Data Performance diagnostic performance were also significantly smaller than In addition to the data distribution, other factors can also if they had a normal distribution (Table 4). The required significantly impact the performance of laboratory data. ORs were comparable in magnitude to those for data For example, lactic dehydrogenase (LDH) is a prognostic following a double-exponential distribution, and were biomarker for survival in adult leukemia–lymphoma much larger than those for data following a Cauchy distri- cases.14 However, the response variable (survival at 5 bution. Therefore, the 2 tails of the distribution curves for years) could have an erroneous value in a small percentage the data in the control and disease groups play an impor- of patients if death were due to a nonleukemia-associated tant role in determining the relationship between biomar- cause (e.g., an automobile accident) or because of a failure ker performance and OR, and this data distribution was in mortality reporting. In the OR example presented in the better approximated by a double-exponential distribution. introductory section of this article, such misclassification The ROC curves of CD19 and protein S were also would represent errors due to patients having awareness compared with the ROC curves of a normally distributed under anesthesia that was not discovered by the inves- case and control population with the same OR. For the tigators, or patients claiming to have had awareness but CD19 data, there was a Ͼ10% gain in the area under the the events recalled did not take place during a general curve (AUC) for the actual CD19 data in relation to that of anesthetic. the data with a normal distribution and the same OR (Fig. In a simulation, we evaluated the impact of such mis- 2B). The performance of the actual data was superior to that classification of the response variable on laboratory data of the normally distributed data over most of the range. The performance. At 80% specificity, a 1% (or 10%) misclas- protein S data did not have good diagnostic utility because sification incidence decreased the average sensitivity the AUC of the ROC curve was close to 0.5. However, the from 74.5% to 73.7% (or 66.3%) (Table 5). The same ROC curve of the actual protein S data had a gain in AUC decrease occurred when the underlying data distribution in relation to that of the normally distributed data (Fig. 2C). was assumed to be normal. This simulation indicates that The distributions of the protein S data in the case and laboratory data performance decreases as the measure- control populations were close to each other, which made ment error in the binary dependent variable increases, the impact of data distribution hard to compare. Therefore, but the decrease is not very large if the measurement the protein S data in the case group was artificially shifted error is small. to the right to create a performance of 80% specificity and In many perioperative medicine studies, data are often 80% sensitivity, to investigate the effect that the data collected from multiple centers to achieve a desired sample

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1033 SPECIAL ARTICLE size. However, laboratories at different centers could pro- patients with diabetes. The scientist has just received a duce data with different statistical distribution parameters, large number of serum samples that were obtained from which introduces a degree of heterogeneity into the pooled control and diabetic individuals. After the A1C values were data. Therefore, we evaluated the impact of intercenter measured, the test developer must establish a threshold differences on a simulation study examining A1C data value that will enable physicians to determine whether an distributions obtained from 10 hypothetical centers. At 80% individual has an abnormal test result. As is shown here, specificity, the sensitivity of the A1C data only decreased the nonnormal distribution of A1C laboratory values re- from 74.5% (if all data were from a single center) to 74.1% sulted in a 13.7% increase in specificity and a 3.1% increase if the SD in the distribution parameter across centers was in sensitivity for the diagnosis of diabetes on the basis of 10% of the control A1C data (Table 5); and the shape of the A1C test results. Thus, for every 1 million diabetic indi- data distribution curve did not affect the size of the viduals that were evaluated by this newly developed test, if decrease in the sensitivity. Thus, combining data from the analyst used threshold values that were based upon a multiple centers had a small effect on laboratory data double-exponential distribution of A1C laboratory data in performance if the tests achieved a reasonable level of the population, 31,000 more test results would be correctly standardization across the centers. If care is taken to classified as abnormal. Similarly, when the test results from minimize the variation among different data centers, the a population of 1 million control individuals were ana- benefit from a larger sample size outweighs any decrease in laboratory data performance resulting from a multicenter lyzed, the use of these cutoff values would avoid misclas- study. sifying 137,000 individuals as having an abnormal test result. DISCUSSION However, diagnoses are not based solely upon a laboratory test result; clinical context and judgment will always The effect that a nonnormal data distribution has on laboratory test performance may at first appear to be a topic be essential for the proper use of any risk stratification tool. that is of interest to statisticians rather than physicians. The prevalence of a particular condition and the level of However, this finding has significant implications for sci- pretest suspicion within the tested population (pretest entists who are discovering and developing new diagnostic probability) will impact laboratory test performance and markers, and subsequently for physicians who will use the the interpretation of test results. For many genomic tests results provided by the next generation of diagnostic tests that an anesthesiologist might use, different clinical sce- to care for their patients. This information is also relevant to narios might lead to requirements for different test perfor- researchers using retrospective AIMS data to develop pre- mance characteristics (sensitivity and specificity). dictors of postoperative outcomes on the basis of quantita- For example, suppose that there was a genomic screen- tive data recorded during anesthesia. ing test that predicted susceptibility to malignant hyper- First, this analysis demonstrates that a nonnormal dis- thermia (MH). For this test, we would want a sensitivity of tribution of laboratory data enables laboratory tests with Ͼ99%, because this condition can be fatal and we do not moderate ORs (6–30) to provide useful diagnostic tools. want to miss anyone. However, we could accept a specific- Second, these ORs are within the range observed for ity of only 50% (for every 2 patients who are resistant to laboratory data identified in etiologic and epidemiologic MH, only 1 will be predicted correctly), because there are studies. This implies that contemporary genomic studies effective alternative methods of providing anesthesia using have the potential to produce clinically useful diagnostic non-MH triggering drugs. Alternatively, suppose another tools. Third, when evaluated within the larger context genetic test predicted whether a patient is likely to experi- of populations that are affected by common diseases ence prolonged sedation when given midazolam. For this Ͼ (prevalence 1% of the population), the improvement in test, we do not need an extremely high sensitivity, but want laboratory test performance due to the nonnormal data a higher level of specificity than for MH. We do not want to distribution assumes substantial significance. Finally, the deprive too many patients of the anxiolytic benefits of this evaluation of the usefulness of a marker (i.e., the indepen- drug, and we can effectively reverse the effect of midazo- dent variable) in predicting an outcome (i.e., the dependent lam with an antagonist (flumazenil), if necessary. However, variable) requires a determination of the probability distri- knowing whether a patient is at risk for prolonged sedation bution of that marker in the population. It is insufficient to would be helpful in assessing a patient who is slow to make assumptions that the marker follows a normal distribution without evaluating this formally, because the failure awaken after an anesthetic, and it could improve outcome. to do this may result in a useful test being inappropriately If the patient were at risk for excessive sedation, the discarded. This caveat applies equally to quantitative fac- practitioner might more quickly reach a decision to admin- tors identified retrospectively from data-mining analyses of ister flumazenil than if the patient were not at risk. Avoid- AIMS databases to be associated with an undesirable ing the administration of flumazenil in situations in which outcome. slow emergence is not likely due to midazolam is desirable, For example, it is estimated that 18.2 million people because provoking a seizure is a potential complication of (6.3% of the population) in the United States had diabetes reversal. These 2 different scenarios indicate how very in 2002, and 5.2 million of these were undiagnosed cases.15 different test performance characteristics can be required to Let us imagine that a scientist has just discovered that address different clinical situations. measurement of A1C could be a potential diagnostic The results from this study are especially important for marker that could be used for long-term monitoring of organizations such as the Anesthesia Quality Institute

1034 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA An Optimistic Prognosis for Clinical Utility of Lab Test

(AQI)b and the Multicenter Perioperative Outcomes group may be redefined using genetic risk factors and laboratory (MPOG),c both of which have recently initiated efforts to test measurements. When emerging laboratory test results improve the quality of anesthesia care through the retro- and genetic data are incorporated into disease classification spective analysis of anesthesia data. The finding that up to criteria, patients with common underlying predispositions a 10% error in survival classification has a small impact on and pathogenesis will be similarly classified. The sensitiv- test performance is fortunate. Typically, the cause of death ity and specificity of the laboratory results that are used for cannot be determined from mortality databases that are diagnosis and prognosis will then improve. publically available, and there is also incomplete reporting to these databases. If small reporting errors adversely ACKNOWLEDGMENTS affected test performance, misclassification of the cause of We thank Dr. Gomathi Krishnan for retrieving the biomarker death would call into question the interpretation of such data, and Dr. Bob Lewis for helpful discussions. studies. Also encouraging for AIMS research involving REFERENCES collaborative databases is that statistical variation in the 1. Guo Y, Shafer S, Weller P, Usuka J, Peltz G. Pharmacogenomics distributions of reported parameters among various con- and drug development. Pharmacogenomics 2005;6:857–64 tributing centers are not likely to have a major impact on 2. Ware JH. The limitations of risk factors as prognostic tools. the analysis. However, AQI and MPOG need to consider New Engl J Med 2006;355:2615–7 3. Wald NJ, Hackshaw AK, Frost CD. When can a risk factor be the implications of the data distribution of identified inde- used as a worthwhile screening test? BMJ 1999;319:1562–5 pendent variables and report on these in their published 4. Pepe MS, Janes H, Longton G, Leisenring W, Newcomb P. reports. The data distributions of the independent variables Limitations of the odds ratio in gauging the performance of a will also affect power analysis calculations (which usually diagnostic, prognostic, or screening marker. Am J Epidemiol 2004;159:882–90 are based on assumptions that the data are normally 5. Sigl JC, Chamoun NG. An introduction to bispectral analysis distributed) for subsequent prospective studies, because for the electroencephalogram. J. Clin Monit 1994;10:392–404 group sizes necessary to demonstrate a specified change in 6. Conover WJ. Practical Nonparametric Statistics. New York: outcome will be influenced. John Wiley & Sons, 1971 7. Jarque C, Bera A. Efficient tests for normality, homoscedastic- The prognosis for laboratory data is also further im- ity and serial independence of regression residuals. Econ Lett proved when viewed through a future-oriented lens. First, 1980;6:255–9 although this analysis evaluated commonly available labo- 8. Royston P. An extension of Shapiro and Wilk’s W test for ratory data, it has significant implications for newly emerging normalit to large samples. Appl Stat 1982;31:115–24 diagnostic tests identified using contemporary transcriptional, 9. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat proteomic, or metabolomic technologies. It is likely that the Soc B 1995;57:289–300 data for these newly discovered tests would also have a 10. Lowe HJ, Ferris TA, Hernandez PM, Weber SC. STRIDE—An similar (nonnormal) distribution, and we also have reason to integrated standards-based translational research informatics be more optimistic about their diagnostic performance. Sec- platform. AMIA Ann Symp Proc 2009;2009:391–5 11. Larsen ML, Horder M, Mogensen EF. Effect of long-term ond, this analysis evaluated the performance of a single monitoring of glycosylated hemoglobin levels in insulin- laboratory test. However, it is likely that many laboratory tests dependent diabetes mellitus. N Engl J Med 1990;323:1021–5 will be identified by contemporary genomic analyses. 12. Norton RM. The double exponential distribution: using calcu- As has been noted for disease-associated gene expres- lus to find a maximum likelihood estimator. Am Stat sion signatures,16 it is likely that a combination of genetic 1984;38:135–6 13. Cassela G, Berger R. Statistical Inference. Pacific Grove, CA: risk factors and laboratory test data will produce better Duxbury Press, 2002 clinical stratification tools. Fortunately, many statistical 14. Tsukasaki K, Hermine O, Bazarbachi A, Ratner L, Ramos JC, learning methods and modeling tools,17 including nonlin- Harrington W Jr, O’Mahony D, Janik JE, Bittencourt AL, Taylor ear mixed-effects modeling,18 have been developed to GP, Yamaguchi K, Utsunomiya A, Tobinai K, Watanabe T. Definition, prognostic factors, treatment, and response criteria identify optimal combinations for clinical stratification. of adult T-cell leukemia–lymphoma: a proposal from an inter- Also, this analysis evaluated test performance within the national consensus meeting. J Clin Oncol 2009;27:453–9 context of existing systems for disease classification. Cur- 15. Centers for Disease Control and Prevention. National Diabetes rently, very large and diverse groups of individuals are Fact Sheet: National Estimates on Diabetes. Atlanta, GA: Centers for Disease Control and Prevention, 2005 included within a disease grouping that is largely based 16. van de Vijver MJ, He YD, van’t Veer LJ, Dai H, Hart AA, upon a collection of symptoms, which often are disparate, Voskuil DW, Schreiber GJ, Peterse JL, Roberts C, Marton MJ, and sometimes incorporating various laboratory variables. Parrish M, Atsma D, Witteveen A, Glas A, Delahaye L, van der However, genetic risk factors and laboratory data have the Velde T, Bartelink H, Rodenhuis S, Rutgers ET, Friend SH, potential to transform our diagnostic and stratification Bernards R. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002;347:1999–2009 practices. In the not too distant future, patient groupings 17. Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning. New York: Springer, 2001 bhttp://www.aqihq.org/resources.aspx, last accessed June 2, 2010. 18. Lindstrom MJ, Bates DM. Nonlinear mixed effects models for chttp://mpog.med.umich.edu/, last accessed June 2, 2010. repeated measures data. Biometrics 1990;46:673–87

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1035 General Article A Comparison of Liver Function After Hepatectomy with Inﬂow Occlusion Between Sevoﬂurane and Propofol Anesthesia

J. C. Song, MD,* Y. M. Sun, MD,* L. Q. Yang, MD,* M. Z. Zhang, MD,† Z. J. Lu, MD,* and W. F. Yu, MD*

BACKGROUND: In this study, we compared liver function tests after hepatectomy with inflow occlusion as a function of propofol versus sevoflurane anesthesia. METHODS: One hundred patients undergoing elective liver resection with inflow occlusion were randomized into a sevoflurane group or a propofol group. General anesthesia was induced with 3 ␮g/kg fentanyl, 0.2 mg/kg cisatracurium, and target-controlled infusion of propofol, set at a plasma target concentration of 4 to 6 ␮g/mL, or sevoflurane initially started at 8%. Anesthesia was maintained with target-controlled infusion of propofol (2–4 ␮g/mL) or sevoflurane (1.5%–2.5%). The primary end point was postoperative liver injury assessed by peak values of liver transaminases. RESULTS: Transaminase levels peaked between the first and the third postoperative day. Peak alanine aminotransferase was 504 and 571 U/L in the sevoflurane group and the propofol group, respectively. Peak aspartate aminotransferase was 435 U/L after sevoflurane and 581 U/L in the propofol group. There were no significant differences in peak alanine aminotransferase or peak aspartate aminotransferase between groups. Other liver function tests including bilirubin and alkaline phosphatase, and peak values of white blood cell counts and creatinine, were also not different between groups. CONCLUSIONS: Sevoflurane and propofol anesthetics resulted in similar patterns of liver function tests after hepatectomy with inflow occlusion. These data suggest that the 2 anesthetics are equivalent in this clinical context. (Anesth Analg 2010;111:1036–41)

nflow occlusion by clamping of the portal triad (Pringle occlusion between July 2009 and December 2009 were maneuver) is routinely used in many centers1–3 to assessed for study eligibility. Exclusion criteria were age Iprevent blood loss during liver transsection.4,5 How- Ͻ18 years, additional ablation therapies (cryosurgery or ever, the Pringle maneuver induces ischemic injury in the radiofrequency ablation), prior liver resection for donation, remnant liver, which is associated with increased morbid- or scheduled resection not requiring inflow occlusion. 6 ity and mortality. Diseased livers such as steatotic or Enrolled patients were randomized at the beginning of the fibrotic livers may be the most vulnerable to temporary operation into a sevoflurane group (inhaled anesthesia 7–9 interruption of blood flow. with sevoflurane) or a propofol group (target-controlled Although volatile anesthetics and propofol have been infusion of propofol). All other anesthetic and surgical studied and compared in ischemia/reperfusion injuries in management was the same. The randomization sequence many organ systems, few studies have compared volatile without any stratification was generated by computer and anesthetics and propofol for liver resection with inflow sealed with consecutively numbered envelopes providing occlusion. In this study, we compared sevoflurane- and concealment of random allocation. The study was ap- propofol-based anesthetics for hepatectomy with inflow proved by our local institutional research ethics committee. occlusion, using indices of postoperative liver function as Written informed patient consent was obtained from all the major end point for comparison. participants. All patients received oral midazolam (5.0 mg) and IM METHODS atropine (0.5 mg) as a premedication. Electrocardiogram, One hundred consecutive ASA physical status I/II/III radial arterial blood pressure, arterial oxygen saturation, patients undergoing elective liver resection with inflow and bispectral index were monitored routinely. Epidural catheters were placed at T7-10 interspaces before anesthe- From the *Department of Anesthesiology, Eastern Hepatobiliary Surgery Hospital, Second Military Medical University; and †Department of Anes- sia, and a test dose of lidocaine was injected. A bolus of 7 to thesiology, Shanghai Children’s Medical Center, Shanghai Jiao Tong Uni- 10 mL ropivacaine 0.75% was administered epidurally after versity School of Medicine, Shanghai, China. induction of anesthesia. In the propofol group, general Accepted for publication June 23, 2010. anesthesia was induced with 3 ␮g/kg fentanyl, followed by Supported by a grant from the National Natural Science Foundation of China (No. 30901394). a target-controlled infusion of propofol, set at a plasma ␮ Disclosure: The authors report no conflicts of interest. target concentration of 4 to 6 g/mL, and 0.2 mg/kg Address correspondence and reprint requests to W.F. Yu, MD, Department cisatracurium. In the sevoflurane group, anesthesia was of Anesthesiology, Eastern Hepatobiliary Surgery Hospital, Second Military induced with 3 ␮g/kg fentanyl, sevoflurane initially started Medical University, Changhai Rd., No. 225, Shanghai, China. Address e-mail to [email protected]. at 8%, and 0.2 mg/kg cisatracurium. After tracheal intuba- Copyright © 2010 International Anesthesia Research Society tion, anesthesia was maintained with a target-controlled DOI: 10.1213/ANE.0b013e3181effda8 infusion of propofol (in the propofol group, as above) or

1036 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 sevoflurane (1.5%–2.5%), fentanyl 1 to 2 ␮g/kg, and cisa- Table 1. Patient Characteristics and Preoperative tracurium 5- to 10-mg boluses according to clinical needs. Laboratory Values To keep arterial mean blood pressure at a target of 60 mm Ϫ Ϫ Hg, we administered 2 to 5 ␮g ⅐ kg 1 ⅐ min 1 dopamine as Sevoflurane group Propofol group (50 ؍ n) (50 ؍ n) indicated. Depth of anesthesia was determined with the Age (y) 48.5 (8.9) 51.4 (7.8) bispectral index with a target range between 35 and 45 Body height (cm) 170.3 (3.7) 167.5 (10.3) during surgery. Epidural catheters were removed after the Weight (kg) 70.3 (9.8) 68.9 (12.1) operation. Intravenous analgesia (continuous IV infusion of Gender, male/female 40/10 36/14 1000 mg tramadol and 100 ␮g sufentanil over 48 hours) was Baseline ALT (U/L) 40 (13) 43 (26) Baseline AST (U/L) 36 (11) 46 (29) continued in all patients postoperatively. Baseline bilirubin (␮mol/L) 14.6 (5.2) 13.5 (5.9) Surgical procedures were performed in a standardized Baseline WBC count (ϫ103/mL) 5.9 (1.8) 6.1 (2.1) manner under the supervision of 2 experienced hepatobili- Baseline creatinine (␮mol/L) 74.6 (10.1) 70.3 (12.3) Baseline ALP (U/L) 92 (41) 103 (57) ary surgeons. After mobilization of the liver, inflow occlu- a sion was achieved by the tourniquet technique around the MELD score 6.85 (1.01) 6.94 (0.82) Cirrhosis, yes/nob 33/17 27/23 portal triad with a 4-mm Mersilene tape. During resections, Malignant/benign disease 45/5 43/7 a low central venous pressure (0–5 mm Hg) was main- Hepatocellular carcinomas 36 36 tained. The length of time for continuous inflow occlusion Intrahepatic 97 was determined by the surgeons. All patients received the cholangiocarcinoma Intrahepatic bile duct stone 2 3 same chemotherapy regimen before and after surgery. Angioleiomyolipoma of liver 3 4 Each patient was followed for the entire hospitalization. The primary end point was postoperative hepatocyte injury Data are expressed as mean (SD). ALT ϭ alanine aminotransferase; AST ϭ aspartate aminotransferase; WBC ϭ defined by peak alanine aminotransferase (ALT) and aspar- white blood cell; ALP ϭ alkaline phosphatase. tate aminotransferase (AST) levels over 6 postoperative days. a MELD, the Model for End-Stage Liver Disease, consists of serum bilirubin Additional end points were peak values of white blood cells, and creatinine levels, international normalized ratio (INR) for prothrombin time, and etiology of liver disease. The formula for the MELD score is 3.8 ϫ ͓ ͔ ϩ ϫ ϩ ϫ bilirubin, alkaline phosphatase, creatinine levels, and length loge(bilirubin mg/dL ) 11.2 loge(INR) 9.6 loge(creatinine of hospital stay. All outcome variables were measured before ͓mg/dL͔) ϩ 6.4 ϫ (etiology: 0 if cholestatic or alcoholic, 1 otherwise). b and 1, 3, and 6 days after surgery. Additionally, cirrhosis Cirrhosis (yes/no) was defined by histologic evaluation. (yes/no) was defined by histologic evaluation. Group sample size was calculated based on differences Table 2. Intraoperative Data from Patients in postoperative peak ALT concentration in a pilot study of Undergoing Hepatectomy with Inflow Occlusion Ϯ patients who received propofol anesthesia (685 392 U/L) Sevoflurane group Propofol group (50 ؍ n) (50 ؍ and sevoflurane anesthesia (487 Ϯ 308 U/L). The following (n formula: n ϭ 15.7/ES2 ϩ 1, where ES ϭ effect size ϭ Pringle time (min) 21.4 (8.5) 18.4 (6.3) (difference between groups)/(mean of the SD between Operation time (min) 136.0 (38.4) 124.3 (29.1) groups), with ␣ ϭ 0.05 and power ϭ 0.8 was used to Blood loss (mL) 302 (269) 291 (187) Bispectral index 40.6 (5.2) 39.1 (6.1) determine that the study would be adequately powered Size of excised liver (cm3) 474.1 (450.4) 527.7 (398.9) with n ϭ 50 per group.10 Major/minor resection We compared peak transaminases (primary outcome) Major resection Ն3 22 23 between groups using a linear regression model with peak segments Minor resection Ͻ3 28 27 transaminases as the dependent and group allocation as the segments independent variable (corresponding to a 2-sample t test). In addition, we adjusted these comparisons for baseline Data are mean (SD). transaminases and bilirubin levels, ischemic (Pringle) time, size of excised liver (cm3), and blood loss in multivariable Table 3. Intraoperative Hemodynamic Data linear regression analyses (analysis of covariance). The other Mean arterial blood measurement data were also compared using a 2-sample t pressure (mm Hg) Heart rate (bpm) ␹2 test, and count data were compared using . Hepatic Sevoflurane Propofol Sevoflurane Propofol function in patients with cirrhosis might be more sensitive group group group group (50 ؍ n) (50 ؍ n) (50 ؍ n) (50 ؍ to inflow occlusion than without cirrhosis. Therefore, we (n conducted a limited number of subgroup analyses to assess T0 94.9 (9.9) 97.1 (13.2) 86.0 (18.7) 83.7 (14.1) T 76.7 (10.2) 80.3 (11.4) 74.6 (13.6) 74.2 (9.4) the effect of cirrhosis (yes/no) on postoperative hepatocyte 1 T 76.2 (10.9) 79.9 (10.9) 72.6 (9.1) 69.5 (8.6) injury (ALT/AST) using a 2-sample t test. All analyses were 2 T3 82.7 (11.8) 80.7 (10.9) 74.9 (10.8) 69.7 (11.5) conducted using SPSS 16.0 (SPSS Inc., Chicago, IL). Results T 80.0 (9.9) 84.6 (9.7) 70.3 (9.6) 66.2 (8.9) Ϯ Ͻ 4 were expressed as the mean SD. A P value 0.05 was Data are mean (SD). considered to represent statistical significance. ϭ ϭ ϭ T0 baseline; T1 5th min after anesthetic induction; T2 5th min before ϭ ϭ Pringle; T3 middle moment of Pringle; T4 5th min after Pringle. RESULTS Fifty patients were included in each group. Table 1 shows were no significant differences. Table 3 shows the hemo- the patient characteristics and baseline values of the out- dynamic data during surgery. Epidural anesthesia was not come variables. Table 2 shows a summary of the important attempted in 2 patients in each group for a platelet count intraoperative data. In both data sets (Tables 1 and 2), there Ͻ80 ϫ 109/L.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1037 Liver Function Tests are Similar After Hepatectomy with Propofol or Sevoflurane

Table 4. Postoperative Laboratory Data and Length of Hospital Stay 95% conﬁdence interval of the difference Sevoﬂurane Propofol (equal variances assumed) group group Lower Upper (50 ؍ n) (50 ؍ n) Peak ALT (U/L) 504 (295) 571 (460) Ϫ159 293 Peak AST (U/L) 435 (275) 581 (494) Ϫ92 383 Peak bilirubin (␮mol/L) 25.2 (9.3) 33.3 (28.7) Ϫ4.3 20.6 Peak ALP (U/L) 121 (35) 144 (83) Ϫ16 61 Peak WBC (ϫ103/mL) 13.1 (2.7) 14.6 (4.6) Ϫ0.6 3.7 Peak creatinine (␮mol/L) 70.3 (11.0) 66.8 (11.7) Ϫ9.7 2.6 Hospital stay (d) 16.1 (4.8) 14.0 (2.9) Ϫ4.6 0.4 Data are mean (SD). ALT ϭ alanine aminotransferase; AST ϭ aspartate aminotransferase; ALP ϭ alkaline phosphatase; WBC ϭ white blood cell.

Table 5. The Peak Values of ALT/AST of the Subgroups over 6 Postoperative Days Sevoﬂurane group Propofol group Cirrhosis Noncirrhosis Cirrhosis Noncirrhosis P value (23 ؍ n) (27 ؍ P value (n (17 ؍ n) (33 ؍ n) Peak ALT (U/L) 558 (310) 398 (245) 0.215 675 (554) 449 (309) 0.231 Peak AST (U/L) 482 (289) 342 (232) 0.218 672 (581) 474 (310) 0.326 Data are mean (SD). ALT ϭ alanine aminotransferase; AST ϭ aspartate aminotransferase.

No patient died in this study. Major complications and intermittent clamping11of the portal triad. Both proce- included sepsis (2 patients in each group), bleeding (2 dures require a surgical intervention and prolong the overall patients in each group), and biloma (1 patient in the time of the surgical procedure. Hence, a pharmacological ap- sevoflurane group). The mean hospital stay was 2 days proach not requiring additional surgical procedures may be shorter in the propofol group (14 vs 16 days) but without a more attractive alternative than the established surgical statistical significance. The degree of ischemia and reper- strategies. In this study, we wanted to address whether the fusion injury of the liver was assessed by postoperative anesthetic affects postoperative hepatic function in patients peak serum ALT and AST levels. Transaminase levels undergoing elective liver resection with inflow occlusion. peaked between the first and the third postoperative day. Many factors contribute to hepatic injury and outcomes The sevoflurane group had slightly lower peak ALT and after liver resection. Among these, patient characteristics AST levels than the propofol group, but these were not such as the severity of liver disease and surgical complica- statistically different (P Ͼ 0.05) (Table 4). Unadjusted and tions are common contributors to poor postoperative out- adjusted results in multivariable linear regression analyses come. In this study, the patient characteristics, baseline (analysis of covariance) were almost identical. Unadjusted values of the outcome variables, and the data of surgery- results are presented in Table 4. Other liver function tests related events showed that the 2 patient groups were truly such as bilirubin and alkaline phosphatase, as well as peak equivalent (Tables 1 and 2). In addition, surgery-related white blood cell levels and creatinine, were not different factors (Pringle time, size of excised liver, and blood loss) between groups (P Ͼ 0.05). The results of subgroup analy- and baseline transaminases and bilirubin levels were con- ses comparing cirrhotic and noncirrhotic patients are pre- sidered in multivariable linear regression analyses. sented in Table 5. Although the serum levels of ALT/AST The effect of sevoflurane on hepatic function has been were higher in cirrhotic patients than in noncirrhotic pa- examined in several studies. The study conducted by Ebert tients, there were no significant differences between these and Arain12 suggested that sevoflurane, but not propofol, subgroups (P Ͼ 0.05). Figures 1 and 2 show ALT and AST was associated with increased liver injury in patients levels over 6 postoperative days, respectively. undergoing hepatectomy without inflow occlusion, but the liver function tests in sevoflurane-exposed patients were in DISCUSSION the upper limits of normal. The study by Beck-Schimmer et We compared the effect of volatile anesthetics with propo- al.,13 however, suggested that sevoflurane preconditioning fol anesthetics on liver function in patients undergoing was protective against ischemia/reperfusion injury during liver resection with inflow occlusion. There were no signifi- liver resection. Sevoflurane preconditioning was shown to cant differences in postoperative liver function as measured prevent hepatic injury, defined by transaminase levels, and by serial transaminase levels, or in clinical outcomes in the improve clinical outcome. In the volatile preconditioning 2 groups. group, the expression of inducible nitric oxide synthase Numerous strategies have been designed to reduce upon reperfusion significantly increased compared with ischemia/reperfusion injury after liver resection. Two pro- the baseline value, which points to a possible protective tective strategies to prevent ischemic-reperfusion injury role of nitric oxide in pharmacological preconditioning. In a have been clinically accepted: ischemic preconditioning4,9 rat hepatic ischemia/reperfusion injury model, clinically

1038 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Figure 1. Alanine aminotransferase (ALT) levels ϭ ϭ over 6 postoperative days. T0 baseline; T1 first ϭ postoperative day; T2 third postoperative day; ϭ ϭ T3 sixth postoperative day; sevoflur sevoflurane; CI ϭ confidence interval. Each bar represents the mean Ϯ SD.

Figure 2. Aspartate aminotransferase (AST) levels ϭ ϭ over 6 postoperative days. T0 baseline; T1 first ϭ postoperative day; T2 third postoperative day; ϭ ϭ T3 sixth postoperative day; sevoflur sevoflurane; CI ϭ confidence interval. Each bar represents the mean Ϯ SD.

relevant concentrations of sevoflurane given before, dur- anesthetics may protect the fasted liver from early, ing, and after hepatic ischemia protected the liver against neutrophil-independent, ischemia/reperfusion injury by ischemia/reperfusion injury. Increased hepatic adenosine acting during the reperfusion phase. Sevoflurane reduced triphosphate and energy levels decreased hepatocyte in- hepatic oxygen consumption and attenuated lactate dehydro- jury, and the hepatic tissue blood flow almost completely genase release during reperfusion. Sevoflurane precondition- recovered after ischemia/reperfusion in the sevoflurane ing may provide a new and easily applicable therapeutic group.14 The study by Imai et al.15 suggested that volatile option to protect the liver in hepatectomy.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1039 Liver Function Tests are Similar After Hepatectomy with Propofol or Sevoflurane

Propofol is an IV sedative-hypnotic drug frequently REFERENCES used in anesthesia and intensive care that not only inhibits 1. Gozzetti G, Mazziotti A, Grazi GL, Jovine E, Gallucci A, lipid peroxidation, but also enhances the cellular antioxi- Gruttadauria S, Frena A, Morganti M, Ercolani G, Masetti M, Pierangeli F. Liver resection without blood transfusion. Br J dant defense system in several tissues including brain, Surg 1995;82:1105–10 16,17 liver, and heart. Propofol pretreatment attenuates 2. Kooby DA, Stockman J, Ben-Porat L, Gonen M, Jarnagin WR, ischemia/reperfusion–induced intestinal epithelial apopto- Dematteo RP, Tuorto S, Wuest D, Blumgart LH, Fong Y. sis, which might be attributable to its antioxidant property Influence of transfusions on perioperative and long-term out- modulating the ceramide pathway.18 Propofol improves come in patients following hepatic resection for colorectal metastases. Ann Surg 2003;237:860–70 survival of liver cells in rat hepatocyte suspensions exposed Ј 3. van der Bilt JD, Livestro DP, Borren A, van Hillegersberg R, to oxidant injury by a free radical generator 2,2 -azobis Borel Rinkes IH. European survey on the application of (2-amidinopropane) dihydrochloride (AAPH).19 Propofol vascular clamping in liver surgery. Dig Surg 2007;24:423–35 protects hepatic L02 cells from hydrogen peroxide–induced 4. Clavien PA, Yadav S, Sindram D, Bentley RC. Protective apoptosis, partly through activating the MEK-ERK path- effects of ischemic preconditioning for liver resection per- 20 formed under inflow occlusion in humans. Ann Surg way and further suppressing Bad and Bax expression. 2000;232:155–62 The study by Shimono et al., however, suggested that 5. Lesurtel M, Selzner M, Petrowsky H, McCormack L, Clavien propofol did not have a protective effect against PA. How should transection of the liver be performed? A hypoxia/reoxygenation injury in rat liver slices.21 The prospective randomized study in 100 consecutive patients: effects of pharmacological preconditioning with propo- comparing four different transection strategies. Ann Surg 2005;242:814–23 fol in patients undergoing liver resection with inflow 6. Clavien PA, Petrowsky H, DeOliveira ML, Graf R. Strategies occlusion remain to be determined. for safer liver surgery and partial liver transplantation. N Engl In our study, patients with biopsy-proven cirrhosis did J Med 2007;356:1545–59 not have worse postoperative liver dysfunction than those 7. Wei AC, Tung-Ping Poon R, Fan ST, Wong J. Risk factors for without cirrhosis, likely because the cirrhotic patients had perioperative morbidity and mortality after extended hepatectomy for hepatocellular carcinoma. Br J Surg 2003;90:33–41 mild disease, indicated by the Model for End-Stage Liver 8. Coelho JC, Claus CM, Machuca TN, Sobottka WH, Gonçalves Disease score. In addition, the period of ischemia was short; CG. Liver resection: 10-year experience from a single institu- longer ischemic stress may have unmasked a difference tion. Arq Gastroenterol 2004;41:229–33 between cirrhotic and noncirrhotic patients. Previous stud- 9. Clavien PA, Selzner M, Ru¨ diger HA, Graf R, Kadry Z, Rousson ies suggest that patients with early-stage cirrhosis can V, Jochum W. A prospective randomized study in 100 consecutive patients undergoing major liver resection with versus tolerate as much as 60 to 75 minutes of inflow occlusion without ischemic preconditioning. Ann Surg 2003;238:843–52 during hepatectomy without serious postoperative liver 10. Lerman J. Study design in clinical research: sample size esti- decompensation.22–24 mation and power analysis. Can J Anaesth 1996;43:184–91 Our study had several limitations. It is not possible to 11. Petrowsky H, McCormack L, Trujillo M, Selzner M, Jochum W, determine from our results whether sevoflurane and Clavien PA. A prospective, randomized, controlled trial comparing intermittent portal triad clamping versus ischemic propofol are protective in the setting of hepatic preconditioning with continuous clamping for major liver ischemia/reperfusion injury. Second, patients were not resection. Ann Surg 2006;244:921–30 systematically assessed for hepatic fibrosis or steatosis, 12. Ebert TJ, Arain SR. Renal responses to low-flow desflurane, both risk factors for liver dysfunction after liver ischemia. sevoflurane, and propofol in patients. Anesthesiology 2000; In conclusion, our data suggest that sevoflurane or propo- 93:1401–6 13. Beck-Schimmer B, Breitenstein S, Urech S, De Conno E, Wit- fol can be used safely in patients undergoing hepatectomy tlinger M, Puhan M, Jochum W, Spahn DR, Graf R, Clavien PA. with inflow occlusion, and that there is no advantage of one A randomized controlled trial on pharmacological precondi- drug over the other. tioning in liver surgery using a volatile anesthetic. Ann Surg 2008;248:909–18 AUTHOR CONTRIBUTIONS 14. Bedirli N, Ofluoglu E, Kerem M, Utebey G, Alper M, Yilmazer D, Bedirli A, Ozlu O, Pasaoglu H. Hepatic energy metabolism JCS helped design and conduct the study, analyze the data, and the differential protective effects of sevoflurane and isoflu- and write the manuscript. This author has seen the original rane anesthesia in a rat hepatic ischemia-reperfusion injury study data, reviewed the analysis of the data, approved the model. Anesth Analg 2008;106:830–7 final manuscript, and is the author responsible for archiv- 15. Imai M, Kon S, Inaba H. Effects of halothane, isoflurane and ing the study files. YMS helped design and conduct the sevoflurane on ischemia-reperfusion injury in the perfused liver of fasted rats. Acta Anasthesiol Scand 1996;40:1242–8 study, analyze the data, and write the manuscript. This 16. De La Cruz JP, Sedenõ G, Carmona JA, Sańchez de la Cuesta F. author has seen the original study data, reviewed the The in vitro effects of propofol on tissular oxidative stress in analysis of the data, and approved the final manuscript. the rat. Anesth Analg 1998;87:1141–6 LQY helped conduct the study and analyze the data. This 17. Kharasch ED, Armstrong AS, Gunn K, Artru A, Cox K, Karol author has seen the original study data, reviewed the MD. Clinical sevoflurane metabolism and disposition. II. The role of cytochrome P450 2E1 in fluoride and hexafluoroisopro- analysis of the data, and approved the final manuscript. panol formation. Anesthesiology 1995;82:1379–88 MZZ helped conduct the study. This author has seen the 18. Liu KX, Chen SQ, Huang WQ, Li YS, Irwin MG, Xia Z. original study data and approved the final manuscript. ZJL Propofol pretreatment reduces ceramide production and helped conduct the study. This author has seen the original attenuates intestinal mucosal apoptosis induced by intesti- study data and approved the final manuscript. WFY helped nal ischemia/reperfusion in rats. Anesth Analg 2008;107: 1884–91 design the study. This author has seen the original study 19. Navapurkar VU, Skepper JN, Jones JG, Menon DK. Propofol data, reviewed the analysis of the data, and approved the preserves the viability of isolated rat hepatocyte suspensions final manuscript. under an oxidant stress. Anesth Analg 1998;87:1152–7

1040 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA 20. Wang H, Xue Z, Wang Q, Feng X, Shen Z. Propofol protects 23. Kim YI, Nakashima K, Tada I, Kawano K, Kobayashi M. hepatic L02 cells from hydrogen peroxide-induced apoptosis Prolonged normothermic ischaemia of human cirrhotic liver via activation of extracellular signal-regulated kinases path- during hepatectomy: a preliminary report. Br J Surg way. Anesth Analg 2008;107:534–40 1993;80:1566–70 21. Shimono H, Goromaru T, Kadota Y, Tsurumaru T, Kanmura Y. 24. Jeppsson B. Liver resection: prolonged inflow occlusion in Propofol displays no protective effect against hypoxia/ human cirrhotic livers. HPB Surg 1996;10:123–5 reoxygenation injury in rat liver slices. Anesth Analg 2003;97:442–8 22. Kim YI, Kobayashi M, Aramaki M, Nakashima K, Mitarai Y, Yoshida T. “Early-stage” cirrhotic liver can withstand 75 minutes of inflow occlusion during resection. Hepatogastroen- terology 1994;41:355–8

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1041 Pain Medicine

Section Editor: Spencer S. Liu

The Costs and Beneﬁts of Extending the Role of the Acute Pain Service on Clinical Outcomes After Major Elective Surgery

Anna Lee, PhD, Simon K. C. Chan, MBBS, Phoon Ping Chen, MBBS, Tony Gin, MD, Angel S. C. Lau, MPhil, and Chun Hung Chiu, MPhil

BACKGROUND: Acute pain services have received widespread acceptance and formal support from institutions and organizations, but available evidence on their costs and benefits is scarce. Although there is good agreement on the provision of acute pain services after many major surgical procedures, there are other procedures for which the benefits are unclear. Data are required to justify any expansion of acute pain services. In this randomized, controlled clinical trial we compared the costs and effects of acute pain service care on clinical outcomes with conventional pain management on the ward. Patients included in the trial were considered by their anesthesiologist to have either arm be suitable for the procedure. METHODS: Four hundred twenty-three patients undergoing major elective surgery were randomized either to an anesthesiologist-led, nurse-based acute pain service group with patient- controlled analgesia or to a control group with IM or IV boluses of opioid analgesia. Both groups were treated with medications to treat opioid-related adverse effects and received the usual care from health professionals assigned to the ward. The main outcome measures were quality of recovery scores, pain intensity measures, global measure of treatment effectiveness, and overall pain treatment cost. Cost-effectiveness acceptability curves were drawn to detect a difference in the joint cost-effect relationship between groups. RESULTS: There was no difference in quality of recovery score on postoperative day 1 between treatment and control groups (mean difference, 0; 95% confidence interval [CI], Ϫ0.7 to 0.7; P ϭ 0.94) or in the rate of improvement in quality of recovery score (mean difference, Ϫ0.1; 95% CI, Ϫ0.4 to 0.1; P ϭ 0.34). The proportion of patients with 1 or more days of highly effective pain management was higher in the acute pain service group than in the control group (86% vs. 75%; P Ͻ 0.01). Costs were higher in the acute pain service group (mean difference, US$46; 95% CI, $44 to $48 per patient; P Ͻ 0.001). A cost-effectiveness acceptability curve showed that the acute pain service was more cost effective than was control for providing highly effective pain management if the decision maker was willing to pay more than US$546 per patient per 1 day with highly effective treatment. CONCLUSION: In extending the role of the acute pain service to a specific group of major surgical procedures, the acute pain service was likely to be cost effective. (Anesth Analg 2010;111: 1042–50)

pproximately 234 million major surgical procedures In some countries, APS is considered to be a standard of are performed every year worldwide, and most of care,4–6 but little is known about its economic benefits. Our Athese will require some form of pain management.1 systematic review of 10 economic evaluations of APS Although patient-controlled analgesia (PCA) and epidural programs (involving 14,774 patients) also showed that analgesia provide better postoperative analgesia than does there was insufficient evidence to draw conclusions about intermittent IM analgesia,2 these techniques require special the cost effectiveness and cost–benefit ratio of APS.7 The care and monitoring from acute pain service (APS) teams. In overall quality of published economic evaluations was poor Hong Kong, it is routine practice for APS,3 rather than ward because most studies were limited to partial economic nurses, to oversee PCA because of a lack of clinical nurse analyses.7 specialists. A limitation of the latest guidelines, “Acute Pain Man- agement: Scientific Evidence,”8 is that they generalized the From the Department of Anaesthesia and Intensive Care, The Chinese evidence and did not present data on a specific procedure.9 University of Hong Kong, Prince of Wales Hospital, Shatin, NT, Hong Kong. There is usually good agreement on the provision of APS Accepted for publication May 28, 2010. after many major surgical procedures (e.g., with upper The work described in this paper was fully supported by a grant from the Central Policy Unit of the Government of HKSAR and the Research Grants abdominal laparotomy). One problem with assessing the Council of the HKSAR, China (project reference: CUHK4004-PPR20051). costs and benefits of an established APS program is that it Address correspondence to Anna Lee, PhD, Department of Anaesthesia and would be unreasonable to revert to standard postoperative Intensive Care, The Chinese University of Hong Kong, Prince of Wales ward care for these patients, making a randomized trial of Hospital, Shatin, NT, Hong Kong. Address e-mail to [email protected]. APS or no APS difficult. However, there are some surgical Copyright © 2010 International Anesthesia Research Society 4 DOI: 10.1213/ANE.0b013e3181ed1317 procedures for which the benefits of APS are unclear. This

1042 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 may be because of new surgical techniques that may for randomizing eligible patients within 24 hours of meet- decrease postoperative pain (laparoscopic assisted proce- ing the study criteria to either APS or CWPS care, by using dures), or patients expected to have significant postopera- a sealed opaque envelope containing a computer-generated tive pain but had not been offered APS in the past (e.g., random treatment allocation. cardiac surgery). Although all anesthesiologists would consider using APS in these patients, mixed views were held Treatment Procedures by staff on the desirability and necessity of providing Patients randomized to the APS group received IV mor- APS for them. Thus, data are required to justify any phine PCA with or without supplementary oral analgesics, expansion of acute pain services to these patients. Be- and medications to treat opioid-related adverse effects. cause they had not traditionally been receiving APS, it They were seen by an APS nurse or an anesthesiologist or was possible to randomize these patients to APS or both once daily (normal practice). The APS team was standard ward care. informed if any of the following occurred: inadequate pain Therefore, we performed prospective cost-effectiveness control (persistent pain score Ն3 of 10), oxygen desatura- Ͻ Ͻ analyses alongside a randomized controlled trial of APS tion (Spo2 90%), bradypnoea (respiratory rate 10/min), care versus conventional pain management on the ward in hypotension (systolic blood pressure Ͻ90 mm Hg), uncon- patients undergoing major elective surgery, for whom the trolled nausea and vomiting with parenteral antiemetic, attending anesthesiologist was uncertain about the benefits severe pruritus if uncontrolled with chlorpheniramine 5 mg of APS care. Our objective was to compare the costs and q8h prn IM/IV, or difficulty awakening the patient. Al- effects of APS care on clinical outcomes with conventional though APS did offer acute pain techniques such as periph- pain management on the ward in patients undergoing eral nerve blocks and epidural PCA, these techniques were major elective surgery from the perspective of the Hospital not used in this study. Patients randomized to the CWPS Authority (a government body funding public health ser- group received opioid (pethidine, morphine, tramadol), non- vices in Hong Kong). opioid (diclofenac, paracetamol/phenyltoloxamine, paracetamol) analgesics by IM, IV boluses and/or oral routes, and METHODS medications to treat opioid-related adverse effects prescribed The study was conducted at the Prince of Wales Hospital in by surgeons. All patients received the usual care from sur- Hong Kong, a large university teaching hospital. The study geons and nursing professionals assigned to the surgical ward was approved by the local Clinical Research Ethics Commit- (and intensive care unit for postoperative care if undergoing tee. The study was registered in the Centre for Clinical Trials coronary artery bypass graft). The patient, attending health Clinical Registry of The Chinese University of Hong Kong professionals, and research staff were not blinded to the (trial no. CUHK_CCT00105) on February 28, 2006, available at treatment assignment. http://www.cct.cuhk.edu.hk/registry/publictriallist.aspx. After giving written informed consent, adult patients were en- Data Collection rolled from April 8, 2006, to February 12, 2009. The APS is All surgical patients were admitted as inpatients on the day staffed by 2.2 anesthesiologists and 1 nurse full-time before surgery. Patients were interviewed by an investiga- equivalent to provide the service continuously 24 h/day, tor (Angel S. C. Lau or Chun Hung Chiu) on 3 consecutive serving an average of 17 patients per day (unpublished days after their surgery with a standardized questionnaire. data for 2009). We collected data on patient’s demographics, ASA Physical Status, and length of hospital stay. We measured pain Patients intensity (worse pain, average pain in the last 24 hours, and Groups of patients were identified that could possibly current pain), pain at rest, pain during movement and pain benefit from APS, but for whom APS care was not often interference with daily activities (walking ability, mood, used in the past. Inclusion criteria were adults, ages 18 sleep, relations with others, and ability to concentrate), and years or older and undergoing major surgery (such as ratings on a numeric rating scale (NRS) using the modified laparoscopic hemicolectomy, open reduction and internal Brief Pain Inventory questionnaire11 on days 1 to 3 after fixation, total abdominal hysterectomy) or complex major surgery. A global measure of treatment effectiveness was surgery (e.g., coronary artery bypass graft, discectomy) as measured by asking patients, “How effective do you think is defined by the Hong Kong Government Gazette.10 Pa- the treatment for pain was?” using a 5-point Likert scale tients recruited to this study were only those identified by (0 ϭ poor,1ϭ fair,2ϭ good,3ϭ very good,4ϭ excellent).12 the attending anesthesiologist as being candidates who The 9-item Quality of Recovery (QOR) score13 was col- may or may not benefit from APS, in comparison with lected to measure the patient’s health-related quality of life conventional ward pain service (CWPS). All patients had after anesthesia and surgery on a daily basis with a score general anesthesia with no additional regional anesthesia. from 0 to 18. The frequency, severity, and distress of Patients were excluded if they were younger than 18 years opioid-related side effects (nausea, vomiting, difficulty in of age, undergoing emergency or obstetric surgery, had a concentrating, drowsiness or difficulty staying awake, feel- history of cognitive impairment or preoperative opioid use, ing confused, and feeling of general fatigue or weakness) or were unable to give consent. were totaled daily into an adverse effect score (from 0 to The principal investigator (Anna Lee) generated the 60). These specific symptoms from the opioid-related random allocation sequence using a computer and was not symptom distress scale14 were chosen because they were involved in the data collection process. Two investigators thought to have a negative effect on patient’s daily activi- (Angel S. C. Lau, and Chun Hung Chiu) were responsible ties and recovery after surgery.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1043 Cost Effectiveness of Acute Pain Service

All calculated direct costs related to postoperative pain social support, and surgeon’s postoperative treatment management were based on the first 3 days after surgery. preferences. From the patient’s drug chart, we recorded the type, dose, and frequency of analgesic drugs and the drugs used to Statistical Analysis treat opioid-related side effects. The medication costs were We calculated the sample size using QOR as the primary estimated from the unit costs of the hospital pharmacy. The outcome because this was a more patient-centered outcome PCA costs, obtained from the hospital administration, than were pain ratings and cost. We calculated that a included the cost for infusion pump, IV tubing sets, car- sample size of 522 would provide 80% power to detect tridges, catheters, batteries, syringes, needles, swabs, dress- a small to moderate effect size (0.25) between groups using ings, saline, and morphine. From the patient’s APS record, a 2-sample t test (EAST 5.2, Cytel Software Corporation, the staff cost was calculated using the total nursing and Cambridge, Massachusetts), allowing for interim analyses. anesthesiologist time spent for each patient. The nursing Two interim analyses were planned after 174 and 348 and anesthesiologists’ staff salaries, obtained from the patients had completed their participation in the study hospital administration, were based on the midpoint of the using the O’Brien–Fleming stopping rules, with ad priori relevant pay scale. The ward nursing cost for APS and boundaries of P Յ 0.0002 and P Յ 0.0121 to reject the null CWPS groups were assumed to be the same as a previous hypothesis (efficacy boundary, if large treatment differ- study at our hospital,15 showing that there was no signifi- ences appear before the end of the study), and P Ն 0.9659 cant difference in total ward nursing time (communication, and P Ն 0.3444 to accept the null hypothesis (futility documentation, administration of drug, and observations) boundary, if there is little chance of finding a significant for patients receiving PCA or IM opioid injection. The total difference between groups). postoperative pain management cost was a total of the costs The primary analyses were performed on a modified for analgesic drugs, drugs to treat opioid-related side intention-to-treat basis (i.e., patients were analyzed accord- effects, PCA, and APS staffing. At the time of reporting the ing to their randomized allocated groups but were ex- study results (October 31, 2009), 1 US$ ϭ HK$7.75. All costs cluded from the analysis if they did not adequately adhere are reported in U.S. dollars. to the protocol after randomization.). We used the 2-sample t test, ␹2 test, Fisher exact test, and Mann–Whitney U test to compare baseline characteristics. The mean difference is Outcome Measures defined as the APS outcome measure minus CWPS outcome The primary outcome measure was QOR scores. Pain measure. For the QOR, pain intensity and interference out- intensity (mean pain ratings for worst pain, average pain in comes, global measure of effectiveness, and adverse effect the last 24 hours, and current pain), pain intensity at rest, score, we used multilevel regression models18,19 to assess the pain intensity during movement, global measure of treat- intervention effect on the change between the measurement effectiveness, and overall pain treatment cost out- ments taken on the first to third day after surgery. An comes were also measured. These outcome measures were advantage of using a multilevel regression model over a used in defining the incremental cost-effectiveness ratios repeated-measure analysis of variance is that it can account and incremental net benefits, the appropriate measures for complex covariance structure and accommodate incom- 16 of reporting results from a cost-effectiveness analysis. Spe- plete data.19 cifically, the effectiveness of the intervention for cost- Given the expected large variability of cost data, the effectiveness analysis purposes was expressed as the number study was underpowered to test the economic hypothesis 17 17 of pain-free days at rest, pain-free days with movement, that APS would be more cost effective than would CWPS. and days with highly effective treatment. A pain-free day However, in the absence of sufficient power to test the was defined as having a NRS Յ3ona0to10scale. The economic hypotheses, there have been methodological ad- analgesic effectiveness was 3 if the patient had 3 pain-free vances in examining sampling uncertainty for incremental days, and 0 if the patient did not experience NRS Յ3 at all. cost-effectiveness ratios, with emphasis on the likelihood The number of days with highly effective treatment was 3 that the intervention represents good value for the cost if the patient rated his or her global measure of effective- rather than on economic hypothesis testing.20,21 ness as excellent or very good on all 3 days, and 0 if the We assumed that APS was cost effective if the extra cost patient did not have days with excellent or very good ratings. of an extra gain in effect was less than the decision maker’s Other secondary end points included pain interference willingness to pay (WTP) for it.22 For example, if the during daily activities, adverse effect score, in-hospital mor- maximum WTP was set at $200, APS would be cost tality, and length of hospital stay. A pain-free interference day effective if the incremental cost-effectiveness ratio (ratio of was defined as having a mean NRS of 0 with pain interference the extra cost to extra benefit, i.e., ⌬C/⌬E) was less than on daily activities scales ranging from 0 to 10. The $200. The 95% confidence interval (95% CI) around the interference-free effect was 3 if the patient had 3 pain-free incremental cost-effectiveness ratio was estimated using interference days, and 0 if the patient experienced pain the Fieller method. Because there is uncertainty on the WTP that interfered with daily activities on all 3 days. The value and the true estimate of the incremental cost- length of stay was not included in the cost-effectiveness effectiveness ratio, the cost-effectiveness acceptability curve analysis because we believed that it was a weak outcome was constructed from net benefit (NB) regressions.23 Under measure. The delay in patient discharge from hospital the NB framework, each subject’s NB is computed from the ϫ Ϫ was often not due to pain or analgesia-related side effects observed data as WTP effecti costi, where effecti and but due to postoperative rehabilitation plans, level of costi are the data for the ith person’s effect and cost,

1044 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA respectively, and WTP is a willingness-to-pay number that must be specified.24 In its simplest form, the NB regression involves fitting the following linear regression model:

؉ ␤ ؉ ␧ ␤ ؍ NBi 0 TX TXi i, ϭ where TXi is the ith person’s treatment indicator (TXi 1 for ␧ 24 APS and 0 for CWPS) and i is a stochastic error term. The equation is fitted several times, each time with a different value of WTP value.24 To generate a cost-effectiveness accept- ␤ Ͼ ability curve, we used the probability that TX 0 for the y axis and WTP for the x axis.24 In other words, the cost- acceptability curves showed the probability that APS was more cost effective than was CWPS for a range of values that decision makers might be willing to pay for 1 day gained of beneficial effect. All analyses were performed with STATA software version 10.0 (StataCorp, College Station, Texas), and cost-effectiveness analysis was performed using the macro “iprogs” available from the University of Pennsylvania (www.uphs.upenn.edu/dgimhsr/stat-cicer.htm; accessed April 28, 2010). A 2-sided P Ͻ 0.05 was considered statistically significant. We declared the trial to be positive if more than 1 of the outcomes, but not all, were significant after correcting for multiplicity using the Holm stepwise approach (corrected overall critical P value was 0.0167).25

RESULTS The trial was stopped at the second interim analysis, after 402 patients had completed the study, on the basis of slower-than- anticipated accrual rate and a prespecified futility stopping Figure 1. Patient ﬂow through clinical trial. APS ϭ acute pain service; rule. On the basis of 398 patients who had complete day 1 CWPS ϭ conventional ward pain service. QOR data (195 in the APS group and 203 in the CWPS group), comparison of the day 1 QOR via a 2-sample t test (P ϭ 0.43) crossed the a priori futility boundary for early stopping with 42, 2-sample t test P Ͻ 0.01). Although the proportion of acceptance of the null hypothesis of no difference between admissions to the intensive care unit after surgery was groups. At interim analysis, it was calculated that if the study similar (␹2 test P ϭ 0.12) between the 2 groups, the had continued to the planned enrollment of 522, the probabil- predicted risk of death from the Acute Physiologic and ity of demonstrating a difference in day 1 QOR between Chronic Health Evaluation II score26 was higher in the APS Ͻ treatment groups was 1% under the alternative hypothesis group (Mann–Whitney U test, P Ͻ 0.01). The median [IQR] on the basis of the observed unadjusted day 1 QOR treatment duration of PCA with morphine was 29.5 [18 to 43] hours. group differences. The median [IQR] time that anesthesiologists (including PCA set-up time in the recovery room) and pain nurses Study Population spent caring for the patients in the APS group were 31 [23 Of the 470 surgical patients screened for the study, 422 met to 39] and 16 [8 to 16] minutes, respectively. the study criteria and were randomized. Two hundred nine patients were allocated to the APS group, and 213 to the Outcomes CWPS group (Fig. 1). Ten patients from each group with- The point estimates on the first day after surgery and the drew from the study after randomization. Although the mean change over 3 days for QOR, pain intensity, interfer- gender and type of surgery distributions in the withdrawal ence with daily activity from pain, global measure of group were similar to those of patients who completed the treatment effectiveness, and adverse effect outcomes are study (Fisher exact test P ϭ 1.00 and ␹2 test P ϭ 0.16, shown in Table 2. There were no differences between respectively), the patients who withdrew were older than groups for outcomes on the first day of surgery: QOR (P ϭ those who completed the study (mean [SD], 58 [8] for those 0.94), pain intensity (P ϭ 0.31), and pain on movement (P ϭ who withdrew and 52 [12] for those who completed; 0.17). However, APS patients had lower pain scores at rest, 2-sample t test P ϭ 0.02). The baseline characteristics at less interference with daily activities because of pain, and enrollment were similar for age, gender, type and magni- better treatment effectiveness than did CWPS patients on tude of surgery, ASA Physical Status, and length of stay in the first day after surgery (Table 2). The rate of improve- the intensive care unit between APS and CWPS patients ment in QOR scores (P ϭ 0.34), daily rate reductions in pain (Table 1). Patients in the APS group had longer duration of intensity (P ϭ 0.20), and pain during movement (P ϭ 0.07) anesthesia than did those in the CWPS group (mean between the 2 groups were similar. The APS group had difference, 26 minutes; 95% confidence interval [CI], 10 to significantly smaller daily reductions in pain scores at rest

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1045 Cost Effectiveness of Acute Pain Service

Table 1. Patient Characteristics at Enrollment Acute pain service group Conventional ward pain service group P value (213 ؍ N) (209 ؍ Characteristic (N Age, mean (SD), years 53 (12.3) 52 (11.3) 0.41 Women, no. (%) 106 (50.7) 108 (50.7) 1.00 Type of surgery, no. (%)a Orthopedic 32 (15.3) 36 (15.9) Gynecology 47 (22.5) 50 (23.5) 0.64 Cardiothoracic 80 (38.3) 69 (32.4) General/other 50 (23.9) 58 (27.2) Magnitude of surgery, no. (%) Major 123 (58.9) 140 (65.7) 0.13 Ultramajor 86 (41.1) 73 (34.3) ASA physical status grade, no. (%)b I 62 (29.8) 78 (40.0) II 78 (37.5) 81 (38.4) 0.37 III 61 (29.3) 45 (21.3) IV 7 (3.4) 7 (3.3) Duration of anesthesia, mean 225 (88.1) 199 (79.6) Ͻ0.01 (SD), minutes Admission to ICU, no. (%) 64 (30.6) 51 (23.9) 0.12 APACHE II score, median (IQR)c 12 (10–16) 11 (8–13) Ͻ0.01 Length of ICU stay, median (IQR), 22 (19–23) 21 (18–22) 0.17 hours

SD, standard deviation; ASA, American Society of Anesthesiologists; ICU, intensive care unit; APACHE II, Acute Physiology and Chronic Health Evaluation II; IQR, interquartile range. a Because of rounding, percentages may not all total 100. b American Society of Anesthesiologists’ physical status could not be determined for 3 patients (1 in acute pain service group and 2 in conventional ward pain service group). c APACHE II score ranges from 0 to 71, with higher scores indicating higher probability of death.

Table 2. Point Estimate on First Day After Surgery and Mean Change Over the First 3 Days After Surgery by Group, and Mean Difference Between Groups (95% Conﬁdence Intervals) for Primary and Secondary Outcomes Acute pain service Conventional ward pain service Mean difference between groups Day 1 Mean change over Day 1 Mean change over Day 1 Mean change over estimate 3 days estimate 3 days estimate 3 days Primary outcome QORa 11.2 (10.7 to 11.7) 1.4 (1.2 to 1.6) 11.2 (10.7 to 11.7) 1.5 (1.3 to 1.7) 0 (Ϫ0.7 to 0.7) Ϫ0.1 (Ϫ0.4 to 0.1) Secondary outcomes Pain intensity 5.3 (4.9 to 5.6) Ϫ0.7 (Ϫ0.9 to Ϫ0.6) 5.5 (5.2 to 5.9) Ϫ0.9 (Ϫ1.0 to Ϫ0.7) Ϫ0.3 (Ϫ0.8 to 0.2) 0.1 (Ϫ0.1 to 0.3) Pain at rest 2.3 (1.9 to 2.8) Ϫ0.2 (Ϫ0.4 to Ϫ0.1) 3.2 (2.8 to 3.6) Ϫ0.6 (Ϫ0.7 to Ϫ0.4) Ϫ0.9 (Ϫ1.4 to Ϫ0.3)* 0.3 (0.1 to 0.5)* Pain on movement 4.5 (4.0 to 4.9) Ϫ0.5 (Ϫ0.7 to Ϫ0.3) 4.9 (4.5 to 5.4) Ϫ0.7 (Ϫ0.9 to Ϫ0.5) Ϫ0.5 (Ϫ1.1 to 0.2) 0.2 (0 to 0.5) Interference 3.0 (2.5 to 3.5) Ϫ0.4 (Ϫ0.6 to Ϫ0.2) 3.9 (3.4 to 4.4) Ϫ0.7 (Ϫ0.9 to Ϫ0.5) Ϫ0.9 (Ϫ1.6 to Ϫ0.2)* 0.3 (0.1 to 0.6)* Global treatment 3.0 (2.8 to 3.2) Ϫ0.1 (Ϫ0.1 to 0) 2.4 (2.2 to 2.6) 0.1 (0.1 to 0.2) 0.6 (0.3 to 0.9)† Ϫ0.2 (Ϫ0.3 to Ϫ0.1)† effectiveness Adverse effectsb 13.9 (12.0 to 15.7) Ϫ2.9 (Ϫ3.6 to Ϫ2.2) 16.3 (14.5 to 18.1) Ϫ4.0 (Ϫ4.6 to Ϫ3.3) Ϫ2.4 (Ϫ5.0 to 0.1) 1.0 (0.0 to 2.0)‡ QOR, Quality of Recovery Score. a Higher QOR scores represent better recovery after anesthesia and surgery; higher global measure of effectiveness scores represent better effectiveness of pain intervention. b Higher adverse effect scores represent worse experience with opioid-related side effects. * P Յ 0.01; † P Յ 0.001; ‡ P ϭ 0.04.

and interference with daily activities than did the CWPS Table 3. Incidence (%; 95% Conﬁdence Interval) group over the first 3 days after surgery (Table 2). of Moderate to Severe Pain at Rest and on The incidence of moderate to severe pain (NRS Ͼ3ona0 Movement After Major Surgery by Groups to 10 scale) at rest and on movement after major surgery in the Conventional 2 groups is shown in Table 3. The incidences of “poor” Acute ward P pain service pain service value treatment effectiveness on the first day after surgery in the Day 1 APS and CWPS groups were 0.5% (95% CI, 0.1 to 2.9) and At rest 25.8 (20.1 to 32.4) 34.7 (28.4 to 41.5) 0.06 4.8% (95% CI, 2.5 to 8.8), respectively. The proportion of On movement 56.1 (49.0 to 63.0) 57.0 (50.1 to 63.7) 0.86 patients with 1 or more days of highly effective pain manage- Day 2 At rest 12.4 (8.3 to 18.0) 20.8 (15.7 to 27.0) 0.03 ment (i.e., treatment effectiveness rated as very good and On movement 41.9 (34.9 to 49.2) 45.1 (38.3 to 52.1) 0.53 excellent) was higher in the APS group than in the control Day 3 group (86% vs. 75%; absolute risk difference 11%; 95% CI, 3% At rest 19.7 (14.6 to 26.0) 14.9 (10.6 to 20.5) 0.22 to 20%; ␹2 test P Ͻ 0.01; Fig. 2). This is equivalent to a number On movement 37.2 (30.5 to 44.5) 33.8 (27.6 to 40.7) 0.50 needed to treat of 9 (95% CI, 5 to 33). There were no significant

1046 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA severity of opioid-related side effects on the first day after surgery tended to be less in the APS group than in the CWPS group (Table 2; P ϭ 0.06). Overall, both groups experienced 1 day of no opioid-related side effects (Table 4).

Costs As was expected, the costs of analgesia, medications to treat opioid-related side effects, and APS staff costs were significantly higher in the APS group than in the CWPS group (Table 5). The mean difference in the total cost of pain treatment was US$46 (95% CI, $44 to $48) per patient (P Ͻ 0.001). Because there was no significant extra day gained for being pain free at rest, pain free during movement, no opioid-related side effects, and no interference with daily activity measures in the APS group over the CWPS group, incremental cost-effectiveness ratios were not estimated. The incremental cost-effectiveness ratio for costs per 1 day Figure 2. The number of days experiencing a highly effective treatment (“very good” and “excellent” ratings) in the acute pain with highly effective treatment gained was US$151 (95% CI, service and conventional ward pain service groups. There was a $87 to $546) per patient. Decision makers who are willing to signiﬁcant difference between the 2 groups (P Ͻ 0.01). pay less than US$87 per patient per 1 day with highly effective treatment can be 95% confident that the APS differences in the other pain-free outcomes (Table 4) between represents bad value; between US$87 and US$546 per the 2 groups. The mean duration of hospital stay (days) was patient per 1 day with highly effective treatment, the similar between the APS and CWPS groups (12 [11] vs. 10 [12], decision maker cannot be 95% confident that the 2 inter- respectively; 2-sample t test P ϭ 0.13). ventions differ in value; for those willing to pay more than US$546 per patient per 1 day with highly effective treat- Adverse Events ment, they can be 95% confident that the APS represents One patient in the APS group had respiratory depression good value in comparison with CWPS (Fig. 3). due to PCA with morphine (incidence 0.5%, 95% CI, 0.1% to 2.8%) and required IV 0.8 mg naloxone treatment. During DISCUSSION the study, 1 patient in the APS group died after coronary We conducted a cost-effectiveness analysis alongside a artery bypass surgery in the intensive care unit because of randomized controlled trial of APS versus CWPS. Previous uncontrolled bleeding from the surgical site. studies included in systematic reviews7,27,28 examining the The risk of opioid-related side effects at any time during effect of APS on postoperative outcomes were likely to be the follow-up was similar between groups (41% in the APS biased because the studies were observational in design group and 40% in the CWPS group; absolute risk difference (before–after studies, matched comparisons). In this trial of 1%, 95% CI, Ϫ9% to 12%; ␹2 test P ϭ 0.76). However, the 422 patients, there were no significant differences in the

Table 4. Number of Outcome-Free Days (Median, IQR) During the First 3 Days After Surgery Acute pain service Conventional ward pain service group group P value (203 ؍ n) (199 ؍ Outcome (n Pain freea 1 (0 to 2) 1 (0 to 2) 0.62 Pain free at resta 3 (2 to 3) 3 (2 to 3) 0.63 Pain free on movementa 2 (0 to 3) 2 (1 to 3) 0.69 Pain-free interference on daily activitiesa 0 (0 to 1) 0 (0 to 1) 0.80 Free of opioid-related side effectsb 1 (0 to 1) 1 (0 to 1) 0.81 IQR, interquartile range. a An outcome-free day was deﬁned as numeric rating Յ3ona0to10scale. b Deﬁned as adverse effect score ϭ 0.

Table 5. Mean Pain Treatment Use and Costs (in US$) per Patient Acute pain service Conventional ward pain service group group Mean difference CI) P value %95) (203 ؍ n) (199 ؍ Cost categorya (n Analgesia 18.74 1.21 17.53 (17.00 to 18.05) Ͻ0.001 Medications to treat side effects 2.19 0.94 1.25 (0.07 to 2.42) 0.04 Acute pain service staff costs 27.34 0.50 26.84 (25.63 to 28.05) Ͻ0.001 Total cost of pain treatment 48.27 2.65 45.62 (43.52 to 47.71) Ͻ0.001 a One patient in the acute pain service group did not receive the intervention because of lack of staff to set up the IV patient-controlled analgesia in the recovery room. Three patients in the conventional ward pain service group received acute pain service intervention after initial pain treatment was inadequate. Refer to METHODS section in text for costing methodology.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1047 Cost Effectiveness of Acute Pain Service

that APS was associated with less pain interference on daily activities on the first day of surgery, but this effect over subsequent days was less than that in the CWPS group and made no difference to the quality of recovery. During the trial, 1 patient died from surgical complications. The risk of respiratory depression associated with PCA in the APS group (0.5%) was similar to that reported in Werner et al.’s systematic review.27 We found no evidence to support APS in preventing or reducing the incidence of opioid-related side effects. However, in comparison with the CWPS group, there was some weak evidence to suggest that APS was associated with milder opioid-related side effects. Figure 3. Cost-effectiveness acceptability curve showing the probability We used a global measure of treatment effectiveness that acute pain service (APS) is cost effective for a range of decision outcome for our cost-effectiveness analysis. The global makers’ maximum willingness to pay (U.S. dollars) for 1 day with highly effective treatment gained. The observed mean incremental cost- measure of treatment effectiveness allows patients to bal- effectiveness ratio of US$151 per patient (x axis) corresponds to a 50% ance the unpleasantness or inconvenience of the pain probability of the acute pain service being cost effective (y axis). service intervention, the personal meaningfulness of improvements in pain and function, and the unpleasantness and meaning of any opioid-related side effects.8 We QOR score on the first day after surgery or in the rate of showed that for every 9 patients treated with the APS, 1 improvement in the QOR score between the 2 groups. It is would experience 1 or more days of highly effective pain possible that the 9-item QOR instrument lacked responsive- treatment. A previous study12 showed that the global ness and discrimination when compared with the 40-item measure of treatment effectiveness is valid and can provide 29 QOR instrument. Nevertheless, we found lower pain estimates of analgesic efficacy equivalent to pain relief. scores at rest, less interference with daily activities because When interpreting whether APS is cost effective, it is of pain, and better global measure of treatment effective- important to ask how much decision makers are willing to ness for the APS group on the first day after surgery. These pay for APS to be highly effective rather than spend funds results suggest that there are some initial benefits associ- on improving analgesic techniques per se for being pain ated with APS over CWPS in the early postoperative free at rest or during movement. Our trial suggests that we period, but that they disappear on the second and third day can be certain that APS is cost effective when the WTP for after surgery. Although we did not measure patients’ locus an extra day gained from a highly effective treatment was of control in pain, we believe that PCA afforded greater more than US$546 per patient. This estimate is higher than 30 perceived control over pain relief and that this benefit the value of what Ͼ90% of Canadian patients were willing extended past the first day after surgery. Not surprisingly, to pay, which was no more than US$245* for APS to APS was associated with additional costs, mainly from staff provide PCA treatment.32 The differences in results may be costs, which were 57% of the total cost of pain treatment. due to differences in median household income, patient Although up to one third of patients in this trial expe- population, and the WTP methodology. It is likely that our rienced moderate to severe pain at rest on the first day after anesthesiologist-led nurse-based APS is cost effective be- Ϫ surgery, the mean difference in pain at rest scores ( 0.9, cause funding of APS reflects public preferences for acute Ϫ Ϫ 95% CI, 1.4 to 0.3) between groups was statistically and pain relief and perceived aversion to major complications clinically significant (28% reduction) if one considers a 20% related to inadequate pain relief.33,34 reduction to represent a minimum clinically important 31 This study has several limitations that need to be difference. Half of our patients experienced moderate to considered. We expanded the coverage of APS care to a severe pain during movement on the first day after surgery. specific group of major surgical procedures for which the This incidence appears to be high in comparison with the benefits of PCA were uncertain. The IV PCA mode ac- results of a meta-analysis of 33 studies by Dolin et al. (95% counted for 92% of the pain techniques given by the APS in CI, 25% to 40%).2 We found differences between the 2 2009 at our hospital. It could be argued that this study groups for pain at rest, but not pain during movement on compared IV PCA to IM analgesia, rather than a compari- the first day after surgery. These conflicting findings may son between APS and CWPS care. However, we believe suggest that many patients are unable to distinguish be- that our study is the latter because we chose to focus on the tween pain at rest and pain during movement, because this health service infrastructure providing the postoperative may be influenced by coughing, need for physiotherapy, pain management rather than the pain technique per se. and dressing changes.2 Therefore, our results for pain In the design of economic analysis alongside clinical during movement require cautious interpretation. trials, it is recommended that the study should be more Pain interference with daily activities—such as walking naturalistic to increase the external validity of the economic ability, mood, sleep, relations with others, and ability to results for which the objective is to determine the value for concentrate—are increasingly being used in conjunction with pain intensity outcomes. This outcome represents the *Conversion of CAN$200 in 1997 to US$245 at 2008 value from physical and mental functions affected by pain during the http://eppi.ioe.ac.uk/costconversion/default.aspx. Accessed April 26, recovery process from anesthesia and surgery. We showed 2010.

1048 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA money (i.e., lowest cost per unit benefit).20,35 That is, the service provided by the APS but many want to main- pragmatic trials evaluate effectiveness in comparison with tain an active role in postoperative pain management.37 We standard care in “real-world” patient populations and have shown that the anesthesiologist-led, nurse-based APS practice settings, and although internal validity may be is likely to be more cost effective than is the CWPS. lower than a standard randomized controlled trial, they are more suitable for collection of health economic data.20,35 RECUSE NOTE Although PCA is a common device used by APS, many Tony Gin is section editor of Anesthetic Clinical Pharmacology patients (63%) who are not cared for by an anesthesiologst- for the Journal. This manuscript was handled by Spencer S. based service also receive PCA.6 Given the same availabil- Liu, section editor of Pain Medicine, and Dr. Gin was not ity of oral/IM analgesia and PCA technology, a dedicated involved in any way with the editorial process or decision. APS will prescribe and use these pain techniques differently than will non-APS physicians.36 For example, if sur- REFERENCES geons prescribed a regular parenteral opioid to patients on the 1. Weiser TG, Regenbogen SE, Thompson KD, Haynes AB, first day after surgery, we would expect pain relief outcomes Lipsitz SR, Berry WR, Gawande AA. An estimation of the global volume of surgery: a modelling strategy based on to be similar to those of the APS group. However, because available data. Lancet 2008;372:139–44 there were significant differences in pain outcomes between 2. Dolin SJ, Cashman JN, Bland JM. Effectiveness of acute post- the groups on the first day after surgery, it is likely that operative pain management: I. Evidence from published data. patients were prescribed PRN analgesia in the CWPS group. Br J Anaesth 2002;89:409–23 Although surgeons want to maintain an active role in post- 3. Cheung CW, Ying CL, Lee LH, Tsang SF, Tsui SL, Irwin MG. 37 An audit of postoperative intravenous patient-controlled anal- operative pain management, extending the role of APS is gesia with morphine: evolution over the last decade. Eur J Pain likely to be more cost effective than the CWPS by surgeons. 2009;13:464–71 The results of this trial are not generalizable to different 4. Hung CT, Lau LL, Chan CK, Chow B, Chui PT, Ho B, Kung organizational structures of APS. Our APS is an MC, Lui J, Hui T, Ho E, Chan SF, Chen PP. Acute pain services in Hong Kong: facilities, volume, and quality. Hong Kong Med J anesthesiologist-led, nurse-based model that is the typical 2002;8:196–201 model found in Hong Kong and Australian hospitals. In 5. McDonnell A, Nicholl J, Read SM. Acute Pain Teams in many centers, the anesthesiologist in the APS may only England: current provision and their role in postoperative pain attend ward rounds once or twice a week, with most care management. J Clin Nurs 2003;12:387–93 6. Miaskowski C, Crews J, Ready LB, Paul SM, Ginsberg B. given by nurses who specialize in pain medicine. This Anesthesia-based pain services improve the quality of postop- would likely reduce the staffing costs associated with an erative pain management. Pain 1999;80:23–9 anesthesiologist-led, nurse-based APS program. An eco- 7. Lee A, Chan S, Chen PP, Gin T, Lau AS. Economic evaluations nomic analysis of a nursed-based APS33 concluded that it of acute pain service programs: a systematic review. Clin J Pain was cost effective. However, we believe that there is still 2007;23:726–33 8. Australian and New Zealand College of Anaesthetists and uncertainty about its cost effectiveness given that a cost- Faculty of Pain Med. Acute Pain Management: Scientific effectiveness acceptability curve was not drawn. We esti- Evidence. 3nd ed. Melbourne: Australian and New Zealand mate that, if we changed our APS model to a nurse-based College of Anaesthetists, 2010 APS, we would reduce the overall pain cost by 25%. 9. Schug SA, Kehlet H, Bonnet F, Camu F, Fischer B, Joshi G, Neugebauer EAM, Rawal N, Simanski C. Procedure specific However, we are uncertain about what corresponding pain management after surgery—“PROSPECT.” Acute Pain changes in clinical benefits would arise. 2007;9:55–7 The nature of the pain service intervention meant that 10. Hong Kong Government. Hong Kong Government Gazette patients, health care professionals and the research assistants 2003;13/2003(Special Suppl 4, Annex IV): D4668–704 could not be blinded. Therefore, we cannot exclude the 11. Mendoza TR, Chen C, Brugger A, Hubbard R, Snabes M, Palmer SN, Zhang Q, Cleeland CS. The utility and validity of possibility of measurement bias and the Hawthorne effect of the modified brief pain inventory in a multiple-dose postop- the APS visits in this trial. However, we believe that selection erative analgesic trial. Clin J Pain 2004;20:357–62 bias and confounding would be minimal in this randomized 12. Collins SL, Edwards J, Moore RA, Smith LA, McQuay HJ. controlled trial. Because the pain score measured every 4 Seeking a simple measure of analgesia for mega-trials: is a single global assessment good enough? Pain 2001;91:189–94 hours was not part of routine postoperative care on the ward 13. Myles PS, Hunt JO, Nightingale CE, Fletcher H, Beh T, Tanil D, at the time of the study, we collected the daily pain scores Nagy A, Rubinstein A, Ponsford JL. Development and psycho- from patients; however, this may be prone to recall bias. We metric testing of a quality of recovery score after general used multiple primary outcomes measures to adequately anesthesia and surgery in adults. Anesth Analg 1999;88:83–90 capture the complexity of the pain experience and how it may 14. Apfelbaum JL, Gan TJ, Zhao S, Hanna DB, Chen C. Reliability and validity of the perioperative opioid-related symptom be modified by pain management interventions consistent distress scale. Anesth Analg 2004;99:699–709 with best practices in pain research.8 Furthermore, we mini- 15. Chang AM, Ip WY, Cheung TH. Patient-controlled analgesia mized the risk of making a false positive conclusion about the versus conventional intramuscular injection: a cost effective- clinical benefits of APS using methods recommended by the ness analysis. J Adv Nurs 2004;46:531–41 16. Hoch JS, Dewa CS. A clinician’s guide to correct cost- IMMPACT (Initiative on Methods, Measurement, and Pain effectiveness analysis: think incremental not average. Can 25 Assessment in Clinical Trials) guidelines. J Psychiatry 2008;53:267–74 Improving postoperative pain management by the use 17. Bartha E, Carlsson P, Kalman S. Evaluation of costs and of APS remains a public health challenge.38 Many APS face effects of epidural analgesia and patient-controlled intravenous analgesia after major abdominal surgery. Br J Anaesth structural, political, cultural, educational, emotional, and 2006;96:111–7 physical/technological challenges despite considerable ef- 18. Singer JD, Willet JB. Applied Longitudinal Data Analysis. New forts by APS members.38 Many surgeons are satisfied with York: Oxford University Press, 2003

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1049 Cost Effectiveness of Acute Pain Service

19. Fitzmaurice GM, Ravichandran C. A primer in longitudinal 28. McDonnell A, Nicholl J, Read SM. Acute pain teams and the data analysis. Circulation 2008;118:2005–10 management of postoperative pain: a systematic review and 20. Marshall DA, Hux M. Design and analysis issues for economic meta-analysis. J Adv Nurs 2003;41:261–73 analysis alongside clinical trials. Med Care 2009;47:S14–20 29. Myles PS, Weitkamp B, Jones K, Melick J, Hensen S. Validity 21. Glick HA, Doshi JA, Sonnad SS, Polsky D. Economic evalua- and reliability of a postoperative quality of recovery score: the tion in clinical trials. New York: Oxford University Press, 2007 QoR-40. Br J Anaesth 2000;84:11–5 22. Hoch JS, Rockx MA, Krahn AD. Using the net benefit regres- 30. Chumbley GM, Hall GM, Salmon P. Patient-controlled analge- sion framework to construct cost-effectiveness acceptability sia: an assessment by 200 patients. Anaesthesia 1998;53:216–21 curves: an example using data from a trial of external loop 31. Cepeda MS, Africano JM, Polo R, Alcala R, Carr DB. What recorders versus Holter monitoring for ambulatory monitoring decline in pain intensity is meaningful to patients with acute of “community acquired” syncope. BMC Health Serv Res pain? Pain 2003;105:151–7 2006;6:68 32. Badner NH, Komar WE, Craen RA. Patient attitudes regarding 23. Hoch JS, Briggs AH, Willan AR. Something old, something pca and associated costs. Can J Anaesth 1997;44:255–8 new, something borrowed, something blue: a framework for 33. Stadler M, Schlander M, Braeckman M, Nguyen T, Boogaerts the marriage of health econometrics and cost-effectiveness JG. A cost-utility and cost-effectiveness analysis of an acute analysis. Health Econ 2002;11:415–30 pain service. J Clin Anesth 2004;16:159–67 24. Hoch JS. Improving efficiency and value in palliative care with 34. Macario A, Fleisher LA. Is there value in obtaining a patient’s net benefit regression: an introduction to a simple method for willingness to pay for a particular anesthetic intervention? cost-effectiveness analysis with person-level data. J Pain Symp- Anesthesiology 2006;104:906–9 tom Manage 2009;38:54–61 35. Ramsey S, Willke R, Briggs A, Brown R, Buxton M, Chawla A, 25. Turk DC, Dworkin RH, McDermott MP, Bellamy N, Burke LB, Cook J, Glick H, Liljas B, Petitti D, Reed S. Good research Chandler JM, Cleeland CS, Cowan P, Dimitrova R, Farrar JT, practices for cost-effectiveness analysis alongside clinical trials: Hertz S, Heyse JF, Iyengar S, Jadad AR, Jay GW, Jermano JA, the ISPOR RCT-CEA Task Force report. Value Health Katz NP, Manning DC, Martin S, Max MB, McGrath P, 2005;8:521–33 McQuay HJ, Quessy S, Rappaport BA, Revicki DA, Rothman 36. Stacey BR, Rudy TE, Nelhaus D. Management of patient- M, Stauffer JW, Svensson O, White RE, Witter J. Analyzing controlled analgesia: a comparison of primary surgeons and a multiple endpoints in clinical trials of pain treatments: dedicated pain service. Anesth Analg 1997;85:130–4 IMMPACT recommendations. Pain 2008;139:485–93 37. Chan SK, Chui PT, Lee A, Lai PB, Li TY, Gin T. Surgeons’ 26. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE attitudes and perception of an acute pain service. Hong Kong II: a severity of disease classification system. Crit Care Med Med J 2008;14:342–7 1985;13:818–29 38. Powell AE, Davies HT, Bannister J, Macrae WA. Challenge of 27. Werner MU, Soholm L, Rotboll-Nielsen P, Kehlet H. Does an improving postoperative pain management: case studies of acute pain service improve postoperative outcome? Anesth three acute pain services in the UK National Health Service. Analg 2002;95:1361–72 Br J Anaesth 2009;102:824–31

1050 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Pain Mechanisms

Section Editor: Tony L. Yaksh/Quinn H. Hogan

The Effect of Ketamine Anesthesia on the Immune Function of Mice with Postoperative Septicemia

Tetsuya Takahashi, MD, PhD,* Manabu Kinoshita, MD, PhD,† Satoshi Shono, MD,† Yoshiko Habu, PhD,† Takahiro Ogura, MD,* Shuhji Seki, MD, PhD,† and Tomiei Kazama, MD, PhD*

BACKGROUND: It is unknown how ketamine anesthesia immunologically affects the outcome of patients with postoperative septicemia. We investigated the effects of ketamine anesthesia on mice with an Escherichia coli or lipopolysaccharide (LPS) challenge after laparotomy, focusing on phagocytosis by liver macrophages (Kupffer cells) and cytokine production. METHODS: C57BL/6 mice received ketamine or sevoflurane anesthesia during laparotomy, which was followed by an E. coli or LPS challenge; thereafter, mouse survival rates and cytokine secretions were examined. The effects of a ␤-adrenoceptor antagonist, nadolol, on ketamine anesthesia were also assessed to clarify the mechanisms of ketamine-induced immunosuppressive effects. RESULTS: Ketamine anesthesia increased the mouse survival rate after LPS challenge after laparotomy compared with sevoflurane anesthesia, whereas such an effect of ketamine was not observed after E. coli challenge. Ketamine suppressed tumor necrosis factor (TNF) and interferon (IFN)-␥ secretion after LPS and E. coli challenge. When bacterial growth was inhibited using an antibiotic, ketamine anesthesia effectively improved mouse survival after E. coli challenge compared with sevoflurane anesthesia. Neutralization of TNF also improved survival and decreased IFN-␥ secretion after bacterial challenge in antibiotic-treated mice with sevoflurane anesthesia, suggesting that ketamine’s suppression of TNF may improve survival. Ketamine also suppressed in vivo phagocytosis of microspheres by Kupffer cells in LPS-challenged mice. Concomitant use of nadolol with an anesthetic dose of ketamine did not restore TNF suppression in LPS-challenged mice, suggesting a mechanism independent of the ␤-adrenergic pathway. However, it restored TNF secretion under low-dose ketamine (10% anesthetic dose). In contrast, nadolol restored the decrease in phagocytosis by Kupffer cells, which was induced by the anesthetic dose of ketamine via the ␤-adrenergic pathway, suggesting distinct mechanisms. CONCLUSION: Ketamine suppresses TNF production and phagocytosis by Kupffer cells/macrophages. Therefore, unless bacterial growth is well controlled (by an antibiotic), postoperative infection might not improve despite reduction of the inflammatory response. (Anesth Analg 2010;111:1051–8)

nesthesia affects the immune functions of patients. but also C3H/HeJ mice initiated by cecal ligation and punc- The decision to use one or a combination of anes- ture.6 Moreover, it has been reported that ketamine thetics might affect both patient outcome and prog- increases mouse mortality after cecal ligation and punc- A 7 nosis.1 A combination of anesthetics also might have ture. In addition, studies of human polymorphonuclear affected the tumor metastasis in experimental animals.2 leukocytes have demonstrated that ketamine attenuates Ketamine is an anesthetic commonly used for patients and not only proinflammatory cytokine secretion but also laboratory animals with septicemia or trauma. Ketamine phagocytosis and bactericidal activity in vitro.8–11 These reduces proinflammatory cytokine production, and thereby equivocal results may be attributed in part to the differ- improves the survival rate in lipopolysaccharide (LPS)- ent septic models used. Nonetheless, a detailed immu- induced septicemia models.3–5 Whereas ketamine reportedly nological analysis of ketamine anesthesia is required to also improves the survival rate of animals after E. coli infec- fully understand the ramifications of its clinical use. tion,4 other reports have demonstrated that ketamine does not Sepsis is defined as a systemic inflammatory response 12 affect the outcome of sepsis using not only C3H/HeN mice syndrome (SIRS) induced by infection. If SIRS occurs in the absence of massive bacterial infection, suppressing From the *Department of Anesthesiology, National Defense Medical Col- inflammatory responses and cytokine production may lege, 3-2 Namiki Tokorozawa Saitama Japan 359-8513, †Department of increase host survival.13,14 In contrast, when the inflamma- Immunology and Microbiology, National Defense Medical College, 3-2 Namiki Tokorozawa Saitama Japan 359-8513. tory response is insufficient and production of proinflam- Accepted for publication June 2, 2010. matory cytokines is inadequate, the host cannot eliminate 15–17 This work was supported in part by a grant-in-aid for Special Research Program the invading bacteria. Thus, inflammatory responses (Host stress responses to internal and external factors) from the National induced in the presence and absence of bacterial infection Defense Medical College (to S.S.). must be considered separately to fully understand the Disclosure: The authors report no conflict of interest. processes that are critical to survival. Address correspondence to Shuhji Seki, MD, Department of Immunology Ketamine stimulates the ␤-adrenergic pathway in and Microbiology, National Defense Medical College, 3-2 Namiki, To- korozawa, 359-8613 Japan. Address e-mail to [email protected] sympathetic nerves and induces the production of cat- 18,19 ␤ Copyright © 2010 International Anesthesia Research Society echolamines. -Adrenoceptor agonists such as adrena- DOI: 10.1213/ANE.0b013e3181ed12fc line and isoproterenol (catecholamine) suppress LPS-induced

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1051 Ketamine Anesthesia and Postoperative Infection

Figure 1. Experimental design of ketamine anesthesia in mice.

tumor necrosis factor (TNF) production20–22 and natural killer Kanagawa, Japan) and sevoflurane (Maruishi, Osaka, Ja- (NK) cell activity.23–25 Therefore, ketamine-induced sympa- pan) were used as anesthetics. ␣-Galactosylceramide thetic nerve stimulation may impair the host immune (␣-GalCer) was kindly provided by Pharmaceutical Re- system. Nevertheless, there is only limited understanding search Laboratory, Kirin Brewery, Takasaki, Japan. of how ketamine anesthesia affects cytokine production and phagocytosis by macrophages and liver Kupffer cells Ketamine Anesthesia and Surgical Intervention and how the ketamine-stimulated ␤-adrenergic pathway Mice were injected intraperitoneally (IP) with an anesthetic affects host immune functions. We hypothesized that dose of ketamine (100 mg/kg/0.5 mL) or phosphate buffer ketamine anesthesia during surgery may change the sodium (PBS) (0.5 mL) 10 minutes before laparotomy. proinflammatory cytokine response against LPS and During laparotomy, mice were anesthetized with approxi- bacteria challenges and may also affect phagocytosis of mately 0.2% to 0.5% sevoflurane (ketamine group, n ϭ 25) bacteria by Kupffer cells. Based on this hypothesis, we or approximately 2% to 3% sevoflurane (ketamine plus found the immune-suppressive effect of ketamine on sevoflurane group, n ϭ 25) for 5 minutes after ketamine Kupffer cells and its possible mechanism. injection or approximately 2% to 3% sevoflurane for 5 minutes (sevoflurane group, n ϭ 25) in room air via a METHODS vaporizer (Fig. 1A). During laparotomy, we must some- Animals and Regents times change the concentration of sevoflurane to regulate All experiments were approved by the National Defense the movement and respiratory conditions of mice during Medical College Institution Animal Care and Use Commit- surgical maneuvers, as described in our previous study.2 tee. Male C57BL/6 mice (10 weeks old, 25 g) were obtained The 3-cm midline incision was made for the laparotomy from SLC, Inc. (Shizuoka, Japan). Escherichia coli strain B and was closed in layers using 4 to 0 silk sutures. An (ATCC11303, Sigma-Aldrich, St. Louis, MO) and LPS (E. anesthetic dose of ketamine alone cannot satisfactorily coli 0111: B4, Sigma-Aldrich) were used for experiments. sedate mouse movement during laparotomy. Also, ket- Ketamine (preservative-free; Daiichi Sankyo Propharma, amine is not clinically used without other anesthetics. We

1052 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA studied the effects of anesthetic methods using ketamine PBS. Homogenates were serially diluted 10-fold with PBS, but not the direct pharmacological effect of ketamine per se. plated on brain heart infusion agar, and then incubated at Therefore, a small dose of sevoflurane inhalant (approxi- 37°C for 24 hours for colony counting. mately 0.2%–0.5%) was administered to ketamine-treated mice during the laparotomy to obtain the appropriate Isolation of Peripheral Blood Leukocytes, Liver, sedative state. and Spleen MNCs, and In Vitro LPS Stimulation Blood samples were obtained from mice by cardiac punc- E. coli or LPS Challenge ture, and leukocytes were separated by hemolysis. Liver Immediately after wound closure, mice were challenged IP MNCs were prepared as previously described.13,26,27 In with a lethal (1 ϫ 109 colony-forming units (CFU)/mouse) brief, liver specimens were minced, placed in 0.05% colla- or sublethal (1 ϫ 108 CFU/mouse) dose of E. coli (n ϭ 25 in genase solution (Wako, Osaka, Japan), shaken for 20 min- each group) (Fig. 1B). Antibiotic-treated mice were also utes in a 37°C water bath, and passed through steel mesh. challenged IP with 2 ϫ 108 CFU/mouse of E. coli immedi- After mixing in 33% Percoll, the samples were centrifuged ately after laparotomy (Fig. 1C). Similarly, mice were at 500g for 20 minutes at room temperature. Liver MNCs challenged IP with LPS (200 ␮g/mouse) after laparotomy were obtained after the red blood cells were lysed. Spleno- (Fig. 1A, n ϭ 25 in each group). cytes were also obtained after the spleen was passed through a mesh and the red blood cells were lysed. The Antibiotic Treatment (Double and cells (5 ϫ 105 per well in 200 ␮L Roswell Park Memorial Single Injections) Institute [RPMI] 1640 supplemented with 10% fetal bovine Tail vein injections of cefazolin sodium (50 mg/kg) were serum in 96 well flat–bottomed plates) were incubated with ϫ Ϫ6 administered to mice in both the ketamine (with approxi- ketamine (1 10 M) at 37°C, in 5% CO2 for 1 hour. mately 0.2%–0.5% sevoflurane) and sevoflurane (ap- Subsequently, the cells were incubated with LPS (200 proximately 2%–3%) groups, one immediately before the ng/mL) for 2 hours. Three individual experiments were laparotomy and the second 6 hours after the E. coli performed, with three or four samples per each group. challenge (n ϭ 20 in each group). A single injection with cefazolin sodium (50 mg/kg) was also given to ketamine- In Vivo Phagocytosis of Microspheres by anesthetized mice just before the laparotomy but not Kupffer Cells ␮ after bacterial challenge (Fig. 1C, n ϭ 25). Mice were injected with 20 L of Fluoresbrite YG (FITC) Carboxylate Microspheres (75-nm diameter; Polysciences Neutralization of TNF with Antibiotic Treatment Europe, Eppelheim, Germany) (hereafter called FITC- Anti-TNF neutralizing antibody (0.5 mg/mouse, MP6-XT3; microspheres) via tail veins, immediately after LPS/PBS BD Pharmingen) or rat IgG was injected via IV 1 hour challenge (n ϭ 5 in each group). One hour later, liver MNCs before E. coli challenge (2 ϫ 108 CFU/mouse) after lapa- were isolated. After incubation for 10 minutes with an rotomy. The mice received double or single injections with Fc-blocker (2.4 G2; Pharmingen) to prevent nonspecific cefazolin sodium, as described above (Fig. 1D, n ϭ 20 in binding, the MNCs were labeled with PE-conjugated anti- each group). CD11b mAb and PC5-conjugated anti-Gr-1 mAb. Kupffer cells stain positively with CD11b but are negative for Gr-1 ϩ Ϫ Pretreatment with Nadolol staining.15 CD11b Gr-1 Kupffer cells were analyzed for in Nadolol is a nonselective ␤-adrenoceptor antagonist. Mice vivo phagocytosis by flow cytometry (EPICS XL, Coulter, were injected with nadolol (10 mg/kg/0.2 mL) or 0.2 mL Miami, FL). PBS via the tail vein 5 minutes before ketamine/PBS treatment. The mice were then laparotomized, followed by Statistical Analysis LPS challenge (Fig. 1E, n ϭ 5 in each group). To precisely The data are presented as means Ϯ SE. Statistical analysis evaluate the relationship between ketamine and nadolol, was performed using Graphpad-Prism software, version some mice did not receive laparotomy, thereby avoiding 4.00 (Graphpad Software, Inc., San Diego, CA). Survival sevoflurane treatment. Ten minutes after ketamine/PBS frequencies were compared using log rank tests. Changes treatment, these mice were similarly challenged IP with in serum cytokine concentrations were compared using LPS (n ϭ 5 in each group). To examine the affect of nadolol two-way analysis of variance (ANOVA). Other cytokine on mice treated with a low dose of ketamine, mice were levels between groups or among several groups were also injected IP with 10 mg/kg ketamine (10% of the compared using the Student t test or one-way ANOVA anesthetic induction dose) 5 minutes after the nadolol/PBS followed by the Newman-Keuls test. The survival of treatment (n ϭ 5 in each group). antibiotic-treated mice was evaluated using Fisher exact test. The level of significance was set at P Ͻ 0.05. Sample Cytokine Measurements and Bacterial Counts sizes were chosen based on data in other papers in this Blood samples of individual mice were obtained from the field.3,5,28–30 retro-orbital sinus immediately before anesthesia administration and at 1, 3, 6, 12, 24, and 48 hours after E. coli or LPS RESULTS challenge. Sera and culture supernatants of mononuclear Ketamine Anesthesia Improves Survival After cells (MNCs) were detected using Enzyme-Linked Immuno- Laparotomy with LPS Challenge and Suppresses Sorbent Assay (ELISA) kits (TNF and interferon [IFN]-␥, Elevation of Serum Proinﬂammatory Cytokines BD Pharmingen, San Diego, CA; IL-12, Endogen, Woburn, Both ketamine-treated groups (anesthetized either with MA). Livers were aseptically removed and homogenized in 0.2%–0.5% sevoflurane or 2%–3% sevoflurane) showed

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1053 Ketamine Anesthesia and Postoperative Infection

Figure 2. The effect of ketamine anesthesia on survival (A), serum tumor necrosis factor (TNF) (B), interleukin (IL)-12 (C), and interferon (IFN)-␥ (D) levels in mice after lipopolysaccharide (LPS) challenge after laparotomy. Mice were anesthetized with ketamine ϩ0.2% to 0.5% sevoflurane (ketamine group), ketamine ϩ2% to 3% sevoflurane (sevoflurane ϩ ketamine group) or 2% to 3% sevoflurane (sevoflurane group). Data are the means Ϯ SE, n ϭ 25 in each group. *P Ͻ 0.01, †P Ͻ 0.05 vs other groups.

Table 1. Serum TNF and IFN-␥ Levels in Mice After E. coli Challenge Following Laparotomy Sevoflurane Ketamine Serum TNF at 1 h 1 ϫ 108 CFU 119 Ϯ 14 66 Ϯ 2* 1 ϫ 109 CFU 404 Ϯ 53 146 Ϯ 19* Serum IFN-␥ at6h 1 ϫ 108 CFU 396 Ϯ 63 145 Ϯ 32* 1 ϫ 109 CFU 1103 Ϯ 181 530 Ϯ 51* TNF ϭ tumor necrosis factor; IFN ϭ interferon. Figure 3. The effects of ketamine anesthesia on mouse survival Sera were obtained from the mice 1 hour and 6 hours after E. coli challenge after E. coli challenge after laparotomy. Mice received ketamine to measure serum TNF and IFN-␥, respectively. Data are means Ϯ SE (pg/mL) anesthesia (ϩ0.2%–0.5% sevoflurane) or sevoflurane (2%–3%) an- from 25 mice in each group. * P Ͻ 0.01 vs sevoflurane. esthesia. Thereafter, they were challenged IP with sublethal (1 ϫ 108 colony-forming units (CFU)/mouse) or lethal (1 ϫ 109 CFU/mouse) doses of E. coli after laparotomy; n ϭ 25 in each group. CFU/mouse) (Fig. 3), despite suppression of TNF and IFN-␥ (Table 1). significantly higher survival rates and greater suppres- Ketamine Anesthesia Plus Antibiotic Therapy sion of serum peaks of TNF, interleukin (IL)-12, and Increases Survival After Laparotomy with IFN-␥ compared with the sevoflurane (approximately E. coli Challenge 2%–3%) group (Fig. 2). However, the two ketamine- Unlike LPS challenge, bacterial growth and proliferation treated groups showed similar survival rates and serum could affect the survival of E. coli–challenged mice after cytokine levels, suggesting that ketamine increases ketamine treatment. Because perioperative septic patients mouse survival and suppresses proinflammatory cyto- are usually treated with antibiotics, E. coli–challenged mice kine secretion. The differences observed between the were given two injections of cefazolin. In addition, TNF mice treated with ketamine/low-dose sevoflurane (ket- was depleted in mice receiving sevoflurane anesthesia and amine group) versus high-dose sevoflurane alone cefazolin injections to determine if TNF levels decreased by (sevoflurane group) might not have been due to the use ketamine might explain the increased survival of cefazolin- of a lower dose of sevoflurane but due to the use of treated mice in postoperative infection. Concomitant use of ketamine. Therefore, the affect of ketamine anesthesia on cefazolin with ketamine anesthesia significantly increased the ketamine (with low-dose sevoflurane) and sevoflu- survival after E. coli challenge with suppression of TNF and rane groups was compared in further experiments. IFN-␥ compared with sevoflurane anesthesia alone (Fig. 4, Table 2). The TNF-depleted mice also showed a marked Ketamine Anesthesia Does Not Increase increase in survival after postlaparotomy E. coli challenge Survival After Laparotomy with E. coli after sevoflurane anesthesia (Fig. 4) with significant sup- Challenge Despite Suppression of Serum pression of IFN-␥ (Table 2). TNF and IFN-␥ Mice in the ketamine group did not survive at higher rates Suppression of TNF Decreases Survival in after laparotomy with E. coli (1 ϫ 108 CFU/mouse) chal- E. coli–Challenged Mice Under Insufficient lenge compared with those in the sevoflurane group (Fig. Regulation of Bacterial Growth 3), despite significant suppression of TNF and IFN-␥ (Table TNF-neutralizing antibodies (Ab) was administered to 1). Ketamine anesthesia also failed to increase survival after single cefazolin-treated mice before laparotomy (but not laparotomy with lethal E. coli challenge (1 ϫ 109 after E. coli challenge) followed by sevoflurane anesthesia.

1054 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 4. The Effect of Antibiotic Treatment on Survival After E. coli Challenge Following Laparotomy in Ketamine-Anesthetized Mice Survival Without injection Single injection Double injections 36 h 10/25* 20/25 21/25 72 h (3d) 5/25* 17/25 20/25 168 h (7d) 4/25* 12/25§ 19/25 Ketamine-anesthetized mice were treated with a single injection of cefazolin, double injections, or without injection. * P Ͻ 0.01 vs other groups, § P Ͻ 0.05 vs double injections. Figure 4. The effect of ketamine anesthesia and neutralizing tumor necrosis factor (TNF) ϩ antibiotic therapy on mouse survival after sublethal E. coli challenge after laparotomy. Mice received ketamine Table 5. The Effect of In Vitro Ketamine anesthesia, sevoflurane anesthesia, or sevoflurane anesthesia after Treatment on LPS-Stimulated TNF Production TNF neutralization. They were then treated twice with cefazolin and challenged IP with 2 ϫ 108 E. coli colony-forming units (CFU)/mouse Ketamine ϭ Ͻ ؊ after laparotomy; n 20 in each group, *P 0.01. Without ketamine With ketamine (10 6 M) Circulating leukocytes 240 Ϯ 20 111 Ϯ 20* Liver MNCs 548 Ϯ 31 278 Ϯ 20* Table 2. Serum TNF and IFN-␥ Levels in Mice Spleen MNCs 200 Ϯ 12 82 Ϯ 5* After Antibiotic-Treated E. coli Challenge LPS ϭ lipopolysaccharide; MNCs ϭ mononuclear cells. Following Laparotomy Data are means Ϯ SE (pg/mL) from three individual experiments with two .Sevoflurane ؉ samples per each group. * P Ͻ 0.05 vs without ketamine Sevoflurane Anti-TNF Ab Ketamine Serum TNF 172.0 Ϯ 23.2* Undetected 58.6 Ϯ 2.2* Serum IFN-␥ 649.3 Ϯ 86.7* 124.5 Ϯ 13.1 146.7 Ϯ 18.1 Table 6. The Effect of Ketamine Treatment TNF ϭ tumor necrosis factor; IFN ϭ interferon; Ab ϭ antibodies. on Phagocytosis by Kupffer Cells in Cefazolin (double injections)-treated mice were injected IV with anti-TNF LPS-Challenged Mice neutralizing Ab or rat IgG, as a control, 1 hour before E. coli challenge. Sera Without laparotomy were obtained from the mice at 1 hour and 6 hour to measure serum TNF and Treatment PBS Ketamine PBS Ketamine IFN-␥, respectively. Data are means Ϯ SE (pg/mL) from 20 mice in each group. * P Ͻ 0.01 vs other groups. Challenge PBS PBS LPS LPS Phagocytosis 13.6 Ϯ 1 7.5 Ϯ 0.8* 20.7 Ϯ 1.1* 14.02 Ϯ 1.1 (%) Laparotomy Table 3. The Effect of Neutralizing Anti-TNF Anesthesia Sevoflurane Ketamine Sevoflurane Ketamine Antibody on Survival After E. coli Challenge of Challenge PBS PBS LPS LPS Mice Treated by a Single Injection of Antibiotic Phagocytosis 14.9 Ϯ 0.5 9.4 Ϯ 0.8* 21.4 Ϯ 0.7* 14.6 Ϯ 0.6 Survival (%) ϭ ϭ .Sevoflurane Sevoflurane ؉ Anti-TNF-Ab PBS phosphate buffer sodium; LPS lipopolysaccharide Proportion of microsphere phagocytosis by Kupffer cells are shown as 36 h 15/20 17/20 means Ϯ SE from 5 mice in each group. * P Ͻ 0.05 vs other groups. 72 h (3 d) 13/20 4/20* 168 h (7 d) 12/20 3/20* TNF ϭ tumor necrosis factor; Ab ϭ antibodies. survival comparable to the mice with two cefazolin injec- Mice were injected IV with anti-TNF neutralizing antibody or rat IgG 1 hour before E. coli challenge following laparotomy. Cefazolin was injected once tions but showed significantly lower survival at 7 days, immediately before laparotomy. * P Ͻ 0.05 vs sevoflurane. although both antibiotic treatments increased mouse survival after bacterial challenge (Table 4). These findings suggest that unless bacterial growth/proliferation is well controlled (by repeated antibiotic injections), suppression Most survived the initial 36-hour period but eventually of TNF (by neutralizing Ab or ketamine) may adversely died 7 days after infection (Table 3). Bacteria were then affect surgical patients with bacterial infections. counted in the livers of TNF-depleted mice treated with cefazolin (one or two injections) or without cefazolin 24 Ketamine Suppresses LPS-Induced TNF hours after E. coli challenge. A single injection of cefazolin Production by Cultured Lymphocytes In Vitro did not effectively suppress bacterial growth (counts) in the Coculturing with ketamine significantly suppressed TNF Ϯ liver compared with double injections (single; 1.1 0.4 production from blood leukocytes, liver, or spleen MNCs Ϯ ϫ 6 Ͻ ϭ versus double; 0.1 0.1 10 CFU, P 0.05, n 6 in each stimulated by LPS in vitro, suggesting that ketamine di- group), although growth was significantly inhibited in rectly suppresses LPS-induced TNF production by these Ϯ antibiotic-treated mice (without antibiotic injection; 5.1 cells (Table 5). 0.3 ϫ 106 CFU, P Ͻ 0.01, n ϭ 6). Because ketamine suppressed TNF secretion, survival after E. coli challenge Ketamine Attenuates In Vivo Phagocytic Activity was examined in the mice receiving ketamine anesthesia of Kupffer Cells in LPS-Challenged Mice that were treated with a single injection of cefazolin. Ketamine treatment significantly decreased phagocytosis During the initial 36 hours, these mice exhibited a rate of by Kupffer cells in mice challenged with PBS (Table 6).

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1055 Ketamine Anesthesia and Postoperative Infection

Table 8. The Effect of Nadolol on Phagocytosis by Kupffer Cells in With or Without Ketamine- Treated Mice Without laparotomy Pretreatment PBS PBS Nadolol Nadolol Anesthesia PBS Ketamine PBS Ketamine Phagocytosis (%) 14.5 Ϯ 0.7 8.4 Ϯ 0.5* 15.8 Ϯ 1.3 15.8 Ϯ 1.5 Laparotomy Pretreatment PBS PBS Nadolol Nadolol Anesthesia Sevoﬂurane Ketamine Sevoﬂurane Ketamine Phagocytosis (%) 15.8 Ϯ 0.9 9.6 Ϯ 0.8* 16.7 Ϯ 1.0 15.0 Ϯ 0.9 PBS ϭ phosphate buffer sodium. Proportion of microsphere phagocytosis by Kupffer cells are shown as means Ϯ SE from 5 mice in each group. * P Ͻ 0.05 vs other groups.

stress such as surgery or traumatic injury induces hyporesponsiveness of TNF secretion to LPS, in particular, the early phase after surgery,31–34 but TNF suppression by ketamine anesthesia (100 mg/kg) did not significantly Figure 5. The effect of nadolol on serum tumor necrosis factor (TNF) differ in mice with (Table 7) or without laparotomy (Fig. 5). levels after lipopolysaccharide (LPS) challenge in mice given the Phagocytosis by Kupffer cells was examined in LPS- anesthetic or low dose of ketamine. After pretreatment with nadolol or phosphate buffer sodium (PBS), mice were treated with ketamine challenged mice (without laparotomy) that received an (100 mg kgϪ1 ϭ anesthetic induction dose; 10 mg kgϪ1 ϭ 10% anesthetic dose of ketamine. Unlike the TNF response, anesthetic induction dose) or PBS followed by LPS challenge. Sera nadolol almost completely restored the ketamine- were obtained from the mice 1 hour after LPS challenge. Data are suppressed phagocytic activity of Kupffer cells (Table 8). the means Ϯ SE from ﬁve mice in each group. §P Ͻ 0.05, *P Ͻ 0.01 vs other groups. This restoration by nadolol was also observed in laparotomized mice (Table 8).

Table 7. The Effect of Nadolol on the Serum TNF DISCUSSION Levels After LPS Challenge Following Laparotomy Ketamine anesthesia improved survival in LPS-challenged Pretreatment PBS PBS Nadolol Nadolol mice but not in mice challenged with viable E. coli. Never- Anesthesia Sevoflurane Ketamine Sevoflurane Ketamine theless, the survival rate after E. coli challenge was Serum TNF 989 Ϯ 74 248 Ϯ 47* 1020 Ϯ 109 239 Ϯ 34* improved by the concomitant use of an antibiotic and levels at 1 h ketamine anesthesia, suggesting that inhibition of bacterial TNF ϭ tumor necrosis factor; LPS ϭ lipopolysaccharide; growth critically influences the ketamine-induced benefi- Data are means Ϯ SE (pg/mL) from 5 mice in each group. * P Ͻ 0.01 vs cial effect on survival. sevoflurane. Since the septic condition induced by LPS is merely a result of SIRS without infection, suppression of proinflammatory cytokines by ketamine improves mouse survival. In Although LPS challenge increased phagocytosis by Kupffer contrast, septicemia after E. coli challenge is a result of cells, ketamine treatment also significantly decreased their infection accompanied by SIRS. TNF and IFN-␥ are impor- phagocytosis (Table 6). Although laparotomy alone did not tant for the elimination of invading bacteria, although they significantly affect the phagocytic activity by Kupffer cells, also induce shock and organ injury.13,14,16 The suppression ketamine anesthesia significantly decreased the phagocyto- of proinflammatory cytokines by ketamine may thereby sis after laparotomy (Table 6). attenuate bacterial elimination. Ketamine also decreased the phagocytic activity of Kupffer cells, indicating further The Effect of Nadolol Pretreatment on impairment of bacterial clearance. These findings may Ketamine-Suppressed Proinflammatory Cytokine explain why ketamine (without antibiotic) could not im- Response or Phagocytosis in LPS-Challenged Mice prove the survival of E. coli–infected mice. A subanesthetic Although the anesthetic dose of ketamine (100 mg/kg) dose of ketamine reportedly suppressed LPS-stimulated markedly suppressed serum TNF levels 1 hour after LPS TNF production by peripheral blood MNCs in patients,11 challenge in mice (without laparotomy), nadolol pretreat- suggesting that ketamine, even at a low dose, suppresses ment did not abolish the ketamine-induced suppression of LPS-induced TNF production in humans and mice. TNF (Fig. 5). A low dose of ketamine (10 mg/kg, 10% of the IL-12 and ␣-GalCer, a synthetic NKT cell ligand, were used anesthetic dose) also suppressed serum TNF levels after to determine whether ketamine directly affects IFN-␥ produc- LPS challenge; however, nadolol pretreatment significantly tion from NK and/or NKT cells. Exogenous IL-12 directly restored TNF levels in mice (Fig. 5). Further examination of induces IFN-␥ production by mouse NKT cells.35–37 ␣-GalCer LPS-induced TNF secretion after laparotomy revealed that also directly stimulates IFN-␥ production from NKT cells and laparotomy obviously suppressed LPS-induced TNF eleva- subsequently from NK cells.2,28,38–41 Ketamine anesthesia did tion in sevoflurane-anesthetized mice (Table 7, Fig. 5). not suppress serum IFN-␥ levels after IL-12 or ␣-GalCer Many investigators have also demonstrated that surgical challenge in mice (data not shown), suggesting that the IFN-␥

1056 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA producing capacities of NK/NKT cells are not directly af- phagocytosis by macrophages/Kupffer cells, which pre- fected by ketamine anesthesia but are suppressed primarily sumably enhance bacterial growth/proliferation in periop- through reduced IL-12 production from macrophages/ erative septic patients. Therefore, antibiotic treatment could Kupffer cells. be important for septic patients who received ketamine Sevoflurane anesthesia for 15 to 30 minutes reportedly anesthesia. It should also be noted that ketamine reportedly suppresses LPS-induced TNF secretion in mice.42,43 In this injures Kupffer cells and endothelial cells and may affect study, sevoflurane was only administered for 5 minutes. liver functions.46 The TNF elevation observed in the ketamine plus high-dose sevoflurane group was similar to that seen after LPS REFERENCES challenge to the ketamine plus low-dose sevoflurane group, 1. Exadaktylos AK, Buggy DJ, Moriarty DC, Mascha E, Sessler DI. suggesting that ketamine-induced suppression of TNF is Can anesthetic technique for primary breast cancer surgery affect recurrence or metastasis? Anesthesiology 2006;105:660–4 independent of sevoflurane dose. 2. Wada H, Seki S, Takahashi T, Kawarabayashi N, Higuchi H, Nadolol did not inhibit suppression of TNF secretion by Habu Y, Sugahara S, Kazama T. Combined spinal and general an anesthetic dose of ketamine (100 mg/kg), although it anesthesia attenuates liver metastasis by preserving TH1/TH2 significantly restored TNF secretion under low-dose ket- cytokine balance. Anesthesiology 2007;106:499–506 3. Taniguchi T, Takemoto Y, Kanakura H, Kidani Y, Yamamoto amine (10 mg/kg). Low-dose ketamine also reportedly K. The dose-related effects of ketamine on mortality and increased serum adenosine levels (0.4 ␮M) and thereby cytokine responses to endotoxin-induced shock in rats. Anesth inhibited secretion of proinflammatory cytokines after LPS Analg 2003;97:1769–72 challenge,5 as evidenced by use of an adenosine receptor 4. Shaked G, Czeiger D, Dukhno O, Levy I, Artru AA, Shapira Y, Douvdevani A. Ketamine improves survival and suppresses antagonist. However, this study did not examine how the IL-6 and TNFalpha production in a model of Gram-negative anesthetic dose of ketamine affects serum adenosine levels bacterial sepsis in rats. Resuscitation 2004;62:237–42 or proinflammatory cytokine levels. Propranolol (another 5. Mazar J, Rogachev B, Shaked G, Ziv NY, Czeiger D, Chaimo- ␤-adrenoceptor blocker) inhibited the secretion of adeno- vitz C, Zlotnik M, Mukmenev I, Byk G, Douvdevani A. sine from the kidneys after electric stimulation of the Involvement of adenosine in the antinflammatory action of 44 ketamine. Anesthesiology 2005;102:1174–81 periarterial sympathetic nerve. Therefore, adenosine and, 6. Imai T, Takahashi K, Masuo F, Goto F. Anaesthesia affects presumably, adrenaline appear to be responsible for the outcome of sepsis in mice. Can J Anaesth 1998;45:360–6 reduction of TNF production by activated-macrophages 7. Hansbrough JF, Zapata-Sirvent RL, Bartle EJ, Anderson JK, Elliott after a low dose of ketamine. However, the direct effect L, Mansour MA, Carter, WH. Alterations in splenic lymphocyte ␤ subpopulations and increased mortality from sepsis following (non- -adrenergic pathway) of an anesthetic dose of ket- anesthesia in mice. Anesthesiology 1985;63:267–73 amine may overcome or overwhelm the inhibitory effect 8. Krumholz W, Endrass J, Hempelmann G. Inhibition of phago- via the ␤-adrenergic pathway by nadolol (as seen with cytosis and killing of bacteria by anaesthetic agents in vitro. low-dose ketamine) and further suppress TNF production. Br J Anaesth 1995;75:66–70 9. Heller A, Heller S, Blecken S, Urbaschek R, Koch T. Effects of The addition of ketamine to cultures consistently suppressed intravenous anesthetics on bacterial elimination in human LPS-induced TNF production from isolated MNCs in vitro, blood in vitro. Acta Anaesthesiol Scand 1998;42:518–26 suggesting a direct suppression of TNF by ketamine. 10. Nishina K, Akamatsu H, Mikawa K, Shiga M, Maekawa N, The results of this study also demonstrated for the first Obara H, Niwa Y. The inhibitory effects of thiopental, midazolam, and ketamine on human neutrophil functions. Anesth time that ketamine anesthesia significantly inhibited in vivo Analg 1998;86:159–65 phagocytic activity by Kupffer cells. Previous in vitro studies 11. Beilin B, Rusabrov Y, Shapira Y, Roytblat L, Greemberg L, had shown that ketamine did not inhibit phagocytosis by Yardeni IZ, Bessler H. Low-dose ketamine affects immune either polymorphonuclear leukocytes or macrophages at clini- responses in humans during the early postoperative period. cal concentrations, although it could suppress at higher con- Br J Anaesth 2007;99:522–7 12. Am College of Chest Physicians/Society of Critical Care Med 9,10,45 ␤ centrations. Interestingly, inhibiting the -adrenergic Consensus Conference: definitions for sepsis and organ failure pathway by nadolol restored ketamine-suppressed phagocy- and guidelines for the use of innovative therapies in sepsis. tosis by Kupffer cells. The anesthetic dose of ketamine might Crit Care Med 1992;20:864–74 directly suppress TNF secretion in LPS-challenged mice inde- 13. Ogasawara K, Takeda K, Hashimoto W, Satoh M, Okuyama R, ␤ Yanai N, Obinata M, Kumagai K, Takada H, Hiraide H, Seki S. pendent of the -adrenergic pathway, while also indirec- Involvement of NK1ϩ T cells and their IFN-gamma produc- tly suppressing phagocytosis by Kupffer cells via the tion in the generalized Shwartzman reaction. J Immunol ␤-adrenergic pathway. Inhibiting the ␤-adrenergic pathway 1998;160:3522–7 by nadolol may be effective in ketamine anesthesia for septic 14. Sato K, Kinoshita M, Motegi A, Habu Y, Takayama E, Nonoyama S, Hiraide H, Seki S. Critical role of the liver CD8ϩ patients because it might abrogate the enhanced catechol- CD122ϩ T cells in the generalized Shwartzman reaction of amine effect. However, when nadolol was administered to E. mice. Eur J Immunol 2005;35:593–602 coli–challenged mice under ketamine anesthesia without us- 15. Kinoshita M, Uchida T, Nakashima H, Ono S, Seki S, Hiraide ing an antibiotic, survival was not improved (unpublished H. Opposite effects of enhanced tumor necrosis factor-alpha production from Kupffer cells by gadolinium chloride on liver observations). Nadolol may induce heart failure and severe injury/mortality in endotoxemia of normal and partially hepa- hypotension in mice with septicemia under ketamine anes- tectomized mice. Shock 2005;23:65–72 thesia and thereby mask restoration of the phagocytic func- 16. Kinoshita M, Seki S, Ono S, Shinomiya N, Hiraide H. Paradoxical tion of macrophages. effect of IL-18 therapy on the severe and mild Escherichia coli infections in burn-injured mice. Ann Surg 2004;240:313–20 Ketamine anesthesia may be preferred for septic patients 17. Watanabe H, Numata K, Ito T, Takagi K, Matsukawa A. Innate because of its suppressive effects of proinflammatory cyto- immune response in Th1- and Th2-dominant mouse strains. kines. However, ketamine also might decrease bacterial Shock 2004;22:460–6

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1057 Ketamine Anesthesia and Postoperative Infection

18. Zsigmond EK, Kelsch RC, Kothary SP. Rise in plasma free- 33. Kawasaki T, Ogata M, Kawasaki C, Tomihisa, T, Okamoto K, norepinephrine during anesthetic induction with ketamine. Shigematsu A. Surgical stress induces endotoxin hyporespon- Behav Neuropsychiatry 1974;6:81–4 siveness and an early decrease of monocyte mCD14 and 19. Clanachan AS, McGrath JC. Effects of ketamine on the periph- HLA-DR expression during surgery. Anesth Analg 2001; eral autonomic nervous system of the rat. Br J Pharmacol 92:1322–6 1976;58:247–52 34. Lahat N, Rahat MA, Brod V, Cohen S, Weber G, Kinarty A, 20. Severn A, Rapson NT, Hunter CA, Liew FY. Regulation of Bitterman H. Abdominal surgery reduces the ability of rat tumor necrosis factor production by adrenaline and beta- spleen cells to synthesize and secrete active tumour necrosis adrenergic agonists. J Immunol 1992;148:3441–5 factor-alpha (TNF-alpha) by a multilevel regulation. Clin Exp 21. Spengler RN, Chensue SW, Giacherio DA, Blenk N, Kunkel SL. Immunol 1999;115:19–25 Endogenous norepinephrine regulates tumor necrosis factor- 35. Seki S, Hashimoto W, Ogasawara K, Satoh M, Watanabe H, alpha production from macrophages in vitro. J Immunol Habu Y, Hiraide H, Takeda K. Antimetastatic effect of NK1ϩ T 1994;152:3024–31 cells on experimental haematogenous tumour metastases in 22. Verhoeckx KC, Doornbos RP, van der Greef J, Witkamp RF, the liver and lungs of mice. Immunology 1997;92:561–6 Rodenburg RJ. Inhibitory effects of the beta-adrenergic recep- 36. Hashimoto W, Takeda K, Anzai R, Ogasawara K, Sakihara H, tor agonist zilpaterol on the LPS-induced production of TNF- Sugiura K, Seki S, Kumagai K. Cytotoxic NK1.1 Agϩ alpha alpha in vitro and in vivo. J Vet Pharmacol Ther 2005;28:531–7 beta T cells with intermediate TCR induced in the liver of mice 23. Schedlowski M, Hosch W, Oberbeck R, Benschop RJ, Jacobs R, by IL-12. J Immunol 1995;154:4333–40 Raab HR, Schmidt RE. Catecholamines modulate human NK 37. Habu Y, Uchida T, Inui T, Nakashima H, Fukasawa M, Seki S. cell circulation and function via spleen-independent beta Enhancement of the synthetic ligand-mediated function of 2-adrenergic mechanisms. J Immunol 1996;156:93–9 liver NK1.1Agϩ T cells in mice by interleukin-12 pretreatment. 24. Shakhar G, Ben-Eliyahu S. In vivo beta-adrenergic stimulation Immunology 2004;113:35–43 suppresses natural killer activity and compromises resistance 38. Nakagawa R, Nagafune I, Tazunoki Y, Ehara H, Tomura H, to tumor metastasis in rats. J Immunol 1998;160:3251–8 Iijima R, Motoki K, Kamishohara M, Seki S. Mechanisms of the 25. Kappel M, Poulsen TD, Galbo H, Pedersen BK. Effects of antimetastatic effect in the liver and of the hepatocyte injury elevated plasma noradrenaline concentration on the immune induced by alpha-galactosylceramide in mice. J Immunol system in humans. Eur J Appl Physiol Occup Physiol 2001;166:6578–84 1998;79:93–8 39. Carnaud C, Lee D, Donnars O, Park SH, Beavis A, Koezuka Y, 26. Dobashi H, Seki S, Habu Y, Ohkawa T, Takeshita S, Hiraide H, Bendelac A. Cutting edge: Cross-talk between cells of the Sekine I. Activation of mouse liver natural killer cells and innate immune system: NKT cells rapidly activate NK cells. NK1.1(ϩ) T cells by bacterial superantigen-primed Kupffer J Immunol 1999;163:4647–50 cells. Hepatology 1999;30:430–6 40. Nakagawa R, Inui T, Nagafune I, Tazunoki Y, Motoki K, 27. Habu Y, Seki S, Takayama E, Ohkawa T, Koike Y, Ami K, Majima Yamauchi A, Hirashima M, Habu Y, Nakashima H, Seki S. T, Hiraide H. The mechanism of a defective IFN-gamma response ϩ ϩ to bacterial toxins in an atopic dermatitis model, NC/Nga mice, Essential role of bystander cytotoxic CD122 CD8 T cells for and the therapeutic effect of IFN-gamma, IL-12, or IL-18 on the antitumor immunity induced in the liver of mice by dermatitis. J Immunol 2001;166:5439–47 alpha-galactosylceramide. J Immunol 2004;172:6550–7 28. Inui T, Nakagawa R, Ohkura S, Habu Y, Koike Y, Motoki K, 41. Kawabata T, Kinoshita M, Inatsu A, Habu Y, Nakashima H, Kuranaga N, Fukasawa M, Shinomiya N, Seki S. Age- Shinomiya N, Seki S. Functional alterations of liver innate associated augmentation of the synthetic ligand- mediated immunity of mice with aging in response to CpG- function of mouse NK1.1 ag(ϩ) T cells: their cytokine produc- oligodeoxynucleotide. Hepatology 2008;48:1586–97 tion and hepatotoxicity in vivo and in vitro. J Immunol 42. Kidani Y, Taniguchi T, Kanakura H, Takemoto Y, Tsuda K, 2002;169:6127–32 Yamamoto K. Sevoflurane pretreatment inhibits endotoxin- 29. Gurfinkel R, Czeiger D, Douvdevani A, Shapira Y, Artru AA, induced shock in rats. Anesth Analg 2005;101:1152–6 Sufaro Y, Mazar J, Shaked G. Ketamine improves survival in 43. Hofstetter C, Boost KA, Flondor M, Basagan-Mogol E, Betz C, burn injury followed by sepsis in rats. Anesth Analg Homann M, Muhl H, Pfeilschifter J, Zwissler B. Anti- 2006;103:396–402 inflammatory effects of sevoflurane and mild hypothermia in 30. Shibakawa YS, Sasaki Y, Goshima Y, Echigo N, Kamiya Y, endotoxemic rats. Acta Anaesthesiol Scand 2007;51:893–9 Kurahashi K, Yamada Y, Andoh T. Effects of ketamine and 44. Mi Z, Jackson EK. Effects of alpha- and beta-adrenoceptor propofol on inflammatory responses of primary glial cell blockade on purine secretion induced by sympathetic nerve cultures stimulated with lipopolysaccharide. Br J Anaesth stimulation in the rat kidney. J Pharmacol Exp Ther 2005;95:803–10 1999;288:295–301 31. Majetschak M, Flach R, Kreuzfelder E, Jennissen V, Heukamp 45. Davidson JA, Boom SJ, Pearsall FJ, Zhang P, Ramsay G. T, Neudeck F, Schmit-Neuerburg KP, Obertacke U, Schade FU. Comparison of the effects of four i.v. anaesthetic agents on The extent of traumatic damage determines a graded depres- polymorphonuclear leucocyte function. Br J Anaesth sion of the endotoxin responsiveness of peripheral blood 1995;74:315–8 mononuclear cells from patients with blunt injuries. Crit Care 46. Thompson JS, Brown SA, Khurdayan V, Zeynalzadedan A, Med 1999;27:313–8 Sullivan PG, Scheff SW. Early effects of tribromoethanol, 32. Ogata M, Okamoto K, Kohriyama K, Kawasaki T, Itoh H, ketamine/xylazine, pentobarbitol, and isoflurane anesthesia Shigematsu A. Role of interleukin-10 on hyporesponsiveness on hepatic and lymphoid tissue in ICR mice. Comp Med of endotoxin during surgery. Crit Care Med 2000;28:3166–70 2002;52:63–7

1058 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Regional Anesthesia

Section Editor: Terese T. Horlocker

Estimation and Pharmacodynamic Consequences of the Minimum Effective Anesthetic Volumes for Median and Ulnar Nerve Blocks: A Randomized, Double-Blind, Controlled Comparison Between Ultrasound and Nerve Stimulation Guidance

Matthieu Ponrouch, MD,* Nicolas Bouic, MD,* Sophie Bringuier, PharmD, PhD,† Philippe Biboulet, MD,* Olivier Choquet, MD,* Miche`le Kassim, MD,* Nathalie Bernard, MD, MSc,* and Xavier Capdevila, MD, PhD‡

BACKGROUND: Nerve stimulation and ultrasound guidance are the most popular techniques for peripheral nerve blocks. However, the minimum effective anesthetic volume (MEAV) in selected nerves for both techniques and the consequences of decreasing the local anesthetic volume on the pharmacodynamic characteristics of nerve block remain unstudied. We designed a randomized, double-blind controlled comparison between neurostimulation and ultrasound guidance to estimate the MEAV of 1.5% mepivacaine and pharmacodynamics in median and ulnar nerve blocks. METHODS: Patients scheduled for carpal tunnel release were randomized to ultrasound guidance (UG) or neurostimulation (NS) groups. A step-up/step-down study model (Dixon method) was used to determine the MEAV with nonprobability sequential dosing based on the outcome of the previous patient. The starting dose of 1.5% mepivacaine was 13 and 11 mL for median and ulnar nerves at the humeral canal. Block success/failure resulted in a decrease/increase of 2 mL. A blinded physician assessed sensory blockade at 2-minute intervals for 20 minutes. Block onset time and duration were noted. RESULTS: The MEAV50 (SD) of the median nerve was lower in the UG group 2 (0.1) mL (95% confidence interval [CI] ϭ [1, 96] to [2, 04]) than in the NS group 4 (3.8) mL (95% CI ϭ [2, 4] to [5, 6]) (P ϭ 0.017). There was no difference for the ulnar nerve between UG group 2 (0.1) mL (95% CI ϭ [1, 96] to [2, 04]) and NS group 2.4 (0.6) mL (95% CI ϭ [2, 1] to [2, 7]). The duration of sensory blockade was significantly correlated to local anesthetic volume, but onset time was not modified. CONCLUSION: Ultrasound guidance selectively provided a 50% reduction in the MEAV of mepivacaine 1.5% for median nerve sensory blockade in comparison with neurostimulation. Decreasing the local anesthetic volume can decrease sensory block duration but not onset time. (Anesth Analg 2010;111:1059–64)

erve stimulation is an indirect technique of nerve vessels, muscles, and needle movements and allows the identification but is still one of the most popular volume distribution of local anesthetic to be controlled.4 Ntechniques for peripheral nerve blocks. The suc- The local anesthetic volume injected near a nerve is a factor cess rate is 91% to 98%, depending on the trials. Ultrasound determining the rate of successful nerve block.5,6 The guidance may be of benefit for peripheral nerve blocks.1–3 volume and concentration affect the absorption of local anes- Ultrasound guidance permits a dynamic vision of nerves, thetic.7–9 The possibility of decreasing local anesthetic volumes for peripheral nerve blocks is a relevant question. Several studies have reported that multiple neurostimulations for locating nerves permitted reduction of the local anesthetic From the *Department of Anesthesiology and Critical Care, Montpellier 10–16 6 University Hospital, Montpellier, France; †Department of Anesthesiology volume. The study of Casati et al. and the systematic and Critical Care Medicine, Lapeyronie University Hospital, and Epidemi- review of Koscielniak-Nielsen2 found that decreasing the local ology and Clinical Research Department, Arnaud de Villeneuve University Hospital Montpellier, France; and ‡Department of Anesthesiology and anesthetic volume seems possible using ultrasound guidance. Critical Care, Montpellier I University and Montpellier University Hospital; O’Donnell and Iohom17 recently reported in a descriptive Institut National de la Sante et de la Recherche Me´dicale, Montpellier, France. study that 1 mL of 2% lidocaine on each nerve component Accepted for publication June 1, 2010. during an ultrasound-guided axillary block was sufficient to The study has been presented in part at the American Society of Anesthe- promote complete anesthetic block. However, we have only siologists meeting, New Orleans, Louisiana, October 17–21, 2009. sparse data supporting the consequences of such low volumes Address correspondence to Prof. Xavier Capdevila, Department of Anesthe- on the pharmacodynamic parameters of sensory blockade.18 siology, Lapeyronie University Hospital, Route de Ganges, 34295 Montpel- lier Cedex 5, France. Address e-mail to [email protected]. Furthermore, some authors have claimed that it is man- Copyright © 2010 International Anesthesia Research Society datory to develop studies comparing neurostimulation 2,6,19–21 DOI: 10.1213/ANE.0b013e3181eb6372 with ultrasound.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1059 Minimum Effective Anesthetic Volumes for Median and Ulnar Nerve Blocks

Therefore, we conducted a prospective, randomized, mA, and stimulation frequency 2 Hz. Nerve blocks were double-blind controlled study to determine the minimum achieved as has been previously demonstrated5 at the effective anesthetic volume (MEAV) necessary to achieve humeral canal by a palmar flexion of the first 3 fingers and median and ulnar nerve blocks, by using neurostimulation a carpal pronation for the median nerve, and a flexion of or ultrasound guidance. We tested the hypothesis that the fifth and fourth fingers, and a thumb adduction and ultrasound guidance reduces MEAV by 20%. The second- flexion of carpi ulnaris for the ulnar nerve. The procedure ary end point provides information on the influence of the started at the median nerve. The puncture site was located local anesthetic volume on the pharmacodynamic charac- at the humeral canal immediately above the brachial artery. teristics of the nerve block. The needle was inserted tangentially to the skin until minimum stimulus intensity between 0.4 and 0.6 mA was METHODS achieved. The starting dose of 1.5% mepivacaine was 13 After obtaining ethics committee approval (Comite´de and 11 mL for median and ulnar nerves. The predefined protection des personnes Sud Me´dite´ranne´e 3) and written local anesthetic volume was injected after an aspiration test informed consent, ASA physical status I–III patients ages 18 repeated between each bolus of 2 mL, until the final to 90 years scheduled to undergo ambulatory endoscopic or volume. The needle was withdrawn without leaving the open pit carpal tunnel release surgery were recruited for skin, and the intensity of stimulation was increased again to this randomized controlled study. Patients who did not 1.5 mA. The needle was redirected medially and posteriorly cooperate and those who had psychological disorders or and the same procedure was performed to locate and block linguistic difficulties that might interfere with sensory the ulnar nerve separately. blockade were excluded. Medical exclusion criteria were In the UG group, localization of both nerves was per- coagulopathies, known allergy to the trial drugs, infection formed with a 50-mm 22-G needle (Uniplex nanoLine Ͼ Ͻ 2 at the puncture site, a body mass index 40 or 19 kg/m , Facet௡, Pajunk©, Germany) and an ultrasound machine diabetes mellitus or known neuropathies, patients who (Logic E௡, GE Healthcare©) with a linear probe set at a received opiates for chronic pain, and cardiac conduction frequency of 12 MHz. After analysis of different anatomical problems (third-degree atrioventricular block). elements, the probe was positioned perpendicularly to the Patients were included in the ultrasound guidance (UG) skin to obtain a cross-section of the humeral canal. The group or neurostimulation (NS) group using a random list nerves were visualized in their short axis. The needle was at the preanesthetic consultation. On the day of surgery, inserted at the lateral end of the probe to keep it in the patients were premedicated with 1 mg/kg hydroxyzine, plane of the sonogram. The needle bevel and shaft were and 500 mL of saline at the rate of 4 to 6 mL/kg/h was viewed throughout the approach to the selected nerve. The infused with an IV 20-G catheter on the contralateral predefined local anesthetic volume was injected after an forearm. Patients were monitored using a noninvasive aspiration test repeated between each bolus of 2 mL, until arterial blood pressure measurement, a continuous electro- the final volume. The injection was slow and at low cardiogram, and pulse oximetry. A high-concentration oxy- pressure. The absence of intraneural injection was con- gen mask at 6 l/minute was put on the patient’s face. The trolled by ultrasound. After a unique needle puncture, patients were sedated with 0.05 ␮g/kg IV sufentanil. The needle repositioning was allowed to optimize the distribu- operating arm was positioned at 80° abduction and external tion of local anesthetic around each nerve. A circumferen- rotation. The blocks of the median and ulnar nerves were tial spread was required without exceeding the defined done at the junction between the upper and middle thirds volume. For both groups the volume of 1 mL was consid- of the arm (i.e., humeral canal) with a 15 mg/mL mepivacaine solution. All blocks were always placed by 1 of the ered the lowest possible volume. same 2 investigators (Miche`le Kassim or Matthieu At the end of the last nerve injection, the sensory Ponrouch), who had substantial expertise in regional anes- blockade was tested every 2 minutes for 20 minutes by an thesia techniques. Patients were blinded to the technique at observer blinded to the technique and the volume injected. the beginning of the procedure by the passage of the The sensory block was assessed by the patient’s ability to ultrasound probe in the NS group and a single stimulation distinguish hot and cold and to discriminate a light touch in at 1 mA of the median nerve in the UG group. In the 2 the center of the skin area innervated by each nerve: the groups the patient could not see the screen of the ultra- thenar eminence and the anterior surface of the distal end sound device. The anesthesiologist in charge of the block of the index finger for the median nerve, and the hypothe- procedure turned off the ultrasound machine in the NS nar eminence and the anterior surface of the distal end of group. The ultrasound probe was placed at the beginning the fifth finger for the ulnar nerve. A comparison with the of the nerve stimulation procedure. A conventional aseptic contralateral area was used to evaluate sensory blockade. A procedure was used for peripheral nerve blocks. The anes- value of 0 was noted if the sensation was the same on both thesiologist wore a mask, cap, and gloves. The puncture sides; 1 in case of decreased sensation in the anesthetized site was prepared with an alcohol povidone–iodine solu- hand and 2 in case of no sensation (complete sensory tion, and surrounding areas were disinfected. The probe block). When the sum of the variables was equal to 4 within was covered with a film-type sterile Tegaderm®. 20 minutes, the block was defined as complete. A sum Ͻ4 In the NS group, nerve location was made with a 50-mm was considered a failure. Adverse events (i.e., paresthesia, 22-G needle (Uniplex nanoLine Facet௡, Pajunk©, Germany) pain during injection, intravascular injection, and cardio- and a nerve stimulator (MultiStim Sensor௡, Pajunk©, Ger- vascular and neurologic events) were noted during the many) initially set at pulse duration 0.1 ms, intensity 1.5 procedure and until the end of the sensory block. The

1060 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Table 1. Patient Characteristics of the Neurostimulation (NS) and Ultrasound Guidance (42 ؍ UG) Groups (n) NS group UG group (21 ؍ n) (21 ؍ n) Age (years) 56 (17) 55 (17) Weight (kg) 70 (22.1) 67.5 (20.2) Height (cm) 164 (36) 167 (36) BMI (kg/m2) 26 (6.9) 25.1 (6.9) Sex (male/female) 8/13 6/15 ASA physical status I/II/III 11/10/0 9/11/1

Data are presented as mean Ϯ SD. P value Ͼ0.05. ASA ϭ American Society of Anesthesiology; BMI ϭ body mass index. There were no signiﬁcant differences between groups.

stimulator was 11 mL and 13 mL for the ulnar and median nerves.5 We used these values as the initial volumes. A difference of 2 mL in this volume in the UG group was considered clinically significant. The Dixon method was 22 Figure 1. Flowchart summarizing the design of both parts of the used. The MEAV50 corresponds to the MEAV that suc- study. A, Comparison of the MEAV measured by the Dixon method for cessfully anesthetized 50% of the patients. Accepting an ␣ both selected nerves in each group. B, Pharmacodynamics of ␤ successful nerve blocks (onset time and duration of blocks) related risk of 5% and a power (1– ) of 80%, 12 subjects in each to local anesthetic volume. MEAV ϭ minimum effective anesthetic group were necessary for a significant difference. It has volume for 50% of the patients; BMI ϭ body mass index; UG ϭ been demonstrated that beyond 11 patients, the signifi- ultrasound guidance; NS ϭ nerve stimulation. cance of the results increases with the number of patients included.22 Considering these factors, the number of patients sensory blockade was assessed on arrival in the postanes- enrolled was arbitrarily set at 42. Patients were divided into 2 thetic care unit, then every 15 minutes until the end of the groups, 21 in the NS group and 21 in the UG group. The block (score returned to 0). distribution was made according to a computer-randomized list in clusters of 6 patients. Statistical analysis was per- The definition of sensory onset time was the time from ௡ performance of the block to a value of 4 in sensory blockade formed using SAS version 9 software (SAS Institute, Cary, in the selected areas, and the definition of duration of North Carolina). Comparison between qualitative variables ␹2 sensory blockade was the time from performance of the was performed by test or Fisher’s exact test, if necessary. block to a value of 0 for the sensory block. Motor blockade Comparisons of quantitative variables were performed was not noted because it was not necessary for surgery, and using the Student’s t test or Mann–Whitney Wilcoxon test, we wanted to avoid interfering with the blinded evaluation if necessary. The relationship between quantitative vari- of the sensory blockade. If the block was ineffective, the ables was evaluated using Spearman’s correlation coeffi- surgeon performed a wrist infiltration with 8 mL of 15 cients. Correlation of the volume of 1.5% mepivacaine in mg/mL mepivacaine. In case of failure, the patient was relation to the sensory onset time and duration of sensory Ͻ anesthetized with sufentanil 0.2 ␮g/kg and propofol in a blockade was done by linear regression. P 0.05 was target-controlled infusion, and a laryngeal mask airway considered significant. was inserted. Postoperative analgesia was provided by 1 g paracetamol every 6 hours for 48 hours. Patients were RESULTS discharged home in the evening of the day of surgery. Forty-two patients were included in the study, 21 in the NS The primary hypothesis was to determine the MEAV group and 21 in the UG group. There were no significant necessary to achieve median and ulnar nerve blocks using differences in the patients’ characteristics in both groups neurostimulation or ultrasound guidance. We tested the (Table 1). The MEAV50 (SD) of the median nerve was lower ϭ hypothesis that ultrasound guidance reduces MEAV by in the UG group 2 (0.1) mL (95% confidence interval [CI] ϭ 20%. We compared the MEAV measured by the Dixon [1, 96] to [2, 04]) than in NS group 4 (3.8) mL (95% CI ϭ method for both selected nerves in each group. Secondary [2, 4] to [5, 6]) (P 0.017). There was no difference for the ϭ end points of the study examined the pharmacodynamic ulnar nerve between UG group 2 (0.1) mL (95% CI [1, 96] ϭ characteristics of all successful nerve blocks (onset time and to [2, 04]) and NS group 2.4 (0.6) mL (95% CI [2, 1] to duration of blocks) related to the local anesthetic volume [2, 7]) (Fig. 2). With both nerve localization techniques, injected. A flowchart summarizing the design of both parts there was a positive significant correlation between dura- of the study is presented in Figure 1. tion of sensory blockade and the volume of mepivacaine 1.5% injected for both ulnar and median nerves. For both Statistics nerves, the duration of sensory block decreased when less The primary end point chosen for the 2 techniques was a local anesthetic was injected. In some patients a volume of 20% decrease in the MEAV in the UG group. The descrip- 1 mL allowed a sensory blockade for Ͻ60 minutes. Con- tive study of Carles et al. reported that the mean volume versely, onset times of sensory blockade of the median and required to block the ulnar nerve successfully using a nerve ulnar nerves were not significantly correlated to the local

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1061 Minimum Effective Anesthetic Volumes for Median and Ulnar Nerve Blocks

Figure 2. Minimum effective anesthetic volume

(MEAV)50 measured by the Dixon up-and-down method for the median (A) and ulnar (B) nerves. A, P value ϭ 0.017. B, P value ϭ 0.441. The volume

X0 was chosen at 13 mL for the median nerve and 11 mL for the ulnar nerve. The interval between volumes in 2 different patients was set at 2 mL.

The next volume X1 was determined by the performance of the previous volume, X0.IfX0 was ϭ effective, X1 X0 – 2 mL; if X0 was ineffective, ϭ ϩ ϭ X1 X0 2 mL. UG ultrasound guidance; NS ϭ nerve stimulation.

Figure 3. A, B, Signiﬁcant correlations between local anesthetic volume and duration of sensory blockade for median and ulnar nerves (*P ϭ 0.03). C, D, No correlation between local anesthetic volume and onset time for median and ulnar nerves (P ϭ 0.71).

anesthetic volume injected (Fig. 3). For 1 patient in the UG general anesthesia was required. No adverse events were group, an unplanned subcutaneous infiltration of local noted in either group. anesthetic in the musculocutaneous nerve skin area was necessary for surgical incision. Fifteen blocks showed a DISCUSSION negative response 20 minutes after block placement. After We report that ultrasound guidance can reach an MEAV50 recording the presence of a negative response to the value lower than neurostimulation for the median nerve up-and-down sequence, these patients received a wrist but not for the ulnar nerve. In addition, the reduction in infiltration with 8 mL of 15 mg/mL mepivacaine. No local anesthetic volume caused a decrease in the duration of

1062 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA sensory blockade but did not modify the onset time of the area had no effect on sensory onset time, whereas the block for both nerves. Ultrasound guidance can reduce the duration of sensory block was shorter. Marhofer et al.23 local anesthetic volume for a selected nerve block in reported that ultrasound can reduce the local anesthetic comparison with neurostimulation. Our results confirm volume and onset time of the femoral sensory blockade. 1–3,6,21,23 data reported in those previous studies. It is inter- However, the authors did not report significant correlations esting to note that if we apply the same methodology for between the local anesthetic volume and onset time of the the calculation of MEAV for both guidance methods, block. Onset time and quality of the blocks also depend on neurostimulation allows the same volume reduction as the nerve and approach.28,29 does ultrasound guidance for some isolated nerves. For the The spread of local anesthetic in contact with the nerve ulnar nerve, the results show that the MEAV50 is compa- is crucial for obtaining the MEAV. It is conceptualized as rable for both techniques of nerve location. For the femoral 6 the minimum circumferential area of nerve exposed to nerve, Casati et al. reported that the MEAV50 was signifi- MEAV associated with the surface of the nerve. Eichen- cantly lower in an UG group (15 mL) than in a NS group (26 21 berger et al. studied the concept of MEAV (in milliliters mL). The same authors reported in another study that the per millimeter) on the basis of the diameter of the nerve. MEAV using neurostimulation was significantly de- 50 They reported a MEAV of 0.11 mL/mm for mepivacaine creased in a multiple-neurostimulations and multiple- 1% for ulnar nerve block at the proximal forearm. This is injections group (14 mL) than in a single-neurostimulation interesting but does not predict MEAV values in other parts and single-injection group (23 mL) for a femoral nerve block.24 In other words, a precise identification of nerves of the same nerve. The variability of the structure of a single nerve from the brachial plexus to its distal part explains the reduces MEAV50. Ultrasound guidance permits direct vi- 30 sualization of the spread of local anesthetic around and in different expected MEAV values. Moayeri et al. reported contact with the nerve.20 from a cadaveric study that the nonneural tissue composi- The location of the median nerve and its anatomical tion of the nerves within the epineurium increased from relationship into the humeral canal may explain the differ- their proximal to distal parts. The neuronal tissue/ nonneural tissue ratio increases from 1:1 to 1:2. These ence in the MEAV50 in the UG group in comparison with the NS group. The median nerve is close to the brachial studies demonstrated that for the same dose of local artery.1,25,26 The circumferential distribution of a small anesthetic (volume and concentration), the onset time de- local anesthetic volume around the median nerve is avail- pends on the quality of nerve localization using ultrasound able with ultrasound guidance through needle redirections guidance or neurostimulation, the nerve studied, and the and visualization of the local anesthetic spread. The inti- local anesthetic spread close to the nerve.25,26 mate relationship of the median nerve and the brachial Some limitations of our study deserve comment. The artery, by using neurostimulation, suggests that the circum- clinical relevance of a MEAV50 value might be questioned, ferential spread around the nerve seems unpredictable and because in the method used, about 50% of the studied difficult to obtain with a low volume of local anesthetic. patients had an incomplete nerve block. Nonetheless, the

Conversely, the ulnar nerve is located farther from the ED50 concept has been used to determine volume–effect artery, allowing that needle movements are not always relationships in the field of peripheral nerve blocks.6,22,24 necessary to obtain a circumferential distribution of local On the other hand, we reported a correlation between 25,26 anesthetic around the ulnar nerve. This may explain volume of local anesthetic solution and duration of the why there was no difference in the MEAV50 for the ulnar blocks. However, the study was not designed for a com- nerve in both guidance groups. Our randomized compara- parison in block duration for fixed high or low volumes of tive study reports that, depending on the nerves and local anesthetics. Further comparative randomized studies surrounding anatomical structure, ultrasound guidance are necessary to determine whether the volume of local allowed a smaller volume than did neurostimulation. The anesthetic solution really influences the duration of nerve MEAV obtained was lower by 50% in the UG group. These blocks. results confirm the results in the Danelli et al. study.27 This double-blind, randomized, clinical trial shows These authors compared neurostimulation and ultrasound that the use of ultrasound guidance was effective in guidance for sciatic nerve blocks using a subgluteal ap- reducing the local anesthetic volume for the median proach. They reported a 37% decrease of MEAV in the UG nerve. Neurostimulation can provide similar results for group. The decrease in volumes of local anesthetic leads to the ulnar nerve. The ability to visualize anatomical struc- changes in some of the nerve block pharmacodynamic tures, to control the position of the needletip, and to control characteristics. When the volume decreases, the onset time the circumferential spread of local anesthetic may partly of the sensory blockade does not vary. Conversely, the explain these results. The decrease in volume helps to duration of nerve sensory blockade is closely correlated to highlight a strong correlation between local anesthetic the local anesthetic volume. Serradell et al., using 20 to 36 volume and duration of the nerve block; the onset time mL of local anesthetic, did not report a difference in the does not vary with the local anesthetic volume. The precise duration of the axillary block, corresponding to the right control of these variables and knowledge of variations of side of the correlation curve in Figure 3.13 In a recent study pharmacodynamic characteristics based on the local anes- of 20 volunteers scheduled for sciatic nerve block with thetic volume injected, concentration, and approach of the ultrasound guidance, Latzke et al.18 reported that a local block should assist practitioners to adapt the procedure to anesthetic volume of 0.10 mL/mm cross-sectional nerve different clinical situations.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1063 Minimum Effective Anesthetic Volumes for Median and Ulnar Nerve Blocks

REFERENCES 17. O’Donnell BD, Iohom G. An estimation of the minimum 1. Abrahams MS, Aziz MF, Fu RF, Horn JL. Ultrasound guidance effective anesthetic volume of 2% lidocaine in ultrasound- compared with electrical neurostimulation for peripheral guided axillary brachial plexus block. Anesthesiology 2009; nerve block: a systematic review and meta-analysis of random- 111:25–9 ized controlled trials. Br J Anaesth 2009;102:408–17 18. Latzke D, Marhofer P, Zeitlinger M, Machata A, Neumann F, 2. Koscielniak-Nielsen ZJ. Ultrasound-guided peripheral nerve Lackner E, Kettner SC. Minimal local anaesthetic volumes for blocks: what are the benefits? Acta Anaesthesiol Scand 2008; sciatic nerve block: evaluation of ED 99 in volunteers. Br J 52:727–37 Anaesth 2009;104:239–44 3. Marhofer P, Chan VW. Ultrasound-guided regional anesthesia: 19. Macaire P, Singelyn F, Narchi P, Paqueron X. Ultrasound- or current concepts and future trends. Anesth Analg 2007;104: nerve stimulation-guided wrist blocks for carpal tunnel re- 1265–9 lease: a randomized prospective comparative study. Reg 4. Retzl G, Kapral S, Greher M, Mauritz W. Ultrasonographic Anesth Pain Med 2008;33:363–8 findings of the axillary part of the brachial plexus. Anesth 20. Hadzic A, Dewaele S, Gandhi K, Santos A. Volume and dose of Analg 2001;92:1271–5 local anesthetic necessary to block the axillary brachial plexus 5. Carles M, Pulcini A, Macchi P, Duflos P, Raucoules-Aime M, using ultrasound guidance. Anesthesiology 2009;111:8–9 Grimaud D. An evaluation of the brachial plexus block at the 21. Eichenberger U, Stockli S, Marhofer P, Huber G, Willimann P, humeral canal using a neurostimulator (1417 patients): the Kettner SC, Pleiner J, Curatolo M, Kapral S. Minimal local efficacy, safety, and predictive criteria of failure. Anesth Analg anesthetic volume for peripheral nerve block: a new 2001;92:194–8 6. Casati A, Baciarello M, Di Cianni S, Danelli G, De Marco G, ultrasound-guided, nerve dimension-based method. Reg Leone S, Rossi M, Fanelli G. Effects of ultrasound guidance on Anesth Pain Med 2009;34:242–6 the minimum effective anaesthetic volume required to block 22. Dixon WJ. Staircase bioassay: the up-and-down method. Neu- the femoral nerve. Br J Anaesth 2007;98:823–7 rosci Biobehav Rev 1991;15:47–50 7. Brown DL, Ransom DM, Hall JA, Leicht CH, Schroeder DR, 23. Marhofer P, Schrogendorfer K, Koinig H, Kapral S, Weinstabl Offord KP. Regional anesthesia and local anesthetic-induced C, Mayer N. Ultrasonographic guidance improves sensory systemic toxicity: seizure frequency and accompanying cardio- block and onset time of three-in-one blocks. Anesth Analg vascular changes. Anesth Analg 1995;81:321–8 1997;85:854–7 8. Becker DE, Reed KL. Essentials of local anesthetic pharmacol- 24. Casati A, Fanelli G, Beccaria P, Magistris L, Albertin A, Torri G. ogy. Anesth Prog 2006;53:98–108 The effects of single or multiple injections on the volume of 9. Mather LE, Copeland SE, Ladd LA. Acute toxicity of local 0.5% ropivacaine required for femoral nerve blockade. Anesth anesthetics: underlying pharmacokinetic and pharmacody- Analg 2001;93:183–6 namic concepts. Reg Anesth Pain Med 2005;30:553–66 25. Sites BD, Neal JM, Chan V. Ultrasound in regional anesthesia: 10. Casati A, Danelli G, Baciarello M, Corradi M, Leone S, Di where should the “focus” be set? Reg Anesth Pain Med Cianni S, Fanelli G. A prospective, randomized comparison 2009;34:531–3 between ultrasound and nerve stimulation guidance for mul- 26. van Geffen GJ, Moayeri N, Bruhn J, Scheffer GJ, Chan VW, tiple injection axillary brachial plexus block. Anesthesiology Groen GJ. Correlation between ultrasound imaging, cross- 2007;106:992–6 sectional anatomy, and histology of the brachial plexus: a 11. Sia S, Bartoli M, Lepri A, Marchini O, Ponsecchi P. Multiple- review. Reg Anesth Pain Med 2009;34:490–7 injection axillary brachial plexus block: a comparison of two 27. Danelli G, Ghisi D, Fanelli A, Ortu A, Moschini E, Berti M, methods of nerve localization–nerve stimulation versus pares- Ziegler S, Fanelli G. The effects of ultrasound guidance and thesia. Anesth Analg 2000;91:647–51 neurostimulation on the minimum effective anesthetic volume 12. Sia S, Lepri A, Ponzecchi P. Axillary brachial plexus block of mepivacaine 1.5% required to block the sciatic nerve using using peripheral nerve stimulator: a comparison between the subgluteal approach. Anesth Analg 2009;109:1674–8 double- and triple-injection techniques. Reg Anesth Pain Med 28. Taboada Muniz M, Rodriguez J, Bermudez M, Valino C, 2001;26:499–503 Blanco N, Amor M, Aguirre P, Masid A, Cortes J, Alvarez J, 13. Serradell A, Herrero R, Villanueva JA, Santos JA, Moncho JM, Atanassoff PG. Low volume and high concentration of local Masdeu J. Comparison of three different volumes of mepivacaine in axillary plexus block using multiple nerve stimulation. anesthetic is more efficacious than high volume and low Br J Anaesth 2003;91:519–24 concentration in Labat’s sciatic nerve block: a prospective, 14. Inberg P, Annila I, Annila P. Double-injection method using randomized comparison. Anesth Analg 2008;107:2085–8 peripheral nerve stimulator is superior to single injection in 29. Casati A, Fanelli G, Borghi B, Torri G. Ropivacaine or 2% axillary plexus block. Reg Anesth Pain Med 1999;24:509–13 mepivacaine for lower limb peripheral nerve blocks. Study 15. Deleuze A, Gentili ME, Marret E, Lamonerie L, Bonnet F. A group on orthopedic anesthesia of the Italian Society of Anes- comparison of a single-stimulation lateral infraclavicular thesia, Analgesia, and Intensive Care. Anesthesiology 1999; plexus block with a triple-stimulation axillary block. Reg 90:1047–52 Anesth Pain Med 2003;28:89–94 30. Moayeri N, Bigeleisen PE, Groen GJ. Quantitative architecture 16. Benhamou D. Axillary plexus block using multiple nerve of the brachial plexus and surrounding compartments, and stimulation: a European view. Reg Anesth Pain Med their possible significance for plexus blocks. Anesthesiology 2001;26:495–8 2008;108:299–304

1064 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA BRIEF REPORT Pulse-Oximetric Measurement of Prilocaine-Induced Methemoglobinemia in Regional Anesthesia

Peter Soeding, MD,* Matthias Deppe,* and Hartmut Gehring, MD, PhD*

BACKGROUND: The Masimo Radical 7® is a new pulse CO oximeter designed to measure methemoglobin. The device has not been evaluated in a clinical setting. METHODS: In this prospective observational study we compared the arterial methemoglobin levels and the corresponding pulse CO-oximetric values of the Radical 7® in regional anesthesia with prilocaine. RESULTS: We analyzed 360 data pairs with methemoglobin values up to 6.6%. The mean bias and limits (Ϯ1.96 SD) of the device were 0.27% (Ϯ1.33%). CONCLUSION: We found a high degree of agreement in measurement of methemoglobin between the 2 methods. (Anesth Analg 2010;111:1065–8)

rilocaine, in comparison with all other local anesthet- MetHb by pulse oximetry (SpMet%).10 These values could ics, has the lowest direct systemic toxicity, but may lead be read using a laptop and stored after conversion into an to an increased formation of methemoglobin Excel spreadsheet. The continuous monitoring of patients P 1,2 ® (MetHb). Whereas in healthy individuals higher concentra- with the Radical 7 was started before the onset of regional tions of MetHb are usually well tolerated, it may endanger anesthesia. oxygen supply in patients with diminished cardiopulmonary At the times 0, 15, 30, 60, 120, 180, 240, 300, and 360 reserves or anemia.3–6 Because of their 2-wavelength technol- minutes after the first injection of prilocaine, 2 mL of ogy, conventional pulse oximeters are not able to identify the arterial blood was taken and immediately analyzed in a extent of dyshemoglobinemias.7–9 The optical analysis of a blood gas analyzer (ABL-625, Radiometer America, Copen- new pulse CO oximeter (Masimo, Radical 7®)) is based on hagen, Denmark) with an integrated CO oximeter as a absorbance measurements at several wavelengths. Conse- reference device. The data for MetHb content (cMetHb) and quently, dyshemoglobinemias might also be recorded by a for Sao2 were compared with the corresponding values pulse oximeter with sufficient precision. This has been dem- obtained with the Radical 7®. onstrated in 1 preclinical trial10 and 2 case reports.11,12 For statistical analysis using the Software Package for This prospective study is the first to evaluate the efficacy Social Sciences (SPSS) 15.0 for Windows, we averaged the 9 of this pulse CO oximeter in a clinical setting. We hypoth- repeated measurements for each of the 40 patients accord- esized that the results for MetHb are the same for the pulse ing to Bland and Altman.13 oximetric method and arterial blood sample measurement A probability of P Ͻ 0.05 was considered statistically using a CO oximeter. significant.

METHODS RESULTS After approval by the ethics committee and written One patient after interscalene block and 1 patient after informed consent, we investigated 40 patients, physical combined femoral–sciatic nerve block required general status ASA I–III, having orthopedic surgery: 20 patients received an interscalene plexus block with 30 mL prilocaine 1% (i.e., 300 mg) and 20 patients a combined femoral–sciatic nerve blockade with 2 ϫ 30 mL prilocaine 1% (i.e., 600 mg in all). All blocks were performed using a nerve stimulator (Stimuplex HNS 11, Braun, Germany). Injection was only performed if a contraction of indicator muscles could be demonstrated at 0.3 to 0.5 mA (stimulus duration: 0.1 ms). The pulse CO oximeter Radical 7® is a device manufac- tured by Masimo Corp. (Irvine, California), which in addition to oxygen saturation (Spo2 in %) can also measure

From the *Department of Anesthesiology, University Clinic of Schleswig— Holstein, Campus Luebeck, Luebeck, Germany. Accepted for publication June 1, 2010. Parts of the work were presented at the 2008 American Society of Anesthe- siologists annual meeting in Baltimore, Maryland, and at the 2008 European Society of Anesthesiologists annual meeting in Copenhagen, Denmark. Address correspondence to Peter Soeding, MD, Department of Anesthesiology, University Clinic of Schleswig—Holstein, Campus Luebeck, Ratzeburger Allee Figure 1. Methemoglobin (MetHb) levels after regional anesthesia 160, D-23,538 Luebeck, Germany. Address e-mail to [email protected]. with prilocaine: plots show mean MetHb levels after 300 mg Copyright © 2010 International Anesthesia Research Society prilocaine for interscalene block (ISB) and after 600 mg prilocaine for DOI: 10.1213/ANE.0b013e3181eb6239 combined femoral–sciatic nerve block (FSNB).

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1065 BRIEF REPORT

Table 1. Methemoglobin Levels and Success Rate After Regional Anesthesia with Prilocaine for Interscalene or Femoral–Sciatic Nerve Blocks Methemoglobin

Peak mean ؎ SD Nerve blockade No. of patients Prilocaine Success rate (range) Time to peak Interscalene 20 300 mg (30 mL) 95% (19/20) 2.3 Ϯ 0.8% (1.1–4.9) 120 minutes Femoral–sciatic 20 600 mg (2 ϫ 30 mL) 95% (19/20) 4.1 Ϯ 1.5% (2.0–6.6) 300 minutes

Table 2. Interclass Statistical Comparison of SpMet (%) Versus cMetHb (%) for Each Subject and for Pooled Data Regression analysis Bland–Altman n Slope y intercept SEE Bias Limits (؎1.96 SD) r Pooled data 360 1.19 Ϫ0.14 0.61 0.27 1.33 0.95 Subject 1 9 1.09 Ϫ0.35 0.21 Ϫ0.12 0.47 0.99 2 9 1.12 Ϫ0.32 0.21 Ϫ0.01 0.48 0.98 3 9 0.84 0.36 0.41 0.13 0.79 0.84 4 9 1.30 Ϫ0.86 0.27 Ϫ0.42 0.59 0.94 5 9 1.05 Ϫ0.004 0.54 0.21 1.05 0.98 6 9 0.79 0.61 0.68 0.37 1.28 0.4 7 9 1.34 Ϫ0.68 0.15 Ϫ0.08 0.64 0.99 8 9 1.21 Ϫ0.04 0.18 0.2 0.39 0.96 9 9 1.24 Ϫ0.17 0.10 0.04 0.23 0.96 10 9 1.20 0.53 0.82 1.16 1.76 0.96 11 9 1.27 Ϫ1.02 0.96 0.04 2.18 0.95 12 9 1.26 Ϫ0.64 0.52 Ϫ0.08 1.09 0.93 13 9 1.17 0.24 1.04 0.74 2.05 0.92 14 9 1.19 Ϫ0.40 0.24 0.13 0.68 0.99 15 9 1.11 Ϫ0.21 0.07 Ϫ0.01 0.23 0.99 16 9 1.16 Ϫ0.31 0.53 0.2 1.15 0.98 17 9 1.18 Ϫ0.34 0.13 Ϫ0.12 0.3 0.98 18 9 1.21 Ϫ0.44 0.47 0.11 1.07 0.97 19 9 0.61 1.27 0.72 0.61 1.43 0.51 20 9 1.38 Ϫ0.39 0.75 0.89 2.09 0.97 21 9 1.15 0.78 0.94 1.34 1.93 0.96 22 9 1.43 Ϫ0.08 0.34 0.77 0.9 0.96 23 9 1.09 Ϫ0.34 0.16 Ϫ0.18 0.36 0.99 24 9 0.88 0.25 0.38 0.06 0.73 0.82 25 9 1.26 Ϫ0.49 0.37 0.1 0.92 0.97 26 9 1.18 Ϫ0.12 0.19 0.36 0.66 0.99 27 9 1.81 Ϫ0.73 0.37 0.4 1.15 0.95 28 9 0.89 0.46 0.94 0.31 1.76 0.44 29 9 0.59 0.23 0.19 Ϫ0.12 0.4 0.57 30 9 0.84 0.70 0.44 0.4 0.87 0.85 31 9 0.65 0.87 0.94 0.38 1.79 0.28 32 9 1.19 Ϫ0.37 0.20 Ϫ0.04 0.46 0.97 33 9 1.12 0.18 0.31 0.38 0.61 0.93 34 9 1.20 0.03 0.41 0.52 0.98 0.98 35 9 1.21 Ϫ0.45 0.41 0.16 1.03 0.98 36 9 1.03 Ϫ0.05 0.44 0 0.82 0.90 37 9 1.22 Ϫ0.28 0.24 0.14 0.57 0.98 38 9 1.46 0.13 0.77 1.26 1.77 0.92 39 9 0.62 0.74 0.42 0.02 0.95 0.75 40 9 1.26 Ϫ0.40 0.38 0.49 1.37 0.99 SpMet ϭ Radical 7® measurement of methemoglobin; cMetHb% ϭ CO oximeter measurement of methemoglobin; n ϭ number of data points for each subject; SEE ϭ SE of the estimate; r ϭ correlation coefﬁcient. anesthesia for block failure (success rate 95%). Figure 1 7® for all 40 patients is shown in Table 2 and Figure 2. shows MetHb levels over time for both types of blocks. According to Bland–Altman analysis (Fig. 2), the bias Peak levels were reached for interscalene blocks after 120 was 0.27%, and the 95% confidence limits (Ϯ1.96 sd) minutes and for combined femoral–sciatic nerve blocks 1.33%. With the increasing rise in MetHb, there is a clear after 5 hours (Table 1). gap between the values reported for functional oxygen The statistical agreement of the MetHb measurement saturation by the Radical 7® and the values reported by between the laboratory method as a reference method the CO oximeter in the blood gas analyzer device (Fig. 3). ® and the pulse oximetric measurement using the Radical The Radical 7 displays Spo2 readings that run in parallel

1066 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Pulse-Oximetric Measurement of Methemoglobin

Figure 4. Regression analysis: scatter plot of the functional satura- Figure 2. Bland and Altman analysis for repeated measurements: tion measured by Radical 7® (SPO ) and the fractional saturation bias plot of the difference of pulse-oximeter estimate of methe- 2 measured by the CO oximeter (O2Hb) versus cMetHb%. Averaged moglobin (MetHb) (SpMet [%]) and cMetHb% versus the average data of the 9 repeated measurements of each of the 40 patients. of SpMet and cMetHb. Averaged data of the 9 repeated measure- cMetHb% ϭ CO oximeter measurement of methemoglobin. SEE ϭ SE ments. Lines show values of bias (mean of the differences) of the estimate and Ϯ1.96 SD. cMetHb% ϭ CO-oximeter measurement of methemoglobin. In comparison with lidocaine and mepivacaine, prilocaine has the advantage of far less cardiac or central nervous system toxicity.14,15 Nevertheless, prilocaine is not available in much of the world. The presence of acquired methemoglobinemia is often assumed.3,16,17 It is triggered not only by prilocaine but also by many other drugs,3,18–21 in particular by benzocaine.17 The symptoms are nonspecific and often unrecognized.3,9 Methemoglobinemia is associated with the reduction in the fractional oxygen saturation. Concentrations Ͻ15% are usually well tolerated by healthy individuals, but in patients with anemia or cardiopulmonary diseases, clinical symptoms can occur when the concentration exceeds 8%.3 Despite administration of the recommended threshold dose of prilocaine, it was only possible to assess MetHb values up to 6.6%. Further investigations are necessary to analyze the accuracy of the pulse oximeter at higher levels. Dyshemoglobinemias cannot be identified by conven- Figure 3. Regression analysis: scatter plot of the functional satura- tional pulse oximeters, because their measurements, based ® tion measured by Radical 7 (SPO2) and the functional saturation on 2 wavelengths of light absorption, only allow the 7,22 measured by the CO oximeter (SAO2) versus cMetHb%. Averaged data recording of oxyhemoglobin and desoxyhemoglobin. of the 9 repeated measurements of each of the 40 patients. The pulse CO oximeter Radical 7® measures the light cMetHb% ϭ CO oximeter measurement of methemoglobin. SEE ϭ SE of the estimate absorption of 8 different wavelengths. In a preclinical study in healthy volunteers, Barker et al. evaluated a predecessor of the pulse oximeter that we used.10 The results (bias 0%, sd Ϯ 0.45%) differ only slightly from the data presented in with and slightly higher than the values for fractional our clinical study. saturation (Fig. 4). ® Although the Radical 7 , according to the manufacturer’s information, only displays the functional oxygen 23 DISCUSSION saturation, the data collected for the Spo2 display during In this prospective study we evaluated a new pulse Co the study period tend to follow the fractional rather than oximeter (Radical 7®, Masimo, Inc.) after regional anesthe- the functional oxygen saturation (Figs. 3 and 4). This sia with high doses of prilocaine. We compared the pulse difference might be a result of the analysis algorithm of the oximetric method for noninvasive monitoring of MetHb device. with direct arterial measurements by using a reference In conclusion, we found a high degree of agreement in procedure. A high degree of agreement between the 2 measurement of MetHb with a CO oximeter and a nonin- methods could be shown for MetHb values up to 6.6% vasive and readily available pulse-oximetric procedure. (mean correlation coefficient ϭ 0.95, bias ϭ 0.27, 95% This may facilitate early diagnosis and treatment, when limits ϭϮ1.33). necessary, of dyshemoglobinemia.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1067 BRIEF REPORT

DISCLOSURE 10. Barker SJ, Curry J, Redford D, Morgan S. Measurement of Financial support for the work: Peter Soeding and Hartmut carboxyhemoglobin and methemoglobin by pulse oximetry. Gehring are employees of the University Clinic of Schleswig— Anesthesiology 2006;105:892–7 11. Annabi EH, Barker SJ. Severe methemoglobinemia detected by Holstein, Campus Luebeck, Luebeck, Germany. Device was pulse oximetry. Anesth Analg 2009;108:898–9 provided by Masimo, Inc., Germany. 12. Macknet M, Kimball-Jones P, Applegate R. Benzocaine in- REFERENCES duced methemoglobinemia after TEE. Resp Care 2007;52:2007 Open Forum Abstracts [e-abstracts] 1. Scott DB, Owen JA, Richmond J. Methaemoglobinaemia due to 13. Bland JM, Altman DG. Agreement between methods of mea- prilocaine. Lancet 1964;3:728–9 surement with multiple observations per individual. J Biop- 2. Wright RO, Lewander WJ, Woolf AD. Methemoglobinemia: harm Stat 2007;17:571–82 etiology, pharmacology, and clinical management. Ann Emerg 14. Zink W, Graf BM. Toxikologie der Lokalana¨sthetika. Anaes- Med 1999;34: 646–56 thesist 2003;52:1102–23 3. Ash-Bernal R, Wise R, Wright SM. Acquired ¨ methemoglobinemia—A retrospective series of 138 cases at 2 15. Scott DB, Jerson. PJR, Braid DP, Ortengren B, Frisch P. Factors teaching hospitals. Med 2004;83:265–73 affecting plasma levels of lignocaine and prilocaine. Br J 4. Bellamy MC, Hopkins PM, Halsall PJ, Ellis FR. A study into Anaesth 1972;44:1040–9 incidence of methemoglobinaemia after “three-in-one” block 16. Weinberg GL. Banning benzocaine: of bananas, bureaucrats, with prilocaine. Anaesthesia 1992;47:1084–5 and blue men. Anesth Analg 2009;108:699–701 5. Knobeloch L, Goldring J, LeMay W, Anderson H. Three cases 17. Guay J. Methemoglobinemia related to local anesthetics: a of methemoglobinemia associated with dental anesthesia. Wis summary of 242 episodes. Anesth Analg 2009;108:837–45 Dent Assoc J 1994;70:34–5 18. Anderson ST, Hajduczek J, Barker SJ. Benzocaine-induced met- 6. Kreeftenberg HG, Braams R, Nauta P. Methemoglobinemia hemoglobinemia in an adult. Anesth Analg 1988;67:1099–101 after low-dose prilocaine in an adult patient receiving barbitu- 19. Rehman HU. Methemoglobinemia. West J Med 2001;175:193–6 rate comedication. Anesth Analg 2007;104:459–60 20. Fung HT, Lai CH, Wong OF, Lam KK, Kam CW. Two cases of 7. Reynolds KJ, Palayiwa E, Moyle JT, Sykes MK, Hahn CE. The methemoglobinemia following zoplicone ingestion. Clin Toxi- effect of dyshemoglobins on pulse oximetry: part I. theoretical col 2008;46:167–70 approach and part II. experimental results using an in vitro test 21. White CD, Weiss LD. Varying presentations of methemoglo- system. J Clin Monit 1993;9:81–90 binemia: two cases. J Emerg Med Suppl 1991;1:45–9 8. Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia 22. Rieder HU, Frei FJ, Zbinden AM, Thomson DA. Pulse oximetry on pulse oximetry and mixed venous oximetry. Anesthesiol- in methaemoglobinaemia. Failure to detect low oxygen satu- ogy 1989;70:112–7 ration. Anaesthesia 1989;44:326–7 9. Yang JJ, Lin N, Lv R, Sun J, Zhao F, Zhang J, Xu JG. 23. Radical-7 color display signal extraction pulse co-oximeter Methemoglobinemia misdiagnosed as ruptured ectopic preg- with Rainbow Technology. Operator‘s manual. Irvine, CA: nancy. Acta Anesth Scand 2005;49:586–8 Masimo Corporation, 2007

1068 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA CASE REPORT Symptomatic Axillary Hematoma After Ultrasound- Guided Infraclavicular Block in a Patient with Undiagnosed Upper Extremity Mycotic Aneurysms

Dave Gleeton, MD, Simon Levesque, MD, FRCPC, Claude A. Tre´panier, MD, FRCPC, Jean-Luc Garie´py, MD, FRCPC, Jean Brassard, MD, FRCPC, and Nicolas Dion, MD, FRCPC

We present a case of axillary hematoma complicating an ultrasound-guided infraclavicular block in a patient with undiagnosed mycotic aneurysms of the peripheral arteries. Mycotic aneurysm is a rare medical condition with well-identified risk factors. When performing regional anesthesia in patients with these risk factors, clinicians should have a high degree of suspicion about the possible existence of vascular anomalies. A preprocedure Doppler study of the block area and real-time guidance of the needle using ultrasound may be useful. (Anesth Analg 2010;111:1069–71)

ltrasound guidance can detect variations in normal offered and accepted by the patient. After sedation with anatomy, thus allowing the anesthesiologist to midazolam 2 mg, standard skin asepsis was accomplished tailor the technique to the patient-specific condi- with 2% wt/vol chlorhexidine gluconate and 70% vol/vol U1–3 tion. Despite these advantages, complications still occur isopropyl alcohol and a sterile sheath was used to cover the during ultrasound-guided peripheral nerve blocks.4 We ultrasound probe. A 5- to 12-MHz linear probe was posi- present a case in which the operator using ultrasound failed tioned in a parasagittal plane, medial to the coracoid to detect mycotic aneurysms of the upper extremity arter- process and adjusted to give a transverse view of the ies, leading to a large hematoma of the axillary region after axillary artery using an ultrasound device (Zone Ultra; an ultrasound-guided infraclavicular block. Zonare Medical Systems, Mountain View, CA). Using an in-plane technique, an 8.89-cm 20-gauge Tuohy needle (B. CASE DESCRIPTION Braun, Bethlehem, PA) was advanced to the posterior side A 44-year-old women, ASA physical status III, was admitted of the axillary artery until a fascial click was perceived and in our hospital with a diagnosis of infected olecranon bursa of 30 mL of mepivacaine 1.5% was injected slowly after the right elbow. Her recent medical history included a mitral multiple negative aspirations. After documentation of suc- valve replacement for mitral endocarditis (Streptococcus viri- cessful sensory block of the upper arm, surgery was dans) 4 months previously. Her medical history revealed IV performed using a tourniquet (pressure of 250 mm Hg). drug abuse, recurrent cellulitis of the right upper limb, high The surgeon opened and drained the olecranon bursa, and blood pressure, Type 2 diabetes, gastroesophageal reflux, and the procedure lasted 10 minutes. The patient was then mild asthma. Her preoperative medication included warfarin, directed to the postanesthesia care unit for a 30-minute rosuvastatin, metformin, diltiazem, acetylsalicylic acid, observation period and was subsequently discharged to the budesonide/formoterol, salbutamol, and pantoprazole. Be- ward. With agreement of the surgeon, IV heparin was cause surgery to drain the bursa was planned, warfarin was resumed 2 hours after the end of the surgery and warfarin discontinued 24 hours before surgery and the patient was was restarted on the first postoperative day. The patient given 5 mg IV of vitamin K after which the international was followed daily and discharged from the hospital 6 days normalized ratio was 1.3. While she awaited surgery, prophy- later, after a therapeutic level of anticoagulation (interna- lactic IV heparin was started and stopped 7 hours before tional normalized ratio 2.6) had been obtained. At that time, surgery. No other coagulation test was done immediately the patient complained of pain in the right shoulder but a before surgery. physical examination did not reveal any abnormality. At her arrival in the operating room, standard monitor- Two weeks later, the patient consulted at emergency ing and an IV line were started. Because the patient did not room for severe pain in her right shoulder. A physical show any sign of systemic infection (normal white blood examination demonstrated a significant swelling of the cell count and temperature), an anesthetic technique con- anterior part of the shoulder, the axilla, and the upper part sisting of an ultrasound-guided infraclavicular block was of the right arm. There was no redness or discoloration of the skin. A neurological examination of the upper arm did From the De´partement d’Anesthe´sie-Re´animation, Centre de Recherche du CHA, Unite´ de Recherche en Traumatologie-Urgence-Soins Intensifs, Hoˆpi- not demonstrate any motor or sensory deficit but the range tal de l’Enfant-Je´sus, Universite´ Laval, Que´bec, Canada. of movement of the right shoulder was limited by severe Accepted for publication June 1, 2010. pain. The radial pulse was present at the right wrist. A Supported by intramural department sources. superficial sonogram of the upper part of the right arm Disclosure: The authors report no conflicts of interest. showed a solid hyperechogenic mass of 7.8 ϫ 3.5 ϫ 4.7 cm Address correspondence and reprint requests to Dave Gleeton, MD, Depart- containing a small amount of liquid. A Doppler study did ment of Anesthesiology, Hoˆpital de l’Enfant-Je´sus, Universite´ Laval, 1401, 18e`me rue, Quebec, QC, G1J 1Z4, Canada. Address e-mail to [email protected]. not reveal any flow in this mass. A computed tomographic Copyright © 2010 International Anesthesia Research Society angiography showed a 5.4-cm hematoma located at the DOI: 10.1213/ANE.0b013e3181ee80b3 axillo-humeral junction of the axillary artery within which

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1069 CASE REPORT

tal the following day without any complication. Three follow-up visits revealed a complete regression of the swelling and the pain. One month later, the patient had no sequel.

DISCUSSION Mycotic aneurysms are defined as an infectious break in the wall of an artery with formation of a blind saccular outpouching that is contiguous with the arterial lumen, usually in an area of bifurcation or narrowing. The aorta, peripheral arteries, cerebral arteries, and visceral arteries are involved in descending order of frequency.5 In the pre–antibiotic era, Ͼ85% of mycotic aneurysms were associated with bacterial endocarditis. Currently, the majority of mycotic aneurysms occur in IV drug users or after invasive medical procedures. Depressed host immunity secondary to systemic disease (diabetes, cirrhosis, collagen vascular disease) and corticosteroid therapy are also contributing factors.6 Only 15 mycotic aneurysms of the subclavian artery have been reported since 1923.7 The diagnosis is often difficult because of the insidious nature of the disease. Pain, erythema, palpable mass, or ischemia distal to the affected area is sometimes present.6 Computed tomographic angiography is the imaging modality of choice for evaluation of mycotic aneurysms but Doppler sonography has both good sensitivity and specificity for detection of mycotic aneurysms located in peripheral arteries.8 The usual treatment is surgical excision but endovascular techniques have also been reported.9 One of the main advantages of ultrasound-guided nerve blocks is the possibility of a real-time visualization of the needle, nerve, and surrounding structures, notably the vessels. It has been demonstrated that ultrasound guidance diminishes the rate of vascular puncture during infraclavicular nerve block compared with a nerve stimulation technique.10 Moreover, it sometimes allows the detection of abnormal anatomy, thus offering the possibility of making adjustments to the anesthetic technique planned.1,3 How- ever, complications have not always been prevented by the use of ultrasound guidance to perform nerve blockade.4,11 The location of the mycotic aneurysms just below the scar Figure 1. A, Angiography of the shoulder area before the emboliza- left by the needle on the skin most likely suggests that in ϭ ϭ tion; black arrow 13-mm mycotic aneurysm; white arrow 5-mm our case 1 of the 2 preexisting mycotic aneurysms was mycotic aneurysm. B, Angiography performed after the embolization; black arrow ϭ previous site of the 13-mm mycotic aneurysm showing punctured by the needle during the technique. This acci- complete exclusion of the aneurysm. dental puncture during the course of an apparently uncomplicated technique can occur in 2 different situations. First, it is possible that the needle was not adequately visualized was a 20-mm pool of contrast dye suggesting an aneurysm by the anesthesiologist (one of the most frequent errors of just underneath the wound left by a recent cutaneous clinicians performing ultrasound-guided nerve block12) puncture below the right clavicle (consistent with the and could have punctured the mycotic aneurysms just puncture site of the infraclavicular block). A second aneu- outside the ultrasound visualization plane. It is also pos- rysm was suspected in the deltoid area. Digital angiogra- sible that the mycotic aneurysms were located within the phy confirmed the presence of 2 aneurysms, 1 near the ultrasound visualization plane but were not recognized humeral neck and the other in the prescapular region (5 and diagnosed, leading to a witnessed but unrecognized and 13 mm). These were supplied by a right lateral branch puncture during the technique. of the dorsoscapular artery and the right thoracoacromial The use of color Doppler study during sonographic artery, respectively. Supraselective catheterization of the examination results in a characteristic yin-yang sign in the feeding artery and embolization with liquid adhesive glue presence of mycotic aneurysms (Fig. 2), thus greatly en- (Indermil; Tyco, Norwalk, CT) and lipiodol was performed. hancing the sensibility of ultrasound to detect mycotic Control angiogram confirmed the occlusion of both aneu- aneurysms.5 The blood stasis inside the mycotic aneurysms rysms (Fig. 1). The patient was discharged from the hospi- produces an unusual gray ultrasound image of arterial

1070 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Undiagnosed Upper Extremity Mycotic Aneurysms

complications. Despite the advantages of this technology, a high degree of suspicion should be maintained to minimize the risk of complications.13 Finally, IV drug users or patients with a history of endocarditis are at a high risk of mycotic aneurysms and may benefit from a formal preprocedure examination before a peripheral nerve block.14

AUTHOR CONTRIBUTIONS All authors helped with manuscript preparation. REFERENCES 1. Duggan E, Brull R, Lai J, Abbas S. Ultrasound-guided brachial plexus block in a patient with multiple glomangiomatosis. Reg Anesth Pain Med 2008;33:70–3 2. Kessler J, Gray AT. Sonography of scalene muscle anomalies for brachial plexus block. Reg Anesth Pain Med 2007;32:172–3 3. Sites BD, Spence BC, Gallagher JD, Beach ML. On the edge of the ultrasound screen: regional anesthesiologists diagnosing nonneural pathology. Reg Anesth Pain Med 2006;31:555–62 4. Zetlaoui PJ, Labbe JP, Benhamou D. Ultrasound guidance for axillary plexus block does not prevent intravascular injection. Anesthesiology 2008;108:761 5. Lee W, Mossop PF, Little AF, Fitt GJ, Vrazas JI, Hoang JK, Hennessy OF. Infected (mycotic) aneurysms: spectrum of imaging appearances and management. Radiographics 2008;28:1853–68 6. Kearney RA, Eisen HJ, Wolf JE. Nonvalvular infections of the Figure 2. Sonograms show a 5.5-cm complex lesion. On the color cardiovascular system. Ann Intern Med 1994;121:219–30 Doppler image, the hypoechoic center has turbulent flow (“ying-yang” 7. Tsao JW, Marder SR, Goldstone J, Bloom AI. Presentation, sign), a finding indicative of a patent aneurysm lumen. The thick, diagnosis, and management of arterial mycotic pseudoaneu- heterogeneous, hypoechoic rind is attributable to hematoma and rysms in injection drug users. Ann Vasc Surg 2002;16:652–62 inflammatory tissue. (Reprinted with permission from Lee et al.5 [Lee 8. Coughlin BF, Paushter DM. Peripheral pseudoaneurysms: W-K, Mossop PJ, Little AF, et al. Infected (mycotic) aneurysms: evaluation with duplex US. Radiology 1988;168:339–42 spectrum of imaging appearances and management. Radiographics 9. Leon LR, Psalms SB, Labropoulos N, Mills JL. Infected upper 2008;28:1853–68].) extremity aneurysms: a review. Eur J Vasc Endovasc Surg 2008;35:320–31 10. Maalouf D, Gordon M, Paroli L, Tong-Ngork S. Ultrasound- lumen that might complicate its recognition when Doppler guidance vs. nerve stimulation for the infraclavicular blockade is not used. In our case, a preprocedure Doppler study was of the brachial plexus: a comparison of the vascular puncture rate. Reg Anesth Pain Med 2006;30:A46 not performed before the technique. Preprocedure diagno- 11. Loubert C, Williams SR, He´lie F, Arcand G. Complication sis of the mycotic aneurysms in the needle path could have during ultrasound-guided regional block: accidental intra- resulted in modification of the anesthetic technique (choice vascular injection of local anesthetic. Anesthesiology of another block site, general anesthesia, use of smaller 2008;108:759–60 needle, etc.) that could have prevented the occurrence of 12. Sites BD, Spence BC, Gallagher JD, Wiley CW, Bertrand ML, Blike GT. Characterizing novice behaviour associated with the complication. However, it should be kept in mind that learning ultrasound-guided peripheral regional anesthesia. ultrasound is a relatively recent technology in the anesthe- Reg Anesth Pain Med 2007;32:107–15 siology field. Because of its rare use for diagnostic pur- 13. Benitez PR, Newell MA. Vascular trauma in drug abuse: poses, certain rare medical conditions such as the one patterns of injury. Ann Vasc Surg 1986;1:175–81 14. Manickam BP, Perlas A, Chan VW, Brull R. The role of a reported herein may still go unrecognized. preprocedure systematic sonographic survey in ultrasound- In conclusion, this case report illustrates that the use of guided regional anesthesia. Reg Anesth Pain Med ultrasound cannot completely eliminate the occurrence of 2008;33:566–70

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1071 COCHRANE CORNER

Infraclavicular Brachial Plexus future comparative studies. There is also a need for additional RCTs comparing ultrasound-guided ICB with other BPBs. Block for Regional Anaesthesia of Chin KJ, Singh M, Velayutham V, Chee V. Infraclavicular brachial plexus block for regional anaesthesia of the lower arm the Lower Arm published, in the Cochrane Database Syst Rev 2010, Issue 2. Art. Ki Jinn Chin, Mandeep Singh, Veerabadran No.: CD005487. DOI: 10.1002/14651858.CD005487.pub2 Velayutham, and Victor Chee

BACKGROUND: Several approaches exist to produce local Heated Humidification Versus Heat anaesthetic blockade of the brachial plexus. It is not clear which is the technique of choice for providing surgical anaes- and Moisture Exchangers for thesia of the lower arm although infraclavicular blockade (ICB) Ventilated Adults and Children has several purported advantages. We therefore performed a systematic review of ICB compared to the other brachial plexus Margaret Kelly, Donna Gillies, David A. Todd, and blocks (BPBs). Catherine Lockwood OBJECTIVES: To evaluate the efficacy and safety of ICB compared to other BPBs in providing regional anaesthesia of the lower arm. BACKGROUND: Humidification by artificial means must be SEARCH STRATEGY: We searched CENTRAL (The Cochrane provided when the upper airway is bypassed during mechanical Library 2008, Issue 3), MEDLINE (1950 to September 22nd ventilation. Heated humidification (HH) and heat and moisture 2008) and EMBASE (1980 to September 22nd 2008). We also exchangers (HMEs) are the most commonly used types of searched conference proceedings (from 2004 to 2008) and artificial humidification in this situation. the www.clinicaltrials.gov registry. No language restriction was OBJECTIVES: To determine whether HHs or HMES are more applied. effective in preventing mortality and other complications in SELECTION CRITERIA: We included any randomized con- people who are mechanically ventilated. trolled trials (RCTs) that compared ICB with other BPBs as the SEARCH STRATEGY: We searched the Cochrane Central sole anaesthetic techniques for surgery on the lower arm. Register of Controlled Trials (The Cochrane Library 2010, Issue DATA COLLECTION AND ANALYSIS: The primary outcome 4) and MEDLINE, EMBASE and CINAHL (January, 2010) to was adequate surgical anaesthesia within 30 minutes of block identify relevant randomized controlled trials. completion. Secondary outcomes included sensory block of SELECTION CRITERIA: We included randomized controlled tri- individual nerves, tourniquet pain, onset time of sensory als comparing HMEs to HHs in mechanically ventilated adults and blockade, block performance time, block-associated pain and children. We included randomized crossover studies. complications related to the block. DATA COLLECTION AND ANALYSIS: We assessed the quality MAIN RESULTS: We identified 15 studies with 1020 partici- of each study and extracted the relevant data. Where appro- pants, of whom 510 received ICB and 510 received other priate, results from relevant studies were meta-analyzed for BPBs. The control group intervention was the axillary block in individual outcomes. MAIN RESULTS: We included 33 trials with 2833 participants; 10 studies, mid-humeral block in two studies, supraclavicular 25 studies were parallel group design (n ϭ 2710) and 8 block in two studies and parascalene block in one study. Three crossover design (n ϭ 123). Only 3 included studies reported studies employed ultrasound-guided ICB. The risk of failed data for infants or children. There was no overall effect on surgical anaesthesia and of complications were low and similar artificial airway occlusion, mortality, pneumonia, or respiratory for ICB and all other BPBs. Tourniquet pain was less likely with ϭ complications; however, the PaCO2 and minute ventilation ICB (risk ratio (RR) 0.47, 95% CI 0.24 to 0.92, P 0.03). were increased when HMEs were compared to HHs and body When compared to a single-injection axillary block, ICB was temperature was lower. The cost of HMEs was lower in all better at providing complete sensory block of the musculocu- studies that reported this outcome. There was some evidence Ͻ taneous nerve (RR for failure 0.46, 95% CI 0.27 to 0.60, P that hydrophobic HMEs may reduce the risk of pneumonia and 0.0001) and the axillary nerve (RR of failure 0.37, 95% CI 0.24 that blockages of artificial airways may be increased with the Ͻ to 0.58, P 0.0001). ICB was faster to perform than use of HMEs in certain subgroups of patients. Ϫ multiple-injection axillary (mean difference (MD) 2.7 min, AUTHORS’ CONCLUSIONS: There is little evidence of an Ϫ Ϫ ϭ 95% CI 4.2 to 1.1, P 0.0006) or midhumeral blocks (MD overall difference between HMEs and HHs. However, hydropho- Ϫ4.8 min, 95% CI Ϫ6.0 to Ϫ3.6, P Ͻ 0.00001) but this was bic HMEs may reduce the risk of pneumonia and the use of an offset by a longer sensory block onset time (MD 3.9 min, 95% HMEs may increase artificial airway occlusion in certain sub- CI 3.2 to 4.5, P Ͻ 0.00001). groups of patients. Therefore, HMEs may not be suitable for AUTHORS’ CONCLUSIONS: ICB is a safe and simple tech- patients with limited respiratory reserve or prone to airway nique for providing surgical anaesthesia of the lower arm, with blockage. Further research is needed relating to hydrophobic an efficacy comparable to other BPBs. The advantages of ICB versus hygroscopic HMEs and the use of HMEs in the pediatric include a lower likelihood of tourniquet pain during surgery, and and neonatal populations. As the design of HMEs evolves, more reliable blockade of the musculocutaneous and axillary evaluation of new generation HMEs will also need to be nerves when compared to a single-injection axillary block. The undertaken. efficacy of ICB is likely to be improved if adequate time is Kelly M, Gillies D, Todd DA, Lockwood C. Heated humidifi- allowed for block onset (at least 30 minutes) and if a volume of cation versus heat and moisture exchangers for ventilated at least 40 ml is injected. Since publication of many of the adults and children. Cochrane Database Syst Rev 2010, Issue trials included in this review, it has become clear that a distal 4. Art. No.: CD004711. posterior cord motor response is the appropriate endpoint for DOI: 10.1002/14651858.CD004711.pub2. electrostimulation-guided ICB; we recommend it be used in all

1072 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 COCHRANE CORNER

1072 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 LETTERS TO THE EDITOR Section Editor: Lawrence Saidman

Airway Management: Standardization, REFERENCES 1. Hung O, Murphy M. Context-sensitive airway management. Simplicity, and Daily Practice Are the Anesth Analg 2010;110:982–3 2. American Society of Anesthesiologists Task Force on Manage- Keys to Success ment of the Difficult Airway (2003). Practice guidelines for management of the difficult airway: an updated report by the American Society of Anesthesiologists Task Force on Manage- ment of the Difficult Airway. Anesthesiology 2003;98:1269–77 To the Editor 3. Hopkins A. Measuring the Quality of Medical Care. 1st ed. he editorial by Hung and Murphy,1 although inter- Oxford: Royal College of Physicians of London, 1990 esting, contains potentially misleading statements. 4. Combes X, Le Roux B, Suen P, Dumerat M, Motamed C, Sauvat S, Duvaldestin P, Dhonneur G. Unanticipated difficult airway in TSome have opined that careful evaluation of the anesthetized patients: prospective validation of a management airway as part of a preplanned strategy may lead to algorithm. Anesthesiology 2004;100:1146–50 improved outcome.2 However, it is potentially dangerous 5. Heidegger T, Gerig HJ, Ulrich B, Kreienbu¨ hl G. Validation of a to suggest that the choice of the technique and hence the simple algorithm for tracheal intubation: daily practice is the key to success in emergencies—an analysis of 13,248 intuba- choice of the equipment is or should be influenced primar- tions. Anesth Analg 2001;92:517–22 ily by different circumstances. Advocating such an ap- 6. Reason J. Human error: models and management. BMJ proach would result in multiple strategies using many 200;320:768–70 different and, at times, unfamiliar airway devices. 7. Heidegger T, Gerig HJ, Henderson JJ. Strategies and algorithms for management of the difficult airway. Best Pract Res Clin Only a few studies have focused on effective airway Anaesthesiol 2005;19:661–74 management and, in these, outcome was investigated un- DOI: 10.1213/ANE.0b013e3181ec312a der the prevailing clinical conditions.3–5 A common characteristic of each of these studies was a limitation of In Response techniques and devices and that deviation from the pre- No one disagrees that the fundamental goal of airway man- defined algorithm was recorded infrequently. agement is oxygenation and ventilation, not devices and tools. A key factor regarding safety recorded by highly reliable A systematic approach to airway management includes air- organizations such as aviation is a standardized process.6 way evaluation, selection of an appropriate course of action However, the “recommendations” of Hung and Murphy likely to succeed (Plan A) and preparation for failure (i.e., Plan that airway management is primarily “context sensitive” B, Plan C, etc.). This approach is, and must be, consistent with would guide us in the wrong direction. There are only a the “context.” In other words, one must accept that airways few situations wherein we might deviate from our difficult present themselves in a variety of forms, circumstances, and airway guidelines. For example, it is very unlikely that you locations, to a panoply of health care providers with varying can perform an awake intubation in an uncooperative skill sets. Although there is a recommended “strategy,” there patient. Importantly, such situations should be managed by are varying tactics or techniques one may use. The tactic or the most experienced physicians. However, if deviation technique is virtually always dependent on the circumstances from, rather than adherence to, the guidelines is current and skill set of the airway practitioner. practice, the guidelines should be modified accordingly. Apart from the oxygenator of a bypass pump, clinicians A second point is the missing commitment to fiberoptic basically use only 4 methods of ventilation and oxygenation: intubation (“unwritten truth”). If the authors mean that through a bag-mask, an extraglottic device (e.g., a laryngeal there are no prospective randomized studies showing the mask airway), a tracheal tube, or a surgical airway. As stated effectiveness of fiberoptic intubation, they are correct. Un- in our editorial, “context-sensitive airway management im- fortunately, there are no prospective randomized studies plies that managing a difficult or failed airway should be demonstrating the effectiveness (not efficacy) of most tech- driven by the principles of ‘gas exchange’ and not be ‘device- niques used in daily practice. However, it is generally dependent.’”1 In addition, we also stated that “clinicians must agreed among airway management practitioners and rec- be trained to understand the basic principles of airway ommended by many anesthesia societies that fiberoptic management using basic techniques and learn how to apply intubation should be used for management of the antici- these techniques properly in an appropriate environment.” pated difficult airway.7 So, questioning this technique is These statements were structured to inform clinicians that probably potentially misleading for the average anesthesi- they must learn basic airway techniques and apply them ologist in practice. properly. Nowhere in the editorial did we advocate that “… Regarding airway management, the message should be: an approach would result in multiple strategies using many Standardization, simplicity, and daily practice are the keys different and at times, unfamiliar airway devices.” to success. Dr. Heidegger correctly states that awake tracheal intubation is considered to be the most prudent approach in a Thomas Heidegger, MD patient with an anticipated difficult airway.2 But, in addi- Department of Anesthesia tion to the flexible bronchoscope, awake intubation can be Spitalregion Rheintal Werdenberg Sarganserland performed safely and effectively by many techniques, in- Walenstadt, Switzerland cluding the rigid fiberoptic laryngoscopes, videolaryngo- [email protected] scopes, and Macintosh laryngoscope with the Eschmann

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1073 LETTERS TO THE EDITOR Section Editor: Lawrence Saidman

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1073 LETTERS TO THE EDITOR

Tracheal Introducer. Dr. Heidegger also correctly states Achieving Full Risk Disclosure in that “. . . it is generally agreed among airway management practitioners and recommended by many anesthesia soci- Pediatric Anesthesia Research eties that fiberoptic intubation should be used for management of the anticipated difficult airway.” However, it would be grossly incorrect and perhaps dangerous for Dr. To the Editor Heidegger to imply that “awake fiberoptic intubation ong et al.’s1 study using fluoroscopy on 73 ASA should be standardized in the management of the antici- physical status I children, ages 1 to 5 years, to pated difficult airway.” In an uncooperative patient, in the Hassess ropivacaine–radiopaque dye solution spread presence of blood, in emergency situations, or in an envi- in caudal injections for postorchiopexy pain raises a vexing ronment with limited resources, fiberoptic intubation question: Were all known risks fully disclosed to their would be difficult, if not impossible. subjects’ parents? The absence of details about fluoroscopy Having taught airway management to thousands of equipment, radiation reduction measures, radiation doses, practitioners, we recognize that there is tremendous vari- and subject radiation risks suggests that they were not. ability within individual skill sets. Because of this variabil- Strauss and Kaste2 stated (and Linet et al.3 concurred) ity, the principles of airway management and the strategy that the ALARA concept means radiation should be “As for managing a difficult airway have been well elucidated Low As Reasonably Achievable.” Children, they noted, by guidelines promulgated by various societies including 3,4 “might be as much as 10 times more radiosensitive than the American Society of Anesthesiologists. It is critical for adults.” Cohen4 compared 2 fluoroscopic machines and all practitioners to recognize that these guidelines are not noted that “the radiologist can greatly increase patient recipes to be rigidly followed. Rather, as stated clearly in exposure by merely altering settings over which they have the ASA Practice Guidelines for Management of the Diffi- immediate control” and cited exposure variations of 3486% cult Airway,3,4 “. . . these recommendations may be adopted, and 4479% on the basis of such factors. Petterson et al.5 modified, or rejected according to clinical needs and constraints” found a 2.23 testicular cancer relative risk for children with (context-sensitive). Furthermore, “Practice guidelines are undescended testes operated on before age 13. Presumably, not intended as standards or absolute requirements. The their risk is augmented by radiation exposure. use of practice guidelines cannot guarantee any specific outcome. Practice guidelines are subject to revision as Hong et al. experimented to answer a clinical question. warranted by the evolution of medical knowledge, technol- Their study conferred no additional benefit to subjects, only 6 ogy, and practice.”3,4 a potential of harm. Science alone benefited. Beecher warned researchers away from such practices. In research Orlando Hung, MD as in clinical practice, Primum Non Nocere still applies. Professor Anesthesiology, Surgery, and Pharmacology Dalhousie University Vincent J. Kopp, MD Queen Elizabeth II Health Sciences Michael G. Danekas, MD Halifax, Nova Scotia, Canada Division of Pediatric Anesthesia [email protected] Department of Anesthesiology Michael Murphy, MD School of Medicine Professor and Chair Anesthesiology University of North Carolina at Chapel Hill Professor Emergency Medicine Chapel Hill, North Carolina Dalhousie University [email protected] District Chief Anesthesiology Capital District Health Authority REFERENCES Queen Elizabeth II Health Sciences 1. Hong J-Y, Han SW, Kim WO, Cho JS, Kil HJ. A comparison of Halifax, Nova Scotia, Canada high volume/low concentration and low volume/high concen- [email protected] tration ropivacaine in caudal analgesia for pediatric orchiopexy. Anesth Analg 2009;109:1073–8 2. Strauss KJ, Kaste SC. The ALARA concept in pediatric interven- REFERENCES tional fluoroscopic imaging: striving to keep radiation doses as 1. Hung O, Murphy M. Context-sensitive airway management. low as possible during fluoroscopy of pediatric patients. A Anesth Analg 2010;110:982–3 white paper executive summary. AJR 2006;187:818–9 2. Heidegger T. Airway management: standardization, simplicity, 3. Linet MS, Kim KP, Rajaraman P. Children’s exposure to diag- and daily practice are the keys to success. Anesth Analg 2010;111:1073 nostic medical radiation and cancer risk: epidemiologic and 3. Practice guidelines for management of the difficult airway: a dosimetric considerations. Pediatr Radiol 2009;39(Suppl 1):S4–26; report by the American Society of Anesthesiologists Task Force Epub 2008 Dec 16 on Management of the Difficult Airway. Anesthesiology 4. Cohen M. Are we doing enough to minimize fluoroscopic 1993;78:597–602 radiation exposure in children? Pediatr Radiol 2007;37:1020–4 4. American Society of Anesthesiologists Task Force on Manage- 5. Petterson A, Richardi L, Nordensklod, Kaijser M, Akre O. Age ment of the Difficult Airway. Practice guidelines for manage- at surgery for undescended testis and risk of testicular cancer. ment of the difficult airway: an updated report by the American N Engl J Med 2007;356:1835–41 Society of Anesthesiologists Task Force on Management of the 6. Beecher HK. Ethics and clinical research. N Engl J Med Difficult Airway. Anesthesiology 2003;98:1269–77 1966;274:367–72 DOI: 10.1213/ANE.0b013e3181ec3153 DOI: 10.1213/ANE.0b013e3181ed17ff

1074 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA LETTERS TO THE EDITOR

1074 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Letters to the Editor

In Response Pulse Dye Densitometry and We agree with Kopp et al.1 that the ALARA (as low as Indocyanine Green Plasma reasonably achievable) concept must be considered in pediatric fluoroscopic imaging. Of course, in our study2 we Disappearance: The Issue of obtained consent from the parents of the children after “Normal” Values explaining details of the study and radiation exposure. The genital area was protected from radiation by a lead plate during fluoroscopy using a new pulsed unit from Phillips To the Editor (Eindhoven, The Netherlands). The study by Cohen3 de- eekers et al.1 investigated normal values of plasma scribed possible radiation exposure variations as much as disappearance rate of indocyanine green (PDR-ICG) 3486% and 4479% with the maximum magnification, great- measured by pulse dye densitometry in a population est pulse rate, intensifier high, grid in, and greatest inten- R of ASA physical status I and II patients, assumed free of sifier exposure rate. We believe that the exposure variations liver disease. They also evaluated the noninvasive mea- were not as great because we did not alter the basal surement of PDR-ICG using a nose and finger probe and magnification setting and used the lowest pulse rate, grid compared these results with those using invasive arterial out, and lowest intensifier exposure rate. The exposure blood measurements. frequency to obtain 1 image (exposure rate) was 1 or 2, and Reekers et al. confirmed that, as has been shown before, Ͻ overall duration of fluoroscopy was 30 seconds. How- PDR-ICG is adequately measured noninvasively by pulse ever, because the average estimated entrance surface dose dye densitometry.2,3 In addition, they found a mean PDR- (ESD) and dose area product increases with age, the ESD ICG value of 23.1%/min (SD Ϯ 7.9%/min), which is in line 4 also increases in children in comparison with adults ;we with previous studies.4–6 However, they found that PDR- investigated the ESD in 5 randomly selected children in our ICG varied widely with a range from 9.7%/min to 5 previous study. The ESD was 0.05 to 0.26 mSv, which was 43.2%/min. This is in contrast to earlier publications in less than the results of a United Kingdom nationwide which PDR-ICG ranged from 18.7%/min to 30.1%/min.4–6 6 survey in 2005. Although our procedures did not diverge As explained by the authors, a PDR-ICG Ͻ18%/min is from the ALARA concept when we substituted the proce- generally considered to indicate impaired hepatic function.5,7 dure to the flow diagram for managing patient dose by Although the authors argue that only patients without overt 7 Strauss and Kaste, we believe that the radiation exposure liver pathology were examined, there is a lack of objective should be strictly limited in children. evidence of normal hepatic function in the entire population studied. Biochemical liver function tests were available in Hae K. Kil, MD only 22% of the patients. The authors recognize this limitation Jeong Y. Hong, MD and argue that the patients studied are representative of a Won O. Kim, MD population with normal liver function based on the absence of Department of Anesthesiology and Pain Medicine any clinical evidence of hepatic dysfunction. This is a rational Yonsei University College of Medicine argument, but the clinical evidence for normal liver function is Seoul, South Korea not a surrogate for laboratory testing because patients with [email protected] elevated liver enzymes do not necessarily present clinical signs of hepatic dysfunction.8 Because PDR-ICG is very REFERENCES sensitive for hepatic dysfunction, changes in PDR-ICG can 1. Kopp VJ, Danekas MG. Achieving full risk disclosure in pedi- even precede changes in serum bilirubin, as the authors atric anesthesia research. Anesth Analg 2010;111:1074 stated.1 Given the fact that the authors ultimately want to 2. Hong JY, Han SW, Kim WO, Cho JS, Kil HK. A comparison of high volume/low concentration and low volume/high concen- argue for a testing threshold to identify normal liver function, tration ropivacaine in caudal analgesia for pediatric orchiopexy. the study would have been more compelling if all patients Anesth Analg 2009;109:1073–8 had received a careful hepatic testing. 3. Cohen M. Are we doing enough to minimize fluoroscopic Unfortunately, Reekers et al. also did not present any data radiation exposure in children? Pediatr Radiol 2007;37: describing hemodynamics of their patients. This is, however, 1020–4 of utmost importance, because PDR-ICG is, as is ICG liver 4. Linet MS, Kim KP, Rajaraman R. Children’s exposure to diagnostic medical radiation and cancer risk: epidemiologic and dosimetric extraction capacity, primarily dependent on splanchnic blood considerations. Pediatr Radiol 2009;39(Suppl 1):S4–26 flow and total plasma volume.2,9 It is also not clear why 5. Shin SK, Hong JY, Kim WO, Koo BN, Kim JE, Kil HK. Ultra- Reekers et al. studied both awake and anesthetized patients. sound evaluation of the acral area and comparison of sacral There are clearly potential differences between the groups interspinous and hiatal approach for caudal block in children. with respect to hemodynamics, which may influence the Anesthesiology 2009;111:1135–40 6. Hart D, Hillier MC, Wall BF. National reference doses for obtained PDR-ICG values. Reekers et al. argue that the results common radiographic, fluoroscopic and dental X-ray examina- in the anesthetized patients were not different from the awake tions in the UK. Br J Radiol 2009;82:1–12 patients when analyzed separately so that the data were 7. Strauss KJ, Kaste SC. The ALARA (as low as reasonably pooled. However, it is well known that induction of anesthe- achievable) concept in pediatric interventional and fluoroscopic sia with propofol results in a significant decrease in cardiac imaging: striving to keep radiation doses as low as possible during fluoroscopy of pediatric patients. A white paper execu- output decreasing splanchnic blood flow and subsequently tive summary. Pediatr Radiol 2006;36(Suppl 2):110–2 impairing PDR-ICG, even in a relatively young group of ASA DOI: 10.1213/ANE.0b013e3181ed1811 physical status I and II patients.10

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1075 LETTERS TO THE EDITOR

In conclusion, we believe the relatively wide and low ranged PDR-ICG “normal” values found by Reekers et al. might be the result of nonhomogeneity of the patient group with nonconstant hemodynamics and thus should not lead to a premature rejection of a well-established method.

Jaap J. Vos, BSc Thomas W. L. Scheeren, MD, PhD Go¨tz J. K. Wietasch, MD, PhD Department of Anesthesiology University Medical Center Groningen University of Groningen Groningen, The Netherlands [email protected]

REFERENCES 1. Reekers M, Simon MJ, Boer F, Mooren RA, van Kleef JW, Figure 1. Representation of cardiac output measurements based on Dahan A, Vuyk J. Pulse dye densitometry and indocyanine arterial blood indocyanine green (ICG) concentrations versus ICG– green plasma disappearance in ASA physical status I-II pa- plasma disappearance rate (PDR) values determined in patients who tients. Anesth Analg 2010;110:466–72 underwent simultaneous arterial ICG sampling. The circles represent 2. von Spiegel T, Scholz M, Wietasch G, Hering R, Allen SJ, Wood the measurements in awake subjects; the diamonds represent the P, Hoeft A. Perioperative monitoring of indocyanine green measurements during propofol induction. clearance and plasma disappearance rate in patients undergoing liver transplantation. Anaesthesist 2002;51:359–66 3. Sakka SG, Reinhart K, Meier-Hellmann A. Comparison of invasive and noninvasive measurements of indocyanine green patients displaying these factors were excluded from our plasma disappearance rate in critically ill patients with me- study population, thus removing those patients that may chanical ventilation and stable hemodynamics. Intensive Care exhibit elevated hepatic enzymes in the absence of physical Med 2000;26:1553–6 4. Rowell LB, Blackmon JR, Bruce RA. Indocyanine green clear- signs or symptoms of liver disease. ance and estimated hepatic blood flow during mild to maximal Regarding the importance of hemodynamics and indocya- exercise in upright man. J Clin Invest 1964;43:1677–90 nine green plasma disappearance rate (ICG-PDR), we origi- 5. Hori T, Iida T, Yagi S, Taniguchi K, Yamamoto C, Mizuno S, nally intended to use the noninvasive method of ICG mea- Yamagiwa K, Isaji S, Uemoto S. K(ICG) value, a reliable real-time estimator of graft function, accurately predicts out- surement, pulse dye densitometry, to measure cardiac output comes in adult living-donor liver transplantation. Liver in our study population. To validate the transcutaneous Transpl 2006;12:605–13 method versus intraarterial measurement of ICG, we per- 6. de Liguori Carino N, O’Reilly DA, Dajani K, Ghaneh P, Poston formed simultaneous measurements in a subpopulation. In GJ, Wu AV. Perioperative use of the LiMON method of this subpopulation, we had to conclude that, for individual indocyanine green elimination measurement for the prediction and early detection of post-hepatectomy liver failure. Eur measurement of cardiac output, the transcutaneous measure- J Surg Oncol 2009;35:957–62 ment of ICG by pulse dye densitometry is not accurate 7. Kuntz H, Schregel W. Indocyanine green: evaluation of liver enough.4 For the purpose of this discussion, the measure- function—application in intensive care medicine. In: Lewis F, ments of cardiac output versus the ICG-PDR value in the Pfeiffer U, eds. Practical Applications of Fiberoptics in Critical Care Monitoring. 2nd ed. New York: Springer, 1990:57–62 patients in whom we performed arterial blood sampling is 8. Hultcrantz R, Glaumann H, Lindberg G, Nilsson LH. Liver shown in Figure 1. The open diamonds represent the patients investigation in 149 asymptomatic patients with moderately receiving a propofol induction. Inspection of Figure 1 leads to elevated activities of serum aminotransferases. Scand J Gastro- the conclusion that the absence or presence of propofol did enterol 1986;21:109–13 not induce significant hemodynamic changes in this other- 9. Caesar J, Shaldon S, Chiandussi L, Guevara L, Sherlock S. The use of indocyanine green in the measurement of hepatic blood wise healthy population nor did it affect ICG-PDR. 4 flow and as a test of hepatic function. Clin Sci 1961;21:43–57 We thus maintain our conclusion that ICG-PDR values 10. Lange H, Stephan H, Rieke H, Kellermann M, Sonntag H, Bircher in a population without clinical signs of liver failure range Ϫ J. Hepatic and extrahepatic disposition of propofol in patients well below 18% min 1, cited as the cutoff value for hepatic undergoing coronary bypass surgery. Br J Anaesth 1990;64:563–70 failure and propagated as criterion for clinical intervention. DOI: 10.1213/ANE.0b013e3181ef35ba This cutoff value needs to be reconsidered as has been suggested elsewhere.5 We agree with the reviewers that In Response further studies are needed and that the final word on this We agree with Vos et al.1 that the absence of liver enzyme subject has not yet been written. measurements in some of our patients may be a drawback in our study. However, biochemical hepatic function test- Marije Reekers, MD ing may not offer the key information on hepatocellular Fred Boer, MD, PhD dysfunction2 as Vos et al. suggest. Hultcrantz et al.3 described Jaap Vuyk, MD, PhD that chronically elevated liver enzymes without symptoms or Department of Anesthesiology physical signs of liver disease correspond with various forms Leiden University Medical Centre of liver disease preferably in the presence of a positive history Leiden, The Netherlands on alcohol consumption, drug abuse, hepatitis, or obesity. All [email protected]

1076 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Letters to the Editor

REFERENCES 1. Vos JJ, Scheeren TWL, Wietasch GJK. Pulse dye densitometry and indocyanine green plasma disappearance: the issue of “normal” values. Anesth Analg 2010;111:1075–6 2. Sakka SG. Assessing liver function. Curr Opin Crit Care 2007;13:207–14 3. Hultcrantz H, Glaumann H, Lindberg G, Nilsson LH. Liver investigation in 149 asymptomatic patients with moderately elevated activities of serum aminotransferases. Scand J Gastro- enterol 1986;21:109–13 4. Reekers M, Simon MJ, Boer F, Mooren FA, van Kleef JW, Dahan A, Vuyk J. Cardiovascular monitoring by pulse dye densitometry or arterial indocyanine green dilution. Anesth Analg 2009;109:441–6 5. Merle U, Sieg O, Stremmel W, Encke J, Eisenbach C. Sensitivity and specificity of plasma disappearance rate of indocyanine green as a prognostic indicator in acute liver failure BMC Gastroenterol- ogy 2009;9:91 DOI: 10.1213/ANE.0b013e3181ef35e7

Figure 1. Testing the effect of a pressure-rated needleless access connector (PNAC) on IV ﬂow rate. Two identical saline bags with IV Serotonin Syndrome in the Perioperative tubing of equal length were primed, suspended 7 feet above the Period: Role of Tramadol catheter tips, then allowed to simultaneously drain to gravity. The only difference between the 2 setups was the presence of a PNAC on the distal aspect of the IV tubing attached to 1 bag but not the To the Editor other. The time required for complete drainage of each saline bag was measured by means of a stopwatch; ﬂow rate was calculated 1 ltman and Jahangiri describe an important com- by dividing the volume of saline drained by time elapsed. plication associated with certain psychiatric as Awell as other serotonergic medications used in the perioperative period. However, in addition to the various drugs mentioned by the authors, an important Table 1. The Effect of Pressure-Rated Needleless drug also implicated in this setting is tramadol. It is a Access Connectors on Gravity-Driven Flow Rate Through Catheters of Various Internal Diameters centrally acting analgesic frequently used for treating moderate to severe postoperative pain, especially in Catheter Control setup, PNAC setup, Difference, mL/min parameter mL/min mL/min (% change) third world countries. Tramadol, a weak agonist at the a Ϫ Ϫ ␮ 22 gauge 40.8 40.0 0.8 ( 2%) -opioid receptor, also has a non-opioid mechanism of 20 gaugea 54.5 56.8 ϩ2.3 (ϩ4%) action that includes release of serotonin and inhibition of 16 gaugeb 214 121 Ϫ93 (Ϫ44%) reuptake of norepinephrine and is therefore likely to 14 gaugeb 223 138 Ϫ85 (Ϫ38%) 2 contribute to the development of serotonin syndrome. PNAC ϭ pressure-rated needleless access connector. Satinder Gombar, MD a Introcan Safety® IV Catheter (B. Braun Medical Inc., Bethlehem, PA). Nidhi Bhatia, MD b CATHLON® IV Catheter (Smiths Medical North America, Dublin, OH). Department of Anaesthesia & Intensive Care Government Medical College and Hospital Chandigarh, India devices. We sought to answer this question by measuring IV [email protected] flow rates through 2 nearly identical IV setups. The only difference between the 2 systems was the presence REFERENCES (“PNAC setup”) or absence (“control setup”) of a PNACa 1. Altman CS, Jahangiri MF. Serotonin syndrome in the perioper- on the distal aspect of the IV tubingb (Fig. 1). Although ative period. Anesth Analg 2010;110:526–8 c 2. Takeshita J, Litzinger MH. Serotonin syndrome associated with our results agree with that of the PNAC manufacturer for tramadol. Prim Care Companion J Clin Psychiatry 2009;11:273 22-gauge catheter sets, we observed a substantially decreased DOI: 10.1213/ANE.0b013e3181eb02e8 gravity-driven flow rate when the device was used with 14- and 16-gauge catheters (Table 1); device performance is not reported by the manufacturer for either of these large-bore Pressure-Rated Needleless Access catheters. Admittedly, these observations merely illustrate a Connectors Slow IV Flow Rate well-known fact: flow rate is a function of various parameters,1–4 yet they also underscore the importance of critically evaluating modifications to IV fluid sets before wholesale To the Editor adoption. Our aim is not to discourage the use of PNACs, but ur hospital recently added pressure-rated needleless access connectors (PNACs) to all of our IV fluid sets to facilitate postoperative withdrawal of blood and aMaxPlus® Clear (Maximus Medical, Ontario, CA). O b ® medication administration. Shortly thereafter, some of our LifeShield Bifurcated Blood Set (Hospira, Morgan Hill, CA) and Clearlink Extension Set (Baxter, Deerfield, IL). anesthesiologists noted decreased flow rate and asked just cAvailable at: http://www.maximusmedical.com/pdf/ML3084FlowRate how much fluid administration was diminished by these Performance.pdf. Accessed May 27, 2010.

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1077 Letters to the Editor

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1077 LETTERS TO THE EDITOR rather to raise awareness of the flow impedance caused by these (and other) IV devices.

ACKNOWLEDGMENTS Lawrence J. Saidman, MD, is acknowledged for his guidance in the preparation of this letter. Jorge A. Caballero, MD Stanford University School of Medicine Stanford, California Frain Rivera, MD Joshua Edwards, MD John G. Brock-Utne, MD, PhD Department of Anesthesia Stanford University School of Medicine Stanford, California [email protected]

REFERENCES 1. Brown N, Duttchen KM, Caveno JW. An evaluation of flow rates of normal saline through peripheral and central venous catheters. American Society of Anesthesiologists Annual Meet- ing, Orlando. Anesthesiology 2008:A1484 2. Andersen HW, Benumof JL, Trousdale FR, Ozaki GT. Increasing the functional gauge on the side port of large catheter sheath introducers. Anesthesiology 1982;56:57–9 3. Benumof JL, Trousdale FR, Alfery DD, Ozaki GT. Larger Figure 1. Arterial anatomy of the hand. catheter sheath introducers and their side port functional gauge. Anesth Analg 1981;60:216–7 4. Benumof JL, Wyte SR, Rogers SN. A large catheter sheath introducer with an increased side-port functional gauge. Crit Care Med 1983;11:660–2 DOI: 10.1213/ANE.0b013e3181f0948c

Anatomic Snuffbox Radial Artery Cannulation To the Editor ailure of radial artery cannulation at the wrist, although uncommon, is not infrequent and is sec- Fondary to vasospasm, hematoma formation, and intimal dissection or thickening. We describe an alternative successful ultrasound-guided approach to dorsal radial artery cannulation at the “anatomic snuffbox.” A 55-year-old woman, with a medical history significant Figure 2. Semiprone position of the hand for cannulation. for diabetes mellitus, coronary artery disease (postcoro- nary artery bypass surgery), and end-stage renal disease receiving hemodialysis via a left arm arteriovenous cannulation site in the anatomic snuffbox is distal to the fistula, was scheduled for renal transplant. After induc- division of the radial artery that provides collateral flow to the tion of general anesthesia, several attempts to percuta- hand through the superficial palmar (volar) arch (Fig. 1).2 neously insert a right radial artery catheter failed be- Cannulation distal to that separation would be expected to cause of hematoma formation. Cannulation of the reduce the potential for digital ischemia. The dorsal radial brachial and axillary arteries of the right arm was artery does provide arterial flow to the deep palmar arch and considered but might have precluded future surgical cannulation of the same may be preferable to a proximal site, fistula formation. We decided not to cannulate the ulnar which could interfere with both the deep and superficial artery because of the radial artery hematoma. The radial palmar arch flows. For successful cannulation, we recom- artery was, however, palpable as the dorsal radial artery mend semiprone position of hand (Fig. 2) and small-sized in the anatomic snuffbox. Using ultrasound guidance, catheters. To conclude, cannulation of the dorsal radial artery the diameter of the dorsal radial artery was measured at is a viable alternative to the radial artery at the wrist and other 2.2 mm and a 22-gauge (0.9-mm outer diameter) cannula possible arterial access sites. was inserted into the snuffbox radial artery. In 1982, Pyles et al.1 published their clinical experi- Krishnaprasad Deepika, MD ence with cannulation of the dorsal radial artery. The Dhamodaran Palaniappan, MD

1078 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA LETTERS TO THE EDITOR rather to raise awareness of the flow impedance caused by these (and other) IV devices.

1078 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA Letters to the Editor

Thomas Fuhrman, MD has been brought to our attention that some of the data that Bruce Saltzman, MD we reported have been included in a series of 4 cases Department of Anesthesiology, Perioperative Medicine and published earlier this year.2 Pain Management We were not aware of this publication at the time our Jackson Memorial Hospital manuscript was accepted for publication and therefore we University of Miami did not cite it in our case report. Therefore, our communi- Miami, Florida cation was inaccurate inasmuch as it was not the second [email protected] time it has been used in an emergency. We would empha- REFERENCES size that the use of the device as described by Strueber et 2 1. Pyles ST, Scher KS, Vega ET, Harrah JD, Rubis LJ. Cannulation al. is a novel application. Using the Novalung for right of the dorsal radial artery: a new technique. Anesth Analg ventricular support mandates central placement as de- 1982;61:876–8 scribed, regardless of the size of the patient. 2. Gray H, Lewis WH. Gray’s Anatomy of the Human Body. 20th US ed. Philadelphia: Lea & Febiger, originally published in 1918 DOI: 10.1213/ANE.0b013e3181ef343a Helen Holtby, MB, BS, FRCPC Director of Cardiac Anesthesia Hospital for Sick Children Correction to the Case Report Toronto, Ontario, Canada “Emergency Interventional Lung [email protected] Assist for Pulmonary Hypertension” REFERENCES Anesth Analg 2009;109:382–5 1. Taylor K, Holtby H. Emergency interventional lung assist for pulmonary hypertension. Anesth Analg 2009;109:382–5 To the Editor 2. Strueber M, Hoeper MM, Fischer S, Cypel M, Warnecke G, Gottlieb J, Pierre A, Welte T, Haverich A, Simon AR, Keshavjee e recently described our experience of the anes- S. Bridge to thoracic organ transplantation in patients with thetic management of a patient using an emer- pulmonary arterial hypertension using a pumpless lung assist Wgency lung assist device.1 Although the above device. Am J Transplant 2009;9:853–7 report focused on the anesthetic considerations for that case, it DOI: 10.1213/ANE.0b013e3181ec2bea

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1079 BOOK, MULTIMEDIA, AND MEETING REVIEWS Section Editor: Paul F. White

An Introductory Curriculum for schematics detailing phenomena such as reverberation artifact. The high-quality images and illustrations are a major Ultrasound-Guided Regional Anesthesia strength of the book. Brian A. Pollard Vincent W. S. Chan; illustrations by Each chapter reads easily, yet with only 81 pages and Diana Kryski; photographs by Michele Dalgarno. Toronto: lacking references, this is not meant to be a comprehensive B. A. Pollard, 2010. ISBN: 978-0-7727-8735-4. $95.00. textbook on the topic. Rather, it nicely complements existing resources, and as such represents one of the finest orty years after Dr. Alon Winnie described the paresthe- presentations on basic UGRA concepts available to practi- Fsia technique for interscalene blockade, evolving tech- tioners. The emphasis on competency with imaging is nology enables clinicians to see and differentiate anatomic appropriate and is reflected in the fact that the bulk of the structures in real time. The growing interest in ultrasound- text material addresses this skill set. Specific clinical recom- guided regional anesthesia (UGRA) has been an exciting, mendations, such as type of local anesthetic and volume landmark development in the practice of regional anesthe- selection, are not emphasized. sia and pain medicine. The authors’ of An Introductory As interest in the use of UGRA in clinical practice and Curriculum for Ultrasound-Guided Regional Anesthesia, Drs. education continues to expand, efforts to optimize patient Pollard and Chan, are well-known and accomplished lead- safety and minimize risk should be at the forefront. Al- ers in this emerging field. though no evidence exists to support this claim, one could With this new technology comes responsibility. Spirited reasonably hypothesize that poor UGRA technique may debate now surrounds the definition of core competencies, increase the risk of associated complications. To that end, training requirements, competency assessment, institu- this text is an essential resource for UGRA educators and tional certification for UGRA practice, and strategies for would be an excellent addition to all libraries on regional quality improvement in UGRA. In writing these introduc- anesthesia. tory curriculums on UGRA, the authors have appropriately side-stepped these issues and focused instead on a more Jonathan C. Beathe, MD important founding principle for their text—patient safety. Gregory A. Liguori, MD As is stated by the authors, the goal in writing this book Weill Medical College of Cornell University was “to create an educational curriculum that will permit a Hospital for Special Surgery self-directed foundational path for clinicians, from novice New York, New York to expert, community and academic-based, to build on [email protected] existing regional anesthesia expertise by integrating essential ultrasound techniques into daily practice.” They accom- plish this task through nine chapters in five well-organized Visionaries and Dreamers: The sections. Story of Founding Fathers of Section 1 (39 pages) is the largest and constitutes almost Anesthesiology in Israel half of the text. Starting with physics as applied to ultrasonography, this section then confronts the fundamentals Gabrial M. Gurman. Beer-Sheva, Israel: Ben-Gurion Uni- of ultrasound scanning and needle techniques. Sections 2, 3, versity of the Negev Press, 2008. ISBN: 978-9-6534- and 4 categorize nerve blocks into introductory, intermedi- 2963-5. 228 pages. $16.53. ate, and advanced designations. Chapters within these sections cover the most commonly performed ultrasound- isionaries and Dreamers is a series of accounts about guided nerve blocks. Finally, section 5 describes the use of Vpioneers of Israeli anesthesiology. The author, Gabriel UGRA to assist with conventionally performed epidural M. Gurman, MD, is a Professor Emeritus of Anesthesiology and subarachnoid blocks. and Critical Care at Ben-Gurion University of the Negev, An ubiquitous debate continues among practitioners as and both Hebrew and English translations were edited by to what constitutes an “introductory” block versus an Lior Granot. Dr. Gurman provides the details of the lives of “advanced” block, which can be seen as a potential weak- innovative anesthesiologists through a series of interviews ness in this presentation. For example, many clinicians may and is able to capture compelling stories of adversity and not find a selective block of the radial nerve “easier” than a eventual triumphs. As the lives of these 12 anesthesiolo- conventional ultrasound-guided approach to the brachial gists are described, the reader is pulled into what seems plexus. like an almost “sacred circle of leaders.” As an anesthesi- Each chapter ends with a salient summary, as well as 4 ologist, it was helpful to see how the book breaks down the to 6 key references for the self-directed learner. The funda- elements of their leadership, and sheds light on each of mental learning points of the chapter are then listed as these qualities separately: the importance of building a useful “knowledge keys.” Esthetics were not overlooked in vision, the value of being bold and unconstrained by the production of the text, which presents superior illustra- established rules, and having the foresight to build a solid, tions and an excellent collection of images. Often, images trustworthy team. These are invaluable lessons that would and illustrations are combined into impressive and thoughtful help the professional careers of all aspiring anesthesiologists.

1080 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 BOOK, MULTIMEDIA, AND MEETING REVIEWS Section Editor: Paul F. White

1080 www.anesthesia-analgesia.org October 2010 • Volume 111 • Number 4 Books, Multimedia, and Meeting Reviews

Dr. Gurman began his career with a generation of There are many stories of success and significant ad- physicians who had already survived a horrific time. Many vances in clinical care in the field of anesthesiology, most of of Dr. Gurman’s friends and colleagues narrowly escaped which were achieved through hard work, perseverance, the front lines of the extermination camps of World War II dedication, and an unwillingness to succumb to the “status through unimaginable acts of courage. These physicians quo.” However, anesthesia was far from ideal in Israel, were in some cases the only survivors of their entire especially from an academic and professional standpoint. families. Most of them migrated to the new state of Israel This is illustrated by Dr. Shamay Cotev, head of the first shortly after its establishment in1948. In the beginning, the intensive care unit (ICU) in Israel, who is quoted in the scarcity of anesthesiologists in the new State of Israel meant book as having said, “In retrospect, I wouldn’t have gone that one anesthesiologist might cover an entire hospital. into anesthesia.” This is perhaps the only part of this bright This is epitomized in the accounts of Dr. Thomas Gesztes, and optimistic book that touches on the harsh reality that who had his own incarcerated hernia repaired under local much more work is still needed—in anesthesia and in anesthesia, and who afterwards anesthetized his patients medicine. Currently, Ͻ2% of the graduates of medical while taking call duties the same night as his own surgery. schools in Israel choose to pursue postgraduate training in This is an example of the ultimate “ambulatory” surgery anesthesiology! experience! Each one of these anesthesiologists was instru- In this well-written book, Dr. Gurman takes the reader mental in building departments (and societies) of anesthe- into the middle of Israeli anesthesia history. The tremen- sia from “the ground up.” When Dr. Gurman migrated to Israel in 1972, anesthesia dous strides of the past 50 years are apparent, but the clinical practices were still in a rapid flux from what summit is yet to be achieved. Dr. Gurman’s book provides seemed archaic practices. The transformation was nothing us with direction regarding where we need to be in the short of miraculous, and clinical anesthesia care is now future. This book is dedicated to the 60th anniversary of the considered among the most advanced specialties in the State of Israel, and is recommended reading for young and field of medicine. In his book, Dr. Gurman interrupts his old anesthesiologists not only for its historical value, but storytelling periodically to offer vignettes and historical more importantly, for the optimistic glimpse it provides commentaries on the various personalities and their per- into the future of anesthesiology, medicine, and humanity. sonal and professional stories. Thus, the action of the book does not proceed chronologically; instead, it moves back Sorin J. Brull, MD and forth in time, depending on which colleague he is Joseph A. Cartwright, MD memorializing. This strategy of tying present-day individuals Mayo Clinic College of Medicine to events in the past is very helpful to those readers who may Jacksonville, Florida not be very familiar with the geography and history of Israel. [email protected]

October 2010 • Volume 111 • Number 4 www.anesthesia-analgesia.org 1081