Preventing Postoperative Pulmonary Complications the Role of the Anesthesiologist David O

1467

Ⅵ CLINICAL CONCEPTS AND COMMENTARY

Richard B. Weiskopf, M.D., Editor

Anesthesiology 2000; 92:1467–72 © 2000 American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc. Preventing Postoperative Pulmonary Complications The Role of the Anesthesiologist David O. Warner, M.D.*

THE Confederate General “Stonewall” Jackson was one widely. Nonetheless, it is still clear that PPCs occur of the earliest known victims of a respiratory complica- relatively frequently. In studies of noncardiac surgery, tion after surgery, dying of pneumonia 10 days after an the frequency of PPCs and cardiac complications (which otherwise successful ether anesthetic in 1863. Despite historically have attracted more attention from the anes- subsequent advances in anesthesia and surgical care, thesia community) are comparable.1 For example, in a postoperative pulmonary complications (PPCs) still are a series of adult men undergoing elective abdominal sur- significant problem in modern practice. This commen- gery, PPCs occurred significantly more frequently than tary examines why PPCs occur and how the anesthesi- cardiac complications (estimated rates of 9.6% and 5.7%, ologist can help prevent them. respectively) and were associated with significantly longer hospital stays.1

Significance of Perioperative Pulmonary Causes of Perioperative Pulmonary Complications Complications Determination of the frequency and clinical impact of A basic understanding of mechanism guides rational PPCs in modern practice is hampered by the lack of a practice. Many PPCs, such as atelectasis and pneumonia, uniform definition of a PPC among studies. Nearly all seem to be related to disruption of the normal activity of investigators include in this definition pneumonia (defi- the respiratory muscles, disruption that begins with the nite or suspected), respiratory failure (usually defined as induction of anesthesia and that may continue into the the need for mechanical ventilatory support), and bron- postoperative period. Breathing is a complex behavior chospasm. Others include unexplained fevers, excessive requiring the coordinated activity of several muscle bronchial secretions, abnormal breath sounds, “produc- groups, both in the upper airway and in the chest wall. tive” cough, atelectasis (often not defined), and hypox- Anesthetics and many other drugs used in the perioper- emia. Even within these categories, definitions vary ative period affect the central regulation of breathing, changing the neural drive to respiratory muscles such as * Professor of Anesthesiology. the diaphragm. At high doses, anesthetics attenuate the Received from the Mayo Medical School, Mayo Clinic, Rochester, activities of all respiratory muscles. However, at moder- Minnesota. Submitted for publication September 8, 1999. Accepted for ate depths of anesthesia, anesthetics may produce respi- publication December 2, 1999. Supported in part by grants no. GM- ratory depression by altering the distribution and timing 40909 and HL-45532 from the National Institutes of Health, Bethesda, of neural drive to the respiratory muscles, rather than by Maryland, and by the Mayo Foundation, Rochester, Minnesota. producing a global depression of activity. For example, Address reprint requests to Dr. Warner: Mayo Medical School, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905. Address at 1.2 minimum alveolar concentration, halothane anes- electronic mail to: [email protected] thesia depresses activity in some respiratory muscles Key words: Atelectasis; epidural analgesia; pneumonia; respiratory (such as the parasternal intercostals) but actually in- failure; respiratory muscles. creases activity in others (such as the transversus abdo- 2 The illustrations for this section are prepared by Dimitri Karetnikov, minis). Thus, perioperative respiratory muscle dysfunc- 7 Tennyson Drive, Plainsboro, New Jersey 08536. tion is, in some cases, more a matter of a lack of

Anesthesiology, V 92, No 5, May 2000 1468

DAVID O. WARNER

Fig. 1. Model showing how incoordination of respiratory muscles impairs lung function. The position of the midpoint of a horizontal bar, suspended between ﬁxed surfaces by inspiratory and expiratory muscles, represents lung volume as denoted on a scale from low (residual volume, RV) to high (total lung capacity, TLC) volumes. During awake, coordinated inspiration (lower left), the bar remains horizontal (representing normal chest wall expansion), and lung volume changes efﬁciently. When anesthetized, muscle activity becomes incoordinated, such that the bar tilts during inspiration (representing chest wall distortion), impairing lung expansion. Incoordination continues into the postoperative period after thoracic and abdominal surgery. Dashed lines in lower panels denote end- expiratory position of the bar.

coordination than a lack of overall activity. As with other nisms (fig. 2). First, functional disruption of respiratory complex systems, this lack of coordination reduces effi- muscles (such as the intercostal or abdominal muscles) ciency (fig. 1), in this instance producing hypoventila- by incisions, even after surgical repair, may impair their tion. In addition, deformation of the chest wall alters the effectiveness. Second, postoperative pain may cause vol- underlying lung, decreasing the functional residual ca- untary limitation of respiratory motion. Finally, stimula- pacity and producing atelectasis in dependent lung re- tion of the viscera, such as provided by mechanical gions. These regions of atelectasis develop in nearly all traction on the gallbladder or esophageal dilation, mark- patients after a few minutes of anesthesia and may sig- edly decreases phrenic motoneuron output and changes nificantly impair pulmonary gas exchange.3 Chest wall the activation of other respiratory muscles, generally distortion and atelectasis also occur when the respira- acting to minimize diaphragmatic descent. These effects tory muscles are inactive during mechanical ventilation are only partially attenuated by vagotomy, suggesting and persist even with positive end-expiratory pressure. that multiple afferent pathways mediate this reflex.4 These intraoperative changes in the pattern of breath- Thus, like anesthesia, surgical trauma can also disrupt ing can persist in the postoperative period as additional normal coordination of respiratory muscle action, lead- effects of surgical trauma come into play. Residual anes- ing to persistent decreases in functional residual capacity thetic effect may also contribute to postoperative and vital capacity, with lung atelectasis that can last for changes in breathing, although the importance of this several days after surgery. The clinical impression is that factor has not been studied. The effects of surgical this atelectasis leads to pneumonia, although this pro- trauma are most pronounced after thoracic and abdom- gression has not been conclusively shown. These post- inal surgery, and they arise from at least three mecha- operative changes in pulmonary function can be partially

Anesthesiology, V 92, No 5, May 2000 1469

POSTOPERATIVE PULMONARY COMPLICATIONS

Fig. 2. Factors producing respiratory muscle dysfunction after surgical trauma. From left to right: (1) surgical trauma stimulates central nervous system (CNS) reflexes mediated by both visceral and somatic nerves that produce reflex inhibition of the phrenic and other nerves innervating respiratory muscle; (2) mechanical disruption of respiratory muscles impairs efficiency; and (3) pain produces voluntary limitation of respiratory motion. These factors all tend to reduce lung volumes and can produce hypoventilation and atelectasis.

ameliorated by using endoscopic techniques to mini- barotrauma and gas exchange abnormalities. Anesthetic mize surgical trauma.5 However, because procedures gasses and tracheal intubation may impair normal muco- such as laparoscopic cholecystectomy still stimulate ab- ciliary transport. Recent studies suggest that prolonged dominal viscera (e.g., through gallbladder traction), pul- anesthesia and surgery may impair the function of lung monary mechanics are still affected; this is probably also inflammatory cells, which could increase susceptibility true for thoracoscopy. to postoperative infections.6 Finally, other respiratory Other factors may also contribute to PPCs. Reflex stim- outcomes are related specifically to surgical or anes- ulation during airway instrumentation and release of thetic interventions, such as acute lung injury after car- inflammatory mediators by drug administration can pro- diopulmonary bypass, pneumothorax caused by baro- duce bronchoconstriction. This increased airway resis- trauma or surgical trauma, negative pressure pulmonary tance limits especially expiratory gas flow from the lung, edema after airway obstruction during spontaneous which, if severe, can produce hyperinflation with risk of breathing, and aspiration pneumonitis.

Anesthesiology, V 92, No 5, May 2000 1470

DAVID O. WARNER

Assessment of Risk factor for PPCs. The optimal timing of quitting is not known and should be the focus of future investigations. Consistent risk factors for PPCs among extant studies Of interest, some studies find that recent cessation or include surgical site (with thoracic and abdominal sur- reduction of smoking (within approximately 2 months gery posing the highest risk), smoking, and the presence before surgery) may actually increase the risk of PPCs.11 of pulmonary disease. Although the results of pulmonary However, these studies are not conclusive because of function testing, including measurements of arterial selection bias, with the sickest patients being more likely blood gasses, have proved useful in predicting pulmo- to reduce smoking. The long-term benefits of smoking nary function after lung resection surgery, they do not 7 cessation to the patient suggest that the anesthesiologist predict PPCs. For example, the degree of airway ob- should seize the preoperative period as an opportune struction, as assessed by the forced expiratory volume in time to encourage quitting, recognizing that the mini- 1 s, is not a significant independent risk factor for the mum duration required to demonstrate benefit is un- development of postoperative respiratory failure after known. abdominal surgery, even in smokers with severe lung 8 Beyond these measures, the single most important disease. Thus, pulmonary function tests should be preoperative intervention is to educate patients regard- viewed as a management tool to optimize preoperative ing the proper performance of maneuvers designed to pulmonary function as appropriate, not as a means to increase lung volumes, which are of proven benefit in assess risk. For example, spirometry or peak flows may the postoperative period (as discussed in the following be useful to monitor the status of a patient with asthma, section). especially the small subset of patients who have difficulty perceiving the status of their disease. Otherwise, routine preoperative pulmonary function testing may be Prevention of Perioperative Pulmonary a waste of resources.9 Complications

Preparation of Patients with Lung Disease for One goal of management is to prevent significant bron- Surgery choconstriction. The reflex irritation produced by laryngoscopy and tracheal intubation is best avoided, if pos- Schemes to prepare patients with lung disease for sible, in patients with reactive airways, although the surgery have included prolonged preoperative hospital majority of these patients, even those with severe admission, inhaled ␤-adrenergic agonists, chest physio- chronic obstructive pulmonary disease, can tolerate this therapy, antibiotics, intravenous aminophylline, and hy- if necessary.8 A variety of measures, especially the pre- dration. Proposed mechanisms of benefit include im- induction administration of inhaled ␤-adrenergic ago- provements in respiratory physiology and elimination of nists and muscarinic antagonists, can effectively attenu- any underlying pulmonary infections. However, there ate airway reflex responses to tracheal intubation. have been no controlled trials demonstrating that these Another approach is to avoid airway manipulation alto- regimens improve outcome in patients with lung dis- gether by using regional anesthesia when feasible, al- ease. Furthermore, some of their components pose risk though whether the use of these techniques actually (e.g., aminophylline) or are impractical in current prac- improves respiratory outcomes is unproven. As a practi- tice (e.g., preoperative hospital admission). Although cal matter, some patients with severe respiratory disease such schemes should not be routinely applied to all use accessory muscles of breathing (such the abdominal patients, preoperative lung function should be opti- muscles) and cannot maintain spontaneous breathing if mized as tailored to the needs of the individual patient. these muscles are paralyzed by techniques such as epi- For example, symptoms should be optimally controlled dural or subarachnoid block. in patients with asthma (usually by ensuring that airway Another goal that has attracted much interest over inflammation is minimized with topical or systemic cor- recent years is to optimize postoperative pulmonary ticosteroids) because patients with more active symp- function by improving postoperative analgesia. In partic- toms have a higher frequency of PPCs.10 However, there ular, regional analgesic techniques have the potential to is no firm evidence that all patients with asthma should improve two of the three mechanisms discussed previ- receive routine systemic corticosteroids.10 All patients ously (fig. 2) that produce postoperative respiratory should be encouraged to quit smoking, a significant risk muscle dysfunction: pain and reflex inhibition of respi-

Anesthesiology, V 92, No 5, May 2000 1471

POSTOPERATIVE PULMONARY COMPLICATIONS

ratory muscles, if the afferent reflex limb can indeed be ative analgesic therapies on pulmonary outcome after a targeted. The expectation is that these effects will im- wide variety of surgical procedures, including abdominal prove respiratory muscle function, increase lung expan- and thoracic surgery.15 In this study, postoperative epi- sion, and prevent PPCs such as atelectasis and pneumo- dural opioids significantly decreased the frequency of nia. Is this expectation true? atelectasis, but not other pulmonary complications, There is no doubt that regional analgesic techniques, at when compared with systemic opioids. Epidural local least those using local anesthetics, can significantly alter anesthetics decreased the incidence of pulmonary infec- postoperative respiratory muscle function. In general, tions and pulmonary complications overall when com- techniques such as segmental epidural blockade with pared with systemic opioids. However, the individual local anesthetics can increase tidal volume and vital studies examined in these analyses were often beset by capacity and improve indices interpreted to reflect dia- the problems noted previously (and others), making phragm activity after thoracic and upper abdominal sur- interpretation of meta-analyses problematic. For exam- gery.12 These effects have been attributed to decreased ple, consider the conclusion that epidural local anesthet- pain and interruption of the afferent limb of reflex dia- ics significantly decrease the rate of pulmonary compli- phragmatic inhibition. However, local anesthetic block cations compared with systemic opioids (controls). Four also can paralyze other respiratory muscles such as the of the eight studies used in the meta-analysis to support intercostal and abdominal muscles, which itself will this conclusion reported a very high frequency of atel- change the pattern of breathing and makes it difficult to ectasis/pneumonia in control groups (60–70%), far interpret various measures of diaphragmatic function.13 greater than the other four studies (which found no Furthermore, blockade may not affect reflex inhibition benefit) and far greater than more recent series,8 making mediated by afferent information carried in nerves such their applicability to current practice questionable. as the phrenic or vagus. In the only direct measurements In one of the few studies that successfully accounts for of postoperative diaphragmatic shortening in human many of these problems and that represents the largest subjects, thoracic epidural blockade after thoracotomy trial to date, Jayr et al.16 examined patients undergoing did not change diaphragm electromyogram or shorten- abdominal cancer surgery randomized to receive either ing, and, in fact, produced paradoxical lengthening dur- continuous epidural bupivacaine and morphine or sub- ing inspiration in half of the patients studied, despite cutaneous morphine infusion via a catheter that simu- improving many more global indices of respiratory func- lated epidural placement. They found that although the tion such as tidal volume.14 Thus, the effect of postop- epidural provided excellent postoperative analgesia, su- erative regional analgesia on the pattern of breathing perior to that afforded by the subcutaneous morphine may be complex and have unintended consequences. infusion, it did not affect the frequency of PPCs as Do these changes in breathing pattern and spirometric carefully defined and prospectively evaluated, either in values produced by regional analgesia translate to im- patients with normal or abnormal lungs. Several other proved clinical outcomes? This straightforward question good studies have also failed to demonstrate significant has proved difficult to answer for a variety of reasons. As clinical benefit of regional techniques. Thus, although previously discussed, there is often little agreement regional techniques can provide excellent analgesia among investigators as to what constitutes a PPC. Pa- when properly applied, in my opinion, it is not yet clear tients and investigators are seldom blinded for tech- that they consistently improve clinical respiratory out- nique, perhaps introducing bias favoring regional anal- comes. gesia. Patient populations, surgical technique, and Postoperative maneuvers to increase mean lung vol- analgesic regimens are heterogeneous, even within stud- umes are of proven benefit in preventing PPCs. Presum- ies. For example, some studies use epidural analgesia ably these techniques increase lung expansile forces and intraoperatively, and others do not. The application of discourage atelectasis. Several methods have been stud- postoperative maneuvers to increase lung volume, ied, including intermittent positive pressure breathing, which are of clear benefit, is often inconsistent or not deep-breathing exercises, incentive spirometry, and reported. Finally, because major PPCs occur with rela- chest physiotherapy. Critical review, as well as a recent tively low frequency, studies often examine insufficient meta-analysis, suggests that all regimens studied are numbers of patients to make conclusions. equally efficacious in reducing the frequency of PPCs (by In an attempt to overcome the latter problem, a recent approximately a factor of two compared with no ther- study reported meta-analyses of the effects of postoper- apy) after upper abdominal surgery.17 Currently, incen-

Anesthesiology, V 92, No 5, May 2000 1472

DAVID O. WARNER

tive spirometry enjoys popularity because it is simple, G, Reiz S: Diaphragmatic activity after laparoscopic cholecystectomy. inexpensive, and provides objective goals for and mon- ANESTHESIOLOGY 1999; 91:406–13 itoring of patient performance. Although there is little 6. Kotani N, Hashimoto H, Sessler DI, Kikuchi A, Suzuki A, Taka- hashi S, Muraoka M, Matsuki A: Intraoperative modulation of alveolar evidence that instituting these measures preoperatively macrophage function during isoflurane and propofol anesthesia. ANES- benefits patient physiology, preoperative education may THESIOLOGY 1998; 89:1125–32 improve patient compliance and performance in the 7. Lawrence VA, Page CP, Harris GD: Preoperative spirometry be- postoperative period. Considering that this intervention fore abdominal operations: A critical appraisal of its predictive value. is the only one proven to reduce the frequency of PPCs, Arch Intern Med 1989; 149:280–5 8. Warner DO, Warner MA, Offord KP, Schroeder DR, Maxson P, both patients and staff must be educated regarding its Scanlon PD: Airway obstruction and perioperative complications in importance. smokers undergoing abdominal surgery. ANESTHESIOLOGY 1999; PPCs are an important clinical problem in modern 90:372–9 practice, rivaling perioperative cardiac morbidity in fre- 9. De Nino LA, Lawrence VA, Averyt EC, Hilsenbeck SG, Dhanda R, quency and severity in some settings. It seems that many Page CP: Preoperative spirometry and laparotomy: Blowing away dol- PPCs are caused by the disruption of the normal coordi- lars. Chest 1997; 111:1536–41 10. Warner DO, Warner MA, Barnes RD, Offord KP, Schroeder DR, nation of respiratory muscle by anesthesia and surgical Gray DT, Yunginger JW: Perioperative respiratory complications in trauma. Although the use of postoperative regional an- patients with asthma. ANESTHESIOLOGY 1996; 85:460–7 algesia can partially improve respiratory muscle func- 11. Bluman LG, Mosca L, Newman N, Simon DG: Preoperative tion, there is no convincing evidence that regional tech- smoking habits and postoperative pulmonary complications. Chest niques decrease the frequency of clinically significant 1998; 113:883–9 PPCs. Rather, maneuvers to encourage deep breathing, 12. Pansard JL, Mankikian B, Bertrand M, Kieffer E, Clergue F, Viars P: Effects of thoracic extradural block on diaphragmatic electrical such as incentive spirometry, are of proven benefit and activity and contractility after upper abdominal surgery. ANESTHESIOL- should be the focus of preventive efforts. In an era of OGY 1993; 78:63–71 relentless high technology, it is perhaps comforting that 13. Duggan JE, Drummond GB: Abdominal muscle activity and in- such a simple technique should prove so effective. traabdominal pressure after upper abdominal surgery. Anesth Analg 1989; 69:598–603 14. Fratacci MD, Kimball WR, Wain JC, Kacmarek RM, Polaner DM, Zapol WM: Diaphragmatic shortening after thoracic surgery in humans: References Effects of mechanical ventilation and thoracic epidural anesthesia. ANESTHESIOLOGY 1993; 79:654–65 1. Lawrence VA, Hilsenbeck SG, Mulrow CD, Dhanda R, Sapp J, 15. Ballantyne JC, Carr DB, deFerranti S, Suarez T, Lau J, Chalmers Page CP: Incidence and hospital stay for cardiac and pulmonary com- TC, Angelillo IF, Mosteller F: The comparative effects of postoperative plications after abdominal surgery. J Gen Intern Med 1995; 10:671–8 analgesic therapies on pulmonary outcome: Cumulative meta-analyses 2. Warner DO, Warner MA: Human chest wall function while awake of randomized, controlled trials. Anesth Analg 1998; 86:598–612 and during halothane anesthesia. ANESTHESIOLOGY 1995; 82:20–31 16. Jayr C, Thomas H, Rey A, Farhat F, Lasser P, Bourgain JL: 3. Tokics L, Hedenstierna G, Strandberg A, Brismar B, Lundquist H: Postoperative pulmonary complications: Epidural analgesia using bu- Lung collapse and gas exchange during general anesthesia: Effects of pivacaine and opioids versus parenteral opioids. ANESTHESIOLOGY 1993; spontaneous breathing, muscle paralysis, and positive end-expiratory 78:666–76 pressure. ANESTHESIOLOGY 1987; 66:157–67 17. Thomas JA, McIntosh JM: Are incentive spirometry, intermittent 4. Ford GT, Grant DA, Rideout KS, Davison JS, Whitelaw WA: Inhi- positive pressure breathing, and deep breathing exercises effective in bition of breathing associated with gallbladder stimulation in dogs. the prevention of postoperative pulmonary complications after upper J Appl Physiol 1988; 65:72–9 abdominal surgery? A systematic overview and meta-analysis. Phys 5. Sharma RR, Axelsson H, Oberg A, Jansson E, Clergue F, Johansson Ther 1994; 74:3–10

Anesthesiology, V 92, No 5, May 2000 Part 10.5: Near-Fatal Asthma Circulation 2005;112;139-142; originally published online Nov 28, 2005; DOI: 10.1161/CIRCULATIONAHA.105.166567 Circulation is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX 72514 Copyright © 2005 American Heart Association. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/cgi/content/full/112/24_suppl/IV-139

Subscriptions: Information about subscribing to Circulation is online at http://circ.ahajournals.org/subsriptions/

Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, 351 West Camden Street, Baltimore, MD 21202-2436. Phone 410-5280-4050. Fax: 410-528-8550. Email: [email protected]

Reprints: Information about reprints can be found online at http://www.lww.com/static/html/reprints.html

Downloaded from circ.ahajournals.org by on January 27, 2006 Part 10.5: Near-Fatal Asthma

sthma accounts for Ͼ2 million emergency department Primary Therapy Avisits and 5000 to 6000 deaths annually in the United Oxygen States, many occurring in the prehospital setting.1 Severe Provide oxygen to all patients with severe asthma, even those asthma accounts for approximately 2% to 20% of admissions with normal oxygenation. Titrate to maintain SaO2 Ͼ92%. As to intensive care units, with up to one third of these patients noted above, successful treatment with ␤-agonists may ini- requiring intubation and mechanical ventilation.2 This section tially cause a decrease in oxygen saturation because the focuses on the evaluation and treatment of patients with resultant bronchodilation may initially increase the near-fatal asthma. ventilation-perfusion mismatch. ␤ Pathophysiology Inhaled 2-Agonists The pathophysiology of asthma consists of 3 key Albuterol (or salbutamol) provides rapid, dose-dependent abnormalities: bronchodilation with minimal side effects. Because the administered dose depends on the patient’s lung volume and ● Bronchoconstriction inspiratory flow rates, the same dose can be used in most ● Airway inflammation patients regardless of age or size. Although 6 adult studies5 ● Mucous impaction and 1 pediatric study6 showed no difference in the effects of Complications of severe asthma, such as tension pneumo- continuous versus intermittent administration of nebulized thorax, lobar atelectasis, pneumonia, and pulmonary edema, albuterol, continuous administration was more effective in the can contribute to fatalities. Cardiac causes of death are less subset of patients with severe exacerbations of asthma,7,8 and common. it was more cost-effective in a pediatric trial.6 A Cochrane meta-analysis showed no overall difference between the Clinical Aspects of Severe Asthma effects of albuterol delivered by metered dose inhaler (MDI)- Wheezing is a common physical finding, but severity does spacer or nebulizer,9 but MDI-spacer administration can be not correlate with the degree of airway obstruction. The difficult in patients in severe distress. The typical dose of absence of wheezing may indicate critical airway obstruction, albuterol by nebulizer is 2.5 or 5 mg every 15 to 20 minutes whereas increased wheezing may indicate a positive response intermittently or continuous nebulization in a dose of 10 to 15 to bronchodilator therapy. mg/h. Oxygen saturation (SaO2) levels may not reflect progressive Levalbuterol is the R-isomer of albuterol. It has recently alveolar hypoventilation, particularly if O2 is being adminis- become available in the United States for treatment of acute tered. Note that the SaO2 may initially fall during therapy asthma. Some studies have shown equivalent or slight im- ␤ because -agonists produce both bronchodilation and vaso- provement in bronchodilation when compared with albuterol dilation and may initially increase intrapulmonary shunting. in the emergency department.10 Further studies are needed Other causes of wheezing are pulmonary edema, chronic before a definitive recommendation can be made. obstructive pulmonary disease (COPD), pneumonia, anaphylaxis,3 foreign bodies, pulmonary embolism, bronchiectasis, Corticosteroids and subglottic mass.4 Systemic corticosteroids are the only proven treatment for the inflammatory component of asthma, but the onset of their Initial Stabilization anti-inflammatory effects is 6 to 12 hours after administra- Patients with severe life-threatening asthma require urgent tion. A comprehensive search of the literature by the Coch- and aggressive treatment with simultaneous administration of rane approach (including pediatric and adult patients) deter- oxygen, bronchodilators, and steroids. Healthcare providers mined that the early use of systemic steroids reduced rates of must monitor these patients closely for deterioration. Al- admission to the hospital.11 Thus, providers should administer though the pathophysiology of life-threatening asthma con- steroids as early as possible to all asthma patients but should sists of bronchoconstriction, inflammation, and mucous im- not expect effects for several hours. Although there is no paction, only bronchoconstriction and inflammation are difference in clinical effects between oral and intravenous amenable to drug treatment. If the patient does not respond to (IV) formulations of corticosteroids,12 the IV route is prefer- therapy, consultation or transfer to a pulmonologist or inten- able because patients with near-fatal asthma may vomit or be sivist is appropriate. unable to swallow. A typical initial adult dose of methylprednisolone is 125 mg (dose range: 40 to 250 mg). Incorporation or substitution of inhaled steroids into this (Circulation. 2005;112:IV-139-IV-142.) scheme remains controversial. A Cochrane meta-analysis of 7 © 2005 American Heart Association. randomized trials (4 adult and 3 pediatric) of inhaled corti- This special supplement to Circulation is freely available at costeroids concluded that steroids significantly reduced the http://www.circulationaha.org likelihood of admission to the hospital, particularly in patients DOI: 10.1161/CIRCULATIONAHA.105.166567 who were not receiving concomitant systemic steroids. But IV-139 IV-140 Circulation December 13, 2005 the meta-analysis concluded that there is insufficient evi- Heliox dence that inhaled corticosteroids alone are as effective as Heliox is a mixture of helium and oxygen (usually a 70:30 systemic steroids.13 helium to oxygen ratio mix) that is less viscous than ambient air. Heliox has been shown to improve the delivery and Adjunctive Therapies deposition of nebulized albuterol.26 Although recent meta- analysis of 4 clinical trials did not support the use of heliox in Anticholinergics 27 Ipratropium bromide is an anticholinergic bronchodilator that the initial treatment of patients with acute asthma, it may be useful for asthma that is refractory to conventional therapy.28 is pharmacologically related to atropine. It can produce a The heliox mixture requires at least 70% helium for effect, so clinically modest improvement in lung function compared if the patient requires Ͼ30% oxygen, the heliox mixture with albuterol alone.14,15 The nebulizer dose is 0.5 mg. It has cannot be used. a slow onset of action (approximately 20 minutes), with peak effectiveness at 60 to 90 minutes and no systemic side effects. Methylxanthines It is typically given only once because of its prolonged onset Although previously a mainstay in the treatment of acute of action, but some studies have shown clinical improvement asthma, methylxanthines are infrequently used because of only with repeated doses.16 Given the few side effects, erratic pharmacokinetics and known side effects. ipratropium should be considered an adjunct to albuterol. Tiotropium is a new, longer-acting anticholinergic that is Leukotriene Antagonists Leukotriene antagonists improve lung function and decrease currently undergoing clinical testing for use in acute the need for short-acting ␤-agonists during long-term asthma asthma.17 therapy, but their effectiveness during acute exacerbations of Magnesium Sulfate asthma is unproven. One study showed improvement in lung IV magnesium sulfate can modestly improve pulmonary function with the addition of IV montelukast to standard function in patients with asthma when combined with nebu- therapy,29 but further research is needed. lized ␤-adrenergic agents and corticosteroids.18 Magnesium Inhaled Anesthetics causes bronchial smooth muscle relaxation independent of Case reports in adults30 and children31 suggest a benefit of the serum magnesium level, with only minor side effects inhalation anesthetics for patients with status asthmaticus (flushing, lightheadedness). A Cochrane meta-analysis of 7 unresponsive to maximal conventional therapy. These anes- studies concluded that IV magnesium sulfate improves pul- thetic agents may work directly as bronchodilators and may monary function and reduces hospital admissions, particu- have indirect effects by enhancing patient-ventilator syn- larly for patients with the most severe exacerbations of chrony and reducing oxygen demand and carbon dioxide 19 asthma. The typical adult dose is 1.2 to2gIVgiven over 20 production. This therapy, however, requires an ICU setting, ␤ minutes. When given with a 2-agonist, nebulized magne- and there have been no randomized studies to evaluate its sium sulfate also improved pulmonary function during acute effectiveness. asthma but did not reduce rate of hospitalization.20

Parenteral Epinephrine or Terbutaline Assisted Ventilation Epinephrine and terbutaline are adrenergic agents that can be Noninvasive Positive-Pressure Ventilation given subcutaneously to patients with acute severe asthma. Noninvasive positive-pressure ventilation (NIPPV) may offer The dose of subcutaneous epinephrine (concentration of short-term support to patients with acute respiratory failure 1:1000) is 0.01 mg/kg divided into 3 doses of approximately and may delay or eliminate the need for endotracheal intu- 0.3 mg given at 20-minute intervals. The nonselective adren- bation.32,33 This therapy requires an alert patient with ade- ergic properties of epinephrine may cause an increase in heart quate spontaneous respiratory effort. Bi-level positive airway rate, myocardial irritability, and increased oxygen demand. pressure (BiPAP), the most common way of delivering Ͼ But its use (even in patients 35 years of age) is well- NIPPV, allows for separate control of inspiratory and expi- 21 tolerated. Terbutaline is given in a dose of 0.25 mg ratory pressures. subcutaneously and can be repeated in 30 to 60 minutes. These drugs are more commonly administered to children with acute asthma. Although most studies have shown them Endotracheal Intubation With to be equally efficacious,22 one study concluded that terbutal- Mechanical Ventilation ine was superior.23 Endotracheal intubation does not solve the problem of small airway constriction in patients with severe asthma. In addi- Ketamine tion, intubation and positive-pressure ventilation can trigger Ketamine is a parenteral dissociative anesthetic that has further bronchoconstriction and complications such as breath bronchodilatory properties. Ketamine may also have indirect stacking (auto-PEEP [positive end-expiratory pressure]) and effects in patients with asthma through its sedative properties. barotrauma. Although endotracheal intubation introduces One case series24 suggested substantial effectiveness, but the risks, elective intubation should be performed if the asthmatic single randomized trial published to date25 showed no benefit patient deteriorates despite aggressive management. of ketamine when compared with standard care. Ketamine Rapid sequence intubation is the technique of choice. The will stimulate copious bronchial secretions. provider should use the largest endotracheal tube available Part 10.5: Near-Fatal Asthma IV-141

(usually 8 or 9 mm) to decrease airway resistance. Immedi- of experienced providers in an intensive care setting. Some ately after intubation, confirm endotracheal tube placement tertiary centers can offer experimental therapies as a last resort, by clinical examination and a device (eg, exhaled CO2 and transfer should be considered for patients with near-fatal detector) and obtain a chest radiograph. asthma that is refractory to aggressive medical management.

Troubleshooting After Intubation References When severe bronchoconstriction is present, breath stacking 1. Division of Data Services. New Asthma Estimates: Tracking Prevalence, (so-called auto-PEEP) can develop during positive-pressure Health Care, and Mortality. Hyattsville, Md: National Center for Health ventilation, leading to complications such as hyperinflation, Statistics; 2001. tension pneumothorax, and hypotension. During manual or 2. McFadden ER Jr. Acute severe asthma. Am J Respir Crit Care Med. 2003;168:740–759. mechanical ventilation use a slower respiratory rate (eg, 6 to 3. Rainbow J, Browne GJ. Fatal asthma or anaphylaxis? Emerg Med J. 10 breaths per minute) with smaller tidal volumes (eg, 6 to 2002;19:415–417. 8 mL/kg),34 shorter inspiratory time (eg, adult inspiratory 4. Kokturk N, Demir N, Kervan F, Dinc E, Koybasioglu A, Turktas H. A flow rate 80 to 100 mL/min), and longer expiratory time (eg, subglottic mass mimicking near-fatal asthma: a challenge of diagnosis. J Emerg Med. 2004;26:57–60. inspiratory to expiratory ratio 1:4 or 1:5) than would typically 5. Rodrigo GJ, Rodrigo C. Continuous vs intermittent beta-agonists in the be provided to nonasthmatic patients. treatment of acute adult asthma: a systematic review with meta-analysis. Mild hypoventilation (permissive hypercapnia) reduces the Chest. 2002;122:160–165. 35 6. Khine H, Fuchs SM, Saville AL. Continuous vs intermittent nebulized risk of barotrauma. Hypercapnia is typically well tolerated. albuterol for emergency management of asthma. Acad Emerg Med. 1996; Sedation is often required to optimize ventilation and mini- 3:1019–1024. mize barotrauma after intubation. Delivery of inhaled medi- 7. Lin RY, Sauter D, Newman T, Sirleaf J, Walters J, Tavakol M. Con- cations may be inadequate before intubation, so continue to tinuous versus intermittent albuterol nebulization in the treatment of acute asthma. Ann Emerg Med. 1993;22:1847–1853. administer inhaled albuterol treatments through the endotra- 8. Rudnitsky GS, Eberlein RS, Schoffstall JM, Mazur JE, Spivey WH. cheal tube. Comparison of intermittent and continuously nebulized albuterol for Four common causes of acute deterioration in any intu- treatment of asthma in an urban emergency department. Ann Emerg Med. bated patient are recalled by the mnemonic DOPE (tube 1993;22:1842–1846. 9. Newman KB, Milne S, Hamilton C, Hall K. A comparison of albuterol Displacement, tube Obstruction, Pneumothorax, and Equip- administered by metered-dose inhaler and spacer with albuterol by neb- ment failure). This mnemonic still holds in the patient with ulizer in adults presenting to an urban emergency department with acute severe asthma. asthma. Chest. 2002;121:1036–1041. If the patient with asthma deteriorates or is difficult to 10. Nowak R. Single-isomer levalbuterol: a review of the acute data. Curr Allergy Asthma Rep. 2003;3:172–178. ventilate, verify endotracheal tube position, eliminate tube 11. Gibbs MA, Camargo CA Jr, Rowe BH, Silverman RA. State of the art: obstruction (eliminate any mucous plugs and kinks), and rule therapeutic controversies in severe acute asthma. Acad Emerg Med. out (or decompress) a pneumothorax. Only experienced 2000;7:800–815. providers should perform needle decompression or insertion 12. Ratto D, Alfaro C, Sipsey J, Glovsky MM, Sharma OP. Are intravenous corticosteroids required in status asthmaticus? JAMA. 1988;260:527–529. of a chest tube for pneumothorax. 13. Rowe BH, Spooner CH, Ducharme FM, Bretzlaff JA, Bota GW. Early Check the ventilator circuit for leaks or malfunction. High emergency department treatment of acute asthma with systemic cortico- end-expiratory pressure can be quickly reduced by separating steroids. Cochrane Database Syst Rev. 2000;CD002178. 14. Aaron SD. The use of ipratropium bromide for the management of acute the patient from the ventilator circuit; this will allow PEEP to asthma exacerbation in adults and children: a systematic review. J dissipate during passive exhalation. To minimize auto-PEEP, Asthma. 2001;38:521–530. decrease inhalation time (this increases exhalation time), 15. Rodrigo G, Rodrigo C, Burschtin O. A meta-analysis of the effects of decrease the respiratory rate by 2 breaths per minute, and ipratropium bromide in adults with acute asthma. Am J Med. 1999;107: 363–370. reduce the tidal volume to 3 to 5 mL/kg. Continue treatment 16. Plotnick LH, Ducharme FM. Acute asthma in children and adolescents: with inhaled albuterol. should inhaled anticholinergics be added to beta(2)-agonists? Am J Respir Med. 2003;2:109–115. 17. Keam SJ, Keating GM. Tiotropium bromide. A review of its use as Cardiac Arrest in the Asthmatic Patient maintenance therapy in patients with COPD. Treat Respir Med. 2004;3: When the asthmatic patient experiences a cardiac arrest, the 247–268. provider may be concerned about modifications to the ACLS 18. Silverman RA, Osborn H, Runge J, Gallagher EJ, Chiang W, Feldman J, Gaeta T, Freeman K, Levin B, Mancherje N, Scharf S. IV magnesium guidelines. There is inadequate evidence to recommend for or sulfate in the treatment of acute severe asthma: a multicenter randomized against the use of heliox during cardiac arrest (Class Indeter- controlled trial. Chest. 2002;122:489–497. minate).36 There is insufficient evidence to recommend com- 19. Rowe BH, Bretzlaff JA, Bourdon C, Bota GW, Camargo CA Jr. Mag- nesium sulfate for treating exacerbations of acute asthma in the pression of the chest wall to relieve gas trapping if dynamic emergency department. Cochrane Database Syst Rev. 2000;CD001490. hyperinflation occurs.37 20. Blitz M, Blitz S, Beasely R, Diner B, Hughes R, Knopp J, Rowe B. Inhaled magnesium sulfate in the treatment of acute asthma. Cochrane Database Syst Rev. 2005;CD003898. Summary 21. Cydulka R, Davison R, Grammer L, Parker M, Mathews J IV. The use of When treating patients with severe asthma, providers should epinephrine in the treatment of older adult asthmatics. Ann Emerg Med. closely monitor patients to detect further deterioration or 1988;17:322–326. 22. Victoria MS, Battista CJ, Nangia BS. Comparison of subcutaneous ter- development of complications. When there is no improve- butaline with epinephrine in the treatment of asthma in children. J Allergy ment and intubation is required, these patients require the care Clin Immunol. 1977;59:128–135. IV-142 Circulation December 13, 2005

23. Victoria MS, Battista CJ, Nangia BS. Comparison between epinephrine 31. Wheeler DS, Clapp CR, Ponaman ML, Bsn HM, Poss WB. Isoflurane and terbutaline injections in the acute management of asthma. J Asthma. therapy for status asthmaticus in children: a case series and protocol. 1989;26:287–290. Pediatr Crit Care Med. 2000;1:55–59. 24. Petrillo TM, Fortenberry JD, Linzer JF, Simon HK. Emergency department use 32. Non-invasive ventilation in acute respiratory failure. Thorax. 2002;57: of ketamine in pediatric status asthmaticus. J Asthma. 2001;38:657–664. 192–211. 25. Howton JC, Rose J, Duffy S, Zoltanski T, Levitt MA. Randomized, 33. Soroksky A, Stav D, Shpirer I. A pilot prospective, randomized, placebo- double-blind, placebo-controlled trial of intravenous ketamine in acute controlled trial of bilevel positive airway pressure in acute asthmatic asthma. Ann Emerg Med. 1996;27:170–175. attack. Chest. 2003;123:1018–1025. 26. Hess DR, Acosta FL, Ritz RH, Kacmarek RM, Camargo CA Jr. The 34. Marik PE, Varon J, Fromm R Jr. The management of acute severe effect of heliox on nebulizer function using a beta-agonist bronchodilator. asthma. J Emerg Med. 2002;23:257–268. Chest. 1999;115:184–189. 35. Mazzeo AT, Spada A, Pratico C, Lucanto T, Santamaria LB. Hyper- 27. Rodrigo GJ, Rodrigo C, Pollack CV, Rowe B. Use of helium-oxygen mixtures in the treatment of acute asthma: a systematic review. Chest. capnia: what is the limit in paediatric patients? A case of near-fatal 2003;123:891–896. asthma successfully treated by multipharmacological approach. Paediatr 28. Reuben AD, Harris AR. Heliox for asthma in the emergency department: Anaesth. 2004;14:596–603. a review of the literature. Emerg Med J. 2004;21:131–135. 36. Rodrigo G, Pollack C, Rodrigo C, Rowe BH. Heliox for nonintubated 29. Camargo CA Jr, Smithline HA, Malice MP, Green SA, Reiss TF. A acute asthma patients. Cochrane Database Syst Rev. 2003;CD002884. randomized controlled trial of intravenous montelukast in acute asthma. 37. Van der Touw T, Mudaliar Y, Nayyar V. Cardiorespiratory effects of Am J Respir Crit Care Med. 2003;167:528–533. manually compressing the rib cage during tidal expiration in mechan- 30. Schultz TE. Sevoflurane administration in status asthmaticus: a case ically ventilated patients recovering from acute severe asthma. Crit Care report. AANA J. 2005;73:35–36. Med. 1998;26:1361–1367. Management of life-threatening asthma in adults

David Stanley MRCP FRCA William Tunnicliffe FRCP

Asthma is a disease of predominantly reversible Guidelines Network published guidelines on Key points airway obstruction characterized by a triad of the management of asthma (http://www.brit- The primary bronchial smooth muscle contraction, airway thoracic.org.uk/asthma-guideline-download.html). pathophysiological problem inflammation, and increased secretions; it is a They were updated in November 2007. In in life-threatening asthma is major health problem for all age groups. For common with other recommendations, these expiratory gas-flow the majority, control of asthma symptoms is give guidance only up to the commencement of limitation. readily achieved; however, in a small minority, intensive care. Evidence for therapies in inten- Always be wary of asthma may cause death. Although the mor- sive care from randomized controlled trials normocapnia and consider tality rate for asthma in those aged less than remains sparse. Ongoing management of life- it in clinical context; it may 65 yrs is now falling, there remain around 1400 threatening episodes (Fig. 1) is often difficult be a sign of patient asthma deaths in the UK each year (http:// with rapid and dynamic changes in physiology. deterioration or exhaustion. www.laia.ac.uk/kf_asthma_03.htm). Most of Complications of treatment are frequent. Initiation of invasive these occur in the pre-hospital setting and, in ventilation is a clinical retrospect, the majority is considered poten- Pathophysiology decision. tially preventable. Factors associated with The safety of permissive asthma death include disease severity, The pathophysiological feature of life- hypercapnia is well inadequate treatment, inadequate monitoring, threatening asthma is gas trapping with established. Striving for the under use of written asthma management dynamic hyperinflation and the generation of normocapnia is likely to plans and adverse psychosocial and behavioural intrinsic positive end-expiratory pressure produce significant factors (www.sign.ac.uk/guidelines/fulltext/63/ (PEEPi). This arises due to disproportionate morbidity. index.html). increases in resistance to expiratory gas flow, Levels of severity of acute asthma exacer- rapid respiratory rates, changes in pulmonary bations have been defined. The features of elastic recoil, and asynchronous respiratory acute severe, life–threatening, and near fatal muscle activity. The consequences include asthma are listed in Table 1. Of note, patients impaired gas exchange, increased work of with life-threatening asthma may not appear breathing with respiratory muscle fatigue, and distressed and might only display one of the increased risk of barotrauma. Hyperinflation features listed. Life-threatening asthma tends to may be so severe that lung volumes approach occur as two sub-types.1 Most commonly, there total lung capacity. Diaphragmatic flattening is a history of progressive worsening of symp- at such volumes reduces the efficiency of ven- David Stanley MRCP FRCA toms over several days; the majority of patients tilation as inspiration becomes primarily by Specialist Registrar in Anaesthesia and in this group report nocturnal dyspnoea in the intercostal muscles rather than the diaphragm. Intensive Care Queen Elizabeth Hospital previous three nights. As a result of greater Together, these factors reduce CO2 elimin- Edgbaston bronchial inflammation and mucous secretion, ation while increasing production. At a point, Birmingham B15 2TH this group (occurring more frequently in production will match and then exceed rate of UK females) tends to respond more slowly to treat- elimination progressing to respiratory failure William Tunnicliffe FRCP when there is inadequate alveolar ventilation. ment. In contrast, a smaller sub-group, more Consultant in Respiratory and Intensive frequently male, presents with a rapidly pro- In addition, airway closure causes Care Medicine gressive condition with highly reactive airways. mismatches in ventilation-perfusion leading to Queen Elizabeth Hospital hypoxaemia. Edgbaston These patients have intense bronchospasm that Birmingham B15 2TH often responds more rapidly to bronchodilator Large negative intrathoracic pressures gener- UK therapy. ated by the augmented inspiratory effort, as Tel: þ44 121 6271627 well as PEEPi through its effects on right atrial Fax: þ44 121 6978291 In February 2003, The British Thoracic E-mail: [email protected] Society together with the Scottish Intercollegiate filling, can impede cardiac output. Dehydration (for correspondence) doi:10.1093/bjaceaccp/mkn012 Continuing Education in Anaesthesia, Critical Care & Pain | Volume 8 Number 3 2008 95 & The Board of Management and Trustees of the British Journal of Anaesthesia [2008]. All rights reserved. For Permissions, please email: [email protected] Management of life-threatening asthma

Table 1 Features of acute severe asthma

P Near fatal asthma Raised aCO2 and/or requiring mechanical ventilation with raised inﬂation pressures Life-threatening Any one of the following in a patient with severe asthma: asthma PEF ,33% best or predicted S , pO2 92% P , aO2 8 kPa P Normal aCO2 (4.6–6.0 kPa) Silent chest Cyanosis Feeble respiratory effort Bradycardia Dysrrhythmia Hypotension Exhaustion Confusion Coma Acute severe asthma Any one of: PEF 33–50% best or predicted Respiratory rate 25 min21 Heart rate 110 min21 Inability to complete sentences in one breath

from reduced intake and increased respiratory losses, along with the development of hypoxaemia, hypokalaemia, and acidosis may further exacerbate this.

Fig. 1 Management of acute severe asthma in adults. Management of life-threatening asthma Initial management management of acute severe asthma and may increase mortality in this setting. A rapid ABC assessment should be undertaken and actioned. Many patients will be hypoxaemic, hypovolaemic, acidotic, and hypokalaemic. Nebulized ipratropium bromide This should be added to nebulized b2 agonists treatment for all patients with life-threatening asthma (500 mg 4 hourly) as it has Oxygen been shown to produce signiﬁcantly greater bronchodilation than

Patients with acute severe asthma are hypoxaemic. This should be b2 agonists alone. Side effects are minimal. corrected urgently with high concentrations of inspired oxygen aiming to achieve oxygen saturations .92%. A FiO of 0.4–0.6 is 2 Steroids often sufficient but, in general, start high and then titrate the Fi O2 Systemic steroids in adequate doses should be given to all patients down. It is important to note that asphyxia remains the most with life-threatening asthma, as early as possible in the episode as common mechanism of death in severe asthma and should never this may improve survival. Steroid tablets (prednisolone 40–50 mg be underestimated. daily) have been shown to be as efficacious as intravenous steroids in acute severe asthma, provided tablets can be swallowed and retained. If in any doubt, the intravenous route should be used Nebulized b2 agonists (hydrocortisone 200 mg stat followed by 100 mg 6 hourly). Short-acting b2-agonists (e.g. salbutamol) should be given repeatedly in 5 mg doses or by continuous nebulization at 10 mg h21 Inhaled/nebulized steroids do not provide additional benefit. driven by oxygen. Importantly, duration of activity and effectiveness are inversely related to severity of asthma; continuous admin- Intravenous magnesium sulphate istration is more efficacious in severe asthma. However, even under A single intravenous dose of magnesium sulphate 1.2–2 g over ideal circumstances only 10% of the nebulized drug will reach the 20 min has been shown to be safe and effective in acute severe bronchioles.2 Administration should continue until there is a sig- asthma. Magnesium is a smooth muscle relaxant, producing nificant clinical response or serious side effects occur, these bronchodilation. Rapid administration may be associated with include tachycardia, arrhythmias, tremor, hypokalaemia, and hypotension. Anecdotal evidence suggests repeated doses or infu- hyperglycaemia. Inhaled longer acting b2-agonists have no role in sions may be of benefit (dose limited by twice normal serum

96 Continuing Education in Anaesthesia, Critical Care & Pain j Volume 8 Number 3 2008 Management of life-threatening asthma

magnesium concentrations); however, hypermagnesaemia is associ- ventilation. Half of the life-threatening complications occur at or ated with muscle weakness and may exacerbate respiratory failure around the time of intubation in patients mechanically ventilated in spontaneously breathing patients. for asthma; consequently, intubation should be performed by the most senior and experienced member of anaesthetic staff available Intravenous bronchodilators with the help of appropriately trained assistance. Where possible,

Parenteral b2 agonists, in addition to nebulized b2 agonists, should pre-oxygenation should be performed diligently followed by a be considered in ventilated patients and those with life-threatening rapid sequence induction. Hypotension at the initiation of venti- asthma; intravenous salbutamol (5–20 mg min21) or terbutaline lation should be anticipated and attenuated by fluid pre-loading. (0.05 mgkg21 min21) should be titrated to response. Lactic acido- Vasopressors should be immediately available for use post- sis will develop in 70% of patients after 2–4 h therapy, requiring induction. Hypotension can be severe enough to result in complete careful monitoring with subsequent reduction or cessation of loss of cardiac output or mimic that occurring with a tension pneu- therapy. In extremis, salbutamol 100–300 mg can be given as an mothorax. Causes of hypotension are multifactorial, including intravenous bolus or via an endotracheal tube. vasodilatation and reduction in sympathetic tone on induction, An additional or alternative intravenous bronchodilator is ami- absolute hypovolaemia and reduction in venous return consequent nophylline. Some patients with life-threatening asthma gain benefit to high intrathoracic pressures precipitated by ventilating against from its intravenous use (5 mg kg21 loading dose over 20 min high airway resistance. Initial hand ventilation is often over- unless on maintenance oral therapy, then infusion of 0.5– enthusiastic and contributory; it should follow the principles out- 0.75 mg kg21 min21). Concern and controversy about its use arises lined for the mechanical ventilation below and be kept to a from its side effects (arrhythmias, restlessness, vomiting, and con- minimum; rate should be kept low and no PEEP should be applied. vulsions) related to a narrow therapeutic window. Trials showing If profound hypotension does occur when assisted ventilation has little overall benefit in lesser degrees of asthma may not be rel- been initiated, consideration should be given to disconnecting the evant when faced with impending asphyxia. Plasma aminophylline patient from the circuit (possibly with the addition of pressure on concentrations should be monitored frequently (therapeutic range the chest wall to assist expiratory flow) to allow full passive 10–20 mgml21). expiration A chest x-ray should be performed following intubation when it is safe to do so, to assess correct positioning of the endo- Epinephrine tracheal tube and exclude pneumothoraces. The additional use of epinephrine (adrenaline) should be considered in patients not responding adequately to the measures out- What are the initial goals of mechanical ventilation lined above via the subcutaneous (0.3–0.4 ml 1:1000 every 20 min and how are they achieved? for three doses), nebulized (2–4 ml of 1% solution hourly) or, in The initial goals of mechanical ventilation are to correct hypoxae- extremis, the intravenous route (0.2–1 mg as a bolus followed by mia, reduce dynamic hyperinflation and to buy time for medical 1–20 mg min21). management to work. Adequate sedation is vital and typically would be morphine and midazolam with ketamine. Morphine has a potential for histamine Mechanical ventilation release so is avoided by some. Propofol and fentanyl, and ketamine Who should be intubated, and when and how should plus midazolam alone are alternatives. The attraction of ketamine mechanical ventilation be initiated? (0.5–2 mg kg21 h21) is its action as a direct bronchodilator. The initiation of invasive ventilation in life-threatening asthma is a Evidence of benefit is equivocal; use maybe limited by its effects bedside clinical decision based on an assessment of the balance of on respiratory tract secretions and its sympathomimetic properties risks and benefits. It can be life saving, but has a higher incidence in a patient already in a heightened sympathetic state. of complications relative to other causes of respiratory failure. Initial neuromuscular blockade is often required. Rocuronium or Absolute indications are coma, respiratory or cardiac arrest and pancuronium is the agents of choice. Atracurium is associated with severe refractory hypoxaemia. Relative indications include an histamine release. Some concern has been raised over the relative adverse trajectory of response to initial management, fatigue and likelihood of developing a neuromyopathy with vecuronium in this somnolence, cardiovascular compromise, and the development of a setting (high-dose steroids, mechanical ventilation, and severe pneumothorax. Prior to initiation, vigilant observation is manda- asthma), though the evidence is limited. The use of neuromuscular tory as fatal apnoea can occur suddenly and unexpectedly.3 blockade should be discontinued as soon as possible. Hypercapnia per se is not an indication for mechanical ventilation. Initial ventilator settings should adopt relatively low rates (12– 21 21 F In life-threatening asthma, the induction of anaesthesia, tracheal 14 breaths min ), tidal volumes of 4–8 ml kg , iO2 sufficient intubation, and initial ventilation are all extremely hazardous as to maintain adequate oxygen saturations (.92%), relatively long- dramatic changes in physiology occur with the induction of anaes- expiratory times (I:E 1:4) and little or no PEEP (,5cmH2O). If thesia, and with the switch from high intrinsically produced nega- using volume controlled ventilation, appropriate goals would be to tive intrathoracic pressures to high positive pressures from extrinsic achieve a Pplat ,35 cm H2O with pH . 7.2. If Pplat . 35 cm H2O,

Continuing Education in Anaesthesia, Critical Care & Pain j Volume 8 Number 3 2008 97 Management of life-threatening asthma

minute ventilation should be reduced (Vt and/or rate), if pH , 7.2 Tromethamine (THAM) has some theoretical advantages over and Pplat , 30 cm H2O, minute ventilation should be increased bicarbonate. Use with either agent is usually complicated by meta-

(rate), if pH , 7.2 and Pplat . 35 cm H2O no change may be bolic alkalosis on resolution of the acute bronchospastic episode. appropriate. This guidance principally derives from Tuxen and Measures to limit CO2 production, such as anti-pyretics and/or Lane’s7 observations of volume controlled ventilation in asthma, active cooling, should be considered as adjuncts or alternatives. namely that minute ventilation is the most important determinant of hyperinflation (inspiratory flow and shape of pressure wave-form Ongoing ventilatory management being of much lesser importance), and that while The use of neuromuscular blockade and deep sedation should be increased-expiratory times are beneficial, the effect of increases discontinued as soon as the clinical situation allows and the return above 3–4 s are minimal. In addition, the role of PEEPe to to spontaneous ventilation should be achieved as soon as is practi- counter PEEPi in controlled mechanical ventilation has no ration- cal. During this process, patient ventilator interactions become ale, although recently potentially beneficial effects of PEEPe have more important and the use of PEEPe and the selection of appro- been reported in ‘obstructive lung disease’ (as opposed to asthma) priate trigger sensitivities assume important roles in reducing the suggesting that a trial of variable PEEPe may be required in some work of breathing. 5 cases. The ventilator care bundle approach should be considered and, Achieving the goals of a pH . 7.2 with a Pplat , 35 cm H2O if not already involved, advice from a respiratory physician should will often not be possible and requires ongoing clinical assessment. be sought with a view to planning the patient’s ongoing manage- High airway pressures should prompt the exclusion of endobron- ment in hospital and in the community. chial intubation and pneumothorax, along with the re-evaluation of the adequacy of sedation. Plateau airway and end-expiratory pressures generally reflect Additional management methods the degree of gas trapping in severe asthma; in addition, total Inhalational anaesthetic agents exhaled volume during an apnoea for 20–60s gives a measure of Volatile inhalational anaesthetic agents (e.g. isoflurane and sevo- 6 the degree of hyperinflation. Not all airways remain patent flurane) are effective bronchodilators and their use is associated throughout expiration, any measurement will tend to be an under- P with reductions in PEEPi and aco2 within hours. Increased benefit estimate and clinical assessment remains vital. Of note, barotrauma may occur with earlier administration and their use should be con- in the mechanically ventilated asthmatic including the risk of pneu- sidered in patients with life-threatening asthma not responding to mothorax is proportional to end inspiratory lung volume. standard treatments. Hypotension is their principal side effect. The Additional complications of mechanical ventilation in life- delivery of volatile anaesthetic agents can be difficult in the inten- threatening asthma include profound hypotension, cardiac stunning, sive care unit setting and may force the use of a lower fidelity ven- arrhythmia, rhabdomyolysis, lactic acidosis, myopathy, and CNS tilator in a sub-optimal environment. In line devices for use with injury. Cardiac stunning is thought to be consequent to massive high fidelity ventilators are available, but scavenging and agent sympathetic activation, arrhythmias generally tend to be benign, monitoring are absolute requirements. rhabdomyolysis is rare and is thought to stem from hypoxaemia in combination with extreme exertion; lactic acidosis in severe asthma is poorly understood. Extra-corporeal support Case reports of the use of extra-corporeal membrane oxygenation Management of hypercarbia in life-threatening asthma suggest that this may be successful, but Permissive hypercapnia is a well-proven protective lung strategy its limited availability and the risk profile limit its applicability. In for limiting the deleterious effects of barotrauma in many causes contrast, the development of less complex systems of extra- of respiratory failure. During the mechanical ventilation for life- pulmonary gas exchange that facilitate CO2 clearance (e.g. threatening asthma, our ability to correct hypercapnia is generally Novalung) brings extra-corporeal CO2 removal (ECCO2R) within limited and fraught with hazards. Although calling it ‘permissive’ bounds. Their use should be considered particularly where the might be a misnomer, hypercapnia is generally well tolerated, and control of hypercarbia is imperative. there are numerous case reports in the literature of asthmatic patients with profound respiratory acidosis of several hours dur- Bronchoscopy ation with no significant adverse effects. The exception is in those Bronchoscopy has a limited role in managing patients with persist- with cerebral anoxia secondary to a respiratory arrest. Control of ent shunt consequent to mucus plugging. Lavage of the obstructed intracranial pressure requires management of hypercarbia; urgent lung segments and the removal of mucus plugs may reduce the consideration should be given to extra corporeal CO2 removal in duration of ventilation, but is often complicated by bronchospasm. these circumstances (see later). In general, patience and persistence with standard therapies and Where respiratory acidosis is extreme and its ventilatory meticulous attention to the hydration and humidification are as management impossible, buffering can be considered acutely. effective.

98 Continuing Education in Anaesthesia, Critical Care & Pain j Volume 8 Number 3 2008 Management of life-threatening asthma

Antibiotics Leukotriene antagonists The majority of episodes of acute severe asthma that has an infec- Currently available leukotriene antagonists (Montelukast, tive precipitant follow a viral infection. The routine use of anti- Zafirlukast) have no role in management of life-threatening biotics in life-threatening asthma has no rationale and they should asthma. be considered only in selected cases. Monoclonal anti-IgE antibodies Non-Invasive ventilation Omalizumab is thought to be effective at reducing the number and Although there is a sound evidence for the use of non-invasive possibly the severity of acute exacerbations of asthma in adults ventilation (NIV) in acute exacerbations of chronic obstructive with moderate to severe allergic asthma that is inadequately con- lung disease, its use in asthma is controversial. Objective evidence trolled by inhaled steroids. It is a preventative measure and has no 7 of benefit is very limited; a recent Cochrane review could only role in the management of acute life-threatening episodes. find one well-conducted prospective, randomized trial of just 30 patients that showed improvements in the respiratory rates in asthmatics with mild to moderate exacerbations when NIV was added References to the standard medical care. Currently, the use of NIV in even 1. Wasserfallen JB, Schaller MD, Feihl F, Perret CH. Am Rev Respir Dis 1990; mild to moderate exacerbations of asthma cannot be recommended 142: 108–11 outside the randomized controlled trials. NIV has no role in the 2. Tuxen D, Leong T. Acute severe asthma. In: Bersten A, Soni N, eds. Oh’s management of life-threatening exacerbations of asthma. Intensive Care Manual. London: Butterworth Heinemann 2003 3. Davidson C, Treacher D, ed. Respiratory Critical Care. Arnold, 2002 Heliox 4. Tuxen D, Leong T. The effects of ventilatory pattern on hyperinflation, airway pressures, and circulation in mechanical ventilation of patients Heliox (oxygen in helium) reduces the density of the gas mix to with severe air-flow obstruction. Am Rev Respir Dis 1987; 142: 872–9 improve turbulent gas flow. Use has been reported in less severe 5. Guerin C, Milic-Emili J, Fournier G. Intensive Care Med 2000; 26: acute exacerbations of asthma where, in spontaneously breathing 1207–14 patients, it may reduce the work of breathing. This use is not, 6. Stather DR, Stewart TE. Clinical review: mechanical ventilation in severe however, supported by an evidence base and its utility is further asthma. Crit Care 2005; 9: 581–7 limited by a maximum oxygen fraction of 0.4 making it unsuitable 7. Ram FSF, Wellington SR, Rowe B, Wedzicha JA. Non-invasive positive for those with life-threatening asthma. Use of heliox with a mech- ventilation for treatment of respiratory failure due to severe acute anical ventilator is complex, requiring the recalibration of the exacerbations of asthma. Cochrane Database Syst Rev 2005; ArtNo.:CD004360. DOI: 10.1002/14651858.CD004360.pub3. pneumotachographs and the use of density independent spirometry of exhaled gas flows. Please see multiple choice questions 14–17

Continuing Education in Anaesthesia, Critical Care & Pain j Volume 8 Number 3 2008 99 ARTICLE IN PRESS

Current Anaesthesia & Critical Care (2007) 18,61–68

www.elsevier.com/locate/cacc FOCUS ON: MEDICAL EMERGENCIES Acute severe asthma: Triage, treatment and thereafter

Felix ChuaÃ, Dilys Lai

Department of Respiratory Medicine, Chelsea & Westminster Hospital, 369 Fulham Road, London SW10 9NH, UK

KEYWORDS Summary Despite an increased understanding of the aetiology and pathogenesis Acute severe of bronchial asthma, attempts to classify its severity have proved difficult. Objective asthma; changes in cardiopulmonary function do not consistently correlate with symptomatic Life threatening; decline during an acute exacerbation of asthma. Whilst the majority of such episodes Bronchodilators; are treatment-responsive, a number of asthmatics still succumb to the severest form Corticosteroids of the disease every year, often despite aggressive medical therapy. Although patients with severe asthma are in the minority within the disease spectrum, they seek and receive a disproportionate share of health care resources and are responsible for considerable morbidity associated with it. In addition to identifying such patients, a major challenge is the prioritisation of medical treatment when they present to the emergency department. Patients with acute severe asthma whose initial treatment is inadequate or delayed risk admission to the intensive care unit; the in-hospital mortality of these individuals remains significant. Timely provision of emergency therapy, vigilance against failure of first-line intervention and ensuring ready availability of escalating measures are critical elements of tertiary asthma care with which hospital physicians must be familiar. & 2007 Elsevier Ltd. All rights reserved.

Background asthma diagnoses. Whilst asthma per se is an unusual cause of all-cause hospital deaths, around 1400 Recent statistics show that asthma as a primary individuals succumbed to fatal asthma in 2004 in 2 diagnosis is responsible for nearly 70,000 hospital the UK. Likewise in the United States, deaths in admissions in the United Kingdom per year.1 In which asthma was reported as an underlying cause 3 general, however, individuals who remain sympto- numbered just over 4000 in 2003. In other words, matic despite high dose inhaled corticosteroids and whilst most patients with this condition respond well long-acting bronchodilators represent under 10% of all to conventional treatment, a subgroup of individuals are susceptible to potentially life threatening exacer- ÃCorresponding author. Tel.: +44 208 746 8472; bations precipitated by exposure to a variety of fax: +44 208 746 8547. external stimuli, some of which are seemingly E-mail address: [email protected] (F. Chua). innocuous.

62 F. Chua, D. Lai

Pathophysiological mechanisms of asthma exacer- episode is often underestimated by patients, their bation appear to differ between patients who are family and health care professionals who attend to prone to episodic decompensation from those with them.12 A consensus definition for severe chronic equally severe but stable disease. Different pheno- (ongoing) asthma has been produced by an Amer- types of the acute presentation have been described, ican Thoracic Society (ATS)-sponsored workshop ranging from previously undiagnosed (‘new’) asthma (Table 1).13 Its primary criteria are determined by to sudden deterioration of previously stable disease, requirements for high dose inhaled and systemic and precipitous episodes that appear ‘out of the corticosteroids. However, this definition has yet to blue’.4 However, catastrophic ‘asphyxic asthma’ is be prospectively evaluated. While there is increas- rare. So-called status asthmaticus manifests when an ing evidence that regular inhaled steroid treatment acute attack becomes refractory to intensive medical might reduce hospitalisation and asthma mortal- therapy. On the whole, most individuals with ity,14,15 the severe asthma phenotype is a broad deteriorating asthma have experienced prodromal entity that encompasses patients with corticoster- symptoms of greater than 24 h, and often several oid resistance who may have fixed airflow limita- days, prior to seeking medical help.5 Nonetheless, tion or unusually unstable airways with excessive the concept of the ‘crashing asthmatic’ helps identify bronchial hyper-reactivity. those patients at risk of progression to respiratory Overall, around 2% of all intensive care unit (ICU) failure or even cardiopulmonary arrest despite admissions are related to acute severe asthma.16,17 aggressive medical attention. This group of patients is characterised by a female to male ratio of 2:1.16,18 The reported mortality of asthmatic patients requiring ICU admission is 17,19 Pathological aspects of the acute around 7–8% although this figure varies widely, ranging from 0%20 to 40%.21 In the UK, there is a asthmatic event suspicion that patients admitted to ICUs with acute severe asthma may be sicker than in some other Acute asthma may be precipitated by a variety of countries.17,22 triggers, including infections (viral more commonly than bacterial), environmental allergens, pollutants, occupational irritants, drugs and treatment disruption. Specific alterations in the airway Table 1 ATS Workshop consensus for the defini- environment also predispose to exacerbations, tion of severe/refractory asthmaa. namely the development of bronchial hyper-reactivity, the presence of chronic airway inflammation Major characteristics and mucus hypersecretion.6 Bronchoalveolar lavage 1. Treatment with continuous or near continuous X fluid collected from mechanically ventilated pa- ( 50% of year) OCS 2. Requirement for treatment with high-dose ICS tients with status asthmaticus show increased numbers of eosinophils and neutrophils as well as Minor characteristics heightened levels of pro-inflammatory mediators 1. Requirement for additional daily treatment with including interleukin-1b, interleukin-6 and tumour a controller medication (e.g. long-acting 7,8 necrosis factor a. At necropsy, the classic b-agonist, theophylline or leukotriene pathological hallmarks of patients dying from antagonist) fulminant asthma are widespread mucus and 2. Asthma symptoms requiring short-acting inflammatory debris impaction, bronchial oedema, b-agonist use on a daily or near-daily basis hyperinflation and regional atelectasis, all conspir- 3. Persistent airway obstruction (FEV1o80% ing to dramatically narrow the airway luminal predicted, diurnal peak flow variability 420%) 4. caliber.9 One or more urgent care visits for asthma per year 5. Three or more oral steroid bursts per year 6. Prompt deterioration with p25% reduction in The road to status asthmaticus oral or ICS dose 7. Near fatal asthma event in the past

Two factors are crucial in understanding the OCS, oral corticosteroids; ICS, inhaled corticosteroids. progression of acute decompensating asthma. Deﬁnition requires 1 or both major criteria and 2 minor Firstly, the severity of bronchial airﬂow limitation criteria. a correlates poorly with classically recognised clin- Requires that other conditions have been excluded, ical signs of worsening disease.10,11 Secondly, and exacerbating factors have been treated, and patient is generally compliant. perhaps more importantly, the severity of an acute ARTICLE IN PRESS

Acute severe asthma: Triage, treatment and thereafter 63

Risk factors for near fatal asthma have been Table 2 BTS/SIGN levels of severity of acute closely studied over the past 30 years. Previous asthma exacerbations. severe asthma attacks, a history of mechanical ventilation and/or ICU admission, prior hospitalisa- Near fatal asthma tion for asthma with respiratory failure and the Raised PCO2 and/or requiring mechanical presence of certain psychosocial factors are ventilation with raised inflation pressures strongly predictive of an enhanced risk of life- Life threatening asthma threatening asthma.23–26 In studies of critical care Any one of the following in a patient with severe patients, receipt of cardiopulmonary resuscitation asthma: (CPR) in the 24 h preceding ICU admission is a key PEF o33% best or predicted Bradycardia determinant of the need for mechanical ventilation SpO2o92% Dysrhythmia and not surprisingly, impacts directly on hospital PaO2o8 kPa Hypotension mortality in patients with this condition.17 Normal PCO2 (4.6–6.0 kPa) Exhaustion Silent chest Confusion Cyanosis Coma Feeble respiratory effort Assessment of the acutely breathless Acute severe asthma asthmatic Any one of PEF 33–50% best or predicted Respiratory rate X25/min Studies have shown that traditionally assessed Heart rate X110/min clinical signs are not sensitive indicators of the 11 Inability to complete severity of airflow obstruction. For this and other sentences in one breath reasons, the adoption of asthma treatment proto- cols and guidelines has gained popularity. While Moderate asthma exacerbation such documents may have streamlined the manage- Increasing symptoms PEF 450–75% best or predicted ment of asthma emergencies, their impact on No features of acute severe concrete objective outcomes including mortality asthma remains unclear. The British Thoracic Society publishes a treat- Brittle asthma ment guideline for acute asthma which was most Type 1: wide PEF variability (440% diurnal recently updated in 2005.27 This guideline classifies variation for 450% of the time over a period 4150 days) despite intensive therapy the severity of acute asthma exacerbations at Type 2: sudden severe attacks on a background of different levels (Table 2) and highlights the apparently well controlled asthma importance of specific features of the medical history to the acute asthma episode. A history of frequent hospital visits, previous tracheal intubation, known brittle asthma with an unpredictable severe episode whereas PEF of less than 33% course, repeated courses of systemic steroids or predicted (or best) carries an enhanced risk of the presence of psychosocial factors should high- developing life-threatening characteristics.27 In the light the need for a more intensive treatment United States, the National Institutes of Health approach. Laboured, fast or shallow breathing, Guidelines for the Diagnosis and Management of fragmented speech due to breathlessness, the use Asthma (NAEPP guidelines) recognise comparable of accessory respiratory muscles, the presence of severity indicators pertaining to the PEF.28 (Na- intercostal retraction and tachycardia are all tional Institutes of Health, 1997). Although less clinical indicators of severe airflow obstruction. commonly available in the emergency department, Auscultatory signs may include wheezing, strained measurement of the forced expiratory volume in or quiet breath sounds, or even respiratory the first second (FEV1) provides additional informa- crackles. In very severe airway obstruction, un- tion of airflow compromise. It has been suggested controlled dynamic hyperinflation may rapidly that a fall in FEV1 to 50% of baseline in acute produce a true ‘silent chest’ that heralds impend- asthma may be associated with as high as a 7- to ing cardiopulmonary collapse. 10-fold increase in the inspiratory muscle work- In the emergency department, measurement of load.29 The measurement of either parameter, peak expiratory flow (PEF) in patients who are able however, becomes difficult and irrelevant in those to perform it provides important information on the who are acutely confused or in extremis. magnitude of airflow obstruction.10,27 PEF of less Arterial blood gas analysis early in the course of than 50% predicted helps to identify an acute an exacerbation typically shows normal or mild ARTICLE IN PRESS

64 F. Chua, D. Lai

hypoxaemia, hypocarbia and a mildly raised blood achieve a high sustained SpO2 (most guidelines pH due to hyperventilation. With significant airflow recommend maintaining an SpO2 of at least 92%). limitation, hypoxaemia worsens and carbon dioxide As hypoxaemia in most cases of acute asthma is due retention occurs, resulting in progressive acidae- to ventilation–perfusion mismatching, modest mia. In the tachypnoeic individual, a normal CO2 amounts of inhaled oxygen are usually sufficient tension signifies ineffective respiratory effort and to achieve the target SpO2. Unlike patients with likely muscle exhaustion. Hence, a normal or rising established COPD, the majority of asthmatics do PCO2 is a key marker of respiratory fatigue and not chronically retain CO2 so the danger of 19 warrants immediate assessment for assisted venti- supplemental O2-induced hypercarbia is minimal. lation. In a proportion of such patients, haemody- As far as possible, oxygen should also be used to namic instability (bradycardia and hypotension) deliver nebulised bronchodilators. In addition, may rapidly ensue, preceding wider circulatory ensuring adequate hydration with intravenous collapse. For practical reasons, assessment for fluids is important, as is baseline evaluation of pulsus paradoxus is no longer recommended. plasma electrolytes in the emergency room. How- A chest radiograph is important in the assessment ever, these measures must not take priority over of acute asthma exacerbation but must not delay the immediate administration of bronchodilators to the initiation of treatment. Common findings relieve airflow obstruction. include hyperinflation, localised atelectasis, bronchial cuffing and consolidation. The physician must also be vigilant for the presence of pneumothorax, Rapidly acting bronchodilators pneumomediastinum, foreign bodies and signs of cardiac failure. The most widely used b-agonist bronchodilator in the UK is salbutamol (albuterol in the United States), followed by terbutaline. It is specific, Criteria for admitting the acute rapidly acting, has a relatively short duration of asthmatic action (approximately 4–6 h) and produces dose- dependent bronchodilatation. In acute severe asthma, salbutamol may be administered either It goes without saying that all patients with any by wet nebulisation or repeated activations of a feature of a severe asthma exacerbation persisting metered dosed inhaler (MDI) via a large volume after initial treatment should be hospitalised. spacer. Most commonly, a nebulised bolus dose of However, it becomes more difficult with patients 2.5–5.0 mg of salbutamol is given at 15–20 min who seemingly recover their lung function (PEF or intervals until the desired effect is obtained or FEV ) within the emergency department. Ideally, 1 treatment escalation is warranted. In some in- all patients with mild to moderate exacerbation stances, the use of an MDI (delivering 100 mcg (PEF 50–75% best or predicted) with ongoing metered doses of the same drug) offers more rapid symptoms, reduced access to health services, a access to treatment and perhaps quicker effective suspicion of poor treatment compliance, social bronchodilatation in mild to moderate asthma.30,31 isolation, psychological problems, physical or However, data comparing the two modes of drug learning disabilities or a history of near fatal or delivery in severe asthma exacerbations are lack- brittle asthma should continue to be closely ing. Certainly, if the prior use of an MDI has been monitored in hospital.27 inadequate, nebulised aerosolisation may be advantageous, not least by enhancing the patient’s perception that treatment is being delivered Emergency treatment of severe acute effectively. asthma A number of clinical trials have also evaluated the use of bolus (intermittent) versus continuous The cornerstone of treatment of acute exacerba- nebuliser treatment, the latter being distinctly tions of asthma is a combination of aerosolised different from ‘back-to-back’ nebulisation. The bronchodilator and systemic corticosteroid therapy. continuous method involves the use of adapted The former is aimed at reversing bronchoconstric- systems to deliver larger quantities of the same tion while the latter is important for resolving bronchodilator agent over a number of hours (e.g. airway inflammation. Supplemental oxygen of at salbutamol at a rate of 10–20 mg/h). Although least 40–60% concentration, delivered via a high earlier studies suggested an increased efficacy of flow device (e.g. Hudson mask), should be adminis- this approach at relieving severe airway obstruc- tered at the earliest opportunity in order to tion, a systematic review of randomised controlled ARTICLE IN PRESS

Acute severe asthma: Triage, treatment and thereafter 65 trials of acute asthma in adults failed to identify BTS/SIGN guideline recommends the use of any significant differences in lung function or 40–60 mg prednisolone daily (versus 120–180 mg hospital admission rates between the continuous methylprednisolone daily). Whilst there is no and intermittent methods.32 evidence that the injection route is superior to Available evidence from the single available meta- oral, parenteral hydrocortisone (100 mg QDS) is analysis does not support the routine use of used in patients in whom swallowing or gastro- intravenous (IV) b-agonists in the management of intestinal absorption may be impaired. Following acute severe asthma.33 The IV route induced more stabilisation, oral prednisolone may be continued adverse sympathetic effects without producing sta- at a dose of 40 mg daily in most patients and tistically significant improvements in clinical signs, stopped after 7–10 days without the need for dose lung function or laboratory indices. However, IV tapering. A small number of patients will require a therapy should still be considered if doubts regarding protracted course due to persistent or recalcitrant efficient drug aerosolisation persist or if the patient airway inflammation. continues to worsen despite repeated nebulisation. A further uncertainty surrounds the substitution of Anti-cholinergic drugs compete with acetylcho- inhaled steroids with systemic therapy during an line to bind to airway muscarinic receptors and exacerbation. Although such a strategy may appear decrease bronchial vagal tone. Thus, the rationale intuitive, studies have shown that the use of high for incorporating such agents into treatment pro- dose inhaled steroids in the emergency department tocols of acute asthma is biologically sound. can produce earlier improvements in pulmonary Indeed, combining an anti-cholinergic drug with function and a decreased likelihood of hospitaliza- nebulised b-agonist has been shown to produce tion.38,39 However, there is no evidence that inhaled greater bronchodilatory benefit compared to steroids alone are adequate for controlling severe b-agonist alone as concluded by a meta-analysis.34 asthma exacerbations. It is reasonable that inhaled The most widely used agent, ipratropium bromide, therapy is not stopped following the commencement is a quartenary anti-cholinergic compound with of systemic steroids but continued to form part of the minimal systemic absorption when delivered into chronic asthma management plan.27 the airways, a slower onset of action than salbutamol (approximately 20 versus 5 min) but a Magnesium sulfate prolonged action profile with few systemic side effects. Tiotropium, a longer acting anti-choliner- Magnesium is a predominantly intracellular cation, gic, has not been properly tested for the treatment many of whose physiological actions involve cal- of either acute or chronic asthma. cium antagonism regardless of the circulating calcium concentration.40 Its potential role in asthma may involve a combination of smooth Corticosteroids muscle relaxation, inhibition of histamine release and acetylcholine release from nerve endings.41 It Corticosteroids principally act by suppressing air- has also been shown to augment b -agonist effects way inflammation. They exert dose-dependent 2 in the airway42 and to attenuate potentially down-regulatory effects on eosinophil, neutrophil deleterious neutrophil oxidative bursts.43 While and lymphocyte function, including the production magnesium sulfate is generally safe when given of growth factors. Although they do not have direct parenterally, its therapeutic potential as an ad- bronchodilatory properties, steroids can upregulate junct to b-agonists in acute asthma remains the number and sensitivity of b-receptors.35 Ster- inconclusive. Two meta-analyses concluded that oids have been shown to reduce rates of hospital magnesium sulfate might be most beneficial in admission and relapse in acute asthma, as well as severe attacks where the PEF is less than 30–40% improve clinical outcome if used early in an predicted and where a failed response to initial exacerbation.36,37 However, unlike the inhaled conventional treatment is evident.44,45 At the route, clinically significant corticosteroid benefits present time, the BTS/SIGN guideline recommends may not be apparent for 4–6 h after oral or using a single IV bolus of 1.2–2.0 g of MgSO as an parenteral administration. These agents should 4 infusion over 20 min. therefore be used early (within 1 h of presentation). Debate continues regarding the optimal dose of Aminophylline corticosteroids in acute severe asthma. Dose–response relationships in vivo have been difficult to Once widely used, the popularity of parenteral prove. In contrast to the NAEPP guidelines, the aminophylline in asthma exacerbations has diminished ARTICLE IN PRESS

66 F. Chua, D. Lai in recent years. Data from two systematic reviews lished so far has failed to show any ketamine- have shown that IV aminophylline in severe specific benefits above those of standard medical acute asthma does not produce additional broncho- treatment.52 Consequently, its use has not been dilatation above that achieved with standard care advocated. using b2-agonists and corticosteroids, at the expense of considerable side effects such as cardiac 46,47 arrhythmias and vomiting. The BTS/SIGN guide- Non-invasive positive-pressure line suggests that a loading dose of 5 mg/kg body ventilation (NIPPV) weight of aminophylline administered intravenously over 20 min, unless already on oral treat- A decade ago, Meduri and colleagues showed, on a ment, followed by an infusion of 0.5 mg/kg/h may small number of subjects, that NIPPV could be help patients with life threatening or near fatal safely used on asthmatic patients with hypercarbic asthma who have responded poorly to nebulised respiratory failure who had not improved despite bronchodilator therapy. Regular blood level mon- medical treatment.53 In contrast to COPD, it has itoring is an important part of using this drug. proved difficult to establish the therapeutic efficacy of NIPPV in asthma despite this promising start. For obvious reasons, the utility of ‘acute’ Antibiotics NIPPV may be limited by patient acceptance. A recent systematic review identified only one While antibiotics are not routinely recommended randomised controlled trial of NIPPV in asthma for cases of asthma exacerbation, a low threshold that stood up to methodological scrutiny.54 In this for treating bacterial infections (and for preventing study, NIPPV added to medical treatment was secondary infections) appears to be a worthwhile associated with significant improvements in the approach. In the absence of pneumonic changes on rate of hospitalisation, emergency department the chest radiograph, it would seem adequate to discharges, percent predicted FEV and PEF.55 On administer a broad-spectrum penicillin (or macro- 1 the basis of such limited data, the wider use of lide) to patients whose blood results or sputum NIPPV to patients presenting with acute asthmatic characteristics raise the possibility of a bacterial exacerbations has been difficult to recommend. In infection. practice, any trial of NIPPV must not delay A number of other treatments remain controver- endotracheal intubation in patients who are rapidly sial. Evidence that leukotriene antagonists (e.g. deteriorating or have developed somnolence due to montelukast) may be beneficial in treating acute either hypoxaemia or hypercarbia. asthma has yet to emerge. While the biological reasons for using IV epinephrine (adrenaline) as an emergency bronchodilator may appear plausible, in reality, its side effect profile on the cardiovascular Whither from the emergency system negates its potentially beneficial role in department? asthma. A recent study reported that major and minor adverse effects occurred in 30% of patients Most patients survive an acute exacerbation of who received this particular treatment.48 Its use is asthma. This observation is in keeping with the best reserved for selected patients on the ICU in natural tendency of acute asthma to resolve, whom invasive cardiovascular monitoring is avail- providing treatment is prompt. Given that the able. Heliox mixtures (helium:oxygen in a ratio of initial symptomatic worsening usually occurs over a 80:20 or 70:30) have physicochemical properties matter of days or even weeks, there should be that potentially reduce airflow resistance, thereby ample opportunity for medical intervention to enhancing the deposition of nebulised drugs within prevent life-threatening complications. For the asthmatic airways. However, in the absence of majority of patients with an asthma exacerbation, proven benefit over conventional gases (air or management in the community or emergency oxygen), heliox has not been endorsed at the department is adequate. However, up to 7–10% of present time.49 Ketamine, a dissociative anaes- individuals discharged from the latter following thetic that works partly as a non-competitive acute treatment may return within a week with antagonist of N-methyl-D-aspartate (NMDA) and further symptomatic decline.56 In part, this may be muscarinic receptors, has also been shown to the result of objective stabilisation being delayed produce bronchodilation in isolated case reports over initial symptomatic improvement, a phenom- and in one small patient series of acute asth- enon that risks inducing an artificial sense of ma.50,51 Even so, a single randomised trial pub- reassurance in both the patient and physician. ARTICLE IN PRESS

Acute severe asthma: Triage, treatment and thereafter 67

Table 3 Hospital discharge check-list (adapted 3. American Lung Association. Trends in asthma morbidity and mortality. Epidemiology and Statistics Unit, Research and from BTS/SIGN guideline, 2005)27. Program Services. /http://www.lungusa.orgS; July 2006 At discharge, patients admitted with acute severe [accessed September 2006]. 4. Cairns CB. Acute asthma exacerbations: phenotypes and asthma should ideally have: management. Clin Chest Med 2006;27(1):99–108. Treatment with oral and inhaled steroids as well 5. Tattersfield AE, Postma DS, Barnes PJ, Svensson K, Bauer CA, as bronchodilators O’Byrne PM, et al. Exacerbations of asthma: a descriptive Been on discharge medication for 24 h and have study of 425 severe exacerbations. The FACET International had inhaler technique checked Study Group. Am J Respir Crit Care Med 1999;160(2):594–9. PEF 475% best (or predicted) and diurnal PEF 6. Djukanovic R. Asthma: a disease of inflammation and repair. variability o25% J Allergy Clin Immunol 2000;105(2 Pt 2):S522–6. Own PEF meter and written asthma discharge 7. Lamblin C, Gosset P, Tillie-Leblond I, Saulnier F, Marquette plan CH, Wallaert B, et al. Bronchial neutrophilia in patients with Agreed ‘SOS’ PEF values below which specific noninfectious status asthmaticus. Am J Respir Crit Care Med action is required or medical attention sought 1998;157(2):394–402. Advice on specific trigger avoidance if these 8. Tillie-Leblond I, Pugin J, Marquette CH, Lamblin C, Saulnier F, Brichet A, et al. Balance between proinflammatory have contributed to admission cytokines and their inhibitors in bronchial lavage from GP follow up arranged within 2 working days patients with status asthmaticus. Am J Respir Crit Care Med Follow-up in Respiratory clinic within 4 weeks 1999;159(2):487–94. 9. Messer JW, Peters GA, Bennett WA. Causes of death and pathologic findings in 304 cases of bronchial asthma. Dis Chest 1960;38:616–24. At the more severe end of the acute asthma 10. Nowak RM, Pensler MI, Sarkar DD, Anderson JA, Kvale PA, spectrum, the dose–response relationship between Ortiz AE, et al. Comparison of peak expiratory flow and FEV1 admission criteria for acute bronchial asthma. Ann Emerg b -agonist therapy and PEF rate becomes flatter; 2 Med 1982;11(2):64–9. thus, in some patients, continued aggressive treat- 11. Baumann UA, Haerdi E, Keller R. Relations between clinical ment is necessary to ensure recovery of even mid- signs and lung function in bronchial asthma: how is acute range (50–60%) predicted PEF values.57 In one bronchial obstruction reflected in dyspnoea and wheezing? recent British study, 9.8% of patients warded with Respiration 1986;50(4):294–300. acute severe asthma died despite having received 12. Wolfenden LL, Diette GB, Krishnan JA, Skinner EA, Stein- 17 wachs DM, Wu AW. Lower physician estimate of underlying critical care on the ICU. In short, individuals who asthma severity leads to undertreatment. Arch Intern Med require hospitalisation for decompensated asthma 2003;163(2):231–6. are particularly vulnerable to symptomatic worsen- 13. Proceedings of the ATS workshop on refractory asthma: ing in the short term, the reasons for which are not current understanding, recommendations, and unanswered entirely clear. Regardless of the location of care questions. American Thoracic Society. Am J Respir Crit Care Med 2000;162(6):2341–51. after transfer out of the emergency department 14. Gerdtham UG, Hertzman P, Jonsson B, Boman G. Impact (ICU, high dependency unit or general ward), of inhaled corticosteroids on acute asthma hospitalization physicians who attend to such patients must be in Sweden 1978 to 1991. Med Care 1996;34(12): vigilant against the rebound of symptoms. Persis- 1188–98. tent airway inflammation, poorly-resolving infec- 15. Neffen H, Baena-Cagnani C, Passalacqua G, Canonica GW, tion and lingering bronchial hyper-reactivity Rocco D. Asthma mortality, inhaled steroids, and changing asthma therapy in Argentina (1990–1999). Respir Med (manifesting as unpredictable swings in the PEF 2006;100(8):1431–5. amplitude) may all play a role in the pathophysiol- 16. Awadh N, Chu S, Grunfeld A, Simpson K, FitzGerald JM. ogy of asthma relapse. Improved and stable Comparison of males and females presenting with acute physiological measurements (e.g. PEF475% best or asthma to the emergency department. Respir Med predicted), sustained clinical well-being and smooth 1996;90(8):485–9. 17. Gupta D, Keogh B, Chung KF, Ayres JG, Harrison DA, Goldfrad conversion to regular inhaled therapy together with C, et al. Characteristics and outcome for admissions to an agreed asthma care plan (Table 3) are some of adult, general critical care units with acute severe asthma: the factors that will help ensure safe discharge from a secondary analysis of the ICNARC Case Mix Programme hospital. Database. Crit Care 2004;8(2):R112–21. 18. Kearney SE, Graham DR, Atherton ST. Acute severe asthma treated by mechanical ventilation: a comparison of the changing characteristics over a 17 yr period. Respir Med References 1998;92(5):716–21. 19. McFadden Jr ER. Acute severe asthma. Am J Respir Crit Care 1. Hospital Episode Statistics, Department of Health. /www. Med 2003;168(7):740–59. hesonline.nhs.ukS; 2005 [accessed September 2006]. 20. Braman SS, Kaemmerlen JT. Intensive care of status 2. Office for National Statistics. Mortality statistics by cause. asthmaticus. A 10-year experience. J Am Med Assoc 1990; Series DH2, no. 31. The London: Stationery Office; 2005. 264(3):366–8. ARTICLE IN PRESS

68 F. Chua, D. Lai

21. Webb AK, Bilton AH, Hanson GC. Severe bronchial asthma 38. Rodrigo G, Rodrigo C. Inhaled flunisolide for acute severe requiring ventilation. A review of 20 cases and advice on asthma. Am J Respir Crit Care Med 1998;157(3 Pt 1): management. Postgrad Med J 1979;55(641):161–70. 698–703. 22. Bion J. Rationing intensive care. Br Med J 1995;310(6981): 39. Edmonds ML, Camargo Jr CA, Pollack Jr CV, Rowe BH. Early 682–3. use of inhaled corticosteroids in the emergency department 23. Crane J, Pearce N, Burgess C, Woodman K, Robson B, Beasley treatment of acute asthma. Cochrane Database Syst Rev R. Markers of risk of asthma death or readmission in the 12 2003(3):CD002308. months following a hospital admission for asthma. Int J 40. Iseri LT, French JH. Magnesium: nature’s physiologic calcium Epidemiol 1992;21(4):737–44. blocker. Am Heart J 1984;108(1):188–93. 24. Turner MO, Noertjojo K, Vedal S, Bai T, Crump S, Fitzgerald 41. Fawcett WJ, Haxby EJ, Male DA. Magnesium: physiology and JM. Risk factors for near-fatal asthma. A case-control study pharmacology. Br J Anaesth 1999;83(2):302–20. in hospitalized patients with asthma. Am J Respir Crit Care 42. Noppen M, Vanmaele L, Impens N, Schandevyl W. Broncho- Med 1998;157(6 Pt 1):1804–9. dilating effect of intravenous magnesium sulfate in acute 25. de Klerk A, van Schalkwyk E, Williams Z, Lee W, Bardin P. severe bronchial asthma. Chest 1990;97(2):373–6. Risk factors for near-fatal asthma—a case-control study in a 43. Cairns CB, Kraft M. Magnesium attenuates the neutrophil Western Cape teaching hospital. S Afr Med J 2002;92(2): respiratory burst in adult asthmatic patients. Acad Emerg 140–4. Med 1996;3(12):1093–7. 26. Kolbe J, Fergusson W, Vamos M, Garrett J. Case-control 44. Rowe BH, Bretzlaff JA, Bourdon C, Bota GW, Camargo Jr CA. study of severe life threatening asthma (SLTA) in adults: Intravenous magnesium sulfate treatment for acute asthma psychological factors. Thorax 2002;57(4):317–22. in the emergency department: a systematic review of the 27. British Thoracic Society and Scottish Intercollegiate Guide- literature. Ann Emerg Med 2000;36(3):181–90. lines Network. British Guideline on the Management of 45. Alter HJ, Koepsell TD, Hilty WM. Intravenous magnesium as Asthma. A national clinical guideline. /www.brit-thoracic. an adjuvant in acute bronchospasm: a meta-analysis. Ann org.ukS; November 2005 [accessed September 2006]. Emerg Med 2000;36(3):191–7. 28. National Institutes of Health. Guidelines for the Diagnosis 46. Littenberg B. Aminophylline treatment in severe, acute asthma. and Management of Asthma, Expert Panel, Report 2. A meta-analysis. J Am Med Assoc 1988;259(11):1678–84. National Heart, Lung and Blood Institute /http://www. 47. Parameswaran K, Belda J, Rowe BH. Addition of intravenous nhlbi.nih.gov/guidelines/asthma/asthgdin.htmS; 1997 [ac- aminophylline to beta2-agonists in adults with acute cessed October 2006]. asthma. Cochrane Database Syst Rev 2000;(4):CD002742. 29. Martin JG, Shore SA, Engel LA. Mechanical load and 48. Putland M, Kerr D, Kelly AM. Adverse events associated with inspiratory muscle action during induced asthma. Am Rev the use of intravenous epinephrine in emergency depart- Respir Dis 1983;128(3):455–60. ment patients presenting with severe asthma. Ann Emerg 30. Summer W, Elston R, Tharpe L, Nelson S, Haponik EF. Aerosol Med 2006;47(6):559–63. bronchodilator delivery methods. Relative impact on pul- 49. Rodrigo GJ, Rodrigo C, Pollack CV, Rowe B. Use of monary function and cost of respiratory care. Arch Intern helium–oxygen mixtures in the treatment of acute asthma: Med 1989;149(3):618–23. a systematic review. Chest 2003;123(3):891–6. 31. Bowton DL, Goldsmith WM, Haponik EF. Substitution of 50. Sarma VJ. Use of ketamine in acute severe asthma. Acta metered-dose inhalers for hand-held nebulizers. Success and Anaesthesiol Scand 1992;36(1):106–7. cost savings in a large, acute-care hospital. Chest 1992; 51. Petrillo TM, Fortenberry JD, Linzer JF, Simon HK. Emergency 101(2):305–8. department use of ketamine in pediatric status asthmaticus. 32. Rodrigo GJ, Rodrigo C. Continuous vs intermittent beta- J Asthma 2001;38(8):657–64. agonists in the treatment of acute adult asthma: a 52. Howton JC, Rose J, Duffy S, Zoltanski T, Levitt MA. systematic review with meta-analysis. Chest 2002;122(1): Randomized, double-blind, placebo-controlled trial of in- 160–5. travenous ketamine in acute asthma. Ann Emerg Med 33. Travers AH, Rowe BH, Barker S, Jones A, Camargo Jr CA. The 1996;27(2):170–5. effectiveness of IV beta-agonists in treating patients with 53. Meduri GU, Cook TR, Turner RE, Cohen M, Leeper KV. acute asthma in the emergency department: a meta- Noninvasive positive pressure ventilation in status asthma- analysis. Chest 2002;122(4):1200–7. ticus. Chest 1996;110(3):767–74. 34. Stoodley RG, Aaron SD, Dales RE. The role of ipratropium 54. Ram FS, Wellington S, Rowe B, Wedzicha JA. Non-invasive bromide in the emergency management of acute asthma positive pressure ventilation for treatment of respiratory exacerbation: a metaanalysis of randomized clinical trials. failure due to severe acute exacerbations of asthma. Ann Emerg Med 1999;34(1):8–18. Cochrane Database Syst Rev 2005;20(3):CD004360. 35. Barnes PJ. Scientific rationale for inhaled combination 55. Soroksky A, Stav D, Shpirer I. A pilot prospective, rando- therapy with long-acting beta2-agonists and corticosteroids. mized, placebo-controlled trial of bilevel positive airway Eur Respir J 2002;19(1):182–91. pressure in acute asthmatic attack. Chest 2003;123(4): 36. Rowe BH, Spooner C, Ducharme FM, Bretzlaff JA, Bota GW. 1018–25. Early emergency department treatment of acute asthma 56. McFadden Jr ER, Elsanadi N, Dixon L, Takacs M, Deal EC, with systemic corticosteroids. Cochrane Database Syst Rev Boyd KK, et al. Protocol therapy for acute asthma: 2001(1):CD002178. therapeutic benefits and cost savings. Am J Med 1995;99(6): 37. Rowe BH, Spooner CH, Ducharme FM, Bretzlaff JA, Bota GW. 651–61. Corticosteroids for preventing relapse following acute 57. McFadden Jr ER, Strauss L, Hejal R, Galan G, Dixon L. exacerbations of asthma. Cochrane Database Syst Rev Comparison of two dosage regimens of albuterol in acute 2001(1):CD000195. asthma. Am J Med 1998;105(1):12–7. Anaesthesia, 2006, 61, pages 1058–1063 doi:10.1111/j.1365-2044.2006.04801.x ......

The role of magnesium as an adjuvant during general anaesthesia

K. Gupta,1 V. Vohra2 and J. Sood3

1 Registrar, 2 Senior Consultant and Vice Chairperson, 3 Senior Consultant and Chairperson, Department of Anaesthesiology, Pain and Perioperative Medicine, Sir Ganga Ram Hospital, Old Rajinder Nagar, New Delhi-110060, India

Summary Magnesium sulphate is used extensively in the treatment of eclampsia, and is also used to treat refractory arrhythmias, asthma, myocardial ischaemia and acute respiratory failure. We studied the interaction between magnesium sulphate and the anaesthetic agents propofol, rocuronium bromide and fentanyl citrate. This randomised, double blind study was conducted in 50 patients. The ) magnesium group A (n = 25) received 30 mg.kg 1 magnesium sulphate before induction of ) anaesthesia and 10 mg.kg 1 continuously intra-operatively until the end of surgery. Group B (n = 25) received the same volume of isotonic saline. Propofol, rocuronium and fentanyl infusions were started and the patients lungs’ were ventilated with 33% oxygen in nitrous oxide. Anaesthetic depth was maintained at a bispectral index value of between 40 and 60. Muscle relaxation was maintained at a train-of-four count of 1 throughout surgery using neuromuscular monitoring. The fentanyl infusion was titrated to haemodynamic variables: heart rate and blood pressure. We concluded that magnesium sulphate has anaesthetic, analgesic and muscle relaxation effects and signiﬁcantly reduces the drug requirements of propofol, rocuronium and fentanyl during anaesthesia.

...... Correspondence to: Dr Vijay Vohra E-mail: [email protected] Accepted: 26 May 2006

Magnesium is the fourth most abundant cation in the a 60% reduction in minimum alveolar concentration of body and the second most abundant intracellular cation halothane was shown in magnesium-treated rats [7], [1, 2]. It activates many of the enzyme systems involved in which was thought to be due to a central effect of the ion. energy metabolism and acts as a natural calcium antag- Magnesium decreases the amount of acetylcholine onist, regulating calcium access into the cell [3]. It has released from the motor nerve terminal, leading to been suggested that magnesium has the potential to treat diminished excitability of the muscle ﬁbre itself and and prevent pain by acting as an antagonist at the reduction in the amplitude of the end plate potential. It N-methyl-D aspartate (NMDA) receptors [2]. therefore potentiates the neuromuscular blockade pro- The role of magnesium in reducing analgesic require- duced by non-depolarising neuromuscular blocking ments during the postoperative period has been demon- agents [8]. It decreases catecholamine release from the strated previously [2]. Magnesium administration also adrenal medulla and adrenergic nerve endings. It also leads to a signiﬁcant reduction in analgesic consumption reduces postoperative shivering and obtunds the pressor during the intra-operative period [3]. response to laryngoscopy and intubation by decreasing Historically, magnesium sulphate has been proposed catecholamine release [1]. as a general anaesthetic [4, 5]. Although magnesium was Magnesium has been extensively used prophylactically regarded as a central nervous system depressant, its for terminating seizures in eclamptic patients [2]. It has anaesthetic effect was shown to result from cerebral also been used in the treatment of ventricular arrhythmias hypoxia following progressive respiratory and cardiac associated with epinephrine, digitalis and bupivacaine depression, whereas on maintaining respiratory support no administration [2]. The use of magnesium in asthma, central nervous system depression was seen even at a very tocolysis, phaeochromocytoma, myocardial ischaemia and high serum concentrations of magnesium [6]. However, acute respiratory failure is under investigation [2].

2006 The Authors 1058 Journal compilation 2006 The Association of Anaesthetists of Great Britain and Ireland Anaesthesia, 2006, 61, pages 1058–1063 K. Gupta et al. Magnesium as an adjuvant ...... Æ

The purpose of this study was to assess the efficacy of stimulus train-of-four neuromuscular stimulation was magnesium sulphate in a general anaesthetic regimen measured. The time to produce a 80% single twitch consisting of propofol, fentanyl citrate and rocuronium depression (T1 = 20%) following rocuronium bromide ) bromide and to record any untoward effects of this drug (0.6 mg.kg 1) was recorded. Orotracheal intubation was in the peri-operative period. performed after complete (T1 = 0%) single twitch depression. Continuous neuromuscular monitoring was commenced using ‘TOF guard’ using a train-of-four Methods (TOF) stimulus every 12 s. Anaesthesia was maintained This randomised, double-blind study was conducted using propofol, rocuronium bromide and fentanyl infu- following approval by the hospital ethics committee and sions. Propofol infusions were started at a rate of ) ) ) after obtaining written informed consent from patients. 10 mg 1.kg 1.h 1 and titrated to maintain BIS values Fifty ASA grade I–II patients between the ages 16 and between 40 and 60. 60 years, undergoing spinal surgery (lumbar laminecto- The fentanyl infusion was titrated according to standard my, lumbar or lumbosacral discectomy) were randomly clinical signs and haemodynamic measurements. Signs of allocated to one of two groups. Group A (n = 25) inadequate analgesia were defined as an increase in heart ) received 30 mg.kg 1 magnesium sulphate and Group B rate and mean arterial pressure of > 20% from baseline (n = 25) received an equal volume of saline. values. Muscle relaxation produced by rocuronium Exclusion criteria included major hepatic, renal or bromide infusion was titrated to maintain a TOF count cardiovascular disease, asthma, chronic pulmonary disease of 1 and a twitch height of < 10%. or haematological disorders, any known allergy to The patients’ lungs were ventilated mechanically with a magnesium sulphate or any other study drugs, and 33% oxygen ⁄ 66% nitrous oxide mixture to maintain pregnancy. Patients receiving treatment with calcium adequate oxygenation and end tidal carbon dioxide levels channel blockers, opioid or anticoagulants, were not between 35 and 40 mmHg (4.6–5.3 kPa). Temperature included in the study. was monitored using an oesophageal probe and normo- Prior to induction of anaesthesia, routine monitoring of thermia was maintained throughout surgery. Approxi- pulse oximetry, non-invasive blood pressure and electro- mately 30 min before the end of surgery, the rocuronium cardiography was commenced and an intravenous line bromide infusion was discontinued. Patients were secured. Muscle relaxation was monitored using the allowed to recover spontaneously, until the return of neuromuscular monitor (TOF guard, Organon teknika T1 = 25% or > 2 responses on neuromuscular monit- ) NV, Turnhout, Belgium). Surface electrodes were posi- oring. A combination of atropine 0.02 mg.kg 1 and ) tioned over the ulnar nerve at the wrist. The depth of neostigmine 0.05 mg.kg 1 was administered to reverse anaesthesia was monitored using the Bispectral index the neuromuscular blockade. (Aspect Medical System, Newton, MA). The target range Time for return of T1 = 25% was recorded. Propofol for BIS was 40–60. Serum magnesium levels were was discontinued on skin closure and patients were measured pre- and postoperatively, 1 h after stopping allowed to recover. The patients’ tracheas were extubated the magnesium sulphate infusion. In the recovery unit, when the BIS value reached 80 and the TOF ratio was deep tendon reflexes, urine output and respiratory rate > 0.7. Times to reach BIS 80 were recorded as recovery were monitored. time. Following transfer of the patient to the recovery area, Group A received 15% magnesium sulphate patients were assessed neurologically for signs of hyper- ) 30 mg.kg 1 (from a blindly labelled vial) as a slow magnesaemia. Data were analysed using Student’s t-test. intravenous infusion prior to induction of anaesthesia and ) ) 10 mg.kg 1.h 1 by continuous infusion during surgery. Results The same volume of isotonic saline was administered to the control group B. The demographic profile of the two groups was similar Following pre-oxygenation for 3 min, anaesthesia was (Table 1). Data are presented as mean (SD). ) ) induced using fentanyl 1 lg.kg 1 and propofol 1 mg.kg 1 The mean age was 43.12 (12.2) years in group A and administered over 15 s. Bispectral index values were 49.36 (10.19) years in group B. The mean weight was monitored; once the BIS values had stabilised, if the 65.76 (12.02) kg in group A and 70 (10.9) kg in group B. ) values were > 60, further increments of propofol 20 mg The propofol induction dose was 1.28 (0.21) mg.kg 1 ) were given over 5 s and the BIS values continued to be in group A and 1.37 (0.28) mg.kg 1 in control group B, monitored. This was repeated until the BIS values which did not achieve statistical significance (p = 0.42). reached 60 and the time to reach this level was recorded. Time to reach a BIS value of 60 (T1) was 63.6 (8.96) s Following induction of anaesthesia, a supramaximal in the magnesium group A compared to 84.6 (12.33) s in

2006 The Authors Journal compilation 2006 The Association of Anaesthetists of Great Britain and Ireland 1059 K. Gupta et al. Magnesium as an adjuvant Anaesthesia, 2006, 61, pages 1058–1063 ...... Æ ......

Table 1 Demographic profile. 41.6 45 40 Group A Magnesium Control Group B Group A, Group B, 35 (3.79) 28.12 mean (SD) mean (SD) p-value 30 25 Age; years 43.12 (12.28) 49.36 (10.19) 0.06 (3.11) Weight; kg 65.76 (12.02) 70.00 (10.91) 0.20 20 Time (min) Sex; male ⁄ female 19 ⁄ 614⁄ 11 0.14 15 9.84 10 7.52 (1.14) 5 (1.16) the control group B, which was statistically significant 0 (p < 0.001). Mean time (T3) Mean time (T4) Time taken for 80% single twitch height depression Figure 2 Mean (SD) time to recovery from muscle relaxation (T2) following administration of rocuronium bromide (T3 – time to return of twitch height to 20%) is significantly ) 0.5 mg.kg 1 was not significantly different between the delayed in Magnesium Group A compared to Control Group B two groups: group A with magnesium (182 (38) s) and (p < 0.001). Recovery from propofol (T4 – time to BIS 80) control group B (179 (28) s) (p = 0.75) (Fig. 1). is significantly delayed in Magnisium Group A compared to Group B (p < 0.001). Time for recovery of the twitch height to 25% (T3) following stopping the rocuronium bromide infusion was significantly shorter in the control group B (28 (3.11) s) 8 than in the magnesium group A (41 (3.79) min) 7 6.87 6.94 6.19 (0.24) (0.24) (p < 0.001). Time to reach a BIS value of 80 following 6 5.86 5.44 (1.13) (0.87) stopping the propofol infusion (T4) was significantly 5 (1.05) longer in the magnesium group A (9.84 (1.14) min) 4 3.37 3.33 3.21 than in the control group group B (7.52 (1.16) min) 3 (0.55) (0.74) 2.95 2.96 (0.50) (0.53) (p < 0.001) (Fig. 2). Dose (mg/kg/h) 2 (0.51) Group A The dose of propofol required to maintain an adequate 1 Group B depth of anaesthesia (BIS = 40–60) was significantly 0 lower in the magnesium group A than in the control 1st h 2nd h 3rd h 4 h 5 h Time (h) group B during the 1st, 2nd, 3rd, 4th and 5th hours (p < 0.001) (Fig. 3). Figure 3 Mean (SD) hourly propofol consumption to maintain The dose of rocuronium bromide required to maintain adequate depth of anaesthesia (BIS 40-60) is significantly lower ) ) adequate neuromuscular blockade (lg.kg 1.h 1) was in Magnesium Group A than Control B in first 2 h (p < 0.001). significantly lower in the magnesium group A than in the control group B (p < 0.001) (Fig. 4). 600 Group A The fentanyl dose required intra-operatively was Group B 500 significantly lower in the magnesium group A than in 434 435 431 418 400 414 the control group B (p < 0.05) (Fig. 5). (46) (41) (40) (10) (7) g/kg/h) µ 300 292 278 254 254 182 (63) (67) 246 (63) (56) (56) Group A 179 200 200 Dose ( Group B 100 150 (28) (38) 0 100 84.6 1st h 2nd h 3rd h 4th h 5th h

Time (s) 63.6 Time (h) 50 (12.33) (8.96) Figure 4 Mean (SD) hourly recoronium for adequate muscle 0 relaxation (to maintain a TOF count of 1) is significantly lower Mean time (T1) Mean time (T2) in Magnesium Group A than Control Group B in first 2 h (p < 0.001). Figure 1 Mean (SD) time to induction by propofol (T1 – time to BIS 60) is significantly faster in Magnesium Group A than In group A, serum magnesium levels increased from the Control Group B (p < 0.001). Time taken for muscle relaxation )1 )1 (T2 – time to 80% single twitch depression) is similar to two pre-operative level (mean) 1.99 mg.dl to 2.17 mg.dl , groups (p = 0.75). whereas in the control group B levels remained stable at

2006 The Authors 1060 Journal compilation 2006 The Association of Anaesthetists of Great Britain and Ireland Anaesthesia, 2006, 61, pages 1058–1063 K. Gupta et al. Magnesium as an adjuvant ...... Æ

0.5 room. Telci et al. [11] studied the interaction between Group A magnesium sulphate and propofol and concluded that 0.45 Group B induction of anaesthesia (BIS = 60) was achieved more 0.4 rapidly in the presence of magnesium. Recovery from propofol was also slightly delayed in the magnesium 0.35 0.337 0.34 g/kg/h) 0.33 group. Choi et al. [12] concluded that intravenous µ (0.12) 0.318 (0.09) (0.06) 0.31 0.3 (0.10) (0.04) administration of magnesium sulphate reduced propofol 0.268 Dose ( infusion requirements, whereas Schulz-Stubner et al. [13] 0.25 (0.05) 0.253 0.25 0.25 0.24 (0.04) (0.06) (0.07) (0.04) observed that propofol consumption remained unchanged 0.2 after using magnesium sulphate. The effects of magnesium on propofol consumption 0.15 1st h 2nd h 3rd h 4th h 5th h could be related to a sedative effect of magnesium, but Time (h) evidence for such an effect is conflicting. Magnesium has Figure 5 Hourly mean (SD) fentanyl consumption for adequate been reported to produce general anaesthesia and to analgesia is significantly lower in Magnesium Group A than enhance the activity of local anaesthetic agents [14–16]. In Control Group B in first 2 h (p < 0.05). those studies, a depressant effect on the CNS of animals injected with magnesium salts was reported. However, ) Aldrete and colleagues suggested that this anaesthetic state Table 2 Magnesium levels (mg.dl 1). was actually a sleep-like state caused by cerebral hypoxia from progressive respiratory and cardiac depression [17]. Group A Group B In our study, both groups of patients were given )1 Pre-operative 1.99 2.02 0.5 mg.kg rocuronium bromide to produce neuromus- Postoperative 2.17 2.06 cular blockade; the onset time for neuromuscular blockade was unaffected by the presence of magnesium sulphate and was similar between the two groups. The ) 2.02 (0.04) mg.dl 1, which is not statistically significant time from the cessation of rocuronium bromide infusion (Table 2). to 25% single twitch recovery was, however, significantly greater in the magnesium group. Time taken from T1 (when neostigmine was administered to reverse neuro- Discussion muscular blockade) to reach TOF 0.7 was similar Balanced anaesthesia incorporates four components: between the two groups. unconsciousness, muscle relaxation, analgesia and amne- Kussman et al. [18] studied the effects of larger doses of ) sia. We studied the possible effects of magnesium sulphate magnesium sulphate (60 mg.kg 1) on the neuromuscular ) in reducing the anaesthetic requirements during total blocking effects of rocuronium bromide 0.6 mg.kg 1 intravenous anaesthesia using propofol, fentanyl citrate during isoflurane anaesthesia. Mean onset time was and rocuronium bromide. Our results demonstrate a unaffected but time to initial 10% and 25% recovery significant reduction in the requirements for each of these from neuromuscular blockade was significantly pro- drugs during balanced anaesthesia in the presence of a longed. Tfuchs [19] investigated the interaction between ) magnesium sulphate infusion. magnesium sulphate 40 mg.kg 1 intravenously and vecu- All patients underwent similar surgery performed by ronium bromide. The ED50 and ED90 of vecuronium the same group of surgeons. No haemodynamic instabil- bromide in the presence of magnesium sulphate were ity was seen associated with magnesium administration. 25% lower than controls. Duration of action was also An objective, quantitative measure of the anaesthetic state prolonged in the magnesium group. (BIS) was used to guide anaesthetic requirements [9, 10] Telci et al. [11] studied the interaction between and to determine the end points. Induction of anaesthesia rocuronium bromide and magnesium sulphate. They using propofol was more rapid in the presence of found that the recovery time was shorter in the control magnesium; however, recovery was found to be slower. group than in the magnesium group. The time from Magnesium did not alter the induction dose of propofol cessation of the rocuronium infusion to T1 = 25% was required, although the induction time was shorter. The significantly longer in the magnesium group than in the rate of fall of the BIS in the two groups was different, control group. Similarly, Schulz-Stubner et al. [13] con- leading to dissimilar induction times. The recovery time cluded from his study that mivacurium consumption was was shorter in the control group B. In both groups, BIS markedly reduced in patients receiving magnesium sul- was > 95 by the time the patient entered the recovery phate infusion.

2006 The Authors Journal compilation 2006 The Association of Anaesthetists of Great Britain and Ireland 1061 K. Gupta et al. Magnesium as an adjuvant Anaesthesia, 2006, 61, pages 1058–1063 ...... Æ ......

In our study, hourly consumption of fentanyl citrate study, was not seen to produce any adverse haemo- was significantly reduced in the magnesium group dynamic effects. compared to the control group. The analgesic effects of magnesium have been demonstrated in several animal References [20, 21] and human studies [22]. Koinig and colleagues [23] studied patients with identical levels of surgical 1 Fawcett WJ, Haxby EJ, Male DA. Magnesium physiology stimulation and evaluated the effect of magnesium on and pharmacology. British Journal of Anaesthesia 1999; 83: peri-operative analgesic requirements. Their data demon- 302–20. strated that magnesium could be used as an adjuvant to 2 James MFM. Clinical use of magnesium infusion in anaes- peri-operative analgesic management by reducing the thesia. Anesthesia and Analgesia 1992; 74: 129–36. 3 Iseri LT, French JH. Magnesium nature physiologic calcium fentanyl requirements. Recently, in a study by Schulz- blocker. American Heart Journal 1984; 108: 188–93. Stubner [13], magnesium sulphate given as a bolus 4 Meltzer SJ, Anes J. Physiological and pharmacological following induction of anaesthesia produced a significant studies on magnesium salts. The toxicity of intravenous reduction in remifentanil consumption during general injection, in particular the effects upon the centers of the anaesthesia. In our study as in previous reports, the medulla. American Journal of Physiology 1906; 15: 387–405. fentanyl requirements were titrated according to haem- 5 Peck CH, Meltzer SJ. Anaesthesia in human beings by odynamic changes. Other work using doses of magnes- intravenous injection of magnesium sulphate. Journal of the ium similar to ours have been shown to produce American Medical Association 1916; 67: 1131–3. reductions in analgesic requirements both during and 6 Somjen G, Hilmy M, Stephen CR. Failure to anaesthetize after surgical interventions [13]. human subjects by intravenous administration of magnesium Significantly lower requirements of propofol, fentanyl sulphate. Journal of Pharmacology and Experimental Therapeu- tics 1966; 154: 652–9. citrate and rocuronium bromide were recorded during 7 Thompson SLO, Mosciki JC, Difazio CA. The anaesthetic each measurement period in the magnesium group A. contribution of magnesium sulphate and ritodrine hydro- When the hourly consumption rates of each drug were chloride in rats. Anesthesia and Analgesia 1988; 20: 1273–5. compared between groups, significant reductions were 8 Savarenese JJ, Caldwell JE, Lien CA, et al. Pharmacology of observed in the first and second hours for propofol, fentanyl muscle relaxants and their antagonists. In: Miller RD, ed. and rocuronium requirements in the magnesium group. Anaesthesia, 5th edn. Philadelphia: Churchill Livingstone, We found no incidence of respiratory depression, 2000: 463. electrocardiographic changes, oliguria, delayed ankle jerk 9 Lui J, Singh H, White PF. Electroencephalographic bi- reflex or bradycardia in any of our patients. Magnesium spectral index correlate with intraoperative recall and depth levels were within safe limits following the administration of propofol induced sedation. Anesthesia and Analgesia 1997; ) of 30 mg.kg 1 magnesium sulphate prior to induction of 84: 185–9. ) ) 10 Glass PSA, Bloom M, Kearse L, et al. Bispectral analysis anaesthesia and a continuous infusion of 10 mg.kg 1.h 1 measures sedation and memory effects of propofol, mida- intra-operatively. In our study, baseline magnesium levels zolam, isoflurane, and alfentanil in healthy volunteers. in the two groups were similar. Levels increased in group Anesthesiology 1997; 86: 836–47. )1 A from 1.99 to 2.17 mg.dl but this rise is within the 11 Telci L, Esen F, Erdon T, et al. Evaluation of effects of normal physiological range. Urinary excretion of mag- magnesium sulphate in reducing intraoperative anaesthetic nesium is very rapid and is a linear function of plasma requirements. British Journal of Anaesthesia 2002; 89: 594– magnesium concentration [24]. Therefore, hypermagnes- 8. aemia does not occur unless renal dysfunction is present. 12 Choi JC, Yoon KB, Wn DJ, et al. Intravenous magnesium Fuchs et al. [19] concluded that plasma magnesium sulphate administration reduces propofol infusion require- ) concentration increased from 0.9 to 1.08 mmol.1 1, ments during maintenance of propofol N2O anaesthesia. ) 15 min after magnesium sulphate 40 mg.kg 1 infusion. Anesthesiology 2002; 97: 1137–41. 13 Schulz-Stubner S, Wettmann G, Reyle-Hahn SM, et al. Telci et al. [11] and Schulz-Stubner et al. [13] demon- Magnesium as part of balanced general anaesthesia with strated that magnesium infusion was well tolerated, with propofol, remifentanil and mivacurium: a double blind, no signs of muscle weakness. In our study, serum randomized prospective study in 50 patients. European Journal )1 magnesium levels rose from 1.99 to 2.17 mg.dl in the of Anaesthesiology 2001; 18: 723–9. )1 magnesium group and were stable at 2.02 (0.04) mg.dl 14 Meltzer SJ, Auer J. Physiological and pharmacological in the control group. studies on magnesium salts. II. Narcotizing effects of mag- In conclusion, we found that the administration of nesium salts upon nerve fibers. American Journal of Physiology magnesium sulphate led to a significant reduction in the 1906; 16: 233–8. anaesthetic drug requirements used during total intraven- 15 Countinho EM. Calcium, magnesium and local anesthesia. ous anaesthesia. Magnesium, in the dosage used in this Journal of General Physiology 1966; 49: 845–6.

2006 The Authors 1062 Journal compilation 2006 The Association of Anaesthetists of Great Britain and Ireland Anaesthesia, 2006, 61, pages 1058–1063 K. Gupta et al. Magnesium as an adjuvant ...... Æ

16 Peck CH, Meltzer SJ. Anesthesia in human beings by 21 Woolf CJ, Thompson SWN. The induction and mainten- intravenous injection of magnesium sulphate. Journal of the ance of central sensitization is dependent on N-methyl D American Medical Association 1916; 67: 1131–3. aspartic acid receptor activation: implications for the treat- 17 Aldrette AJ, Barbes DR, Aikawa K. Does magnesium pro- ment of post injury pain and hypersensitivity states. Pain duce anesthesia? Evaluation of its effects on the cardiovas- 1991; 44: 293–9. cular and neurologic systems. Anesthesia and Analgesia 1968; 22 Tramer MR, Schneider J, Marti RA, et al. Role of mag- 47: 428–33. nesium sulfate in postoperative analgesia. Anesthesiology 18 Kussman B, Shorten G, Uppugton J, et al. Administration of 1996; 84: 340–7. magnesium sulphate before recuronium: effects on speed 23 Koinig H, Wallner T, Marhofer P, et al. Magnesium sul- of onset and duration of neuromuscular block. British Journal phate reduced intra- and postoperative analgesic require- of Anaesthesia 1997; 79: 122–4. ments. Anesthesia and Analgesia 1998; 87: 206–10. 19 Buder TF, Wilder Smith OHG, Borgeat A, et al. Interaction 24 Chesley LC. Parenteral magnesium sulphate and the distri- of magnesium sulphate with vecuronium bromide-induced bution plasma levels, and excretion of magnesium. American neuromuscular block. British Journal of Anaesthesia 1995; 74: Journal of Obstetrics and Gynaecology 1979; 133: 1–7. 405–9. 20 Feria M, Abad F, Sanchez A, et al. Magnesium sulphate injected subcutaneously suppresses autonomy in peripherally deafferented rats. Pain 1993; 53: 287–93.

2006 The Authors Journal compilation 2006 The Association of Anaesthetists of Great Britain and Ireland 1063 MINERVA ANESTESIOL 2006;72

A C Anaesthetic managementI in asthma S. M. BURBURAN 1, 2D, D. G. XISTO 2, P. R. M. ROCCO 2 E Anaesthetic management in asthmatic patients 1®Division of Anaesthesiology has been focused in avoiding bronchocon- Department of Surgery striction and inducing bronchodilation. Faculty of Medicine, Federal University However, the definition of asthma has changed of Rio de Janeiro, Rio de Janeiro, Brazil M 2 over the past decade. Asthma has been defined Laboratory of Pulmonary Investigation as a clinical syndrome characterized by an Carlos ChagasT Filho Biophysics Institute inflammatory process that extends beyond the Federal University of Rio de Janeiro central airways to the distal airways and lung Rio de Janeiro, Brazil parenchyma. With this concept in mind, and knowing that asthma is a common disorderA H with increasing prevalence rates and severity worldwide, a rational choice of anaesthetic agents and procedures is mandatory. VThus, we ing the risk of bronchospasm during anaes- pursued an update on the pharmacologic and thesiaG induction. Propofol yields central air- technical anaesthetic approach for the asth- way dilation, and is more reliable than etomi- matic patient. When feasible, regional anaes- dateI or thiopental. Halothane, enflurane and thesia should be preferred becauseR it reduces isoflurane are potent bronchodilators and can airway irritation and postoperative complica- be helpful even in status asthmaticus. Sevo- tions. If general anaesthesia is unavoidable, a flurane has shown controversial results in asth- laryngeal mask airway is safer than endotra-Rmatic patients. Vecuronium, rocuronium, cheal intubation. LidocaineE inhalation, alone or cisatracurium, and pancuronium do not induce combined with albuterol, minimizes histamine- bronchospasm, while atracurium and mivac- induced bronchoconstriction. Propofol and urium can dose-dependently release histamine ketamine inhibit bronchoconstriction, decreas-Y and should be cautiously administered in those N patients. Further knowledge about the sites of Supported by The Centres of Excellence Program (PRO- action of anaesthetic agents in the lung, allied NEX-FAPERJ), BrazilianI Council for Scientific and Technolo- to our understanding of asthma pathophysi- gical Development (CNPq), Carlos Chagas FilhoP Rio de Janeiro ology will establish the best anaesthetic State Research Supporting Foundation (FAPERJ). approach for people with asthma. Acknowledgments.—The authors would like to express their gratitude to A. Benedito da Silva and S. Cezar da Silva Key words: Asthma - Anaesthesia - Anaesthe- Junior for theirM technical assistance.O tics - Bronchial spasm - Bronchial hypereac- Received January 26, 2006. tivity. Accepted for publication October 24, 2006.

Address reprint requestsC to: P. R. M. Rocco, MD, PhD, sthma is a major public health issue with Laboratório de Investigação Pulmonar, Instituto de Biofísica Ahigh and increasing prevalence rates1 Carlos Chagas Filho, C.C.S., Universidade Federal do Rio de Janeiro, Ilha do Fundão. CEP. 21.949-900 – Rio de Janeiro, RJ, and a concomitant increase in morbidity and Brazil. E-mail: [email protected] mortality.2 Studies have showed that preva-

Vol. 72, 2006 MINERVA ANESTESIOLOGICA 1 BURBURAN ANAESTHETIC MANAGEMENT IN ASTHMA

lence of lifetime asthma among adults is 11%,3 In view of these new pathophysiological being even higher among children.4 These findings we attempted this review to outline data contribute to make challenging adverse available data and the criteria for the anaes- events due to asthma a widespread situation thetic management in asthmatic patients. in anaesthesiological practice. A Therefore, in order to avoid increasing the risk of perioperative complications, a good Methods understanding of asthma pathophysiology, an adequate preoperative evaluation, and A computerized searchC of the English- optimisation of patient’s condition, allied to language literature Iof PUBMED, between the best pharmacological and technical 1995 and 2005 was conducted using the approach are imperative. terms airway obstruction, asthma, bronchial The pathophysiological hallmark of asth- hyper-reactivity, anaesthesia and anaes- ma is a reduction in airway diameter due to thetics. VariousD combinations of the terms the contraction of smooth muscle, vascular were used to maximize the results. congestion, oedema of the bronchial wall, Bibliographies of original research, com- and tenacious secretions.5 The chronic inflam- mentaries,E textbooks, and symposia were matory process leads to tissue injury and sub- reviewed for additional® relevant references. sequent reorganization. The term “airway These publications were abstracted and remodelling” is widely used to refer to the compiled into tabular form under types of development of those structural changes,6, 7 studyM design. Two-hundred twenty-two arti- such as: epithelial shedding, subepithelial cles were selectedT for critical review. Of fibrosis, increased numbers and volume of the articles excluded, most were reports mucous cells in the epithelium, airway smooth that described anaphylactic reactions, muscle hyperplasia and hypertrophy, and bronchial hyper-responsiveness, and bron- A8 increased vascularization of the airway wall. chospasm inH nonasthmatic patients. These changes in the extracellular matrix, smooth muscle, and mucous glands Vhave the capacity to influence airway function and Anaesthetic management strategy reactivity in asthmatic patients manifested by G a decline in forced expiratory volume in 1 s PreoperativeI evaluation (FEV ) and bronchial hyper-responsiveness.R 9 1 Bronchospasm and mucous plugging The anaesthesiologist’s responsibility starts obstruct both inspiratory and expiratory air- at the preoperative phase with the evalua- flow. Resistance to expiratoryE airflow resultsRtion of the patient’s pulmonary function. in positive alveolar pressures at the end of Warner et al.15 observed that the frequency of expiration, that causes air-trapping and hyper- perioperative bronchospasm and laryn- inflation of the lungs and thorax, increasedY gospasm was surprisingly low in asthmatic work of breathing,N and alterations in respi- patients. They identified 3 factors correlated ratory muscle function.I Airflow obstruction is with perioperative bronchospasm: the use not uniform and the mismatchingP of ventila- of antibronchospastic medications; recent tion to perfusion occurs leading to changes in symptomatic exacerbation; and recent visit arterial blood gases.10, 11 to a medical facility for treatment of asthma. The definitionM of asthma has changed over Therefore, they have concluded the follow- the past decade. Asthma hasO been defined ing statements: 1) persons with asthma but no as a clinical syndrome characterized by an symptoms are at low risk for severe morbid- inflammatory process that extends beyond ity from anaesthesia; 2) persons with asth- the central airways toC the distal airways and ma are, however, at a low but increased risk the lung parenchyma. Small airways have for severe morbidity; 3) adverse outcomes recently been recognized as a site of airflow from bronchospasm occur in patients with obstruction and hyper-responsiveness.12-14 no previous history of asthma.16

2 MINERVA ANESTESIOLOGICA Vol. 72, 2006 ANAESTHETIC MANAGEMENT IN ASTHMA BURBURAN

In view of this, the approach to the asth- means of assessing presence and severity of matic patient should include a detailed history airway obstruction as well as the degree of of the patient’s experience with reactive air- reversibility in response to bronchodilator way disease, searching for the following: 1) therapy. An increase of 15% in FEV1 is con- a recent upper respiratory infection; 2) aller- sidered clinically significant.20 gies; 3) possible precipitating factors for asth- In asthmatic patients, chest Aroentgenogram ma; 4) use of medications, including drugs even in acute asthma attacks often shows that could precipitate the attack, as well as normal findings.17 those used to prevent an attack, and 5) the C occurrence of dyspnoea at night or in the Preoperative management of asthma early morning hours. Furthermore, to better I understand a patient’s bronchial reactivity, it In patients with reversible airway obstruc- is important to know whether he or she can tion and bronchial reactivity, preoperative β tolerate cold air, dust or smoke and if he has treatment withD 2-adrenergic agonists and β ever undergone general anaesthesia under corticosteroids should be considered. 2- tracheal intubation.17 Documentation of an adrenergic agonists have been shown to episode of status asthmaticus requiring intu- attenuateE the reflex bronchoconstriction fol- bation portends of a difficult perioperative lowing endotracheal ®intubation. Even though, course.10 significant bronchoconstriction and wheezing Drugs are commonly associated with the occurs following intubation.17, 21, 22 Combined treatment with corticosteroids induction of acute episodes of asthma and it Mβ is important to recognize drug-induced and a 2-adrenergicT agonist can improve pre- bronchial narrowing because its presence is operative lung function and decrease the often associated with great morbidity. Even incidence of wheezing following endotra- β cheal intubation.21, 23 Concerns about negative the selective 1-adrenergic antagonists regularly obstruct the airways in individuals withA effects of perioperativeH treatment with corti- asthma and should be avoided.5 costeroids in terms of wound healing and The triad of bronchial asthma, nasal poly- infection were not supported by studies in posis, and intolerance to aspirin or aspirin-likeV asthmatic patients receiving prophylactic chemicals are designated aspirin-induced treatmentG with corticosteroids perioperative- asthma (AIA) or Samter’s syndrome. In these ly Iand there is evidence that asthmatic patients, marked cross-sensitivityR with nons- patients who are treated with corticosteroids teroidal anti-inflammatory drugs may pre- can undergo surgical procedures with a low cipitate severe bronchospasm and adverse incidence of complications.24 Thus, preop- reactions. Hence, it is importantE to refer theseRerative treatment with combined corticos- patients to allergy clinics to evaluate possible teroids [methylprednisolone (40 mg/day oral- analgesic cross reactivity and intolerance to ly)] and salbutamol minimizes intubation- anaesthetic agents.18 Y evoked bronchoconstriction in patients with Physical examinationN of the lungs may be reversible airway obstruction or with a history normal or reveal wheezing and/or other of severe bronchial hyper-reactivity.17, 21 I 11 adventitial sounds. Preoperative wheezingP is According to Enright, preoperative man- predictive of a difficult perioperative course. agement in asthmatics should include the fol- Indeed, if a severe asthmatic history and aus- lowing measures: cultatory wheezing are encountered during 1) Bronchospasm should be treated with M O β the initial screening, a consultation with the inhaled 2-agonists. pulmonologist may be recommended. In 2) If a patient is at risk for complications, severe cases of asthma, laboratory studies preoperative treatment with 40-60 mg of such as arterial bloodC gases and pulmonary prednisone/day or hydrocortisone 100 mg function tests should be indicated in order every 8 h intravenously is suggested. Anyone to analyse the degree of respiratory impair- with a preoperative FEV1<80% of predicted ment.19 Preoperative spirometry is a simple should receive steroids.

Vol. 72, 2006 MINERVA ANESTESIOLOGICA 3 BURBURAN ANAESTHETIC MANAGEMENT IN ASTHMA

3) Infections should be eradicated using strated by Groeben et al.28 in a randomized antibiotics. double-blind fashion study of 10 volunteers 4) Fluid and electrolyte imbalances should with mild asthma who underwent endotra- β be corrected, given that high dose 2-ago- cheal intubation under local anaesthesia. nists can cause hypokalemia, hyperglycemia They performed lung functionA test before and hypomagnesemia. In addition to that and after intubation and observed over a 50% imbalance, there may be a decreased res- reduction in FEV1 after the procedure. β ponse to 2-agonists and predisposition to However, after prophylactic administration β C cardiac arrhythmias. of a 2-adrenergic agonist and topical lido- 5) Prophylactic cromolyn treatment to pre- caine, the reduction inI FEV1 was lower (20%). vent degranulation of mast cells and release In summary, it is preferable to avoid airway of mediators should be continued. instrumentation in asthmatic patients and 6) Chest physiotherapy improves sputum regional anaesthesia should always be con- clearance and bronchial drainage. sidered for thisD purpose, as well as to reduce 7) Other conditions such as cor pulmonale postoperative complications.17 should be treated. Pregnancy can adversely affect the course 8) The patient should stop smoking in of asthma,E increasing perioperative risk in order to reduce carboxyhemoglobin levels. these patients. For this® reason regional anaesthesia is the technique of choice for preg- nant asthmatics and parturients, especially if Choice of anaesthetic technique prostaglandinsM and its derivates are adminis- Bronchial hyper-reactivity associated with tered for abortion Tor operative delivery.20, 29 asthma is an important risk factor of periop- When regional anaesthesia is not feasible erative bronchospasm. The occurrence of and general anaesthesia is required, pro- this potentially life-threatening condition in phylactic antiobstructive treatment, volatile anaesthesia practice varies from 0.17%A to anaesthetics,H propofol, opioids, and an ade- 4.2%.20 quate choice of muscle relaxants minimize During general anaesthesia, withV or with- the anaesthetic risk.17 Other than that, the out tracheal intubation, there is a reduction of use ofG face masks and laryngeal mask air- tone in either the palatal or pharyngeal mus- ways have been reported to cause less airway cles accompanied by a lung volume reduction irritation.I Kim and Bishop30 randomized 52 and an augmentation of the layerR of liquid on nonasthmatic patients to receive an endotra- the airway wall. These factors predispose to cheal tube or a laryngeal mask airway under unstable airway conditions, airflow obstruc-Rgeneral anaesthesia; and observed that res- tion and considerably greaterE airway resis- piratory system resistance (Rrs) was lower in tance.25 patients receiving laryngeal mask airways, Airway instrumentation causes reflex bron- than in those submitted to endotracheal intu- choconstriction mediated by the parasym-Y bation. This result supports that laryngeal N 17, 26 pathetic nervous system. In addition, evi- mask might be a more reliable alternative dences suggest Ithat mechanical stimulation of than endotracheal intubation. the airways may activate the peripheralP ter- minals of C-fibres afferents. These nerve fibres PREMEDICATION release substance P and neurokinin A which can causeM an increase in vascular perme- An adequate sedation of the patient should ability, bronchial smooth Omuscle constric- be achieved in order to avoid perioperative tion, and local vasodilatation.27 The anaes- complications. For this purpose, benzodi- thesiologists’ goal should be to minimize the azepines are safe and do not alter bronchial risk of inciting bronchospasmC and to avoid tone.17 In this context, Kil et al.31 noticed that triggering stimuli. oral midazolam (0.5 mg/kg) did not change The effect of endotracheal intubation, even oxygen saturation, respiratory rate, and pulse in symptom-free asthmatics, was demon- rate in children with mild to moderate asth-

4 MINERVA ANESTESIOLOGICA Vol. 72, 2006 ANAESTHETIC MANAGEMENT IN ASTHMA BURBURAN

ma undergoing dental treatment. Hence, mechanics and lung histology in normal rats midazolam at a dose of 0.5 mg/kg is a safe anaesthetised with sevoflurane. They obser- and effective mean for sedation of patients ved that sevoflurane anaesthesia did not act with mild to moderate asthma. at airway level but at lung periphery, stiff- ening lung tissues and increasingA mechanical 37 INHALATIONAL ANAESTHETICS inhomogeneities. Besides, Takala et al. evaluated pulmonary inflammatory mediators in Inhalational agents possess bronchodilato- bronchoalveolar lavage fluid after sevoflu- ry effects, decrease airway responsiveness and rane anaesthesia in pigsC and reported that attenuate histamine-induced bronchospasm.10 sevoflurane increased pulmonary leukotriene β I The mechanism is thought to be -adrenergic - - C4, NO3 , and NO2 production suggesting an receptor stimulation leading to increased intra- inflammatory response. cellular cyclic-AMP. This has a direct bronchial muscle relaxing effect. Increased cAMP may D INTRAVENOUS ANAESTHETICS bind free calcium within bronchial myoplasm and cause relaxation by negative feedback. It Ketamine is an i.v. general anaesthetic that may impede antigen-antibody mediated is consideredE an attractive choice because of enzyme production and the release of hista- its sympathomimetic® bronchodilatory prop- mine from leukocytes as well.11 erties and its effectiveness at preventing and For all these reasons, volatile agents such reversing wheezing in patients with asthma as halothane and isoflurane have been rec- whoM require anaesthesia and intubation.38 ommended for general anaesthetic techniques Ketamine relaxes the bronchiolar musculature in patients with obstructive airway diseases and prevents the bronchoconstrictionT induced for many years and they are even helpful to by histamine, decreasing the risk of bron- treat status asthmaticus. Exception should chospasm during the induction of anaesthe- be made to desflurane that can leadA to sia. These effectsH derive from a direct action increased secretions, coughing, laryngospasm on bronchial muscle as well as a potentia- and bronchospasm.32 tion of cathecolamines. Nonetheless, keta- Hitherto, studies using sevofluraneV have mine increases bronchial secretions and it is shown controversial results. Rooke et al.33 usual Gto administer an anticholinergic agent compared the bronchodilating efficacy of suchI as atropine or glycopyrrolate in con- sevoflurane, isoflurane or halothaneR after tra- junction. Hallucinations are the most unpleas- cheal intubation in patients without asthma. ant side effect of ketamine and can be min- In their study, halothane was not significantly imized with concomitant sedation with ben- better than isofluraneE at reducing Rrs.Rzodiazepines.10 However, its effectiveness Nonetheless, sevoflurane decreased Rrs more has not been demonstrated in a controlled tri- than either halothane or isoflurane. al.39, 40 Although previous studies analysed Habre et al.34 studied lung function inY chil- the effects of ketamine on central airways, dren with and withoutN asthma receiving Alves-Neto et al.41 observed in rats without anaesthesia withI sevoflurane and concluded pre-existing airway constriction that keta- that, in children with mild to moderateP asth- mine did not act at airway level but at lung ma, endotracheal intubation during sevoflu- periphery increasing mechanical inhomo- rane anaesthesia was associated with an geneities, which may result from dilation of increaseM in Rrs that was not seen in nonasth- distal airways and alveolar collapse. matic children. In spite of Othat, no apparent Brown and Wagner 38 examined the local clinical adverse event was observed, and airway effects of ketamine and propofol on according to the Scalfaro et al. study,35 a pre- attenuating direct and reflex induced airway anaesthetic treatment Cwith inhaled salbutamol constriction. They developed a nonasthmat- administered before sevoflurane anaesthesia ic animal model in which they administered can prevent that increase of Rrs. ketamine and propofol directly to the air- Correa et al.36 analysed the respiratory ways via the bronchial artery and concluded

Vol. 72, 2006 MINERVA ANESTESIOLOGICA 5 BURBURAN ANAESTHETIC MANAGEMENT IN ASTHMA

that the major mechanism of bronchopro- caused the problem. Accordingly, propofol tection of ketamine and propofol is deter- should be used with caution in patients with mined by an inhibition of neurally-induced allergic disease or drug-induced asthma. bronchoconstriction. Propofol, a widely used short-acting i.v. OPIOIDS anaesthetic, has been associated with less A bronchoconstriction during anaesthetic induc- Although opioids could release histamine, tion than other anaesthetic agents.42 In vitro they are considered safe for patients with data suggest that propofol has a direct air- increased bronchial reactivity.C Fentanyl and way smooth muscle relaxant action.43 Pizov et its analogues are frequentlyI used in the induc- al.44 in a randomized controlled clinical trial tion of anaesthesia and they can lead to tho- evaluated the incidence of wheezing in rax rigidity that can be misinterpreted as bron- asymptomatic asthmatic and nonasthmatic chospasm. With slow injection this effect is patients receiving i.v. anaesthetic agents for hardly observed.D induction of anaesthesia. They observed that Moreover, the suppression of cough reflex both asthmatic and nonasthmatic patients and the deepening of anaesthesia level who received a thiobarbiturate for induction achievedE after opioids administration can be had a greater incidence of wheezing than did helpful in asthmatic® patients.17 patients receiving propofol. Similarly, Eames et al.45 assessed Rrs in a nonasthmatic patient MUSCLE RELAXANTS population with a high incidence of smok- MDepending on which type of muscarinic ing, and compared thiopental, etomidate, and receptor is stimulated,T increased or decreased propofol. They observed that tracheal intu- bronchial tone and reactivity can be expect- bation with propofol anaesthesia produced ed. It has been shown that muscle relaxants, lower Rrs than when thiopental or a relative- A which affect M2 receptors more than M3 ly high dose of etomidate was used. To com- receptors (gallamin,H pipecuronium, rapacuro- pare the effects of propofol anaesthesia in nium) can cause and enhance bronchocon- children with and without asthma, Habre et striction.49 Otherwise, muscle relaxants which 46 V al. induced anaesthesia with propofol, fen- seem to bind M receptors more or at least the tanyl, and atracurium and maintained with G 3 same way as M2 receptors do not induce an infusion of propofol and 50% nitrous oxide bronchospasm.I Among those, vecuronium, in oxygen. Final results showedR that respira- rocuronium, cisatracurium and pancuroni- tory mechanics was not altered by propofol um are considered safe.11 anaesthesia in both children with and without Other than these direct effects on mus- 47R asthma. In this context,E Peratoner et al. carinic receptors, atracurium, and mivacuri- analysed the effects of propofol on respiratory um dose-dependently release histamine and mechanics in normal rats and correlated these have been identified as triggers of bron- parameters with lung histology, to defineY the N choconstriction and should be used careful- sites of action of propofol. They observed ly in asthmatic patients.17 that propofol actsI at the airway level decreas- Furthermore, the reversal of muscle relaxing respiratory system and lung impedancesP ation at the end of surgery should be avoid- as a result of central airway dilation. ed since neostigmine or physostigmine cause On the basis of the aforementioned, propo- bradycardia, increased secretion and fol is consideredM safe for patientsO with asth- bronchial hyper-reactivity. For this purpose, ma, who require timely intubation. Neverthe- doses of muscle relaxants should be timed so less, Nishiyama and Hanaoka48 reported 2 as to be worn off at the end of surgery.11 cases of bronchoconstriction following propofol induction. Both patientsC had allergic prob- LOCAL ANAESTHETICS lems and it was postulated that it was the particular formulation of propofol which con- Local anaesthetics of the amide type atten- tained yolk lecithin and soybean oil that uate and even block afferent and efferent

6 MINERVA ANESTESIOLOGICA Vol. 72, 2006 ANAESTHETIC MANAGEMENT IN ASTHMA BURBURAN

nerve conduction of autonomic nerve fibres reappeared 55 min after termination of the and autonomic reflexes, such as the cough- continuous epidural injection of lidocaine. It ing or bronchoconstriction reflex, can be sup- corroborates the hypothesis that regional pressed with plasma concentrations of lido- anaesthesia in asthmatic patients, alone or caine below the toxic threshold of 5 mg/mL.50 in combination with general anaesthesia is In asthmatic volunteers i.v. lidocaine doses of advantageous. A 1-2 mg/kg of body weight significantly attenuated histamine-induced bronchoconstriction and can be used to attenuate the Treatment of intraoperativeC response to airway irritations like endotra- bronchospasmI cheal suction or intubation. Groeben et al.51 demonstrated that, in awaken humans, both If intraoperative wheezing should develop, i.v. lidocaine and inhaled albuterol signifi- nonbronchospastic causes of wheezing cantly increased the histamine threshold (mechanical obstructionD of the endotracheal when given alone. They recommended pre- tube, endobronchial intubation, pulmonary operative treatment with inhaled albuterol aspiration, pulmonary embolism, pulmonary and i.v. lidocaine to prevent reflex bron- oedema,E tension pneumothorax and nega- chospasm with tracheal intubation. tive pressure inspiration)® must be ruled out.19 Alternatively, Maslow et al.22 studied 60 asth- The first step is to deepen the level of anaes- matic patients and found that inhaled thesia via the i.v. or inhalational route or albuterol attenuated airway response to tra- both.M Administration of 100% oxygen should cheal intubation in asthmatic patients, while be instituted to prevent hypoxemia. β T i.v. lidocaine did not. 2-agonists via metered dose inhaler should Inhalation of lidocaine can attenuate the be administered through the airway. It is response to airway irritation with plasma con- important to consider the fact that delivery of centrations lower than those following sys-A aerosolized Hagents during mechanical venti- temic administration.17 Nevertheless, the lation is not adequate, being estimated that as attenuation of bronchial reactivity is preced- little as 1% to 3% of a dose of nebulized med- ed by a mild airway irritation 52, 53 Vthat can ication actually reaches the lungs of a patient β G be avoided with a pretreatment with a 2- on positive pressure ventilation. The amount adrenergic agonist or minimized by using of Iaerosol reaching the lungs could be lidocaine inhalation in a doseR of 2 mg/kg as improved by means of increase in respirato- a 4% solution for topical anaesthesia.50 This ry time, reduction of respiratory rate, increase is the regimen that attenuates bronchial in the volume of nebulizer fill, and position- hyper-reactivity with theE least airway irrita-Ring of the nebulizer between the Y-piece and tion.54 In addition, Hunt et al.55 recruited 50 catheter mount or on the inspiratory limb of subjects with mild to moderate asthma to the ventilator circuit Y-piece when jet nebu- receive either an inhaled placebo or inhaledY lizers are used.27 lidocaine 4% forN 8 weeks. Their results Epinephrine, either subcutaneously or showed that nebulizedI lidocaine was an intravenously, can help in severely bron- effective therapy in those patients.P chospastic patients. Corticosteroids can be Moreover, local anaesthetics, absorbed utilized, but its onset of action takes place from the epidural space to the blood, atten- within 4 to 8 h of administration.27 uate bronchialM hyper-reactivity to chemical Leukotriene receptor antagonists and mast stimuli. Shono et al.56 reportedO a case of a cell inhibitors have no use in acute bron- man with bronchial asthma under continuous chospasm. Intravenous aminophylline can epidural anaesthesia with 2% lidocaine in be started but side effects such as tachycar- which wheezing graduallyC diminished after dia and hypertension may limit its useful- the epidural injection and completely disap- ness. As a result, methylxantines are no longer peared over 155 min during continuous recommended for acute exacerbations. epidural injection of lidocaine. Wheezing A smooth, slow emergence minimizes the

Vol. 72, 2006 MINERVA ANESTESIOLOGICA 7 BURBURAN ANAESTHETIC MANAGEMENT IN ASTHMA

risk of bronchospasm. Deep extubation can ment in status asthmaticus and subsequent- be attempted if no airway difficulties were ly failed conventional therapy. Sevoflurane encountered during induction. If deep extu- was administered for approximately 150 min, bation is contraindicated, the patient may be stabilizing the patient’s condition enough to taken to the postanesthesia care unit intu- allow fixed-wing air transportA to a tertiary bated and opioids administered to facilitate facility. Likewise, Mazzeo et al.57 reported an tolerance to the endotracheal tube. When 8-year-old boy treated with ketamine and the patient is awake and possesses appro- sevoflurane and observed no episode of priate airway reflexes, extubation can occur. haemodynamic instabilityC despite severe pro- Intravenous lidocaine may be of use in pre- longed hypercapnia.I Oxygenation was main- venting bronchospasm with extubation.19 tained and successful recovery followed without sequelae.

Inhalation anaesthesia D in status asthmaticus Conclusions

Status asthmaticus refers to acute asthma Some Eanaesthetics have been used even attacks in which the degree of bronchial for intractable status® asthmaticus. Despite obstruction is either severe from the onset that, and the fact that the understanding of or worsens and is not relieved by usual ther- pathophysiology of asthma has changed, apy in 30 to 60 min. The term refractory sta- thereM are only a few studies describing the tus asthmaticus describes those cases in sites and mechanismsT of action of our fre- which the patient’s condition continues to quently used anaesthetic agents in a chron- deteriorate despite aggressive pharmacologic inflamed, hyper-responsive and remod- ic interventions and persists for more than elled lung, as seen in asthmatic patients. 10 A 24 h. Knowledge Hof these mechanisms will stimu- When conventional bronchodilators fail, late a great revolution in anaesthetic man- the intensivist may resort to the useV of drugs agement, allowing anaesthesiologists and such as ketamine, and inhalation anaesthesia. intensivistsG to optimise the use of anaesthet- In this context, deep sedation is important ic drugs and techniques in those patients, as not only to improve oxygenation, but to wellI as it will direct researchers to develop reduce cerebral metabolic requirementsR as new drugs effective not only on inhibiting well.57 hyper-responsiveness and producing bron- Therapy with inhalational anaestheticRchodilation, but also on reducing or even agents such as halothane,E enflurane, isoflu- suppressing the underlying inflammation rane and diethyl ether has been successfully process and remodelling. Thereupon, the used in the management of refractory status incidence of anaesthetic complications and asthmaticus.10, 58 Iwaku et al.59 treated patientsY N adverse events will be strongly minimized with status asthmaticus with isoflurane inhala- which will be very beneficial to patients and tion and observedI that tidal volume, pH, and physicians. PaCO2 improved after anaesthesiaP and that these patients stayed in the Intensive Care Unit (ICU) and underwent mechanical ventilation Mfor a shorter period than those who Riassunto were not treated with isoflurane.O Gestione anestesiologica dell’asma Que and Lusaya60 successfully used sevoflurane inhalation in a parturient in sta- La gestione anestesiologica dei pazienti asmatici è tus asthmaticus forC caesarean section. stata focalizzata a evitare la broncocostrizione e a indurre la broncodilatazione. Tuttavia, la definizione 61 Schultz reported the use of sevoflurane in a di asma è mutata negli ultimi dieci anni. L’asma è 26-year-old woman who presented to a rur- stata definita come una sindrome clinica caratterizzata al critical access hospital emergency depart- da un processo infiammatorio che si estende oltre le

8 MINERVA ANESTESIOLOGICA Vol. 72, 2006 ANAESTHETIC MANAGEMENT IN ASTHMA BURBURAN

vie aeree centrali, interessando quelle periferiche e il apeutic implications of mechanisms. Physiology parenchima polmonare. Tenendo conto di questo 2005;20:28-35. concetto e sapendo che l’asma è un disturbo comu- 7. Chen FH, Samson KT, Miura K, Ueno K, Odajima Y, Shougo T et al. Airway remodeling: a comparison ne con tassi di prevalenza e gravità in aumento in between fatal and nonfatal asthma. J Asthma tutto il mondo, è obbligatoria una scelta razionale 2004;41:631-8. dei farmaci e delle procedure anestesiologiche. Di 8. Nakagawa T, Hoshino M. Airway Aremodeling in asth- conseguenza, abbiamo cercato di fornire un aggior- ma: an introduction. Clin Rev Allergy Immunol 2004; namento circa l’approccio farmacologico e le tecniche 27:1-2. 9. Vignola AM, Mirabella F, Costanzo G, Di Giorgi R, anestesiologiche per il paziente asmatico. Quando Gjomarkaj M, Bellia V et al. Airway remodeling in asth- fattibile, si dovrebbe preferire l’anestesia regionale, dal ma. Chest 2003;123:S417-22.C momento che essa riduce l’irritazione delle vie aeree 10. Jagoda A, Shepherd SM, Spevitz A, Joseph MM. e le complicanze postoperatorie. Se non si può evi- Refractory asthma, PartI 1: Epidemiology, pathophysi- tare il ricorso all’anestesia generale, la maschera respi- ology, pharmacologic interventions. Ann Emerg Med ratoria laringea è più sicura rispetto all’intubazione 1997;29:262-74. 11. Enright A. Bronchospastic disease and emergency endotracheale. L’inalazione di loidocaina, da sola o in surgery. Middle East J Anesthesiol 2004;17:927-38. combinazione con l’albuterolo, minimizza la bron- 12. Xisto DG, FariasD LL, Ferreira HC, Picanco MR, Amitrano cocostrizione istamino-indotta. Il propofol e la keta- D, Lapa ES Jr et al. Lung parenchyma remodeling in a mina inibiscono la broncocostrizione, diminuendo il murine model of chronic allergic inflammation. Am J rischio di broncospasmo durante l’induzione dell’a- Respir Crit Care Med 2005;171:829-37. nestesia. Il propofol favorisce la dilatazione delle vie 13. Tulic MK,E Hamid Q. Contribution of the distal lung to the pathologic and physiologic changes in asthma: aeree centrali ed è più affidabile rispetto all’etomidato potential therapeutic ®target Roger S. Mitchell lecture. o al tiopentale. L’alotano, l’enflurano e l’isoflurano Chest 2003;123:S348-55. sono potenti broncodilatatori e possono essere utili 14. Tulic MK, Christodoulopoulos P, Hamid Q. Small air- anche nello status asmatico. Il sevoflurano ha evi- way inflammation in asthma. Respir Res 2001;2:333-9. denziato risultati contrastanti nei pazienti asmatici. Il 15.M Warner DO, Warner MA, Barnes RD, Offord KP, Schroeder DR, Gray DT et al. Perioperative respirato- vecuronio, il rocuronio, il cisatracurio e il pancuronio ry complications Tin patients with asthma. non inducono broncospasmo, mentre l’atracurio e il Anesthesiology 1996;85:460-7. mivacurio possono far rilasciare, con meccanismo 16. Bishop MJ, Cheney FW. Anesthesia for patients with dose-dipendente, istamina e dovrebbero essere som- asthma. Low risk but not no risk. Anesthesiology ministratti con cautela in questo tipo di pazienti. I 1996;85:455-6. A 17. Groeben H.H Strategies in the patient with compromised dati futuri sui siti d’azione polmonari dei farmaci ane- respiratory function. Best Pract Res Clin Anaesthesiol stetici, assieme alla nostra comprensione della fisio- 2004;18:579-94. patologia dell’asma, ci renderanno in grado di sce- 18. Celiker V, Basgul E. Anaesthesia in aspirin-induced gliere il miglior approccio anestesiologicoV per la asthma. Allergol Immunopathol 2003;31:338-41. popolazione con asma. 19. StasicG AF. Perioperative implications of common respiratory problems. Semin Pediatr Surg 2004;13:174-80. Parole chiave: Asma bronchiale - Anestesia - Anesteti- 20.I Bremerich DH. Anesthesia in bronchial asthma. ci - Spasmo bronchiale - IpereattivitàR bronchiale. Anasthesiol Intensivmed Notfallmed Schmerzther 2000;35:545-58. 21. Silvanus MT, Groeben H, Peters J. Corticosteroids and inhaled salbutamol in patients with reversible airway References R obstruction markedly decrease the incidence of bron- E chospasm after tracheal intubation. Anesthesiology 1. Beasley R, Crane J, Lai CK, Pearce N. Prevalence and 2004;100:1052-7. etiology of asthma. J Allergy Clin Immunol 2000;105: 22. Maslow AD, Regan MM, Israel E, Darvish A, Mehrez M, S466-72. Boughton R et al. Inhaled albuterol, but not intra- Y venous lidocaine, protects against intubation-induced 2. Sunyer J, Anto JM, Tobias A, Burney P. Generational N bronchoconstriction in asthma. Anesthesiology increase of self-reported first attack of asthma in fifteen industrialized countries. European Community Respira- 2000;93:1198-204. tory Health StudyI (ECRHS). Eur Respir J 1999;14:885- 23. Barnes PJ. Mechanisms of action of glucocorticoids in 91. P asthma. Am J Respir Crit Care Med 1996;154:S21-6; dis- 3. CDC. From the Centers for Disease Control and cussion S6-7. Prevention. Self-reported asthma prevalence and con- 24. Kabalin CS, Yarnold PR, Grammer LC. Low complica- trol among adults-United States, 2001. JAMA 2003;289: tion rate of corticosteroid-treated asthmatics undergo- 2639-40. ing surgical procedures. Arch Intern Med 1995;155: 4. Arif AA,M Borders TF, PattersonO PJ, Rohrer JE, Xu KT. 1379-84. Prevalence and correlates of paediatric asthma and 25. Burwell DR, Jones JG. The airways and anaesthesia I. wheezing in a largely rural USA population. J Paediatr Anatomy, physiology and fluid mechanics. Anaesthesia Child Health 2004;40:189-94. 1996;51:849-57. 5. McFadden Jr ER. Asthma. In: Kasper DL, Braunwald EL, 26. Barnes PJ. What is the role of nerves in chronic asth- Fauci AS, Hauser SL, LongoC DL, Jameson JL et al. edi- ma and symptoms? Am J Respir Crit Care Med tors. Harrison’s Principles of Internal Medicine. 16th 1996;153:S5-8. ed. New York: McGraw-Hill Companies, Inc.; 2005. p. 27. Jagoda A, Shepherd SM, Spevitz A, Joseph MM. 1510-1. Refractory asthma, Part 2: Airway interventions and 6. Homer RJ, Elias JA. Airway remodeling in asthma: ther- management. Ann Emerg Med 1997;29:275-81.

Vol. 72, 2006 MINERVA ANESTESIOLOGICA 9 BURBURAN ANAESTHETIC MANAGEMENT IN ASTHMA

28. Groeben H, Schlicht M, Stieglitz S, Pavlakovic G, Peters 45. Eames WO, Rooke GA, Wu RS, Bishop MJ. Comparison J. Both local anesthetics and salbutamol pretreatment of the effects of etomidate, propofol, and thiopental on affect reflex bronchoconstriction in volunteers with respiratory resistance after tracheal intubation. asthma undergoing awake fiberoptic intubation. Anesthesiology 1996;84:1307-11. Anesthesiology 2002;97:1445-50. 46. Habre W, Matsumoto I, Sly PD. Propofol or halothane 29. Kuczkowski KM. Labor analgesia for the parturient anaesthesia for children with asthma: effects on respi- with respiratory disease: what does an obstetrician ratory mechanics. Br J Anaesth 1996;77:739-43.A need to know? Arch Gynecol Obstet 2005;272:160-6. 47. Peratoner A, Nascimento CS, Santana MC, Cadete RA, 30. Kim ES, Bishop MJ. Endotracheal intubation, but not Negri EM, Gullo A et al. Effects of propofol on respi- laryngeal mask airway insertion, produces reversible ratory mechanic and lung histology in normal rats. Br bronchoconstriction. Anesthesiology 1999;90:391-4. J Anaesth 2004;92:737-40. 31. Kil N, Zhu JF, VanWagnen C, Abdulhamid I. The effects 48. Nishiyama T, Hanaoka K.C Propofol-induced bron- of midazolam on pediatric patients with asthma. Pediatr choconstriction: two case reports. Anesth Analg 2001;93: Dent 2003;25:137-42. 645-6. I 32. Goff MJ, Arain SR, Ficke DJ, Uhrich TD, Ebert TJ. 49. Jooste E, Klafter F, Hirshman CA, Emala CW. A mech- Absence of bronchodilation during desflurane anes- anism for rapacuronium-induced bronchospasm: M2 thesia: a comparison to sevoflurane and thiopental. muscarinic receptor antagonism. Anesthesiology Anesthesiology 2000;93:404-8. 2003;98:906-11. 33. Rooke GA, Choi JH, Bishop MJ. The effect of isoflurane, 50. Groeben H, SchwalenD A, Irsfeld S, Stieglitz S, Lipfert P, halothane, sevoflurane, and thiopental/nitrous oxide on Hopf HB. Intravenous lidocaine and bupivacaine dose- respiratory system resistance after tracheal intubation. dependently attenuate bronchial hyperreactivity in Anesthesiology 1997;86:1294-9. awake volunteers. Anesthesiology 1996;84:533-9. 34. Habre W, Scalfaro P, Sims C, Tiller K, Sly PD. Respiratory 51. GroebenE H, Silvanus MT, Beste M, Peters J. Combined mechanics during sevoflurane anesthesia in children intravenous lidocaine and inhaled salbutamol protect with and without asthma. Anesth Analg 1999;89:1177-81. against bronchial hyperreactivity® more effectively than 35. Scalfaro P, Sly PD, Sims C, Habre W. Salbutamol pre- lidocaine or salbutamol alone. Anesthesiology 1998;89: vents the increase of respiratory resistance caused by 862-8. tracheal intubation during sevoflurane anesthesia in 52. Bulut Y, Hirshman CA, Brown RH. Prevention of lido- asthmatic children. Anesth Analg 2001;93:898-902. Mcaine aerosol-induced bronchoconstriction with intra- 36. Correa FC, Ciminelli PB, Falcao H, Alcantara BJ, venous lidocaine. Anesthesiology 1996;85:853-9. Contador RS, Medeiros AS et al. Respiratory mechan- 53. Groeben H, GrosswendtT T, Silvanus MT, Pavlakovic ics and lung histology in normal rats anesthetized with G, Peters J. Airway anesthesia alone does not explain sevoflurane. J Appl Physiol 2001;91:803-10. attenuation of histamine-induced bronchospasm by 37. Takala RS, Soukka HR, Salo MS, Kirvela OA, Kaapa local anesthetics: a comparison of lidocaine, ropiva- PO, Rajamaki AA et al. Pulmonary inflammatory medi- caine, and dyclonine. Anesthesiology 2001;94:423-8. ators after sevoflurane and thiopentone anaesthesiaA in 54. Groeben H,H Grosswendt T, Silvanus M, Beste M, Peters pigs. Acta Anaesthesiol Scand 2004;48:40-5. J. Lidocaine inhalation for local anaesthesia and atten- 38. Brown RH, Wagner EM. Mechanisms of bronchopro- uation of bronchial hyper-reactivity with least airway tection by anesthetic induction agents: propofol versus irritation. Effect of three different dose regimens. Eur ketamine. Anesthesiology 1999;90:822-8. V J Anaesthesiol 2000;17:672-9. 39. Howton JC, Rose J, Duffy S, Zoltanski T, Levitt MA. 55. HuntG LW, Frigas E, Butterfield JH, Kita H, Blomgren J, Randomized, double-blind, placebo-controlled trial of Dunnette SL et al. Treatment of asthma with nebulized intravenous ketamine in acute asthma. Ann Emerg Med Ilidocaine: a randomized, placebo-controlled study. J 1996;27:170-5. R Allergy Clin Immunol 2004;113:853-9. 40. Youssef-Ahmed MZ, Silver P, Nimkoff L, Sagy M. 56. Shono S, Higa K, Harasawa I, Sugano H, Dan K. Continuous infusion of ketamine in mechanically ven- Disappearance of wheezing during epidural lidocaine tilated children with refractory bronchospasm. Intensive anesthesia in a patient with bronchial asthma. Reg Care Med 1996;22:972-6. E R Anesth Pain Med 1999;24:463-6. 41. Alves-Neto O, Tavares P, Rocco PR, Zin WA. Respiratory 57. Mazzeo AT, Spada A, Pratico C, Lucanto T, Santamaria mechanics and morphometric changes during anes- LB. Hypercapnia: what is the limit in paediatric patients? thesia with ketamine in normal rats. Braz J Med Biol Res A case of near-fatal asthma successfully treated by mul- 2001;34:1217-23. tipharmacological approach. Paediatr Anaesth 2004;14: 42. Wu RS, Wu KC, Sum DC, Bishop MJ. ComparativeY 596-603. effects of thiopentoneN and propofol on respiratory 58. Miyagi T, Gushima Y, Matsumoto T, Okamoto K, Miike resistance after tracheal intubation. Br J Anaesth T. Prolonged isoflurane anesthesia in a case of cata- 1996;77: 735-8.I strophic asthma. Acta Paediatr Jpn 1997;39:375-8. 43. Ouedraogo N, Roux E, Forestier F, RossettiP M, Savineau 59. Iwaku F, Otsuka H, Kuraishi H, Suzuki H. [The inves- JP, Marthan R. Effects of intravenous anesthetics on tigation of isoflurane therapy for status asthmaticus normal and passively sensitized human isolated air- patients]. Arerugi 2005;54:18-23. Japanese. way smooth muscle. Anesthesiology 1998;88:317-26. 60. Que JC, Lusaya VO. Sevoflurane induction for emer- 44. Pizov R, Brown RH, Weiss YS, Baranov D, Hennes H, gency cesarean section in a parturient in status asth- BakerM S et al. Wheezing duringO induction of general maticus. Anesthesiology 1999;90:1475-6. anesthesia in patients with and without asthma. A ran- 61. Schultz TE. Sevoflurane administration in status asth- domized, blinded trial. CAnesthesiology 1995;82:1111-6. maticus: a case report. AANA J 2005;73:35-6.

10 MINERVA ANESTESIOLOGICA Vol. 72, 2006

MEDICAL PROGRESS

Review Article

Medical Progress ple in the United States have COPD. In the third U.S. National Health and Nutrition Examination Sur- vey, airflow obstruction was found in approximately CHRONIC OBSTRUCTIVE PULMONARY 14 percent of white male smokers, as compared with DISEASE approximately 3 percent of white male nonsmokers; the figures for white female smokers and for black PETER J. BARNES, D.SC. smokers were slightly lower than those for white male smokers.3 COPD is now the fourth leading cause of HRONIC obstructive pulmonary disease death in the United States, and it is the only com- (COPD) is characterized by the progressive mon cause of death that is increasing in incidence. development of airflow limitation that is not The importance of COPD as a cause of death is like- C 1 ly to be underestimated, since COPD probably con- fully reversible. The term COPD encompasses chronic obstructive bronchitis, with obstruction of small tributes to other common causes of death. airways, and emphysema, with enlargement of air There has been an increase in the prevalence of and spaces and destruction of lung parenchyma, loss of mortality from COPD, even in industrialized coun- 1 lung elasticity, and closure of small airways. Chronic tries. The World Health Organization predicts that bronchitis, by contrast, is defined by the presence of by 2020 COPD will rise from its current ranking as a productive cough of more than three months’ du- the 12th most prevalent disease worldwide to the 5th ration for more than two successive years. The cough and from the 6th most common cause of death to the 4 is due to hypersecretion of mucus and is not neces- 3rd. Reasons for the dramatic increase in COPD in- sarily accompanied by airflow limitation. However, clude reduced mortality from other causes, such as there is some epidemiologic evidence that mucus cardiovascular diseases in industrialized countries and hypersecretion is accompanied by airflow obstruc- infectious diseases in developing countries, along with tion, perhaps as a result of obstruction of peripheral a marked increase in cigarette smoking and environ- airways.2 Most patients with COPD have all three mental pollution in developing countries. pathologic conditions (chronic obstructive bronchi- MOLECULAR GENETICS tis, emphysema, and mucus plugging), but the relative extent of emphysema and obstructive bronchitis Longitudinal monitoring of lung function reveals within individual patients can vary (Fig. 1 and 2). that substantial airflow obstruction due to an accel- Although there has been an explosion of research on erated decline in lung function (two to five times the asthma and a revolution in asthma therapy, COPD normal yearly decline of 15 to 30 ml in forced expiratory volume in one second [FEV ]) occurs in only a has been surprisingly neglected, with little research 1 into cellular mechanisms and few advances in thera- minority of cigarette smokers (approximately 15 per- py. However, it is increasingly apparent that this is an cent of whites and 5 percent of Asians). This strong- important disease whose incidence is rising worldwide ly suggests that genetic factors may determine in which and that there is a need to develop new treatments smokers airflow limitation will develop. Further evi- to prevent the progression of the disease, which re- dence that genetic factors are important comes from the familial clustering of patients with early-onset sults in a large consumption of health care resources. 5 This recognition has led to a revival of research on COPD and from the differences in the prevalence mechanisms of COPD, involving new molecular ap- of COPD among different racial groups. In patients with a -antitrypsin deficiency, as shown by a protein- proaches and a search for treatments that will halt 1 ase inhibitor phenotype (PiZZ) with a -antitrypsin progression of the disease. 1 levels below 10 percent of normal values, early em- EPIDEMIOLOGY physema develops that is exacerbated by smoking, Precise figures on prevalence are surprisingly scanty, indicating a clear genetic predisposition to COPD. but it is estimated that approximately 14 million peo- However, less than 1 percent of patients with COPD have a1-antitrypsin deficiency, and many other genetic variants of a1-antitrypsin that are associated with lower-than-normal serum levels of this proteinase in- From the Department of Thoracic Medicine, National Heart and Lung Institute, Imperial College School of Medicine, London. Address reprint hibitor have not been clearly associated with an in- requests to Professor Barnes at the Department of Thoracic Medicine, Na- creased risk of COPD.6 This uncertainty has led to tional Heart and Lung Institute, Dovehouse St., London SW3 6LY, Unit- ed Kingdom, or at [email protected]. a search for associations between COPD and poly- ©2000, Massachusetts Medical Society. morphisms of other genes that may be involved in

Volume 343 Number 4 · 269

The New England Journal of Medicine

Normal Chronic Obstructive Pulmonary Disease

Disrupted alveolar attachments Mucus hypersecretion (emphysema) (luminal obstruction)

Airway held open by alveolar Mucosal and peribronchial attachments inflammation and fibrosis (obliterative bronchiolitis)

Figure 1. Mechanisms of Airflow Limitation in Chronic Obstructive Pulmonary Disease. In the peripheral airways of patients with chronic obstructive pulmonary disease, as compared with normal peripheral airways, there is airflow limitation due to a variable mixture of loss of alveolar attachments, inflammatory obstruction of the airway, and luminal obstruction with mucus.

its pathophysiology.7, 8 So far, few associations have ticulates associated with cooking in confined spaces, been detected, and even those reported have not been are important causes.12 It is likely that there are im- confirmed in other studies. portant interactions between environmental factors The reported risk of COPD is 10 times the nor- and a genetic predisposition to COPD. Air pollution mal level in a Taiwanese population with a polymor- (particularly with sulfur dioxide and particulates), phism in the promoter region of the gene for tumor exposure to certain occupational chemicals (such as necrosis factor a (TNF-a) that is associated with in- cadmium), and passive smoking may all be risk fac- creased TNF-a production.9 However, members of tors. The role of airway hyperresponsiveness and al- a British population with the same polymorphism do lergy as risk factors for COPD is still uncertain. Atopy, not have an increased risk of COPD.10 A polymor- serum IgE concentrations, and blood eosinophilia phic variant of microsomal epoxide hydrolase, an en- are not important risk factors,13 but the Lung Health zyme involved in the metabolism of epoxides that may Study showed that increased airway responsiveness be generated in tobacco smoke, has been associated to inhaled methacholine was a predictor of an accel- with a quintupling of the risk of COPD.11 Matched erated decline in lung function over a period of five cigarette smokers with and without COPD are being years.14 However, this is not necessarily the same compared by techniques such as DNA microarray type of abnormal airway responsiveness that is seen (gene chips) to detect single-nucleotide polymor- in asthma. Low birth weight is also a risk factor for phisms, two-dimensional gel electrophoresis to detect COPD, probably because poor nutrition in fetal life novel proteins (proteomics), and differential display results in small lungs, so that the normal decline in to assess which known and novel genes are expressed. lung function with age starts from a lower peak value.15 These techniques may not only identify markers of risk, but also reveal novel molecular targets for fu- INFLAMMATION ture treatments. The recognition that chronic airway inflammation has a critical role in producing the symptoms of asth- RISK FACTORS ma led to the earlier and more widespread use of an- In industrialized countries, cigarette smoking ac- tiinflammatory treatments, particularly inhaled corti- counts for most cases of COPD, but in developing costeroids, which have become the mainstay of asthma countries other environmental pollutants, such as par- management. It is now apparent that there is a chron-

270 · July 27, 2000

MEDICAL PROGRESS ic inflammatory process in COPD, but it differs mark- the greater excretion of desmosine, derived from elas- edly from that seen in asthma, with different inflam- tin cross-links, in smokers with rapid decline in lung matory cells, mediators, inflammatory effects, and function than in those with a normal decline.29 Neu- 16,17 responses to treatment. trophil elastase is inhibited by a1-antitrypsin in the lung parenchyma and almost certainly accounts for Inflammatory Cells and Mediators the emphysema in a1-antitrypsin deficiency, but its Histopathological studies show that most inflam- role in smoking-related emphysema is less certain. The mation in COPD occurs in the peripheral airways concentrations of neutrophil elastase in complex with

(bronchioles) and lung parenchyma. The bronchioles a1-antitrypsin are elevated in patients with emphyse- are obstructed by fibrosis and infiltration with mac- ma.30 In addition, serine proteases are potent stimu- rophages and T lymphocytes. There is destruction of lants of mucus secretion and may have an important lung parenchyma and an increased number of mac- role in the mucus hypersecretion seen in chronic rophages and T lymphocytes, which are predomi- bronchitis.31,32 nantly CD8+ (cytotoxic) T cells.18,19 Bronchial-biopsy However, there is now increasing evidence that results are similar to the histopathological findings: matrix metalloproteinases derived from macrophag- patients with severe COPD have infiltration of mac- es and neutrophils have a role.26 In patients with em- rophages and CD8+ T cells and an increased number physema, there is an increase in concentrations in of neutrophils.20,21 There is a marked increase in mac- bronchoalveolar-lavage fluid and expression by macrophages and neutrophils in bronchoalveolar-lavage rophages of matrix metalloproteinase-1 (collagenase) fluid and induced sputum.22,23 In contrast to the sit- and matrix metalloproteinase-9 (gelatinase B).33-35 uation with asthma, eosinophils are not prominent There is an increase in the activity of matrix metal- except during exacerbations or in patients with con- loproteinase-9 and matrix metalloproteinase-2 in the comitant asthma.20,24 lung parenchyma of patients with emphysema.36 The The inflammatory mediators involved in COPD interest in matrix metalloproteinases has been further are less well defined than those in asthma. The con- heightened by the demonstration in mice that em- centration of leukotriene B4, which is chemotactic for physema induced by chronic cigarette exposure is neutrophils, is increased in the sputum of patients with prevented by deletion of the gene encoding matrix COPD.25 Concentrations of the cytokines TNF-a metalloproteinase-12 (macrophage metalloelastase).37 and the neutrophil-chemotactic chemokine interleu- Although there are doubts about the importance of kin-8 are also increased in the sputum of patients matrix metalloproteinase-12 in human macrophages, with COPD.23 There is probably a complex interac- this result demonstrates the capacity of matrix me- tion between cells and mediators in COPD, resulting talloproteinases to induce emphysema. Matrix metal- in progressive obstructive changes in small airways loproteinases may generate chemotactic peptides that and destruction of lung parenchyma. Macrophages ap- promote recruitment of macrophages to the paren- pear to play a critical part, since these cells are 5 to chyma and airways. 10 times more numerous, are activated, are localized Normally, all of these proteolytic enzymes are coun- to sites of damage, and also have the capacity to pro- teracted by antiproteases. The major inhibitors of 26 duce all of the pathologic changes of COPD. Mac- serine proteases are a1-antitrypsin in lung parenchyma rophages may be activated by cigarette smoke and and airway-epithelium–derived secretory leukopro- other irritants to release neutrophil-chemotactic fac- tease inhibitor in the airways. At least three tissue in- tors, such as leukotriene B4 and interleukin-8. Neu- hibitors of matrix metalloproteinases (called TIMP-1, trophils and macrophages release multiple proteinases TIMP-2, and TIMP-3) counteract matrix metallopro- that break down connective tissue in the lung paren- teinases. Cigarette smoking may induce inflammation chyma, resulting in emphysema, and stimulate mucus and increased release of proteases that are counter- secretion (Fig. 3). The role of cytotoxic T cells is not acted by antiproteases in amounts sufficient to prevent yet clear, but they may be involved in the apoptosis parenchymal injury, but in smokers in whom COPD and destruction of alveolar-wall epithelial cells through develops, the production of antiproteases may be in- the release of perforins and TNF-a.27 adequate to neutralize the effects of multiple proteases, perhaps because of genetic polymorphisms that Protease–Antiprotease Imbalance impair the function or production of these proteins. It has long been proposed that various proteases break down connective-tissue components, particular- Oxidative Stress ly elastin, in lung parenchyma to produce emphysema There is accumulating evidence that oxidative stress (Fig. 4). Most attention has focused on neutrophil may have an important role in COPD.38 There is an elastase and proteinase 3, which are neutrophil-derived increase in the concentration of hydrogen peroxide serine proteases, and on cathepsins, all of which can in the exhaled breath condensates of patients with produce emphysema in laboratory animals.28 Evidence COPD, particularly during exacerbations,39 and in- of degradation of elastin in COPD is provided by creased breath and urinary concentrations of 8-iso-

Volume 343 Number 4 · 271

The New England Journal of Medicine

A B

520 µm 280 µm

Figure 2. Histopathological Features of Chronic Obstructive Pulmonary Disease. Panel A shows a transverse section of a small airway of normal appearance, with a patent lumen and a relatively thin airway wall with several surrounding alveolar attachments. The elastic recoil of the alveolar attachments helps to maintain the patency of the airway lumen, particularly during expiration. Panel B shows chronic obstructive bronchiolitis, with thickening of the airway wall and infiltration with lymphocytes, macrophages, and neutrophils. Panel C (facing page) shows emphysema, with peribronchiolar destruction of alveolar walls, resulting in loss of alveolar attachments, airway collapse, and enlargement of air spaces distal to the terminal bronchioles. Each section has been stained with hematoxylin and eosin to show the cell nuclei and cytoplasm, respectively. (Courtesy of Professor Peter Jeffery, National Heart and Lung Institute, London.)

prostane, a marker of oxidative stress.40,41 Oxidative species and expression of adhesion molecules in cir- stress may exacerbate COPD through several mech- culating neutrophils.43,44 Circulating concentrations anisms, including the activation of the transcription of interleukin-6 and of acute-phase proteins, such as factor nuclear factor-kB (NF-kB), which switches on C-reactive protein, are also increased even in the sta- the genes for TNF-a, interleukin-8, and other inflam- ble state, although they are further increased during matory proteins,42 and oxidative damage of antipro- exacerbations.44-46 Some patients, particularly those teases, such as a1-antitrypsin and secretory leuko- with predominant emphysema, have profound weight protease inhibitor, thus enhancing inflammation and loss; this factor is a predictor of increased mortality proteolytic injury (Fig. 5). that is independent of poor lung function.47 Weight loss in COPD, as in other chronic inflammatory diseas- Systemic Effects es, has been associated with increased circulating lev- There is increasing evidence of systemic (nonpul- els of TNF-a and soluble TNF receptors and with in- monary) effects in patients with COPD, which may creased release of TNF-a from circulating cells.45,46,48,49 have an important negative influence on the quality These increases have been associated with increased of life. There is evidence of systemic oxidative stress circulating levels of leptin, which may contribute to in COPD, with increased release of reactive oxygen weight loss in these patients.49 The weight loss in

272 · July 27, 2000

MEDICAL PROGRESS

C Cigarette smoke and other irritants

MCP-1

Alveolar macrophage Epithelial cells

Neutrophil chemotactic factors Interleukin-8, leukotriene B4

CD8+ lymphocytes Neutrophil

520 µm

Proteases Protease Neutrophil elastase COPD is due to increased metabolism and is largely inhibitors Cathepsins explained by a loss of skeletal muscle and wasting of Matrix metalloproteinases limb muscles.50 Skeletal-muscle weakness is a common feature of COPD and exacerbates dyspnea. The weakness is due to a combination of chronic hypoxia, immobility, and increased metabolic rate. There is a Alveolar-wall Mucus profound decrease in myosin heavy chain in these destruction hypersecretion muscles.51 These systemic effects of COPD are an (emphysema) (chronic bronchitis) important target for skeletal- and respiratory-muscle training as part of pulmonary rehabilitation. Improved nutrition and short courses of anabolic steroids have Figure 3. Inflammatory Mechanisms in Chronic Obstructive been shown to improve lung function in patients Pulmonary Disease. 52 Cigarette smoke and other irritants activate macrophages and with COPD. airway epithelial cells in the respiratory tract, which release neutrophil chemotactic factors, including interleukin-8 and leu- Amplifying Mechanisms kotriene B4. Neutrophils and macrophages then release proteas- The inflammatory changes and protease imbalance es that break down connective tissue in the lung parenchyma, that occur in COPD are also seen in cigarette smok- resulting in emphysema, and also stimulate mucus hypersecretion. Proteases are normally counteracted by protease inhibitors, ers without COPD, but to a lesser extent.23,34 This including a1-antitrypsin, secretory leukoprotease inhibitor, and observation suggests that the accelerated decline in tissue inhibitors of matrix metalloproteinases. Cytotoxic T cells lung function may be due to amplification of the nor- (CD8+ lymphocytes) may also be involved in the inflammatory mal pulmonary response to irritants, either because cascade. MCP-1 denotes monocyte chemotactic protein 1, which of increased production of inflammatory proteins and is released by and affects macrophages.

Volume 343 Number 4 · 273

The New England Journal of Medicine

ACUTE EXACERBATIONS Neutrophil elastase Although acute exacerbations of COPD, defined Proteinase 3 by increased symptoms and worsening lung function, Cathepsins Matrix metalloproteinases (1, 2, 9, 12) are a common cause of hospital admission, their mech- Others anism is far from clear. Acute exacerbations may be prolonged and may have a profound effect on the quality of life.57 It was always assumed that the in- Decrease creased amount and increased purulence of sputum were due to bacterial infection of the respiratory tract. It is now evident that many exacerbations in COPD, as in asthma, are due to upper respiratory tract viral infections (such as rhinovirus infection) and to envi- Increase ronmental factors, such as air pollution and temperature.58,59 There is an increase in neutrophils and in the concentrations of interleukin-6 and interleukin-8 a1-Antitrypsin in sputum during an exacerbation, and patients who Secretory leukoprotease inhibitor have frequent exacerbations have higher levels of in- Elafin 60 Tissue inhibitors of matrix terleukin-6, even when COPD is stable. Bronchial metalloproteinases biopsies show an increase in eosinophils during exacerbations in patients with mild COPD,24 but there is no increase in sputum eosinophils during exacer- Figure 4. Protease–Antiprotease Imbalance in Chronic Obstruc- bations in patients with severe COPD.60 An increase tive Pulmonary Disease. in markers of oxidative stress and exhaled nitric ox- In chronic obstructive pulmonary disease the balance appears to be tipped in favor of increased proteolysis, because of either an ide, presumably reflecting increased airway inflamma- increase in proteases, including neutrophil elastase, cathepsins, tion, is observed during exacerbations.39,61,62 and matrix metalloproteinases, or a deficiency of antiproteases, which may include a1-antitrypsin, elafin, secretory leukoprotease ADVANCES IN DRUG THERAPY inhibitor, and tissue inhibitors of matrix metalloproteinases. Antismoking Measures Smoking cessation is the only measure that will slow the progression of COPD, as confirmed in the large Lung Health Study.63 Nicotine-replacement ther- enzymes or because of defective endogenous antiin- apy (by gum, transdermal patch, or inhaler) provides flammatory or antiprotease mechanisms. These dif- help to patients in quitting smoking,64 but the use ferences might be explained by polymorphisms in the of the recently introduced drug bupropion, a nor- genes encoding cytokines, proteases, antiinflamma- adrenergic antidepressant, has proved to be the most tory proteins, and antiproteases. effective strategy to date. A recent controlled trial Another hypothesis is that these differences are due showed that after a 9-week course of bupropion, ab- to latent viral infection. The latent adenovirus se- stinence rates were 30 percent at 12 months, as com- quence E1A is more commonly detected in the lungs pared with only 15 percent with placebo. The absti- of patients with COPD than in matched control nence rate was slightly improved with the addition smokers.53 Transfection of E1A into a human epi- of a nicotine patch.65 thelial cell line results in increased activation of the transcription factor NF-kB, with consequent increased New Bronchodilators release of interleukin-8 in response to endotoxin and Bronchodilators are the mainstay of current drug increased expression of intercellular adhesion mole- therapy for COPD. Bronchodilators cause only a small cule 1, providing a molecular mechanism for the am- (<10 percent) increase in FEV1 in patients with plification during the inflammatory response.54 COPD, but these drugs may improve symptoms by Although smoking is the principal cause of COPD, reducing hyperinflation and thus dyspnea, and they quitting smoking does not appear to result in reso- may improve exercise tolerance, despite the fact that lution of the inflammatory response in the airways.55,56 there is little improvement in spirometric measure- This suggests that there are perpetuating mechanisms ments.66 Several studies have demonstrated the use- that maintain the chronic inflammatory process once fulness of the long-acting inhaled b2-agonists sal- it has become established. Such mechanisms may ac- meterol and formoterol in COPD.67-69 An additional count for the presentation of COPD in patients who benefit of long-acting b2-agonists in COPD may be have stopped smoking many years before their first a reduction in infective exacerbations, since these symptoms develop. The mechanisms of disease per- drugs reduce the adhesion of bacteria such as Hae- sistence are currently unknown. mophilus influenzae to airway epithelial cells.70

274 · July 27, 2000

Cigarette smoke Inflammatory cells (neutrophils, macrophages) Activation of nuclear factor-B

Decrease in antiproteases Tumor necrosis Interleukin-8 a1-Antitrypsin and secretory factor a leukoprotease inhibitor

– O2 , H2O2 OH˙, ONOO– Neutrophil recruitment

Increased mucus Bronchoconstriction secretion

Plasma leak Isoprostanes

Figure 5. Oxidative Stress in Chronic Obstructive Pulmonary Disease. Reactive oxygen species from cigarette smoke or from inflammatory cells have several damaging effects, including decreased antiprotease defenses; activation of nuclear factor-kB, resulting in increased secretion of the cytokines interleukin-8 and tumor necro- ¡ sis factor a; increased production of isoprostanes; and direct effects on airway functions. O2 denotes superoxide anion, H2O2 hydrogen peroxide, OH hydroxyl radical, and ONOO¡ peroxynitrate.

COPD appears to be more effectively treated by terms of clinical outcome and lung function.73 Al- anticholinergic drugs than by b2-agonists, in sharp though antibiotics are still widely used for exacerba- contrast to asthma, for which b2-agonists are more tions of COPD, methods to diagnose bacterial infec- effective.71 A new anticholinergic drug, tiotropium tion reliably in the respiratory tract are needed so that bromide, which is not yet available for prescription, antibiotics are not used inappropriately. There is no has a prolonged duration of action and is suitable for evidence that prophylactic antibiotics prevent acute once-daily inhalation in COPD.72 exacerbations.

Antibiotics Oxygen Acute exacerbations of COPD are commonly as- Home oxygen therapy accounts for a large pro- sumed to be due to bacterial infection, since they portion of the costs of treating COPD (over 30 per- may be associated with increased volume and puru- cent in the United States, where expensive liquid ox- lence of the sputum. However, as noted above, it is ygen is widely used). Long-term oxygen therapy was increasingly recognized that exacerbations may be justified by two large trials that showed reduced mor- due to viral infections of the upper respiratory tract tality and improvement in quality of life in patients or may be noninfective,58 so that antibiotic treatment with severe COPD and chronic hypoxemia (partial is not always warranted. A meta-analysis of con- pressure of arterial oxygen, <55 mm Hg).74 More re- trolled trials of antibiotics in COPD showed a statis- cent studies have demonstrated that oxygen does not tically significant but small benefit of antibiotics in increase survival in patients with less severe hypox-

Volume 343 Number 4 · 275

Downloaded from www.nejm.org at UNIV NATAL MED SCH on February 16, 2007 . Copyright © 2000 Massachusetts Medical Society. All rights reserved. The New England Journal of Medicine emia, so that the selection of patients is important in reduce hospital admissions in patients with severe prescribing this expensive therapy.75 Similarly, in pa- COPD and hypercapnia and may improve long-term tients with COPD who have nocturnal hypoxemia, survival, although large, controlled trials are now nocturnal treatment with oxygen does not appear to needed.88 In one small, controlled trial, noninvasive increase survival or delay the prescription of contin- positive-pressure ventilation was not well tolerated uous oxygen therapy.76 and produced only marginal benefits.89 The combination of noninvasive positive-pressure ventilation and CORTICOSTEROIDS long-term oxygen therapy may be more effective,90 Inhaled corticosteroids are now the mainstay of but larger trials are needed before this combined therapy for chronic asthma, and the recognition that therapy can be routinely recommended. chronic inflammation is also present in COPD pro- Pulmonary Rehabilitation vided a rationale for their use in COPD. Indeed, inhaled corticosteroids are now widely prescribed for Pulmonary rehabilitation consisting of a structured COPD, and in North America they are used as fre- program of education, exercise, and physiotherapy quently in patients with COPD as in those with has been shown in controlled trials to improve exer- asthma. However, the inflammation in COPD is not cise capacity and quality of life among patients with suppressed by inhaled or oral corticosteroids, even at severe COPD91 and to reduce the amount of health 92 high doses.34,77 This lack of effect may be due to the care needed. fact that corticosteroids prolong the survival of neu- Lung-Volume–Reduction Surgery trophils78 and do not suppress neutrophilic inflammation in COPD.79 Approximately 10 percent of pa- There has been considerable interest in the surgi- tients with stable COPD have some symptomatic and cal removal of the most emphysematous parts of the objective improvement with oral corticosteroids. It lung to improve ventilatory function.93,94 The reduc- is likely that these patients have concomitant asthma, tion in hyperinflation improves the mechanical effi- since both diseases are very common. Indeed, airway ciency of the inspiratory muscles. Careful selection hyperresponsiveness, a characteristic of asthma, may of patients after a period of pulmonary rehabilitation is essential. Patients with localized upper-lobe em- predict an accelerated decline in FEV1 in patients with COPD.14 Recent studies found no evidence physema appear to do best; relatively low lung resist- that long-term treatment with high doses of inhaled ance during inspiration appears to be a good predictor 95 corticosteroids reduced the progression of COPD, of improved FEV1 after surgery. Functional improve- even when treatment was started before the disease ments include increased FEV1, reduced total lung ca- became symptomatic.80-82 Inhaled corticosteroids may pacity and functional residual capacity, improved func- slightly reduce the severity of acute exacerbations,83 tion of respiratory muscles, improved exercise capacity, but it is unlikely that their use can be justified in and improved quality of life.96,97 The benefits persist view of the risk of systemic side effects in these sus- for at least a year in most patients, but careful ex- ceptible patients and the expense of using high-dose tended follow-up is needed to evaluate the long-term inhaled corticosteroids for several years. benefits of this therapy. A large, multicenter study, By contrast, two recent studies have demonstrated the National Emphysema Treatment Trial, is now un- a beneficial effect of systemic corticosteroids in treat- der way in the United States to investigate the effec- ing acute exacerbations of COPD, with improved tiveness and cost of lung-volume–reduction surgery clinical outcome and reduced length of hospitaliza- in comparison with conventional therapy.98 84,85 tion. The reasons for this discrepancy between NEW TREATMENTS the responses to corticosteroids in acute and chronic COPD may be related to differences in the inflam- Apart from smoking cessation, no treatment, in- matory response (such as increased numbers of eo- cluding corticosteroids, slows the progression of sinophils) or airway edema in exacerbations. COPD. Yet this disease is associated with an active inflammatory process and progressive proteolytic in- NONPHARMACOLOGIC TREATMENTS jury of lung tissues, suggesting that pharmacologic intervention may be possible.99 A better understanding Noninvasive Ventilation of the cellular and molecular mechanisms involved The use of noninvasive positive-pressure ventila- in COPD provides new molecular targets for the de- tion with a simple nasal mask, which eliminates the velopment of drugs,100 and several classes of new necessity for endotracheal intubation, reduces the need drug are now in development (Table 1). for mechanical ventilation in acute exacerbations of COPD in the hospital,86 although some studies have Mediator Antagonists shown little or no benefit.87 Uncontrolled studies Several inflammatory mediators are implicated in have shown that noninvasive positive-pressure venti- COPD, providing logical targets for the develop- lation used at home may improve oxygenation and ment of synthesis inhibitors or receptor antagonists.

276 · July 27, 2000

therapy, although there are doubts that sufficient pro- TABLE 1. NEW TREATMENTS FOR CHRONIC OBSTRUCTIVE tein can be delivered by either approach. PULMONARY DISEASE. New Antiinflammatory Drugs Mediator antagonists Although corticosteroids are ineffective, other an- Leukotriene B4 antagonists (LY 29311, SC-53228, CP-105,696, SB 201146, BIIL284) tiinflammatory treatments, particularly those that in- 5'-Lipoxygenase inhibitors (zileuton, Bay x1005) hibit neutrophilic inflammation, might be effective. Interleukin-8 antagonists (SB 225002, CXCR2 antagonist) Tumor necrosis factor inhibitors (monoclonal antibodies, soluble recep- The most promising of these are phosphodiesterase tors, tumor necrosis factor convertase inhibitors) 4 inhibitors, which have an inhibitory effect on key Antioxidants (stable glutathione analogues) inflammatory cells involved in COPD, including mac- Protease inhibitors rophages, neutrophils, and cytotoxic T lymphocytes.105 Neutrophil elastase inhibitors (ICI 200,355, ONO-5046, MR-889, L-658,758) One phosphodiesterase 4 inhibitor (SB 207499) has Cathepsin inhibitors (suramin) shown promising results in a clinical trial of treatment Nonselective matrix metalloproteinase inhibitors (batimastat, marimastat) of COPD, with improvement in symptoms, quality and selective inhibitors of life, and lung function.106 Several other phospho- a1-Antitrypsin (purified, human recombinant, gene-transfer) Secretory leukoprotease inhibitor (human recombinant, gene-transfer) diesterase 4 inhibitors are in development for the Elafin (human recombinant, gene-transfer) treatment of COPD, although a limitation of drugs Antiinflammatory drugs in this class is the common side effect of nausea. Phosphodiesterase 4 inhibitors (SB 207499, CP-80,633, CDP-840) Other novel antiinflammatory approaches under de- Nuclear factor-kB inhibitors (IkB kinase inhibitors, IkB-a gene-transfer) Adhesion-molecule inhibitors (anti-CD11/CD18, anti–intercellular adhe- velopment include inhibitors of NF-kB, inhibitors of sion molecule 1, E-selectin inhibitors, glycoprotein selectin inhibitors) p38 mitogen-activated protein kinase, and interleu- Interleukin-10 100 p38 Mitogen–activated protein kinase inhibitors (SB 203580, SB 220025, kin-10, although none have reached clinical trials. RWJ 67657) Drug Delivery Current inhaler devices have been designed to deliver drugs optimally to the conducting airways in patients with asthma. However, COPD predominant- These include 5-lipoxygenase inhibitors, which pre- ly affects the peripheral airways and lung parenchyma, vent the synthesis of leukotriene B4, and specific leu- which may not be optimally targeted by current in- kotriene B4 antagonists, several of which are now being halers. It is likely that systemic drugs or new inhaler evaluated for the treatment of COPD.100 Specific an- devices delivering smaller aerosolized particles will tagonists of CXCR2, one of the receptors on neu- be more useful in COPD. In the future, targeted de- trophils that are activated by interleukin-8, have been livery to specific cells, such as macrophages, may be developed,101 and humanized antibodies and soluble possible, particularly if new treatments are found to receptors that block TNF-a have already been devel- have systemic toxicity. oped for use in other chronic inflammatory diseases.102 More potent and stable antioxidants are also FUTURE DEVELOPMENTS under development. However, it is not certain that In view of the high and increasing global preva- antagonizing a single mediator will have a substan- lence of COPD, its continuing high morbidity and tial clinical effect, since many mediators with over- mortality, and the consequent high health care costs, lapping actions are involved in COPD. more attention should be focused on the prevention and treatment of the disease. The previous view of Protease Inhibitors COPD as untreatable should now be replaced by a Several inhibitors of neutrophil elastase are now in positive approach to management with several meas- clinical development. Only one has been tested clin- ures that improve the quality of life in patients with ically.103 The study showed no benefit, but this may symptomatic disease. COPD remains insufficiently reflect the fact that several proteases are involved in recognized in family practice and is often still treated COPD or the fact that it may be difficult to monitor as poorly responsive asthma. Smoking cessation is of efficacy in such a slowly progressive disease. Several the utmost importance, but COPD may also be due nonselective matrix metalloproteinase inhibitors have to other causes. Why only some people are suscep- been developed, and there is now a search for inhib- tible to COPD is not yet understood, but it is likely itors that will be more selective and that therefore that advances in molecular genetics will provide the will not have the musculoskeletal side effects that have means to identify those at risk in the future. At present, hampered the development of this class of drug.104 smoking cessation is the only strategy that prevents Another approach is supplementation of endogenous the relentless progression of airflow obstruction, but antiproteases, such as a1-antitrypsin, secretory leu- it is likely that new drugs will be effective in the fu- koprotease inhibitor, or elafin, by the administration ture. Once these drugs become available, it will be of human recombinant products or even by gene important to detect the disease at an early stage —

Volume 343 Number 4 · 277

before symptoms begin — to prevent the disease bronchial lavage of patients with chronic obstructive pulmonary disease. Eur Respir J 1998;12:380-6. from progressing to the point at which it consumes 23. Keatings VM, Collins PD, Scott DM, Barnes PJ. Differences in inter- a large amount of medical resources. There are now leukin-8 and tumor necrosis factor-a in induced sputum from patients with exciting research opportunities, and investigators can chronic obstructive pulmonary disease or asthma. Am J Respir Crit Care Med 1996;153:530-4. have a substantial impact on this increasingly impor- 24. Saetta M, Di Stefano A, Maestrelli P, et al. Airway eosinophilia in tant disease. chronic bronchitis during exacerbations. Am J Respir Crit Care Med 1994; 150:1646-52. 25. Hill AT, Bayley D, Stockley RA. The interrelationship of sputum in- I am indebted to Neil Pride for his helpful comments on the flammatory markers in patients with chronic bronchitis. Am J Respir Crit manuscript. Care Med 1999;160:893-8. 26. Shapiro SD, Senior RM. Matrix metalloproteinases: matrix degradation and more. Am J Respir Cell Mol Biol 1999;20:1100-2. REFERENCES 27. Liu AN, Mohammed AZ, Rice WR, et al. Perforin-independent CD8(+) T-cell-mediated cytotoxicity of alveolar epithelial cells is preferen- 1. American Thoracic Society. Standards for the diagnosis and care of pa- tially mediated by tumor necrosis factor-alpha: relative insensitivity to Fas tients with chronic obstructive pulmonary disease. Am J Respir Crit Care ligand. Am J Respir Cell Mol Biol 1999;20:849-58. Med 1995;152:S77-S121. 28. Stockley RA. Neutrophils and protease/antiprotease imbalance. Am 2. Vestbo J, Prescott E, Lange P. Association of chronic mucus hyperse- J Respir Crit Care Med 1999;160:S49-S52. cretion with FEV1 decline and chronic obstructive pulmonary disease mor- 29. Gottlieb DJ, Stone PJ, Sparrow D, et al. Urinary desmosine excretion bidity. Am J Respir Crit Care Med 1996;153:1530-5. in smokers with and without rapid decline of lung function: the Normative 3. Centers for Disease Control and Prevention. Current estimates from the Aging Study. Am J Respir Crit Care Med 1996;154:1290-5. National Health Interview Survey, 1995. Vital and health statistics. Wash- 30. Yoshioka A, Betsuyaku T, Nishimura M, Miyamoto K, Kondo T, ington, D.C.: Government Printing Office, 1996. (DHHS publication no. Kawakami Y. Excessive neutrophil elastase in bronchoalveolar lavage fluid (PHS) 96-1527.) in subclinical emphysema. Am J Respir Crit Care Med 1995;152:2127-32. 4. Lopez AD, Murray CC. The global burden of disease, 1990-2020. Nat 31. Sommerhoff CP, Nadel JA, Basbaum CB, Caughey GH. Neutrophil Med 1998;4:1241-3. elastase and cathepsin G stimulate secretion from cultured bovine airway 5. Silverman EK, Chapman HA, Drazen JM, et al. Genetic epidemiology gland serous cells. J Clin Invest 1990;85:682-9. of severe, early-onset chronic obstructive pulmonary disease: risk to rela- 32. Witko-Sarsat V, Halbwachs-Mecarelli L, Schuster A, et al. Proteinase tives for airflow obstruction and chronic bronchitis. Am J Respir Crit Care 3, a potent secretagogue in airways, is present in cystic fibrosis sputum. Am Med 1998;157:1770-8. J Respir Cell Mol Biol 1999;20:729-36. 6. Mahadeva R, Lomas DA. Genetics and respiratory disease. 2. Alpha 33. Finlay GA, O’Driscoll LR, Russell KJ, et al. Matrix metalloproteinase 1-antitrypsin deficiency, cirrhosis and emphysema. Thorax 1998;53:501- expression and production by alveolar macrophages in emphysema. Am 5. J Respir Crit Care Med 1997;156:240-7. 7. Sandford AJ, Weir TD, Pare PD. Genetic risk factors for chronic ob- 34. Culpitt SV, Maziak W, Loukidis S, Nightingale JA, Matthews JL, structive pulmonary disease. Eur Respir J 1997;10:1380-91. Barnes PJ. Effect of high dose inhaled steroid on cells, cytokines, and pro- 8. Barnes PJ. Genetics and pulmonary medicine. 9. Molecular genetics of teases in induced sputum in chronic obstructive pulmonary disease. Am chronic obstructive pulmonary disease. Thorax 1999;54:245-52. J Respir Crit Care Med 1999;160:1635-9. 9. Huang SL, Su CH, Chang SC. Tumor necrosis factor-alpha gene poly- 35. Betsuyaku T, Nishimura M, Takeyabu K, et al. Neutrophil granule pro- morphism in chronic bronchitis. Am J Respir Crit Care Med 1997;156: teins in bronchoalveolar lavage fluid from subjects with subclinical emphy- 1436-9. sema. Am J Respir Crit Care Med 1999;159:1985-91. 10. Higham MA, Pride NB, Alikhan A, Morrell NW. Tumour necrosis fac- 36. Ohnishi K, Takagi M, Kurokawa Y, Satomi S, Konttinen YT. Matrix tor-a gene promoter polymorphism in chronic obstructive pulmonary dis- metalloproteinase-mediated extracellular matrix protein degradation in hu- ease. Eur Respir J 2000;15:281-4. man pulmonary emphysema. Lab Invest 1998;78:1077-87. 11. Smith CAD, Harrison DJ. Association between polymorphism in gene 37. Hautamaki RD, Kobayashi DK, Senior RM, Shapiro SD. Requirement for microsomal epoxide hydrolase and susceptibility to emphysema. Lancet for macrophage elastase for cigarette smoke-induced emphysema in mice. 1997;350:630-3. Science 1997;277:2002-4. 12. Dennis RJ, Maldonado D, Norman S, Baena E, Martinez G. Wood- 38. Repine JE, Bast A, Lankhorst I. Oxidative stress in chronic obstructive smoke exposure and risk for obstructive airways disease among women. pulmonary disease. Am J Respir Crit Care Med 1997;156:341-57. Chest 1996;109:115-9. 39. Dekhuijzen PNR, Aben KK, Dekker I, et al. Increased exhalation of 13. Hospers JJ, Schouten JP, Weiss ST, Rijcken B, Postma DS. Asthma at- hydrogen peroxide in patients with stable and unstable chronic obstructive tacks with eosinophilia predict mortality from chronic obstructive pulmo- pulmonary disease. Am J Respir Crit Care Med 1996;154:813-6. nary disease in a general population sample. Am J Respir Crit Care Med 40. Pratico D, Basili S, Vieri M, Cordova C, Violi F, Fitzgerald GA. 1999;160:1869-74. Chronic obstructive pulmonary disease is associated with an increase in uri-

14. Tashkin DP, Altose MD, Connett JE, Kanner RE, Lee WW, Wise RA. nary levels of isoprostane F2a-III, an index of oxidant stress. Am J Respir Methacholine reactivity predicts changes in lung function over time in Crit Care Med 1998;158:1709-14. smokers with early chronic obstructive pulmonary disease. Am J Respir 41. Montuschi P, Corradi M, Ciabattoni G, et al. Breath condensate analy- Crit Care Med 1996;153:1802-11. sis of 8-isoprostane: a new approach for assessment of oxidative stress in 15. Strachan DP, Seagroatt V, Cook DG. Chest illness in infancy and patients with chronic obstructive pulmonary disease. Am J Respir Crit Care chronic respiratory disease in later life: an analysis by month of birth. Int Med 1999;159:Suppl:A798. abstract. J Epidemiol 1994;23:1060-8. 42. Barnes PJ, Karin M. Nuclear factor-kB — a pivotal transcription factor 16. Jeffery PK. Structural and inflammatory changes in COPD: a compar- in chronic inflammatory diseases. N Engl J Med 1997;336:1066-71. ison with asthma. Thorax 1998;53:129-36. 43. Rahman I, Morrison D, Donaldson K, MacNee W. Systemic oxidative 17. Barnes PJ. Mechanisms in COPD: differences from asthma. Chest stress in asthma, COPD, and smokers. Am J Respir Crit Care Med 1996; 2000;117:Suppl:10S-14S. 154:1055-60. 18. Finkelstein R, Fraser RS, Ghezzo H, Cosio MG. Alveolar inflamma- 44. Noguera A, Busquets X, Sauleda J, Villaverde JM, MacNee W, Agusti tion and its relation to emphysema in smokers. Am J Respir Crit Care Med AG. Expression of adhesion molecules and G proteins in circulating neu- 1995;152:1666-72. trophils in chronic obstructive pulmonary disease. Am J Respir Crit Care 19. Saetta M, Di Stefano A, Turato G, et al. CD8+ T-lymphocytes in pe- Med 1998;158:1664-8. ripheral airways of smokers with chronic obstructive pulmonary disease. 45. Di Francia M, Barbier D, Mege JL, Orehek J. Tumor necrosis factor- Am J Respir Crit Care Med 1998;157:822-6. a levels and weight loss in chronic obstructive pulmonary disease. Am 20. O’Shaughnessy TC, Ansari TW, Barnes NC, Jeffery PK. Inflammation J Respir Crit Care Med 1994;150:1453-5. in bronchial biopsies of subjects with chronic bronchitis: inverse relation- 46. Schols AM, Buurman WA, Staal van den Brekel AJ, Dentener MA, ship of CD8+ T lymphocytes with FEV1. Am J Respir Crit Care Med Wouters EF. Evidence for a relation between metabolic derangements and 1997;155:852-7. increased levels of inflammatory mediators in a subgroup of patients with 21. Di Stefano A, Capelli A, Lusuardi M, et al. Severity of airflow limita- chronic obstructive pulmonary disease. Thorax 1996;51:819-24. tion is associated with severity of airway inflammation in smokers. Am 47. Schols AM, Slangen J, Volovics L, Wouters EF. Weight loss is a revers- J Respir Crit Care Med 1998;158:1277-85. ible factor in the prognosis of chronic obstructive pulmonary disease. Am 22. Pesci A, Balbi B, Majori M, et al. Inflammatory cells and mediators in J Respir Crit Care Med 1998;157:1791-7.

278 · July 27, 2000

48. de Godoy I, Donahoe M, Calhoun WJ, Mancino J, Rogers RM. Ele- 73. Saint S, Bent S, Vittinghoff E, Grady D. Antibiotics in chronic ob- vated TNF-alpha production by peripheral blood monocytes of weight-los- structive pulmonary disease exacerbations: a meta-analysis. JAMA 1995; ing COPD patients. Am J Respir Crit Care Med 1996;153:633-7. 273:957-60. 49. Schols AM, Creutzberg EC, Buurman WA, Campfield LA, Saris WH, 74. Tarpy SP, Celli BR. Long-term oxygen therapy. N Engl J Med 1995; Wouters EF. Plasma leptin is related to proinflammatory status and dietary 333:710-4. intake in patients with chronic obstructive pulmonary disease. Am J Respir 75. Gorecka D, Gorzelak K, Sliwinski P, Tobiasz M, Zielinski J. Effect of Crit Care Med 1999;160:1220-6. long-term oxygen therapy on survival in patients with chronic obstructive 50. Bernard S, LeBlanc P, Whittom F, et al. Peripheral muscle weakness in pulmonary disease with moderate hypoxaemia. Thorax 1997;52:674-9. patients with chronic obstructive pulmonary disease. Am J Respir Crit Care 76. Chaouat A, Weitzenblum E, Kessler R, et al. A randomized trial of Med 1998;158:629-34. nocturnal oxygen therapy in chronic obstructive pulmonary disease pa- 51. Maltais F, Sullivan MJ, LeBlanc P, et al. Altered expression of myosin tients. Eur Respir J 1999;14:1002-8. heavy chain in the vastus lateralis muscle in patients with COPD. Eur Res- 77. Keatings VM, Jatakanon A, Worsdell YM, Barnes PJ. Effects of inhaled pir J 1999;13:850-4. and oral glucocorticoids on inflammatory indices in asthma and COPD. 52. Schols AM, Soeters PB, Mostert R, Pluymers RJ, Wouters EF. Physi- Am J Respir Crit Care Med 1997;155:542-8. ologic effects of nutritional support and anabolic steroids in patients with 78. Meagher LC, Cousin JM, Seckl JR, Haslett C. Opposing effects of chronic obstructive pulmonary disease: a placebo-controlled randomized glucocorticoids on the rate of apoptosis in neutrophilic and eosinophilic trial. Am J Respir Crit Care Med 1995;152:1268-74. granulocytes. J Immunol 1996;156:4422-8. 53. Matsuse T, Hayashi S, Kuwano K, Keunecke H, Jefferies WA, Hogg 79. Nightingale JA, Rogers DF, Fan Chung K, Barnes PJ. No effect of in- JC. Latent adenoviral infection in the pathogenesis of chronic airways ob- haled budesonide on the response to inhaled ozone in normal subjects. Am struction. Am Rev Respir Dis 1992;146:177-84. J Respir Crit Care Med 2000;161:479-86. 54. Keicho N, Higashimoto Y, Bondy GP, Elliott WM, Hogg JC, Hayashi 80. Pauwels RA, Löfdahl C-G, Laitinen LA, et al. Long-term treatment S. Endotoxin-specific NF-kB activation in pulmonary epithelial cells har- with inhaled budesonide in persons with mild chronic obstructive pulmo- boring adenovirus E1A. Am J Physiol 1999;277:L523-L532. nary disease who continue smoking. N Engl J Med 1999;340:1948-53. 55. Turato G, Di Stefano A, Maestrelli P, et al. Effect of smoking cessation 81. Vestbo J, Sorensen T, Lange P, Brix A, Torre P, Viskum K. Long-term on airway inflammation in chronic bronchitis. Am J Respir Crit Care Med effect of inhaled budesonide in mild and moderate chronic obstructive pul- 1995;152:1262-7. monary disease: a randomised controlled trial. Lancet 1999;353:1819-23. 56. Rutgers SR, Postma DS, ten Hacken NH, et al. Ongoing airway in- 82. Burge PS, Calverley PMA, Jones PW, Spencer S, Anderson JA, Maslen flammation in patients with COPD who do not currently smoke. Thorax TK. Randomised, double-blind, placebo-controlled study of fluticasone 2000;55:12-8. propionate in patients with moderate to severe chronic obstructive pulmo- 57. Seemungal TA, Donaldson GC, Paul EA, Bestall JC, Jeffries DJ, nary disease: the ISOLDE trial. BMJ 2000;320:1297-303. Wedzicha JA. Effect of exacerbation on quality of life in patients with 83. Paggiaro PL, Dahle R, Bakran I, Frith L, Hollingworth K, Efthimou chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998; J. Multicentre randomised placebo-controlled trial of inhaled fluticasone 157:1418-22. propionate in patients with chronic obstructive pulmonary disease. Lancet 58. Seemungal TA, Harper-Owen R, Bhowmik A, Sapsford R, Jefferies 1998;351:773-80. DJ, Wedzicha JA. Rhinovirus infects the lower respiratory tract during 84. Niewoehner DE, Erbland ML, Deupree RH, et al. Effect of systemic COPD exacerbation. Thorax 1999;54:Suppl 3:A70. abstract. glucocorticoids on exacerbations of chronic obstructive pulmonary disease. 59. Donaldson GC, Seemungal T, Jeffries DJ, Wedzicha JA. Effect of tem- N Engl J Med 1999;340:1941-7. perature on lung function and symptoms in chronic obstructive pulmonary 85. Davies L, Angus RM, Calverley PM. Oral corticosteroids in patients disease. Eur Respir J 1999;13:844-9. admitted to hospital with exacerbations of chronic obstructive pulmonary 60. Bhowmik A, Seemungal TA, Sapsford RJ, Wedzicha JA. Relation of disease: a prospective randomised controlled trial. Lancet 1999;354:456- sputum inflammatory markers to symptoms and lung function changes in 60. COPD exacerbations. Thorax 2000;55:114-20. 86. Antonelli M, Conti G, Rocco M, et al. A comparison of noninvasive 61. Maziak W, Loukides S, Culpitt S, Sullivan P, Kharitonov SA, Barnes positive-pressure ventilation and conventional mechanical ventilation in pa- PJ. Exhaled nitric oxide in chronic obstructive pulmonary disease. Am tients with acute respiratory failure. N Engl J Med 1998;339:429-35. J Respir Crit Care Med 1998;157:998-1002. 87. Barbe F, Togores B, Rubi M, Pons S, Maimo A, Agusti AG. Noninva- 62. Agusti AG, Villaverde JM, Togores B, Bosch M. Serial measurements sive ventilatory support does not facilitate recovery from acute respiratory of exhaled nitric oxide during exacerbations of chronic obstructive pulmo- failure in chronic obstructive pulmonary disease. Eur Respir J 1996;9: nary disease. Eur Respir J 1999;14:523-8. 1240-5. 63. Anthonisen NR, Connett JE, Kiley JP, et al. Effects of smoking inter- 88. Jones SE, Packham S, Hebden M, Smith AP. Domiciliary nocturnal vention and the use of an inhaled anticholinergic bronchodilator on the intermittent positive pressure ventilation in patients with respiratory failure

rate of decline of FEV1: the Lung Health Study. JAMA 1994;272:1497- due to severe COPD: long-term follow up and effect on survival. Thorax 505. 1998;53:495-8. 64. Silagy C, Mant D, Fowler G, Lodge M. Meta-analysis on efficacy of 89. Strumpf DA, Millman RP, Carlisle CC, et al. Nocturnal positive-pres- nicotine replacement therapies in smoking cessation. Lancet 1994;343: sure ventilation via nasal mask in patients with severe chronic obstructive 139-42. pulmonary disease. Am Rev Respir Dis 1991;144:1234-9. 65. Jorenby DE, Leischow SJ, Nides MA, et al. A controlled trial of sus- 90. Meecham Jones DJ, Paul EA, Jones PW, Wedzicha JA. Nasal pressure tained-release bupropion, a nicotine patch, or both for smoking cessation. support ventilation plus oxygen compared with oxygen therapy alone in hy- N Engl J Med 1999;340:685-91. percapnic COPD. Am J Respir Crit Care Med 1995;152:538-44. 66. O’Donnell DE, Lam M, Webb KA. Measurement of symptoms, lung 91. Lacasse Y, Wong E, Guyatt GH, King D, Cook DJ, Goldstein RS. hyperinflation, and endurance during exercise in chronic obstructive pul- Meta-analysis of respiratory rehabilitation in chronic obstructive pulmonary monary disease. Am J Respir Crit Care Med 1998;158:1557-65. disease. Lancet 1996;348:1115-9. 67. Boyd G, Morice AH, Pounsford JC, Siebert M, Peslis N, Crawford C. 92. Griffiths TL, Burr ML, Campbell IA, et al. Results at 1 year of out- An evaluation of salmeterol in the treatment of chronic obstructive pulmo- patient multidisciplinary pulmonary rehabilitation: a randomised con- nary disease. Eur Respir J 1997;10:815-21. [Erratum, Eur Respir J 1997; trolled trial. Lancet 2000;355:362-8. 10:1696.] 93. Doyle RL, Mark JB. Lung volume reduction surgery for the treatment 68. Mahler DA, Donohue JF, Barbee RA, et al. Efficacy of salmeterol xin- of chronic obstructive pulmonary disease. Adv Intern Med 1998;43:233- afoate in the treatment of COPD. Chest 1999;115:957-65. 52. 69. Maesen BL, Westermann CJ, Duurkens VA, van den Bosch JM. Ef- 94. Geddes D, Davies M, Koyama H, et al. Effect of lung-volume–reduc- fects of formoterol in apparently poorly reversible chronic obstructive pul- tion surgery in patients with severe emphysema. N Engl J Med 2000;343: monary disease. Eur Respir J 1999;13:1103-8. 239-45. 70. Dowling RB, Johnson M, Cole PJ, Wilson R. Effect of salmeterol on 95. Ingenito EP, Evans RB, Loring SH, et al. Relation between preoper- Haemophilus influenzae infection of respiratory mucosa in vitro. Eur Respir ative inspiratory lung resistance and the outcome of lung-volume–reduc- J 1998;11:86-90. tion surgery for emphysema. N Engl J Med 1998;338:1181-5. 71. Rennard SI, Serby CW, Ghafouri M, Johnson PA, Friedman M. Ex- 96. Cooper JD, Patterson GA, Sundaresan RS, et al. Results of 150 con- tended therapy with ipratropium is associated with improved lung function secutive bilateral lung volume reduction procedures in patients with severe in patients with COPD: a retrospective analysis of data from seven clinical emphysema. J Thorac Cardiovasc Surg 1996;112:1319-29. trials. Chest 1996;110:62-70. 97. Criner GJ, Cordova FC, Furukawa S, et al. Prospective randomized tri- 72. Disse B, Speck GA, Rominger KL, Witek TJ Jr, Hammer R. Tiotro- al comparing bilateral lung volume reduction surgery to pulmonary reha- pium (Spiriva): mechanistical considerations and clinical profile in obstruc- bilitation in severe chronic obstructive pulmonary disease. Am J Respir Crit tive lung disease. Life Sci 1999;64:457-64. Care Med 1999;160:2018-27.

Volume 343 Number 4 · 279

98. The National Emphysema Treatment Trial Research Group. Rationale sis of the mechanism of action predicts utility in other diseases. Transplant and design of the National Emphysema Treatment Trial: a prospective ran- Proc 1998;30:4126-7. domized trial of lung volume reduction surgery. Chest 1999;116:1750- 103. Luisetti M, Sturani C, Sella D, et al. MR889, a neutrophil elastase 61. inhibitor, in patients with chronic obstructive pulmonary disease: a double- 99. Barnes PJ. New therapies for chronic obstructive pulmonary disease. blind, randomised, placebo-controlled clinical trial. Eur Respir J 1996;9: Thorax 1998;53:137-47. 1482-6. 100. Idem. Novel approaches and targets for treatment of chronic ob- 104. Cawston TE. Metalloproteinase inhibitors and the prevention of con- structive pulmonary disease. Am J Respir Crit Care Med 1999;160:S72- nective tissue breakdown. Pharmacol Ther 1996;70:163-82. S79. 105. Torphy TJ. Phosphodiesterase isoenzymes: molecular targets for nov- 101. White JR, Lee JM, Young PR, et al. Identification of a potent, selec- el antiasthma agents. Am J Respir Crit Care Med 1998;157:351-70. tive non-peptide CXCR2 antagonist that inhibits interleukin-8-induced 106. Torphy TJ, Barnette MS, Underwood DC, et al. Ariflo (SB 207499), neutrophil migration. J Biol Chem 1998;273:10095-8. a second generation phosphodiesterase 4 inhibitor for the treatment of 102. Feldman M, Taylor P, Paleolog E, Brennan FM, Maini RN. Anti- asthma and COPD: from concept to clinic. Pulm Pharmacol Ther 1999; TNFa therapy is useful in rheumatoid arthritis and Crohn’s disease: analy- 12:131-5.

280 · July 27, 2000

The new england journal of medicine

original article

The Body-Mass Index, Airflow Obstruction, Dyspnea, and Exercise Capacity Index in Chronic Obstructive Pulmonary Disease

Bartolome R. Celli, M.D., Claudia G. Cote, M.D., Jose M. Marin, M.D., Ciro Casanova, M.D., Maria Montes de Oca, M.D., Reina A. Mendez, M.D., Victor Pinto Plata, M.D., and Howard J. Cabral, Ph.D.

abstract background Chronic obstructive pulmonary disease (COPD) is characterized by an incompletely re- From the COPD Center at St. Elizabeth’s versible limitation in airflow. A physiological variable — the forced expiratory volume Medical Center, Tufts University School of Medicine, Boston (B.R.C., V.P.P.); Bay Pines in one second (FEV1) — is often used to grade the severity of COPD. However, patients Veterans Affairs Medical Center, Bay Pines, with COPD have systemic manifestations that are not reflected by the FEV1. We hypoth- Fla. (C.G.C.); Hospital Miguel Servet, Zara- esized that a multidimensional grading system that assessed the respiratory and sys- goza, Spain (J.M.M.); Hospital Nuestra Senora de La Candelaria, Tenerife, Spain temic expressions of COPD would better categorize and predict outcome in these pa- (C.C.); Hospital Universitario de Caracas tients. and Hospital Jose I. Baldo, Caracas, Vene- zuela (M.M.O., R.A.M.); and Boston Uni- methods versity School of Public Health, Boston (H.J.C.). Address reprint requests to Dr. We first evaluated 207 patients and found that four factors predicted the risk of death in Celli at Pulmonary and Critical Care Medi- this cohort: the body-mass index (B), the degree of airflow obstruction (O) and dys- cine, St. Elizabeth’s Medical Center, 736 Cambridge St., Boston, MA 02135, or at pnea (D), and exercise capacity (E), measured by the six-minute–walk test. We used [email protected]. these variables to construct the BODE index, a multidimensional 10-point scale in which higher scores indicate a higher risk of death. We then prospectively validated the N Engl J Med 2004;350:1005-12. index in a cohort of 625 patients, with death from any cause and from respiratory caus- Copyright © 2004 Massachusetts Medical Society. es as the outcome variables. results There were 25 deaths among the first 207 patients and 162 deaths (26 percent) in the validation cohort. Sixty-one percent of the deaths in the validation cohort were due to respiratory insufficiency, 14 percent to myocardial infarction, 12 percent to lung cancer, and 13 percent to other causes. Patients with higher BODE scores were at higher risk for death; the hazard ratio for death from any cause per one-point increase in the BODE score was 1.34 (95 percent confidence interval, 1.26 to 1.42; P<0.001), and the hazard ratio for death from respiratory causes was 1.62 (95 percent confidence interval, 1.48 to 1.77; P<0.001). The C statistic for the ability of the BODE index to predict the risk of death was larger than that for the FEV1 (0.74 vs. 0.65). conclusions The BODE index, a simple multidimensional grading system, is better than the FEV1 at predicting the risk of death from any cause and from respiratory causes among patients with COPD.

n engl j med 350;10 www.nejm.org march 4, 2004 1005

The new england journal of medicine

hronic obstructive pulmonary and six-minute–walk tests; a myocardial infarction c disease (COPD), a common disease char- within the preceding four months; unstable angi- acterized by a poorly reversible limitation na; or congestive heart failure (New York Heart As- in airflow,1 is predicted to be the third most fre- sociation class III or IV). quent cause of death in the world by 2020.2 The risk of death in patients with COPD is often graded variables selected for the bode index with the use of a single physiological variable, the We determined the following variables in the first 1,3,4 forced expiratory volume in one second (FEV1). 207 patients who were recruited between 1995 and However, other risk factors, such as the presence 1997: age; sex; pack-years of smoking; FVC; FEV1, of hypoxemia or hypercapnia,5,6 a short distance measured in liters and as a percentage of the pre- walked in a fixed time,7 a high degree of functional dicted value according to the guidelines of the breathlessness,8 and a low body-mass index (the American Thoracic Society11; the best of two six- weight in kilograms divided by the square of the minute–walk tests performed at least 30 minutes height in meters),9,10 are also associated with an apart12; the degree of dyspnea, measured with the increased risk of death. We hypothesized that a mul- use of the modified Medical Research Council tidimensional grading system that assessed the res- (MMRC) dyspnea scale13; the body-mass index9,10; piratory, perceptive, and systemic aspects of COPD the functional residual capacity and inspiratory would better categorize the illness and predict the capacity11; the hematocrit; and the albumin level. outcome than does the FEV1 alone. We used data The validated Charlson index was used to deter- from an initial cohort of 207 patients to identify mine the degree of comorbidity. This index has four factors that predicted the risk of death: the been shown to predict mortality.14 The differenc- body-mass index (B), the degree of airflow ob- es in these values between survivors and nonsur- struction (O) and functional dyspnea (D), and exer- vivors are shown in Table 1. cise capacity (E) as assessed by the six-minute– Each of these possible explanatory variables walk test. We then integrated these variables into a was independently evaluated to determine its as- multidimensional index — the BODE index — and sociation with one-year mortality in a stepwise for- validated the index in a second cohort of 625 pa- ward logistic-regression analysis. A subgroup of tients, with death from any cause and death from four variables had the strongest association — the respiratory causes as the outcome variables. body-mass index, FEV1 as a percentage of the predicted value, score on the MMRC dyspnea scale, methods and the distance walked in six minutes (generalized r2=0.21, P<0.001) — and these were includ- Between January 1997 and June 2002, a total of ed in the BODE index (Table 2). All these variables 859 outpatients with a wide range in the severity predict important outcomes, are easily measured, of COPD were recruited from clinics in the United and may change over time. We chose the post-bron- States, Spain, and Venezuela. The study was ap- chodilator FEV1 as a percent of the predicted value, proved by the human-research review board at each classified according to the three stages identified site, and all patients provided written informed con- by the American Thoracic Society, because it can be sent. COPD was defined by a history of smoking used to predict health status,15 the rate of exacer- 16 that exceeded 20 pack-years and a ratio of FEV1 to bation of COPD, the pharmacoeconomic costs of forced vital capacity (FVC) of less than 0.7 measured the disease,17 and the risk of death.18,19 We chose 20 minutes after the administration of albuterol.1 the MMRC dyspnea scale because it predicts the All patients were in clinically stable condition and likelihood of survival among patients with COPD8 receiving appropriate therapy. Patients who were and correlates well with other scales and health- receiving inhaled oxygen had to have been taking status scores.20,21 We chose the six-minute–walk a stable dose for at least six months before study test because it predicts the risk of death in patients entry. The exclusion criteria were an illness other with COPD,7 patients who have undergone lung- than COPD that was likely to result in death within reduction surgery,22 patients with cardiomyopa- three years; asthma, defined as an increase in the thy,23 and those with pulmonary hypertension.24 12 FEV1 of more than 15 percent above the base-line In addition, the test has been standardized, the value or of 200 ml after the administration of a bron- clinically significant thresholds have been deter- chodilator; an inability to take the lung-function mined,25 and it can be used to predict resource uti-

1006 n engl j med 350;10 www.nejm.org march 4, 2004

a multidimensional grading system in chronic obstructive pulmonary disease lization.26 Finally, there is an inverse relation between body-mass index and survival9,10 that is not Table 1. Characteristics of the First 207 Patients, According to Whether They Survived.* linear but that has an inflection point, which was 10 21 in our cohort and in another study. Survived Died Characteristic (N=182) (N=25) P Value validation of the bode index mean ±SD The BODE index was validated prospectively in Age (yr) 66±9 70±7 0.03 two ways in a different cohort of 625 patients who FVC (liters) 2.78±0.89 2.27±0.57 0.04 were recruited between January 1997 and January 2003. First, we used the empirical model: for each FEV1 Liters 1.31±0.63 0.84±0.33 0.002 threshold value of FEV1, distance walked in six min- Percent of predicted 43±19 28±12 0.001 utes, and score on the MMRC dyspnea scale shown FRC (%) 150±43 170±52 0.12 in Table 2, the patients received points ranging Inspiratory capacity (liters) 2.0±0.7 1.6±0.5 0.007 from 0 (lowest value) to 3 (maximal value). For body-mass index the values were 0 or 1, because of MMRC dyspnea scale† 2.7±0.89 3.3±0.87 0.001 the unique relation between body-mass index and Distance walked in 6 min (m) 264±113 175±86 0.001 survival described above. The points for each varia- Body-mass index‡ 26±5 23±5 0.002 ble were added, so that the BODE index ranged Hematocrit (%) 42±5 39±5 0.01 from 0 to 10 points, with higher scores indicating a Albumin (g/ml) 4.0±0.3 3.8±0.4 0.08 greater risk of death. In an exploratory analysis, the Smoking history (pack-yr) 88±48 77±48 0.36 various components of the BODE index were assigned different weights, with no corresponding Charlson index§ 2.9±1.3 5.9±1.9 0.02 increase in predictive value. * FVC denotes forced vital capacity, FEV1 forced expiratory volume in one second, and FRC functional residual capacity. study protocol † Scores on the modified Medical Research Council (MMRC) dyspnea scale can In the cohort, patients were evaluated with the use range from 0 to 4, with a score of 4 indicating that the patient is too breathless to leave the house or becomes breathless when dressing or undressing. of the BODE index within six weeks after enroll- ‡ The body-mass index is the weight in kilograms divided by the square of the ment and were seen every three to six months for height in meters. at least two years or until death. The patient and § Scores on the Charlson index can range from 0 to 33, with higher scores indicating more coexisting conditions. family were contacted if the patient failed to return for appointments. Death from any cause and from specific respiratory causes was recorded. The cause of death was determined by the investigators at each Table 2. Variables and Point Values Used for the Computation of the Body- site after reviewing the medical record and death Mass Index, Degree of Airflow Obstruction and Dyspnea, and Exercise certificate. Capacity (BODE) Index.* statistical analysis Variable Points on BODE Index Data for continuous variables are presented as 0 1 2 3 ± means SD. Comparison among the three coun- FEV1 (% of predicted)† ≥65 50–64 36–49 ≤35 tries was completed with the use of one-way analy- Distance walked in 6 min (m) ≥350 250–349 150–249 ≤149 sis of variance. The differences between survivors MMRC dyspnea scale‡ 0–1 2 3 4 and nonsurvivors in pulmonary-function variables and other pertinent characteristics were established Body-mass index§ >21 ≤21 with the use of t-tests for independent samples. * The cutoff values for the assignment of points are shown for each variable. To evaluate the capacity of the BODE index to pre- The total possible values range from 0 to 10. FEV1 denotes forced expiratory dict the risk of death, we performed Cox propor- volume in one second. tional-hazards regression analyses.27 We estimat- † The FEV1 categories are based on stages identified by the American Thoracic Society. ed the hazard ratio, 95 percent confidence interval, ‡ Scores on the modified Medical Research Council (MMRC) dyspnea scale can and P value for the BODE score, before and after range from 0 to 4, with a score of 4 indicating that the patient is too breathless adjustment for coexisting conditions as measured to leave the house or becomes breathless when dressing or undressing. § The values for body-mass index were 0 or 1 because of the inflection point in by the Charlson index. We repeated these analyses the inverse relation between survival and body-mass index at a value of 21. using the BODE index as the predictor of interest in

n engl j med 350;10 www.nejm.org march 4, 2004 1007

The new england journal of medicine

dummy-variable form, using the first quartile as the reference group. These analyses yielded estimates Table 3. Characteristics of the Patients in the Validation Cohort. of risk similar to those obtained from analyses using the BODE score as a continuous variable. Thus, No. of Patients (%) we focus our presentation on the predictive charac- Characteristic (N=625) teristics of the BODE index and present only bivari- Severity of COPD* ate results for survival according to quartiles of the Stage I (FEV1 >50% of predicted) 186 (30) Stage II (FEV1, 36–50% 204 (33) BODE index in a Kaplan–Meier analysis. The statis- of predicted) tical significance was evaluated with the use of the Stage III (FEV1 ≤35 percent 235 (38) log-rank test. We also performed bivariate analysis of predicted) on the stage of COPD according to the validated BODE index score† 3 0 33 (5) staging system of the American Thoracic Society. 1 55 (9) In the Cox regression analysis, we assessed the 2 81 (13) reliability of the model with the body-mass index, 3 95 (15) 4 92 (15) degree of airflow obstruction and dyspnea, and ex- 5 75 (12) ercise capacity score as the predictor of the time to 6 55 (9) death by computing bootstrap estimates using the 7 60 (10) 8 43 (7) full sample for the hazard ratio and its 95 percent 9 24 (4) confidence interval (according to the percentile 10 12 (2) method). This approach has the advantage of not requiring that the data be split into subgroups and * Because of rounding, percentages do not total 100. The three stages of chronic obstructive pulmonary disease is more precise than alternative methods, such as (COPD) were defined by the American Thoracic Society. 28 cross-validation. FEV1 denotes forced expiratory volume in one second. Finally, in order to determine how much more † Higher scores on the body-mass index, degree of airflow obstruction and dyspnea, and exercise capacity (BODE) precise the BODE index is than the FEV1 alone, we index indicate a greater risk of death. Quartile 1 was de- 29 computed the C statistics for a model containing fined by a score of 0 to 2, quartile 2 by a score of 3 to 4, FEV1 or the BODE score as the sole independent quartile 3 by a score of 5 to 6, and quartile 4 by a score variable. We compared the survival times and esti- of 7 to 10. mated the probabilities of death up to 52 months. In these analyses, the C statistic is a mathematical function of the sensitivity and specificity of the percent died of myocardial infarction, 12 percent BODE index in classifying patients by means of the of lung cancer, and the rest of miscellaneous Cox model as either dying or surviving. The null causes. The BODE score was lower among survi- value for the C statistic is 0.5, with a maximum of vors than among those who died from any cause 1.0 (with higher values being better).29 (3.7±2.2 vs. 5.9±2.6, P<0.005). The score was also lower among survivors than among those who results died of respiratory causes, and the difference between the scores was larger (3.6±2.2 vs. 6.7±2.3, The validation cohort consisted primarily of elderly P<0.001). patients (Tables 3 and 4) with all degrees of severity Table 5 shows the BODE index as a predictor of of COPD. The FEV1 was slightly lower among pa- death from any cause after correction for coexisting tients in the United States than among those in Ven- conditions. There were significantly more deaths ezuela or Spain. The U.S. patients also had more in the United States (32 percent) than in Spain (15 functional impairment, more severe dyspnea, and percent) or Venezuela (13 percent) (P<0.001). How- more coexisting conditions. The 27 patients (4 per- ever, when the analysis was done separately for cent) lost to follow-up were evenly distributed ac- each country, the predictive power of the BODE in- cording to the severity of COPD and did not differ dex was similar; therefore, the data are presented significantly from the rest of the cohort with respect together. Table 5 shows that the BODE index was to any measured characteristic. There were 162 also a predictor of death from respiratory causes deaths (26 percent) over a median follow-up of 28 after correction for coexisting conditions (hazard months (range, 4 to 68). The majority of patients ratio, 1.63; 95 percent confidence interval, 1.48 to (61 percent) died of respiratory insufficiency, 14 1.80; P<0.001). The Kaplan–Meier analysis of sur-

1008 n engl j med 350;10 www.nejm.org march 4, 2004

a multidimensional grading system in chronic obstructive pulmonary disease vival (Fig. 1A) shows that each quartile increase in the BODE score was associated with increased mor- Table 4. Characteristics of the Validation Cohort According to Country.* tality (P<0.001). Thus, the highest quartile (a BODE United score of 7 to 10) was associated with a mortality rate Spain Venezuela States of 80 percent at 52 months. These same data are Characteristic (N=223) (N=54) (N=348) P Value† shown in Figure 1B in relation to the severity of Age (yr) 66±8 64±10 67±9 0.02 COPD according to the staging system of the Amer- Body-mass index 27.5±4.5 23.4±4.7 26.2±4.7 <0.001 ican Thoracic Society. The C statistic for the ability FEV1 (%) 47±17 47±19 39±15 <0.001 of the BODE index to predict the risk of death was MMRC dyspnea scale‡ 1.7±1.2 2.1±1.1 2.7±0.8 <0.001 0.74, as compared with a value of 0.65 with the use Distance walked in 6 min (m) 446±99 225±40 311±121 <0.001 of FEV1 alone (expressed as a percentage of the pre- FEV1 (liters) 1.29±0.52 1.4±0.64 1.20±0.57 0.01 dicted value). The computation of 2000 bootstrap samples for these data and estimation of the haz- FVC (liters) 2.92±0.98 2.8±0.94 2.72±0.8 0.09 ard ratios for death indicated that for each one-point Charlson index§ 2.9±1.3 3.9±1.5 5.3±3.1 <0.001 increment in the BODE score the hazard ratio for BODE index¶ 2.9±2.2 4.9±2.1 5.1±2.4 <0.001 death from any cause was 1.34 (95 percent confidence interval, 1.26 to 1.42) and the hazard ratio * Plus–minus values are means ±SD. † Analysis of variance was used to calculate the P values. for death from a respiratory cause was 1.62 (95 per- ‡ Scores on the modified Medical Research Council (MMRC) dyspnea scale can cent confidence interval, 1.48 to 1.77). range from 0 to 4, with a score of 4 indicating that the patient is too breathless to leave the house or becomes breathless when dressing or undressing. § Scores on the Charlson index can range from 0 to 33, with higher scores indi- discussion cating more coexisting conditions. ¶Scores on the body-mass index, degree of airflow obstruction and dyspnea, We devised a simple grading system for COPD — and exercise capacity (BODE) index can range from 0 to 10, with higher scores indicating a greater risk of death. the BODE index — and validated its use by showing that it is a better predictor of the risk of death from any cause and from respiratory causes than is the FEV1 alone. We believe that the BODE index Table 5. Risk of Death from Any Cause and from Respiratory Failure, is useful because it includes one domain that quan- Pneumonia, or Pulmonary Embolism.* tifies the degree of pulmonary impairment (FEV1), one that captures the patient’s perception of symp- Hazard Ratio Variable (95% CI) P Value toms (the MMRC dyspnea scale), and two independent domains (the distance walked in six minutes Risk of death from all causes and the body-mass index) that express the systemic Model I consequences of COPD. The FEV1 is essential for BODE score 1.34 (1.26–1.42) <0.001 the diagnosis and quantification of the respirato- Model II 1,3,4 ry impairment resulting from COPD. In addi- BODE score 1.32 (1.23–1.40) <0.001 tion, the rate of decline in FEV1 is a good marker Charlson index 1.05 (1.00–1.10) 0.06 of disease progression and mortality.18,19 Howev- Death from respiratory failure, pneumonia, er, the FEV1 does not adequately reflect all the sys- or pulmonary embolism temic manifestations of the disease. For example, Model I the FEV1 correlates weakly with the degree of dys- 20 BODE score 1.62 (1.48–1.77) <0.001 pnea, and the change in FEV1 does not reflect the rate of decline in patients’ health.30 More impor- Model II tant, prospective observational studies of patients BODE score 1.63 (1.48–1.80) <0.001 with COPD have found that the degree of dyspnea8 Charlson index 0.99 (0.93–1.07) 0.97 and health-status scores31 are more accurate predictors of the risk of death than is the FEV1. Thus, * The Cox proportional-hazards models for death from any cause include 162 deaths. The Cox proportional-hazards models for death from specific respira- although the FEV1 is important to obtain and essen- tory causes include 96 deaths. Model I includes the body-mass index, degree tial in the staging of disease in any patient with of airflow obstruction and dyspnea, and exercise capacity (BODE) index alone. COPD, other variables provide useful information The hazard ratio is for each one-point increase in the BODE score. Model II includes coexisting conditions as expressed by each one-point increase in the that can improve the comprehensibility of the eval- Charlson index. CI denotes confidence interval. uation of patients with COPD. Each variable should

n engl j med 350;10 www.nejm.org march 4, 2004 1009

The new england journal of medicine

are simply obtained and independently predict the A risk of death.7,9,10 It is likely that they share some 1.0 common underlying physiological determinants, Quartile 1 but the distance walked in six minutes contains a 0.8 degree of sensitivity not provided by the body-mass Quartile 2 index. The six-minute–walk test is simple to per- 0.6 form and has been standardized.12 Its use as a clin- Quartile 3 ical tool has gained acceptance, since it is a good 0.4 predictor of the risk of death among patients with other chronic diseases, including congestive heart

Probability of Survival Quartile 4 23 24 0.2 failure and pulmonary hypertension. Indeed, the P<0.001 distance walked in six minutes has been accepted 0.0 as a good outcome measure after interventions such 0 4 8 12 16 20 24 28 32 36 40 44 48 52 as pulmonary rehabilitation.32 The body-mass in- Months dex was also an independent predictor of the risk No. at Risk 625 611 574 521 454 322 273 159 80 of death and was therefore included in the BODE index. We evaluated the independent prognostic B power of body-mass index in our cohort using dif- 1.0 ferent thresholds and found that values below 21 were associated with an increased risk of death, an 0.8 observation similar to that reported by Landbo and Stage I coworkers in a large population study.10 Stage II 0.6 The Global Initiative for Chronic Obstructive Stage III Lung Disease and the American Thoracic Society 0.4 recommend that a patient’s perception of dyspnea 1,3

Probability of Survival be included in any new staging system for COPD. 0.2 Dyspnea represents the most disabling symptom P<0.001 of COPD; the degree of dyspnea provides informa- 0.0 0 4 8 12 16 20 24 28 32 36 40 44 48 52 tion regarding the patient’s perception of illness Months and can be measured. The MMRC dyspnea scale is No. at Risk 625 611 574 521 454 322 273 159 80 simple to administer and correlates with other dyspnea scales20 and with scores of health status.21 Figure 1. Kaplan–Meier Survival Curves for the Four Quartiles of the Body-Mass Furthermore, in a large cohort of prospectively fol- Index, Degree of Airflow Obstruction and Dyspnea, and Exercise Capacity Index lowed patients with COPD, which used the thresh- (Panel A) and the Three Stages of Severity of Chronic Obstructive Pulmonary old values included in the BODE index, the score Disease as Defined by the American Thoracic Society (Panel B). on the MMRC dyspnea scale was a better predictor In Panel A, quartile 1 is a score of 0 to 2, quartile 2 is a score of 3 to 4, quartile 3 8 of the risk of death than was the FEV1. a score of 5 to 6, and quartile 4 a score of 7 to 10. Survival differed significantly among the four groups (P<0.001 by the log-rank test). In Panel B, stage I is de- The BODE index combines the four variables by fined by a forced expiratory volume in one second (FEV1) that is more than 50 means of a simple scale. We also explored whether percent of the predicted value, stage II by an FEV1 that is 36 to 50 percent of the weighting the variables included in the index im- predicted value, and stage III by an FEV1 that is no more than 35 percent of the proved the predictive power of the BODE index. In- predicted value. Survival differed significantly among the three groups (P<0.001 terestingly, it failed to do so, most likely because by the log-rank test). each variable included has already proved to be a good predictor of the outcome of COPD. Our study had some limitations. First, relative- correlate independently with the prognosis of ly few women were recruited, even though enroll- COPD, should be easily measurable, and should ment was independent of sex. It probably reflects serve as a surrogate for other potentially important the problem of the underdiagnosis of COPD in variables. women. Second, there were differences among the In the BODE index, we included two descriptors three countries. For example, patients in the United of systemic involvement in COPD: the body-mass States had a higher mortality rate, more severe dys- index and the distance walked in six minutes. Both pnea, more functional limitations, and more co-

1010 n engl j med 350;10 www.nejm.org march 4, 2004

Downloaded from www.nejm.org at UNIV OF MEDICINE AND DENTISTRY OF NJ on February 17, 2007 . Copyright © 2004 Massachusetts Medical Society. All rights reserved. a multidimensional grading system in chronic obstructive pulmonary disease existing conditions than patients in Venezuela or that the FEV1, the degree of dyspnea, and exercise Spain, even though the severity of airflow obstruc- performance provide independent information tion was relatively similar among the patients as a regarding the degree of compromise in patients whole. The reasons for these differences are un- with COPD. known, because there have been no systematic com- Besides its excellent predictive power with re- parisons of the regional manifestations of COPD. gard to outcome, the BODE index is simple to cal- In all three countries, the BODE index was the best culate and requires no special equipment. This predictor of survival, an observation that renders makes it a practical tool of potentially widespread our findings widely applicable. applicability. Although the BODE index is a predic- Three studies have reported the effects of the tor of the risk of death, we do not know whether it grouping of variables to express the various do- will be a useful indicator of the outcome in clinical mains affected by COPD.33-35 These studies did not trials, the degree of utilization of health care re- include variables now known to be important pre- sources, or the clinical response to therapy. dictors of outcome, such as the body-mass index. We are indebted to Dr. Gordon L. Snider, whose guidance, com- However, as we found in our study, they showed ments, and criticisms were fundamental to the final manuscript. references 1. Pauwels RA, Buist AS, Calverley PM, ease. Am J Respir Crit Care Med 1999;160: da A, Koyama H, Izumi T. Comparison of Jenkins CR, Hurd SS. Global strategy for the 1856-61. discriminative properties among disease- diagnosis, management, and prevention of 11. American Thoracic Society Statement. specific questionnaires for measuring health- chronic obstructive pulmonary disease: Lung function testing: selection of reference related quality of life in patients with chronic NHLBI/WHO Global Initiative for Chronic values and interpretative strategies. Am Rev obstructive pulmonary disease. Am J Respir Obstructive Lung Disease (GOLD) Work- Respir Dis 1991;144:1202-18. Crit Care Med 1998;157:785-90. shop summary. Am J Respir Crit Care Med 12. ATS Committee on Proficiency Stan- 22. Szekely LA, Oelberg DA, Wright C, et al. 2001;163:1256-76. dards for Clinical Pulmonary Function Lab- Preoperative predictors of operative mor- 2. Murray CJL, Lopez AD. Mortality by oratories. ATS statement: guidelines for the bidity and mortality in COPD patients under- cause for eight regions of the world: Global six-minute walk test. Am J Respir Crit Care going bilateral lung volume reduction sur- Burden of Disease Study. Lancet 1997;349: Med 2002;166:111-7. gery. Chest 1997;111:550-8. 1269-76. 13. Mahler D, Wells C. Evaluation of clinical 23. Shah M, Hasselblad V, Gheorgiadis M, 3. Definitions, epidemiology, pathophys- methods for rating dyspnea. Chest 1988;93: et al. Prognostic usefulness of the six-min- iology, diagnosis, and staging. Am J Respir 580-6. ute walk in patients with advanced conges- Crit Care Med 1995;152:Suppl:S78-S83. 14. Charlson M, Szatrowski T, Peterson J, tive heart failure secondary to ischemic and 4. Siafakas NM, Vermeire P, Pride NB, et Gold J. Validation of a combined comor- nonischemic cardiomyopathy. Am J Car- al. Optimal assessment and management bidity index. J Clin Epidemiol 1994;47:1245- diol 2001;88:987-93. of chronic obstructive pulmonary disease 51. 24. Miyamoto S, Nagaya N, Satoh T, et al. (COPD). Eur Respir J 1995;8:1398-420. 15. Ferrer M, Alonso J, Morera J, et al. Chron- Clinical correlates and prognostic signifi- 5. Nocturnal Oxygen Therapy Trial Group. ic obstructive pulmonary disease stage and cance of six-minute walk test in patients with Continuous or nocturnal oxygen therapy in health-related quality of life. Ann Intern primary pulmonary hypertension: compari- hypoxemic chronic obstructive pulmonary Med 1997;127:1072-9. son with cardiopulmonary exercise testing. disease: a clinical trial. Ann Intern Med 1980; 16. Dewan NA, Rafique S, Kanwar B, et al. Am J Respir Crit Care Med 2000;161:487-92. 93:391-8. Acute exacerbation of COPD: factors associ- 25. Redelmeier DA, Bayoumi AM, Gold- 6. Intermittent positive pressure breathing ated with poor treatment outcome. Chest stein RS, Guyatt GH. Interpreting small dif- therapy of chronic obstructive pulmonary 2000;117:662-71. ferences in functional status: the Six Minute disease: a clinical trial. Ann Intern Med 1983; 17. Friedman M, Serby CW, Menjoge SS, Walk test in chronic lung disease patients. 99:612-20. Wilson JD, Hilleman DE, Witek TJ Jr. Phar- Am J Respir Crit Care Med 1997;155:1278- 7. Gerardi DA, Lovett L, Benoit-Connors macoeconomic evaluation of a combination 82. ML, Reardon JZ, ZuWallack RL. Variables re- of ipratropium plus albuterol compared with 26. Decramer M, Gosselink R, Troosters T, lated to increased mortality following out- ipratropium alone and albuterol alone in Verschueren M, Evers G. Muscle weakness patient pulmonary rehabilitation. Eur Res- COPD. Chest 1999;115:635-41. is related to utilization of health care resourc- pir J 1996;9:431-5. 18. Anthonisen NR, Wright EC, Hodgkin es in COPD patients. Eur Respir J 1997;10: 8. Nishimura K, Izumi T, Tsukino M, Oga JE. Prognosis in chronic obstructive pulmo- 417-23. T. Dyspnea is a better predictor of 5-year sur- nary disease. Am Rev Respir Dis 1986;133: 27. Cox DR. Regression models and life- vival than airway obstruction in patients with 14-20. tables. J R Stat Soc [B] 1972;34:187-220. COPD. Chest 2002;121:1434-40. 19. Burrows B. Predictors of loss of lung 28. Harrell FE Jr, Lee KL, Mark DB. Multi- 9. Schols AM, Slangen J, Volovics L, Wout- function and mortality in obstructive lung variate prognostic models: issues in devel- ers EF. Weight loss is a reversible factor in diseases. Eur Respir Rev 1991;1:340-5. oping models, evaluating assumptions and the prognosis of chronic obstructive pulmo- 20. Mahler DA, Weinberg DH, Wells CK, adequacy, and measuring and reducing er- nary disease. Am J Respir Crit Care Med Feinstein AR. The measurement of dyspnea: rors. Stat Med 1996;15:361-87. 1998;157:1791-7. contents, interobserver agreement, and phys- 29. Nam B-H, D’Agostino R. Discrimina- 10. Landbo C, Prescott E, Lange P, Vestbo J, iologic correlates of two new clinical index- tion index, the area under the ROC curve. In: Almdal TP. Prognostic value of nutritional es. Chest 1984;85:751-8. Huber-Carol C, Balakrishnan N, Nikulin MS, status in chronic obstructive pulmonary dis- 21. Hajiro T, Nishimura K, Tsukino M, Ike- Mesbah M, eds. Goodness-of-fit tests and

n engl j med 350;10 www.nejm.org march 4, 2004 1011

model validity. Boston: Birkhäuser, 2002: mortality in male patients with chronic ob- sis of dyspnea ratings, respiratory muscle 273-7. structive pulmonary disease. Am J Respir strength, and lung function in patients with 30. Burge PS, Calverley PM, Jones PW, Spen- Crit Care Med 2002;166:680-5. chronic obstructive pulmonary disease. Am cer S, Anderson JA, Maslen TK. Random- 32. Pulmonary rehabilitation — 1999. Am Rev Respir Dis 1992;145:467-70. ised, double blind, placebo controlled study J Respir Crit Care Med 1999;159:1666-82. 35. Wegner RE, Jorres RA, Kirsten DK, Mag- of fluticasone propionate in patients with 33. Ries AL, Kaplan RM, Blumberg E. Use nussen H. Factor analysis of exercise capaci- moderate to severe chronic obstructive pul- of factor analysis to consolidate multiple out- ty, dyspnoea ratings and lung function in pa- monary disease: the ISOLDE trial. BMJ 2000; come measures in chronic obstructive pul- tients with severe COPD. Eur Respir J 1994; 320:1297-303. monary disease. J Clin Epidemiol 1991;44: 7:725-9. 31. Domingo-Salvany A, Lamarca R, Ferrer 497-503. Copyright © 2004 Massachusetts Medical Society. M, et al. Health-related quality of life and 34. Mahler DA, Harver A. A factor analy-

1012 n engl j med 350;10 www.nejm.org march 4, 2004

Downloaded from www.nejm.org at UNIV OF MEDICINE AND DENTISTRY OF NJ on February 17, 2007 . Copyright © 2004 Massachusetts Medical Society. All rights reserved. Factors Associated with Postoperative Pulmonary Complications in Patients with Severe Chronic Obstructive Pulmonary Disease

David H. Wong, PharmD, MD, Erich C. Weber, MD, Michael J. Schell, PhD, Anne B. Wong, MD, Cynthia T. Anderson, MD, and Steven J. Barker, Pm, MD Departments of Anesthesiology and Medicine, University of California at Irvine, and the Long Beach Department of Veteran’s Affairs, Long Beach, California

The purpose of this study was to determine the inci- by analyzing potential preoperative and intraoperative dence of different postoperative pulmonary complica- risk factors. Pulmonary factors alone do not predict the tions (PIG) and their associated risk factors in patients likelihood of PPCs in severe COPD patients. Multiple with severe chronic obstructive pulmonary disease logistic regression identified composite scoring sys- (COPD) (forced expiratory volume in 1 s lFEV,l ~1.2 L tems, such as the ASA physical status, as the best preop- and FEV,/forced vital capacity (F’VC) <75%) undergo- erative predictors of PPCs, probably because they in- ing noncardiothoracic operations. Thirty-nine of 105 clude both pulmonary and nonpulmonary factors. patients (37%) had one or more PPCs (death, pneumo- During the intraoperative period, avoiding general an- nia, prolonged intubation, refractory bronchospasm, or esthesia with tracheal intubation may decrease the risk prolonged intensive care unit (ICU) stay). Thirty-eight of postoperative bronchospasm. Shortening the dura- of 39 patients (97%) with a PPC had an anesthetic duration of surgery and anesthesia may decrease the risk of tion >2 h. Our study patients had a 47% 2-yr mortality prolonged ICU stay. rate. We determined specific risk factors for each PPC (Anesth Analg 1995;80:276-84)

ostoperative pulmonary complications (PIG) including all the different kinds of PPCs together have been reported to occur in 5%-10% of the effectively treats them all as equivalent, and they P general patient population (1,2) and in 4%-22% are not. Atelectasis, for example, is less clinically sig- of patients undergoing abdominal surgery (3-5). PPCs nificant than respiratory failure and pneumonia may be associated with significant mortality (6). Many which, in turn, are less clinically significant than authors have studied the incidence of PPCs in both death. Each PPC should be analyzed individually for thoracic and nonthoracic surgery patients (1,2,4-S). risk factors. They have shown that PPCs have multiple risk factors, One problem with this kind of analysis is sample such as anesthesia duration, incision site, anesthesia size. For example, the number of postoperative pneu- type, and chronic obstructive pulmonary disease monias will be smaller than total number of all the (COPD). However, different authors have defined different kinds of PK. Therefore, determining the risk PPCs differently. Some have defined PPCs as atelec- factors for pneumonia requires a larger sample size than determining the risk factors for a group of PPCs tasis and pneumonia (l-3,9); others have included all that includes atelectasis and pneumonia. A larger PPCs together into one group for statistical analysis, sample size is also required when studying risk factors sometimes defining pleural effusion (lo), pulmonary for the more clinically significant PPC, since the more embolism (lo), cough (ll), or aspiration pneumonia severe PPCs are less common than atelectasis. For (12) as a PK. However, we believe that all PPCs example, only eight instances of respiratory failure should not be analyzed together as one group, because occurred in a study of anesthetic complications in 13,693 routine surgery patients (13). Presented in part at the International Anesthesia Research Society COPD patients have been found to be at higher risk Meeting, March 1987. for postoperative atelectasis or pneumonia (2,7,14), Accepted for publication September 9, 1994. and death (6,15-17). Therefore, perhaps the risk fac- Address correspondence and reprint requests to David H. Wong, PharmD, MD, Long Beach VAMC, 5901 E. 7th Street (139), Long tors for the more severe PPCs may be determined by Beach, CA 90822. studying a relatively small group of COPD patients.

01995 by the International Anesthesia Research Society 276 An&h Analg 1995;80:276-84 0003-2999/95/$5.00 ANESTH ANALG CRITICAL CARE AND TRAUMA WONG ET AL. 277 1995;80:276-84 PULMONARY COMPLICATIONS IN SEVERE COPD

The purpose of this study was to determine the wheezing upon auscultation that acutely required ad- incidence of different PPC in patients with severe ditional parenteral drug administration (intravenous COPD. We also wanted to determine the specific risk or subcutaneous injection; not inhaled) in addition to factors for each kind of PK. the patient’s preoperative drug regimen. Prolonged ICU stay has been proposed to be a more objective indicator of perioperative complications than other Methods PPCs (21) and was defined as an ICU stay 24 days. Atelectasis (i.e., isolated fever or abnormal chest ra- In compliance with institutional review board guide- diograph without a positive sputum culture) was not lines and without written patient informed consent, considered a PK. demographic and clinical data were collected on every Patients who received combined regional and gen- patient with a clinical diagnosis of COPD, a forced eral anesthesia were included in the general anesthe- expiratory volume in 1 s (FEV,) ~1.2 L and a sia group. Similarly, patients with lower abdominal FEV,/forced vital capacity (FVC) ratio ~75% who and upper abdominal incisions were combined into underwent noncardiothoracic surgery at the Long one group, defined as patients with abdominal inci- Beach Veteran’s Administration Medical Center from sions. Only patients who had received general anes- April 1986 to May 1990. A FEV, threshold of 1.2 L was thesia were analyzed for the risk of prolonged post- selected because it has been reported previously that operative intubation. these patients are at higher risk for PPC (s). The pre- Statistical analysis was done in three phases. First, dicted FEV, for all our patients was 2.50 L or larger; univariate analysis regarding the PPC incidence for thus all our patients had FEV, ~50% of their calcu- each potential risk factor was evaluated by Fisher’s lated predicted normal value (181, which is one indi- exact test. Optimal cutoff points for univariate analy- cator of severe pulmonary impairment (19). Patients sis were determined using reduced monotonic regres- with preexisting tracheostomies and thoracic surgery sion analysis (M. J. Schell, B. Singh, personal commu- patients were excluded. If a patient had more than one nication, 1994). We considered a P value co.05 as operation during the study, only data from the first statistically significant. All potential risk factors were operation was included. also analyzed by Spearman rank (nonparametric) cor- Potential preoperative risk factors obtained were: relation coefficient analysis to determine whether the age, smoking history (pack-years), sputum production potential risk factors were truly independent vari- and color, ASA physical status grade, a postoperative ables. Most of the variables were not highly correlated. pulmonary complication risk score as described by For example, Shapiro score and ASA physical status Shapiro (Shapiro score, see Appendix) (20), FEV,, only had a correlation coefficient (r) of 0.22 with each FEV,/FVC, and arterial blood gas (ABG). The anes- other. Only three variables had a r >0.50 with another thesia care team assigned the patient’s ASA physical variable: ASA physical status and emergency opera- status grade. A research team member determined tion (r = 0.56), prebronchodilator FEV, and postbron- and recorded the patient’s Shapiro score. FEV, and chodilator FEV, (r = 0.62), and duration of anesthesia FVC were measured with a Vitalographm spirometer and duration of surgery (r = 0.93). Since duration of (Vitalograph, Buckingham, England). Potential intra- anesthesia and duration of surgery were highly corre- operative risk factors recorded were: type of anesthe- lated, duration of anesthesia was included in the lo- sia, duration of anesthesia, duration of surgery, and gistic regression model and duration of surgery was incision site. Each patient’s attending anesthesiologist excluded. selected the anesthetic technique. We defined an op- Next, the univariate risk factors with P < 0.25 were eration as emergent if the patient had a suffix “E” then evaluated by forward stepwise multiple logistic assigned to their ASA physical status grade by the regression using SAS computer programs (SAS Insti- attending anesthesiologist. The occurrence of five dif- tute, Cary, NC). A P < 0.05 was required to stay in the ferent postoperative complications was specifically logistic regression model. Age, smoking (pack-years), sought: death, pneumonia, prolonged intubation, re- duration of anesthesia, FEV,, and FEV,/FVC were fractory bronchospasm, and prolonged intensive care examined both as continuous variables and catego- unit (ICU) stay. Death as a postoperative complication rized by cutoff points as in Table 2. Optimal cutoff was defined as death occurring during the same hos- points were determined using reduced monotonic re- pital stay. Postoperative pneumonia was defined as a gression analysis (M. J. Schell, B. Singh, personal com- new infiltrate on a chest radiograph combined with munication, 1994). No optimal cutoff points for age, fever, leukocytosis, and a positive sputum Gram stain smoking, or FEV,/FVC could be determined so the or culture. Prolonged intubation was defined as the cutoff points for these variables listed in Table 2 were failure to extubate within 24 h after the end of the selected arbitrarily. Anesthesia duration >2 h was a operation. Refractory bronchospasm was defined as unique cutoff point; almost all the PPCs occurred in 278 CRITICAL CARE AND TRAUMA WONG ET AL. ANESTH ANALG PULMONARY COMPLICATIONS IN SEVERE COPD 1995;80:276-84

patients with anesthesia duration >2 h. This caused factors for different individual PPC by multivariate the effect of anesthesia duration >2 h to overwhelm analysis. Emergency operation, abdominal incision, the effect of the other single-risk factors in the regres- anesthesia duration, and general anesthesia were the sion models. Therefore anesthesia duration >2 h was intraoperative risk factors which were significant accepted as a risk factor for PPC and excluded as a independent risk factors for different individual PPCs cutoff point in the logistic regression models. It may by multivariate analysis. The significant independent be unfair to compare composite classification systems risk factors for each PPC and their odds ratios are such as ASA physical status and Shapiro score with listed in Table 3. individual risk factors because several individual risk When the composite classification systems (ASA factors may be incorporated into one classification physical status and Shapiro score) were excluded from system. Therefore, the logistic regression models were the multiple logistic regression models, FEV, became computed two ways: with and without the composite the only significant preoperative risk factor for differ- classification systems as risk factors. Lastly, to evalu- ent individual PPCs. Emergency operation, anesthesia ate the effect of second order interactions, another duration, and general anesthesia remained significant stepwise logistic regression analysis was computed intraoperative risk factors. The significant indepen- which included second order interaction terms for all dent risk factors for PPCs and their odds ratios, when dichotomous (two-valued) risk factors. the composite classification systems are excluded Long-term followup for determining survival was from the models, are listed in Table 4. No statistically accomplished by examining hospital admission and significant second order interactions were found in discharge records, clinic visit records, personal tele- any of the multiple logistic regression models. phone contact, and Social Security Administration of- The long-term survival of patients with severe fice registration. The long-term survival curve was COPD undergoing noncardiothoracic surgery was created using Epistat 4.0 (Epistat Services, Richardson, poor. Figure 1 shows the study group’s long-term TX). survival curve. At least 2 yr of followup was available on all 105 patients in the study. The 2-yr mortality rate was 47%. Results One hundred five patients were studied; 53 had gen- Discussion eral anesthesia (38 had general anesthesia alone and 15 had combined [general and epidurall anesthesia), 51 This study shows that patients with severe COPD had regional anesthesia (38 spinal, 13 epidural), and 1 undergoing surgery and anesthesia have a high inci- had local anesthesia. All but four were male; other dence of PPCs and poor long-term survival. Pulmo- patient characteristics are summarized in Table 1. In- nary risk factors alone do not predict the risk of PPCs. formation about all preoperative and intraoperative A high numerical value on a composite classification risk factors studied, except for sputum quality, ABG system was the most common preoperative risk factor data, and postbronchodilator lung spirometry data, for any individual PK. Including composite classifi- were available in all patients. Room air ABG analyses cation systems in the multiple logistic regression mod- were available in 77 patients. Of 60 patients who had els improves the ability of the models to predict PPCs. spirometry before and after bronchodilator therapy, Composite classification systems are probably more 35 increased their FEV, ~15% after bronchodilator useful than single-risk factors in predicting PPCs be- therapy. cause nonpulmonary variables are also important. Thirty-nine patients (37%) had one or more PPCs. It is not surprising that patients with severe COPD Refractory bronchospasm (17 patients) and prolonged have a high incidence of PPCs; they have underlying ICU stay (24 patients) were the most common PPCs. pulmonary disease. In addition, both surgery and an- Seven patients died during their hospitalization, seven esthesia can adversely affect lung function (9,23). In patients developed pneumonia, and eight patients re- general, the risk factors associated with PPCs in severe mained intubated postoperatively. COPD patients are some of the same factors that have Emergency operation was the univariate risk factor been also been associated with postoperative atelecta- most frequently associated with a PK. The incidence sis, i.e., ASA physical status (4,5), duration of surgery of all the potentially significant univariate risk factors (41, duration of anesthesia (71, and abdominal incision for predicting individual PPCs is summarized in Table (4). Since higher ASA physical status has also been 2. All other potential risk factors failed to show any associated with postoperative pneumonia (41, pro- relationship with any PPC by univariate analysis. longed postoperative intubation (131, and higher ASA physical status ?IV, Shapiro score 2 5, and mortality (24,25), it is not surprising that high ASA FEV, were the only significant preoperative risk physical status would be associated with PPCs. ANESTH ANALG CRITICAL CARE AND TRAUMA WONG ET AL. 279 1995;80:27684 PULMONARY COMPLICATIONS IN SEVERE COl’D

Table 1. Patient Characteristics Characteristics n Mean % SD (range) Age (yr) 68.0 ? 8.9 (51-93) Anesthesia duration (h) 3.4 + 2.0 (0.75-10.2) Operation duration (h) 2.3 + 1.7 (0.5-9.0) Pulmonarv Smoking (pack-years) 69.5 + 33.9 (O-190) <25 5 25-50 31 50-75 >75 zJ Prebronchodilator FEV, (L) 0.85 + 0.20 (0.4-1.2) 51.20 105 neck incisions 18 Major cancer surgery 6 Panendoscopy 4 All other head and neck surgery 8 Coexisting disease Cardiac dysrhythmia 18 Previous myocardial infarction 15 Congestive heart failure or cardiomyopathy 10 Peripheral vascular disease 14 Cerebral vascular accident 2 Hypertension 25 Diabetes 2 Cancer 29

FEV, = forced expiratory volume in 1 s; FVC = forced vital capacity. 280 CRITICAL CARE AND TRAUMA WONG ET AL. ANESTH ANALG PULMONARY COMPLICATIONS IN SEVERE COI’D 1995;80:276-84

Table 2. Postoperative Pulmonary Complications (PITS) by Individual Risk Factors Prolonged Prolonged Death Pneumonia intubation Bronchospasm ICU stay Any PIT Risk factor (n, n-GA)” (n = 7) (n = 7) (n = 8) (n = 17Y (n = 24) (n = 39) FEV, 10.75 L (36,20) 7 (19%)§ 2 (6%) 5 (25%Y 9 (25%)* 15 (42%)§ 20 (56%)$ ~0.60 L (15,8) 2 (13%) 1(7%) 2 (13%) 2 (13%) 9 (60%)§ 10 (67%)+ FEV,/FVC < 0.5 (73,391 5 (7%) 6 (8%) 5 (13%) 11 (15%) 17 (23%) 24 (33%)” General anesthesia (53,531 3 (6%) 5 (9%) 8 (15%) 13 (25%H 17 (32%)t 29 (55%& Abdominal incision 5 (14%)t 4 (ll%y+ 6 (29%)t 9 (25%)* 19 (53%)§ 25 (69%& (36,211 Smoking (pack-years) 2 (5%)* 4 (10%) 5 (27%)* 8 (20%) 9 (23%) 17 (43%) >75 (40,19) ASA physical status zIV 5 (18%)$ 5 (18%)$ 6 (43%)§ 7 (25%)* 18 (64%)§ 21 (75%)§ (28,14 Age 265 (62,241 4 (6%) 4 (6%) 3 (13%) 9 (15%) 15 (24%) 21 (34%) Emergency operation 4 (31 %o)§ 4 (31 %)gj 4 (50%)§ 4 (31%) 8 (61%)§ 10 (77%)$ (13,8) Shapiro score 24 (37,21) 5 (14%)” 2 (5%) 4 (19%) 9 (24%)* 14 (38%)$ 19 (51%H 25 (11,5) 4 (36%)§ 1(9%‘0) 1(20%) 6 (55%& 5 (46%)* 8 (73%)$ Anesthesia duration >2 h (73,46) 6 (8%) 7 (10%) 8 (17%)* 17 (23%)$ 23 (32%)$ 38 (52%)§ 24 h (33,26) 4 (12%) 3 (9%) 5 (19%) 9 (27%H 12 (36%)t 20 (61%)§ ~6.5 h (8,7) 0 (0%) 1(12%) 1(14%) 2 (25%) 7 (SS%)§ 7 (SS%)$

Q n = the number of patients with risk factor; n-GA = the number of patients with factor who underwent general anesthesia (GA) except for prolonged intubation where (%) = % patients with listed risk factor and GA who had that PK. b % incidence of prolonged intubation was analyzed only in those patients who underwent GA. * 0.05 < P 5 0.25; t 0.01 < P c 0.05; $0.001 < P 5 0.01; 5 P 5 0.001.

A high numerical value on a composite classifica- factor particularly improved the logistic regression tion system, either a ASA physical status zIV or a model for bronchospasm. Why is a Shapiro score 25, Shapiro score ~5, was generally the most common which includes spirometry measurements, more use- and strongest risk factor for PPCs. Besides anesthesia ful than FEV, values in predicting death and broncho- duration >2 h, the single-risk factors that were asso- spasm? A Shapiro score ~5 signifies that at least two ciated with individual PPC were mostly nonpulmo- organ systems are diseased: the pulmonary system nary, intraoperative risk factors (emergency surgery, and one other. Thus, a Shapiro score 25 becomes more abdominal incision, and general anesthesia). Pulmo- like an ASA physical status ?IV, i.e., an indicator of nary risk factors (i.e., smoking history, FEV,, or severe systemic disease. Of the preoperative risk fac- FEV,/FVC) were less common and less significant tors that were predictive of PPCs, the Shapiro score is risk factors. This implies that nonpulmonary risk fac- the hardest to determine, spirometry is easier, and the tors were important in predicting the risk of PK. ASA physical status is the easiest to determine. The Nonpulmonary risk factors may have been important ASA physical status classification system also pro- in our study because our study group had a muder- vides for emergency surgery, which was the best sin- ately high incidence of nonpulmonary disease (Table gle risk factor for PPC in this study. Composite clas- 1). The presence of both pulmonary and nonpulmo- sification systems such as the ASA physical status nary disease in our patients may have favored select- should be included in the logistic regression models ing composite classification systems as risk factors for PPCs. because classification systems can include both pul- The most serious PPC is death. Besides same-hos- monary and nonpulmonary risk factors. pitalization mortality, we looked at long-term mortal- Composite classification systems enhanced the lo- ity. The long-term mortality of COPD patients under- gistic regression models, as evidenced by the 2 values going noncardiothoracic surgery is very high. Patients associated with the models. ,$ values were generally with severe COPD undergoing nonthoracic surgery higher when the composite classification systems were have long-term mortality rates comparable to similar included in the model than when they were not (Ta- patients with severe coronary artery disease who un- bles 3 and 4). Including Shapiro score 25 as a risk dergo noncardiac surgery (26). Our mortality rates are ANESTH ANALG CRITICAL CARE AND TRAUMA WONG ET AL. 281 1995;80:276-84 PULMONARY COMPLICATIONS IN SEVERE COPD

Table 3. Independent Risk Factors for Postoperative Pulmonary Complications (PITS) by Multivariate Analysis: ASA Physical Status and Shapiro Score Included in the Model PIT (n; model 2 value) 95% Confidence independent risk factors Odds ratio interval P value Death (7; 32.7) Shapiro score ~5 17 2.5-114.5 0.0001 Emergency operation 13 1.9-86.6 0.0034 FEV, 10.75 L co 0.0179 Pneumonia (7; 13.9) Emergency operation 13 2.5-68.5 0.0002 Prolonged intubation (8; 11 .4Jb ASA physical status zIV 14 2.4-81.8 0.0007 Bronchospasm (17; 19.4) Shapiro score ~5 14 2.9-65.9 0.0003 General anesthesia 5.7 1.4-22.9 0.0082 Prolonged ICU stay (24; 63.8) ASA physical status >IV 153 12.0-1976.1 0.0001 Anesthesia duration >6.5 h 564 10.4-30,502 0.0001 Abdominal incision 14 2.1-102.1 0.0004 General anesthesia 9.5 1.3-70.0 0.0272 FEV, 10.60 L 21 1.1406.6 0.0291 Any PIT (39; 48.6) Abdominal incision 18 1.2-97.1 0.0001 Anesthesia duration (per h) 1.78 1.14-2.71 0.0001 ASA physical status zIV 41 5.8-292.4 0.0002 General anesthesia 20 3.1-127.0 0.0008 FEV,/FVC 10.5 16 2.7-88.4 0.0020 Shapiro score 25 10 1.1-89.4 0.0345

FEV, = forced expiratory volume in 1 s; WC = forced vital capacity; ICU = intensive care unit. a The model 2 value, which is based on the score statistic, represents how well the logistic regression model explains the data; the higher the 2 value, the better the model. b Analyzed only in the 53 patients who had general anesthesia. c For the continuous variable “anesthesia duration,” the odds ratio represents the risk per hour. To obtain the odds ratio of an at-risk individual to a selected baseline, raise the odds ratio given to the difference in time between the at-risk individual and the baseline. For example, the odds ratio that a patient will have a prolonged ICU stay given a 5-h anesthetic compared to a Z-h anesthetic is (1.78j3 = 5.64, where 1.78 is the odds ratio listed and the exponent 3 is the difference in anesthesia duration, in hours.

higher than those reported by other authors who per- pulmonary toilet or postoperative ICU observation), formed longitudinal mortality studies of severe COPD although this study did not specifically evaluate the patients who did not undergo surgery at the time of benefit of these interventions. Regarding the intraop- study entry (27-29). The added mortality in our erative period, our results suggest that avoiding gen- patients may be due to the risk of surgery and anes- eral anesthesia should decrease the risk of broncho- thesia, or due to the underlying disease that made the spasm and that shortening the duration of surgery operation necessary. (and anesthesia) and avoiding abdominal incision Knowing the incidence of WCs, the risk factors for may decrease the risk of prolonged ICU stay in these PPCs, and the long-term survival of patients with patients. General anesthesia has been reported previ- severe COPD is useful for three reasons. First, al- ously to be a risk factor for bronchospasm (30), prob- though there may be no spirometry value or specific ably because endotracheal tubes can directly irritate risk factor which absolutely contraindicates surgery airways. Anesthesia duration >2 h is a risk factor for and anesthesia, knowing the incidence of WCs and PPCs because only patients having simpler surgeries long-term mortality allows the clinician to better (n = 32) had anesthesia duration <2 h. The duration weigh the risk versus benefit of a surgical procedure of surgery (and anesthesia) probably depends on in any given patient preoperatively. Second, knowing the underlying surgical disease, and the true risk the risk factors for PPC helps identify which patients factor for PPCs may be severe surgical disease. are at risk for severe WCs so that their management Unfortunately, avoiding general anesthesia with can be individualized. Patients who are identified to tracheal intubation, shortening the duration of sur- be at higher risk for severe PPCs might deserve extra gery (and anesthesia), and avoiding abdominal preoperative or postoperative care (e.g., preoperative incision may not always be possible. Last, knowing 282 CRITICAL CARE AND TRAUMA WONG ET AL. ANESTH ANALG PULMONARY COMPLICATIONS IN SEVERE COPD 1995;80:276-84

Table 4. Independent Risk Factors for Postoperative Pulmonary Complications @‘PCs) by Multivariate Analysis: ASA Physical Status and Shapiro Score Excluded from the Model PIT (n; model 2 value)” 95% Confidence independent risk factors Odds ratio interval P value Death (7; 14.4) FEV, so.75 L co 0.0001 Pneumonia (7; 13.9) Emergency operation 13 2.5-68.5 0.0002 Prolonged intubation (8; 11 .l)b Emergency operation 14 2.3-86.0 0.0009 Bronchospasm (17; 6.0) Duration of anesthesia (per h) 1.33 1.05-1.71 0.0142 Prolonged ICU stay (24; 46.2) Abdominal incision 21 5.0-87.8 0.0001 Anesthesia duration >6.5 h 43 3.2-591.5 0.0002 FEV, 10.75 L 5.5 5.0-20.2 0.0063 Any PPC (39; 48.6) Abdominal incision 21 5.0-87.6 0.0001 Anesthesia duration (per h) 1.60 1.13-2.24 0.0001 General anesthesia 5.9 1.6-21.4 0.0085 FEV, SO.75 L 4.7 1.4-16.3 0.0099 FEV, /Fvc so.5 7.6 1.9-31.3 0.0150

FEV, = forced expiratory volume in 1 s; FVC = forced vital capacity; ICU = intensive care unit. n The model ,$ value, which is based on the score statistic, represents how well the logistic regression model explains the data; the higher the 2 value, the better the model. b Analyzed only in the 53 patients who had general anesthesia. ’ For the continuous variable “anesthesia duration,” the odds ratio represents the risk per hour. To obtain the odds ratio of an at-risk individual to a selected baseline, raise the odds ratio given to the difference in time between the at-risk individual and the baseline. For example, the odds ratio that a patient will have any PPC stay given a 5-h anesthetic compared to a 2-h anesthetic is (1.60j3 = 4.10, where 1.60 is the odds ratio listed and the exponent 3 is the difference in anksthesia d&&ion, in hours.

Expected outcome comparisons look at more than the overall outcome (e.g., overall PPC rates). Expected outcome comparisons also account for expected risk. For example, the relatively high rate of PPCs in our study could be interpreted two ways: either the patients were at a relatively high risk for PPCs or they received relatively low quality perioperative medical care. We believe that our patients received good perioperative medical care. The way to determine whether our patients received good or poor quality health care would be to study several institutions with comparable risk patients and compare our incidence of PPCs with the other institutions’ incidence of PPCs. Lower quality health care would be 10- associated with worse outcomes than expected (an incidence of PPCs greater than the other institutions’ 0 , ,,,,,,,),,,,, , , ,,,I incidence of PITS). Years 0 ’ 1 2 3 4 ' 5 6 Before the risk factors reported here can be used to y;y'k" 105 77 56 34 21 12 4 predict the risk of PITS in other patients at other Figure 1. Cumulative % survival of chronic obstructive pulmonary institutions, several potential problems with this study disease patients over time. Upper numbers along the x axis refer to should be considered: sample bias, assumptions inher- years of followup; lower numbers along the x axis refer to the number of patients at risk at each time-point. The mortality rate at ent with our logistic regression model, small sample 1 yr was 27%; 2 yr, 47%; 3 yr, 64%; 4 yr, 70%; 5 yr, 72%; and 6 yr, size, and a nonrandomized study design. 84%. Our logistic regression model has several inherent problems. First, our identification of PPC risk factors is the incidence of PPCs in severe COPD patients pro- limited to the risk factors that we studied. In reality, vides some basis for observed versus expected out- coexisting diseases that we did not examine may have come comparisons. contributed to PCs, particularly death and prolonged ANESTH ANALG CRITICAL CARE AND TRAUMA WONG ET AL. 283 1995;80:276-84 PULMONARY COMPLICATIONS IN SEVERE COPD

ICU stay. Another problem with the logistic regres- decrease the risk of postoperative bronchospasm. sion model is that it sequentially selects factors one at Shortening the duration of surgery and anesthesia a time. The logistic regression model includes the most may decrease the risk of prolonged ICU stay. significant factor first and then discounts (excludes) the lesser significant factors. This could exclude a Appendix. Classification of Risk of Pulmonary variable that was really a significant risk factor. For Complications of Thoracic and Abdominal Procedures example, both ASA physical status ?IV and a Shapiro score ~5 indicate severe systemic medical disease Category Points” which could increase the risk of death. Selection of the I. Expiratory spirogram most significant scoring system first (i.e., Shapiro A. Normal (%EVC + %FEV,/FVC >150) 0 B. %FVC + %FEV,/FVC = loo-150 1 score ~5) in the logistic regression model excluded C. %FVC + %FEV,/FVC 50 mm Hg or Pao, <60 mm Hg 1 stay. The small sample size results in large confidence on room air C. Metabolic pH abnormality; >7.50 or <7.30 1 intervals for our odds ratios. A larger sample size V. Postoperative ambulation would have resulted in smaller confidence intervals A. Expected ambulation (minimum, sitting at 0 for our odds ratios and also might have allowed us to bedside) within 36 h evaluate more factors simultaneously. However, a B. Expected complete bed confinement for at 1 larger sample would have required a study period least 36 h longer than 4 yr or a larger, multicenter trial. Despite a A maximum total score of 7 points is possible. (Reprinted with permission from Shapiro BA, et al. Clinical Application of Respiratory Care. 3rd ed. our relatively small sample size, we noted almost as Chicago: Yearbook Publishers, 1985.) many serious WCs as larger studies with a wider spectrum of pulmonary disease: 8 instances of respiratory failure and 17 instances of bronchospasm were The authors would like to thank Ms. Diana Lugo for excellent noted in 13,696 patients in a study by Forrest et al. (13) secretarial support. and 16 instances of pneumonia, using our definition of pneumonia, were noted in 1000 patients in a study by Hall et al. (5). References In conclusion, patients with FEV, 51.2 L undergo- 1. King DS. Post-operative pulmonary complications. Surg Gy- ing noncardiothoracic surgery have a 37% incidence of necol Obstet 1933;56:43-50. WCs, excluding atelectasis, and have a 47% 2-yr mor- 2. Wightman JAK. A prospective survey of the incidence of postoperative pulmonary complications. Br J Surg 1968;55:85-91. tality rate. However, preoperative pulmonary factors 3. Dripps RD, Deming MV. Post-operative atelectasis and pneu- alone do not predict the likelihood of WCs in severe monia. Ann Surg 1946;124:94-110. COPD patients. Composite classification systems, 4. Garibaldi RA, Britt MR, Coleman ML, et al. Risk factors for post-operative pneumonia. Am J Med 1981;70:677-80. such as the ASA physical status, probably predict the 5. Hall IC. Talala RA, Hall IL. Mander 1. A multivariate analvsis of likelihood of PPCs better than individual risk factors the risk of pulmonary complications after laparotomy. ‘Chest because they include both pulmonary and nonpulmo- 1991;99:923-7. 6. Tarhan S, Moffitt EA, Sessler AD, et al. Risk of anesthesia and nary factors. During the intraoperative period, avoid- surgery in patients with chronic bronchitis and chronic obstruc- ing general anesthesia with tracheal intubation may tive pulmonary disease. Surgery 1973;74:720-6. 284 CRITICAL CARE AND TRAUMA WONG ET AL. ANESTH ANALG PULMONARY COMPLICATIONS IN SEVERE COPD 1995;80:276&34

7. Latimer RG, Dickman M, Day WC, et al. Ventilatory patterns 19. Gold WM, Boushey HA. Pulmonary function testing. In: Mur- and pulmonary complications after upper abdominal surgery ray JF, Nadel JA, eds. Textbook of respiratory medicine. determined by pre-operative and post-operative computerized Philadelphia: WB Saunders, 1988:667-70. spirometry and-blood gas analysis. Am J Surg 1971;122:622-32. 20. Shapiro BA, Harrison RA, Kacmarek RM, Cane RD. Clinical 8. Lockwood I’. The urinciules of uredictinz the risk of oost- application of respiratory care. Chicago: Year Book Medical thoracotomy function-related comblicationg in bronchial barci- Publishers, 1985:518. noma. Respiration 1973;30:329-44. 21. Zibrak JD, O’Donnell CR, Marton K. Indications for pulmonary 9. Tisi GM. Pre-operative evaluation of pulmonary function. Am function testing. Ann Intern Med 1990;112:763-71. Rev Respir Dis 1979;119:293-310. 22. Deleted in proof. 10. Gold MI, Schwam SJ, Goldberg M. Chronic obstructive pulmo- 23. Craig DB. Postoperative recovery of pulmonary function. nary disease and respiratory complications. Anesth Analg 1983; Anesth Analg 1981;60:46-52. 62:975-81. 11. Kroenke K, Lawrence VA, Theroux JF, Tuley MR. Operative risk 24. Vacanti CJ, Van Houten RJ, Hill RC. A statistical analysis of the in patients with severe obstructive pulmonary disease. Arch relationship of physical status to postoperative mortality in Intern Med 1992;152:967-71. 68,388 cases. Anesth Analg 1970;49:564-6. 12. Pedersen T, Eliasen K, Henriksen E. A prospective study of 25. Marx GH, Matte0 CV, Orkin LR. Computer analysis of post mortality associated with anaesthesia and surgery: risk indica- anesthetic deaths. Anesthesiology 1973;39:54-8. tors of mortality in hospital. Acta Anaesthesiol Stand 1990;34: 26. Browner WS, Li J, Mangano DT. In hospital and long-term 176-82. mortality in male veterans following noncardiac surgery. JAMA 13. Forrest JB, Rehder K, Cahalan MK, Goldsmith CH. Multicenter 1992;268:228-32. Study of General Anesthesia, Anesthesiology 1992;76:3-15. 27. Burrows B, Earle RH. Course and prognosis of chronic obstruc- 14. Stein M, Koota GM, Simon M, Frank HA. Pulmonary evaluation tive lung disease. A prospective study of 200 patients. N Engl J of surgical patients. JAMA 1962;181:765-70. ’ Med 1969;280:397-404. 15. Gracev DR, Divertie MB, Didier ED. Pre-ouerative nulmonarv 28. Postma DS, Burema J, Gimeno F, et al. Prognosis in severe preparation of patients with chronic obstruc’tive pulmonary dis- chronic obstructive pulmonary disease. Am Rev Respir Dis ease, a prospective study. Chest 1979;76:123-9. 1979;119:357-67. 16. Stein M, Cassara EL. Preoperative pulmonary evaluation and 29. Renzetti AD, McClement JH, Litt BD. The Veterans Adminis- therapy for surgery patients. JAMA 1970;211:787-90. tration cooperative study of pulmonary function. Mortality in 17. Grover EL, Hammermeister KE, Burchfiel C. Initial report of the relation to respiratory function in chronic obstructive pulmo- Veterans Administration preoperative risk assessment study for cardiac surgery. Ann Thorac Surg 1990;50:12-28. nary disease. Am J Med 1966;41:115-29. 18. Manfreda J, Nelson N, Chemiak RM. Prevalence of respiratory 30. Olsson GL. Bronchospasm during anaesthesia. A computer- abnormalities in a rural and an urban community. Am Rev aided incidence study of 136,929 patients. Acta Anaesthesiol Respir Dis 1978;117:215-6. Stand 1987;31:244-52.

The new england journal of medicine

review article

current concepts Management of Chronic Obstructive Pulmonary Disease

E. Rand Sutherland, M.D., M.P.H., and Reuben M. Cherniack, M.D.

hronic obstructive pulmonary disease (copd) is a syndrome of From the Department of Medicine, Nation- progressive airflow limitation caused by chronic inflammation of the air- al Jewish Medical and Research Center; and c 1 the Division of Pulmonary Sciences and ways and lung parenchyma. The primary physiological abnormality in COPD is Critical Care Medicine, Department of Med- an accelerated decline in the forced expiratory volume in one second (FEV1) from the icine, University of Colorado Health Sci- normal rate in adults over 30 years of age of approximately 30 ml per year to nearly 60 ences Center — both in Denver. Address 2 reprint requests to Dr. Cherniack at 1400 ml per year. As shown in Figure 1, the disease course begins with an asymptomatic Jackson St., J-208, Denver, CO 80206. phase in which lung function deteriorates without associated symptoms. The onset of the subsequent symptomatic phase is variable but often does not occur until the N Engl J Med 2004;350:2689-97. 3 Copyright © 2004 Massachusetts Medical Society. FEV1 has fallen to approximately 50 percent of the predicted normal value. Since substantial deterioration in airflow has already occurred by the time most patients present with symptoms, it is reasonable to conclude that the degree of airflow limitation is only one of many factors that govern the onset of symptoms. Hyperinflation, which occurs at rest and worsens with exercise (Fig. 2), is an additional physiological abnormality that is commonly seen in patients with moderate- to-severe COPD. It is manifested primarily by an increase in the functional residual capacity, which places the muscles of respiration at a mechanical disadvantage, thereby increasing the work of breathing and reducing exercise tolerance. Additional physiological abnormalities include a reduction in the diffusing capacity for carbon monoxide, hypoxemia, and alveolar hypoventilation.

diagnosis, staging, and prognosis

Because the majority of cases occur in patients who have smoked,4 all current or former smokers should be considered at increased risk for COPD. Other risk factors, which 5 account for far fewer cases, include a1-antitrypsin deficiency, airway hyperresponsiveness,6 and indoor air pollution.7 Since symptoms may not occur until lung function is substantially reduced, early detection is enhanced by spirometric evaluation of FEV1 and forced vital capacity (FVC). Guidelines from the Global Initiative for Chronic Obstructive Lung Disease (GOLD) state that the airflow limitation in COPD is characterized by an FEV1 value that is less than 80 percent of the predicted normal value and 3 an FEV1:FVC ratio of less than 0.70. Currently, most guidelines recommend that practitioners use a combination of information about symptoms and evidence of impairment of physiological function in determining the severity of the disease,3,8-10 although the guidelines differ somewhat with regard to setting thresholds for mild, moderate, and severe disease (Table 1). The stage of the disease suggests the prognosis, and follow-up data from longitudinal studies indicate that moderate and severe stages of the disease are associated with higher mortality.11 However, in the largely asymptomatic group of patients that GOLD3 cate- gorizes as “stage 0, at risk,” only 18.5 percent of the patients progress to more severe

n engl j med 350;26 www.nejm.org june 24, 2004 2689

The new england journal of medicine

airflow limitation at 15 years,12 which suggests (including both the indolent and progressive na- that more information is required to predict which ture of symptoms), and a knowledge of the distri- patients with incipient disease will progress rapidly bution and potential overlap of physiological dis- to a more advanced stage. turbances. Most guidelines also state that in addition to airflow limitation, patients with COPD have an in- management of stable copd complete response to albuterol (change in FEV1, <200 ml and 12 percent)3,13 and typically do not The major goals of therapy include smoking ces- have evidence of airway hyperresponsiveness (i.e., sation, symptom relief, improvement in physio- an abnormal bronchoconstrictor response to a logical function, and limitation of complications, stimulus such as methacholine). Although these such as abnormal gas exchange and exacerbations features are helpful in distinguishing COPD from of the disease. As summarized in Figure 3, an inte- asthma,3 the distinctions are not entirely clear-cut. grated approach to treatment combines health care Indeed, there is responsiveness to a bronchodila- maintenance and use of drug and supplemental tor in 23 to 42 percent of patients with COPD, de- therapies in a stepwise fashion as the disease pro- pending on the criteria used.14 Furthermore, data gresses. from the Lung Health Study indicate that 59 percent of men and 85 percent of women with mod- health care maintenance erate disease (mean [±SD] FEV1:FVC ratio, 0.63± Regular Assessment of Lung Function 0.055 percent) have airway hyperresponsiveness.15 It is not yet known whether spirometric screening Thus, although guideline-based spirometric criteria for COPD is cost effective, and evidence-based cri- are useful starting points, differentiation of COPD teria for the optimal frequency of such testing in from asthma requires careful integration of epi- patients with established disease need to be es- demiologic risk factors (including the patient’s age, tablished. Until more data become available, we smoking status, and family history), clinical status recommend that spirometry be performed in all

100

Severe

50 (% of predicted) Symptoms 1 FEV

Asymptomatic

20 Lung function Lung function normal reduced

Mild

Axis of Progression

Figure 1. Deterioration in Lung Function in Patients with COPD. Symptoms generally develop only after a significant decline in forced expiratory volume in one second (FEV1) has occurred; they progress as lung function deteriorates further.

2690 n engl j med 350;26 www.nejm.org june 24, 2004

current concepts patients at risk to detect asymptomatic airflow lim- drug therapy itation; in patients with established disease, spi- Inhaled Bronchodilators rometry should be performed at least annually, and Inhaled bronchodilators are the foundation of more frequently if needed, to assess clinical status pharmacotherapy for COPD because of their ca- or the response to therapy. pacity to alleviate symptoms, decrease exacerbations of disease, and improve the quality of life.21-25 Smoking Cessation These drugs also improve airflow and hyperinfla- Abstinence from smoking results in a sustained tion (Fig. 2),26,27 thereby decreasing the work of 50 percent reduction in the rate of lung-function decline in patients with COPD,2 and smoking cessation is the only intervention known to be so ef- Total Lung fective in modifying the disease. Unfortunately, Capacity achieving and maintaining smoking cessation in patients with COPD is a challenge. Approximately 35 percent of the subjects in the Lung Health Study achieved abstinence at one year, but only 22 percent Tidal reported continued abstinence at five years with a Ventilation regimen combining nicotine replacement (available in the form of chewing gum, inhaler, spray, and transcutaneous patch), behavioral counseling, and frequent maintenance visits.16 Sustained- release bupropion is also effective, although the likelihood of sustained abstinence among patients Healthy Patients No Broncho- With Broncho- who have COPD is lower with bupropion than with Patients with COPD dilator dilator 17 Patients with COPD nicotine replacement. during Exercise

Vaccination Figure 2. Pulmonary Hyperinflation in Patients with COPD. Although there is little evidence of a direct benefit As compared with healthy patients, patients with COPD have pulmonary hy- of vaccination in patients with COPD, we recom- perinflation with an increase in functional residual capacity (red) and a de- mend that pneumococcal vaccination and annual crease in inspiratory capacity (blue). This condition increases the volume at influenza vaccination be offered to all patients in which tidal ventilation (oscillating line) occurs and places the muscles of respiration at mechanical disadvantage. Hyperinflation worsens with exercise an attempt to reduce both disease-specific mortali- 18,19 and therefore reduces exercise tolerance (dynamic hyperinflation). Inhaled ty and mortality from all causes. Administra- bronchodilators improve dynamic hyperinflation, as well as hyperinflation at tion of the influenza vaccine does not appear to in- rest (not shown), thereby reducing the work of breathing and increasing exer- crease adverse outcomes in patients with COPD in cise tolerance. the short term.20

Table 1. A Comparison of Four Sets of Staging Criteria for COPD.* American Thoracic Society European Respiratory Society British Thoracic Society Stage (1995)8 (1995)9 (1997)10 GOLD (2003)3

FEV1 Symptoms FEV1 Symptoms FEV1 Symptoms FEV1 Symptoms % % % % 0 (at risk) ≥80 – I (mild) ≥50 NA 70 NA 60–80 ± ≥80 ± II (moderate) 35–49 NA 50–69 NA 40–59 + 50–79 + III (severe) <35 NA <50 NA <40 ++ 30–49 ++ IV (very severe) <30 +++

* GOLD denotes Global Initiative for Chronic Obstructive Lung Disease, and FEV1 forced expiratory volume in one second (shown as a percentage of the predicted normal value). In the Symptoms columns, NA denotes not applicable (staging is based on physiology only), – no symptoms, ± variable symptoms, + mild-to-moderate symptoms, ++ symptoms that limit exertion, and +++ symptoms that limit daily activities.

n engl j med 350;26 www.nejm.org june 24, 2004 2691

The new england journal of medicine

breathing and improving exercise tolerance. Para- Inhaled bronchodilators can be grouped ac- doxically, improvement in function resulting from cording to mechanism or duration of action (Ta- the administration of bronchodilators is not always ble 2). Short-acting b2-adrenergic–receptor agonists reflected by changes in FEV1 and FVC, and mea- (e.g., albuterol sulfate) and cholinergic-receptor surement of lung volumes or inspiratory capacity antagonists (e.g., ipratropium bromide) result in may be necessary to document physiological im- bronchodilation for four to six hours.28 Long-act- 27 provement. ing b2-adrenergic–receptor agonists such as for-

Supplemental Therapy

Surgery

FEV1 Evaluation and treatment of hypoxemia

Pulmonary rehabilitation

Stepwise Drug Therapy Combination of inhaled corticosteroid and long-acting b-agonist

Consider theophylline

Combination of anticholinergic and b-agonist bronchodilator

Long-acting inhaled bronchodilator

Single short-acting Short-acting inhaled bronchodilator inhaled bronchodilator for acute relief of symptoms

Health Care Maintenance Pneumococcal and annual influenza vaccination

Smoking cessation

Regular assessment of lung function

Symptoms

Figure 3. An Algorithm for the Treatment of COPD. The components of COPD therapy include health care maintenance, drug therapy, and supplemental therapy. Because patients with reduced lung function may be asymptomatic, spirometry is indicated to diagnose asymptomatic reduction in lung function in at-risk patients. Treatment should be initiated when reduced lung function is demonstrated, with or without the presence of symptoms. Smoking cessation should be aggressively pursued in patients across the severity spectrum, and vaccination is an important addition to health care maintenance. Patients may initially require only as- needed therapy with a single short-acting anticholinergic agent or b-agonist. For patients with moderate-to-severe disease, or for those with persistent or increasing symptoms with as-needed bronchodilators, a single regularly scheduled, long-acting inhaled bronchodilator of either pharmacologic class or the regularly scheduled combination of a short- or long-acting anticholinergic agent and a b-agonist is preferred. For patients treated with a long-acting inhaled bronchodilator, a short-acting agent should be prescribed concurrently for rapid treatment of acute symptoms (box with dashed outline). The addition of pulmonary rehabilitation to treatment regimens will reduce symptoms and improve exercise performance, and the addition of theophylline or an inhaled corticosteroid (or both) to optimal inhaled bronchodilator therapy may provide additional benefits. Patients with moderate or severe disease should be tested for hypoxemia, and it should be aggressively treated if present. Lung-volume–reduction surgery and transplantation are options for a subgroup of patients with very severe disease.

2692 n engl j med 350;26 www.nejm.org june 24, 2004

current concepts moterol fumarate and salmeterol xinafoate have al improvement in lung function and symptoms an effect for 8 to 12 hours, and the long-acting an- when added to inhaled bronchodilators.35,36 Since ticholinergic agent tiotropium bromide has a dura- theophylline may be toxic, frequent monitoring for tion of effect of more than 24 hours. In stable dis- supratherapeutic levels, adverse drug reactions, and ease, administration by means of a metered-dose drug interactions is critical. or dry-powder inhaler is preferred. For patients with mild airflow limitation and in- Inhaled Corticosteroids termittent symptoms, a single short-acting inhaled The appropriate role of inhaled corticosteroids in bronchodilator relieves symptoms and improves COPD is controversial. Many studies have shown airflow. Albuterol and ipratropium are equally ef- that inhaled corticosteroids do not substantially fective with regard to bronchodilation, symptom modify airway inflammation in COPD, and four scores, and the rates of treatment failure and can be large, long-term clinical trials37-40 comparing in- used interchangeably for mild disease as the first haled corticosteroids with placebo found that these step in a series of measures for treating patients drugs do not appreciably alter the rate of decline with COPD (Fig. 3).28-31 in lung function. This absence of physiological Since most patients have at least moderate air- effects, as well as differences in inflammatory phe- flow limitation when first evaluated, they are likely notype between COPD and asthma,41 has led many to require regularly scheduled bronchodilation3 investigators to conclude that these drugs are inef- and to derive benefit from a long-acting broncho- fective in COPD. dilator as initial therapy. Formoterol, salmeterol, However, some of the same trials have demon- and tiotropium all have an equivalent peak bron- strated that treatment with inhaled corticosteroids chodilator effect but have a prolonged duration of alleviates patients’ symptoms,40 reduces the fre- effect, which may explain the superiority of these quency of exacerbations,39 and improves health sta- 42 drugs to short-acting bronchodilators in reducing tus. These studies relied on FEV1 as the primary symptoms and the frequency of exacerbations and outcome variable and did not evaluate such physio- in improving the quality of life.22,23,25 Treatment logical variables as hyperinflation, which may have may be initiated with either a long-acting anticho- a greater effect on clinical status than the FEV1 does. linergic agent or a b-agonist, since there is little Moreover, these studies may not have been pow- evidence to suggest clinically significant differenc- ered adequately to detect small differences in the 43 es between pharmacologic classes. Long-acting in- rate of decline in the FEV1. In fact, the effect of haled bronchodilators are not appropriate for the cigarette smoking in blunting the spirometric re- treatment of acute symptoms, so a short-acting sponse to corticosteroids, as was recently observed bronchodilator should also be prescribed for acute in a study involving patients with asthma,44 may relief of symptoms (Fig. 3). have affected the response to inhaled corticoste- Combination bronchodilator therapy (an anticholinergic agent plus a b-agonist) may be considered for patients in whom a single inhaled bron- Table 2. Duration and Administration of Inhaled Bronchodilators. chodilator has failed to provide adequate relief. Drug Duration Usual Dose* The combination of albuterol and ipratropium Short-acting provides greater bronchodilation than either drug 28 Albuterol sulfate 4–6 hr Two puffs every 4 hr used alone, and similar benefits are obtained by (MDI, 90 µg/puff) combining long-acting b-agonists with ipratro- 32,33 Ipratropium bromide 4–6 hr Two puffs every 4 hr pium. Although the combination of albuterol (MDI, 18 µg/puff) and ipratropium is commonly prescribed for regu- Long-acting larly scheduled use, regularly scheduled ipratropi- Formoterol fumarate 8–12 hr One inhalation twice daily um combined with albuterol on an as-needed ba- (DPI, 12 µg/inhalation) sis has been reported to be equally effective.34 Salmeterol xinafoate 8–12 hr One inhalation twice daily (DPI, 50 µg/inhalation) Theophylline Tiotropium bromide More than 24 hr One inhalation once daily (DPI, 18 µg/inhalation) If symptoms continue despite combined inhaled- bronchodilator therapy, theophylline may be pre- * MDI denotes metered-dose inhaler, and DPI dry-powder inhaler. scribed because of its capacity to provide addition-

n engl j med 350;26 www.nejm.org june 24, 2004 2693

The new england journal of medicine

roids in the studies of COPD, in which smokers received oral prednisolone before fluticasone.39 made up 39 to 100 percent of the study population. The overall response to prednisolone was minimal Thus, the medical literature suggests that inhaled (a mean increase in FEV1 of 69 ml) and was unre- corticosteroids provide clinical benefit to some pa- lated to the patients’ baseline FEV1, responsiveness tients with COPD and that this effect is indepen- to a bronchodilator, subsequent decline in FEV1, 48 dent of the patients’ FEV1 response, perhaps op- or response to inhaled fluticasone. Thus, although erating through an improvement in hyperinflation a trial of oral corticosteroids may be useful in de- or a reduction in the frequency of exacerbations. tecting coexisting asthma, it is a poor predictor Guidelines recommend that inhaled cortico- of the response to inhaled corticosteroids among steroids be considered for patients with moder- patients with COPD. Oral corticosteroids should ate-to-severe airflow limitation who have persistent not be used in the routine management of stable symptoms despite optimal bronchodilator thera- disease. py.3 This recommendation is based in large part on the Inhaled Steroids in Obstructive Lung Dis- supplemental therapy ease in Europe (ISOLDE) trial, in which subjects Pulmonary Rehabilitation with a mean FEV1 of approximately 50 percent of Pulmonary rehabilitation improves patients’ exer- the predicted normal value had a 25 percent rela- cise capacity, reduces dyspnea, improves the qual- tive reduction in frequency of exacerbations when ity of life,49 and reduces the number and duration treated with inhaled fluticasone propionate.39 Ex- of hospitalizations related to respiratory disease.50 acerbations appear to accelerate the rate of lung- It is appropriate for patients with clinically signif- function decline in COPD,45 and the reduction in icant exertional symptoms and is most effective exacerbations seen in the ISOLDE trial supports when delivered as a multifaceted program incor- the use of inhaled corticosteroids to modify the porating individually tailored aerobic physical train- frequency of exacerbations, independently of the ing, comprehensive education about the disease, drugs’ effects on underlying airway inflammation. psychosocial counseling, and nutritional support.51 In addition, the observation that the combination Although having a low body-mass index is associat- of inhaled corticosteroids and long-acting b-ago- ed with increased mortality from respiratory disease nists is superior to placebo or either drug alone among patients with COPD,52 there is no evidence with regard to lung function, frequency of exacer- that enhanced nutrition improves body weight, lung bations, symptoms, and health status46,47 suggests function, exercise capacity, or survival.53 that the use of inhaled corticosteroids should be restricted to patients in whom optimal bronchodilator Treatment of Abnormal Gas Exchange therapy has failed to improve the symptoms, physi- Hypoxemia develops as a result of a worsening ven- ological findings, or frequency of exacerbations. tilation–perfusion mismatch, and aggressive test- It is important to recognize that in older pa- ing for hypoxemia is critical, since clinical trials tients the side effects of inhaled corticosteroids are have shown that mortality is reduced by treatment not well understood, and the use of these drugs with supplemental oxygen for 15 or more hours per should be carefully considered. Since it is difficult to day.54,55 In stable patients, Medicare guidelines predict accurately which patients will benefit from suggest that oxygen therapy should be initiated therapy, clinical and spirometric responses should if the resting partial pressure of arterial oxygen is be assessed in the months after the initiation of in- 55 mm Hg or lower or if the oxygen saturation is haled corticosteroids. Treatment should be discon- 88 percent or less. However, these recommenda- tinued if no substantial clinical or physiological tions are based in large part on the inclusion crite- improvement is seen, since there is no evidence ria for the Medical Research Council study54 and that continuing treatment with inhaled corticoste- the Nocturnal Oxygen Therapy Trial55 and may not roids provides any long-term benefit in such cases. identify all patients who would benefit from supplemental oxygen. For example, supplemental ox- Oral Corticosteroids ygen substantially improves training intensity and Assessing the spirometric response to a trial of oral exercise tolerance even in patients in whom desat- corticosteroids has been advocated as a means of uration does not occur during exercise.56 identifying patients who have a response to inhaled Supplemental oxygen should be adjusted to corticosteroids. In the ISOLDE trial, all subjects maintain an oxygen saturation of at least 90 per-

2694 n engl j med 350;26 www.nejm.org june 24, 2004

Downloaded from www.nejm.org at UNIV NATAL MED SCH on February 16, 2007 . Copyright © 2004 Massachusetts Medical Society. All rights reserved. current concepts cent at all times. Since patients may have normal therapy alone. In 2000, a small randomized, con- oxygen saturation at rest but hypoxemia with ex- trolled trial that compared lung-volume–reduction ertion or sleep, pulse oximetry and oxygen titra- surgery with medical therapy in patients with se- tion should be performed during all three condi- vere emphysema demonstrated improved lung tions. Worsening hypoxemia during air travel must function, exercise capacity, and quality of life 6 to be considered, and a general recommendation is 12 months after surgery.61 Subsequently, the Na- that patients requiring oxygen should increase tional Emphysema Treatment Trial found that the their oxygen flow rate by 2 liters per minute during addition of lung-volume–reduction surgery to op- flight.57 timal medical therapy and rehabilitation led to In advanced disease, hypoxemia and hypercap- an overall improvement in exercise tolerance and nia (alveolar hypoventilation) may occur as a result survival in a subgroup of patients with reduced ex- of an increase in the dead-space fraction, a ventila- ercise tolerance and predominantly upper-lobe em- tion–perfusion mismatch, or an increase in the work physema.62 Overall mortality did not improve, of breathing with enhanced production of carbon however, and mortality was increased in a sub- dioxide. Inhaled bronchodilators can help reduce group of patients with severe physiological impair- the work of breathing and improve gas exchange ment (FEV1, ≤20 percent of the predicted normal in some patients with alveolar hypoventilation. value) and homogeneous emphysema or a carbon Trials of noninvasive positive-pressure ventilation monoxide diffusing capacity no more than 20 per- have been conducted in patients with stable COPD, cent of the predicted normal value.63 and although hypercapnia can be improved,58 im- Single-lung transplantation is an alternative sur- provement often comes at the cost of increased gical option for patients with end-stage emphyse- 59 hyperinflation. A two-year trial of noninvasive ma who have an FEV1 that is less than 25 percent positive-pressure ventilation in addition to supple- of the predicted normal value after the administra- mental oxygen in patients with alveolar hypoventi- tion of a bronchodilator and who have such com- lation demonstrated improvements in dyspnea and plications as pulmonary hypertension, marked hy- the quality of life but only small improvements in poxemia, and hypercapnia.64 The surgery does not arterial carbon dioxide levels.60 appear to improve survival significantly in these patients, however.65 Surgery Supported by a grant (K23 HL04385) from the National Heart, Lung-volume–reduction surgery can reduce hyper- Lung, and Blood Institute. Dr. Cherniack reports having served as a consultant for Glaxo- inflation and should be considered in patients with SmithKline, and Dr. Sutherland having served as a consultant for severe upper-lobe emphysema and reduced exer- Schering-Plough and having received grant support from Glaxo- cise tolerance who are not faring well with medical SmithKline. references 1. Barnes PJ. Chronic obstructive pulmo- obstructive pulmonary disease. Am J Respir National Health and Nutrition Examination nary disease. N Engl J Med 2000;343:269- Crit Care Med 1996;153:1802-11. Survey follow up study. Thorax 2003;58: 80. 7. Perez-Padilla R, Regalado J, Vedal S, et 388-93. 2. Anthonisen NR, Connett JE, Murray RP. al. Exposure to biomass smoke and chronic 12. Vestbo J, Lange P. Can GOLD Stage 0 Smoking and lung function of Lung Health airway disease in Mexican women: a case- provide information of prognostic value in Study participants after 11 years. Am J Res- control study. Am J Respir Crit Care Med chronic obstructive pulmonary disease? Am pir Crit Care Med 2002;166:675-9. 1996;154:701-6. J Respir Crit Care Med 2002;166:329-32. 3. Fabbri LM, Hurd SS. Global strategy for 8. Standards for the diagnosis and care of 13. Lung function testing: selection of ref- the diagnosis, management and prevention patients with chronic obstructive pulmo- erence values and interpretative strategies. of COPD: 2003 update. Eur Respir J 2003; nary disease. Am J Respir Crit Care Med Am Rev Respir Dis 1991;144:1202-18. 22:1-2. 1995;152:S77-S121. 14. Calverley PM, Burge PS, Spencer S, 4. Fletcher C, Peto R. The natural history 9. Siafakas NM, Vermeire P, Pride NB, et al. Anderson JA, Jones PW. Bronchodilator of chronic airflow obstruction. Br Med J Optimal assessment and management of reversibility testing in chronic obstructive 1977;1:1645-8. chronic obstructive pulmonary disease pulmonary disease. Thorax 2003;58:659-64. 5. Eriksson S. Studies in alpha 1-anti- (COPD). Eur Respir J 1995;8:1398-420. 15. Tashkin DP, Altose MD, Bleecker ER, et trypsin deficiency. Acta Med Scand Suppl 10. BTS guidelines for the management of al. The Lung Health Study: airway respon- 1965;432:1-85. chronic obstructive pulmonary disease. Tho- siveness to inhaled methacholine in smok- 6. Tashkin DP, Altose MD, Connett JE, rax 1997;52:Suppl 5:S1-S28. ers with mild to moderate airflow limita- Kanner RE, Lee WW, Wise RA. Methacho- 11. Mannino DM, Buist AS, Petty TL, Enright tion. Am Rev Respir Dis 1992;145:301-10. line reactivity predicts changes in lung func- PL, Redd SC. Lung function and mortality 16. Anthonisen NR, Connett JE, Kiley JP, et tion over time in smokers with early chronic in the United States: data from the First al. Effects of smoking intervention and the

n engl j med 350;26 www.nejm.org june 24, 2004 2695

use of an inhaled anticholinergic broncho- solution in patients with chronic obstruc- Venn AJ, Martin RJ. Inhaled corticosteroids dilator on the rate of decline of FEV1: the tive pulmonary disease: a three-center study. reduce the progression of airflow limitation Lung Health Study. JAMA 1994;272:1497- Am Rev Respir Dis 1989;139:1188-91. in chronic obstructive pulmonary disease: 505. 30. Ram FS, Sestini P. Regular inhaled short a meta-analysis. Thorax 2003;58:937-41. 17. Tashkin D, Kanner R, Bailey W, et al. acting beta2 agonists for the management 44. Chaudhuri R, Livingston E, McMahon Smoking cessation in patients with chronic of stable chronic obstructive pulmonary dis- AD, Thomson L, Borland W, Thomson NC. obstructive pulmonary disease: a double- ease: Cochrane systematic review and meta- Cigarette smoking impairs the therapeutic blind, placebo-controlled, randomised trial. analysis. Thorax 2003;58:580-4. response to oral corticosteroids in chronic Lancet 2001;357:1571-5. 31. Petrie GR, Palmer KN. Comparison of asthma. Am J Respir Crit Care Med 2003;168: 18. Nichol KL, Margolis KL, Wuorenma J, aerosol ipratropium bromide and salbuta- 1308-11. Von Sternberg T. The efficacy and cost effec- mol in chronic bronchitis and asthma. Br 45. Donaldson GC, Seemungal TA, Bhow- tiveness of vaccination against influenza Med J 1975;1:430-2. mik A, Wedzicha JA. Relationship between among elderly persons living in the commu- 32. Wadbo M, Lofdahl CG, Larsson K, et al. exacerbation frequency and lung function nity. N Engl J Med 1994;331:778-84. Effects of formoterol and ipratropium bro- decline in chronic obstructive pulmonary dis- 19. Nichol KL, Nordin J, Mullooly J, Lask R, mide in COPD: a 3-month placebo-con- ease. Thorax 2002;57:847-52. Fillbrandt K, Iwane M. Influenza vaccina- trolled study. Eur Respir J 2002;20:1138-46. 46. Calverley PM, Pauwels RA, Vestbo J, et al. tion and reduction in hospitalizations for 33. van Noord JA, de Munck DR, Bantje TA, Combined salmeterol and fluticasone in cardiac disease and stroke among the elderly. Hop WC, Akveld ML, Bommer AM. Long- the treatment of chronic obstructive pulmo- N Engl J Med 2003;348:1322-32. term treatment of chronic obstructive pul- nary disease: a randomised controlled trial. 20. Tata LJ, West J, Harrison T, Farrington P, monary disease with salmeterol and the Lancet 2003;361:449-56. [Erratum, Lancet Smith C, Hubbard R. Does influenza vacci- additive effect of ipratropium. Eur Respir 2003;361:1660.] nation increase consultations, corticosteroid J 2000;15:878-85. 47. Szafranski W, Cukier A, Ramirez A, et al. prescriptions, or exacerbations in subjects 34. Cook D, Guyatt G, Wong E, et al. Regu- Efficacy and safety of budesonide/formot- with asthma or chronic obstructive pulmo- lar versus as-needed short-acting inhaled erol in the management of chronic obstruc- nary disease? Thorax 2003;58:835-9. beta-agonist therapy for chronic obstructive tive pulmonary disease. Eur Respir J 2003; 21. Higgins BG, Powell RM, Cooper S, Tat- pulmonary disease. Am J Respir Crit Care 21:74-81. [Erratum, Eur Respir J 2003;21: tersfield AE. Effect of salbutamol and ipra- Med 2001;163:85-90. 912.] tropium bromide on airway calibre and 35. McKay SE, Howie CA, Thomson AH, 48. Burge PS, Calverley PM, Jones PW, Spen- bronchial reactivity in asthma and chronic Whiting B, Addis GJ. Value of theophylline cer S, Anderson JA. Prednisolone response bronchitis. Eur Respir J 1991;4:415-20. treatment in patients handicapped by chronic in patients with chronic obstructive pulmo- 22. Dahl R, Greefhorst LA, Nowak D, et al. obstructive lung disease. Thorax 1993;48: nary disease: results from the ISOLDE study. Inhaled formoterol dry powder versus ipra- 227-32. Thorax 2003;58:654-8. tropium bromide in chronic obstructive pul- 36. ZuWallack RL, Mahler DA, Reilly D, et 49. Lacasse Y, Wong E, Guyatt GH, King D, monary disease. Am J Respir Crit Care Med al. Salmeterol plus theophylline combina- Cook DJ, Goldstein RS. Meta-analysis of 2001;164:778-84. tion therapy in the treatment of COPD. Chest respiratory rehabilitation in chronic obstruc- 23. Rennard SI, Anderson W, ZuWallack R, 2001;119:1661-70. tive pulmonary disease. Lancet 1996;348: et al. Use of a long-acting inhaled beta2- 37. Pauwels RA, Löfdahl C-G, Laitinen LA, 1115-9. adrenergic agonist, salmeterol xinafoate, in et al. Long-term treatment with inhaled 50. Griffiths TL, Burr ML, Campbell IA, et al. patients with chronic obstructive pulmonary budesonide in persons with mild chronic Results at 1 year of outpatient multidisci- disease. Am J Respir Crit Care Med 2001;163: obstructive pulmonary disease who continue plinary pulmonary rehabilitation: a random- 1087-92. smoking. N Engl J Med 1999;340:1948-53. ised controlled trial. Lancet 2000;355:362-8. 24. Aalbers R, Ayres J, Backer V, et al. For- 38. Vestbo J, Sorensen T, Lange P, Brix A, [Erratum, Lancet 2000;355:1280.] moterol in patients with chronic obstruc- Torre P, Viskum K. Long-term effect of 51. British Thoracic Society Standards of tive pulmonary disease: a randomized, con- inhaled budesonide in mild and moderate Care Subcommittee on Pulmonary Reha- trolled, 3-month trial. Eur Respir J 2002;19: chronic obstructive pulmonary disease: bilitation. Pulmonary rehabilitation. Tho- 936-43. [Erratum, Eur Respir J 2002;20:245.] a randomised controlled trial. Lancet 1999; rax 2001;56:827-34. 25. Casaburi R, Mahler DA, Jones PW, et al. 353:1819-23. 52. Gray-Donald K, Gibbons L, Shapiro SH, A long-term evaluation of once-daily inhaled 39. Burge PS, Calverley PM, Jones PW, Spen- Macklem PT, Martin JG. Nutritional status tiotropium in chronic obstructive pulmo- cer S, Anderson JA, Maslen TK. Random- and mortality in chronic obstructive pulmonary disease. Eur Respir J 2002;19:217-24. ised, double blind, placebo controlled study nary disease. Am J Respir Crit Care Med 1996; 26. Belman MJ, Botnick WC, Shin JW. of fluticasone propionate in patients with 153:961-6. Inhaled bronchodilators reduce dynamic moderate to severe chronic obstructive pul- 53. Ferreira IM, Brooks D, Lacasse Y, Gold- hyperinflation during exercise in patients monary disease: the ISOLDE trial. BMJ 2000; stein RS, White J. Nutritional supplementa- with chronic obstructive pulmonary disease. 320:1297-303. tion for stable chronic obstructive pulmo- Am J Respir Crit Care Med 1996;153:967-75. 40. The Lung Health Study Research Group. nary disease. Cochrane Database Syst Rev 27. O’Donnell DE, Lam M, Webb KA. Spiro- Effect of inhaled triamcinolone on the decline 2003;1:CD000998. metric correlates of improvement in exercise in pulmonary function in chronic obstruc- 54. Long term domiciliary oxygen therapy performance after anticholinergic therapy tive pulmonary disease. N Engl J Med 2000; in chronic hypoxic cor pulmonale compli- in chronic obstructive pulmonary disease. 343:1902-9. cating chronic bronchitis and emphysema: Am J Respir Crit Care Med 1999;160:542-9. 41. Sutherland ER, Martin RJ. Airway inflam- report of the Medical Research Council 28. Combivent Inhalation Aerosol Study mation in chronic obstructive pulmonary Working Party. Lancet 1981;1:681-6. Group. In chronic obstructive pulmonary disease: comparisons with asthma. J Allergy 55. Nocturnal Oxygen Therapy Trial Group. disease, a combination of ipratropium and Clin Immunol 2003;112:819-27. Continuous or nocturnal oxygen therapy in albuterol is more effective than either agent 42. Spencer S, Calverley PM, Sherwood hypoxemic chronic obstructive lung dis- alone: an 85-day multicenter trial. Chest Burge P, Jones PW. Health status deteriora- ease: a clinical trial. Ann Intern Med 1980; 1994;105:1411-9. tion in patients with chronic obstructive pul- 93:391-8. 29. Gross NJ, Petty TL, Friedman M, Sko- monary disease. Am J Respir Crit Care Med 56. Emtner M, Porszasz J, Burns M, Som- rodin MS, Silvers GW, Donohue JF. Dose 2001;163:122-8. fay A, Casaburi R. Benefits of supplemental response to ipratropium as a nebulized 43. Sutherland ER, Allmers H, Ayas NT, oxygen in exercise training in nonhypox-

2696 n engl j med 350;26 www.nejm.org june 24, 2004

Downloaded from www.nejm.org at UNIV NATAL MED SCH on February 16, 2007 . Copyright © 2004 Massachusetts Medical Society. All rights reserved. current concepts emic chronic obstructive pulmonary disease lung volume in severe stable COPD. Thorax medical therapy for severe emphysema. patients. Am J Respir Crit Care Med 2003; 2002;57:533-9. N Engl J Med 2003;348:2059-73. 168:1034-42. 60. Clini E, Sturani C, Rossi A, et al. The Ital- 63. Idem. Patients at high risk of death after 57. Johnson AOC. Chronic obstructive pul- ian multicentre study on noninvasive venti- lung-volume–reduction surgery. N Engl J monary disease · 11: fitness to fly with COPD. lation in chronic obstructive pulmonary dis- Med 2001;345:1075-83. Thorax 2003;58:729-32. ease patients. Eur Respir J 2002;20:529-38. 64. Arcasoy SM, Kotloff RM. Lung trans- 58. Meecham Jones DJ, Paul EA, Jones PW, [Erratum, Eur Respir J 2002;20:1617.] plantation. N Engl J Med 1999;340:1081- Wedzicha JA. Nasal pressure support venti- 61. Geddes D, Davies M, Koyama H, et al. 91. lation plus oxygen compared with oxygen Effect of lung-volume–reduction surgery in 65. Hosenpud JD, Bennett LE, Keck BM, therapy alone in hypercapnic COPD. Am patients with severe emphysema. N Engl Edwards EB, Novick RJ. Effect of diagnosis J Respir Crit Care Med 1995;152:538-44. J Med 2000;343:239-45. on survival benefit of lung transplantation 59. O’Donoghue FJ, Catcheside PG, Jordan 62. National Emphysema Treatment Trial for end-stage lung disease. Lancet 1998; AS, Bersten AD, McEvoy RD. Effect of CPAP Research Group. A randomized trial com- 351:24-7. on intrinsic PEEP, inspiratory effort, and paring lung-volume-reduction surgery with Copyright © 2004 Massachusetts Medical Society.

electronic access to the journal’s cumulative index

At the Journal’s site on the World Wide Web (www.nejm.org), you can search an index of all articles published since January 1975 (abstracts 1975–1992, full text 1993–present). You can search by author, key word, title, type of article, and date. The results will include the citations for the articles plus links to the full text of articles published since 1993. For nonsubscribers, time-limited access to single articles and 24-hour site access can also be ordered for a fee through the Internet (www.nejm.org).

n engl j med 350;26 www.nejm.org june 24, 2004 2697

Downloaded from www.nejm.org at UNIV NATAL MED SCH on February 16, 2007 . Copyright © 2004 Massachusetts Medical Society. All rights reserved. British Journal of Pharmacology (2006) 147, S297–S303 & 2006 Nature Publishing Group All rights reserved 0007–1188/06 $30.00 www.nature.com/bjp Drugs for asthma

*Peter J. Barnes

Department of Thoracic Medicine, National Heart and Lung Institute, Imperial College School of Medicine, Dovehouse St, SW3 6LYLondon

Current drug therapy for asthma is highly effective and has evolved from naturally occurring substances through logical pharmaceutical developments. Pharmacology has played a critical role in asthma drug development and several key experimental observations have been published in this journal. Understanding the pharmacology of effective drug therapies has also taught us much about the underlying mechanisms of asthma. b2-Adrenoceptor agonists are the most effective bronchodilators and evolved from catecholamines from the adrenal medulla, whereas corticosteroids, from the adrenal cortex, are by far the most effective controllers of the underlying inflammatory process in the airways. The current ‘gold standard’ of asthma therapy is a combination inhaler containing a long- acting b2-agonist with a corticosteroid – an improved form of adrenal gland extract. Cromoglycate, derived from a plant product and theophylline, a dietary methyl xanthine, have also been extensively used in the therapy of asthma, but we still do not understand their molecular mechanisms. Pharmacology has played an important role in improving natural products to make effective long lasting and safe asthma therapies, but has so far been challenged to produce new classes of antiasthma therapy. The only novel class of antiasthma therapy introduced in the last 30 years are leukotriene antagonists, which are less effective than existing treatments. New, more specific, therapies targeted at specific cytokines are less effective than corticosteroids, whereas more effective therapies carry a risk of side effects that may not be acceptable. It seems likely that pharmacology, rather than molecular genetics, will remain the main approach to the further improvement of treatment for asthma. British Journal of Pharmacology (2006) 147, S297–S303. doi:10.1038/sj.bjp.0706437 Keywords: Asthma; bronchodilator; b-adrenoceptor agonist; muscarinic antagonist; theophylline; cromoglycate; corticosteroids Abbreviations: ACTH, adrenocorticotrophic hormone; BDP, beclomethasone dipropionate; COPD, chronic obstructive pulmonary disease; Cys-LT, cysteinyl leukotriene; DSCG, disodium cromoglycate; PDE, phosphodiesterase; SRS-A, slow-reacting substance of anaphylaxis

Introduction

We have now evolved highly effective drugs for the manage- role in the development of bronchodilators and British ment of asthma that have led to a marked reduction in hospital pharmaceutical companies have played (and continue to play) admissions and mortality for this increasingly common a leading role in the development of asthma medications, disease. Most patients with asthma are now able to lead supported by strong interactions with basic and clinical a normal life through the use of medications that are virtually pharmacologists. free of side effects. Indeed, current therapies are so effective It is of interest that many of our effective therapies for that it has so far been proved impossible to develop any asthma were originally derived from natural substances. Many new classes of drug that are more effective than existing agents. were isolated from plants through the discovery of herbal The advances in asthma therapy have been largely through remedies, including atropine, dietary xanthines such as improving the selectivity and duration of action of existing theophylline and chromones from a Mediterranean medicinal effective classes of drugs. With respect to this, pharmacology herb. The most effective treatments for asthma are derived has played an important role in validating drug targets from hormones, b-adrenoceptor agonists from adrenaline and and drug design. Pharmacology, particularly through the corticosteroids from cortisone, both secreted by the adrenal use of selective agonists and antagonists, has also played gland. Indeed, the most effective therapies available for asthma an important role in increasing our understanding of the so far are combination inhalers containing a long-acting underlying inflammatory mechanisms of asthma, providing b-agonist and a corticosteroid. a rational basis for the use of current drug therapy. The history of asthma treatment goes back to thousands of years, but most of the important advances have been made during Muscarinic receptor antagonists the last 75 years. As indicated in this review several key studies have been published in the British Journal of Pharmacology. The leaves of Datura, commonly known as Jimson weed or Autonomic pharmacology, which evolved and has always been thorn apple, which were smoked in India for several centuries very strong in the U.K., has played a particularly important as a treatment for respiratory disorders (including asthma), contain a muscarinic receptor antagonist, atropine. The ancient Egyptians also inhaled the vapour of heated henbane *Author for correspondence; E-mail: [email protected] (Hyoscyamus muticus), which contains another antimuscarinic S298 P.J. Barnes Drugs for asthma alkaloid, scopolamine, for the treatment of asthma-like the earliest antiasthma agent known. It acts indirectly, by conditions. These therapies were available until well into the releasing endogenous catecholamines, resulting in bronchodi- last century but fell into disuse with the introduction of more latation. It was shown to be effective by inhalation by Dale as effective bronchodilators derived from adrenaline. An impor- early as 1910 (Barger & Dale, 1910). Oliver and Sharpey- tant advance in the use of muscarinic receptor antagonists for Shafer were the first to describe the effect of adrenal gland asthma was the development of quaternary ammonium extract on blood pressure but they did not study any airway derivatives, which did not pass the blood–brain barrier and effects. It was Solis-Cohen (1900), a physician from Philadel- thus were devoid of the central side effects, such as hallucina- phia, who first showed that orally administered adrenal extract tions, of naturally occurring atropine-like compounds. (adrenal substance pills) was beneficial in asthma and the Although these quaternary derivatives are not absorbed from direct bronchodilator effect of adrenaline was first demon- the gastrointestinal tract, they are effective when inhaled and strated by Kahn in 1907 using precontracted tracheal strips ipratropium bromide, a synthetic quaternary antimuscarinic in vitro (Brewis, 1990). Adrenaline given by subcutaneous compound, is still used as a bronchodilator in patients with injection became a widely used treatment, particularly severe asthma. However, it is less effective than a b-agonist, as for acute exacerbations of asthma. Of course, adrenergic cholinergic bronchoconstriction is only a relatively small agonists are now given preferably by inhalation and the component of the bronchoconstriction in asthma compared first known description of inhaled adrenaline in asthma was by to the direct bronchoconstrictor action of other inflammatory Percy Camps, a general practitioner from Teddington, who mediators in most patients. However, antimuscarinic agents described the efficacy of nebulising an adrenaline solution with have turned out to be the bronchodilators of choice in the oxygen in patients with acute exacerbations of asthma (Brewis, treatment of chronic obstructive pulmonary disease (COPD), 1990). where the only reversible component appears to be cholinergic Isoprenaline was synthesised by German chemists in the tone in the airways. The most recent advance has been the 1940s and was shown to have less cardiovascular side effects introduction of the long-acting antimuscarinic, tiotropium, than adrenaline and became the most widely used inhaled developed by Boehringer Ingelheim, which induces broncho- treatment for asthma for about 20 years. It was the synthesis dilatation lasting for several days (Hansel & Barnes, 2002). of isoprenaline that allowed Ahlquist in 1948 to distinguish A key development in muscarinic pharmacology has been between a- and b-adrenergic receptors, based on the difference the recognition of distinct muscarinic receptor subtypes, which in bronchial response to isoprenaline and noradrenaline have different functions and distribution. An important (Ahlquist, 1948). In 1967 Lands et al. (1967) demonstrated, development in lung pharmacology made by Fryer & using the rank order of potency of natural and synthetic

Maclagan (1984) was the recognition that M3 receptors sympathomimetic amines, that b-receptors could be further mediate the bronchoconstrictor effect of cholinergic tone, subdivided into b1-receptors in the heart and b2-receptors whereas M2 receptors functioned as feedback inhibitory in the airways. Isoetharine had been found by Lands to be receptors (autoreceptors) in parasympathetic nerves of a highly selective agonist at b2-adrenoceptors and this was animals. This was subsequently confirmed in human airways (Minette & Barnes, 1988). The clinical consequence is that the nonselective muscarinic antagonists, such as atropine and ipratropium, will also increase acetylcholine production from cholinergic nerves by blocking the M2 autoreceptors and may thus overcome the blockade of the M3 receptors on airway smooth muscle cells. This led to the idea that M3-selective antagonists may be more effective as bronchodilators. Indeed, tiotropium has a kinetic selectivity for M3 receptors as it dissociates much more slowly from M3 receptors than from M2 receptors. However, it has not been convincingly shown that

M3 receptor selectivity has any important clinical advantage and the long duration of action is a much more important advantage. New developments include more long-acting muscarinic antagonists which will be used alone and in combination with long-acting b2-agonists, mainly for COPD patients, but also for patients with asthma. Interestingly, an old antimuscarinic drug, glycopyrrolate, used for many years by anaesthetists to dry upper airway secretions, has recently been found to have a similar pharmacology to tiotropium with kinetic selectivity for M3 receptors and a long duration of action when given by inhalation. b-Adrenoceptor agonists

Ephedrine, derived from the plant Ephedra and known in Chinese medicine as Ma Huang, has been used in the treatment of respiratory diseases for over 5000 years. Ephedrine is thus Figure 1 Sir David Jack, FRS.

British Journal of Pharmacology vol 147 (S1) P.J. Barnes Drugs for asthma S299 confirmed in humans by Collier & Dornhorst (1969). How- Hopkins University had shown that it had equally good ever, isoetharine was short-lived in its effects, like isoprenaline, effects in patients with asthma. They described five patients due to rapid metabolism of the catechol ring. A major with asthma, interestingly all of whom had eosinophilic breakthrough was the discovery of the first b2-selective agonist sputum, who improved rapidly with intramuscular injections with a longer duration of action than isoprenaline by the team of ACTH over a 3-week period with disappearance of the at Glaxo led by David Jack and Roy Brittain (Figure 1; sputum. They later confirmed these observations in a larger Brittain et al., 1968), and its pharmacology was elegantly group of patients. Subsequently, oral cortisone, widely used at described in the British Journal of Pharmacology by Cullum that time to treat several inflammatory diseases, was shown to et al. (1969) (later Alabaster). This compound, salbutamol, be an effective replacement for the injections, in patients with remains the most widely used antiasthma drug in the world difficult-to-control asthma. However, there was scepticism in today. the U.K. leading to a Medical Research Council multicentre The next logical development was to extend the duration of trial of cortisone in asthma patients, which was the first action of salbutamol by substitution in the side chain and this placebo-controlled trial performed in asthma (Medical resulted in the discovery by Brittain and Jack of salmeterol, Research Council, 1956). Surprisingly the results were the first long-acting b2-agonist with a bronchodilator action disappointing with few clinical improvements that were not of over 12 h (Ball et al., 1991). Inhaled salmeterol was shown sustained during the 2 months of therapy. This may have to have a prolonged duration of action in patients with reflected the low dose of cortisone used, the lack of objective asthma (Ullman & Svedmyr, 1988) and was introduced into measurements of lung function and the inclusion of many clinical practice in 1990. Another long-acting b2-agonist, patients who had COPD. Despite this poor result, oral steroids formoterol, was initially used in tablet form in Japan, with were used more and more in patients with severe asthma, but it no indication of a long duration of action. This was was clear that side effects were a major problem, resulting in only discovered when formoterol was given by inhalation to stunting of growth in children, osteoporosis and metabolic asthmatic patients and shown to have a similar duration of disturbances. action to salmeterol. These long-acting b2-agonists have found This immediately suggested the need to give corticosteroids an important place in the management of asthma in by inhalation as a way of reducing systemic side effects, yet combination with a corticosteroid. These combination inhalers cortisone and dexamethasone given by inhalation proved to are the most effective asthma therapies currently available, as be of little benefit. This was because of their lack of topical the long-acting b2-agonist and corticosteroids exert comple- efficacy and led to a search for topically active steroids. mentary actions and, in some situations, can show synergism McKenzie & Stoughton (1962) discovered that this topical (Barnes, 2002). Combination inhalers are now the most rapidly efficacy was correlated with skin blanching, although the grown segment of the antiasthma market, with the most recent cellular basis for this test is still uncertain. Hydrocortisone development being the synthesis of even longer acting b2- turned out to be weak in the McKenzie test, but two synthetic agonists, such as indacaterol (QAB149), which has a duration steroids, beclomethasone dipropionate (BDP) and betametha- of over 24 h making it suitable for once-daily dosing (Beeh sone-17-valerate gave good skin blanching responses. Both of et al., 2005). these steroids were effective as topical treatments for eczema, predicting that they would also be effective by inhalation. Both of these new steroids were developed for inhalation and an Corticosteroids important paper by Harry Morrow Brown and co-workers in 1972 established that inhaled BDP was very effective in Corticosteroids are the most effective controllers of asthma reducing the need for oral corticosteroids and in many patients and it has been proved to be extremely difficult to find any achieved better control (Brown et al., 1972). Interestingly, new treatment that comes close to providing an equal Brown reported that the patients who did best had high therapeutic benefit. It is likely that the benefits of orally numbers of eosinophils in their sputum, an observation that administered adrenal extract described by Solis-Cohen in 1900 has been confirmed in many subsequent studies. The wide- was, in fact, due to the steroid content rather than spread use of inhaled corticosteroids in asthma has been the any adrenaline present, as the adrenaline would be extensively major reason why asthma morbidity and recently mortality metabolised on its absorption from the gastrointestinal tract. have fallen. There is now a search for inhaled corticosteroids The gut wall and liver contain high levels of monoamine with improved therapeutic ratios and less systemic side effects, oxidase, the major enzyme inactivating endogenous mono- with the introduction of budesonide and fluticasone propio- amines including catecholamines (see also Youdim & Bakhle, nate which have reduced oral bioavailability and recently this issue). This was not recognised at the time and it was ciclesonide, which is a prodrug, activated by esterases in the almost 50 years later, when cortisol had been isolated from the lower airways. adrenal cortex, that the idea of corticosteroids as therapy for There has recently been a much better understanding of the asthma became clear. The Nobel Prize for Medicine and molecular mechanisms involved in the anti-inflammatory Physiology was awarded in 1950 to Kendall and Reichstein effects of corticosteroids in asthma, with particular emphasis who had independently isolated and synthesised cortisol and on the effects of corticosteroids on chromatin remodelling then adrenocorticotropic hormone (ACTH) and Philip Hench, through increased recruitment of histone deacetylase-2 to a rheumatologist working at the Mayo Clinic, who had activated inflammatory genes (Barnes & Adcock, 2003). In the demonstrated its dramatic efficacy when given by intravenous future, there is a prospect for selective glucocorticoid receptor injection in patients with rheumatoid arthritis. Only 6 months agonists or dissociated steroids, which have improved anti- after Hench’s demonstration of the clinical efficacy of ACTH inflammatory effects through repression of activated inflam- in rheumatoid patients, Boardley et al. (1949) at Johns matory genes and reduced binding of glucocorticoid receptors

British Journal of Pharmacology vol 147 (S1) S300 P.J. Barnes Drugs for asthma to DNA, which is thought to mediate side effects (see also sensory nerves. Their lack of side effects implied that their Buckingham, this issue). Interestingly, the topically active effect was specific for the abnormality of asthma, but their corticosteroids now used in inhalers for asthma already show a molecular target has not yet been identified, although there is degree of dissociation. some evidence that they act on certain chloride channels (Norris & Alton, 1996). Identification of the molecular mechanism of action of chromones may be an important Chromones approach to finding new antiasthma medications and the development of longer acting and perhaps orally active drugs Khellin is a naturally occurring chromone, extracted from the that target the same mechanism. medicinal plant Amni visnaga and long used in Egypt and the Eastern Mediterranean countries for the treatment of respiratory disorders. Khellin has bronchodilator properties but also Theophylline caused nausea and a research group at Fisons Pharmaceuticals decided to test related chromone derivatives as potential Theophylline, a methyl xanthine found in tea, was isolated at antiasthma drugs. As there was no satisfactory animal model the end of the 19th century but its use in asthma was not seen these compounds were tested on allergen challenge in until Hirsch (1922) described its bronchodilator effect in three asthmatic volunteers, including the leader of the team, Roger asthmatic patients and its relaxant effect in bovine airways Altounyan (Figure 2). Altounyan identified the most active in vitro. This confirmed the in vitro observations made in the compounds, leading eventually to the synthesis of a bis- previous year by Macht & Ting (1921). The soluble ethylene chromone, disodium cromoglycate (DSCG). This remarkable diamine salt of theophylline, aminophylline, was developed for drug inhibited not only antigen challenge but also challenges intravenous administration and shown to be very effective in due to exercise and irritant gases. DSCG was orally inactive acute severe asthma, particularly in patients who had not and had to be given by a dry powder inhaler device (Spinhaler) responded well to adrenaline (Hermann et al., 1937). that was devised by Altounyan. DSCG proved to be effective Intravenous aminophylline remained a standard treatment in clinical trials in asthmatic patients and was without side for acute exacerbations of asthma until displaced by nebulised effects (Howell & Altounyan, 1967). However, DSCG had a b2-agonists over the last 20 years. It is still used in occasional short duration of action, prompting the search for compounds patients who fail to respond to adrenergic bronchodilators. of longer duration or that were orally active. Nedocromil The main limitations of theophylline are its side effects, such as sodium was introduced as a slightly longer-acting inhaled nausea, headache and diuresis, which occurred within the cromone but had little advantage over DSCG. Chromones therapeutic range and occasionally the very serious adverse have now largely been replaced by inhaled corticosteroids, but effects of cardiac arrhythmias and seizures. Indeed, overdosage they remain a fascinating novel therapy with an unknown of aminophylline was to become the commonest cause of death mode of action. Although it was believed that chromones due to asthma in hospital. worked as mast cell stabilisers (Cox, 1967), it later became This led to several studies relating the efficacy and side clear that they also worked on other cell types, including effects of theophylline to plasma concentrations. In a classical pharmacokinetic study, Mitenko & Ogilvie (1973) demonstrated that the bronchodilator effect of theophylline was related to plasma concentration between 5 and 20 mg l1, but above 20 mg l1, side effects were very common. This led to recommendations for a therapeutic range of 10–20 mg l1, although even within this range side effects were relatively common. Plasma monitoring became routine, particularly in view of the variable pharmacokinetics of theophylline and the multiplicity of factors that affected plasma concentrations. Oral theophylline was a very popular treatment which was inexpensive, eventually becoming the most widely used therapy worldwide, but it has a short duration of action. This limitation led to the formulation of slow-release theophylline and aminophylline preparations that could be given once or twice daily, which were successful due to their convenience and greater tolerability. Side effects limited the use of theophylline as a bronchodilator and the introduction of inhaled

b2-agonists as bronchodilators led to a decline in its use as b-agonists were more effective and better tolerated. Theophylline is still used in the management of severe asthma as an additional therapy added to inhaled corticosteroids and there has recently been a revival of interest in its mechanisms of action (Barnes, 2003). The bronchodilator effect of theophylline appears to be due to inhibition of phosphodiesterases (principally PDE3 and 4) in airway smooth muscle and this may also account for the Figure 2 Dr Roger Altounyan (1922–1987). nausea, headaches and some of the cardiovascular side effects,

British Journal of Pharmacology vol 147 (S1) P.J. Barnes Drugs for asthma S301 explaining why these side effects are commonly seen at guinea-pig lungs perfused with cobra venom or as shown bronchodilator doses. Theophylline acts as a functional later, allergen, released a substance that contracted smooth antagonist in airway smooth muscle and has greater efficacy muscle preparations more slowly than histamine, which they than b2-agonists when airway smooth muscle is strongly designated slow-reacting muscle-stimulant substance (Feldberg contracted (Karlsson & Persson, 1981). Theophylline is also an & Kellaway, 1938). It was subsequently shown to be released adenosine receptor antagonist at relatively high concentrations with histamine during anaphylactic shock and termed slow- and this may account for serious side effects such as cardiac reacting substance of anaphylaxis (SRS-A) (Brocklehurst, arrhythmias and seizures. However, there is increasing 1960). We made great efforts to identify this substance but it evidence that at lower plasma concentrations (5–10 mg l1) was not until 1979 that the chemical structure of SRS-A was theophylline has nonbronchodilator effects that include anti- identified as a mixture of lipid mediators called leukotrienes, inflammatory actions and immunomodulatory effects. These from the fact that they could be derived from leukocytes and cannot be accounted for by PDE inhibition or adenosine the carbon backbone had three double bonds (Murphy antagonism suggesting that there is some other molecular basis et al., 1979). The biosynthetic pathway for the cysteinyl- for these effects. Recently, theophylline in low therapeutic leukotrienes (cys-LT) involved the initial oxidation of arachi- concentrations has been shown to activate the nuclear enzyme donic acid by 50-lipoxygenase and subsequent conjugation histone deacetylase and thereby to switch off activated with glutathione. The effects of cys-LTs were blocked by inflammation genes in an action that is synergistic with the previously discovered SRS-A anatgonist called FPL 55712. corticosteroids (Ito et al., 2002). This interaction may open Priscilla Piper and her co-workers carried out a series of up possibilities for novel anti-inflammatory therapies in the studies on the pharmacological properties of cys-LTs (Piper & future. Samhoun, 1982), which were shown to be potent bronchocon- strictors when given by inhalation to asthmatic patients. This then led to a search for more potent and selective cys-LT Antagonists of inflammatory mediators antagonists or 50-lipoxygenase inhibitors, culminating in the development of several highly potent cys-LT1-receptor antago- Many mediators have been implicated in the pathophysiology nists, such as zafirlukast and montelukast. These were of asthma with the result that even more antagonists and introduced into clinical practice in the 1990s and were the synthesis inhibitors have been developed as potential anti- first new class of anti-asthma therapy for over 30 years. asthma therapies. Inflammatory mediators come and go in Although cys-LT antagonists have some clinical efficacy in popularity, but most have proved to be disappointing as asthma patients, they have proved to be relatively weak therapeutic targets for asthma. There are over 100 mediators compared to inhaled corticosteroids. However, they have some already implicated in asthma, making it unlikely (but not value as an add-on therapy to inhaled corticosteroids and have impossible) that blocking a single mediator would have a the advantage of oral administration without significant side major clinical effect (Barnes et al., 1998). effects. Histamine was the first mediator implicated in the patho- Other mediator antagonists, including antagonists of kinins, physiology of asthma since Dale’s demonstration that it platelet-activating factor or inhibitors of prostaglandin synth- mimicked anaphylactic bronchoconstriction in guinea pigs esis provide no clinical benefit in asthma. More recently (Dale & Laidlaw, 1910). Schild went on to show that sensitised inhibition of key cytokines, such as interleukin (IL)-4 and IL-5 asthmatic lung tissue and airways caused bronchoconstriction have also been found to have little or no clinical efficacy in through the release of histamine (Schild et al., 1951). Curry asthma patients (Barnes, 2004). This lack of efficacy of single (1946) showed that intravenous and inhaled histamine caused mediator antagonists presumably reflects the multiplicity of bronchoconstriction in patients with asthma but not in normal mediators in asthma and the redundancy of their effects. It subjects. He was thus the first to demonstrate airway hyper- suggests that effective asthma therapies need to have a broad responsiveness in patients with asthma, which is the defining spectrum of anti-inflammatory activity. physiological abnormality of this disease and remains an important target of asthma therapy. All this research suggested that antihistamines might be a useful therapy for asthma. In a Future directions in pharmacological therapy classic paper published in the British Journal of Pharmacology, Ash and Schild showed that the classical antihistamine There has been great pressure from the pharmaceutical mepyramine blocked the contraction of guinea-pig trachea, industry to develop new drugs for asthma as this is an but not the gastric acid secretion, deducing that there were enormous and expanding global market. However, it has different types of histamine receptors (Ash & Schild, 1966; see proved to be a difficult challenge as existing therapies are also Parsons & Ganellin, this issue). Mepyramine blocks highly effective and safe. Combination inhalers with a histamine H1 receptors and antihistamines, such as mepyr- corticosteroid and a long-acting b2-agonist are the most amine and chlorpheniramine, were tested in asthma but with effective treatment so far available and it is likely that several disappointing results. Even with the development of much combination inhalers will become available, including once- more potent nonsedating H1-receptor antagonists, there is no daily drugs. There is a need to find more effective therapies for clinical benefit in patients with asthma. The reason for this patients with more severe asthma, who are not well controlled disappointing finding is that there are other bronchoconstric- by current therapies. Although this is a small minority of tor mediators produced in asthma. patients (o5%) they account for more than half of the health There was particular interest in another mediator released care spending on asthma. New treatments in development for from lungs that was clearly different from histamine. asthma include inhibitors of the proinflammatory enzymes, Kellaway, Feldberg and co-workers demonstrated that such as PDE4, p38 mitogen-activated kinase and nuclear-

British Journal of Pharmacology vol 147 (S1) S302 P.J. Barnes Drugs for asthma factor-kB activating kinase (IKK2) (Barnes, 2004). More 75 years and it will continue to be a vital part of progress in specific approaches include inhibiting chemokine receptors on the future. Human pharmacology studies have been eosinophils and T lymphocytes, inhibiting adhesion molecules of particular value, in view of the disappointing predict- that recruit key inflammatory cells and inhibiting mast cells ability of animal models of asthma. Allergen challenge in with Syk kinase inhibitors. Antibodies that block IgE have asthmatic volunteers has proved to be particularly valuable. now been introduced in some countries and have clinical All of the treatments that have been effective in asthma efficacy, especially in patients with severe allergic asthma. have demonstrated an effect on some aspect of the allergen None of the treatments currently available for asthma change response, whereas drugs that have not shown antiallergic the natural history of the disease or are curative. There is now activity have failed in subsequent clinical trials for interest in vaccination approaches that divert the immune asthma. Novel genes associated with asthma have been system in asthmatic patients back to normal, but the potential identified by molecular genetic techniques, but this has not dangers of such approaches have not been explored. yet resulted in any new treatment approaches. In the future, it may be possible to identify subsets of asthmatic patients through genetic analysis so that more specific therapies can be Final comments administered, but so far pharmacogenetics has had little clinical impact. A more significant contribution of pharmaco- Pharmacology has played an important role in identifying genetics in the future may be in selecting patients that respond and validating new targets in asthma therapy over the last best to more specific therapies.

References

AHLQUIST, R.P. (1948). A study of adrenotropic receptors. Am. J. CURRY, J.J. (1946). The action of histamine on the repiratory tract in Physiol., 153, 586. normal and asthmatic subjects. J. Clin. Invest., 25, 785–791. ASH, A.S.P. & SCHILD, H.O. (1966). Receptors mediating some actions DALE, H.H. & LAIDLAW, P.P. (1910). The physiological action of of histamine. Br.J.Pharmac.Chemother.,27, 427–439. b-iminazoly-ethylamine. J. Physiol., 41, 318–341. BALL, D.I., BRITTAIN, R.T., COLEMAN, R.A., DENYER, L.H., FELDBERG, W. & KELLAWAY, C.H. (1938). Liberation of histamine JACK, D., JOHNSON, M., LUNTS, L.H., NIALS, A.T., SHELDRICK, and formation of lysocithin-like substances by cobra venom. K.E. & SKIDMORE, I.F. (1991). Salmeterol, a novel, long-acting J. Physiol., 94, 187–226. b2-adrenoceptor agonist: characterization of pharmacological FRYER, A.D. & MACLAGAN, J. (1984). Muscarinic inhibitory activity in vitro and in vivo. Br. J. Pharmacol., 104, 665–671. receptors in pulmonary parasympathetic nerves in the guinea-pig. BARGER, G. & DALE, H.H. (1910). Chemical structure and sympatho- Br. J. Pharmacol., 83, 973–978. mimetic action of amines. J. Physiol., 41, 19–59. HANSEL, T.T. & BARNES, P.J. (2002). Tiotropium bromide: a novel BARNES, P.J. (2002). Scientific rationale for combination inhalers with once-daily anticholinergic bronchodilator for the treatment of along-actingb2-agonists and corticosteroids. Eur. Respir. J., 19, COPD. Drugs Today, 38, 585–600. 182–191. HERMANN, G.R., AYNESWORTH, M.B. & MARTIN, J. (1937). BARNES, P.J. (2003). Theophylline: new perspectives on an old drug. Succesful treatment of persistent extreme dyspnea ‘Status asthma- Am. J. Respir. Crit. Care Med., 167, 813–818. ticus’ – use of theophylline ethylene diamine. J. Lab. Clin. Med., 23, BARNES, P.J. (2004). New drugs for asthma. Nat. Rev. Drug Discov., 3, 135–148. 831–844. HIRSCH, S. (1922). Klinischer und experimenteller Beitrag zur BARNES, P.J. & ADCOCK, I.M. (2003). How do corticosteroids work in krampfloseneden Wirkung der Purin Derivativ. Klin. Wochenschr., asthma? Ann Intern. Med., 139, 359–370. 1, 615–618. BARNES, P.J., CHUNG, K.F. & PAGE, C.P. (1998). Inflammatory HOWELL, J.B. & ALTOUNYAN, R.E. (1967). A double-blind trial of mediators of asthma: an update. Pharmacol. Rev., 50, 515–596. disodium cromoglycate in the treatment of allergic bronchial BEEH, K.M., SCHELFOUT, V., GRONKE, F., KANNIESS, F., CAMERON, asthma. Lancet, 2, 539–542. R. & VAN AS, A. (2005). QAB149: the first once-daily b2-agonist with ITO, K., LIM, S., CARAMORI, G., COSIO, B., CHUNG, K.F., ADCOCK, 24-h bronchodilation. Proc. Amer. Thorac. Soc., 2, A356. I.M. & BARNES, P.J. (2002). A molecular mechanism of action of BOARDLEY, J.E., CAREY, R.A. & HARVEY, A.M. (1949). Preliminary theophylline: Induction of histone deacetylase activity to decrease observations on the effect of adrenocorticotropic hormone in inflammatory gene expression. Proc. Natl. Acad. Sci. U.S.A., 99, allergic diseases. Bull. Johns. Hopkins. Hosp., 85, 396–410. 8921–8926. BREWIS, R.A.L. (1990). Classical Papers in Asthma. London: Science KARLSSON, J.A. & PERSSON, C.G. (1981). Influence of tracheal Press. contraction on relaxant effects in vitro of theophylline and BRITTAIN, R.T., FARMER, J.B., JACK, D., MARTIN, L.E. & isoprenaline. Br. J. Pharmacol., 74, 73–79. SIMPSON, W.T. (1968). Alpha-[(t-Butylamino)methyl]-4-hydroxy- LANDS, A.M., ARNOLD, A., MCAULIFF, J.P., LUDUENA, F.P. & m-xylene-alpha 1,alpha 3-diol (AH.3365): a selective beta- BROWN JR, T.G. (1967). Differentiation of receptor adrenergic stimulant. Nature, 219, 862–863. systems activated by sympathomimetic amines. Nature, 214, BROCKLEHURST, W.E. (1960). The release of histamine and formation 597–598. of a slow-reacting substance (SRS-A) during anaphylactic shock. MACHT, D. & TING, G.C. (1921). A study of antispasmodic drugs on J. Physiol., 151, 416–435. the bronchus. J. Pharmacol. Exp. Ther., 18, 373–398. BROWN, H.M., STOREY, G. & GEORGE, W.H. (1972). Beclomethasone MCKENZIE, A.W. & STOUGHTON, R.B. (1962). Method for compar- dipropionate: a new steroid aerosol for the treatment of allergic ing percutaneous absorption of steroids. Arch. Dermatol., 86, asthma. BMJ, 1, 585–590. 608–610. COLLIER, J.G. & DORNHORST, A.C. (1969). Evidence for MEDICAL RESEARCH COUNCIL. (1956). Controlled trial of two different types of beta-receptors in man. Nature, 223, effects of cortisone acetate in status asthmaticus; report to the 1283–1284. Medical Research Council by the subcommittee on clinical trials in COX, J.S.G. (1967). Disodium cromoglycate (FPL 670) (‘Intal’): a asthma. Lancet, 271, 803–806. specific inhibitor of reaginic antibody-antigen mechanisms. Nature, MINETTE, P.A. & BARNES, P.J. (1988). Prejunctional inhibitory 216, 1328–1329. muscarinic receptors on cholinergic nerves in human and guinea- CULLUM, V.A., FARMER, J.B., JACK, D. & LEVY, G.P. (1969). pig airways. J. Appl. Physiol., 64, 2532–2537. Salbutamol: a new, selective beta-adrenoceptive receptor stimulant. MITENKO, P.A. & OGILVIE, R.I. (1973). Rational intravenous doses of Br. J. Pharmacol., 35, 141–151. theophylline. New Engl. J. Med., 289, 600–603.

British Journal of Pharmacology vol 147 (S1) P.J. Barnes Drugs for asthma S303

MURPHY, R.C., HAMMARSTROM, S. & SAMUELSSON, B. (1979). SCHILD, H.O., HAWKINS, D.F., MONGAR, J.L. & HERXHEIMER, H. Leukotriene C: a slow-reacting substance from murine mastocyto- (1951). Reaction of isolated human and asthmatic lung and ma cells. Proc. Natl. Acad. Sci. U.S.A., 76, 4275–4279. bronchial tissue to a specifci antigen. Histamine release and NORRIS, A.A. & ALTON, E.W. (1996). Chloride transport and the muscular contraction. Lancet, 2, 376–382. action of sodium cromoglycate and nedocromil sodium in asthma. SOLIS-COHEN, S. (1900). The use of adrenal substances in the Clin.Exp.Allergy,26, 250–253. treatmentofasthma.JAMA, 34, 1164–1169. PIPER, P.J. & SAMHOUN, M.N. (1982). Stimulation of arachidonic acid ULLMAN, A. & SVEDMYR, N. (1988). Salmeterol, a new long acting

metabolism and generation of thromboxane A2 by leukotrienes B4, C4 inhaled b2-adrenoceptor agonist: comparison with salbuterol in and D4 in guinea-pig lung in vitro. Br.J.Pharmacol.,77, 267–275. adult asthmatic patients. Thorax, 43, 674–678.

British Journal of Pharmacology vol 147 (S1) INTENSIVE CARE

Asthma and chronic Learning objectives obstructive pulmonary After reading this article you should be able to: disease in the intensive C outline the medical management of acute exacerbations of chronic obstructive pulmonary disease (COPD) and asthma care unit C list the appropriate initial investigations of an exacerbation of asthma or COPD, interpret their results, and initiate appropriate Daniel J Garner interventions such as non-invasive ventilation (NIV) C understand the limitations of NIV and describe features in David Tuxen patients who are failing treatment or likely to fail treatment with NIV C list specific complications associated with intubation and mechanical ventilation in patients with these conditions, and Abstract know how to appropriately manage these complications if they There are many pitfalls in the management of patients with asthma or arise chronic obstructive pulmonary disease, especially when their condition C utilize appropriate ventilation strategies to minimize compli- becomes severe enough to warrant intensive care unit level care. cations commonly associated with the intubated COPD or Mortality in both groups remains significant. Standard principles of asthma patient. oxygen and drug administration and mechanical ventilation technique used for typical critically ill patients can all cause problems. Recognition of the presence of airflow obstruction, the potential for dynamic hyperinflation and careful alteration of the principles and targets of therapy are approximately 1500 (0.3%) deaths per year in England and required to avoid complications. In this article we examine the nature and Wales attributed to acute exacerbations. Acute asthma attacks scope of the challenge facing intensivists, highlighting difficulties in are attributed to a variety of causes including antigen exposure, management and outlining specific strategies to aid in managing both non-specific irritants (cold air, smoke, pollution), anxiety and conditions. unknown factors (up 25% of acute attacks). Guidelines written and published by various expert bodies Keywords Asthma; COPD; dynamic hyperinflation; invasive mechanical (British Thoracic Society (BTS), National Asthma Council of ventilation; non-invasive ventilation Australia) have been well implemented in primary care, leading to a diminishing burden on acute hospital services. These guidelines can easily be accessed online and are simple to understand and implement. Their key components include earlier recognition of asthma severity, more widespread use of inhaled steroids, and pre-set plans to ensure prompt treatment of Demographics and medical treatment exacerbations. Asthma and chronic obstructive pulmonary disease (COPD) are There are comparable and excellent resources guiding the conditions characterized by airflow limitation which when medical management of COPD (Global Initiative for Chronic sufficiently severe, mandates critical care management. Mortality Obstructive Lung Disease, BTS). In either case it is important that from COPD is increasing worldwide, exacerbations of COPD are these medical measures are instituted as soon as an exacerbation triggered by infection (50%) and episodes of heart failure (25%) of either condition is identified. but in others no obvious cause is identified. COPD also commonly presents as a co-morbidity of another illness where mechanical ventilation is required, such as major elective or Involvement of critical care services and the use of non-invasive emergency surgery, or trauma. ventilation (NIV) There is also an increasing prevalence of asthma amongst Most referrals to critical care services centre on requests for populations worldwide, with significant morbidity and impact on consideration of ventilatory support in patients with severe quality of life.1 Mortality accounts for approximately 1 in 250 presentation or patients who fail to improve despite optimal deaths worldwide.1 The prevalence of asthma is highest amongst medical therapy. Non-invasive ventilation (NIV) has become first world countries.1 Mortality remains significant with a care standard in the management of acute exacerbations of COPD, with good evidence proving its role.2 A large Cochrane meta-analysis demonstrated that NIV significantly reduced Daniel J Garner MB BCHIR MRCP is a Clinical Research Registrar in mortality and invasive ventilation rates and led to a shortened Respiratory Medicine at The Alfred Hospital, Melbourne, Australia. hospital length of stay. NIV improves respiratory physiology with 2 Conflicts of interest: none declared. improvements in both PaO2 and PaCO2. Excellent, concise specific NIV guidelines exist,3 focusing on the targeted use of NIV David Tuxen MBBS FRACP Dip DHM MD FJFICM FCICM is a Consultant in in patients with exacerbations of COPD and concomitant respi- Intensive Care at The Alfred Hospital, Melbourne, Australia. Conflicts of ratory acidosis, which has failed to respond to optimal medical interest: none declared. therapy.

Acute hypercapnic respiratory acidosis may also be triggered (increasing respiratory acidosis, patient exhaustion, hypoxia, or worsened by excess oxygen therapy in a patient subset whose reduced conscious level or respiratory arrest). The life expec- chronic relative hypoxaemia is essential for stable ventilatory tancy in patients with COPD falls with deteriorating FEV1 and function. The majority of patients with severe COPD (FEV1 episodes of acute respiratory failure. Of all patients with COPD, <30% predicted), especially those with pre existing hypercapnia those admitted to ICU for invasive ventilatory support have the are at risk of an increasing PaCO2 when oxygen therapy results in highest mortality, even those patients who survive to discharge SpO2 >95%. This is due to a combination of factors (Figure 1). from ICU have a decreased life expectancy in comparison to the Improvement may occur with titration of FiO2 to a SpO2 target of general COPD population. Careful consideration of any course of 88e92%. If an oxygen-induced rise in PaCO2 has occurred in invasive treatment needs to be undertaken in patients who have a patient with an exacerbation of COPD then hypercapnia may severe COPD and end-stage lung disease (often in association improve but not resolve when FiO2 is reduced, such that NIV with significant co-morbidity). This assessment should include may still be necessary e hence the need for constant monitoring age, severity of COPD, the presence of a reversible component and regular review (Table 1). and functional status. Functional status prior to the exacerbation The aims of NIV are to unload respiratory muscles, whilst is perhaps the most important factor; this often-omitted infor- augmenting ventilation and oxygenation, also offsetting the mation should be carefully collected. Patients with severely adverse effects of sleep on ventilation and airway resistance limited mobility (housebound or worse) and compromised (Table 2). self-care (e.g. showering, dressing) may require limitation of full There is a growing consensus, but no strong supporting intensive care management. This limitation is most commonly to evidence, that the judicious use of NIV in a critical care setting provide active non-invasive care including NIV, all drug, fluid, for severe acute asthma can obviate the need for invasive nutritional and physical therapies but not invasive ventilation. It ventilation.4 may also include limitations on other life supports systems such NIV services will vary between hospitals, with emergency as renal replacement therapy and inotropic support. departments and respiratory physicians increasingly imple- menting treatment outside intensive care unit (ICU) settings. The Ventilation and dynamic hyperinflation application of NIV is a significant intervention, patients may The primary challenge in the ventilation of patients with COPD deteriorate in spite of its institution, the evidence suggests it is or severe asthma is that of dynamic hyperinflation (DHI). best delivered in a dedicated critical care environment, where Incomplete expiration of inspired gas (gas trapping), due to slow experienced staff can easily escalate care as required. In high-risk expiratory airflow and airway closure, leads to increased end- groups (Table 2) this type of setting is mandated. expiratory lung volume.5 When DHI is excessive, this can lead to Despite optimal medical management and the correct usage of compromise of cardiac output, circulatory collapse and increase NIV, a proportion of patients with COPD and acute asthma will the risk of pneumothorax.5,6 The three primary determinants of still require endotracheal intubation for their respiratory failure DHI are (i) the severity of airflow obstruction (the expiratory

Mechanism contributing to rising PaCO2

Attenuation of hypoxic Oxygen-induced vasoconstriction atelectasis

Haldane effect Decreased central respiratory drive

Anxiolysis

Figure 1 Mechanisms contributing to rising PaCO2 as a result of excessive oxygen therapy, ﬁve major causes in red.

Comparison of medical treatments of exacerbations of COPD and acute severe asthma

Treatment Severe COPD* Severe Asthma** Inhaled B2 Agonists Frequently administered, may need to be driven Driven with oxygen and given “back to back” with air Inhaled anticholinergics Ipratropium bromide 500 mcg qid Ipratropium bromide 500 mcg qid

Controlled oxygen Important to titrate FiO2 to SpO2 (aim 88–92%), Usually no adverse effects of high flow O2, therapy monitoring of ventilation and PaCO2 important hypoxia is a late sign of severe respiratory

failure and the impending need for ventilatory OTADNAM support R Y Glucocorticoids Hydrocortisone 100–200 mg IV stat then 100 Hydrocortisone 100–200 mg IV stat then 100 mg qid, to ensure adequate/rapid dosing mg qid, to ensure adequate/rapid dosing Majority of sputum cultures negative, but Small minority of severe acute asthma triggered commonly of benefit if features of infective by bacterial infection – only indicated if Antibiotics exacerbation (fever, increasing sputum volume, compelling features of infection e.g. fever, purulent sputum) neutrophilia, consolidation Magnesium infusion Lacks evidence but unlikely to cause adverse Lacks evidence but unlikely to cause adverse effects effects O

Intravenous theophylline Effective bronchodilator but side effects close Effective bronchodilator but side effects close TP to clinical range and uncertain additional to clinical range and uncertain additional benefit in presence of optimal beta agonists benefit in presence of optimal beta agonists and

and steroids. Can be tried if bronchospasm steroids. Can be tried if bronchospasm LANOI refractory to mandatory treatment refractory to mandatory treatment Intravenous B2 agonists Rarely used Can be used if asthma refractory but evidence lacking and associated with increased risk of lactic acidosis Advanced care Careful assessment of pre-morbid functional Usually not required directives state and patient wishes should be undertaken

Table 1

time constant) (ii) the volume inspired (Vt) and (iii) the time for excessive, external PEEP (5e10 cmH2O) may reduce the work of expiration (Te), determined by both the respiratory rate and the breathing for spontaneous breaths. inspiratory flow (or I:E ratio). Auto-PEEP or intrinsic PEEP (PEEPi) can be assessed during an In severe asthma, initial ventilation should target low minute end-expiratory pause (0.5 seconds) (Figure 2). This is useful to ventilation (<115 ml/kg/minute), low tidal volumes (<8 ml/kg) detect the presence of airflow obstruction, however it is known to and low respiratory rate (<14 breaths/minute), high inspiratory underestimate the severity of DHI and to be insensitive to change. flow rates (>70 litres/minute) with controlled ventilation in This is because PEEPi only measures the pressure in lung units still synchronized intermittent mandatory ventilation (SIMV) mode. communicating with the central airways at the end of expiration, Externally applied positive end-expiratory pressure (PEEP), and cannot measure the pressure in more severely obstructed, high during controlled ventilation, worsens hyperinflation and should pressure lung units that have airway closure during expiration. not be used. Significant sedation and in some patients, inter- A better measure of DHI is the plateau airway pressure (Pplat) mittent neuromuscular blockade, may be required to achieve measured during an end-inspiratory pause. This manoeuvre is best these targets to control DHI. Hypercapnia is not a reason to performed after 0.5 seconds of end-inspiratory pause and should increase ventilation as this will increase DHI. Although muscle target a Pplat <25 cmH2O. If the application of an end-inspiratory relaxants may be required, their use should be minimized as they pause reduces the subsequent expiratory time, it is important to risk acute necrotizing myopathy which can prolong weaning avoid applying this pause to sequential breathes as this will worsen from ventilation. Attempts to limit high peak inspiratory pressure DHI. by decreasing inspiratory flow rate will paradoxically increase If the Pplat is <25 cmH2O then the ventilator rate can be the risk of barotrauma through decreasing expiratory time (lower increased (provided Pplat remains <25 cmH2O) and, when I:E ratio), worsening DHI.5 In severe COPD, the same ventilation appropriate, sedation decreased and spontaneous ventilation principles apply but in these patients it is usually safe to allow commenced. If the Pplat is >25 cmH2O or if there is hypotension spontaneous ventilation within SIMV mode (e.g. controlled rate present that responds to transient ventilator disconnection then <10). When spontaneous breathing is occurring and DHI is not DHI should be reduced by decreasing the respiratory rate,

ensuring the inspiratory flow rate is high and accepting the Indications, complications and contraindications for the presence of significant respiratory acidosis. The Pplat should be application of NIV in COPD rechecked after these adjustments to ensure it has decreased. Uncommonly, despite optimal management and profound Indications for NIV in COPD hypoventilation, some patients may continue to have excessive Respiratory rate >28/minute DHI resulting in refractory hypotension. This mainly applies to þ 8 PaCO2 >45 mmHg (6 kPa) with pH<7.35 (H 4.46710 ) e not patients with severe asthma and is rare in COPD. In this situa- due to excessive supplemental oxygen therapy tion, special measures may be helpful; these include adminis- Respiratory distress with impending fatigue tering intravenous adrenaline and/or intravenous magnesium and/or introducing helium (improved laminar flow of gases) or Cautious use of NIV i.e. best undertaken in an ICU/critical care anaesthetic gases (direct bronchodilatory effect) into the venti- setting lator circuit. Patients in extremis may require extra corporeal pH<7.25 (Hþ 5.623108) membrane oxygenation (ECMO) to achieve adequate gas Mechanical ventilation inappropriate exchange when even minimal ventilation still results in circula- Drowsiness/reduced conscious level, secondary to respiratory tory failure and/or insufficient gas exchange, though this failure remains the preserve of paediatric practice in all but exceptional Hypoxaemia e failure is common if in combination with normo- circumstances. However not all centres may be able to offer such capnia or hypocapnia specialist intervention and the only practical option may be to After extubation e mechanical ventilation to assist in ‘stepping tolerate profound hypoventilation with muscle paralysis. down’ (Figure 3 e effects of ventilatory strategies). Possible NIV complications The particular challenges of trying to achieve ventilator targets Pressure ulcers/necrosis from mask e compliance may be improved in DHI lead to a constellation of specific problems including with full face, helmet or other mask interfaces myopathy, the risk of barotrauma/pneumothorax and circulatory Gastric distension/aspiration collapse with pulseless electrical activity (PEA). These in turn may Discomfort and intolerance increase the number of days a patient is mechanically ventilated Contraindications and affect how easily they wean from the ventilator. Coma/severe encephalopathy Cardiac/respiratory arrest or severe haemodynamic instability Pneumothorax Facial surgery/trauma Though spontaneous pneumothorax occurs more commonly Inability to clear secretions, tolerate mask or protect airway amongst the asthma and COPD populations it rarely occurs Severe GI bleeding or vomiting during or as the precipitant of an acute exacerbation. Barotrauma during an exacerbation of asthma or COPD is usually from COPD, chronic obstructive pulmonary disease; GI, gastrointestinal; ICU, intensive care unit; NIV, non-invasive ventilation. excessive DHI or needle introduction to the thorax (attempts at gaining central venous access or attempted treatment of sus- Table 2 pected tension pneumothorax). Tension pneumothorax is the rule rather than the exception as airflow obstruction itself acts as a valve. This usually presents during mechanical ventilation with a worsening of respiratory state including hypoxia, increased airway pressures, hypotension and sometimes circulatory collapse or even cardiac arrest. Furthermore, a tension pneumothorax on one side can redistribute ventilation to the contra- lateral lung, worsening DHI within the second lung and risking the development of bilateral pneumothoraces. Thus, when a pneumothorax is suspected and hypotension is absent or mild, the ventilator rate should be decreased (to protect the second lung) and an urgent chest X-ray obtained to ensure a pneumothorax is present before a careful blunt dissection intercostal catheter is inserted. If severe hypotension mandates needle decompression using a vascular cannula in the intercostal space, a formal blunt dissection intercostal catheter should always follow, as the attempted needle decompression will always produce a pneumothorax if one was not already present.

Figure 2 Ventilator trace during severe airﬂow limitation. Note the small Pulseless electrical activity (PEA) and cardiac arrest tidal volume, that high inspiratory ﬂow (Vi) results in high measured peak Tension pneumothorax is a commonly recognized cause of circu- airway pressures, and that the low respiratory rate in combination with the high Vi results in a long expiratory time. However the ventilatory latory collapse and cardiac arrest in the ventilated COPD or asthma pattern has resulted in a comfortably safe Pplat during an end-inspiratory patient. However DHI, even with very low minute ventilation, can pause (white section of pressure curve). lead to cardiac arrest with PEA, sometimes after only a very short

Effects of ventilatory strategies

Figure 3 Strategy 1: Ventilating to a normal PaCO2 (40 mmHg) leads to DHI with dangerous lung volumes and their associated complications (red shaded area). Strategy 2: Marked hypoventilation eliminates the problems of strategy 1 but leads to severe respiratory acidosis and requires a significant sedation/paralysis associated with complications (blue shaded area). Strategy 3: Ventilation just below the safety limit (Pplat 25 cmH2O) results in only 5 mild-moderately elevated PaCO2 and minimizes the complications associated with the alternative strategies. period of mechanical ventilation. It is important to recognize to ventilator support progressively reduced until a trial of extuba- avoid well-intentioned but unnecessary interventions aimed at tion can be considered. This usually happens at the commence- other causes of PEA (pericardiocentesis, chest decompression etc), ment of mechanical ventilation in COPD but commonly requires which may complicate the situation further. Disconnecting the significant improvement in airflow in asthma. The application of ventilator for over 1 minute or ventilating with fewer than four NIV immediately following extubation in COPD may reduce the breaths per minute should be performed in this circumstance. If this need for reintubation and ongoing mechanical support but is manoeuvre results in spontaneous return of circulation, it is likely usually not required in asthma except where there is a large that DHI was the underlying cause and ventilation should be irreversible component. resume, but at a much lower rate and tidal volume. Although overfeeding can lead an increase in CO2 production potentially impairing weaning, the long term benefits from adequate Myopathy nutrition are important on COPD and should be given priority. Tracheostomy is rarely required in patients with asthma but The need for profound hypoventilation and muscle paralysis may be required in patients with COPD who are still ventilator combined with the usual regime of significant glucocorticoid dependent after 7e10 days. Extubation to NIV support can be usage may lead to an acute necrotizing myopathy consisting of successful and may be worth trying before performing a trache- subjective and objective weakness, EMG changes, increased CK ostomy. A tracheostomy is better tolerated than naso/orotracheal and characteristic biopsy findings. The length of ventilation and tube with patients requiring less or no sedation. Other benefits the usage of neuromuscular blockers strongly correlate to the include reduction in airway resistance to facilitate weaning, phenomenon and emphasize the need to avoid excessive seda- easier upper and lower airway toileting, easier swallowing and tion and paralysis where possible. speaking, and facilitates ICU discharge. In a small minority of patients weaning from mechanical Lactic acidosis ventilation may not be possible despite optimization of all The use of salbutamol in frequent schedules, high doses or reversible factors, in these cases long-term institutional or home intravenously carries a significant risk of lactic acidosis. Serum ventilation is an option if the service is available. However the lactate levels of 4e12 mmol/litre are not uncommon and at these huge impact of such a measure may represent such a poor quality levels significantly add to respiratory distress e infusions should of life that this option may not be desirable. be limited to doses below 10 mg/minute and stat doses to less than 250 mg. Ceasing infusions for 4e6 hours will usually lead to Post ICU care resolution of the problem. An admission to ICU and the requirement for non-invasive or invasive ventilation represent a life-threatening event in both Ventilator management and weaning asthma and COPD. It should act as a trigger to review and opti- When DHI becomes manageable, spontaneous ventilation should mize all aspects of an individual’s primary and preventative care. be encouraged to promote respiratory muscle activity and In both asthma and COPD this will usually include regular

ANAESTHESIA AND INTENSIVE CARE MEDICINE 11:11 485 Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved. INTENSIVE CARE long-acting bronchodilators and inhaled steroids. In both groups by recognizing patients with severe airflow limitation and initi- a plan for early recognition and treatment of future exacerbations ating appropriate treatment strategies, improved outcomes are should be given to the patient prior to discharge. This plan may achievable. A include assessment of peak expiratory flow, additional inhaled bronchodilators, and possibly immediate commencement of pre- prescribed oral steroids or antibiotics, and a trigger level and path for medical presentation. In COPD pulmonary rehabilitation REFERENCES programs improve physiology, lung function and quality of life 1 GINA. Global initiative for asthma (GINA) report, Global burden of whilst reducing hospitalization rates. Influenza and pneumo- asthma; 2010. coccal vaccines should be given to asthmatics and those with 2 Lightowler JV, Wedzicha JA, Elliott MW, Ram FS. Non-invasive positive COPD when well. Lung reduction surgery and lung trans- pressure ventilation to treat respiratory failure resulting from exacer- plantation may also be considered for a select population of those bations of chronic obstructive pulmonary disease: cochrane system- with advanced COPD. atic review and meta-analysis. BMJ 2003; 326: 185. 3 Roberts CM, Brown JL, Reinhardt AK, et al. Non-invasive ventilation in Prognosis chronic obstructive pulmonary disease: management of acute type 2 Survival at 180 days in patients admitted to ICU for COPD respiratory failure. Clin Med 2008; 8: 517e21. exceeds 60%. Importantly, the majority (73%) of these patients 4 Murase K, Tomii K, Chin K, et al. The use of non-invasive ventilation for reported equivalent or improved quality of life, of survivors 96% life-threatening asthma attacks: changes in the need for intubation. would choose similar treatment again. Features such as, organ Respirology 2010; 15(4): 714e20. dysfunction other than respiratory failure, advanced disease with 5 Tuxen DV, Lane S. The effects of ventilatory pattern on hyperinflation, severe respiratory dysfunction and prolonged hospital stay prior airway pressures, and circulation in mechanical ventilation of to admission to ICU are associated with worse outcomes. In the patients with severe air-flow obstruction. Am Rev Respir Dis 1987; longer term outcome data suggest that COPD combined with 136: 872e9. severe respiratory failure has 1- and 5-year survival rates 6 Williams TJ, Tuxen DV, Scheinkestel CD, Czarny D, Bowes G. Risk comparable to many solid organ tumours, with studies reporting factors for morbidity in mechanically ventilated patients with acute e e 7 21 39% mortality at 1 year and 55 76% mortality at 5 years. severe asthma. Am Rev Respir Dis 1992; 146: 607e15. Less than 0.2% of hospital attendances for asthma result in 7 McGhan R, et al. Predictors of rehospitalization and death after death, risk factors statistically associated with such a cata- a severe exacerbation of COPD. Chest 2007; 132: 1748e55. strophic outcome are a history of recurrent admissions for 8 Stow PJ, et al. Improved outcomes from acute severe asthma in asthma and a prior episode of asthma requiring mechanical Australian intensive care units (1996e2003). Thorax 2007; 62: ventilation. 842e7. Recent data from Australia indicate that around one-third of patients presenting with severe asthma require mechanical venti- FURTHER READING/USEFUL RESOURCES lation. Among these the mortality rate has fallen progressively British Thoracic Society, www.brit-thoracic.org.uk. (from 10e3% amongst ventilated patients over the eight years to David V Tuxen, Matthew T Naughton. Acute respiratory failure in chronic 20038), other Australian data show that in-hospital mortality rates obstructive pulmonary disease and acute severe asthma. In: Andrew D for hypercapnic COPD patients can be as low as 11%, with all Bersten, Neil Soni, eds. Oh’s intensive care manual, 6th edn. Elsevier: deaths occurring with palliative intent. These results suggest that Butterworth Heinemann, 2008.

The pulmonary physician in critical care • 12: Acute severe asthma in the intensive care unit

P Phipps and C S Garrard

Thorax 2003;58;81-88 doi:10.1136/thorax.58.1.81

Updated information and services can be found at: http://thorax.bmj.com/cgi/content/full/58/1/81

These include: References This article cites 101 articles, 47 of which can be accessed free at: http://thorax.bmj.com/cgi/content/full/58/1/81#BIBL

3 online articles that cite this article can be accessed at: http://thorax.bmj.com/cgi/content/full/58/1/81#otherarticles Email alerting Receive free email alerts when new articles cite this article - sign up in the box at the service top right corner of the article

Topic collections Articles on similar topics can be found in the following collections

Critical Care / Intensive Care (304 articles) Asthma (1218 articles)

Notes

To order reprints of this article go to: http://journals.bmj.com/cgi/reprintform To subscribe to Thorax go to: http://journals.bmj.com/subscriptions/ Downloaded from thorax.bmj.com on 21 August 2007

REVIEW SERIES The pulmonary physician in critical care c 12: Acute severe asthma in the intensive care unit P Phipps, C S Garrard ......

Thorax 2003;58:81–88 Most deaths from acute asthma occur outside hospital, within a year.25 Interestingly, the association but the at-risk patient may be recognised on the basis of between asthma deaths and β agonist use is still debated and there has been concern that the use prior ICU admission and asthma medication history. of long acting β agonists may increase asthma Patients who fail to improve significantly in the mortality.6 This has not been confirmed in studies emergency department should be admitted to an HDU monitoring their use.78In contrast, there is a consensus that underuse of anti-inflammatory treat- or ICU for observation, monitoring, and treatment. ment in the period leading up to the acute severe Hypoxia, dehydration, acidosis, and hypokalaemia attack worsens prognosis.9 render the severe acute asthmatic patient vulnerable to Unfortunately, a proportion of asthmatic patients die despite reaching hospital alive. Such cardiac dysrrhythmia and cardiorespiratory arrest. deaths can usually be attributed either to Mechanical ventilation may be required for a small inadequate observation or treatment. Sadly, a proportion of patients for whom it may be life saving. number of patients suffer unobserved respiratory failure that progresses to cardiac arrest and Aggressive bronchodilator (continuous nebulised β anoxic brain damage.10 A small number of agonist) and anti-inflammatory therapy must continue patients are resistant to the most aggressive treatments and interventions. throughout the period of mechanical ventilation. Necroscopic studies of patients dying of acute Recognised complications of mechanical ventilation severe asthma have found extensive mucus plug- include hypotension, barotrauma, and nosocomial ging of bronchi that has been termed “endobronchial mucus suffocation” (fig 1).11 Microscopic pneumonia. Low ventilator respiratory rates, long examination reveals extensive inflammatory expiratory times, and small tidal volumes help to prevent changes that involve all airway wall components hyperinflation. Volatile anaesthetic agents may produce and the pulmonary arterioles.12 13 The degree of β bronchial occlusion is much greater than in con- bronchodilation in patients resistant to agonists. trol asthmatic subjects, with mucus, desqua- Fatalities in acute asthmatics admitted to HDU/ICU are mated epithelium, inflammatory cells and plasma 11 rare. exudate all contributing. Sudden asphyxic asthma may be a distinct pathological subtype in ...... which intense bronchoconstriction causes respiratory failure, often over the course of 1–2 14 ost asthma exacerbations are managed in hours. Recovery appears to be rapid, which the community or emergency department suggests that bronchoconstriction may be the 15 Mwhile the more severe cases that fail to predominant pathophysiological factor. respond to bronchodilator and anti-inflammatory therapy require admission to high dependency INTENSIVE THERAPY AND MONITORING (HDU) or intensive care units (ICU). Patients who fail to improve with optimal medical Worldwide asthma prevalence is increasing, treatment in the emergency department should and with that the total number of admissions to be considered for HDU or ICU admission to facili- hospital and intensive care. Although the time tate continuous monitoring of physiological pa- between the onset of symptoms and the require- rameters such as pulse oximetry, ECG, and ment for ventilation is becoming shorter, the out- arterial and central venous pressure. Equipment come is improving with fewer deaths and lower and experienced staff are also available for urgent complication rates.1 procedures such as endotracheal intubation or the insertion of thoracostomy tubes.

MORTALITY Clinical, physiological and laboratory See end of article for A history of mechanical ventilation or ICU assessment authors’ affiliations admission is a well documented indicator of sub- The immediate assessment of patients with ...... sequent near fatal asthma.23Women and smokers asthma should include the degree of respiratory Correspondence to: are also over-represented in both life threatening distress (ability to speak, respiratory rate, use of 3–5 Dr C S Garrard, Intensive attacks and asthma deaths. It is believed that accessory muscles, air entry), degree of hypoxia Care Unit, John Radcliffe patients who have had a life threatening attack (cyanosis, pulse oximetry, level of consciousness), Hospital, Oxford and those who die are from a similar demo- and cardiovascular stability (arrhythmias, blood OX3 9DU, UK; graphic group. In a large study of patients admit- pressure). Accessory muscle use, wheeze, para- chris.garrard@ clinical-medicine.oxford.ac.uk ted with a near fatal episode, two thirds of subse- dox, and tachypnoea may diminish as the patient ...... quent severe attacks or deaths had occurred tires.16

www.thoraxjnl.com Downloaded from thorax.bmj.com on 21 August 2007

82 Phipps, Garrard

Expiration Inspiration A Residual expiratory flow at onset of inspiratory effort

0 Flow _

Onset of inspiratory effort B Onset of inspiratory flow

PEEPi Figure 1 Mucus cast of bronchial tree coughed up by an asthmatic + patient during an exacerbation. Reproduced with permission of E Klatt, Utah.

Poes 0

_ Forced expiratory volume in 1 second (FEV1) and peak expiratory flow rate are the most used and convenient measures of airflow obstruction that support clinical findings and quantify the response to treatment.17 Occasionally patients are too distressed to perform forced expiratory manoeuvres or C there is a risk of precipitating further bronchoconstriction.18 Expiratory airflow limitation results in a dynamic increase in + end expiratory lung volume which interferes with inspiratory muscle function, both of the diaphragm and the chest wall. 0 The positive alveolar pressure at end expiration (PEEP) due to residual elastic recoil has been termed intrinsic (PEEPi) and

Lung volume _ its presence is suggested by residual expiratory flow at the onset of inspiration. In spontaneously breathing patients the Increasing FRC magnitude of PEEPi can be estimated by the change in intrapleural pressure (usually measured by an oesophageal Time pressure probe) between the onset of inspiratory effort and the onset of inspiratory flow (fig 2). The work of breathing is Figure 2 (A) Schematic diagram of an asthmatic patient exhibiting significant residual flow at end expiration. (B) The oesophageal increased in the presence of PEEPi because the residual alveo- pressure (Poes), an estimate of intrapleural pressure, shows the lar pressure must be overcome by muscle effort before inspira- degree of pressure change required to overcome intrinsic pressure tory flow commences. Many patients also use expiratory mus- (PEEPi) and initiate inspiratory flow. (C) A progressive increase in cles to aid expiration, which may paradoxically worsen lung volume (breath stacking) occurs if expiratory time is insufficient dynamic airway collapse and PEEPi. Inspiratory muscle activ- to allow complete exhalation of the tidal volume. ity may also persist during expiration,19–21 which contributes to increased expiratory work of breathing. In mechanically ven- There are other causes of wheeze and respiratory distress tilated paralysed patients the magnitude of PEEPi is estimated that must be considered in the differential diagnosis. These by performing an end expiratory breath hold and measuring include left ventricular failure, upper airway obstruction, the airway pressure with reference to that of the atmosphere. inhaled foreign body, and aspiration of stomach contents. In the presence of PEEPi a short expiratory time may lead to “breath stacking” and progressive hyperinflation as the next breath is initiated before the previous tidal volume has been Treatment completely exhaled. The pathological effects of PEEPi include Intensive care treatment of the poorly responsive asthmatic hypotension due to reduced venous return and an increased patient should include high concentrations of inspired 22 β risk of pneumothorax. oxygen, continuous nebulisation of agonists, intravenous 27–29 Pulse oximetry is an invaluable adjunct to monitoring corticosteroids, and respiratory support. Clinicians must be since the avoidance or abolition of hypoxia is a prime goal of aware of the need to optimise oxygenation and avoid treatment. Regular arterial blood gas measurements provide dehydration and hypokalaemia. Unrestricted high concentra- a measure of gas exchange and facilitate the monitoring of tions of oxygen (60–100%) must be administered to abolish 27 30 serum potassium levels. The arterial carbon dioxide tension hypoxaemia, unlike the patient with chronic obstructive lung disease where controlled limited oxygen is indicated. (PaCO2) and acid-base status help to identify the presence of pre-existing respiratory or metabolic acidosis, and the trend Hypokalaemia is common and may be exaggerated by fluid resuscitation and the administration of β agonist bronchodila- in PaCO2 is helpful when assessing the response to treatment. The degree of hypokalaemia and lactic acidosis may also tors. Repeated infusions of potassium chloride may be guide treatment.23 A chest radiograph is indicated to identify required with careful monitoring of serum levels and continu- pneumothorax, areas of segmental or lobar collapse, or infil- ous ECG monitoring. trates that may suggest pneumonia. However, the yield is low.24 25 An ECG may detect myocardial ischaemia or identify Specific asthma drug treatment arrhythmias, especially in older patients.26 Right axis devia- On admission to the ICU there should be a rapid review of tion and right heart strain are common findings. Potassium, earlier asthma treatment to identify elements that can be magnesium, calcium and phosphate deficiencies should be intensified or deficiencies remedied. Drugs contraindicated in corrected to reduce the risk of arrhythmia and respiratory asthma include β blockers, aspirin, non-steroidal anti- muscle weakness. inflammatory drugs, and adenosine.

www.thoraxjnl.com Downloaded from thorax.bmj.com on 21 August 2007

Acute severe asthma in the intensive care unit 83

Corticosteroids Box 1 Contraindications to a trial of mask CPAP or Evidence continues to accumulate that early treatment with NIV in acute severe asthma adequate doses of corticosteroid improves outcome in severe acute asthma. There does not appear to be any benefit from • Need for immediate endotracheal intubation high doses of hydrocortisone exceeding 400 mg/day, and no • Poor patient cooperation particular advantage of the intravenous over the oral route • Inability of the ventilator to supply high FiO 31 2 provided there is reliable gastrointestinal absorption. Inhaled • Hypercapnia (CPAP less likely to benefit than NIV) corticosteroids have not been fully evaluated in this setting. • Excess respiratory secretions • Lack of experienced staff and/or a high dependency area β agonists Salbutamol (albuterol in North America) is the most commonly prescribed β agonist for the treatment of acute Ipratropium asthma.29 It appears to be more effective and induces less Ipratropium bromide has a mild additional bronchodilating hypokalaemia when delivered by the inhaled route, although effect when added to β agonists that may only be significant in there is a theoretical rationale for administering salbutamol severe asthma.46 The safety profile and the fact that individual intravenously to bypass obstructed airways.32 Concerns over patients may obtain benefit have resulted in aerosolised iprat- blood levels and potential cardiotoxicity may inhibit more ropium (500 µg 6 hourly) being recommended for the aggressive use of nebulised salbutamol.33 However, continuous treatment of acute severe asthma.47 administration of nebulised salbutamol in doses approaching 20 mg/h can achieve bronchodilation without toxicity; indeed, Aminophylline the patient’s heart rate may fall with alleviation of airway 34 The addition of aminophylline does not add to the bronchodi- obstruction. lating effect of optimal doses of β agonists.48 Other reported Interpretation of the literature on continuous nebulised benefits of aminophylline such as improving diaphragmatic salbutamol is hampered by differences in the definitions of endurance, stimulating ventilatory drive, and anti- “continuous” (length of time) and the delivered doses used in inflammatory effect do not seem to improve outcome in acute individual studies. One study compared 27.5 mg salbutamol severe asthma.49–53 Currently, aminophylline is not recom- by either continuous or intermittent nebulisation over 6 mended as a first line drug in acute asthma management and hours35 and, not surprisingly, showed little difference between its inclusion as a second line agent is still debated.46 However, intermittent and continuous regimens. Another study sug- when other agents fail to achieve bronchodilation, aminophyl- gested benefit from prolonged continuous aerosol use only in severe asthma.36 Doses of 0.3 mg/kg/h nebulised salbutamol line can be used providing dosing regimens are adhered to. Typically, a loading dose of 5 mg/kg by slow intravenous infu- have been safely used in children without significant µ toxicity.37 Higher inhaled doses may be required in mechani- sion over 20 minutes is followed by an infusion of 500 g/kg/ cally ventilated patients due to aerosol losses in the ventilator h. If prolonged administration is required, daily monitoring of circuit. blood theophylline levels is essential. The therapeutic range is µ Salbutamol delivered by metered dose inhaler (MDI) with a 55–110 mol/l (10–20 mg/l). spacer device is at least as effective as nebulised drug in the management of acute asthma in the emergency ASSISTED VENTILATION department.38 In patients with hypoxia and respiratory If there is inadequate response to drug treatment or if the distress, however, nebulised drugs may be easier to administer. patient is in extremis at presentation, mechanical ventilation Even intranasal and intratracheal instillation of β agonists may be required. may be effective in an emergency situation.39 40 Salbutamol is currently a racemic mixture of R and L forms. The Mask continuous positive airway pressure (CPAP) and S-enantiomer does not have β agonist effects and competes for non-invasive ventilation (NIV) the binding sites of the R form, levosalbutamol.41 Formula- In spontaneously breathing patients the application of low tions of levosalbutamol have fewer side effects and greater levels of mask CPAP (3–8 cm H2O) may improve respiratory efficacy than the racemic mixture, suggesting that it may be rate, dyspnoea, and work of breathing in asthma, particularly preferable in acute severe asthma.42 Terbutaline is an alterna- if there is evidence of smoking related lung disease.20 54 55 There tive β agonist that has been less widely studied but is effective is a danger that CPAP may worsen lung hyperinflation. If by the inhaled, intravenous, and subcutaneous routes. Long patients are intolerant of the mask or do not derive benefit, acting β agonists such as salmeterol and eformoterol are not CPAP should be withdrawn. In hypercapnic patients CPAP recommended because of their slow onset of action. alone may not improve ventilation. Both nebulised and intravenous adrenaline (epinephrine) Few studies have looked specifically at NIV in asthma. Low are effective in the treatment of acute asthma. The putative levels of CPAP and pressure support of 10–19 cm H2O in acute benefits of the α-adrenergic component of adrenaline include severe asthma improved gas exchange and prevented endotra- reduced microvascular permeability and airway wall oedema, cheal intubation in all but two of 17 hypercapnic patients.56 and less impairment of ventilation/perfusion matching than However, the rate of intubation in patients with acute asthma, with more selective β agonists.43 However, there does not even in the presence of hypercapnia, is low at 3–8%.28 57 It is appear to be any clinical benefit over β agonists such as reasonable to give asthmatic patients a trial of NIV over 1–2 salbutamol.44 hours in an HDU or ICU if there are no contraindications (box Some of the metabolic and cardiovascular complications of 1).56 Deciding when to initiate NIV, when a trial of NIV has acute severe asthma may be exacerbated by high dose β ago- failed, and optimising NIV in this setting require considerable nist therapy. Lactic acidosis as a result of parenteral β agonist expertise. In our experience, high flow ventilators specifically use, anaerobic metabolism due to high work of breathing, tis- designed for NIV (such as the BiPAP Vision; Respironics, Pitts- sue hypoxia, intracellular alkalosis, and reduced lactate clear- burg, USA) that allow significant mask and mouth leaks are ance due to liver congestion all contribute to the complex better tolerated than many conventional ICU ventilators. metabolic disturbances of acute severe asthma. Haemody- namically significant arrhythmias are relatively infrequent, Endotracheal intubation even with the combination of methylxanthines and β receptor Cardiopulmonary arrest and deteriorating consciousness are agonists45; however, β agonist induced hypokalaemia height- absolute indications for intubation and assisted ventilation. ens the risk. Hypercapnia, acidosis, and clinical signs of severe disease at

www.thoraxjnl.com Downloaded from thorax.bmj.com on 21 August 2007

84 Phipps, Garrard

Box 2 Summary of recommendations for the process Ketamine is a general anaesthetic agent that has been used of intubation before, during, and after intubation in patients with acute severe asthma.61–63 It has sympathomimetic and bronchodilat- • Performed/supervised by experienced anaesthetists or ing properties. The usual dose for intubation is 1–2 mg/kg intensivists given intravenously over 2–4 minutes. It may increase blood • Skilled assistants in an appropriate environment pressure and heart rate, lower seizure threshold, alter mood, • Good preparation and understanding of the pathophysiol- and cause delirium. Inhalational anaesthetics used for gas ogy induction have the advantage of bronchodilation and may • Correct electrolyte disturbances and rehydrate make muscle relaxation unnecessary. However, specialised • Obtain reliable large bore venous access anaesthetic equipment is required for this approach. • Continuous ECG and pulse oximetry Opioids are a useful addition to sedatives and provide anal- • Continuous arterial monitoring not essential, but helpful gesia during intubation and mechanical ventilation. Morphine • Pre-oxygenate • Use familiar method of intubation in large boluses causes histamine release, which may worsen • Use familiar sedatives and muscle relaxants bronchoconstriction and hypotension. Some intravenous • Prepare for the rapid correction of hypotension, arrhyth- preparations also contain metabisulphite, to which some mias and barotrauma. asthmatics are sensitive. Fentanyl is a better choice of opioid • Ventilator set up and ready to monitor airway pressures for intubation as it inhibits airway reflexes and is short acting. early It causes less histamine release than morphine but large • Get aerosol delivery system for the ventilator connected or boluses may cause bronchospasm and chest wall rigidity. commence parenteral bronchodilator therapy • Plan ongoing sedation/paralysis before intubation Neuromuscular blocking drugs Rapid sequence induction with cricoid pressure should be used to prevent aspiration of gastric contents. Suxametho- presentation may not require immediate intubation before an 57 58 nium, a depolarising muscle relaxant, is widely used. It has a aggressive trial of conventional bronchodilator therapy. rapid onset and short duration of action but may cause hyper- Conversely, progressive deterioration with increasing distress kalaemia and increased intracranial pressure. Rocuronium, a or physical exhaustion may warrant intubation and mechani- non-depolarising muscle relaxant with an acceptably rapid cal ventilation without the presence of hypercapnia. onset, offers an alternative. Allergic sensitivity may occur to Once it has been decided that mechanical ventilation is any neuromuscular blocking agent and most may also cause required, the necessary medications, suitable monitoring histamine release and the potential for bronchospasm, equipment, and expert help should be sought. An understand- particularly in bolus doses. Atracurium boluses should be ing of the pathophysiology and anticipation of difficulties can avoided because of this possibility and vecuronium or minimise the complications associated with endotracheal pancuronium infusions used for longer term maintenance of intubation and ventilation. The best technique for intubation muscle relaxation. is generally that most familiar to the clinician performing the Myopathy and muscle weakness are well recognised procedure. complications of the long term administration of non- The process of intubation begins with explanation and depolarising neuromuscular blocking agents in asthmatic reassurance for the patient, followed by pre-oxygenation. The patients with an incidence of about 30%.64–66 In most cases the asthmatic patient is often dehydrated and the combination of PEEPi, the loss of endogenous catecholamines, and the myopathy is reversible, but may take weeks to resolve. There is vasodilating properties of the anaesthetic agents can cause an association between neuromyopathy and the duration of 59 60 muscle relaxant drug use that is independent of corticosteroid catastrophic hypotension. Volume resuscitation before 65 induction of anaesthesia can limit the degree of hypotension therapy. The use of neuromuscular blocking agents should but vasoconstrictors such as ephedrine or metaraminol should therefore be kept to a minimum. be at hand. Intubation is best performed by direct laryngoscopy after Mechanical ventilation induction of general anaesthesia. Endoscopic methods have Mechanical ventilation provides respiratory support while 16 58 drug therapy reverses bronchospasm and airway been advocated with either oral or nasal intubation, but 28 laryngeal spasm and further bronchoconstriction may occur. inflammation. Abolishing hypoxia is the most important Satisfactory local anaesthesia of the oropharynx, nasophar- aim. ynx, and larynx is therefore essential. Longer term sedation is required once the airway has been secured. Some recommen- Modes of ventilation dations for successful and safe endotracheal intubation are Commonly adopted ventilation modes include pressure summarised in box 2. limited and time cycled (pressure control or bilevel ventilation) or volume limited and time cycled modes (synchronised Drug therapy for intubation and mechanical ventilation intermittent mandatory ventilation, SIMV).28 In pressure lim- Anaesthetic agents and sedatives ited modes, maximum airway pressure is set and the tidal vol- Etomidate and thiopentone are short acting imidazole and ume delivered depends on respiratory system compliance. Vol- barbiturate drugs, respectively, that are commonly used for ume cycled modes, however, deliver a set tidal volume and can intubation although rarely bronchospasm and anaphylactoid be successfully and safely used provided an airway pressure reactions have been reported. Longer term sedation may be limit is set appropriately. As the patient improves and begins obtained by infusion of midazolam (2–10 mg/h); metabolites to breathe, spontaneously triggered modes of ventilation such may accumulate in renal and hepatic impairment. Propofol is as pressure support ventilation (PSV) can be introduced. a useful drug for intubation and intermediate term sedation, mainly because of its rapid onset and offset of action. It is eas- Ventilator settings ily titratable for intubation, providing deep sedation rapidly, The major variables that need to be set when placing a patient although it has no analgesic properties. However, vasodilata- on mechanical ventilation include the oxygen concentration tion and hypotension occur, especially in dehydrated patients. of inspired gas (FiO2), tidal volume (VT) or inspiratory Relatively little literature regarding its specific use in asthma is pressure, ventilator rate, inspiratory to expiratory time ratio available. The doses of all the above agents need to be adjusted (I:E ratio), and PEEP. The aims are to minimise airway for patient size and pre-existing level of consciousness. pressure, allowing sufficient time for completion of expiration

www.thoraxjnl.com Downloaded from thorax.bmj.com on 21 August 2007

Acute severe asthma in the intensive care unit 85

Box 3 Initial ventilator settings in paralysed patients Box 4 Recommendations for aerosol delivery to (adapted from Finfer and Garrard109) mechanically ventilated patients Metered dose inhaler (MDI) system •FiO2= 1.0 (initially) • Long expiratory time (I:E ratio >1:2) • Spacer or holding chamber • Low tidal volume 5–7 ml/kg • Location in inspiratory limb rather than Y piece • Low ventilator rate (8–10 breaths/min) • No humidification (briefly discontinue)

• Set inspiratory pressure 30–35 cm H2O on pressure control • Actuate during lung inflation ventilation or limit peak inspiratory pressure to • Large endotracheal tube internal diameter

<40 cm H2O • Prolonged inspiratory time • Minimal PEEP <5 cm H O 2 Jet nebuliser system • Mount nebuliser in inspiratory limb • Delivery may be improved by inspiratory triggering while achieving adequate alveolar ventilation. Suggested • Increase inspiratory time and decrease respiratory rate initial ventilator settings are shown in box 3. • Use a spacer Outcome is improved in mechanically ventilated asthmatics • High flow to generate aerosol by limiting airway pressure using a low respiratory rate and • High volume fill tidal volume while permitting a moderate degree of hypercar- • Stop humidification bia and respiratory acidosis.67 Hypercarbia has not been found • Consider continuous nebulisation to be detrimental except in patients with raised intracranial Ultrasonic nebulisers pressure or severe myocardial depression. Moderate degrees of hypercarbia with an associated acidosis (pH 7.2–7.15) are • Position in inspiratory limb prior to a spacer device • Use high power setting generally well tolerated. Reducing the respiratory rate to 8 or • Use a high volume fill 10 breaths/min prolongs expiratory time so that I:E ratios of • Maximise inspiratory time greater than 1:2 can be achieved. An attempt to increase • Drugs must be stable during ultrasonic nebulisation minute ventilation (to reduce PaCO2) by increasing the ventilator respiratory rate invariably reduces the expiratory time and I:E ratio, increases air trapping, and may paradoxically cause Metered dose inhalers have been widely used and may pro- an increased PaCO2. This has resulted in perceived failure of 68 vide at least as good drug delivery as nebulisers, depending on mechanical ventilation. actuator design and the presence of humidiﬁcation and spacer Humidiﬁcation of inspired gas is particularly important in devices.77 78 The recommended characteristics of aerosol deliv- asthmatic patients to prevent thickening of secretions and ery systems used in ventilated patients are shown in box 4. drying of airway mucosa, a stimulus for bronchospasm in 69 Ideally, each aerosol delivery system should be evaluated for itself. each type of ventilator circuit used.73

Ventilator alarms THERAPEUTIC OPTIONS IN THE NON-RESPONDING These include peak pressure and low tidal volume/low minute PATIENT ventilation alarms. If exceptionally high airway pressures A proportion of patients improve rapidly following the occur or there is a sudden fall in VT, blockage of the endotra- introduction of mechanical ventilation, and weaning should cheal tube, pneumothorax, or lobar collapse should be occur in line with this improvement. Unfortunately, for some excluded. Plateau rather than peak airway pressure may pro- there is difficulty in achieving adequate ventilation or there is vide the best measure of alveolar pressure and provide the best persistent hypoxia. Difficulty in ventilation may be due to predictor of barotrauma, together with measures of hyperin- refractory bronchospasm, extreme hyperinflation, or mucus flation such as PEEPi.70 plugging.

Extrinsic PEEP Manual compression Low level CPAP may be beneficial in spontaneously breathing, This technique was first described anecdotally by Watts in mechanically ventilated patients, especially if expiratory mus- 1984.79 Hyperinflation is relieved by manual compression of cle activity is contributing to dynamic airways collapse. How- the chest wall during expiration.80 The technique has been ever, in mechanically ventilated paralysed patients extrinsic advocated and used with success in both intubated and non- PEEP was of no benefit at low levels and was detrimental at intubated patients, although it has not been fully evaluated by high levels because the fall in gas trapping was outweighed by a controlled clinical study in humans.81 80 the rise in functional residual capacity (FRC).22 However, in this study large VT were used (up to 18 ml/kg); furthermore Mucolytics PEEPi and arterial blood gases were not measured. Changes in There is often a striking degree of mucus impaction in both FRC and gas trapping may guide the level of PEEP. Applied large and small airways that contributes to hyperinflation, extrinsic PEEP should not exceed PEEPi. segmental and lobar collapse with shunting, increased airway pressure, and barotrauma. Chest physiotherapy and mucolyt- Topical drug delivery to the ventilated patient ics have no proven benefit. Bronchoscopic lavage with locally Mechanical ventilation, whether invasive or non-invasive, may applied acetylcysteine may be used to help clear impacted compromise the delivery of bronchodilator aerosols. The secretions in selected refractory patients but its routine use is amount of nebulised drug reaching the airways depends on not advocated.82 Recently, recombinant DNase, a mucolytic the nebuliser design, driving gas flow, characteristics of the prescribed for sputum liquefaction in cystic fibrosis, has been ventilator tubing, and the size of the endotracheal tube.47 71 used to treat mucus impaction in asthma but there are no Drug delivery may vary from 0% to 42% in ventilated clinical trials.83 patients.72 The presence of humidification alone may reduce drug deposition by as much as 40%, but may be reversed by the Inhalational anaesthetic agents addition of a spacer device.73 74 Both ultrasonic and jet nebulis- Halothane, isoflurane, and sevoflurane are potent bronchodi- ers are effective in ventilated patients.75 Nebulisers may, how- lators in asthmatic patients receiving mechanical ventilation ever, be a source of bacterial contamination.76 who have failed to respond to conventional β adrenergic

www.thoraxjnl.com Downloaded from thorax.bmj.com on 21 August 2007

86 Phipps, Garrard agents.84 Experimental evidence indicates a direct effect on Nitric oxide (NO) bronchial smooth muscle mediated via calcium dependent NO exerts a weak bronchodilator effect.107 It dilates pulmonary channels as well as by modulating vagal, histamine, allergen, arteries and, when inhaled, may improve ventilation/perfusion and hypoxia induced bronchoconstrictor mechanisms.85 86 matching. Furthermore, these agents reduce pulmonary vascular tone resulting in lower pulmonary artery pressures in acute 87 OUTCOME AND FOLLOW UP asthma. Bronchodilator responses are seen in the form of ICU admission identifies an asthmatic patient as a member of reduced peak airway pressures within minutes, associated a poor prognostic group.3 5 108 Follow up should include a focus with improved ventilation distribution (lower PaCO2) and 88 on anti-inflammatory therapy and a written management reduced air trapping. Although bronchodilator effects are plan that may include the emergency use of intramuscular seen at sub-anaesthetic concentrations, these agents also adrenaline. Issues such as access to health care services, com- offer a relatively expensive method of sedation. A few ICU pliance with treatment, avoidance of triggers, socioeconomic ventilators, such as the Seimens Servo 900 series, can be fit- and psychosocial factors also need to be addressed. ted with a vaporiser allowing anaesthetic gases to be administered. Effective exhaled gas scavenging systems are required when using inhalational anaesthetics in the ICU. If this facil- ACKNOWLEDGEMENT ity is not available a Cardiff canister can be added to the The authors would like to thank Dr Duncan Youngfor his review of the expiratory port of the ventilator to remove effluent anaes- manuscript and helpful suggestions. thetic gases. Significant side effects such as hypotension and myocardial irritability exist, and prolonged administration of ...... some agents may result in bromide or fluoride toxicity.89 Authors’ affiliations Sevoflurane, a halogenated ether, is largely devoid of P Phipps, C S Garrard, Intensive Care Unit, John Radcliffe Hospital, cardiorespiratory side effects and may be the preferred agent. Oxford OX3 9DU, UK Administration of sub-anaesthetic concentrations of these agents via face mask may relieve bronchospasm refractory to REFERENCES conventional treatment.90 1 Kearney SE, Graham DR, Atherton ST. Acute severe asthma treated by mechanical ventilation: a comparison of the changing characteristics over One of the difficult aspects of mechanical ventilation of the a17yrperiod.Respir Med 1998;92:716–21. acute asthmatic patient is the weaning and extubation 2 Richards GN, Kolbe J, Fenwick J, et al. Demographic characteristics of process. The presence of the endotracheal tube within the lar- patients with severe life threatening asthma: comparison with asthma deaths. Thorax 1993;48:1105–9. ynx and trachea induces bronchoconstriction which becomes 3 Turner MO, Noertjojo K, Vedal S, et al. Risk factors for near-fatal troublesome as the sedation is withdrawn in preparation for asthma. A case-control study in hospitalized patients with asthma. Am J extubation.91 92 Use of an inhalational anaesthetic agent allows Respir Crit Care Med 1998;157:1804–9. 4 Sandford AJ, Pare PD. The genetics of asthma. The important questions. the endotracheal tube to be removed under anaesthesia with Am J Respir Crit Care Med 2000;161:S202–6. the confident expectation of rapid recovery once the anaes- 5 Marquette CH, Saulnier F, Leroy O, et al. Long-term prognosis of thetic is discontinued. near-fatal asthma. A 6-year follow-up study of 145 asthmatic patients who underwent mechanical ventilation for a near-fatal attack of asthma. Am Rev Respir Dis 1992;146:76–81. 6 Abramson MJ, Bailey MJ, Couper FJ, et al. Are asthma medications and Helium management related to deaths from asthma? Am J Respir Crit Care Med A mixture of helium and oxygen (heliox) may reduce the work 2001;163:12–8. of breathing and improve gas exchange because of its low 7 Castle W, Fuller R, Hall J, et al. Serevent nationwide surveillance study: comparison of salmeterol with salbutamol in asthmatic patients who density that reduces airway resistance and hyperinflation. require regular bronchodilator treatment. BMJ 1993;306:1034–7. However, the benefits are marginal and the concentration of 8 Donohue JF. The expanding role of long-acting beta-agonists. Chest inspired oxygen is consequently decreased. Flow meters and 2000;118:283–5. 9 Kallenbach JM, Frankel AH, Lapinsky SE, et al. Determinants of near nebuliser generator systems must be adapted for heliox use in 95 93 fatality in acute severe asthma. Am J Med 1993; :265–72. ventilated patients. The use of heliox to prevent intubation 10 Shugg AW, Kerr S, Butt WW. Mechanical ventilation of paediatric has not been studied, but dyspnoea scores were improved in patients with asthma: short and long term outcome. J Paediatr Child one study, possibly by reducing the work of breathing.94 Health 1990;26:343–6. 11 Saetta M, Di Stefano A, Rosina C, et al. Quantitative structural analysis of peripheral airways and arteries in sudden fatal asthma. Am Rev Respir Dis 1991;143:138–43. Magnesium sulphate 12 Carroll N, Elliot J, Morton A, et al. The structure of large and small Early anecdotal reports suggested benefit from intravenous airways in nonfatal and fatal asthma. Am Rev Respir Dis magnesium sulphate, which has been inconsistently sup- 1993;147:405–10. 95–99 13 Faul JL, Tormey VJ, Leonard C, et al. Lung immunopathology in cases of ported by randomised studies. A significant benefit was sudden asthma death. Eur Respir J 1997;10:301–7. recently observed in children receiving intravenous magne- 14 Reid LM. The presence or absence of bronchial mucus in fatal asthma. J sium sulphate (40 mg/kg) during acute asthma attacks.100 Allergy Clin Immunol 1987;80:415–6. Overall, the case for magnesium sulphate in acute asthma 15 Wasserfallen JB, Schaller MD, Feihl F, et al. Sudden asphyxic asthma: a distinct entity? Am Rev Respir Dis 1990;142:108–11. requires further evaluation in both adults and children. 16 Corbridge TC, Hall JB. The assessment and management of adults with status asthmaticus. Am J Respir Crit Care Med 1995;151:1296–316. 17 Rodrigo G, Rodrigo C. Assessment of the patient with acute asthma in Leukotriene inhibitors the emergency department. A factor analytic study. Chest Leukotrienes are inflammatory mediators known to be active 1993;104:1325–8. 18 Lim TK. Status asthmaticus in a medical intensive care. Singapore Med J in the airway inflammation of asthma. Leukotriene receptor 1989;30:334–8. antagonists (zafirlukast, montelukast) and synthesis blockers 19 Hill AR. Respiratory muscle function in asthma. J Assoc Acad Minor Phys (zilueton) currently have a relatively minor role in the 1991;2:100–8. 20 Lougheed DM, Webb KA, O’Donnell DE. Breathlessness during induced management of poorly controlled and aspirin sensitive lung hyperinflation in asthma: the role of the inspiratory threshold load. 101 asthma. However, recent work has suggested a role for leu- Am J Respir Crit Care Med 1995;152:911–20. kotriene antagonists in acute asthma.102–105 21 De Troyer A, Pride NB. The chest wall and respiratory muscles. In: Roussos C, ed. The thorax. New York: Dekker, 1995: 1975–2006. 22 Tuxen DV. Detrimental effects of positive end-expiratory pressure during Platelet activating factor (PAF) inhibitors controlled mechanical ventilation of patients with severe airflow obstruction. Am Rev Respir Dis 1989;140:5–9. PAF inhibitors attenuate the late response in asthma but have 23 Appel D, Rubenstein R, Schrager K, et al. Lactic acidosis in severe limited clinical efficacy.106 asthma. Am J Med 1983;75:580–4.

www.thoraxjnl.com Downloaded from thorax.bmj.com on 21 August 2007

Acute severe asthma in the intensive care unit 87

24 Findley LJ, Sahn SA. The value of chest roentgenograms in acute asthma 57 Mountain RD, Sahn SA. Clinical features and outcome in patients with in adults. Chest 1981;80:535–6. acute asthma presenting with hypercapnia. Am Rev Respir Dis 25 White CS, Cole RP, Lubetsky HW, et al. Acute asthma. Admission chest 1988;138:535–9. radiography in hospitalized adult patients. Chest 1991;100:14–6. 58 Leatherman J. Life-threatening asthma. Clin Chest Med 26 Myrianthefs MM, Nicolaides EP, Pitiris D, et al. False positive 1994;15:453–79. ST-segment depression during exercise in subjects with short PR segment 59 Mansel JK, Stogner SW, Petrini MF, et al. Mechanical ventilation in and angiographically normal coronaries: correlation with patients with acute severe asthma. Am J Med 1990;89:42–8. exercise-induced ST depression in subjects with normal PR and normal 60 Zimmerman JL, Dellinger RP, Shah AN, et al. Endotracheal intubation coronaries. J Electrocardiol 1998;31:203–8. and mechanical ventilation in severe asthma. Crit Care Med 27 British Thoracic Society, British Paediatric Association, Research Unit of 1993;21:1727–30. the Royal College of Physicians of London, et al. Guidelines on the 61 L’Hommedieu CS, Arens JJ. The use of ketamine for the emergency management of asthma. Thorax 1993;48(Suppl):S1–24. intubation of patients with status asthmaticus. Ann Emerg Med 28 Garrard CS, Nolan K. Intensive care management of asthma. In: Webb 1987;16:568–71. AR, Shapiro MJ, Singer M, et al, eds. Oxford textbook of intensive care. 62 Hemming A, MacKenzie I, Finfer S. Response to ketamine in status Oxford: Oxford University Press, 1999:90–5. asthmaticus resistant to maximal medical treatment. Thorax 29 Garrard CS, Benham S. Bronchodilators in intensive care. In: Webb AR, 1994;49:90–1. Shapiro MJ, Singer M, et al,eds.Oxford textbook of intensive care. 63 Howton JC, Rose J, Duffy S, et al. Randomized, double-blind, Oxford: Oxford University Press, 1999:1241–4. placebo-controlled trial of intravenous ketamine in acute asthma. Ann 30 Lipworth BJ, Jackson CM, Ziyaie D, et al. An audit of acute asthma Emerg Med 1996;27:170–5. admissions to a respiratory unit. Health Bull (Edinb) 1992;50:389–98. 64 Griffin D, Fairman N, Coursin D, et al. Acute myopathy during treatment 31 Manser R, Reid D, Abramson M. Corticosteroids for acute severe asthma of status asthmaticus with corticosteroids and steroidal muscle relaxants. in hospitalised patients. Cochrane Database Syst Rev 2000;2. Chest 1992;102:510–4. 32 Salmeron S, Brochard L, Mal H, et al. Nebulized versus intravenous 65 Leatherman JW, Fluegel WL, David WS, et al. Muscle weakness in albuterol in hypercapnic acute asthma. A multicenter, double-blind, mechanically ventilated patients with severe asthma. Am J Respir Crit randomized study. Am J Respir Crit Care Med 1994;149:1466–70. Care Med 1996;153:1686–90. 33 Lin RY, Smith AJ, Hergenroeder P. High serum albuterol levels and 66 Margolis BD, Khachikian D, Friedman Y, et al. Prolonged reversible tachycardia in adult asthmatics treated with high-dose continuously quadriparesis in mechanically ventilated patients who received long-term aerosolized albuterol. Chest 1993;103:221–5. infusions of vecuronium. Chest 1991;100:877–8. 34 Lin RY, Sauter D, Newman T, et al. Continuous versus intermittent 67 Darioli R, Perret C. Mechanical controlled hypoventilation in status albuterol nebulization in the treatment of acute asthma. Ann Emerg Med asthmaticus. Am Rev Respir Dis 1984;129:385–7. 1993;22:1847–53. 68 Shapiro MB, Kleaveland AC, Bartlett RH. Extracorporeal life support for 35 Besbes-Ouanes L, Nouira S, Elatrous S, et al. Continuous versus status asthmaticus. Chest 1993;103:1651–4. intermittent nebulization of salbutamol in acute severe asthma: a 69 Scharf SM, Heimer D, Walters M. Bronchial challenge with room randomized, controlled trial. Ann Emerg Med 2000;36:198–203. temperature isocapnic hyperventilation. A comparison with histamine 36 Rudnitsky GS, Eberlein RS, Schoffstall JM, et al. Comparison of challenge. Chest 1985;88:586–93. intermittent and continuously nebulized albuterol for treatment of asthma 70 Tuxen DV, Williams TJ, Scheinkestel CD, et al. Use of a measurement of in an urban emergency department. Ann Emerg Med 1993;22:1842–6. pulmonary hyperinflation to control the level of mechanical ventilation in 37 Papo MC, Frank J, Thompson AE. A prospective, randomized study of patients with acute severe asthma. Am Rev Respir Dis 1992;146:1136– continuous versus intermittent nebulized albuterol for severe status 42. asthmaticus in children. Crit Care Med 1993;21:1479–86. 71 Thomas SH, O’Doherty MJ, Fidler HM, et al. Pulmonary deposition of a 38 Idris AH, McDermott MF, Raucci JC, et al. Emergency department nebulised aerosol during mechanical ventilation. Thorax treatment of severe asthma. Metered-dose inhaler plus holding chamber 1993;48:154–9. is equivalent in effectiveness to nebulizer. Chest 1993;103:665–72. 72 Dhand R, Tobin MJ. Inhaled bronchodilator therapy in mechanically 39 Weksler N, Brill S, Tarnapolski A, et al. Intranasal salbutamol instillation ventilated patients. Am J Respir Crit Care Med 1997;156:3–10. 17 in asthma attack. Am J Emerg Med 1999; :686–8. 73 Lange CF, Finlay WH. Overcoming the adverse effect of humidity in Steedman DJ 40 , Robertson CE. Emergency endotracheal drug aerosol delivery via pressurized metered-dose inhalers during mechanical 15 administration using aerosol. Resuscitation 1987; :135–9. ventilation. Am J Respir Crit Care Med 2000;161:1614–8. 41 Boulton DW, Fawcett JP. The pharmacokinetics of levosalbutamol: what 74 Harvey CJ, O’Doherty MJ, Page CJ, et al. Effect of a spacer on are the clinical implications? Clin Pharmacokinet 2001;40:23–40. pulmonary aerosol deposition from a jet nebuliser during mechanical 42 Handley DA. Single-isomer beta-agonists. Pharmacotherapy ventilation. Thorax 1995;50:50–3. 2001;21:21–7S. 75 Harvey CJ, O’Doherty MJ, Page CJ, et al. Comparison of jet and 43 Baldwin DR, Sivardeen Z, Pavord ID et al. Comparison of the effects of ultrasonic nebulizer pulmonary aerosol deposition during mechanical salbutamol and adrenaline on airway smooth muscle contractility in vitro ventilation. Eur Respir J 1997;10:905–9. and on bronchial reactivity in vivo. Thorax 1994;49:1103–8. 76 Craven DE, Lichtenberg DA, Goularte TA, et al. Contaminated 44 Abroug F, Nouira S, Bchir A, et al. A controlled trial of nebulized medication nebulizers in mechanical ventilator circuits. Source of salbutamol and adrenaline in acute severe asthma. Intensive Care Med bacterial aerosols. Am J Med 1984;77:834–8. 1995;21:18–23. Gay PC 45 Molfino NA, Nannini LJ, Martelli AN, et al. Respiratory arrest in 77 , Patel HG, Nelson SB, et al. Metered dose inhalers for near-fatal asthma. N Engl J Med 1991;324:285–8. bronchodilator delivery in intubated, mechanically ventilated patients. 99 46 Silverman R. Treatment of acute asthma. A new look at the old and at Chest 1991; :66–71. the new. Clin Chest Med 2000;21:361–79. 78 Duarte AG, Momii K, Bidani A. Bronchodilator therapy with 47 The COPD Guidelines Group of the Standards of Care Committee metered-dose inhaler and spacer versus nebulizer in mechanically of the BTS. BTS guidelines for the management of chronic obstructive ventilated patients: comparison of magnitude and duration of response. pulmonary disease. Thorax 1997;52(Suppl 5):S1–28. Respir Care 2000;45:817–23. 48 Parameswaran K, Belda J, Rowe BH. Addition of intravenous 79 Watts JI. Thoracic compression for asthma. Chest 1984;86:505. aminophylline to beta2-agonists in adults with acute asthma (Cochrane 80 Fisher MM, Bowey CJ, Ladd-Hudson K. External chest compression in Review). Cochrane Database Syst Rev 2000;4. acute asthma: a preliminary study. Crit Care Med 1989;17:686–7. 49 Murciano D, Auclair MH, Pariente R, et al. A randomized, controlled 81 Eason J, Tayler D, Cottam S, et al. Manual chest compression for total trial of theophylline in patients with severe chronic obstructive pulmonary bronchospasm. Lancet 1991;337:366. disease. N Engl J Med 1989;320:1521–5. 82 Lang DM, Simon RA, Mathison DA, et al. Safety and possible efficacy of 50 Lakshminarayan S, Sahn SA, Weil JV. Effect of aminophylline on fiberoptic bronchoscopy with lavage in the management of refractory ventilatory responses in normal man. Am Rev Respir Dis asthma with mucous impaction. Ann Allergy 1991;67:324–30. 1978;117:33–8. 83 Patel A, Harrison E, Durward A, et al. Intratracheal recombinant human 51 Ward AJ, McKenniff M, Evans JM, et al. Theophylline: an deoxyribonuclease in acute life-threatening asthma refractory to immunomodulatory role in asthma? Am Rev Respir Dis conventional treatment. Br J Anaesth 2000;84:505–7. 1993;147:518–23. 84 Eger EI. The pharmacology of isoflurane. Br J Anaesth 52 Rodrigo C, Rodrigo G. Treatment of acute asthma. Lack of therapeutic 1984;56:71–99S. benefit and increase of the toxicity from aminophylline given in addition 85 Hirshman CA, Edelstein G, Peetz S, et al. Mechanism of action of to high doses of salbutamol delivered by metered-dose inhaler with a inhalational anesthesia on airways. Anesthesiology 1982;56:107–11. spacer. Chest 1994;106:1071–6. 86 Korenaga S, Takeda K, Ito Y. Differential effects of halothane on airway 53 Zainudin BM, Ismail O, Yusoff K. Effect of adding aminophylline nerves and muscle. Anesthesiology 1984;60:309–18. infusion to nebulised salbutamol in severe acute asthma. Thorax 87 Saulnier FF, Durocher AV, Deturck RA, et al. Respiratory and 1994;49:267–9. hemodynamic effects of halothane in status asthmaticus. Intensive Care 54 Mansel JK, Stogner SW, Norman JR. Face-mask CPAP and sodium Med 1990;16:104–7. bicarbonate infusion in acute, severe asthma and metabolic acidosis. 88 Maltais F, Reissmann H, Navalesi P, et al. Comparison of static and Chest 1989;96:943–4. dynamic measurements of intrinsic PEEP in mechanically ventilated 55 Shivaram U, Miro AM, Cash ME, et al. Cardiopulmonary responses to patients. Am J Respir Crit Care Med 1994;150:1318–24. continuous positive airway pressure in acute asthma. J Crit Care 89 Echeverria M, Gelb AW, Wexler HR, et al. Enflurane and halothane in 1993;8:87–92. status asthmaticus. Chest 1986;89:152–4. 56 Meduri GU, Turner RE, Abou-Shala N, et al. Noninvasive positive 90 Padkin AJ, Baigel G, Morgan GA. Halothane treatment of severe pressure ventilation via face mask. First-line intervention in patients with asthma to avoid mechanical ventilation. Anaesthesia 1997;52:994–7. acute hypercapnic and hypoxemic respiratory failure. Chest 91 Gal TJ. Pulmonary mechanics in normal subjects following endotracheal 1996;109:179–93. intubation. Anesthesiology 1980;52:27–35.

www.thoraxjnl.com Downloaded from thorax.bmj.com on 21 August 2007

88 Phipps, Garrard

92 Habre W, Scalfaro P, Sims C, et al. Respiratory mechanics during 101 Drazen JM, Israel E, O’Byrne PM. Treatment of asthma with drugs sevoflurane anesthesia in children with and without asthma. Anesth modifying the leukotriene pathway. N Engl J Med 1999;340:197–206. Analg 1999;89:1177–81. 102 Kuitert LM, Barnes NC. Leukotriene receptor antagonists: useful in acute 93 Gluck EH, Onorato DJ, Castriotta R. Helium-oxygen mixtures in intubated asthma? Thorax 2000;55:255–6. patients with status asthmaticus and respiratory acidosis. Chest 103 Drazen JM, O’Brien J, Sparrow D, et al. Recovery of leukotriene E4 from 1990;98:693–8. the urine of patients with airway obstruction. Am Rev Respir Dis 94 Kass JE, Terregino CA. The effect of heliox in acute severe asthma: a 146 randomized controlled trial. Chest 1999;116:296–300. 1992; :104–8. 95 McNamara RM, Spivey WH, Skobeloff E, et al. Intravenous magnesium 104 Dworski R, Fitzgerald GA, Oates JA, et al. Effect of oral prednisone on sulfate in the management of acute respiratory failure complicating airway inflammatory mediators in atopic asthma. Am J Respir Crit Care asthma. Ann Emerg Med 1989;18:197–9. Med 1994;149:953–9. 96 Green SM, Rothrock SG. Intravenous magnesium for acute asthma: 105 Dockhorn RJ, Baumgartner RA, Leff JA, et al. Comparison of the effects failure to decrease emergency treatment duration or need for of intravenous and oral montelukast on airway function: a double blind, hospitalization. Ann Emerg Med 1992;21:260–5. placebo controlled, three period, crossover study in asthmatic patients. 97 Bloch H, Silverman R, Mancherje N, et al. Intravenous magnesium Thorax 2000;55:260–5. sulfate as an adjunct in the treatment of acute asthma. Chest 106 Evans DJ, Barnes PJ, Cluzel M, et al. Effects of a potent 1995;107:1576–81. platelet-activating factor antagonist, SR27417A, on allergen-induced 98 Rowe BH, Bretzlaff JA, Bourdon C, et al. Magnesium sulfate for treating asthmatic responses. Am J Respir Crit Care Med 1997;156:11–6. exacerbations of acute asthma in the emergency department. Cochrane 107 Hogman M, Frostell CG, Hedenstrom H, et al. Inhalation of nitric oxide Database Syst Rev 2000;2. 99 Chande VT, Skoner DP. A trial of nebulized magnesium sulfate to modulates adult human bronchial tone. Am Rev Respir Dis reverse bronchospasm in asthmatic patients. Ann Emerg Med 1993;148:1474–8. 1992;21:1111–5. 108 Molfino NA, Nannini LJ, Rebuck AS, et al. The fatality-prone asthmatic 100 Ciarallo L, Brousseau D, Reinert S. Higher-dose intravenous magnesium patient. Follow-up study after near-fatal attacks. Chest 1992;101:621–3. therapy for children with moderate to severe acute asthma. Arch Pediatr 109 Finfer SR, Garrard CS. Ventilatory support in asthma. Br J Hosp Med Adolescent Med 2000;154:979–83. 1993;49:357–60.

www.thoraxjnl.com

The New England Journal of Medicine

Review Article

Current Concepts sory effect (such as asbestos) can be inhaled for prolonged periods and result in serious injury. PATHOPHYSIOLOGY AND PATHOGENESIS OCCUPATIONAL RESPIRATORY Rhinitis and Laryngitis DISEASES Nasal hairs and turbinates filter particles and gases (Fig. 1). The initial nasal mucosal response is vascu- WILLIAM S. BECKETT, M.D. lar dilation and increased permeability, rhinorrhea, and congestion.8,9 Allergens cause sneezing, itching, rhinorrhea, and congestion, whereas irritants cause burning or irritation and congestion.10 Numerous ATERIALS inhaled in the workplace can substances in the workplace, from latex-coated corn- lead to all the major chronic lung diseases starch particles to common cleaning solutions, cause Mexcept those due to vascular disease. The allergic and irritant upper-airway disease (Table 1). physician should consider the possibility of occu- The presence of several risk factors, both occupa- pational exposure when a working or retired adult tional and nonoccupational, can contribute to the de- presents with unexplained respiratory illness. Even ex- velopment of disease. In the general population, an posure in office buildings1 and hospitals2 occasional- increased risk of rhinosinusitis is associated with a fam- ly causes illness. Identifying a workplace-related cause ily history and previous allergic disease. Occupational of disease is important because it can lead to cure and rhinitis is distinguished from perennial and seasonal to prevention for others. Because of differences in the rhinitis by an improvement in symptoms when the metabolism and susceptibility of hosts, one occupa- person is away from work. Some people have en- tional agent may cause many diseases, just as cigarette hanced responses to environmental tobacco smoke smoke causes several distinct disorders. Conversely, and other substances.12 Nonirritant rhinorrhea occurs one respiratory disease may have several occupational with cold air, exposure to cholinergic substances, causes.3 The recognition of occupational causes can be or exposure to cholinesterase-inhibiting substances. made difficult by delayed responses that occur at home Allergic occupational rhinitis often precedes occupa- after work and by years of latency between exposure tional asthma. Chromic acid and other substances in the workplace and the occurrence of disease. produce nasal ulceration and perforation of the sep- The incidence of diseases caused by mineral dust tum (Table 1), whereas nickel and certain wood dusts has declined recently in postindustrial countries, and cause nasal carcinoma.13,14 asthma has emerged as the principal occupational lung The larynx, which has the smallest cross-sectional disease.4,5 Each year, new substances are introduced area in the respiratory tract, forms a narrows in the into the workplace, and some are found to cause lung airstream where air velocity increases, eddies form, and disease.6,7 particulate matter is deposited. Inflammation and ede- The airways, from nares to alveoli, come into con- ma of the vocal cords result from irritants and aller- tact with 14,000 liters of air in the workplace during gens or from the drainage of inflammatory mediators a 40-hour workweek. Physical activity can increase from nasal passages. The inadvertent mixing of bleach ventilation, and thus exposure to contaminants, up to (hypochlorous acid) with common institutional de- 12 times the levels at rest. As ventilation increases, tergent solutions that contain phosphoric acid gener- breathing shifts from nasal to a combination of oral ates chlorine gas, a potent irritant. The mixing of de- and nasal, allowing a greater volume of air to bypass tergents with ammonia generates monochloramine the cleansing nasopharynx and further increasing the and dichloramine, which are also potent irritants.11,15 exposure of the lower airways to inhaled materials Such exposures have also been associated with vocal- (Fig. 1). Strong irritants (such as ammonia) produce cord dysfunction, which is a narrowing of the vocal- an aversive response, whereas materials with little sen- cord aperture during inspiration that produces asthma-like symptoms.16

Tracheitis, Bronchitis, and Bronchiolitis From the Department of Environmental Medicine and Pulmonary and Critical Care Division, University of Rochester School of Medicine and Den- Soluble gases are absorbed by the upper-airway tistry, Rochester, N.Y. Address reprint requests to Dr. Beckett at the Univer- mucosa, whereas less soluble gases penetrate to the sity of Rochester School of Medicine and Dentistry, Box EHSC, 575 Elm- wood Ave., Rochester, NY 14642, or at [email protected]. alveoli. The location of particle deposition in the air- ©2000, Massachusetts Medical Society. ways is determined by the concentration and size of

406 · February 10, 2000

CURRENT CONCEPTS

Rhinitis and laryngitis Nasal cavity Large particles are deposited in the nose, pharynx, and larynx. More soluble gases (e.g., sulfur dioxide) are absorbed by upper respiratory tract mucous membranes, causing Pharynx edema and mucus hypersecretion.

Tracheitis, bronchitis, and bronchiolitis Larynx Large particles (more than 10 µm in diameter) are deposited and then cleared by cilia. Small particles and fine fibers are deposited in bronchioles Trachea and bifurcations of alveolar ducts. Less soluble gases penetrate to deeper, small airways. Bronchus Asthma and chronic obstructive pulmonary disease Allergens and irritants are deposited in large airways by turbulent flow, causing chronic inflammatory changes.

Cancer Carcinogens (asbestos and polycyclic aromatic hydrocarbons) come into contact with bronchial epithelial cells, causing mutations in proto-oncogenes and tumor-suppressor genes. More than one such contact results in malignant transformation.

Interstitial disease Small particles (less than 10 µm in diameter) and fibers are deposited in terminal bronchioles, alveolar ducts, and alveoli. Penetration Bronchiole to the interstitium results in fibrosis and the formation of Alveolus granulomas.

Figure 1. Occupational Respiratory Diseases. Shown are categories of occupational respiratory disease, their anatomical locations within the respiratory system, examples of common causative substances, and their pathophysiologic effects.

the particles. Particles that are 10 µm or more in di- respirators that filter particles or chemicals.23 Bron- ameter are deposited in the nose and pharynx, where- chitis can often be reversed by reducing the amount of as particles that are 5 µm in diameter or smaller may exposure. Chronic bronchitis is treated with inhaled penetrate to the alveoli. Particles of intermediate size ipratropium bromide taken as two puffs every four to are deposited in differing proportions at intervening six hours, with each puff delivering 18 µg. levels (Fig. 1). Bronchiolitis obliterans — the narrowing and fill- Acute bronchitis results from mucosal inflammation ing of bronchioles — was first described in soldiers after exposure. The condition is treated with inhaled who had been exposed to chlorine and phosgene gas- corticosteroids (e.g., taken as two puffs of beclo- es, and it was later described in survivors of acci- methasone four times a day, with each puff deliver- dental workplace gassings with chlorine and in working 50 µg) until it subsides. Asthma may be a sequela ers exposed to nitrogen dioxide generated by freshly after bronchitis has resolved.17 Chronic irritation can stored hay (silo filler’s disease). It is manifested by lead to hyperplasia and hypersecretion of the mucous dyspnea, chest tightness, and irreversible airflow ob- glands. A cough with mucus production on most days struction, with onset several hours to days after a heavy for three months per year over two consecutive years exposure to irritant gas (Table 2).24 Bronchiolitis ob- is diagnostic of chronic bronchitis and correlates with literans may present with radiographic infiltrates or histologically confirmed mucous hyperplasia.18 Many may be clinically confused with asthma, which pre- occupational substances produce industrial bronchi- sents similarly with wheezing and a clear chest x-ray tis, just as cigarette smoke causes smoker’s bronchi- film but which responds to bronchodilators. High- tis.19 Bronchitis is found in persons whose occupation resolution computed tomography may show a patchy involves exposure to dust or who work in chemical interstitial pattern. Early therapy with systemic corti- and food processing, mining, storage and processing costeroids may reduce long-term morbidity. of grains and feeds, cotton-textile milling, and welding.20-22 The degree of exposure can be reduced by the Asthma substitution of materials, wetting of the dust, better More than 250 substances found in the workplace ventilation of the workstation, or the use of mask cause occupational asthma (Table 2). Work-related

Volume 342 Number 6 · 407

The New England Journal of Medicine

TABLE 1. SELECTED COMMON CAUSES TABLE 2. SELECTED COMMON CAUSES OF OCCUPATIONAL UPPER RESPIRATORY OF OCCUPATIONAL AIRWAY DISEASE.* TRACT DISEASE.* Bronchitis Rhinitis and laryngitis† Sulfur dioxide (used in chemical manufacturing) Allergic sensitizers8 Rock and mineral dusts (used in road construction and Dust from flour used in baking (wheat, rye, soy, and buck- digging of foundations) wheat flour and gluten and amylase), which can also Cement dust contain mites and fungi Smoke from welding or cutting with acetylene torch Dusts from animal feeds and grains Bronchiolitis Ethylenediamine in adhesives Latex on cornstarch granules from latex gloves Acetaldehyde Pollens and mold spores (found in plants and on moist, Ammonia (used in farm-crop preservation) dark surfaces) Chlorine gas Proteins from laboratory animals (such as guinea pigs Hydrogen fluoride and rats) Hydrogen sulfide (used in oil refining) Acid anhydrides (used in adhesives and coatings with Nitrogen dioxide (generated by freshly stored hay in silos), epoxy resin, circuit boards, and plasticizers) nitric acid, nitrous acid, and nitric oxide Psyllium (a pharmaceutical stool-bulking agent) Phosgene (used in chemical manufacturing) Irritants Asthma 11 Acidic or alkaline cleaning solutions and powders Asthma with latency Ammonia Acid anhydrides (used in epoxy adhesives and paints, 12 Environmental tobacco smoke coatings, circuit boards, polymers, polyesters, and 11 Hypochlorous acid (bleach) plasticizers) Metalworking fluids (cutting oils) Aldehydes Ozone (in aluminum welding) Acrylates (used in paints and adhesives) Sulfur dioxide Animal proteins (in laboratory animals, farming, and Volatile organic compounds (in paints, thinners, solvents, veterinary medicine) and industrial cleaning solutions) Cobalt (used in carbide-tipped tools) Rhinorrhea Dusts from flours and grains (found in bakeries) Cold air Dusts from wood (used in furniture making and Certain pesticides (carbaryl, malathion, parathion, cabinetry) mevinphos, pyrethrum) Ethylenediamine, monoethanolamine, and other amines Formaldehyde and glutaraldehyde (used in sterilizing Nasal ulceration and perforation of the septum medical instruments) Arsenic Isocyanates (hexamethylene diisocyanate, diphenyl- Chromic acid and chromates methane diisocyanate, and toluene diisocyanate) used Copper dusts and mists in polyurethane paint (used in auto-body repair) and the manufacture and application of foam (used in *Causes of upper respiratory tract disease are discussed in roofing foams) Bascom and Shusterman.10 Latex (used in health care facilities) Asthma without latency (irritants that cause the reactive air- †Many substances that cause allergic rhinitis may also cause way dysfunction syndrome) asthma. Contaminants in metalworking fluids Chlorine gas (pulp from paper mills) Bleach (sodium hypochlorite) Strong acids Chronic obstructive pulmonary disease and chronic air- asthma can be an exacerbation of asthma that was flow limitation previously subclinical or in remission (work-aggravat- Coal dust (causes emphysema with nodular fibrosis) Crystalline silica (causes chronic airflow limitation) ed asthma), a new onset of asthma caused by a sensi- Cotton dust (causes chronic airflow limitation) tizing exposure (asthma with latency), or asthma that Cadmium (causes emphysema) (used in electronics, metal results from a single heavy exposure to a potent res- plating, and batteries) piratory irritant (asthma without latency, irritant asth- Toluene diisocyanate (causes chronic airflow obstruction) ma, or the reactive airways dysfunction syndrome). 29 *Causes of bronchitis are discussed in Morgan19 and Fish- Five to 15 percent of new cases of asthma in working wick et al.,20 causes of bronchiolitis in Wright,24 causes of asthma in Chan-Yeung and Malo,25 and causes of chronic ob- adults are caused by occupational exposure. structive pulmonary disease and chronic airflow limitation in Work-aggravated asthma is caused by mechanical Hendrick,26 Christiani et al.,27 and Davison et al.28 irritation of the airways from nonallergenic dust and by chemical irritation. In asthma with latency, high- molecular-weight biologic proteins (e.g., particles from wood products, textiles, grains, crustaceans, or weight substances (less than 5000 daltons) can cause latex that are of 5000 daltons or more) stimulate the sensitization without producing specific IgE. Among production of specific IgE. A later reexposure trig- these substances, the isocyanates, which are used wide- gers the release of cytokines, whose effector cells par- ly in polyurethane foams and paints, are a frequently ticipate in an inflammatory response. Persons who recognized cause of asthma. Exposed French workers have a history of atopy or who smoke are at greater in whom the disease developed were more likely to risk than others for asthma that develops in response be positive for HLA-DQB1*0503 and the allelic com- to high-molecular-weight antigens. Low-molecular- bination HLA-DQB1*0201/0301, whereas HLA-

408 · February 10, 2000

CURRENT CONCEPTS

At work; 700 At home

600

500

400 (liters/min) 300 Daytime Peak Flow; 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Day of Measurement

Figure 2. Rates of Peak Expiratory Flow in a 42-Year-Old Man with Asthma and Work-Related Symptoms. The patient used a portable peak expiratory flowmeter and a daily calendar. The early-morning peak flows showed the diurnal morning dip usually seen with asthma. The peak flow values were consistently lower when the patient was at work than when he was at home, a finding that confirmed that the asthma was related to work.

DQB1*0501 and the HLA-DQA1*0101, DQB1*0501, vent chronic symptoms. Removal of the substance to DR1 haplotype appeared to be in part protective. 30 which the patient has been exposed usually brings Because some patients continue to have asthma some improvement.34 for years after the discontinuation of exposure, therapy emphasizes prevention, early diagnosis, and re- Chronic Obstructive Pulmonary Disease moval of the substance responsible for the exposure. or Chronic Airflow Limitation For most patients, the best therapeutic intervention Cigarette smoking remains the predominant cause is to eliminate exposure. The treating physician must of chronic obstructive pulmonary disease, but many ultimately address this issue, because prolonging ex- occupational dusts can cause or contribute to chron- posure can worsen bronchial hyperreactivity, symp- ic airflow limitation or emphysema (Table 2).26,35 toms, lung function, and prognosis. If the diagno- When chronic airflow limitation is caused by silica or sis of asthma is in question, nonspecific testing of beryllium dust,36 there may be accompanying radio- bronchial responsiveness (e.g., the methacholine chal- graphic evidence of pneumoconiosis. With most oth- lenge) can establish the presence or absence of airway er substances that cause chronic obstructive pulmo- hyperresponsiveness, which is almost always present nary disease (Table 2) there can be airflow obstruction in asthma. with a clear chest film. With many substances (includ- A history of onset after a new workplace exposure, ing coal and silica dust), the disease may progress for or of preexisting asthma made worse in a new work decades after the exposure has ceased.37 Cadmium environment, indicates that the asthma may be work causes emphysema after prolonged exposure at levels related. Checking whether the substance in question lower than those for other mineral dusts.28 Cotton is on published lists of the known causes of asthma dust, a complex mixture that contains bacterial en- can be helpful.25 The measurement of specific IgE, dotoxin, can cause chronic bronchitis,38 occupation- by epicutaneous (skin-prick) testing or a radioaller- al asthma,39 inhalation fever, and chronic airflow lim- gosorbent test, is now possible for a growing list of itation.27 known causes of asthma. A positive test confirms systemic sensitization but does not demonstrate sen- Lung Cancer sitization of the respiratory tract. Airflow obstruc- At least 12 substances found in the workplace are tion that is temporally related to exposure in the classified as human lung carcinogens (Table 3).40 Oc- workplace can be identified by having the patient cupational exposure is estimated to account for ap- record peak expiratory flows both at work and away proximately 5 percent of lung cancers in the United from work (Fig. 2).31,32 Diagnostic criteria that fol- States.41 The majority of these cancers are caused by low a physiologically based approach are now avail- asbestos, followed by radon, silica, chromium, cadmi- able.33 Airflow obstruction is treated with inhaled um, nickel, arsenic, and beryllium.42 Cigarette smoke beta-agonists for acute symptoms (e.g., taken as two and asbestos interact strongly in causing broncho- puffs of albuterol sulfate every four to six hours, genic carcinomas, and the risk of carcinoma is great- with each puff delivering 120 µg) and with inhaled er in persons with the interstitial fibrosis of asbes- corticosteroids and other drugs as needed to pre- tosis.43 Many asbestos-related lung cancers occur in

Volume 342 Number 6 · 409

The New England Journal of Medicine

TABLE 3. SELECTED CAUSES OF OCCUPATIONAL LUNG CANCER.* TABLE 4. SELECTED COMMON CAUSES OF OCCUPATIONAL INTERSTITIAL DISEASE. Asbestos (used in boiler and pipe insulation) Arsenic compounds (formerly used in pesticide sprays for orchards) Pulmonary fibrosis Bis(chloromethyl)ether and chloromethyl methyl ether (used in the man- Asbestos ufacture of ion-exchange resins, bactericides, pesticides, water repellents, Crystalline silica (produced by stone cutting, drilling, and tunneling) and flame repellents) Kaolin (a clay used in china, ceramics, and pharmaceuticals) Cadmium and cadmium compounds Talc (magnesium silicate, which is used in the paint, ceramics, leather, fab- ric, and paper industries) Chromium and certain (hexavalent) chromium compounds (used in alloys Tungsten carbide with cobalt (a hard metal used in carbide-tipped tools) and metal plating) Crystalline silica (produced by stone cutting, drilling, and tunneling) Alveolar proteinosis Mustard gas Fine crystalline silica dust (found in silica flour and produced by sand- blasting) Nickel in nickel refining Radon progeny (products of decay) and ionizing radiation Lipoid pneumonia 47 Soots, tars, and mineral oils (polycyclic aromatic hydrocarbons) Oily metalworking fluids (used in machining shops) Hypersensitivity pneumonitis *Causes of occupational lung cancer are discussed in the National Tox- Amebae icology Program Report on Carcinogens.40 Animal proteins (from pigeon, chicken, turkey, duck, and rat) Fungi (e.g., Aureobasidium pullulans) Metalworking fluid aerosols (used in metal-parts machining shops) Thermophilic bacteria (e.g., thermoactinomyces, Saccharopolyspora rectivir- gula) Other bacteria (Bacillus subtilis, B. cereus, and Pseudomonas fluorescens) pipe insulators, building-construction workers, ship- Toluene diisocyanate, diphenylmethane diisocyanate, and hexamethylene diisocyanate (in polyurethane paints, adhesives, and foam production) yard workers, and others with a history of prolonged, Trimellitic anhydride and phthalic anhydride (in epoxy resins, coatings, and heavy exposure before the current workplace controls paints) were instituted. Establishing a history of exposure Granulomatous disease that began more than 10 years before diagnosis and Beryllium (used in the aerospace industry and in beryllium copper alloy identifying indicators of the severity of the exposure machining) (such as pleural plaques or a large number of as- Inhalation fever Amebae, mixed bacteria, and fungi (endotoxin and beta glucans) from hu- bestos bodies or uncoated asbestos fibers in resect- midifiers and other sources of water aerosols 48 ed lung tissue) are helpful in determining the cause Cotton dust of cancer. Freshly generated zinc oxide fumes (from vaporized, galvanized metal)49 Heated fluorocarbon monomers and polymers (e.g., polytetrafluoroethyl- Homozygous deletion of the glutathione S-trans- ene [Teflon]) ferase M1 (GSTM1) gene, or an N-acetyltransferase 2 (NAT2) slow acetylator genotype, may confer additional risk of lung cancer among persons exposed to asbestos.44 Diffuse malignant mesothelioma of the pleura, peritoneum, and pericardium is also caused by asbestos, but the risk is not affected by smoking. Fibrotic Disease The epidemic of asbestos-related lung cancers among In the 25 years from 1968 through 1992, a total persons heavily exposed through the 1970s is esti- of 100,890 death certificates in the United States list- mated to have peaked in the early 1990s at 1200 cas- ed pneumoconiosis (dust-associated disease of the es per year in the United States45 and is expected to lungs) as a contributing or primary cause of death. peak in the United Kingdom by 2010.46 Reductions The majority of these cases were coal worker’s pneu- in the amount of exposure to these substances in the moconiosis, silicosis, and asbestosis.4 The rate of death past 25 years, along with the beneficial effect of pre- from pneumoconiosis is declining, but silicosis re- vious reductions in cigarette smoking, will contrib- mains prevalent in many regions and industries and ute to the eventual decline in the incidence of lung has recently been recognized as a disease of con- cancer in these countries, whereas the frequency of struction workers who perform surface-rock drilling the disease is expected to increase in countries with and cutting. Dusts such as coal, cobalt, talc, and ka- increasing occupational exposure and increasing prev- olin,47 which are encountered less frequently in in- alence of smoking. dustry than silica, are important causes of new cases of disease. Interstitial Lung Disease The number of deaths from asbestosis has increased Most inhaled dust is filtered out by the upper air- over the past 20 years but appears to have plateaued. ways or cleared by the ciliated epithelium of large Benign pleural disease is often the only manifesta- airways. If these defenses are overwhelmed by fine tion of exposure to asbestos. Asbestosis usually has dust (less than 10 µm in diameter), however, the lung an indolent course and may be present subclinically reacts with an alveolar and interstitial inflammation for many years before becoming symptomatic in late that may culminate in disease (Table 4). middle age, with dyspnea, inspiratory crackles, and

410 · February 10, 2000

Downloaded from www.nejm.org at UNIV OF MEDICINE AND DENTISTRY OF NJ on February 16, 2007 . Copyright © 2000 Massachusetts Medical Society. All rights reserved. CURRENT CONCEPTS bilateral basilar interstitial densities. A diagnosis is involves indoor work in the winter in which an or- based on the history of exposure, chest examina- ganism-contaminated water-spray system is used to tion and film, and pulmonary-function tests. Because humidify air (humidifier fever)48,54; disturbing an ac- there is no effective therapy, open or thoracoscopic cumulation of moldy hay, grass, compost, mulch, or lung biopsy is reserved for patients in whom a treat- wood chips; or heating zinc to the point of vaporiza- able disease such as pulmonary vasculitis (often action (e.g., in welding galvanized steel). Inhaled com- companied by evidence of systemic inflammation) is bustion products of polytetrafluoroethylene (Teflon) part of the differential diagnosis or for patients with may cause fever and hemorrhagic pneumonitis. 55 advanced disease who are being evaluated for lung transplantation. DIAGNOSIS, DETERMINATION OF CAUSE, Diffuse alveolar-filling disease (presenting as dysp- AND TREATMENT nea with interstitial abnormalities on the chest x-ray Most occupational respiratory illnesses can be diag- film) due to the overproduction of surfactant by nosed on the basis of the history,56 physical exami- type II alveolar pneumocytes may be caused by in- nation, chest x-ray film, and pulmonary-function tests. tense exposure to fine silica dust, and it has been re- An effective approach is first to determine the diag- ported after several other types of occupational expo- nosis and then to use the occupational history to iden- sure to dust. Lipoid, alveolar-filling pneumonia can tify possible causes (Tables 1, 2, 3, and 4). Higher- be caused by the inhalation of fine oil mist.50 risk workplaces are those with obvious dust, smoke, or vapor or those in which there is spraying, paint- Granulomatous Disease ing, or drying of coated surfaces. Heavier exposure Hypersensitivity pneumonitis, an acute febrile occurs when there is friction, grinding, heat, or blast- pneumonitis with peripheral leukocytosis, is caused ing, when very small particles are generated, and in by immunologic reaction to a variety of inhaled sub- enclosed spaces. The association of the illness with stances. It may have acute, subacute, and chronic work is suggested by a history of symptomatic im- forms, depending on the degree of exposure, the du- provement on weekends or vacations, followed by in- ration of exposure, and the susceptibility of the pa- creased symptoms on return to work; it can also be tient. A growing list of workplace substances have suggested by the onset of disease after a new job is been identified as capable of causing this response51 started or after a new material is introduced at the (Table 4). Cigarette smoking and intense exposure workplace. Asthmatic responses may be immediate predispose workers to sensitization to some substanc- (occurring within 2 hours after exposure), delayed es.52 An acute episode often has its onset late in the (4 to 12 hours after exposure), or both, but after asth- day or in the evening after exposure in the workplace. ma is established, the symptoms may be constant. Recurrent episodes become more severe as sensitiza- Episodes with increasing severity of symptoms are tion increases. Early episodes resolve spontaneously, suggestive of immunologic sensitization. and corticosteroids (60 mg of prednisone taken orally The patient’s history can be supplemented by Ma- every day or the equivalent, tapered over a period of terial Safety Data Sheets (from the employer or poi- 10 days) may hasten their resolution, but if episodes son-control centers), which list the chemical com- are allowed to recur repeatedly, permanent damage ponents of commercial materials. The results of caused by interstitial fibrosis may result. previous measurements of air from the patient’s work Clinically and pathologically similar to sarcoidosis, area may be obtained,57 or the physician may advise beryllium disease can be diagnosed on the basis of a the employer to consult a certified industrial hygien- history of exposure, the finding of noncaseating gran- ist for air measurements and remediation.58 Referral ulomas in lung tissue, and T-lymphocyte sensitiza- to an occupational-medicine specialist may be help- tion to beryllium in peripheral blood or fluid from ful for patients whose illness requires detailed inves- bronchoalveolar lavage.53 Systemic corticosteroids are tigation. beneficial for relieving acute symptoms, but their Employers often wonder why disease develops in long-term efficacy is uncertain. only one of many similarly exposed persons. Poly- morphisms in genes produce variation in the rate Inhalation Fever and pathway of metabolism, which results in marked Inhalation fever is characterized by self-limiting fe- differences in susceptibility to occupational substances ver after a single exposure, with peripheral leukocy- among workers, just as polymorphisms affect respons- tosis but minimal or no lung inflammation (Table 4). es to some medications. Interactions among occu- Cytokines from resident lung cells cause demargin- pational exposure, atopic predisposition, nutrition,59 ation of pulmonary vascular leukocytes, and prosta- home or avocational exposure, and host factors (such noids that are produced by these cells act at the hy- as gastroesophageal reflux, cigarette smoking, and vi- pothalamus to produce the febrile episode, which ral infections) are common, and they help to explain usually begins up to 6 hours after exposure and lasts the occurrence of disease in certain persons. A life- less than 12 hours.49 The typical history of exposure time occupational history is essential for the recog-

Volume 342 Number 6 · 411

nition of some exposures, because 10 or more years 9. Woodin M, Hauser R, Liu Y, et al. Molecular markers of acute upper airway inflammation in workers exposed to fuel-oil ash. Am J Respir Crit pass from the initial exposure until the first manifes- Care Med 1998;158:182-7. tations of asbestosis, silicosis, disease from coal dust, 10. Bascom R, Shusterman D. Occupational and environmental exposures lung disease from beryllium, and other illnesses. Im- and the upper respiratory tract. In: Naclerio RM, Durham SR, Mygind N, eds. Rhinitis: mechanisms and management. Vol. 123 of Lung biology in migrants may have had heavy exposure while work- health and disease. New York: Marcel Dekker, 1999:65-94. ing in unregulated industry in developing countries. 11. Chlorine gas toxicity from mixture of bleach with other cleaning products — California. MMWR Morb Mortal Wkly Rep 1991;40:619-21, 627- Treatment is usually identical to treatment for non- 9. [Errata, MMWR Morb Mortal Wkly Rep 1991;40:646, 819.] occupational forms of the illness. With the patient’s 12. Bascom R, Kesavanathan J, Permutt T, Fitzgerald TK, Sauder L, Swift permission, the employer should be notified of the DL. Tobacco smoke upper respiratory response relationships in healthy nonsmokers. Fundam Appl Toxicol 1996;29:86-93. hazard in order to prevent further occurrences, to 13. Welding. In: Chromium, nickel, and welding. IARC monographs on initiate compensation, and to allow recording of the the evaluation of carcinogenic risks to humans. Vol. 49. Lyons, France: In- illness in logs, an action that is required by the Oc- ternational Agency for Research on Cancer, 1990:447-525. 14. Imbus HR, Dyson WL. A review of nasal cancer in furniture manu- cupational Safety and Health Administration. If an facturing and woodworking in North Carolina, the United States, and oth- ongoing hazard is suspected, health departments, the er countries. J Occup Med 1987;29:734-40. 15. Litovitz TL, Bailey KM, Schmitz BF, Holm KC, Klein-Schwart W. Occupational Safety and Health Administration, and 1990 Annual report of the American Association of Poison Control Cen- the National Institute for Occupational Safety and ters National Data Collection System. Am J Emerg Med 1991;9:461-509. Health can be invited by the employer, employee, or 16. Perkner JJ, Fennelly KP, Balkissoon R, et al. Irritant-associated vocal cord dysfunction. J Occup Environ Med 1998;40:136-43. physician to conduct an investigation. Physicians are 17. Chan-Yeung M, Malo J-L. Occupational asthma. N Engl J Med 1995; required by public-health law in 23 states to report 333:107-12. certain specific illnesses or all occupational lung dis- 18. De Haller R, Reid L. Adult chronic bronchitis: morphology, histo- chemistry, and vascularisation of the bronchial mucous glands. Med Tho- 60 eases to the state health department. rax 1965;22:549-67. Management of the disease includes either a change 19. Morgan WKC. Industrial bronchitis. Br J Ind Med 1978;35:285-91. 20. Fishwick D, Bradshaw LM, D’Souza W, et al. Chronic bronchitis, in the work practices that led to the disease or re- shortness of breath, and airway obstruction by occupation in New Zealand. strictions regarding what the patient can and cannot Am J Respir Crit Care Med 1997;156:1440-6. do. These measures can be discussed with both the 21. Viegi G, Prediletto R, Paoletti P, et al. Respiratory effects of occupational exposure in a general population sample in north Italy. Am Rev Res- patient and the employer in order to prevent future pir Dis 1991;143:510-5. harmful exposures. Reasonable accommodation to al- 22. Korn RJ, Dockery DW, Speizer FE, Ware JH, Ferris BG Jr. Occupa- low a patient with a disability to keep working is now tional exposures and chronic respiratory symptoms: a population-based study. Am Rev Respir Dis 1987;136:298-304. encouraged by the Americans with Disabilities Act. 23. Harber P, Barnhart S, Boehlecke B, et al. Respiratory protection A work-related illness often threatens the patient’s guidelines: this official statement of the American Thoracic Society was adopted by the ATS board of directors, March 1996. Am J Respir Crit Care livelihood and health, and physicians can be instru- Med 1996;154:1153-65. mental in providing medical guidance for a transfer 24. Wright JL. Inhalational lung injury causing bronchiolitis. Clin Chest within the same workplace or, if necessary, job re- Med 1993;14:635-44. 25. Chan-Yeung M, Malo J-L. Aetiologic agents in occupational asthma. training through services provided by the govern- Eur Respir J 1994;7:346-71. ment or the insurance company. 26. Hendrick DJ. Occupation and chronic obstructive pulmonary disease. Thorax 1996;51:947-55. 27. Christiani DC, Ye TT, Wegman DH, Eisen EA, Dai HL, Lu PL. Cot-

Supported by grants from the National Institute of Environmental ton dust exposure, across-shift drop in FEV1, and five-year change in lung Health Sciences (P30 ES01247) and the New York State Occupational function. Am J Respir Crit Care Med 1994;150:1250-5. Health Clinic Network. 28. Davison AG, Fayers PM, Taylor AJ, et al. Cadmium fume inhalation and emphysema. Lancet 1988;1:663-7. 29. Brooks SM, Hammad Y, Richards I, Giovinco-Barbas J, Jenkins K. I am indebted to Drs. Martin Cherniack, David Hendrick, and The spectrum of irritant-induced asthma: sudden and not-so-sudden onset Lee Petsonk for critical review of the manuscript. and the role of allergy. Chest 1998;113:42-9. 30. Bignon J, Aron Y, Ju LY, et al. HLA class II alleles in isocyanate- REFERENCES induced asthma. Am J Respir Crit Care Med 1994;149:71-5. 31. Moscato G, Godnic-Cvar J, Maestrelli P, Malo JL, Burge PS, Coifman 1. Menzies D, Bourbeau J. Building-related illnesses. N Engl J Med 1997; R. Statement on self-monitoring of peak expiratory flows in the investiga- 337:1524-31. tion of occupational asthma. Eur Respir J 1995;8:1605-10. 2. Sussman GL, Beezhold DH. Allergy to latex rubber. Ann Intern Med 32. Burge PS. Single and serial measurements of lung function in the di- 1995;122:43-6. agnosis of occupational asthma. Eur J Respir Dis Suppl 1982;123:47-59. 3. McKusick V. Diseases of the genome: an interview with Victor A. 33. Chan-Yeung M. Assessment of asthma in the workplace: ACCP con- McKusick. JAMA 1984;252:1041-8. sensus statement. Chest 1995;108:1084-117. 4. National Institute for Occupational Safety and Health. Work-related 34. Paggiaro PL, Vagaggini B, Bacci E, et al. Prognosis of occupational lung disease surveillance report, 1996. Washington, D.C.: Government asthma. Eur Respir J 1994;7:761-7. Printing Office, 1996. (DHHS (NIOSH) publication no. 96-134.) 35. Sunyer J, Kogevinas M, Kromhout H, et al. Pulmonary ventilatory de- 5. Ross DJ. Ten years of the SWORD project: Surveillance of Work-related fects and occupational exposures in a population-based study in Spain. Am and Occupational Respiratory Disease. Clin Exp Allergy 1999;29:750-3. J Respir Crit Care Med 1998;157:512-7. 6. Kern DG, Crausman RS, Durand KT, Nayer A, Kuhn C III. Flock 36. Andrews JL, Kazemi H, Hardy HL. Patterns of lung dysfunction in worker’s lung: chronic interstitial lung disease in the nylon flocking indus- chronic beryllium disease. Am Rev Respir Dis 1969;100:791-800. try. Ann Intern Med 1998;129:261-72. [Erratum, Ann Intern Med 1999; 37. Cowie RL. The influence of silicosis on deteriorating lung function in 130:246.] gold miners. Chest 1998;113:340-3. 7. Chronic interstitial disease in nylon flocking industry workers — Rhode 38. Beckett WS, Pope CA, Xu X-P, Christiani DC. Women’s respiratory Island, 1992–1996. MMWR Morb Mortal Wkly Rep 1997;46:897-901. health in the cotton textile industry: an analysis of respiratory symptoms in [Erratum, MMWR Morb Mortal Wkly Rep 1997;46:974.] 973 non-smoking female workers. Occup Environ Med 1994;51:14-8. 8. Naclerio R, Solomon W. Rhinitis and inhalant allergens. JAMA 1997; 39. Bouhuys A, Heaphy LJ Jr, Schilling RSF, Welborn JW. Byssinosis in 278:1842-8. the United States. N Engl J Med 1967;277:170-5.

412 · February 10, 2000

40. Report on carcinogens. 8th ed. Research Triangle Park, N.C.: National caused by oil mist exposure in a steel rolling tandem mill. Am J Ind Med Toxicology Program, 1998. 1981;2:51-8. 41. Doll R, Peto R. The causes of cancer: quantitative estimates of avoid- 51. Reynolds HY. Hypersensitivity pneumonitis: correlation of cellular and able risks of cancer in the United States today. J Natl Cancer Inst 1981;66: immunologic changes with clinical phases of disease. Lung 1991;169: 1191-308. S109-S114. 42. Steenland K, Loomis D, Shy C, Simonsen N. Review of occupational 52. Barker RD, van Tongeren MJ, Harris JM, Gardiner K, Venables KM, lung carcinogens. Am J Ind Med 1996;29:474-90. Newman Taylor AJ. Risk factors for sensitisation and respiratory symptoms 43. Weiss W. Cigarette smoke, asbestos, and small irregular opacities. Am among workers exposed to acid anhydrides: a cohort study. Occup Environ Rev Respir Dis 1984;130:293-301. Med 1998;55:684-91. 44. Hirvonen A, Saarikoski ST, Linnainmaa K, et al. Glutathione S-trans- 53. Newman LS, Rose CS, Maier LA. Sarcoidosis. N Engl J Med 1997; ferase and N-acetyltransferase genotypes and asbestos-associated pulmonary 336:1224-34. [Erratum, N Engl J Med 1997;337:139.] disorders. J Natl Cancer Inst 1996;88:1853-6. 54. Pickering CA. Humidifier fever. Eur J Respir Dis Suppl 1982;123:104-7. 45. Lilienfeld DE, Mandel JS, Coin P, Schuman LM. Projection of asbes- 55. Brubaker RE. Pulmonary problems associated with the use of polytet- tos related diseases in the United States, 1985-2009. I. Cancer. Br J Ind rafluoroethylene. J Occup Med 1977;19:693-5. Med 1988;45:283-91. 56. Newman LS. Occupational illness. N Engl J Med 1995;333:1128-34. 46. Peto J, Hodgson JT, Matthews FE, Jones JR. Continuing increase in 57. Himmelstein JS, Frumkin H. The right to know about toxic exposures: mesothelioma mortality in Britain. Lancet 1995;345:535-9. implications for physicians. N Engl J Med 1985;312:687-90. 47. Morgan WKC, Donner A, Higgins ITT, Pearson MG, Rawlings W Jr. 58. American Industrial Hygiene Association membership directory: CIH The effects of kaolin on the lung. Am Rev Respir Dis 1988;138:813- after the name designates a certified industrial hygienist. Fairfax, Va.: Amer- 20. ican Industrial Hygiene Association, 1999. (See http://aiha.ags.com.) 48. Taylor AN, Pickering CAC, Turner-Warick M, Pepys J. Respiratory al- 59. Chuwers P, Barnhart S, Blanc P, et al. The protective effect of beta- lergy to a factory humidifier contaminant presenting as pyrexia of undeter- carotene and retinol on ventilatory function in an asbestos-exposed cohort. mined origin. BMJ 1978;2:94-5. Am J Respir Crit Care Med 1997;155:1066-71. 49. Blanc P, Wong H, Bernstein MS, Boushey HA. An experimental hu- 60. Freund E, Seligman PJ, Chorba TL, Safford SK, Drachman JG, Hull man model of metal fume fever. Ann Intern Med 1991;114:930-6. HF. Mandatory reporting of occupational diseases by clinicians. JAMA 50. Cullen MR, Balmes JR, Robins JM, Smith GJW. Lipoid pneumonia 1989;262:3041-4.

POSTING PRESENTATIONS AT MEDICAL MEETINGS ON THE INTERNET Posting an audio recording of an oral presentation at a medical meeting on the Internet, with selected slides from the presentation, will not be considered prior publication. This will allow students and physicians who are unable to attend the meeting to hear the presentation and view the slides. If there are any questions about this policy, authors should feel free to call the Journal’s Editorial Offices.

Volume 342 Number 6 · 413

Downloaded from www.nejm.org at UNIV OF MEDICINE AND DENTISTRY OF NJ on February 16, 2007 . Copyright © 2000 Massachusetts Medical Society. All rights reserved. Preoperative Assessment of Pulmonary Risk

Mark K. Ferguson

Chest 1999;115;58-63 DOI 10.1378/chest.115.suppl_2.58S The online version of this article, along with updated information and services can be found online on the World Wide Web at: http://chestjournals.org/cgi/content/abstract/115/suppl_2/58S

CHEST is the official journal of the American College of Chest Physicians. It has been published monthly since 1935. Copyright 2007 by the American College of Chest Physicians, 3300 Dundee Road, Northbrook IL 60062. All rights reserved. No part of this article or PDF may be reproduced or distributed without the prior written permission of the copyright holder (http://www.chestjournal.org/misc/reprints.shtml). ISSN: 0012-3692.

Mark K. Ferguson, MD, FCCP

Study objectives: A summary of current modalities for and the utility of preoperative assessment of pulmonary risk. Design: Review of recent literature published in the English language. Setting: Not applicable. Patients or participants: Patients who undergo elective cardiothoracic or abdominal operations. Interventions: Not applicable. Measurements and results: Postoperative pulmonary complications occur after 25 to 50% of major surgical procedures. The accuracy of the preoperative assessment of the risk of such complications is only fair. The routine assessment for all preoperative patients includes age, general physiologic status, and the nature of the planned operation. Specific tests such as measurement of spirometric values and diffusing capacity are indicated routinely only for patients who are candidates for major lung resection or esophagectomy. Conclusions: Pulmonary complications are an important form of postoperative morbidity after major cardiothoracic and abdominal operations. The appropriate preoperative assessment of the risk of such complications is well defined for lung resection and esophagectomy operations, but it requires refinement for general surgical and cardiovascular operations. (CHEST 1999; 115:58S–63S)

ulmonary complications are the most common patients are at increased risk for postoperative pul- P form of postoperative morbidity experienced by monary complications and mortality. patients who undergo general surgical abdominal procedures and thoracotomy, and frequently occur Pathophysiology of Postoperative after cardiac surgical operations. The cost of postop- Respiratory Complications erative pulmonary complications was well recognized at the beginning of the 20th century, at which time a A prescient commentary in 1910 by W. Pasteur9 number of clinical reviews identified the mortality pointed the direction to our current understanding rate associated with the development of postopera- of the etiology of postoperative pulmonary compli- tive pneumonia among Ͼ 40,000 patients to be in cations. He noted that “when the true history of excess of 40%.1–8 In addition to pneumonia, postop- postoperative lung complications comes to be writ- erative pulmonary complications include massive ten, active collapse of the lung, from deficiency of lobar collapse due to mucus plugging of a central inspiratory power, will be found to occupy an impor- airway, pneumonitis, atelectasis, and a combination tant position among determining causes.”9 Most of one or more of these or other less common postoperative pulmonary complications develop as a problems that results in respiratory insufficiency. result of changes in lung volumes that occur in Because of the high incidence of these complica- response to dysfunction of muscles of respiration and tions and their associated costs such as prolonged other changes in chest wall mechanics. Abdominal hospital stay and mortality, substantial effort has and thoracic surgical procedures cause large reduc- been made during the 20th century to predict which tions in vital capacity and smaller but crucial reduc- patients are at increased risk for developing such tions in functional residual capacity (FRC), which complications and to identify techniques that can be has been recognized for decades as the single most used to prevent them. This article will focus on important lung volume measurement involved in the methods that are currently used to predict which etiology of respiratory complications.10 Although no consistent changes occur in FRC after nonabdomi- *From the Department of Surgery, the University of Chicago, nal, nonthoracic surgery, FRC decreases after lower Chicago, IL. abdominal operations by 10 to 15%, by 30% after Correspondence to: Mark K. Ferguson, MD, FCCP, 5841 S Maryland Ave MC5035, Chicago, IL 60637; e-mail: mferguso@ upper abdominal operations, and by 35% after tho- surgery.bsd.uchicago.edu racotomy and lung resection.11–16 Other factors that

58S Perioperative Cardiopulmonary Evaluation and Management Downloaded from chestjournals.org on February 17, 2007 Copyright © 1999 by American College of Chest Physicians decrease FRC include the supine position, obesity, accurate for surgical candidates at the far ends of the the presence of ascites, the development of perito- body mass spectrum, further refinement of these nitis, and general anesthesia. measurements has included expressing them as a The other important element in the etiology of percentage of predicted based on patient age, sex, postoperative respiratory complications is the closing and height.22,24,25 The calculation of postoperative volume (CV), which is the lung volume at which the predicted values for both spirometric raw numbers flow from the dependent parts of the lungs stops and percentage of predicted values has further in- during expiration because of airway closure. Factors creased the accuracy of spirometry as a preoperative that promote an increase in CV include advanced tool for evaluating pulmonary risk preoperative- age, tobacco use, fluid overload, bronchospasm, and ly.25–27 This calculation is usually performed by the presence of airway secretions. estimating the number of functional lung segments Under normal circumstances, FRC is about 50% that will remain postoperatively. Quantitative venti- and CV is about 30% of total lung capacity. When lation-perfusion scans used to assess regional lung FRC is reduced or CV is increased, portions of the function have aided considerably in the calculation of lung are subject to premature airway closure and predicted postoperative spirometric function in pa- atelectasis. This causes ventilation-perfusion mis- tients who are considered borderline candidates for match resulting in hypoxemia and promotes the operation based on standard techniques.28,29 trapping of secretions resulting in pneumonitis, all of In addition to these standard methods, other which may combine to cause respiratory insuffi- measures of gas exchange and oxygen consumption ciency. have also proved useful in the preoperative assessment of risk. These include clinical assessments such as the 6-min walk distance and stair climbing effort Thoracotomy and Lung Resection and laboratory measures of exercise capacity such as maximum oxygen consumption during exercise The incidence of postoperative pulmonary compli- 30–33 (V˙ o2max). All have shown some promise in the cations after thoracotomy and lung resection is about prediction of postoperative pulmonary complications 30% and is related not only to the removal of lung and, in some settings, postoperative mortality. Mea- tissue but is also caused by alterations in chest wall surement of gas exchange capacity using diffusing 17–21 mechanics due to the thoracotomy itself. All capacity of the lung for carbon monoxide (Dlco) has spirometric measurements fall precipitously imme- proved to be an independent and useful means of diately postoperatively and do not return toward estimating operative risk for patients undergoing 16 normal until 6 to 8 weeks postoperatively. major lung resection. Preoperative raw values or Knowledge about the utility of preoperative as- values expressed as a percent of predicted (Dlco%) sessment of the lung resection candidate was first as well as calculated postoperative values expressed developed in the 1950s and further refinement has as a percent of predicted function have all been taken place since then (Table 1). Early methods of shown to be useful, although the best value to use is evaluating risk included the measurement of bellows the calculated postoperative Dlco expressed as a function of the lungs such as maximum voluntary percent of predicted (ppoDlco%).26,34,35 In patients 22 ventilation and FRC. The latter continues to be preselected as adequate candidates for lung resec- important for this purpose. Air flow parameters that tion on the basis of spirometry, the risk of pulmonary are useful include FEV1 and forced expiratory flow complications is best defined by patient age and rate in the middle 50% of the forced expiratory flow ppoDlco%.35 A direct comparison between the use curve.23,24 of Dlco% and V˙ o2max revealed that Dlco% was a Because raw spirometric values are relatively in- better predictor of pulmonary complications after lung resection.21 There have been important advances in the selec- Table 1—Preoperative Tests for Assessing Pulmonary tion and postoperative care of the lung resection Risk Prior to Major Lung Resection patient since the time most of the data noted previ- Value Range for ously were derived. Postoperative analgesia with Test Low-risk Patients epidural catheters or patient-controlled delivery de- Ͼ vices has substantially reduced surgical pain. Vigor- FEV1% 60 % Dlco% Ͼ 60 % ous pulmonary toilet exercises are used more rou- Ͼ ppoFEV1 800 mL tinely and frequently. Experience in lung volume Ͼ ppoFEV1% 40 % reduction surgery and lung transplantation has in- ppoDlco% Ͼ 40 % creased our knowledge of how to treat critically ill ˙ Ͼ Vo2max during exercise 15 mL/kg/min patients with end-stage emphysema. There have also

CHEST / 115/5/MAY, 1999 SUPPLEMENT 59S Downloaded from chestjournals.org on February 17, 2007 Copyright © 1999 by American College of Chest Physicians been changes in the way in which lung resection an incidence of left lower lobe collapse of Ͼ 80% operations are performed. The use of muscle-sparing compared with only 32% in patients in whom no thoracotomy reduces postthoracotomy pain, retains slush is used.42 shoulder girdle muscle strength, and may permit The other unique factor that is associated with the improved spirometric function in the early postop- development of pulmonary complications is the use erative period compared with a standard lateral of cardiopulmonary bypass. Within 24 h of surgery, thoracotomy.17,36 Further improvements such as there is a reduction in arterial oxygen tension of Ͼ these may be evident with additional experience 30%, an increase in the alveolar-arterial oxygen Ͼ using thoracoscopic lung resection techniques. gradient of 150%, and an increase in the pulmo- At the present time, the risk of postoperative nary shunt fraction from a baseline of 3% to 19%. pulmonary complications in the candidate for lung These changes only partially resolve by the end of the first postoperative week and eventually return to resection should be evaluated with age and perfor- baseline values after 6 weeks.39 The only predictor of mance status during the initial history and physical this complication is a preoperative abnormality of the examination. Based on the extent of planned lung alveolar-arterial oxygen gradient. The presumed eti- resection, postoperative predicted spirometry and ology of this profound dysfunction is the activation of diffusing capacity are calculated (Table 1). For high- a multitude of inflammatory mediators in addition to ˙ risk patients, an additional assessment of Vo2max the factors mentioned above. may be useful. Conclusions about the utility of The overall preoperative assessment of pulmonary muscle-sparing and thoracoscopic approaches await risk in a patient who is to undergo cardiac surgery is further data. based more on the planned operation and less on the patient’s preoperative status than for any other preoperative assessment. Issues of critical importance Cardiac Surgery other than patient age and performance status are the choice of conduit if the patient is having coronary The incidence of pulmonary complications after artery bypass grafting, the technique used for myo- cardiac surgical procedures is high and includes cardial protection, and possibly the duration of car- pneumonitis, bronchospasm, or lobar collapse in diopulmonary bypass. Whether the minimally inva- 40%, prolonged mechanical ventilation in 5 to 10%, sive approaches to bypass grafting and valve repair or and generalized respiratory dysfunction in most pa- replacement will reduce the incidence of postoper- tients who undergo cardiopulmonary bypass.37–39 ative pulmonary complications is as yet unknown. The etiology of pulmonary complications in patients who undergo cardiac surgery has some factors that are similar to those that have been identified for Esophagectomy pulmonary complications that develop after lung resection, specifically alterations in chest wall me- Postoperative pulmonary complications occur in chanics due to the incision. FRC is decreased by 25 to 50% of patients after esophagectomy.43–45 nearly 20% at the time of hospital discharge but is These complications arise from a number of factors, normal at 3 months after the operation. Interest- including the type of incision used, the extent of ingly, whether an internal mammary artery is used mediastinal dissection, the development of a recur- for bypass grafting has an important impact on rent laryngeal nerve injury that may impair coughing respiratory function postoperatively. Increasing age efficiency postoperatively, and the presence of an and the use of an internal mammary artery graft have intrathoracic reconstructive organ or pleural effusion significant and independent negative impacts on that may directly impair ventilation in the early spirometric values postoperatively.40,41 In contrast to postoperative period. lung resection patients, however, the prediction of The risk of pulmonary complications after esoph- pulmonary complications after cardiac surgery is not agectomy is predicted on the basis of a number of aided by preoperative measurement of lung volumes preoperative factors, including patient age, spiromet- and flows.37 ric values, diffusing capacity, performance status, Two unique factors contribute to the development nutritional status, and a diagnosis of COPD.43–45 of pulmonary complications after cardiac surgery. Intraoperative factors also strongly predict the like- The first of these is the use of topical slush to protect lihood of pulmonary complications. An increase in the myocardium, which results in phrenic nerve complications is associated with an increased volume paralysis in Ͼ 30% of patients compared with an of blood loss, use of the substernal rather than the incidence of Ͻ 5% in patients in whom no topical posterior mediastinal route for esophageal recon- slush is used. The use of slush is also associated with struction, and routine use of ventilatory support

60S Perioperative Cardiopulmonary Evaluation and Management Downloaded from chestjournals.org on February 17, 2007 Copyright © 1999 by American College of Chest Physicians rather than early extubation postoperatively.43,44,46 which patients are at greatest risk has enjoyed wide- The type of incision used to perform the resection is spread popularity, but its predictive value when used also a predictor of the likelihood of postoperative routinely is unproved.55 Clinical factors that have pulmonary complications. Use of an isolated left been shown to be useful in the prediction of postthoracotomy results in fewer complications than operative pulmonary complications include a history does an Ivor Lewis approach combining a right of smoking, chronic bronchitis, airflow obstruction, thoracotomy and laparotomy. The Ivor Lewis ap- obesity, and a prolonged preoperative hospital proach is associated with fewer complications than is stay.50,56,57 The presence of colonizing bacteria in the a transhiatal approach, in which a laparotomy and stomach and the use of nasogastric intubation in- cervical incision are performed and no thoracotomy crease the specific risk of postoperative pneumo- is necessary.44,47 The development of pulmonary nia.50,58 Smaller incisions and the use of laparoscopic complications is associated with a sevenfold increase techniques promise to reduce the incidence of pul- in the risk of operative mortality, and pulmonary monary complications by preventing substantial re- complications account for 40 to 60% of operative ductions in pulmonary function postoperatively, but mortality.43–45,48 the data supporting these outcomes are scant at the Because of the high incidence of pulmonary com- present time.59 The most important predictive fac- plications and associated operative mortality after tors appear to be the overall condition of the patient esophagectomy, a thorough preoperative evaluation (based on the classification of the American Society of pulmonary risk is appropriate in candidates for of Anesthesiologists) and patient age.52 esophagectomy. The evaluation should include a Based on available information, the preoperative general assessment of age, performance and nutri- evaluation of pulmonary risk in the candidate for tional status, measurement of spirometric values, and abdominal surgery should include an assessment of an assessment of diffusing capacity. Knowledge of patient age, general performance status, relative the planned approach to resection and the route to weight, pulmonary comorbid conditions, the planned be used for reconstruction will also provide useful operation, and the incision that is to be used. information regarding the risk of postoperative pul- Spirometry is indicated in patients in whom severe monary complications. pulmonary dysfunction is evident as a means to assess whether a period of pulmonary rehabilitation is indicated to improve the preoperative pulmonary Abdominal Surgery condition prior to an elective operation. The incidence of pulmonary complications after abdominal surgery is about 30%, a frequency that is References high enough to have stimulated considerable re- 1 Armstrong GE. Lung complaints after operations with anaes- 49–53 search into the etiology of this problem. In thesia. BMJ 1906; 1:1141–1142 addition to dysfunction of abdominal wall muscula- 2 Bancroft FW. A review of the postoperative cases of pneu- ture, the supine position, the development of ascites, monia in the second surgical service of the New York and other factors that reduce FRC postoperatively Hospital. Med Record 1916; 89:583 3 Lawen. Uber lungenkomplikationen nach bauchoperationen. after laparotomy, abdominal surgery has the unique Zentralbl Chir 1907; 34:96–97 propensity to impair diaphragmatic function, an ef- 4 Beckman EH. Complications following surgical operations. fect that further contributes to the reduction in Surg Gynecol Obstet 1914; 18:551–555 FRC. Transdiaphragmatic pressure decreases by al- 5 Booth LS. Discussion of LeWald LT. Differential diagnosis most 70% on the first postoperative day and does not between pneumonia and acute appendicitis made by roent- gen examination before physical signs of pneumonia ob- return to normal until at least 1 week postopera- tained. Med Record 1916; 89:582–583 54 tively. Adequate relief of postoperative pain does 6 Cutler EC, Morton JJ. Postoperative pulmonary complica- not reduce this impairment that appears to result tions. Surg Gynecol Obstet 1917; 25:621–649 from dysfunction of the diaphragm itself rather than 7 Whipple AO. A study of postoperative pneumonitis. Surg from phrenic nerve or central neural sources. Upper Gynecol Obstet 1918; 26:29–47 8 Cleveland M. Further studies in postoperative pneumonitis. abdominal operations are associated with substan- Surg Gynecol Obstet 1919; 28:282–293 tially worse diaphragmatic function postoperatively 9 Pasteur W. Active lobar collapse of the lung after abdominal than are lower abdominal operations, and the risk of operations. Lancet 1910; 2:1080–1083 postoperative pulmonary complications is accord- 10 Beecher HK. The measured effect of laparotomy on the ingly higher by a factor of 1.5.50 respiration. J Clin Invest 1933; 12:639–650 11 Ali J, Weisel RD, Layug AB, et al. Consequences of postop- The accurate preoperative prediction of pulmo- erative alterations in respiratory mechanics. Am J Surg 1974; nary risk associated with abdominal surgery has been 128:376–382 somewhat elusive. The use of spirometry to assess 12 Meyers JR, Lembeck L, O’Kane H, et al. Changes in

CHEST / 115/5/MAY, 1999 SUPPLEMENT 61S Downloaded from chestjournals.org on February 17, 2007 Copyright © 1999 by American College of Chest Physicians functional residual capacity of the lung after operation. Arch predicts morbidity and mortality after pulmonary resection. Surg 1975; 110:576–583 J Thorac Cardiovasc Surg 1988; 96:894–900 13 Alexander JI, Spence AA, Parikh RK, et al. The role of airway 35 Ferguson MK, Reeder LB, Mick R. Optimizing selection of closure in postoperative hypoxaemia. Br J Anaesth 1973; patients for major lung resection. J Thorac Cardiovasc Surg 45:34–40 1995; 109:275–283 14 Vaughan RW, Wise L. Choice of abdominal operative incision 36 Lemmer JH Jr, Gomez MN, Symreng T, et al. Limited lateral in the obese patient: a study using blood gas measurements. thoracotomy. Arch Surg 1990; 125:873–877 Ann Surg 1975; 181:829–835 37 Warner MA, Divertie MB, Tinker JH. Preoperative cessation 15 Craig DB. Postoperative recovery of pulmonary function. of smoking and pulmonary complications in coronary artery Anesth Analg 1981; 60:46–52 bypass patients. Anesthesiology 1984; 60:380–383 16 Bastin R, Moraine J-J, Bardocsky G, et al. Incentive spirom- 38 Hammermeister KE, Burchfiel C, Johnson R, et al. Identifi- etry performance. Chest 1997; 111:559–563 cation of patients at greatest risk for developing major 17 Hazelrigg SR, Landreneau RJ, Boley TM, et al. The effect of complications at cardiac surgery. Circulation 1990; 82 (suppl muscle-sparing versus standard posterolateral thoracotomy IV):IV-380–IV-389 on pulmonary function, muscle strength, and postoperative 39 Taggart DP, El-Fiky M, Carter R, et al. Respiratory dysfunc- pain. J Thorac Cardiovasc Surg 1991; 101:394–401 tion after uncomplicated cardiopulmonary bypass. Ann Tho- 18 Busch E, Verazin G, Antkowiak JG, et al. Pulmonary compli- rac Surg 1993; 56:1123–1128 cations in patients undergoing thoracotomy for lung carci- 40 Shapira N, Zabatino SM, Ahmed S, et al. Determinants of noma. Chest 1994; 105:760–766 pulmonary function in patients undergoing coronary bypass 19 Bolliger CT, Wyser C, Roser H, et al. Lung scanning and operations. Ann Thorac Surg 1990; 50:268–273 exercise testing for the prediction of postoperative perfor- 41 Berrizbeitia LD, Tessler S, Jacobowitz IJ, et al. Effect of mance in lung resection candidates at increased risk for sternotomy and coronary bypass surgery on postoperative complications. Chest 1995; 108:341–348 pulmonary mechanics. Chest 1989; 96:873–876 20 Epstein SK, Faling J, Daly BDT, et al. Inability to perform 42 Efthimiou J, Butler J, Woodham C, et al. Diaphragm paralysis bicycle ergometry predicts increased morbidity and mortality following cardiac surgery: role of phrenic nerve cold injury. after lung resection. Chest 1995; 107:311–316 Ann Thorac Surg 1991; 52:1005–1008 21 Wang J, Ultman R, Olak J, et al. Prospective trial of diffusing 43 Law SYK, Fok M, Wong J. Risk analysis in resection of capacity and oxygen consumption in the prediction of pulmo- squamous cell carcinoma of the esophagus. World J Surg nary complications after lung resection. Chest 1997; 112:153S 1994; 18:339–346 22 Gaensler EA, Cugell DW, Lindgren I, et al. The role of 44 Ferguson MK, Martin TR, Reeder LB, et al. Determinants pulmonary insufficiency in mortality and invalidism following of pulmonary complications following esophagectomy. In: surgery for pulmonary tuberculosis. J Thorac Cardiovasc Surg Peracchia A, Rosati R, Bonavina L, et al, eds. Recent 1955; 29:163–187 advances in diseases of the esophagus. Milan, Italy: Monduzzi 23 Boushy SF, Billig DM, North LB, et al. Clinical course Editore, 1996; 527–532 related to preoperative and postoperative pulmonary function 45 Hennessy TPJ. Respiratory complications in oesophageal in patients with bronchogenic carcinoma. Chest 1971; 59: surgery. In: Peracchia A, Rosati R, Bonavina L, et al, eds. 383–391 Recent advances in diseases of the esophagus. Milan, Italy: 24 Olsen GN, Block AJ, Swenson EW, et al. Pulmonary function Monduzzi Editore, 1996; 533–535 evaluation of the lung resection candidate: a prospective 46 Bartels H, Stein HJ, Siewert JR. Early extubation versus study. Am Rev Respir Dis 1975; 111:379–387 prolonged ventilation after esophagectomy: a randomized 25 Putnam JB Jr, Lammermeier DE, Colon R, et al. Predicted prospective study. In: Peracchia A, Rosati R, Bonavina L, et pulmonary function and survival after pneumonectomy for al, eds. Recent advances in diseases of the esophagus. Milan, primary lung carcinoma. Ann Thorac Surg 1990; 49:909–915 Italy: Monduzzi Editore, 1996; 537–539 26 Markos J, Mullan BP, Hillman DR, et al. Preoperative 47 Stark SP, Romberg MS, Pierce GE, et al. Transhiatal versus assessment as a predictor of mortality and morbidity after transthoracic esophagectomy for adenocarcinoma of the distal lung resection. Am Rev Respir Dis 1989; 139:902–910 esophagus and cardia. Am J Surg 1996; 172:478–482 27 Wahi R, McMurtrey MJ, DeCaro LF, et al. Determinants of 48 Ferguson MK, Martin TR, Reeder LB, et al. Mortality after perioperative morbidity and mortality after pneumonectomy. esophagectomy: risk factor analysis. World J Surg 1997; Ann Thorac Surg 1989; 48:33–37 21:599–604 28 Ali MK, Ewer MS, Atallah MR, et al. Regional and overall 49 Lyager S, Wernberg M, Rajani N, et al. Can postoperative pulmonary function changes in lung cancer. J Thorac Cardio- pulmonary conditions be improved by treatment with the vasc Surg 1983; 86:1–8 Bartlett-Edwards incentive spirometer after upper abdominal 29 Bria WF, Kanarek DJ, Kazemi H. Prediction of postoperative surgery? Acta Anaesthesiol Scand 1979; 23:312–319 pulmonary function following thoracic operations. J Thorac 50 Mitchell C, Garrahy P, Peake P. Postoperative respiratory Cardiovasc Surg 1983; 86:186–192 morbidity: identification and risk factors. AustNZJSurg 30 Marshall MC, Olsen GN. The physiologic evaluation of the 1982; 52:203–209 lung resection candidate. Clin Chest Med 1993; 14:305–320 51 Celli BR, Rodriguez KS, Snider GL. A controlled trial of 31 Olsen GN, Bolton JW, Weiman DS, et al. Stair climbing as an intermittent positive pressure breathing, incentive spirome- exercise test to predict the postoperative complications of try, and deep breathing exercises in preventing pulmonary lung resection. Chest 1991; 99:587–590 complications after abdominal surgery. Am Rev Respir Dis 32 Smith TP, Kinasewitz GT, Tucker WY, et al. Exercise capac- 1984; 130:12–15 ity as a predictor of postthoracotomy morbidity. Am Rev 52 Hall JC, Tarala RA, Hall JL, et al. A multivariate analysis of Respir Dis 1984; 129:730–734 the risk of pulmonary complications after laparotomy. Chest 33 Bechard D, Wetstein L. Assessment of exercise oxygen 1991; 99:923–927 consumption as preoperative criterion for lung resection. Ann 53 Kocabas A, Kara K, Ozgur G, et al. Value of preoperative Thorac Surg 1987; 44:344–349 spirometry to predict postoperative pulmonary complications. 34 Ferguson MK, Little L, Rizzo L, et al. Diffusing capacity Respir Med 1996; 90:25–33

62S Perioperative Cardiopulmonary Evaluation and Management Downloaded from chestjournals.org on February 17, 2007 Copyright © 1999 by American College of Chest Physicians 54 Simonneau G, Vivien A, Sartene R, et al. Diaphragm dys- upper abdominal surgery: an important problem for smokers. function induced by upper abdominal surgery. Am Rev Respir Med 1992; 86:205–210 Respir Dis 1983; 128:899–903 58 Ephgrave KS, Kleiman-Wexler R, Pfaller M, et al. Postoper- 55 Lawrence VA, Page CP, Harris GD. Preoperative spirometry ative pneumonia: a prospective study of risk factors and before abdominal operations. Arch Intern Med 1989; 149: morbidity. Surgery 1993; 114:815–821 280–285 59 McMahon AJ, Russell IT, Ramsay G, et al. Laparoscopic and 56 Garibaldi RA, Britt MR, Coleman ML, et al. Risk factors for minilaparotomy cholecystectomy: a randomized trial compar- postoperative pneumonia. Am J Med 1981; 70:677–680 ing postoperative pain and pulmonary function. Surgery 1994; 57 Dilworth JP, White RJ. Postoperative chest infection after 115:533–539

CHEST / 115/5/MAY, 1999 SUPPLEMENT 63S Downloaded from chestjournals.org on February 17, 2007 Copyright © 1999 by American College of Chest Physicians Preoperative Assessment of Pulmonary Risk Mark K. Ferguson Chest 1999;115;58-63 DOI 10.1378/chest.115.suppl_2.58S This information is current as of February 17, 2007

Updated Information Updated information and services, including & Services high-resolution figures, can be found at: http://chestjournals.org/cgi/content/full/115/suppl_2/58S References This article cites 44 articles, 23 of which you can access for free at: http://chestjournals.org/cgi/content/full/115/suppl_2/58S# BIBL Citations This article has been cited by 5 HighWire-hosted articles: http://chestjournals.org/cgi/content/full/115/suppl_2/58S Permissions & Licensing Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://chestjournals.org/misc/reprints.shtml Reprints Information about ordering reprints can be found online: http://chestjournals.org/misc/reprints.shtml Email alerting service Receive free email alerts when new articles cite this article sign up in the box at the top right corner of the online article. Images in PowerPoint format Figures that appear in CHEST articles can be downloaded for teaching purposes in PowerPoint slide format. See any online article figure for directions.

Preoperative assessment: pulmonary Peter Rock, MD, MBA*, Anthony Passannante, MD Departments of Medicine and Anesthesiology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA

Postoperative pulmonary complications (PPCs) include atelectasis, pneumonia, bronchitis, bronchospasm, hypoxemia, respiratory failure, and prolonged mechanical ventilation. These conditions are a significant source of morbidity and mortality [1,2]. Postoperative changes include diaphragmatic dysfunction, V/Q mismatch, and reductions in functional residual capacity (FRC), which, while measurable, have an unclear relation to morbidity and mortality. The etiology of PPCs is complex and poorly understood. Because it is not practically or ethically possible to study the isolated effects of general anesthesia (GA) on the development of pulmonary complications after surgery, it is difficult to determine the dominant contribution (from anesthesia or from surgery). GA and surgery produce significant changes in the respiratory system, and the interaction of these changes is responsible, along with underlying respiratory conditions, for PPCs. GA decreases the number and activity of alveolar macrophages, inhibits mucociliary clearance, increases alveolar-capillary permeability, inhibits surfactant release, increases the activity of pulmonary nitric oxide synthetase, and enhances the sensitivity of the pulmonary vasculature to neurohumoral mediators [3–22]. GA also results in mechanical and functional changes to the respiratory system that may contribute to the development of PPCs. After induction of GA, FRC decreases and atelectatic plaques form in dependent portions of the lungs [23]. GA also alters diaphragmatic movement with near uniform motion of the diaphragm along the ventral-dorsal axis [24]. This may result in more ventilation of the superior (poorly perfused) portion of the lung and less ventilation of the lung in the dependent (better perfused) portion. The resulting ventilation- perfusion (V/Q) inequality leads to shunt and dead-space ventilation. Regional diaphragmatic anatomic and neural differences result in abnormalities in diaphragmatic position and movement during GA [24]. These changes result in an

* Corresponding author. Department of Anesthesiology, University of North Carolina at Chapel Hill, Campus Box 7010, Chapel Hill, NC 27599. E-mail address: [email protected] (P. Rock).

0889-8537/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/S0889-8537(03)00112-3 78 P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 increase in the alveolar-arterial oxygen gradient and explain why supplemental oxygen is necessary during surgery and GA. Even small concentrations of intravenous or volatile anesthetics (0.1 MAC) blunt the ventilatory response to both hypoxemia and hypercarbia by depression of the peripheral chemoreceptors (which play an important role in the response to hypoxemia) and the central nervous system (CNS) regions that regulate PaCO2 [25–30]. Respiratory drive is determined by the state of arousal, metabolic factors, including pHa, PaCO2, and PaO2, and the interaction between the response to hypoxia and hypercarbia [31]. Depending on the state of arousal and anesthetic agent, sedative concentrations of volatile anesthetics may result in either no change or a depressed response to hypoxemia and hypercarbia. Depression of the ventilatory response to hypoxemia or hypercarbia by subanesthetic concentrations of volatile anesthetics may be important in the pathogenesis and maintenance of postoperative hypoxia. In addition, neuromuscular blockers may reduce the hypoxic drive to ventilate [32]. Because a low level of volatile anesthetic persists for hours, patients are at risk for anesthetic-induced depression of the response to hypoxemia or hypercarbia after they leave the highly monitored operating room and postanesthesia care unit (PACU) environment.

Effects of surgery: postoperative changes Patients who undergo abdominal (upper > lower) and thoracic surgery have a decreased postoperative vital capacity and functional residual capacity (FRC). This decrease in FRC results in V/Q mismatch and contributes to the development of hypoxemia. In patients undergoing laparotomy, FRC decreases to approximately 50% of baseline, returning toward normal over 1 to 2 weeks [33]. Diaphragmatic dysfunction occurs after upper abdominal or thoracic surgery and may help explain PPCs. The diaphragmatic ‘‘weakness’’ is not caused by residual neuromuscular blockade or inadequate pain relief. Phrenic nerve pacing restores diaphragm function suggesting that reflex inhibition of phrenic nerve output results in diaphragmatic dysfunction [34,35]. The hypothesis that phrenic nerve-induced diaphragmatic dysfunction may contribute to atelectasis has been tested and phrenic nerve stimulation during GA reduced atelectasis in human volunteers [36]. Surgical trauma may also increase airway tone and reactivity [37]. Exposure to airway irritants (secretions, infection) in this setting of increased airway reactivity could result in bronchospasm that could lead to atelectasis or pneumonia. Based on this evidence, bronchodilators have a potential role in the postoperative respiratory care of even those patients without asthma or chronic obstructive lung disease (COPD). Because it is ‘‘minimally invasive,’’ laparoscopic surgery might have advantages in patients with underlying lung disease. There is improved FEV1 and FVC, better arterial oxygenation, and improved ventilation after laparoscopic, as compared with open, procedures [38]. But, upper abdominal laparoscopic surgery is associated with dysfunction of the diaphragm [39]. The site of surgery rather than P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 79 the surgical technique is critical in determining whether there will be diaphragmatic dysfunction. Importantly, no studies have demonstrated a reduction in clinical respiratory complications such as pneumonia, bronchitis, or respiratory failure after laparoscopic surgery.

Preoperative assessment To prevent or reduce the incidence of PPC, there must be an understanding of patient conditions that increase the risk of developing PPCs and of effective interventions available to reduce the impact of pre-existing patient conditions on the subsequent development of PPCs. PPCs occur commonly in patients who undergo anesthesia and surgery, especially after upper abdominal or thoracic surgery. Risk factors for PPCs also include cigarette smoking, underlying chronic respiratory disease, emergency surgery, anesthetic time of 180 minutes or more, and perhaps advanced age [40]. Depending on the series, definition of pulmonary complications, and methods used to detect such complications, PPCs occur in approximately 10% to 30% (or even higher in some series) of patients undergoing major, nonthoracic surgery [1,2]. While perioperative cardiac complications are easy to define, a standard definition of exactly what constitutes a PPC does not exist. The lack of a standard definition for PPCs hinders comparisons between series. Numerous studies have confirmed that the presence of underlying lung disease, specifically COPD, places the patient at risk for the development of PPCs. What is it about COPD that increases the risk of PPCs? In smokers undergoing abdominal surgery the presence of obstructive lung disease predicts bronchospasm but not prolonged endotracheal intubation or ICU admission [41]. Thus, the increased frequency of PPCs in patients with COPD may be caused by the combination of pre-existing pulmonary disease in addition to other co-morbidities (eg, cardiovascular disease) rather than by the isolated presence of airway obstruction. This finding suggests there are additional, currently unidentified risk factors that contribute to the development of PPCs. It also suggests that in the future one may be able to better predict which patients will develop PPCs, and the frequency with which an individual patient will develop PPC. Similarly, the care of patients undergoing surgery, and resource allocation decisions regarding the necessity for postoperative ICU care would be improved if there were an effective way to triage to the intensive care unit patients that are at high risk of developing PPCs. Patients with a longer history of smoking, lower preoperative arterial PaO2, and greater intraoperative blood loss have been found to require mechanical ventilation for more than 24 hours after abdominal vascular surgery [42]. Perhaps surprisingly, risk factors for the development of PPCs are primarily clinical in nature. Although laboratory testing can predict patients at risk for PPCs, these tests have not been shown to have superior sensitivity or specificity compared with clinical observations. No studies have suggested a pulmonary function test (PFT) value that would contraindicate surgery. In fact, empiric ob- 80 P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 servations suggest that even patients with significant obstructive lung disease can successfully undergo anesthesia and major surgery [43]. Lung volume reduction surgery (LVRS) is successfully performed on patients with severe lung impairment [44]. Consideration of the foregoing risk factors for the development of PPCs suggests patients undergoing LVRS are at high risk: they are older, have significant underlying lung disease, and are undergoing surgery associated with a significant risk for PPCs. The results of such surgery emphasize that assessment of patients with lung disease must take into account not only the risks of the proposed surgery but also the potential benefits.

Necessity for pulmonary function testing and arterial blood gases Is preoperative laboratory evaluation of pulmonary function (PFTs, spirometry) required for proper assessment of patients with lung disease? For more than a decade studies have cast doubt on the necessity for such testing [45]. Clinical identification of lung disease is thought to be comparable to spirometry in terms of assessing a patient’s risk for developing PPCs. Reduction in the forced vital capacity in 1 second (FEV1), or other spirometric indices of abnormal lung function and abnormalities in arterial blood gases such as hypercapnia or hypoxemia suggest that the patient is at increased risk for developing PPCs, but their presence does not improve the ability to stratify risk if the patient is judged to have lung disease on clinical grounds. However, the concept that clinical identification of underlying lung disease is equivalent to laboratory testing is flawed. Several studies suggest that clinical identification of pre-existing chronic lung disease is inadequate for the purposes of risk assessment [46,47]. Some asthmatic patients are unaware of significant changes in their lung function, and in these patients symptoms are unreliable for assessing severity and optimization of function [48]. PFTs (FEV1, peak flow) can determine a change from baseline or response to therapy. Thus, spirometry continues to have a role in the preoperative risk assessment process. Spirometry is useful when there is uncertainty about the presence of lung impairment. A proper determination of whether or not the patient has underlying lung disease can have a profound effect on subsequent management decisions. If a patient is inaccurately assessed as having COPD, their risk will be judged to be higher and the patient may be denied surgery or exposed to unnecessary treatment. Failure to diagnose underlying lung disease when it is present may suggest the patient is at lower risk and result in forgoing treatment that might prevent PPCs. An ambiguous clinical picture regarding the severity of bronchospasm, presence of COPD, response to bronchodilators, or unexplained shortness of breath can be clarified by spirometry. Spirometry can complement a clinician’s ability to detect changes in asthmatic status or resolution of a COPD flare. Spirometry is non-invasive, easily performed and interpreted, inexpensive, and provides important information about lung disease. PFTs and exercise testing provide information that is useful in determining whether a patient is a candidate P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 81 for lung resection. Spirometry does not allow calibration of a patient’s risk, but may enhance diagnosis of lung disease. It should not be used indiscriminately nor should it be avoided; rather, it should be used selectively when the information it provides will change management or improve risk stratification. Obtaining routine preoperative baseline arterial blood gases (ABGs) is usually not helpful as postoperative patients frequently have abnormalities in oxygenation and ventilation. Baseline ABGs do not improve risk assessment nor add to risk stratification. Since the need for postoperative supplemental oxygen is determined by the level of oxygenation and hemoglobin after surgery, and because supplemental oxygen is titrated to effect, baseline ABGs are not necessary in most circumstances. Pulse oximetry is non-invasive, cost-effective, and supplements the history and physical examination. Patients with hypoxemia will be detected by an abnormal arterial oxygenation saturation.

Cigarette smoking Cigarette smoking is a significant preoperative risk factor [42,49]. This effect is primarily related to the resulting chronic lung disease although smoking has other effects on lung function. Cessation of cigarette smoking for 48 hours before surgery decreases carboxyhemoglobin levels to normal, abolishes the stimulant effect of nicotine on the cardiovascular system, and improves respiratory ciliary beating. However, 1 to 2 weeks are required to decrease sputum volume and 4 to 6 weeks are required to improve symptoms and lung function [50]. More than 20 pack-years of cigarette smoking is associated with an increased incidence of PPCs. Patients who stop smoking more than 8 weeks before surgery have a reduced rate of PPCs compared with those who continue to smoke [51,52].

A pulmonary risk index Perioperative management of patients would be improved if one could predict with more assurance, not only which patients will develop PPCs but the magnitude of the risk faced by an individual patient (ie, perform risk stratification). Most of the data in the literature have suggested that although one can identify patients at increased risk for the development of PPCs, it is not possible to quantitate with more precision their risk (excluding thoracic surgery patients for whom risk predictors are more well accepted) [45]. Efforts to use laboratory testing to aid in risk stratification have not been helpful but have been, as discussed, useful in identifying underlying lung disease when the clinical diagnosis of lung disease is not clear. Recently, there have been renewed efforts to develop a pulmonary risk index predictive of the development of postoperative respiratory failure or pneumonia [53,54]. In contrast to previous studies where attempts to develop a pulmonary risk index were limited to analysis of small numbers of patients, the risk factors in these studies were based on the analysis of thousands 82 P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 of patients in the veteran affairs patient database. Points were assigned to clinical and other factors based on the significance of these factors as assessed by logistical regression. The major contributors to patients’ risk remain clinical, and PFT does not seem to contribute to assessment of a patient’s risk. The major factors that increase a patients risk of postoperative pneumonia or respiratory failure are: age (>70), type of surgery (vascular and thoracic), presence of underlying lung disease, renal failure, poor nutritional status, and the amount of blood loss during surgery (>4 units packed blood cells).

Decreasing postoperative complications A variety of techniques have been reported to decrease PPCs, including incentive spirometry, pain relief, and agents such as theophylline, which in addition to its bronchodilating properties, has a stimulant effect on the respiratory system and is an inotrope to the diaphragm. None of these maneuvers is entirely effective in preventing PPCs. Incentive spirometry can decrease rates of PPCs and hospital length of stay [55]. There is conflicting evidence regarding the usefulness of regional analgesic techniques in preventing PPCs [56,57]. Studies that show a reduction of PPCs with the use of regional anesthesia after surgery examine high-risk patients (upper abdominal surgery, COPD) and use epidural local anesthetics. A recent meta-analysis suggests that postoperative epidural analgesia may reduce clinical PPCs [58]. However, the lack of a standard definition for PPCs and the age of the studies included in the meta analysis limit the applicability of the results of that study. A more recent double-masked randomized trial failed to show a benefit of regional anesthesia or analgesia with respect to reintubation, prolonged intubation, or pneumonia [59]. Other studies have also yielded contradictory results regarding the value of epidural analgesia in reducing the incidence of PPCs [60,61]. The conflicting data suggest that mechanisms underlying PPCs involve more than pain and effects of the stress response on the respiratory system. Additional well-performed randomized clinical trials are necessary to establish the role of epidural analgesia in the treatment or prevention of PPCs. Episodic or sustained arterial desaturation occurring in the days after surgery may play a role in the development of other complications. If postoperative hypoxemia develops, tachycardia and hypertension will ensue at a time when postoperative anemia decreases arterial oxygen content. This may induce myocardial ischemia in patients with coronary artery stenoses. A recent study showed postoperative oxygen therapy decreased heart rate and increased arterial oxygen saturation several days after abdominal surgery [62]. Perioperative supplemental oxygen also has been shown to decrease both postoperative nausea and vomiting and surgical wound infections by approximately 50% [63,64]. Oxygen therapy may be beneficial even if administered for only several hours after surgery, rather than administered only in the PACU, especially in patients who have undergone major abdominal surgery. P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 83

Types of anesthesia The results of many studies do not demonstrate a clear advantage of one anesthetic technique over another in lessening PPCs [56]. Regional anesthesia has the advantage of not requiring airway manipulation or affecting ventilatory control. Neuromuscular block is not necessary and there may be less impact on ventilatory control than with a GA. Neuraxial block with local anesthetics may result in respiratory muscle weakness and impaired cough. GA has the advantage of ensuring patient cooperation, control of the airway and the ability to suction secretions through an endotracheal tube. Disadvantages of GA include the common adjunct use of neuromuscular block, alteration of ventilatory control and airway manipulation, with its potential for secretions and bronchospasm. Residual neuromuscular blockade occurs more commonly after pancuronium than after vecuronium or atracurium, and residual neuromuscular blockade from pancuronium is a risk factor for development of PPCs [65]. Use of shorter-acting neuromuscular blocking agents should be considered in patients at risk for PPCs. Further studies are required to define the role of residual neuromuscular blockade in the pathogenesis of PPCs and to determine the role of regional anesthesia in the patient with lung disease.

Patients with asthma Bronchospasm is one of the most significant respiratory events that can occur during anesthesia [66]. Complication rates in asthmatic patients are lower than rates reported 30 years ago, suggesting that current management has reduced the risk of perioperative bronchospasm, pneumonia, respiratory failure, and death. Risk factors for the development of PPCs in asthmatics include recent asthma symptoms, recent use of anti-asthma drugs or therapy in a medical facility for asthma symptoms, and history of tracheal intubation for asthma [67]. The type of anesthesia has not been demonstrated to be a risk factor for PPCs in asthmatics. In Warner’s study [67] of over 1500 patients with asthma, the complication rates for general and regional anesthesia were similar, refuting the notion that regional anesthesia was safer for patients with asthma. The risk for bronchospasm in the perioperative period is low in stable asthmatic patients and when it occurs is usually not associated with serious morbidity. Corticosteroids are effective in attenuating bronchospasm [68–71]. Asthmatic patients at risk for PPCs may be treated with steroids in the perioperative period starting 24 to 48 hours before surgery, because the beneficial effect of steroids on airway reactivity occurs over a period of hours. An appropriate daily dose of prednisone in an adult would be 40 to 60 mg. Although there is limited evidence regarding the appropriate treatment regiment, if corticosteroids are administered preoperatively in asthmatic individuals, they may be given 1 to 2 days in advance of planned surgery either orally in the does suggested earlier or intravenously (hydrocortisone in a dose of 100 mg every 8 hours) in those unable to take 84 P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 medications by mouth. Steroids can be discontinued after surgery without tapering doses in the absence of bronchospasm. Use of short courses of systemic steroids in the perioperative period is not associated with increased wound infections or poor wound healing [72]. Inhaled steroids might be a reasonable alternative to parenteral steroids for the purpose of reducing perioperative bronchospasm in patients with asthma. Patients who are wheezing before surgery should receive treatment with inhaled beta-2 adrenergic agents and corticosteroids. Theophylline is no longer a front-line drug in the management of asthma and it has the potential for significant toxicity [73]. Those who improve with treatment may be allowed to proceed with surgery. Elective surgery should be deferred if the patient does not improve. Airway hyperreactivity persists for several weeks after an episode of asthma [74]. Improvement in asthma symptoms does not preclude the development of bronchospasm in response to various stimuli. Volatile anesthetic agents are bronchodilators and the differences between them with respect to their efficacy in treating bronchospasm are probably clinically insignificant [75]. Propofol is useful in patients with bronchospasm and it is associated with reductions in wheezing during induction [76,77]. Common sense suggests that regional anesthesia should be used, where appropriate, to avoid instrumentation of the airway. Regional anesthesia does not lead to unopposed parasympathetic effects or enhanced bronchoconstriction [67]. The combination of lidocaine and a beta-2 aerosol has a synergistic effect in attenuating the airway response to bronchocon- strictors. [78] Laryngeal mask airways are associated with less airway reaction than endotracheal tubes suggesting they may be useful in patients with reactive airways disease [79,80]. Not all wheezes require treatment and not all wheezing is related to pre-existing reactive airways disease. Other causes of wheezing include pulmonary edema, pneumothorax, drug reactions, aspiration, carinal irritation, and endobronchial intubation. Assessment of the significance of wheezing requires measurement of tidal volume and associated airway pressures, arterial oxygenation, and vital signs. Oxygenation is more important than elimination of CO2 in the patient who develops severe bronchospasm with increased airway pressures. In this situation permissive hypercapnia may be useful allowing PaCO2 to rise above normal levels, using supplemental oxygen to insure adequate oxygenation and avoiding high airway pressures that can cause barotrauma [81–83]. ‘‘ICU- type’’ ventilators can also help improve gas exchange in patients with high airway pressures.

Patients with chronic obstructive pulmonary disease Patients with underlying chronic lung disease are at risk for the development of PPCs. Respiratory infections should be treated with antibiotics. Some patients with COPD have bronchospasm, in addition to their fixed disease. Inhaled beta-2 adrenergic agonists, anticholinergic agents, or a course of steroids may be useful. P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 85

Patients with COPD may have chronically fatigued respiratory muscles. Impaired nutrition, electrolyte, and endocrine disorders contribute to respiratory muscle weakness and should be corrected before surgery. Patients with COPD should be examined for unrecognized cor pulmonale; if present, it should be treated before surgery. One modality of potential value in patients with COPD is respiratory muscle training. Results of LVRS are consistent with the hypothesis that pulmonary rehabilitation is helpful in reducing PPCs in high-risk patients. Although this has not been tested with a randomized clinical trial, measurement of respiratory muscle strength may identify patients at risk for development of PPCs. Patients that demonstrate increased respiratory muscle strength after respiratory muscle training have fewer PPCs than those who do not increase their respiratory muscle strength [84]. Determination of exercise capacity may also be of benefit in identifying patients at risk for PPCs [85]. There is little published information to guide management of patients requiring chronic oxygen administration. Patients with chronic hypoxemia benefit from short-term oxygen administration, which usually results in lessening of pulmonary hypertension, reduction in signs and symptoms of heart failure, and improvement of mentation. A preoperative finding of hypoxemia should prompt further investigation. Even if hypoxemia is chronic, but the patient is not receiving oxygen at home, continuous oxygen administration should be started and elective surgery deferred to allow improvement in pulmonary hypertension and heart function.

Patients with obstructive sleep apnea Obstructive sleep apnea (OSA) is a breathing disorder characterized by repeated collapse of the upper airway during sleep with cessation of breathing [86]. The loss of upper airway muscle tone, particularly during rapid eye movement (REM) sleep, results in a narrow floppy airway becoming narrower. Airway obstruction results in arousal, sleep is interrupted, muscle tone is restored, and the airway becomes patent again. Almost all patients with OSA have a history of snoring. OSA is more common in men, obese individuals, and the elderly. OSA is associated with hypertension, arrhythmias, congestive heart failure, coronary artery disease, and stroke. A preoperative history of snoring or apneas during sleep strongly suggest OSA [87]. Preoperative snoring is a risk factor for postoperative apnea and lower postoperative mean oxygen saturation [88]. GA alone results in transient and minimal alterations in sleep architecture although anesthetic agents reduce upper airway muscle tone (which acts to oppose airway collapse) to a greater extent than diaphragmatic strength thus increasing the propensity for obstruction. In the initial nights after major surgery, sleep is fragmented with decreased REM sleep. In succeeding nights REM sleep is increased (REM rebound). Sleep disturbances are less after laparoscopic surgery suggesting that the magnitude of surgery or use of opioid analgesics may be important factors in the development of postoperative sleep disturbances [89–92]. 86 P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91

Postoperative hypoxemia develops quickly after emergence from GA [93]. Diffusion hypoxemia may contribute to early postoperative hypoxemia. How- ever, postoperative hypoxemia is usually related to V/Q inequality or hypoventilation. The elderly are prone to more severe hypoxemia. Multiple episodes of more acute and severe hypoxemia associated with airway obstruction have been observed days after surgery. There are interactions between natural ‘‘sleep,’’ age, the postoperative period, and narcotic analgesia that produce ventilatory disturbances [94]. Preoperative sleep-related oxygenation disturbances need not be present for postoperative hypoxemia to occur [95]. Sleep disordered breathing may occur after major surgery even in patients that do not have OSA. Patients with sleep apnea are likely to have worsening of their disease process after anesthesia and surgery and may be at risk for development of more apneas and more severe episodes of hypoxemia postoperatively. Sedatives, anesthetic, and analgesic agents may worsen OSA by decreasing pharyngeal tone and attenuating ventilatory and arousal responses to hypoxia, hypercarbia, and obstruction. Postoperatively patients are often supine, which can aggravate OSA. OSA patients frequently have redundant pharyngeal tissue, which makes airway management difficult. OSA patients may need expensive treatments (eg, continuous positive airway pressure [CPAP] and bilevel positive airway pressure [BiPAP]), may need special monitoring, and may have cardiovascular disease. A recent retrospective study demonstrated a higher rate of adverse postoperative complications in patients with OSA undergoing hip and knee replacements compared with case-matched controls [96]. It seems prudent and cost-effective to diagnose OSA preoperatively so these measures are applied appropriately and other risk-reduction measures can be taken. Preoperative treatment of OSA can improve heart function, reduce pulmonary artery pressures, and allow blood pressure to normalize, all of which may reduce complications. The limited evidence available regarding postoperative risks in patients with OSA suggests the type of surgery is more important than the anesthetic. Regional anesthesia may reduce or eliminate problems with airway maintenance, depression of ventilation and arousal responses but may not affect REM rebound. Out- patient surgery may be appropriate for extremity or ‘‘non-invasive’’ procedures, when rapid and complete restoration of postoperative consciousness is possible, if there is a limited need for narcotic analgesics, or if the patient’s OSA does not require CPAP. A postoperative hospital stay is appropriate for abdominal surgery, when narcotic analgesia is required, if the patient is not awake enough to manage CPAP, snoring or airway obstruction occurs postoperatively, or episodic desatu- rations are evident in the PACU. Special monitoring or an ICU admission is warranted if, in addition to symptoms mentioned earlier, preoperative assessment of OSA reveals it to be severe, the patient cannot manage CPAP by themselves, airway management was difficult in the OR, severe obesity exists, the patient is very sedated postoperatively, or pain cannot be managed without opioids. There is no evidence to support risk stratification or the necessity for a period of observation in the ICU depending on whether the patient with OSA is having airway surgery [97,98]. P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 87

Management of patients with OSA requires judicious use of sedatives. Diffi- cult airway precautions are appropriate including availability of laryngeal mask airway, fiberoptic bronchoscope, and a tracheostomy kit. Short acting agents are preferred. Non-invasive ventilatory devices that can deliver positive pressure to the airway (eg, BiPAP) should be available postoperatively. Non-steroidal analgesics may reduce the need for narcotic-based analgesia.

Summary Understanding the risk factors for the development of PPCs allows targeted interventions aimed at reducing the frequency and severity of PPCs. The broad categories of what increases the likelihood of developing a PPC are understood but specific understanding of how individual risk factors act to cause PPCs is lacking, and there is little information regarding the interaction or synergy between risk factors. Further research is needed to define the nature of risk factors and develop better predictive models of patients at risk for developing PPCs. It is clear that anesthetic agents produce significant changes in the respiratory system but further information is needed to define how such changes contribute, if at all, to the subsequent development of PPCs. The ongoing controversy regarding the value of regional analgesia or anesthetic techniques, especially epidural analgesia and anesthesia, in reducing or preventing PPCs requires well-done randomized clinical trials. Further research is also needed in the area of postoperative care such as interventions in patients with OSA or the use of inventive spirometric techniques.

References

[1] Warner DO. Preventing postoperative pulmonary complications: the role of the anesthesiologist. Anesthesiology 2000;92:1467–72. [2] Smetana GW. Preoperative pulmonary evaluation. N Engl J Med 1999;340:937–44. [3] Forbes AR, Gamsu G. Mucociliary clearance in the canine lung during and after general anesthesia. Anesthesiology 1979;50:26–9. [4] Cavaliere F, Schiavello R, Masieri S, Passali D. Mucociliary flow in the nose during general and epidural anesthesia. Acta Anaesthesiol Belg 1983;34:33–9. [5] Cervin A, Lindberg S, Mercke U. Effects of halothane on mucociliary activity in vivo. Otolar- yngol Head Neck Surg 1995;112:714–22. [6] Forbes AR, Horrigan RW. Mucociliary flow in the trachea during anesthesia with enflurane, ether, nitrous oxide, and morphine. Anesthesiology 1977;46:319–21. [7] Forbes AR, Gamsu G. Depression of lung mucociliary clearance by thiopental and halothane. Anesth Analg 1979;58:387–9. [8] Kotani N, Lin CY, Wang JS, Gurley JM, Tolin FP, Michelassi F, et al. Loss of alveolar macrophages during anesthesia and operation in humans. Anesth Analg 1995;81:1255–62. [9] Kotani N, Hashimoto H, Sessler DI, Kikuchi A, Suzuki A, Takahashi S, et al. Intraoperative modulation of alveolar macrophage function during isoflurane and propofol anesthesia. Anes- thesiology 1998;89:1125–32. [10] Bruegger D, Bauer A, Finsterer U, Bernasconi P, Kreimeier U, Christ F. Microvascular changes during anesthesia: sevoflurane compared with propofol. Acta Anaesthesiol Scand 2002;46: 481–7. 88 P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91

[11] Nielsen VG, Baird MS, McAdams ML, Freeman BA. Desflurane increases pulmonary alveolar- capillary membrane permeability after aortic occlusion-reperfusion in rabbits: evidence of oxidant- mediated lung injury [see comments]. Anesthesiology 1998;88:1524–34. [12] Wollmer P, Schairer W, Bos JA, Bakker W, Krenning EP, Lachmann B. Pulmonary clearance of 99mTc-DTPA during halothane anaesthesia. Acta Anaesthesiol Scand 1990;34:572–5. [13] Ward HE, Nicholas TE. Effect of artificial ventilation and anaesthesia on surfactant turnover in rats. Respir Physiol 1992;87:115–29. [14] Tobin WR, Kaiser HE, Groeger AM, Lema MJ. The effects of volatile anesthetic agents on pulmonary surfactant function. In Vivo 2000;14:157–63. [15] Pajewski TN, DiFazio CA, Moscicki JC, Johns RA. Nitric oxide synthase inhibitors, 7-nitro indazole and nitroG-L- arginine methyl ester, dose dependently reduce the threshold for isoflurane anesthesia. Anesthesiology 1996;85:1111–9. [16] Zuo Z, Johns RA. Inhalational anesthetics up-regulate constitutive and lipopolysaccharide-induced inducible nitric oxide synthase expression and activity. Mol Pharmacol 1997;52:606–12. [17] Chen BB, Nyhan DP, Fehr DM, Goll HM, Murray PA. Halothane anesthesia causes active flow- independent pulmonary vasoconstriction. Am J Physiol 1990;259:H74–83. [18] Chen BB, Nyhan DP, Fehr DM, Murray PA. Halothane anesthesia abolishes pulmonary vascular responses to neural antagonists. Am J Physiol 1992;262:H117–22. [19] Gambone LM, Murray PA, Flavahan NA. Isoflurane anesthesia attenuates endothelium-dependent pulmonary vasorelaxation by inhibiting the synergistic interaction between nitric oxide and prostacyclin. Anesthesiology 1997;86:936–44. [20] Gambone LM, Fujiwara Y, Murray PA. Endothelium-dependent pulmonary vasodilation is selectively attenuated during isoflurane anesthesia. Am J Physiol 1997;272:H290–8. [21] Lennon PF, Murray PA. Isoflurane and the pulmonary vascular pressure-flow relation at baseline and during sympathetic alpha- and beta-adrenoreceptor activation in chronically instrumented dogs. Anesthesiology 1995;82:723–33. [22] Murray PA. Anesthetic modulation of pulmonary vascular regulation. Adv Pharmacol 1994;31: 485–504. [23] Tokics L, Hedenstierna G, Strandberg A, Brismar B, Lundquist H. Lung collapse and gas exchange during general anesthesia: effects of spontaneous breathing, muscle paralysis, and positive end-expiratory pressure. Anesthesiology 1987;66:157–67. [24] Krayer S, Rehder K, Vettermann J, Didier EP, Ritman EL. Position and motion of the human diaphragm during anesthesia-paralysis. Anesthesiology 1989;70:891–8. [25] Lockhart SH, Rampil IJ, Yasuda N, Eger 2nd EI, Weiskopf RB. Depression of ventilation by desflurane in humans. Anesthesiology 1991;74:484–8. [26] Canet J, Sanchis J, Zegri A, Llorente C, Navajas D, Casan P. Effects of halothane and isoflurane on ventilation and occlusion pressure. Anesthesiology 1994;81:563–71. [27] Dahan A, van den Elsen MJ, Berkenbosch A, DeGoede J, Olievier IC, Burm AG, et al. In- fluence of a subanesthetic concentration of halothane on the ventilatory response to step changes into and out of sustained isocapnic hypoxia in healthy volunteers. Anesthesiology 1994;81: 850–9. [28] Ide T, Sakurai Y, Aono M, Nishino T. Contribution of peripheral chemoreception to the depression of the hypoxic ventilatory response during halothane anesthesia in cats. Anesthesiology 1999;90:1084–91. [29] Bailey PL, Pace NL, Ashburn MA, Moll JW, East KA, Stanley TH. Frequent hypoxemia and apnea after sedation with midazolam and fentanyl. Anesthesiology 1990;73:826–30. [30] Nieuwenhuijs D, Sarton E, Teppema L, Dahan A. Propofol for monitored anesthesia care: implications on hypoxic control of cardiorespiratory responses. Anesthesiology 2000;92:46–54. [31] Robotham JL. Do low-dose inhalational anesthetic agents alter ventilatory control? Anesthesi- ology 1994;80:723–6. [32] Jonsson M, Kim C, Yamamoto Y, Runold M, Lindahl SG, Eriksson LI. Atracurium and vecuronium block nicotine-induced carotid body chemoreceptor responses. Acta Anaesthesiol Scand 2002;46:488–94. P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 89

[33] Beecher HK. The measured effect of laparotomy on the respiration. J Clin Invest 1933;88: 639–50. [34] Simonneau G, Vivien A, Sartene R, Kunstlinger F, Samii K, Noviant Y, et al. Diaphragm dysfunction induced by upper abdominal surgery. Role of postoperative pain. Am Rev Respir Dis 1983;128:899–903. [35] Dureuil B, Viires N, Cantineau JP, Aubier M, Desmonts JM. Diaphragmatic contractility after upper abdominal surgery. J Appl Physiol 1986;61:1775–80. [36] Hedenstierna G, Tokics L, Lundquist H, Andersson T, Strandberg A, Brismar B. Phrenic nerve stimulation during halothane anesthesia. Effects of atelectasis. Anesthesiology 1994;80:751–60. [37] Rock P, Freed AN, Nyhan DP, Murray PA. Thoracotomy increases peripheral airway tone and reactivity. Am J Respir Crit Care Med 1995;151:1047–52. [38] Putensen-Himmer G, Putensen C, Lammer H, Lingnau W, Aigner F, Benzer H. Comparison of postoperative respiratory function after laparoscopy or open laparotomy for cholecystectomy. Anesthesiology 1992;77:675–80. [39] Erice F, Fox GS, Salib YM, Romano E, Meakins JL, Magder SA. Diaphragmatic function before and after laparoscopic cholecystectomy. Anesthesiology 1993;79:966–75; discussion 27A–28. [40] Pedersen T, Viby-Mogensen J, Ringsted C. Anaesthetic practice and postoperative pulmonary complications. Acta Anaesthesiol Scand 1992;36:812–8. [41] Warner DO, Warner MA, Offord KP, Schroeder DR, Maxson P, Scanlon PD. Airway obstruction and perioperative complications in smokers undergoing abdominal surgery. Anesthesiology 1999;90:372–9. [42] Jayr C, Matthay MA, Goldstone J, Gold WM, Wiener-Kronish JP. Preoperative and intraoperative factors associated with prolonged mechanical ventilation. A study in patients following major abdominal vascular surgery. Chest 1993;103:1231–6. [43] Nunn JF, Milledge JS, Chen D, Dore C. Respiratory criteria of fitness for surgery and anaesthesia. Anaesthesia 1988;43:543–51. [44] Yusen RD, Trulock EP, Pohl MS, Biggar DG. Results of lung volume reduction surgery in patients with emphysema. Semin Thorac Cardiovasc Surg 1996;8:99–109. [45] American College of Physicians. Preoperative pulmonary function testing. Ann Intern Med 1990; 112:793–4. [46] Badgett RG, Tanaka DJ, Hunt DK, Jelley MJ, Feinberg LE, Steiner JF, et al. Can moderate chronic obstructive pulmonary disease be diagnosed by historical and physical findings alone? Am J Med 1993;94:188–96. [47] Emerman CL, Lukens TW, Effron D. Physician estimation of FEV1 in acute exacerbation of COPD. Chest 1994;105:1709–12. [48] Banzett RB, Dempsey JA, O’Donnell DE, Wamboldt MZ. Symptom perception and respiratory sensation in asthma. Am J Respir Crit Care Med 2000;162:1178–82. [49] Bluman LG, Mosca L, Newman N, Simon DG. Preoperative smoking habits and postoperative pulmonary complications. Chest 1998;113:883–9. [50] Pearce AC, Jones RM. Smoking and anesthesia: preoperative abstinence and perioperative morbidity. Anesthesiology 1984;61:576–84. [51] Warner MA, Offord KP, Warner ME, Lennon RL, Conover MA, Jansson-Schumacher U. Role of preoperative cessation of smoking and other factors in postoperative pulmonary complications: a blinded prospective study of coronary artery bypass patients. Mayo Clin Proc 1989;64:609–16. [52] Warner MA, Divertie MB, Tinker JH. Preoperative cessation of smoking and pulmonary complications in coronary artery bypass patients. Anesthesiology 1984;60:380–3. [53] Arozullah AM, Daley J, Henderson WG, Khuri SF. Multifactorial risk index for predicting postoperative respiratory failure in men after major noncardiac surgery. The National Veterans Administration Surgical Quality Improvement Program. Ann Surg 2000;232:242–53. [54] Arozullah AM, Khuri SF, Henderson WG, Daley J. Development and validation of a multifactorial risk index for predicting postoperative pneumonia after major noncardiac surgery. Ann Intern Med 2001;135:847–57. [55] Celli BR, Rodriguez KS, Snider GL. A controlled trial of intermittent positive pressure breath- 90 P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91

ing, incentive spirometry, and deep breathing exercises in preventing pulmonary complications after abdominal surgery. Am Rev Respir Dis 1984;130:12–5. [56] Liu S, Carpenter RL, Neal JM. Epidural anesthesia and analgesia. Their role in postoperative outcome. Anesthesiology 1995;82:1474–506. [57] Jayr C, Thomas H, Rey A, Farhat F, Lasser P, Bourgain JL. Postoperative pulmonary complications. Epidural analgesia using bupivacaine and opioids versus parenteral opioids. Anesthesi- ology 1993;78:666–76; discussion 22A. [58] Ballantyne JC, Carr DB, deFerranti S, Suarez T, Lau J, Chalmers TC, et al. The comparative effects of postoperative analgesic therapies on pulmonary outcome: cumulative meta-analyses of randomized, controlled trials. Anesth Analg 1998;86:598–612. [59] Norris EJ, Beattie C, Perler BA, Martinez EA, Meinert CL, Anderson GF, et al. Double-masked randomized trial comparing alternate combinations of intraoperative anesthesia and postoperative analgesia in abdominal aortic surgery. Anesthesiology 2001;95:1054–67. [60] Rigg JR, Jamrozik K, Myles PS, Silbert BS, Peyton PJ, Parsons RW, et al. Epidural anaesthesia and analgesia and outcome of major surgery: a randomised trial. Lancet 2002;359:1276–82. [61] Rodgers A, Walker N, Schug S, McKee A, Kehlet H, van Zundert A, et al. Reduction of postoperative mortality and morbidity with epidural or spinal anaesthesia: results from overview of randomised trials. BMJ 2000;321:1493. [62] Rosenberg-Adamsen S, Lie C, Bernhard A, Kehlet H, Rosenberg J. Effect of oxygen treatment on heart rate after abdominal surgery. Anesthesiology 1999;90:380–4. [63] Greif R, Laciny S, Rapf B, Hickle RS, Sessler DI. Supplemental oxygen reduces the incidence of postoperative nausea and vomiting. Anesthesiology 1999;91:1246–52. [64] Greif R, Akca O, Horn EP, Kurz A, Sessler DI. Supplemental perioperative oxygen to reduce the incidence of surgical-wound infection. Outcomes Research Group. N Engl J Med 2000;342: 161–7. [65] Berg H, Roed J, Viby-Mogensen J, Mortensen CR, Engbaek J, Skovgaard LT, et al. Residual neuromuscular block is a risk factor for postoperative pulmonary complications. A prospective, randomised, and blinded study of postoperative pulmonary complications after atracurium, vecuronium and pancuronium. Acta Anaesthesiol Scand 1997;41:1095–103. [66] Bishop MJ, Cheney FW. Anesthesia for patients with asthma. Low risk but not no risk. Anes- thesiology 1996;85:455–6. [67] Warner DO, Warner MA, Barnes RD, Offord KP, Schroeder DR, Gray DT, et al. Perioperative respiratory complications in patients with asthma. Anesthesiology 1996;85:460–7. [68] Darowski MJ, Hannon VM, Hirshman CA. Corticosteroids decrease airway hyperresponsiveness in the Basenji-Greyhound dog model of asthma. J Appl Physiol 1989;66:1120–6. [69] Emala CW, Clancy J, Hirshman CA. Glucocorticoid treatment decreases muscarinic receptor expression in canine airway smooth muscle. Am J Physiol 1997;272:L745–51. [70] Parker SD, Brown RH, Hirshman CA. Differential effect of glucocorticoids on pulmonary responses and eosinophils. Respir Physiol 1991;83:323–31. [71] Tobias JD, Sauder RA, Hirshman CA. Methylprednisolone prevents propranolol-induced airway hyperreactivity in the Basenji-greyhound dog. Anesthesiology 1991;74:1115–20. [72] Kabalin CS, Yarnold PR, Grammer LC. Low complication rate of corticosteroid-treated asthmatics undergoing surgical procedures. Arch Intern Med 1995;155:1379–84. [73] Hart SP. Should aminophylline be abandoned in the treatment of acute asthma in adults? QJM 2000;93:761–5. [74] Whyte MK, Choudry NB, Ind PW. Bronchial hyperresponsiveness in patients recovering from acute severe asthma. Respir Med 1993;87:29–35. [75] Brown RH, Zerhouni EA, Hirshman CA. Comparison of low concentrations of halothane and isoflurane as bronchodilators. Anesthesiology 1993;78:1097–101. [76] Eames WO, Rooke GA, Wu RS, Bishop MJ. Comparison of the effects of etomidate, propofol, and thiopental on respiratory resistance after tracheal intubation. Anesthesiology 1996;84: 1307–11. [77] Pizov R, Brown RH, Weiss YS, Baranov D, Hennes H, Baker S, et al. Wheezing during P. Rock, A. Passannante / Anesthesiology Clin N Am 22 (2004) 77–91 91

induction of general anesthesia in patients with and without asthma. A randomized, blinded trial. Anesthesiology 1995;82:1111–6. [78] Groeben H, Silvanus MT, Beste M, Peters J. Combined intravenous lidocaine and inhaled salbutamol protect against bronchial hyperreactivity more effectively than lidocaine or salbutamol alone. Anesthesiology 1998;89:862–8. [79] Kim ES, Bishop MJ. Endotracheal intubation, but not laryngeal mask airway insertion, produces reversible bronchoconstriction. Anesthesiology 1999;90:391–4. [80] Tait A, Pandit U, Voepel-Lewis T, Munro H, Malviya S. Use of the laryngeal mask airway in children with upper respiratory tract infections: a comparison with endotracheal intubation. Anesth Analg 1998;86(4):706–11. [81] Mutlu GM, Factor P, Schwartz DE, Sznajder JI. Severe status asthmaticus: management with permissive hypercapnia and inhalation anesthesia. Crit Care Med 2002;30:477–80. [82] Hickling KG. Permissive hypercapnia. Respir Care Clin N Am 2002;8:155–69 [v]. [83] Papiris S, Kotanidou A, Malagari K, Roussos C. Clinical review: severe asthma. Crit Care 2002; 6:30–44. [84] Nomori H, Kobayashi R, Fuyuno G, Morinaga S, Yashima H. Preoperative respiratory muscle training. Assessment in thoracic surgery patients with special reference to postoperative pulmonary complications. Chest 1994;105:1782–8. [85] Gerson MC, Hurst JM, Hertzberg VS, Baughman R, Rouan GW, Ellis K. Prediction of cardiac and pulmonary complications related to elective abdominal and noncardiac thoracic surgery in geriatric patients. Am J Med 1990;88:101–7. [86] Loadsman JA, Hillman DR. Anaesthesia and sleep apnoea. Br J Anaesth 2001;86:254–66. [87] Kapuniai LE, Andrew DJ, Crowell DH, Pearce JW. Identifying sleep apnea from self-reports. Sleep 1988;11:430–6. [88] Gentil B, Lienhart A, Fleury B. Enhancement of postoperative desaturation in heavy snorers. Anesth Analg 1995;81:389–92. [89] Knill RL, Moote CA, Skinner MI, Rose EA. Anesthesia with abdominal surgery leads to intense REM sleep during the first postoperative week. Anesthesiology 1990;73:52–61. [90] Moote CA, Knill RL. Isoflurane anesthesia causes a transient alteration in nocturnal sleep. Anesthesiology 1988;69:327–31. [91] Rosenberg-Adamsen S, Kehlet H, Dodds C, Rosenberg J. Postoperative sleep disturbances: mechanisms and clinical implications. Br J Anaesth 1996;76:552–9. [92] Rosenberg-Adamsen S, Skarbye M, Wildschiodtz G, Kehlet H, Rosenberg J. Sleep after laparoscopic cholecystectomy. Br J Anaesth 1996;77:572–5. [93] Xue FS, Li BW, Zhang GS, Liao X, Zhang YM, Liu JH, et al. The influence of surgical sites on early postoperative hypoxemia in adults undergoing elective surgery. Anesth Analg 1999;88: 213–9. [94] Catley DM, Thornton C, Jordan C, Lehane JR, Royston D, Jones JG. Pronounced, episodic oxygen desaturation in the postoperative period: its association with ventilatory pattern and analgesic regimen. Anesthesiology 1985;63:20–8. [95] Reeder MK, Goldman MD, Loh L, Muir AD, Casey KR, Lehane JR. Late postoperative nocturnal dips in oxygen saturation in patients undergoing major abdominal vascular surgery. Predictive value of pre-operative overnight pulse oximetry. Anaesthesia 1992;47:110–5. [96] Gupta RM, Parvizi J, Hanssen AD, Gay PC. Postoperative complications in patients with obstructive sleep apnea syndrome undergoing hip or knee replacement: a case-control study. Mayo Clin Proc 2001;76:897–905. [97] Ulnick KM, Debo RF. Postoperative treatment of the patient with obstructive sleep apnea. Otolaryngol Head Neck Surg 2000;122:233–6. [98] Terris DJ, Fincher EF, Hanasono MM, Fee Jr WE, Adachi K. Conservation of resources: indications for intensive care monitoring after upper airway surgery on patients with obstructive sleep apnea. Laryngoscope 1998;108:784–8. Preoperative pulmonary evaluation: facts and myths Andreas Zollingera, Christoph K. Hofera and Thomas Paschb

This review summarizes recent reports on preoperative Introduction pulmonary evaluation focusing on the impact on outcome in Pulmonary complications are among the major causes of thoracic and non-thoracic surgery. Data suggest that hitherto perioperative morbidity and mortality in thoracic as well widely accepted pulmonary function tests do not predict as in non-thoracic surgical patients. For several decades, perioperative complications. Therefore, they may not be preoperative pulmonary evaluation has been considered considered alone to decide on the patient's operability. So- to be most important with respect to prevention of called prohibitive lung function parameters should no longer pulmonary complications. However, the de®nition of a be used to deny a potentially curative lung resection. A more pulmonary complication is different not only in clinical clinically oriented, interdisciplinary approach to severely use but also between innumerable studies dealing with compromised patients may be best suited to discuss and this topic for nearly 50 years [1]. Most earlier studies solve their problems. Curr Opin Anaesthesiol 14:59±63. # 2001 have analysed the correlation between pulmonary risk Lippincott Williams & Wilkins. factors and complications in thoracic surgical patients. So-called prohibitive lung function parameters were derived from these reports (e.g. cut-off values for predicted postoperative forced expiratory volume in 1s aInstitute of Anaesthesiology, Triemli City Hospital ZuÈ rich, and bInstitute of (ppoFEV1) at 0.8 L [2], forced vital capacity of less than Anaesthesiology, University Hospital, ZuÈ rich, Switzerland 70% or FEV1 of less than 70% of the predicted value, Correspondence to Dr Andreas Zollinger M.D., Institute of Anaesthesiology, Triemli FEV1/forced vital capacity ratio of less than 65% [3], City Hospital ZuÈ rich, Birmensdorferstrasse 497, CH-8063 ZuÈ rich, Switzerland etc.). They were used to determine the feasibility of Tel: +41 1 466 22 09; fax: +41 1 466 27 43; e-mail: [email protected] lung resection in patients with lung cancer, tuberculosis, or other severe pulmonary infection. However, the Current Opinion in Anaesthesiology 2001, 14:59±63 results and conclusions of these studies were frequently Abbreviations also applied to non-thoracic procedures or to cardiovas- FEV1 forced expiratory volume in 1s cular operations. LVRS lung volume reduction surgery Furthermore, in recent years, various new anaesthetic # 2001 Lippincott Williams & Wilkins 0952-7907 techniques, including monitoring devices, were developed and alternative surgical methods as well as modern intensive care concepts are currently being applied. One might assume that they all contribute to a better quality of patient management. Therefore, it is time to separate myths from facts in this ®eld. It is the aim of this paper to review recent reports on preoperative pulmonary assessment focusing on the impact of the evaluation results on outcome in thoracic and non-thoracic surgical patients.

Pulmonary complication and pulmonary risk assessment: definitions The de®nition of a pulmonary complication is different between studies, and may also differ in clinical use. As for other complications, perioperative pulmonary events become clinically relevant if they affect patient outcome; that is, if they contribute to morbidity or mortality or prolong intensive care and/or hospital stay. Hence, postoperative events like atelectasis, pneumonia, bronchospasm, acute exacerbation of underlying chronic lung disease and respiratory failure with the need to prolonged mechanical ventilation are summarized in general as postoperative pulmonary complications [4].

59 60 Thoracic anaesthesia

When assessing the operative risk for the individual of pulmonary complications, however, was the site of patient, one should also consider the risk of denying or operation. The complication rate of upper abdominal postponing a potentially curative operation, which still and thoracic incisions was reported to be 10±40% of all remains the only therapeutic option in many cases. cases, compared with 0.3±0.4% in laparoscopic cholecys- However, unfortunately risk factors may be known in tectomy [4]. Furthermore, the author analysed the general but the speci®c risk for the individual patient predictive value of pulmonary-function testing. He cannot be predicted reliably. Too many variables, which stated that current studies `suggest no threshold beyond are potentially interrelated, are involved (i.e. the under- which the risk of surgery is prohibitive' and concluded lying disease; comorbid conditions; preexisting medica- that `the results of preoperative pulmonary-function tion; drug, alcohol and nicotine addiction; age; sex; type testing should not be used to deny a patient surgery' and duration of surgery and anaesthesia etc.). A [4] (p. 941). This paper provoked a considerable number statement made by Mittman 40 years ago remains valid of letters and comments [10±13] but the primary until today: `The ideal assessment of operative risk messages were not challenged. would identify every patient who could safely tolerate surgery. This ideal is probably unattainable...' [5] A similar review based on data published between 1995 (p. 203). The role of the anaesthetist in preoperative and 1998 resulted in similar conclusions [14]. None of evaluation, risk assessment and perioperative care is the studies included in the analysis could identify any currently being updated [6]. Traditionally, surgeons and lung function parameter obtained by spirometry, or a internists, sometimes cardiologists and pneumologists, combination of such parameters, as an independent risk decided on operability of patients. It is still the surgeon's factor for postoperative pulmonary complications. On the responsibility today to decide on the indication for a contrary, clinical scoring systems to assess overall surgical therapy. However, the role of the anaesthetist as comorbidity, such as the American Society of Anesthe- a perioperative physician [6,7] includes his coordinator siologists classi®cation, the Goldman cardiac risk index, function for preoperative assessment and preoperative the Charlson comorbidity index or the Shapiro Score, therapy if indicated. In a recent review, Slinger and were the best predictors of complications. This may be Johnston [8 .] stated that `anesthesiologists are not partly explained by the high prevalence of comorbidity gatekeepers' (p. 202), and therefore would rarely decide in patients with severe respiratory impairment. Post- who is or is not an operative candidate. However, operative respiratory compromise may often be accom- preoperative assessment is an interdisciplinary approach panied or followed by cardiac dysfunction and vice versa. to the speci®c problems of a severely compromised or In fact, Lawrence et al. [15] reported both cardiac and comorbid patient. The anaesthetist should be ready to respiratory complications in 33% of all cases. Hence, the take charge of his responsibility for the patient not only isolated identi®cation of pulmonary or cardiovascular risk during anaesthesia but in the perioperative period as a factors alone may be impossible and in many cases even whole [6]. inappropriate. Combined prediction of both types of complications rather than a single complication appears Non-thoracic surgery: are pulmonary risk to better meet the clinical requirements of preoperative factors really known? risk assessment in these patients. Patients with severe pulmonary impairment frequently present for general, non-thoracic surgery. Pulmonary The perioperative complications of a speci®c high-risk complications were found to be at least as common as group of smokers with preoperatively determined cardiac complications after abdominal surgery [9]. Pre- signi®cant airway obstruction (FEV1 540%) undergoing operative identi®cation of individual risk factors that abdominal surgery were studied by Warner et al. [16]. predict perioperative pulmonary complications would Although bronchospasm was more likely to develop in appear to be of great value. In March 1999, Smetana [4] these patients compared with a group without airway reviewed recent data on preoperative pulmonary evalua- obstruction, outcome regarding the need for prolonged tion and presented an overview on risk-reduction endotracheal intubation, prolonged intensive care stay or strategies. The author summarized the most current the need to intensive care readmission was not different and signi®cant patient and procedure-related risk factors between groups. The overall complication rate was less for postoperative pulmonary complications. Smoking than 5% [16]. An accompanying editorial to this article, even without chronic lung diseases, American Society however, strikes a note of caution regarding the of Anesthesiologists physical status grade higher than II, importance of spirometry [17]. Altered actions of the age over 70 years, and chronic obstructive pulmonary care team based on knowledge of such tests could be the disease were the only patient-related factors associated reason for a relatively favourable outcome in this group with a high relative risk. Obesity itself was associated of high-risk patients [17]. Hence, spirometry, if not with only a small increase in risk for postoperative predictive for pulmonary complications, would at least pulmonary complications. The most important predictor contribute to their prevention. Preoperative pulmonary evaluation: facts and myths Zollinger et al. 61

In summary, the detailed assessment of the patient's Anesthesiologists class (III), 36% in the higher age group history, including smoking, sputum production and the (470 years), and 27% in patients with a low FEV1/ presence of chronic bronchitic symptoms, and simple forced vital capacity ratio (555%). One-year survival in physical and radiological lung examination appear to be the American Society of Anesthesiologists class III-group good predictors for complications. Moreover, this type of was very poor (30%). Higher age, pathological stage of clinical risk assessment is cost effective and may reduce the tumour and also low predicted postoperative FEV1 the patient's stress associated with most of the more were independent factors to predict adverse outcome in technical examinations. Furthermore, while performing a another study in 193 patients following resection of non- careful physical examination, the anaesthetist is directly small cell lung carcinoma [24 .]. However, this study was involved in the preoperative evaluation and risk assess- also retrospective and data go back to the years 1994 to ment of the patient undergoing surgery, thereby 1997. signi®cantly improving his/her own impact on perioperative patient management. In recent years, lung volume reduction surgery (LVRS) has become a widely performed therapeutic option for Prohibitive lung function for pulmonary patients with end-stage pulmonary emphysema. The resection: are there any facts? results published ®ve years after the ®rst reported series Lung function parameters are traditionally used to are encouraging [25]. The surgical principles as well as identify patients who might not tolerate a planned lung the anaesthesiological management have been described resection. Therefore, the de®nition of a clear-cut so- and reviewed previously [26±28]. Brie¯y, bilateral called prohibitive lung function appears to be manda- wedge-resections of hyperin¯ated functionless areas of tory. In the light of current data, however, this issue the lung are performed in patients with preoperative needs to be reconsidered. FEV1 of less than 0.35% (50.8 L), marked hyperin¯a- tion and severe disability despite optimal medical The traditional above-cited cut-off values have been treatment. This procedure has been shown to improve questioned by many authors in recent years. Cerfolio et dyspnoea by, among others, restoration of the elastic al. [18] stated that `The previous concept of a minimal recoil pressure and diaphragm position and function [29± ppoFEV1 of 0.8 L may no longer be applicable with new 31]. anesthetic and critical care techniques. We were unable to identify any speci®c preoperative pulmonary function LVRS requires median sternotomy or thoracoscopy as a test as a predictor of postoperative morbidity' (p. 350). surgical approach with one-lung ventilation and immediate postoperative extubation following pulmonary resec- Preoperative pulmonary function testing, when used to tion. This appears to contradict most previously predict postoperative lung function, is subject to a great established opinions and standards of thoracic surgery deal of uncertainty. First, the quality of spirometry itself and anaesthesia. Interestingly enough, the postoperative may be questioned. The results of a survey in primary signi®cant improvement in dyspnoea, exercise perfor- care practice revealed alarming nonacceptability and mance and over-all quality of life following LVRS could nonreproducibility of tests [19 .]. Furthermore, Bolliger only partly be explained by improved conventional lung and Perruchoud [20] described signi®cant differences function parameters [28,32,33]. A number of alternative between preoperative predicted and postoperative mea- parameters better re¯ect the changes associated with sured spirometric values. In their investigations and in a surgery. For example, the ratio of residual volume to study by Larsen et al. [21] conventional preoperative total lung capacity appears to be a major determinant of lung function testing overestimated the functional loss air¯ow limitation, and was the single most important after lung resection. Furthermore, the time interval of parameter of improvement in lung function following the analysis following lung resection (three or six months LVRS in a study by Fessler and Permutt [34]. after operation) appears to be crucial. Signi®cant Furthermore, preoperative measurement of inspiratory improvement was observed in all tests (spirometry, lung resistance may be useful to select patients suitable diffusion capacity and exercise testing) six months after for LVRS [32]. Radiologic emphysema morphology and lobectomy or pneumonectomy [20]. In a retrospective lung perfusion scans were the most important predictors case study, Wang et al. [22] found a low predicted of a favourable clinical outcome in some studies [35,36]. postoperative diffusing capacity to be associated with a A morphologic grading system was shown to allow the high operative mortality (7.8%) but not with an adverse quanti®cation and hence comparison of results between effect on long-term survival among operative survivors. different patient groups and different therapeutic con- The interpretation of this ®nding remains speculative. In cepts [37]. Finally, breathing characteristics, predomi- another retrospective analysis of patients with pneumo- nantly the contribution of the diaphragm to tidal volume, nectomy, the hospital mortality was 15% [23]. However, may be suitable for outcome prediction [30]. An increase it was 40% in patients with higher American Society of in diaphragmatic neuromechanical coupling after surgery 62 Thoracic anaesthesia was directly related with an increase in exercise capacity References and recommended reading and a relieve of dyspnoea [31]. Papers of particular interest, published within the annual period of review, have been highlighted as: . of special interest Recent reports suggest that the limits for operability in .. of outstanding interest patients with severe pulmonary emphysema have not yet been reached. Eugene et al. [38] performed LVRS 1 Gaensler E, Cugell D, Lindgren I, et al. The role of pulmonary insufficiency in mortality and invalidism following surgery for pulmonary tuberculosis. J successfully in 44 patients with a mean preoperative Thorac Surg 1954; 24:163±187. FEV1 0.41 L (range 0.23±0.50 L). Subjective improve- 2 Olsen G, Block A, Swenson E, et al. Pulmonary function evaluation of the ment of dyspnoea was noted by 89% of patients, and lung resection candidate: a prospective study. Am Rev Respir Dis 1975; 111:379±387. mean FEV1 and mean forced vital capacity increased 3 Gass G, Olsen G. Preoperative pulmonary function testing to predict by 51% and 56% respectively. There was only one postoperative morbidity and mortality. Chest 1986; 89:127±135. death within 30 days after surgery in this study. 4 Smetana G. Preoperative pulmonary evaluation. N Engl J Med 1999; Furthermore, LVRS was performed in long-term 340:937±944. ventilator-dependent chronic obstructive pulmonary 5 Mittman C. Assessment of operative risk in thoracic surgery. Am Rev Respir disease patients, who were then successfully weaned Dis 1961; 84:197±207. from the ventilator and discharged home [39,40]. It was 6 Van Aken H, Thomson D, Smith G, Zorab J. 150 years of anaesthesia: a long way to perioperative medicine: the modern role of the anaesthesiologist. Eur J also used in lung transplant candidates with end-stage Anaesthesiol 1998; 15:520±523. pulmonary emphysema as an alternative to transplanta- 7 Hoellerich V. Acute care medicine. A community hospital practice. tion [41] and in patients with severe pulmonary Anesthesiol Clin North America 2000; 18:539±550. emphysema and lung cancer: `Surgical treatment for 8 Slinger P, Johnston M. Preoperative assessment for pulmonary resection. J lung cancer is now possible for patients who, prior to . Cardiothorac Vasc Anesth 2000; 14:202±211. This paper updates preanaesthetic assessment, perioperative complications and LVRS, had inoperable conditions due to poor pulmon- some of the most relevant anaesthetic techniques in patients with lung cancer. ary function' [42] (p. 887). 9 Lawrence V, Hilsenbeck S, Mulrow C, et al. Incidence and hospital stay for cardiac and pulmonary complications after abdominal surgery. J Gen Intern In summary, the experience with LVRS indicates that Med 1995; 10:671±678. conventional, so-called prohibitive lung function para- 10 Marx J, Punjabi N, Schwartz A. Preoperative pulmonary evaluation [letter]. N Engl J Med 1999; 341:613. meters should no longer be used alone to deny a 11 Varon J, Marik P. Preoperative pulmonary evaluation [letter]. N Engl J Med potentially curative treatment or a surgical therapy 1999; 341:613. which would signi®cantly increase the quality of life. 12 Shulman M. Preoperative pulmonary evaluation [letter]. N Engl J Med 1999; Low FEV1 is more likely to be a symptom of present 341:613±614. pulmonary impairment than an outcome predictor. A 13 Smetana G. Preoperative pulmonary evaluation [letter]. N Engl J Med 1999; number of potential alternative parameters to predict 341:614. perioperative morbidity and mortality have been 14 Zollinger A, Pasch T. The pulmonary risk patient. In: Clinical anaesthesiology. International practice and research: pre-anaesthetic care. List W, Metzler H identi®ed. However, they need to be veri®ed on the (editors). London Philadelphia Sydney Tokyo Toronto: BaillieÁ re Tindall; 1998. basis of large multicentre trials, which still have to be pp. 391±403. performed. 15 Lawrence V, Dhanda R, Hilsenbeck S, Page C. Risk of pulmonary complications after elective abdominal surgery. Chest 1996; 110:744±750. 16 Warner D, Warner M, Offord K, et al. Airway obstruction and perioperative Conclusion complications in smokers undergoing abdominal surgery. Anesthesiology Current data suggest that hitherto widely accepted 1999; 90:372±379. pulmonary function tests do not predict perioperative 17 Kimball W. The role of spirometry in predicting pulmonary complications after complications. As they rely on myths rather than facts, abdominal surgery. Anesthesiology 1999; 90:356±357. they alone should not be taken into account to decide on 18 Cerfolio R, Allen M, Trastek V, et al. Lung resection in patients with compromised pulmonary function. Ann Thorac Surg 1996; 62:348±351. the patient's operability. So-called prohibitive lung function parameters should no longer be used to deny 19 Eaton T, Withy S, Garrett J, et al. Spirometry in primary care practice. The . importance of quality assurance and the impact of spirometry workshops. Chest a potentially curative lung resection. Patients at risk in 1999; 116:416±423. The quality of spirometry performed in primary care practice in New Zealand and non-thoracic surgery are best identi®ed by a detailed the influence of a spirometry workshop on quality were assessed in this history, careful clinical examination and overall comor- prospective, controlled, randomized study. bidity scoring systems. An interdisciplinary approach to 20 Bolliger C, Perruchoud A. Functional evaluation of the lung resection the speci®c problems of a severely compromised or candidate. Eur Respir J 1998; 11:198±212. comorbid patient appears to be mandatory, since the 21 Larsen K, Svendsen U, Milman N, et al. Cardiopulmonary function at rest and during exercise after resection for bronchial carcinoma. Ann Thorac Surg methods and concepts of modern anaesthesia, surgery 1997; 64:960±964. and intensive care medicine are highly interrelated. Any 22 Wang J, Olak J, Ferguson M. Diffusing capacity predicts operative mortality further development in one speciality will in¯uence the but not long-term survival after resection for lung cancer. J Thorac other. Thus, a continuous reevaluation of indications, Cardiovasc Surg 1999; 117:581±586. contraindications, limitations, and conventions is re- 23 Groenendijk R, Croiset van Uchelen F, Mol S, et al. Factors related to outcome after pneumonectomy: retrospective study of 62 patients. Eur J Surg quired. 1999; 165:193±197. Preoperative pulmonary evaluation: facts and myths Zollinger et al. 63

24 Varela G, Novoa N, Jimenez M. Influence of age and predicted forced expiratory 33 Moy M, Ingenito E, Mentzer S, et al. Health-related quality of life improves . volume in 1 s on prognosis following complete resection for non-small cell lung following pulmonary rehabilitation and lung volume reduction surgery. Chest carcinoma. Eur J Cardiothorac Surg 2000; 18:2±6. 1999; 115:383±389. This retrospective study analysed the course of 193 patients after lung resection 34 Fessler H, Permutt S. Lung volume reduction surgery and airflow limitations. regarding long-term survival. The predictors for inverse outcome were identified. Am J Respir Crit Care Med 1998; 157:715±722. 25 Cooper J, Lefrak S. Lung-reduction surgery: 5 years on. Lancet 1999; 353 35 McKenna R, Brenner M, Fischel R, et al. Patient selection for lung volume (Suppl I):26±27. reduction surgery. J Thorac Cardiovasc Surg 1997; 114:957±964. 26 Zollinger A, Pasch T. Anaesthesia for lung volume reduction surgery. Curr Opin Anaesthesiol 1998; 11:45±49. 36 Weder W, Thurnheer R, Stammberger U, et al. Radiologic emphysema morphology is associated with outcome after surgical lung volume reduction. 27 Cooper J, Trulock E, Triantafillou A, et al. Bilateral pneumectomy (volume Ann Thorac Surg 1997; 64:313±319. reduction) for chronic obstructive pulmonary disease. J Thorac Cardiovasc Surg 1995; 109:106±119. 37 Wisser W, Klepetko W, Kontrus M, et al. Morphologic grading of the emphysematous lung and its relation to improvement after lung reduction 28 Bingisser R, Zollinger A, Hauser M, et al. Bilateral volume reduction surgery surgery. Ann Thorac Surg 1998; 65:793±799. for diffuse pulmonary emphysema by video-assisted thoracoscopy. J Thorac Cardiovasc Surg 1996; 112:875±882. 38 Eugene J, Dajee A, Kayaleh R, et al. Reduction pneumoplasty for patients with a forced expiratory volume in 1 second of 500 milliliters or less. Ann 29 Sciurba F, Goger R, Keenan R, et al. Improvement in pulmonary function and Thorac Surg 1997; 96:894±900. elastic recoil after lung-reduction surgery for diffuse emphysema. N Engl J Med 1996; 334:1095±1099. 39 Criner G, O'Brien G, Furukawa S, et al. Lung volume reduction surgery in ventilator-dependent COPD patients. Chest 1996; 110:877±884. 30 Bloch K, Li Y, Zhang J, et al. Effect of surgical lung volume reduction on breathing patterns in severe pulmonary emphysema. Am J Respir Crit Care 40 Schmid R, Vogt P, Stocker R, et al. Lung volume reduction surgery for a Med 1997; 156:553±560. patient receiving mechanical ventilation after a complex cardiac operation. J Thorac Cardiovasc Surg 1998; 115:236±237. 31 Laghi F, Jubran A, Topeli A, et al. Effect of lung volume reduction surgery on neuromechanical coupling of the diaphragm. Am J Respir Crit Care Med 41 Zenati M, Keenan R, Sciurba F, et al. Role of lung reduction in lung transplant 1998; 157:474±483. candidates with pulmonary emphysema. Ann Thorac Surg 1996; 62:994± 999. 32 Ingenito E, Evans R, Loring S, et al. Relation between preoperative inspiratory lung resistance and the outcome of lung-volume-reduction surgery 42 McKenna R, Fischel R, Brenner M, et al. Combined operations for lung for emphysema. N Engl J Med 1998; 338:1181±1185. volume reduction surgery and lung cancer. Chest 1996; 110:885±888.

CURRENT CONCEPTS

Review Article

Current Concepts PATIENT-RELATED RISK FACTORS Potential patient-related risk factors that may contribute to the risk of postoperative pulmonary complications include smoking, poor general health sta- PREOPERATIVE PULMONARY tus, older age, obesity, chronic obstructive pulmonary EVALUATION disease, and asthma (Table 1).

Smoking GERALD W. SMETANA, M.D. Smoking is a risk factor for postoperative pulmonary complications, as has been demonstrated re- OSTOPERATIVE pulmonary complications are peatedly since the first report in 1944.8 Smoking in- an important part of the risk of surgery and creases risk even among those without chronic lung prolong the hospital stay by an average of one disease.3 The relative risk of pulmonary complica- P 1 to two weeks. Much of the literature on the assess- tions among smokers as compared with nonsmokers ment of perioperative risk has focused on identifying ranges from 1.4 to 4.3. Unfortunately, the risk de- the now well-defined cardiac risk factors. However, clines only after eight weeks of preoperative cessa- clinically significant postoperative pulmonary compli- tion. Warner et al.20 prospectively studied 200 smok- cations are as common as postoperative cardiac com- ers preparing for coronary bypass surgery and found plications. According to one review, pulmonary com- a lower risk of pulmonary complications among plications were at least as common as or more those who had stopped smoking at least eight weeks common than cardiac complications in 17 of 25 before surgery than among current smokers (14.5 studies of postoperative complications.2 This article percent vs. 33 percent). Paradoxically, those who reviews patient- and procedure-related risk factors, had stopped smoking less than eight weeks earlier clinical evaluation, pulmonary-function testing, and had a higher risk than current smokers. risk-reduction strategies. The evaluation of candi- General Health Status dates for lung resection has been reviewed extensively and is not discussed here. The Goldman cardiac-risk index includes factors from the patient’s history, the physical examination, DEFINITION OF POSTOPERATIVE and laboratory data.21 This index predicts pulmonary PULMONARY COMPLICATIONS as well as cardiac complications.1,22 In addition, the Definitions of postoperative pulmonary complica- widely used American Society of Anesthesiologists tions have varied widely. This variation, along with classification, which was developed to evaluate the differences in the selection of patients, accounts for risk of overall perioperative mortality, is strongly pre- 23,24 most of the differences in the reported incidence of dictive of postoperative pulmonary complications. 25 pulmonary complications. Many early definitions in- Poor exercise capacity also identifies patients at risk. cluded complications of no clinical significance. For In a multivariate analysis of patients over 65 years of example, one study included all patients with a tem- age who were undergoing abdominal or nonresec- perature of 37.2°C (99°F) or higher, productive tive thoracic surgery, the inability to perform two cough, and any new physical signs on chest exami- minutes of supine bicycle exercise sufficient to raise nation.3 Recent studies have generally included only the heart rate to 99 beats per minute was the strong- 22 complications known either to prolong the hospital est predictor of pulmonary complications. The in- stay or to contribute to morbidity and mortality.4-6 ability to exercise predicted 79 percent of pulmonary These important pulmonary complications include complications; only 33 percent of patients without pneumonia, respiratory failure with prolonged me- pulmonary complications were unable to exercise. chanical ventilation, bronchospasm, atelectasis, and Age exacerbation of underlying chronic lung disease. Studies suggesting an increased risk of pulmonary complications with older age have not generally con-

From the Division of General Medicine and Primary Care, Beth Israel trolled for coexisting conditions. When data are strat- Deaconess Medical Center and Harvard Medical School, Boston. Address ified according to the American Society of Anesthe- reprint requests to Dr. Smetana at the Division of General Medicine and siologists class, the overall perioperative mortality for Primary Care, Beth Israel Deaconess Medical Center, 330 Brookline Ave., 26 Boston, MA 02215. classes II through V is the same in all age groups. ©1999, Massachusetts Medical Society. In a study of 500 patients over 80 years of age, the

The New England Journal of Medicine

30-day mortality was 6.2 percent for all patients and Physicians should aggressively treat patients with less than 1 percent for patients in American Society chronic obstructive pulmonary disease who, before of Anesthesiologists class II.27 Most deaths were from surgery, do not have optimal reduction of symptoms myocardial infarction or infection. Age was not a pre- and airflow obstruction on physical examination and dictor of postoperative pulmonary complications in who do not have optimal exercise capacity. Elective two studies of patients with severe chronic obstruc- surgery should be deferred if acute exacerbation is tive pulmonary disease.4,23 Pulmonary complications present. The treatments are the same as those for pa- are more strongly related to coexisting conditions tients who are not about to have an operation. than to chronologic age, and therefore advanced age Combinations of bronchodilators, physical therapy, alone is not a reason to withhold surgery. antibiotics, smoking cessation, and corticosteroids reduce the risk of postoperative pulmonary compli- Obesity cations.19,30 There are few data on the preoperative Despite a common assumption that obesity in- benefit of individual agents. creases the risk of pulmonary complications, most Patients with symptomatic chronic obstructive pul- studies have found no such association (Table 1). For monary disease should be treated with inhaled ipra- example, a review of 10 series of morbidly obese pa- tropium. Inhaled beta-adrenergic–receptor agonists, tients undergoing gastric bypass surgery found an used up to four times daily, should be added as incidence of pneumonia and atelectasis of only 3.9 needed for symptoms. The two agents have an addi- percent, a rate similar to that for nonobese patients.28 tive effect.31 Theophylline reduces dyspnea in pa- More recently, Phillips and colleagues reported no tients in stable condition who continue to have symp- difference between obese and nonobese patients in toms despite the use of inhaled beta-agonists and the risk of pulmonary complications after laparo- anticholinergic drugs32 and may be added preopera- scopic cholecystectomy.29 Obesity is not a significant tively for such patients. Although not all patients risk factor for pulmonary complications. with chronic obstructive pulmonary disease respond to corticosteroid therapy, a two-week preoperative Chronic Obstructive Pulmonary Disease course of systemic corticosteroids is reasonable for Patients with chronic obstructive pulmonary disease patients who continue to have symptoms despite have an increased risk of postoperative pulmonary bronchodilator therapy and who are not at their best complications. The incidence varies according to the personal base-line level as determined by symptoms, definition of a complication and the severity of lung chest examination, and spirometry.33 disease; relative risks range from 2.7 to 4.7 (Table 1). Clinicians should reserve preoperative antibiotics

TABLE 1. POTENTIAL PATIENT-RELATED RISK FACTORS FOR POSTOPERATIVE PULMONARY COMPLICATIONS.

UNADJUSTED RELATIVE RISK POTENTIAL INCIDENCE OF ASSOCIATED RISK FACTOR*TYPE OF SURGERY STUDY PULMONARY COMPLICATIONS WITH FACTOR WHEN FACTOR WHEN FACTOR WAS PRESENT WAS ABSENT

percent

Smoking Coronary bypass Warner et al.7 39 11 3.4 Abdominal Wightman,3 Morton,8 15–46 6–21 1.4–4.3 Brooks-Brunn9 ASA class >II Unselected Wolters et al.10 26 16 1.7 Thoracic or abdominal Brooks-Brunn,9 Kroenke 26–44 13–18 1.5–3.2 et al.,11 Hall et al.,12 Garibaldi et al.13 Age >70 yr Unselected Wightman,3 Pedersen14 9–17 4–9 1.9–2.4 Thoracic or abdominal Garibaldi et al.,13 Thomas 17–22 12–21 0.9–1.9 et al.,15 Calligaro et al.16 Obesity Unselected Wightman3 11 9 1.3 Thoracic or abdominal Brooks-Brunn,9 Hall et al.,12 19–36 17–27 0.8–1.7 Garibaldi et al.,13 Moulton et al.,17 Dales et al.18 COPD Unselected Wightman,3 Pedersen et al.,5 6–26 2–8 2.7–3.6 Tarhan et al.19 Thoracic or abdominal Kroenke et al.11 18 4 4.7

*ASA denotes American Society of Anesthesiologists, and COPD chronic obstructive pulmonary disease.

CURRENT CONCEPTS for outpatients in whom the presence of infection is incidence of infection or other postoperative com- suggested by a change in the character or amount of plications in patients with asthma.37,38 sputum. The indiscriminate use of preoperative antibiotics does not reduce the risk of postoperative PROCEDURE-RELATED RISK FACTORS pneumonia in patients undergoing nonthoracic sur- The surgical site is the most important predictor gery.34 The risk of postoperative pulmonary compli- of pulmonary risk (Table 2). Risk increases as the in- cations in patients with viral upper respiratory infec- cision approaches the diaphragm. Upper abdominal tions is unknown. A study of patients with asthma and thoracic surgery carries the greatest risk of post- found no increase in postoperative pulmonary com- operative pulmonary complications, ranging from plications among patients who had had symptoms of 10 to 40 percent. The risk is much lower for laparo- upper respiratory infection within the previous 30 scopic cholecystectomy (0.3 to 0.4 percent) than for days.24 No studies, however, have evaluated high-risk open cholecystectomy (13 to 33 percent). Postoper- patients undergoing major abdominal or thoracic ative pulmonary complications are rare after opera- surgery. It is reasonable to delay truly elective sur- tions outside the thorax or abdomen. gery in patients with upper respiratory infection. Table 3 summarizes studies of other procedure- related risk factors. Surgical procedures lasting more Asthma than three hours are associated with a higher risk An early study reported overall rates of postoper- of pulmonary complications.9,44 Whenever possible, ative complications of 24 percent in patients with one should consider shorter procedures for patients asthma and 14 percent in controls.35 Recent studies who have unmodifiable factors that place them at have not confirmed this risk. For example, in a study high risk. of 706 patients with asthma, 33 percent of whom Most studies have reported a lower risk of pulmo- had received bronchodilators within the previous 30 nary complications with epidural or spinal anesthesia days, the overall incidence of perioperative broncho- than with general anesthesia (Table 3), although the spasm was only 1.7 percent.24 Before surgery, pa- results are mixed.25 For example, in an early retro- tients should be free of wheezing, with a peak flow spective study of patients with chronic obstructive greater than 80 percent of the predicted or personal pulmonary disease, 8 percent of 464 patients under- best value.36 If necessary, the patient should receive going general anesthesia died of respiratory failure, oral corticosteroids (60 mg of prednisone daily or whereas there were no deaths among 121 patients the equivalent) to achieve this goal. A short course receiving spinal or epidural anesthesia.19 No studies of perioperative corticosteroids does not increase the reported a higher risk of pulmonary complications

TABLE 2. INFLUENCE OF SURGICAL SITE ON RATES OF POSTOPERATIVE PULMONARY COMPLICATIONS.

STUDY YEAR TYPE OF SURGERY UPPER LOWER LAPAROSCOPIC ABDOMINAL ABDOMINAL CHOLECYSTECTOMY THORACIC ALL OTHER

% of cases with complications (total no. of cases)

Pooler39 1949 19 (331) 11 (1334) 0.7 (4204) Wightman3 1968 19 (130) 6 (323) 0.6 (330) Tarhan et al.19* 1973 13 (75) 7 (45) 10 (112) 3 (396) Gracey et al.6† 1979 25 (57) 0 (7) 19 (21) 17 (72) Garibaldi et al.13‡ 1981 17 (201) 5 (208) 40 (102) Pedersen et al.5 1990 33 (419) 16 (200) 3 (6687) Southern Surgeons 1991 0.3 (1518) Club40 Phillips et al.29 1994 0.4 (841) Brooks-Brunn9 1997 28 (238) 15 (162)

*Patients had moderate-to-severe chronic obstructive pulmonary disease. Complication was defined as postoperative death. †Patients had chronic obstructive pulmonary disease. ‡Complication was defined as pneumonia.

The New England Journal of Medicine

TABLE 3. PROCEDURE-RELATED RISK FACTORS FOR POSTOPERATIVE PULMONARY COMPLICATIONS.

UNADJUSTED RELATIVE RISK INCIDENCE OF ASSOCIATED RISK FACTOR TYPE OF SURGERY STUDY PULMONARY COMPLICATIONS WITH FACTOR WHEN FACTOR WHEN FACTOR WAS PRESENT WAS ABSENT

percent

Surgery lasting Unselected Kroenke et al.,4 Pedersen 10–53 3–15 1.6–5.2 >3 hr et al.,5 Tarhan et al.19 Thoracic or abdominal Garibaldi et al.13 40 11 3.6 General anesthesia Unselected Gracey et al.,6 Tarhan et al.19 8–19 0–17 1.2–∞ Thoracic, abdominal, Christopherson et al.,41 28–32 11–12 2.2–3.0 or vascular Yeager et al.42 Intraoperative Unselected Berg et al.43 17* 5 3.2 pancuronium

*The value is the incidence among patients with residual neuromuscular blockade.

with spinal anesthesia. One should consider spinal or sensus that all candidates for lung resection should epidural anesthesia, if possible, for surgery in pa- undergo preoperative pulmonary-function testing. tients at high risk for pulmonary complications. Re- Such testing should be performed selectively in pa- gional anesthesia, such as an axillary block, carries a tients undergoing other surgical procedures. It has lower risk than either spinal or general anesthesia. been suggested that an increased risk of pulmonary A recent study reported a higher risk of pulmo- complications is associated with a forced expiratory nary complications among patients receiving the long- volume in one second (FEV1) or forced vital capacity acting neuromuscular blocker pancuronium than (FVC) of less than 70 percent of the predicted value, 46 among those receiving the shorter-acting atracurium or a ratio of FEV1 to FVC of less than 65 percent. or vecuronium.43 This prospective study of 691 pa- In 1990 a position paper of the American College tients found three times as many pulmonary compli- of Physicians recommended spirometry for patients cations among the patients receiving pancuronium with a history of tobacco use or dyspnea who are who had postoperative residual neuromuscular block undergoing coronary bypass or upper abdominal as among those who did not. The increased rate of surgery, patients with unexplained dyspnea or pul- pulmonary complications was due to more frequent monary symptoms who are undergoing head and and prolonged residual neuromuscular blockade, neck, orthopedic, or lower abdominal surgery, and which resulted in postoperative hypoventilation. The all patients who are undergoing lung resection.47 intensity of monitoring was equal in all groups. The However, a critical appraisal of 22 early studies of use of pancuronium should be avoided in patients at preoperative spirometry found methodologic flaws high risk for pulmonary complications. in every study.48 Recent studies show that spirometry has a variable PREOPERATIVE CLINICAL EVALUATION predictive value (Table 4). In addition, clinical find- A careful history taking and physical examination ings are generally more predictive of pulmonary com- are the most important parts of preoperative pulmo- plications than spirometric results in the few studies nary risk assessment. One should seek a history of that have evaluated both factors.11,25,57 However, no exercise intolerance,22 chronic cough, or unexplained prospective, randomized trials have addressed this dyspnea. The physical examination may identify find- question. For example, in a case–control study of ings suggestive of unrecognized pulmonary disease. 164 patients undergoing elective abdominal surgery, Among such findings, decreased breath sounds, dull- abnormal findings on chest examination were highly ness to percussion, wheezes, rhonchi, and a pro- associated with the risk of pulmonary complications, longed expiratory phase predict an increase in the whereas abnormal spirometric findings were not pre- risk of pulmonary complications.1,45 dictive (odds ratios, 5.8 and 1.0, respectively).1 One potential rationale for preoperative pulmo- PREOPERATIVE PULMONARY-FUNCTION nary-function testing is to identify patients in whom TESTING the risk is so high that surgery should be canceled. The value of routine preoperative pulmonary- However, even patients at very high risk as defined function testing remains controversial. There is con- spirometrically can undergo surgery with an accept-

TABLE 4. USE OF PREOPERATIVE SPIROMETRY TO PREDICT POSTOPERATIVE PULMONARY COMPLICATIONS.

RR ASSOCIATED WITH ABNORMAL INCIDENCE OF FINDINGS STUDY YEAR TYPE OF SURGERY PULMONARY COMPLICATIONS (95% CI)* AMONG PATIENTS AMONG PATIENTS WITH ABNORMAL WITH NORMAL SPIROMETRIC SPIROMETRIC FINDINGS FINDINGS

% (total no. of cases)

Collins et al.49 1968 Upper abdominal 44 (25) 17 (95) 2.6 (1.4–4.9) Stein and Cassara30 1970 Unselected 42 (48) 10 (29) 4.0 (1.3–12.4) Appleberg et al.50 1974 Unselected 16 (76) 8 (24) 1.9 (0.5–7.9) Fogh et al.51 1987 Abdominal 27 (22) 16 (100) 1.7 (0.8–3.9) Poe et al.52 1988 Cholecystectomy 34 (38) 11 (171) 3.3 (1.8–6.1) Svensson et al.53†‡ 1991 Aortic 49 (47) 32 (44) 1.5 (0.9–2.6) Kispert et al.54 1992 Vascular 23 (62) 6 (85) 3.8 (1.5–10.1) Kroenke et al.11§ 1993 Abdominal and 52 (52) 39 (52) 1.4 (0.9–2.1) thoracic Kocabas et al.45 1996 Upper abdominal 50 (16) 30 (44) 1.7 (0.9–3.3) Bando et al.55‡ 1997 Cardiac 11 (67) 10 (301) 1.0 (0.5–2.2) Jacob et al.56 1997 Coronary bypass 45 (107) 50 (56) 0.9 (0.6–1.3)

*RR denotes relative risk, and CI confidence interval.

†Abnormal findings were defined as an FEV1 of less than 73 percent of the predicted value. ‡Complication was defined as the need for mechanical ventilation for more than 48 hours.

§Abnormal findings were defined as an FEV1 that was 50 to 79 percent of the predicted value and an FEV1:FVC of less than 70 percent.

able risk of pulmonary complications. Kroenke and obstruction has been optimally reduced. The results colleagues studied 107 operations in patients with of preoperative pulmonary-function testing should severe chronic obstructive pulmonary disease as de- not be used to deny a patient surgery. fined by an FEV1 of less than 50 percent of the pre- On the basis of small case series, several authors dicted value and a ratio of FEV1 to FVC of less than have found a partial pressure of arterial carbon diox- 4 70 percent. Postoperative pulmonary complications ide (PaCO2) greater than 45 mm Hg to be a strong occurred in only 29 percent of the patients, with risk factor for pulmonary complications.58,59 In these only 1 death among 97 patients undergoing noncor- studies, all patients with an elevated PaCO2 also had onary procedures. This study and others of patients substantial airway obstruction on spirometry, al- at high risk suggest no threshold beyond which the though symptoms and exercise capacity were not re- risk of surgery is prohibitive.11,23 ported. Physicians would probably identify these pa- Physicians should use a variety of strategies to re- tients as being at high risk on clinical grounds alone. duce the risk of pulmonary complications in patients A more recent prospective study did not find an el- at high risk after a thorough clinical evaluation and evated PaCO2 to be a risk factor among candidates review of risk factors. No data suggest that spirom- for lung resection.60 Clinicians should not use arte- etry identifies a high-risk group with abnormal spi- rial-blood gas analyses to identify patients for whom rometric findings but no clinical evidence of pulmo- the risk of surgery is prohibitive. nary disease or other risk factors for pulmonary complications. Clinicians may reserve preoperative PULMONARY-RISK INDEXES spirometry for patients who are to undergo thoracic Clinicians have used cardiac-risk indexes to esti- or upper abdominal surgery and who have symp- mate perioperative risk for more than two decades.21 toms of cough, dyspnea, or exercise intolerance that In 1993, Epstein and colleagues developed a car- remain unexplained after a careful history taking and diopulmonary-risk index that is the sum of a modi- physical examination. In addition, spirometry may fied Goldman index and pulmonary risk factors, in- be helpful in a patient with chronic obstructive pul- cluding obesity, smoking, productive cough, diffuse monary disease or asthma if, after clinical assess- wheezing or rhonchi, an FEV1:FVC ratio of less than 61 ment, it is uncertain whether the degree of airflow 70 percent, and a PaCO2 greater than 45 mm Hg.

Volume 340 Number 12 · 941 Downloaded from www.nejm.org at LEICESTER GEN HOSP on August 26, 2004. Copyright © 1999 Massachusetts Medical Society. All rights reserved. The New England Journal of Medicine

In the original series of candidates for lung resection, the odds ratio for pulmonary complications was TABLE 5. RISK-REDUCTION STRATEGIES. 13 for patients with scores of 4 or more (of 10 possible points), as compared with those with scores of PREOPERATIVE 3 or less. This index has not been validated in pa- Encourage cessation of cigarette smoking for at least tients undergoing abdominal surgery. A drawback of 8 wk this index is the need for routine preoperative pulmo- Treat airflow obstruction in patients with chronic nary-function testing and arterial-blood gas analysis. obstructive pulmonary disease or asthma In a case–control study of patients undergoing Administer antibiotics and delay surgery if respiratory infection is present abdominal surgery,1 a multivariate analysis found that, Begin patient education regarding lung-expansion in descending order of importance, abnormal results maneuvers on chest examination, an abnormal chest radiograph, a high score on the Goldman cardiac-risk in- INTRAOPERATIVE dex, and a high score on the Charlson comorbidity Limit duration of surgery to less than 3 hr index were independently associated with an elevat- Use spinal or epidural anesthesia* ed pulmonary risk. This scheme has several advan- Avoid use of pancuronium tages. It does not require routine spirometry or ar- Use laparoscopic procedures when possible Substitute less ambitious procedure for upper terial-blood gas analysis, and it was developed in a abdominal or thoracic surgery when possible population of patients undergoing abdominal surgery. POSTOPERATIVE RISK-REDUCTION STRATEGIES Use deep-breathing exercises or incentive spirometry Table 5 summarizes established risk-reduction strat- Use continuous positive airway pressure egies that can be used throughout the perioperative Use epidural analgesia* period. Postoperative strategies include lung-expan- Use intercostal nerve blocks* sion maneuvers and pain control. *This strategy is recommended, although variable efficacy has been reported in the literature. Lung-Expansion Maneuvers Lung-expansion maneuvers are the mainstay of postoperative prevention in patients at high risk. These techniques lower the risk of atelectasis by increasing patients who are unable to perform deep-breathing lung volume. Deep-breathing exercises, which are a exercises or incentive spirometry. component of chest physical therapy, and incentive Nasal bilevel positive airway pressure minimizes spirometry have been studied most extensively. A the expected decline in lung volume after surgery critical review of these techniques found that they and may reduce postoperative pulmonary complica- consistently reduced the relative risk of pulmonary tions. In a study of 29 medical and postoperative pa- complications by 50 percent.62 Most studies have tients with impending respiratory failure, 76 percent found no difference in efficacy between these two of those treated with bilevel positive airway pressure techniques. For example, a recent meta-analysis of did not require intubation.66 Bilevel positive airway 14 studies of incentive spirometry and deep-breath- pressure is effective for the treatment of postopera- ing exercises, as compared with no intervention, found tive atelectasis and respiratory failure. Because less odds ratios of 0.44 and 0.43 for pulmonary compli- costly lung-expansion maneuvers are available, phy- cations, respectively.63 Preoperative education in lung- sicians should not use bilevel positive airway pres- expansion maneuvers reduces pulmonary complica- sure routinely for the primary prevention of compli- tions to a greater degree than instruction that begins cations. after surgery.44,64 Intermittent positive-pressure breathing is also ef- Pain Control fective, although most physicians have abandoned its Postoperative epidural analgesia reduces the rate routine use for primary prevention because of its rel- of pulmonary complications in patients at high risk. atively high cost and incidence of complications.44 In one study, for example, 75 patients undergoing Postoperative continuous positive airway pressure is cholecystectomy, including 31 with pulmonary dis- at least as effective as deep breathing and incentive ease, were randomly assigned to receive epidural an- spirometry.65 The principal advantage of continuous algesia or parenteral narcotics.67 The rates of pulmo- positive airway pressure is that it is not dependent on nary complications were 24 percent and 56 percent, the patient’s effort. It is costly, however, and requires respectively. Not all reports have shown a similar re- more intensive involvement by hospital personnel duction in risk, although no studies have found ex- than other methods. Clinicians should restrict the cess pulmonary morbidity with epidural analgesia. use of continuous positive airway pressure for the Epidural analgesia is safe. Among 1106 patients treat- primary prevention of pulmonary complications to ed with postoperative epidural morphine in one study,

942 · March 25, 1999 Downloaded from www.nejm.org at LEICESTER GEN HOSP on August 26, 2004. Copyright © 1999 Massachusetts Medical Society. All rights reserved. CURRENT CONCEPTS there were only two episodes of respiratory depres- factors in postoperative pulmonary complications: a blinded prospective 68 study of coronary artery bypass patients. Mayo Clin Proc 1989;64:609-16. sion. Hypotension is more common among pa- 21. Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of tients receiving epidural analgesia, but it responds to cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297: treatment and is not a cause of morbidity. Postoper- 845-50. 22. Gerson MC, Hurst JM, Hertzberg VS, Baughman R, Rouan GW, Ellis ative epidural analgesia is recommended after high- K. Prediction of cardiac and pulmonary complications related to elective risk thoracic, abdominal, and major vascular pro- abdominal and noncardiac thoracic surgery in geriatric patients. Am J Med 1990;88:101-7. cedures. 23. Wong D, Weber EC, Schell MJ, Wong AB, Anderson CT, Barker SJ. A meta-analysis of the use of postoperative inter- Factors associated with postoperative pulmonary complications in patients costal nerve blocks reported a relative risk of 0.47 with severe chronic obstructive pulmonary disease. Anesth Analg 1995;80: 276-84. for all pulmonary complications, but the risk reduc- 24. Warner DO, Warner MA, Barnes RD, et al. Perioperative respiratory tion did not achieve statistical significance.69 This complications in patients with asthma. Anesthesiology 1996;85:460-7. treatment is an option if epidural analgesia is ineffec- 25. Williams-Russo P, Charlson ME, MacKenzie CR, Gold JP, Shires GT. Predicting postoperative pulmonary complications: is it a real problem? tive or technically difficult. Arch Intern Med 1992;152:1209-13. 26. Marx GF, Mateo CV, Orkin LR. Computer analysis of postanesthetic deaths. Anesthesiology 1973;39:54-8. 27. Djokovic JL, Hedley-Whyte J. Prediction of outcome of surgery and I am indebted to Mark Aronson, M.D., Roger Davis, Sc.D., Erin anesthesia in patients over 80. JAMA 1979;242:2301-6. Hartman, M.S., and Russell Phillips, M.D., for their review of the 28. Pasulka PS, Bistrian BR, Benotti PN, Blackburn GL. The risks of sur- manuscript and their helpful suggestions. gery in obese patients. Ann Intern Med 1986;104:540-6. 29. Phillips EH, Carroll BJ, Fallas MJ, Pearlstein AR. Comparison of lap- REFERENCES aroscopic cholecystectomy in obese and non-obese patients. Am Surg 1994;60:316-21. 1. Lawrence VA, Dhanda R, Hilsenbeck SG, Page CP. Risk of pulmonary 30. Stein M, Cassara EL. Preoperative pulmonary evaluation and therapy complications after elective abdominal surgery. Chest 1996;110:744-50. for surgery patients. JAMA 1970;211:787-90. 2. Lawrence VA, Hilsenbeck SG, Mulrow CD, Dhanda R, Sapp J, Page 31. The COMBIVENT Inhalation Aerosol Study Group. In chronic ob- CP. Incidence and hospital stay for cardiac and pulmonary complications structive pulmonary disease, a combination of ipratropium and albuterol is after abdominal surgery. J Gen Intern Med 1995;10:671-8. more effective than either agent alone: an 85-day multicenter trial. Chest 3. Wightman JA. A prospective survey of the incidence of postoperative 1994;105:1411-9. pulmonary complications. Br J Surg 1968;55:85-91. 32. Vaz Fragoso CA, Miller MA. Review of the clinical efficacy of theo- 4. Kroenke K, Lawrence VA, Theroux JF, Tuley MR. Operative risk in pa- phylline in the treatment of chronic obstructive pulmonary disease. Am tients with severe obstructive pulmonary disease. Arch Intern Med 1992; Rev Respir Dis 1993;147:S40-S47. 152:967-71. 33. Mendella LA, Manfreda J, Warren CP, Anthonisen NR. Steroid re- 5. Pedersen T, Eliasen K, Henriksen E. A prospective study of risk factors sponse in stable chronic obstructive pulmonary disease. Ann Intern Med and cardiopulmonary complications associated with anaesthesia and sur- 1982;96:17-21. gery: risk indicators of cardiopulmonary morbidity. Acta Anaesthesiol 34. Lazlo G, Archer GG, Darrell JH, Dawson JM, Fletcher CM. The di- Scand 1990;34:144-55. agnosis and prophylaxis of pulmonary complications of surgical operations. 6. Gracey DR, Divertie MB, Didier EP. Preoperative pulmonary prepara- Br J Surg 1973;60:129-34. tion of patients with chronic obstructive pulmonary disease: a prospective 35. Gold MI, Helrich M. A study of complications related to anesthesia in study. Chest 1979;76:123-9. asthmatic patients. Anesth Analg 1963;42:238-93. 7. Warner MA, Divertie MB, Tinker JH. Preoperative cessation of smok- 36. Guidelines for the diagnosis and management of asthma. National ing and pulmonary complications in coronary artery bypass patients. An- Heart, Lung, and Blood Institute, National Asthma Education Program, esthesiology 1984;60:380-3. expert panel report. X. Special considerations. J Allergy Clin Immunol 8. Morton HJV. Tobacco smoking and pulmonary complications after sur- 1991;88:Suppl:523-34. gery. Lancet 1944;1:368-70. 37. Pien LC, Grammer LC, Patterson R. Minimal complications in a sur- 9. Brooks-Brunn JA. Predictors of postoperative pulmonary complications gical population with severe asthma receiving prophylactic corticosteroids. following abdominal surgery. Chest 1997;111:564-71. J Allergy Clin Immunol 1988;82:696-700. 10. Wolters U, Wolf T, Stutzer H, Schroder T. ASA classification and peri- 38. Kabalin CS, Yarnold PR, Grammer LC. Low complication rate of cor- operative variables as predictors of postoperative outcome. Br J Anaesth ticosteroid-treated asthmatics undergoing surgical procedures. Arch Intern 1996;77:217-22. Med 1995;155:1379-84. 11. Kroenke K, Lawrence VA, Theroux JF, Tuley MR, Hilsenbeck S. Post- 39. Pooler HE. Relief of post-operative pain and its influence on vital ca- operative complications after thoracic and major abdominal surgery in pa- pacity. BMJ 1949;2:1200-3. tients with and without obstructive lung disease. Chest 1993;104:1445-51. 40. The Southern Surgeons Club. A prospective analysis of 1518 laparo- 12. Hall JC, Tarala RA, Hall JL, Mander J. A multivariate analysis of the scopic cholecystectomies. N Engl J Med 1991;324:1073-8. [Erratum, risk of pulmonary complications after laparotomy. Chest 1991;99:923-7. N Engl J Med 1991;325:1517-8.] 13. Garibaldi RA, Britt MR, Coleman ML, Reading JC, Page NL. Risk 41. Christopherson R, Beattie C, Frank SM, et al. Perioperative morbidity factors for postoperative pneumonia. Am J Med 1981;70:677-80. in patients randomized to epidural or general anesthesia for lower extrem- 14. Pedersen T. Complications and death following anaesthesia: a prospec- ity vascular surgery. Anesthesiology 1993;79:422-34. tive study with special reference to the influence of patient-, anaesthesia-, 42. Yeager MP, Glass DD, Neff RK, Brink-Johnsen T. Epidural anesthesia and surgery-related risk factors. Dan Med Bull 1994;41:319-31. and analgesia in high-risk surgical patients. Anesthesiology 1987;66:729- 15. Thomas P, Doddoli C, Neville P, et al. Esophageal cancer resection in 36. the elderly. Eur J Cardiothorac Surg 1996;10:941-6. 43. Berg H, Viby-Mogensen J, Roed J, et al. Residual neuromuscular 16. Calligaro KD, Azurin DJ, Dougherty MJ, et al. Pulmonary risk factors block is a risk factor for postoperative pulmonary complications: a prospec- of elective abdominal aortic surgery. J Vasc Surg 1993;18:914-21. tive, randomised, and blinded study of postoperative pulmonary complica- 17. Moulton MJ, Creswell LL, Mackey ME, Cox JL, Rosenbloom M. tions after atracurium, vecuronium, and pancuronium. Acta Anaesthesiol Obesity is not a risk factor for significant adverse outcomes after cardiac Scand 1997;41:1095-103. surgery. Circulation 1996;94:Suppl II:II-87–II-92. 44. Celli BR, Rodriguez KS, Snider GL. A controlled trial of intermittent 18. Dales RE, Dionne G, Leech JA, Lunau M, Schweitzer I. Preoperative positive pressure breathing, incentive spirometry, and deep breathing exer- prediction of pulmonary complications following thoracic surgery. Chest cises in preventing pulmonary complications after abdominal surgery. Am 1993;104:155-9. Rev Respir Dis 1984;130:12-5. 19. Tarhan S, Moffitt EA, Sessler AD, Douglas WW, Taylor WF. Risk of 45. Kocabas A, Kara K, Ozgur G, Sonmez H, Burgut R. Value of preop- anesthesia and surgery in patients with chronic bronchitis and chronic ob- erative spirometry to predict postoperative pulmonary complications. Res- structive pulmonary disease. Surgery 1973;74:720-6. pir Med 1996;90:25-33. 20. Warner MA, Offord KP, Warner ME, Lennon RL, Conover MA, Jans- 46. Gass GD, Olsen GN. Preoperative pulmonary function testing to pre- son-Schumacher U. Role of preoperative cessation of smoking and other dict postoperative morbidity and mortality. Chest 1986;89:127-35.

Volume 340 Number 12 · 943 Downloaded from www.nejm.org at LEICESTER GEN HOSP on August 26, 2004. Copyright © 1999 Massachusetts Medical Society. All rights reserved. The New England Journal of Medicine

47. American College of Physicians. Preoperative pulmonary function test- 59. Stein M, Koota GM, Simon M, Frank HA. Pulmonary evaluation of ing. Ann Intern Med 1990;112:793-4. surgical patients. JAMA 1962;181:765-70. 48. Lawrence VA, Page CP, Harris GD. Preoperative spirometry before 60. Kearney DJ, Lee TH, Reilly JJ, DeCamp MM, Sugarbaker DJ. Assess- abdominal operations: a critical appraisal of its predictive value. Arch Intern ment of operative risk in patients undergoing lung resection: importance Med 1989;149:280-5. of predicted pulmonary function. Chest 1994;105:753-9. 49. Collins CD, Darke CS, Knowelden J. Chest complications after upper 61. Epstein SK, Faling LJ, Daly BD, Celli BR. Predicting complications abdominal surgery: their anticipation and prevention. BMJ 1968;1:401- after pulmonary resection: preoperative exercise testing vs a multifactorial 6. cardiopulmonary risk index. Chest 1993;104:694-700. 50. Appleberg M, Gordon L, Fatti LP. Preoperative pulmonary evaluation 62. Brooks-Brunn JA. Postoperative atelectasis and pneumonia. Heart of surgical patients using the vitalograph. Br J Surg 1974;61:57-9. Lung 1995;24:94-115. 51. Fogh J, Wille-Jorgensen P, Brynjolf I, et al. The predictive value of pre- 63. Thomas JA, McIntosh JM. Are incentive spirometry, intermittent pos- operative perfusion/ventilation scintigraphy, spirometry and x-ray of the itive pressure breathing, and deep breathing exercises effective in the pre- lungs on postoperative pulmonary complications: a prospective study. Acta vention of postoperative pulmonary complications after upper abdominal Anaesthesiol Scand 1987;31:717-21. surgery? A systematic overview and meta-analysis. Phys Ther 1994;74:3-16. 52. Poe RH, Kallay MC, Dass T, Celebic A. Can postoperative pulmonary 64. Castillo R, Haas A. Chest physical therapy: comparative efficacy of complications after elective cholecystectomy be predicted? Am J Med Sci preoperative and postoperative in the elderly. Arch Phys Med Rehabil 1988;295:29-34. 1985;66:376-9. 53. Svensson LG, Hess KR, Coselli JS, Safi HJ, Crawford ES. A prospec- 65. Stock MC, Downs JB, Gauer PK, Alster JM, Imrey PB. Prevention of tive study of respiratory failure after high-risk surgery on the thoracoab- postoperative pulmonary complications with CPAP, incentive spirometry, dominal aorta. J Vasc Surg 1991;14:271-82. and conservative therapy. Chest 1985;87:151-7. 54. Kispert JF, Kazmers A, Roitman L. Preoperative spirometry predicts 66. Pennock BE, Kaplan PD, Carlin BW, Sabangan JS, Magovern JA. Pres- perioperative pulmonary complications after major vascular surgery. Am sure support ventilation with a simplified ventilatory support system ad- Surg 1992;58:491-5. ministered with a nasal mask in patients with respiratory failure. Chest 55. Bando K, Sun K, Binford RS, Sharp TG. Determinants of longer du- 1991;100:1371-6. ration of endotracheal intubation after adult cardiac operations. Ann Tho- 67. Cuschieri RJ, Morran CG, Howie JC, McArdle CS. Postoperative pain rac Surg 1997;63:1026-33. and pulmonary complications: comparison of three analgesic regimens. Br 56. Jacob B, Amoateng-Adjepong Y, Rasakulasuriar S, Manthous CA, J Surg 1985;72:495-8. Haddad R. Preoperative pulmonary function tests do not predict outcome 68. Ready LB, Loper KA, Nessly M, Wild L. Postoperative epidural mor- after coronary artery bypass. Conn Med 1997;61:327-32. phine is safe on surgical wards. Anesthesiology 1991;75:452-6. 57. Cain HD, Stevens PM, Adaniya R. Preoperative pulmonary function 69. Ballantyne JC, Carr DB, deFerranti S, et al. The comparative effects and complications after cardiovascular surgery. Chest 1979;76:130-5. of postoperative analgesic therapies on pulmonary outcome: cumulative 58. Milledge JS, Nunn JF. Criteria of fitness for anaesthesia in patients meta-analyses of randomized, controlled trials. Anesth Analg 1998;86:598- with chronic obstructive lung disease. BMJ 1975;3:670-3. 612.

EFFECT OF SYSTEMIC GLUCOCORTICOIDS ON EXACERBATIONS OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE

DENNIS E. NIEWOEHNER, M.D., MARCIA L. ERBLAND, M.D., ROBERT H. DEUPREE, PH.D., DOROTHEA COLLINS, SC.D., NICHOLAS J. GROSS, M.D., PH.D., RICHARD W. LIGHT, M.D., PAULA ANDERSON, M.D., AND NANCY A. MORGAN, R.PH., M.B.A., FOR THE DEPARTMENT OF VETERANS AFFAIRS COOPERATIVE STUDY GROUP*

ABSTRACT ATIENTS with chronic obstructive pulmo- Background and Methods Although their clinical nary disease (COPD) frequently have exac- efficacy is unclear and they may cause serious ad- erbations that require hospitalization. Hos- verse effects, systemic glucocorticoids are a stand- pital treatment for this common condition is ard treatment for patients hospitalized with exacer- Passociated with high costs and relatively poor out- bations of chronic obstructive pulmonary disease comes.1 In addition to antibiotics, oxygen, and bron- (COPD). We conducted a double-blind, randomized chodilators, most hospitalized patients receive sys- trial of systemic glucocorticoids (given for two or temic glucocorticoids. Less severely ill patients often eight weeks) or placebo, in addition to other thera- receive oral glucocorticoids as outpatients. pies, for exacerbations of COPD. Most other care was standardized over the six-month period of fol- Systemic glucocorticoids improve outcomes in pa- 2 low-up. The primary end point was treatment failure, tients with acute asthma, but their clinical efficacy in defined as death from any cause or the need for the treatment of COPD is less clear. Two small trials intubation and mechanical ventilation, readmission suggested that several days of therapy with systemic to the hospital for COPD, or intensification of drug glucocorticoids improved the forced expiratory vol- therapy. ume in one second (FEV1) during exacerbations of Results Of 1840 potential study participants at 25 COPD.3,4 Another trial found that a single dose of Veterans Affairs medical centers, 271 were eligible methylprednisolone did not improve spirometric re- for participation and were enrolled; 80 received an sults over the succeeding five hours.5 None of these eight-week course of glucocorticoid therapy, 80 re- trials were explicitly designed to evaluate clinical out- ceived a two-week course, and 111 received placebo. comes. The role of systemic glucocorticoids in pa- About half the potential participants were ineligible 6 because they had received systemic glucocorticoids tients with stable COPD is similarly unclear. in the previous 30 days. Rates of treatment failure Adverse effects of the short-term administration were significantly higher in the placebo group than of systemic glucocorticoids include secondary infec- in the two glucocorticoid groups combined at 30 days tions, hyperglycemia, and a range of mood and be- (33 percent vs. 23 percent, P=0.04) and at 90 days havioral changes.7 Long-term therapy may cause os- (48 percent vs. 37 percent, P=0.04). Systemic gluco- teoporosis, cataracts, hypertension, myopathy, and corticoids (in both groups combined) were associat- adrenal insufficiency. ed with a shorter initial hospital stay (8.5 days, vs. We conducted a randomized, double-blind, place- 9.7 days for placebo; P=0.03) and with a forced ex- bo-controlled, multicenter trial to evaluate the effi- piratory volume in one second that was about 0.10 cacy of systemic glucocorticoids for exacerbations of liter higher than that in the placebo group by the first day after enrollment. Significant treatment benefits COPD. The principal objective was to determine rates were no longer evident at six months. The eight- of treatment failure. A secondary goal was to deter- week regimen of therapy was not superior to the mine the optimal duration of treatment. two-week regimen. The patients who received glucocorticoid therapy were more likely to have hyper- METHODS glycemia requiring therapy than those who received The Human Rights Committee of the Veterans Affairs Coop- placebo (15 percent vs. 4 percent, P=0.002). erative Studies Program and the institutional review boards of the Conclusions Treatment with systemic glucocorti- participating medical centers approved this study. All patients coids results in moderate improvement in clinical out- gave written informed consent. comes among patients hospitalized for exacerbations of COPD. The maximal benefit is obtained during the first two weeks of therapy. Hyperglycemia of suffi- From the Veterans Affairs medical centers in Minneapolis (D.E.N.), Little cient severity to warrant treatment is the most fre- Rock, Ark. (M.L.E., P.A.), Hines, Ill. (N.J.G.), and Long Beach, Calif. quent complication. (N Engl J Med 1999;340:1941-7.) (R.W.L.); the Cooperative Studies Program Coordinating Center, West Haven, Conn. (R.H.D., D.C.); and the Veterans Affairs Cooperative Studies ©1999, Massachusetts Medical Society. Program Clinical Research Pharmacy Coordinating Center, Albuquerque, N.M. (N.A.M.). Address reprint requests to Dr. Niewoehner at the Pul- monary Section (111N), Veterans Affairs Medical Center, 1 Veterans Dr., Minneapolis, MN 55417, or at [email protected]. *Other investigators who participated in the study are listed in the Ap- pendix.

Volume 340 Number 25 · 1941

The New England Journal of Medicine

Study Design pium bromide (two puffs from a metered-dose inhaler or a nebulizer treatment at least four times daily), and starting on day 4, We designed this study to assess the equivalence of two ap- inhaled triamcinolone acetonide (eight puffs daily in divided doses) proaches to the treatment of COPD.8,9 Systemic glucocorticoids or its equivalent. Use of theophylline, high-dose inhaled gluco- are the standard therapy for hospitalized patients with COPD, corticoids (more than eight puffs daily of triamcinolone acetonide even though they have adverse effects. Therefore, the withholding or its equivalent), and open-label systemic glucocorticoids was of glucocorticoids may be viewed as an experimental intervention not allowed. Treatment was considered to have failed if any of the associated with no glucocorticoid-related complications. The plan- forbidden medications were prescribed. Other medications were ning committee settled on a 7.5 percent absolute difference in the permitted according to medical need. We evaluated the patients rate of treatment failure as the clinically meaningful upper limit. on each of the first three hospital days and at two weeks, eight In other words, withholding glucocorticoids would be consid- weeks, and six months. We continued to obtain follow-up data ered the preferred treatment if the results showed a difference in for patients in whom the study drug had been withdrawn because the failure rate (the rate with placebo minus the rate with active of treatment failure or for other reasons. If a patient missed a visit, treatment) of 7.5 percent or less. The secondary objective was to we collected data by mail, telephone, or a review of medical records. assess the equivalence of two different periods of therapy (two and eight weeks). The follow-up period lasted for six months End Points from the time of enrollment. A detailed description of the rationale for the study, its design, the protocol, and the planned analy- The primary end point, a first treatment failure, was defined as ses is provided elsewhere.10 death from any cause or the need for intubation and mechanical ventilation, readmission because of COPD, or intensification of Study Population pharmacologic therapy. The patients’ primary physicians made all the clinical decisions. We defined intensification of pharmacologic All patients admitted to participating Veterans Affairs medical therapy as the prescription of open-label systemic glucocorti- centers for exacerbations of COPD were potential subjects. The coids, high-dose inhaled glucocorticoids (more than eight puffs principal inclusion criteria were a clinical diagnosis of exacerba- per day of triamcinolone acetonide or its equivalent), theophyl- tion of COPD, an age of 50 years or more, a history of 30 pack- line, or any combination of these three therapies. When multiple years or more of cigarette smoking, and either an FEV of 1.50 1 failures occurred on the same day, the assignment to the category liters or less or an inability to undergo spirometry because of of first failure was hierarchical, in the following descending order: dyspnea. The principal exclusion criteria were a diagnosis of asth- death, intubation, readmission, and intensification of therapy. When ma, use of systemic glucocorticoids within the preceding 30 days, a primary end point (other than death) was reached, the study coexisting medical conditions that made survival for at least 1 year treatment was terminated, and usual medical care was resumed. unlikely, and inability to give informed consent. We obtained Secondary end points were a change in FEV , the length of the base-line data on respiratory disease and other pertinent aspects 1 11 hospital stay, and death from any cause during the six months of of the medical history by means of a questionnaire. follow-up. The patients underwent spirometry at base line; on days 1, 2, and 3; and at the two-week, eight-week, and six-month Treatments visits. All centers performed spirometry (model 922, SensorMedics, We randomly assigned patients within 12 hours after presenta- Yorba Linda, Calif.) according to standard recommendations.12 We tion to one of three treatment groups. The first group received calculated the initial hospital stay as the period from the day of eight weeks of glucocorticoid therapy, consisting of intravenous admission to the day of discharge or transfer to an extended-care methylprednisolone (Solu-Medrol, Pharmacia & Upjohn, Kala- facility. mazoo, Mich.) (given in a dose of 125 mg every 6 hours for 72 hours) followed by once-daily oral prednisone (60 mg on study Complications days 4 through 7, 40 mg on days 8 through 11, 20 mg on days 12 We evaluated the patients for any possible adverse effects of through 43, 10 mg on days 44 through 50, and 5 mg on days 51 treatment at each visit. As described elsewhere,10 the diagnosis of through 57). The second group received two weeks of gluco- hyperglycemia, hypertension, secondary infection, upper gastro- corticoid therapy, consisting of intravenous methylprednisolone intestinal bleeding, or acute psychiatric illness required a consul- (125 mg every 6 hours for 72 hours), followed by oral predni- tation, an invasive procedure, or initiation of a specific therapy. sone (60 mg on days 4 through 7, 40 mg on days 8 through 11, We also questioned the patients about other possible adverse and 20 mg on days 12 through 15), with placebo capsules on events. study days 16 through 57. The third group received placebo, consisting of an equivalent volume of intravenous 5 percent dextrose Statistical Analysis solution (every 6 hours for 72 hours), followed by placebo capsules on days 4 through 57. Randomization was stratified accord- The base-line characteristics of the patients in the three treating to hospital with a permuted-block scheme; 40 percent of the ment groups were compared by means of analysis of variance for patients were assigned to the placebo group, 30 percent to the continuous variables and the chi-square test for categorical vari- eight-week glucocorticoid group, and 30 percent to the two- ables.13 All comparisons of results were based on the intention-to- week glucocorticoid group. treat principle. Treatment comparisons were made with the use of The Veterans Affairs Cooperative Studies Clinical Research a two-step procedure: if the findings for the two-week and the Pharmacy Coordinating Center distributed the study medica- eight-week groups were found to be equivalent, these two groups tions. Designated research pharmacists dispensed the intravenous were combined into a single active-treatment group for comparison medications in a blinded fashion. All patients received the same with placebo. Comparisons were made at 30, 90, and 182 days after number of identical-appearing study capsules in blister packs. We the start of treatment. Treatment failure, the primary end point, assessed compliance on the basis of capsule counts. was analyzed with use of the upper limit of a one-sided 95 percent The patients remained hospitalized for at least three days for confidence interval to determine therapeutic equivalence14 and a intravenous therapy, after which they received capsules of predni- two-sided log-rank test to compare differences between curves for 15 sone or placebo for eight weeks. Hospital staff decided the date the cumulative failure rate. Values for FEV1 in the glucocorticoid of discharge after three days of intravenous therapy. All the pa- and placebo groups were compared by analysis of variance, and hos- tients received a broad-spectrum antibiotic for seven days. For the pital stays were compared with use of the Wilcoxon two-sample entire six-month period, the patients were required to use an in- rank test.13 A complication rate was defined as the proportion of haled b-adrenergic agonist (two puffs from a metered-dose inhaler patients who had one or more episodes of a complication during or a nebulizer treatment at least four times daily), inhaled ipratro- the six months of follow-up. Logistic-regression analysis was used

1942 · June 24, 1999

EFFECT OF SYSTEMIC GLUCOCORTICOIDS ON EXACERBATIONS OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE

TABLE 1. BASE-LINE CHARACTERISTICS OF THE 271 PATIENTS ACCORDING TO TREATMENT ASSIGNMENT.*

GLUCOCORTICOIDS GLUCOCORTICOIDS PLACEBO FOR 2 WK FOR 8 WK CHARACTERISTIC (N=111) (N=80) (N=80)

Age — yr 67.8±10.0 67.1±10.6 68.1±6.8 Male sex — no. 111 80 77 White race — no. 99 59 68 Cough — no. (%) 65 (59) 41 (51) 43 (54) Sputum production — no. (%) 74 (67) 54 (68) 52 (65) Wheezing — no. (%) 98 (88) 70 (88) 72 (90) No. of chest colds per year — no. (%) None 17 (15) 17 (21) 14 (18) 1 or 2 64 (58) 51 (64) 50 (62) »3 30 (27) 12 (15) 16 (20) Smoked in past 3 mo — no. (%) 56 (50) 42 (52) 40 (50) Total cigarette smoking — pack-yr 77.0±35.5 67.3±31 80.9±43.8† Regular medications — no. (%) Inhaled beta-adrenergic agonist 96 (86) 66 (83) 72 (90) Inhaled anticholinergic drug 81 (73) 47 (59) 58 (72) Oral beta-adrenergic agonist 12 (11) 4 (5) 7 (9) Theophylline 37 (33) 26 (32) 30 (38) Inhaled glucocorticoids 49 (44) 39 (49) 40 (50) Use of oxygen at home — no. (%) 20 (18) 12 (15) 15 (19) Hospitalization for COPD in previous 2 yr 73 (66) 51 (64) 60 (75) — no. (%) Prior use of systemic glucocorticoids — no. (%) 52 (47) 30 (38) 46 (58)† Other illnesses — no. (%) Diabetes mellitus 5 (5) 12 (15) 11 (14)† History of ulcer 23 (21) 19 (24) 16 (20) Hypertension 44 (40) 39 (49) 39 (49) Disabling heart disease 13 (12) 8 (10) 8 (10) Disabling arthritis 12 (11) 10 (12) 15 (19) History of psychiatric disorder requiring 13 (12) 9 (11) 9 (11) hospitalization

FEV1 — ml‡ 750±271 772±286 785±288 Time from presentation to randomization — hr 3.7±2.6 3.8±2.6 3.7±2.2

*Plus–minus values are means ±SD. †P«0.05 for differences among groups by analysis of variance for continuous variables and by the chi-square test for categorical variables. ‡Data were available for 101 patients in the placebo group, 73 in the two-week glucocorticoid group, and 72 in the eight-week glucocorticoid group.

to identify variables that predicted treatment failure within six percent had taken systemic glucocorticoids in the pre- months.15 All reported P values are two-tailed. vious 30 days. Other common reasons for exclusion RESULTS included unwillingness or inability to participate (23.2 percent), a history of less than 30 pack-years of smok- Enrollment began in November 1994 and con- ing (14.6 percent), and coexisting medical conditions cluded in October 1996, one year ahead of schedule. expected to limit survival (18.4 percent). On the basis of interim analyses, the Veterans Affairs Eighty patients were assigned to receive glucocor- Cooperative Studies Evaluation Committee recom- ticoid therapy for eight weeks, 80 were assigned to mended termination of enrollment at that time. receive glucocorticoid therapy for two weeks, and Study Population 111 were assigned to receive placebo. The three treatment groups were similar with respect to base-line A total of 1840 potential patients at 25 Veterans characteristics (Table 1). There were small differenc- Affairs medical centers were screened for the study, es in total pack-years of cigarette smoking, prior use of whom 271 were found to be eligible and were en- of systemic glucocorticoids, and the prevalence of rolled. The enrollment rate was lower than had been diabetes mellitus. projected,10 largely because of a substantial decline in admissions for COPD throughout the Veterans Af- Discontinuation of Study Drugs and Compliance fairs medical system and an unexpectedly high rate of Study drugs were discontinued for reasons other exclusion because of recent use of systemic glucocor- than a primary end point in 10 patients assigned ticoids. Among the patients who were screened, 49.9 to placebo (9 percent), 10 assigned to two weeks of

Volume 340 Number 25 · 1943

The New England Journal of Medicine

20 Glucocorticoids, 8 wkG Glucocorticoids, 2 wkG 10 Placebo Rate of Treatment Failure (%) 0 061 2 3 4 5 Month

NO. AT RISK Glucocorticoids, 8 wkG 80G 61G 50G 21G Glucocorticoids, 2 wkG 80G 59G 46G 20G Placebo 111 74 58 39

Figure 1. Kaplan–Meier Estimates of the Rate of First Treatment Failure at Six Months, According to Treatment Group.

glucocorticoids (12 percent), and 5 assigned to eight As compared with placebo, glucocorticoids signif- weeks of glucocorticoids (6 percent). Follow-up icantly reduced the rate of first treatment failure at data were complete for 19 of these 25 patients. All 30 days (23 percent vs. 33 percent, P=0.04) and 90 available data were included in the analyses. On the days (37 percent vs. 48 percent, P=0.04) (Table 2). basis of counts of returned study capsules, the com- Treatment-failure rates did not differ significantly at pliance rate was 89 percent in the placebo group, six months (51 percent in the combined glucocorti- 85 percent in the two-week glucocorticoid group, coid groups vs. 54 percent in the placebo group, and 87 percent in the eight-week glucocorticoid P=0.58). The duration of glucocorticoid therapy group. (two weeks or eight weeks) had no significant effect on the rate of treatment failure at any time. Primary Outcomes Figure 1 shows Kaplan–Meier estimates of rates Length of Hospitalization of first treatment failure for the three study groups, The average length of the initial hospitalization was and Table 2 shows the reasons for treatment failure significantly longer in the placebo group than in the at 30, 90, and 182 days. At least one treatment fail- combined glucocorticoid groups (9.7 vs. 8.5 days, ure occurred in approximately half the patients. In- P=0.03). After the initial hospitalization, patients in tensification of therapy was the most common rea- the placebo group spent an average of 2.0 days in son for treatment failure, accounting for 70 percent the hospital because of COPD, as compared with of the total failures at 30 days, 62 percent at 90 1.9 days for patients in the glucocorticoid groups days, and 58 percent at 182 days. When therapy was (P=0.98). Glucocorticoid-treated patients, on aver- intensified, physicians administered open-label sys- age, spent more time in the hospital for reasons other temic glucocorticoids in more than 75 percent of than COPD than did patients receiving placebo (4.4 cases. vs. 1.2 days, P=0.07). The trial did not demonstrate equivalence of outcomes at any time. When the upper limits of one- Spirometric Findings sided confidence intervals are used to compare fail- FEV1 improved significantly faster in the patients ure rates between groups, the results show that the who received systemic glucocorticoids than in those withholding of glucocorticoids may have increased who received placebo (Fig. 2). The maximal differ- treatment-failure rates by as much as 20 percent at ence, approximately 0.10 liter, was evident by the 30 days, 21 percent at 90 days, and 12 percent at 182 first day after enrollment. By the end of two weeks, days. All values exceeded the limit of 7.5 percent set FEV1 did not differ significantly between the active- by the protocol. treatment and placebo groups.

1944 · June 24, 1999

EFFECT OF SYSTEMIC GLUCOCORTICOIDS ON EXACERBATIONS OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE

1.2 TABLE 2. CUMULATIVE PRIMARY OUTCOMES ACCORDING TO TREATMENT ASSIGNMENT. 127 67 1.1 GLUCO- GLUCO- Glucocorticoid 120 CORTICOIDS CORTICOIDS PLACEBO FOR 2 WK FOR 8 WK P 137 130 OUTCOME (N=111) (N=80) (N=80) VALUE* 1.0 * * number (percent) 140 82 (liters)

1 103 30 days 0.9 * Placebo

Death 3 (3) 0 2 (2) FEV Intubation 3 (3) 2 (2) 1 (1) 77 Readmission for COPD 5 (5) 4 (5) 2 (2) 145 92 86 Intensification of therapy 26 (23) 13 (16) 13 (16) 0.8 Total 37 (33) 19 (24) 18 (22) 0.04 90 days Death 4 (4) 2 (2) 2 (2) 95 Intubation 3 (3) 3 (4) 1 (1) 101 Readmission for COPD 13 (12) 8 (10) 6 (8) 0.7 Intensification of therapy 33 (30) 17 (21) 20 (25) Total 53 (48) 30 (38) 29 (36) 0.04 Day 0 Day 1 Day 2 Day 3 Week 2Week 8 182 days Week 26 Death† 4 (4) 2 (2) 3 (4)

Intubation 3 (3) 3 (4) 2 (2) Figure 2. Mean Forced Expiratory Volume in One Second (FEV1) Readmission for COPD 17 (15) 12 (15) 13 (16) at Selected Times According to Treatment Group. Intensification of therapy 36 (32) 22 (28) 24 (30) The two-week and eight-week glucocorticoid groups have been Total 60 (54) 39 (49) 42 (52) 0.58 combined. The numbers at each time point are the numbers of *P values are for comparisons of the placebo group with the combined patients in each group for whom data were available. The as- glucocorticoid groups, by the log-rank test. terisks denote P<0.05 for the comparison with placebo. The bars indicate standard errors. †Only deaths that were counted as primary outcomes are listed. The total numbers of deaths during six months of follow-up were 11 in the placebo group and 13 in the glucocorticoid groups.

Death from All Causes TABLE 3. COMPLICATIONS OF TREATMENT Over the six months of follow-up, 11 of the 111 DURING THE SIX-MONTH FOLLOW-UP PERIOD. patients receiving placebo and 13 of the 160 receiving glucocorticoids died (P=0.61). Seven deaths in GLUCOCORTICOIDS GLUCOCORTICOIDS the placebo group and six in the combined gluco- PLACEBO FOR 2 WK FOR 8 WK P corticoid groups were attributed to COPD. COMPLICATION (N=111) (N=80) (N=80) VALUE* number (percent) Complications Hyperglycemia 4 (4) 14 (18) 10 (12) 0.002 Table 3 shows the reported complications for each Gastrointestinal 5 (5) 0 3 (4) 0.21 treatment group over the six months. A greater pro- bleeding portion of patients in the glucocorticoid groups Secondary infection 19 (17) 12 (15) 18 (22) 0.73 than in the placebo group had hyperglycemia requir- Hypertension 4 (4) 6 (8) 4 (5) 0.33 ing treatment (15 percent vs. 4 percent, P=0.002). Psychiatric disorder 3 (3) 5 (6) 2 (2) 0.47 Other adverse 16 (14) 18 (22) 21 (26) 0.04 Twenty-two of the 24 episodes in the glucocorticoid events† groups occurred during the first 30 days of follow- up. Sixteen of the 24 glucocorticoid-treated patients *P values are for comparisons of the placebo group with the combined with hyperglycemia were known to have diabetes. glucocorticoid groups, by the chi-square test. The patients who received glucocorticoids also had †This category includes 41 different symptoms or conditions. more adverse events classified as “other” (P=0.04); these included 41 separate symptoms or conditions, most of which were not thought to be caused by glucocorticoids. Reported rates of secondary infection group and one in the two-week glucocorticoid group did not differ significantly among the three groups, were rehospitalized for infection. but the eight-week glucocorticoid group had the highest proportion of patients with serious infections. Subgroup Analyses Eleven of the patients in this group were rehospital- As specified by the protocol, we performed sub- ized with a primary diagnosis of infection; 9 of the group analyses for the following variables: base-line

11 had pneumonia. Only four patients in the placebo FEV1, theophylline use before randomization, hospi-

Volume 340 Number 25 · 1945

Downloaded from www.nejm.org at UNIV OF MEDICINE AND DENTISTRY OF NJ on February 17, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. The New England Journal of Medicine talization because of COPD in the previous two Osteoporosis was not evaluated in this trial, but years, a history of cough, a history of wheezing, a his- even relatively brief courses of systemic glucocorti- tory of sputum production, and a history of chest coids cause reductions in trabecular bone mineral colds. Multiple logistic regression indicated that a density.19 The cumulative effects of long-term ther- base-line value for FEV1 that was less than the median apy confer a substantial risk of painful vertebral frac- value of 0.73 liter predicted a higher rate of treat- tures and other long-term complications.7, 2 0 ment failure at 182 days (odds ratio, 1.8; 95 percent Intensification of pharmacologic therapy account- confidence interval, 1.1 to 3.1), as did theophylline ed for more than half of all treatment failures at six use before randomization (odds ratio, 2.3; 95 per- months and an even higher proportion during the cent confidence interval, 1.3 to 4.0). Only prior hos- early weeks of follow-up in our study. Open-label pitalization because of COPD had a significant in- glucocorticoids were administered in most of these teraction with the treatment assignment (P=0.01). cases. Thus, the principal consequence of withhold- Treatment with glucocorticoids was associated with ing glucocorticoids in patients receiving placebo a more favorable outcome in the group of 184 pa- was to delay their administration to about half of tients who had previously been hospitalized because these patients. The other half recovered and received of COPD than in the group of 87 with no history no glucocorticoids during the full six months of of hospitalization because of COPD (odds ratio, follow-up. 4.6; 95 percent confidence interval, 1.4 to 14.8). In The overall exposure to glucocorticoids among pa- the group of previously hospitalized patients, the tients hospitalized for COPD would be substantially failure rate at six months was 66.7 percent for those decreased if the drug were withheld until it was ev- who received placebo and 49.5 percent for those who ident that other therapy had failed. The disadvantag- received glucocorticoids. es of this option are a delay in the administration of effective therapy to patients with severe dyspnea and DISCUSSION a prolongation of the average hospital stay by slight- We found that the withholding of systemic gluco- ly more than a day. corticoids was not equivalent to active treatment for Recent use of systemic glucocorticoids disqualified hospitalized patients with COPD. Glucocorticoids half the patients screened for this study, and these were marginally superior to placebo in reducing rates patients might have had different responses to glu- of treatment failure at 30 and 90 days, but not at cocorticoids. We designed this study specifically for 6 months. Glucocorticoid therapy also shortened the hospitalized patients, reasoning that the effect of initial hospital stay by an average of 1.2 days. This dif- treatment would be most evident in the sickest pa- ference may be an underestimate, because the proto- tients. However, systemic glucocorticoids are also col required a hospital stay of at least three days and frequently used for outpatient treatment of COPD, because some patients assigned to receive placebo also and the clinical profiles of nonhospitalized patients received open-label glucocorticoids. may be different.

Glucocorticoid-induced improvements in FEV1 We conclude that systemic glucocorticoids decrease provide a plausible basis for the better clinical out- the rate of treatment failure by about 10 percentage comes. The magnitude of the early effect of treatment points for up to 90 days when used for patients hos- on FEV1, approximately 0.10 liter, is similar to that pitalized with exacerbations of COPD. A two-week found in a previous study.3 More patients received regimen was as effective as an eight-week regimen; open-label glucocorticoids as the study progressed, so this result was consistent with those of small trials we may have underestimated the true differences at involving patients with acute asthma.21,22 In addi- later times. tion, subgroup analyses suggest that the treatment Hyperglycemia of sufficient severity to require ther- benefit may be restricted largely to patients who apy was the major complication of glucocorticoids have previously been hospitalized because of COPD. that we identified. This finding may be due in part to the higher proportions of patients with diabetes in Conducted under the aegis of the Cooperative Studies Program of the the glucocorticoid groups than in the placebo group, Department of Veterans Affairs Office of Research and Development and supported by a grant from Boehringer Ingelheim. but hyperglycemia is a known complication of glu- Presented in part at the annual meeting of the American Thoracic Soci- cocorticoid therapy.16,17 We also noted a trend toward ety, Chicago, April 24–29, 1998, and the annual meeting of the European longer hospital stays for causes other than COPD in Respiratory Society, Geneva, September 19–23, 1998. both glucocorticoid groups. Careful review of these APPENDIX data revealed an unusual number of infections re- The other study participants were as follows: Planning Committee: S. quiring hospital readmission in the eight-week glu- Weiss, J. Stoller, I. Tager, M. Antonelli, and M. Buchanan; Data Monitor- cocorticoid group. Controlled trials of treatment for ing Board: D. Dantzker (chair), D. Tashkin, R. Simon, and M. Lebowitz; other diseases have shown an increased risk of seri- Pharmacy coordinator: J. Day; investigators at individual Veterans Affairs medical centers: J. Curtis (principal investigator) and C. Siegert, Ann Ar- ous infection in patients receiving systemic gluco- bor, Mich.; G. Emmanuel (coprincipal investigator), S. Carranza (coprin- corticoids.16,18 cipal investigator), and D. Johnson, Bay Pines, Fla.; P. Romano (coprincipal

1946 · June 24, 1999

Downloaded from www.nejm.org at UNIV OF MEDICINE AND DENTISTRY OF NJ on February 17, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. EFFECT OF SYSTEMIC GLUCOCORTICOIDS ON EXACERBATIONS OF CHRONIC OBSTRUCTIVE PULMONARY DISEASE

investigator), P. Kaul (coprincipal investigator), and R. Varano, Brooklyn, 8. Jones B, Jarvis P, Lewis JA, Ebbutt AF. Trial to assess equivalence: the N.Y.; S. Sethi (principal investigator) and P. DiMarzia, Buffalo, N.Y.; R. importance of rigorous methods. BMJ 1996;313:36-9. [Erratum, BMJ Keller, Hines, Ill.; M. Reinoso (principal investigator) and P. Guillet, Hous- 1996;313:550.] ton; G. Bhaskar (principal investigator) and H. Hermesman, Lake City, 9. Ware JH, Antman EM. Equivalence trials. N Engl J Med 1997;337: Fla.; G. San Pedro (coprincipal investigator) and L. Frazier, Little Rock, 1159-61. Ark.; J. Despars, Long Beach, Calif.; K. Rice (principal investigator) and F. 10. Erbland ML, Deupree RH, Niewoehner DE. Systemic Corticosteroids Lebahn, Minneapolis; A. Fulambarker (principal investigator) and D. Fer- in Chronic Obstructive Pulmonary Disease Exacerbations (SCCOPE): guson, North Chicago, Ill.; P. Krumpe (principal investigator) and R. Wel- rationale and design of an equivalence trial. Control Clin Trials 1998;19: dermuth, Reno, Nev.; J. Liu (principal investigator) and T. Thompson, Sa- 404-17. lem, Va.; M. Habib (principal investigator) and T. Vincent, Tucson, Ariz.; 11. Epidemiology Standardization Project. II. Recommended respiratory S. Santiago (principal investigator), D. Boyd, and L. Robinson, West Los disease questionnaires for use with adults and children in epidemiological Angeles, Calif.; J. Sampson (principal investigator), Alexandria, La.; F. Al- research. Am Rev Respir Dis 1978;118:Suppl:7-53. Bazzaz (principal investigator), Chicago (West Side); M. Nelson (principal 12. Standardization of spirometry — 1987 update: statement of the Amer- investigator), Kansas City, Mo.; J. McCormick (principal investigator) and ican Thoracic Society. Am Rev Respir Dis 1987;136:1285-98. S. Shariaty, Lexington, Ky.; M. Tenholder (principal investigator), Mem- 13. Snedecor GW, Cochran WG. Statistical methods. 8th ed. Ames: Iowa phis, Tenn.; W. Davis (principal investigator) and Z. She, Augusta, Ga.; P. State University Press, 1989. Caralis (principal investigator), Miami; B. Gray (principal investigator) and 14. Durrleman S, Simon R. Planning and monitoring of equivalence stud- K. Laughlin, Oklahoma City; and C. Atwood (principal investigator), ies. Biometrics 1990;46:329-36. Pittsburgh. 15. Cox DR, Oakes D. Analysis of survival data. London: Chapman & Hall, 1984. REFERENCES 16. Conn HO, Poynard T. Corticosteroids and peptic ulcer disease: meta- analysis of adverse events during steroid therapy. J Intern Med 1994;236: 1. Connors AF Jr, Dawson NV, Thomas C, et al. Outcomes following 619-32. acute exacerbation of severe chronic obstructive lung disease. Am J Respir 17. Smyllie HC, Connolly CK. Incidence of serious complications of cor- Crit Care Med 1996;154:959-67. [Erratum, Am J Respir Crit Care Med ticosteroid therapy in respiratory disease: a retrospective survey of patients 1997;155:386.] in the Brompton Hospital. Thorax 1968;23:571-81. 2. Rowe BH, Keller JL, Oxman AD. Effectiveness of steroid therapy in acute 18. Stuck AE, Minder CE, Frey FJ. Risk of infectious complications in pa- exacerbations of asthma: a meta-analysis. Am J Emerg Med 1992;10:301-10. tients taking glucocorticosteroids. Rev Infect Dis 1989;11:954-63. 3. Albert RK, Martin TR, Lewis SW. Controlled clinical trial of methyl- 19. Laan RFJM, van Riel PLCM, van de Putte LBA, van Erning LJTO, prednisolone in patients with chronic bronchitis and acute respiratory in- van’t Hof MA, Lemmens JAM. Low-dose prednisone induces rapid revers- sufficiency. Ann Intern Med 1980;92:753-8. ible axial bone loss in patients with rheumatoid arthritis: a randomized, 4. Thompson WH, Nielson CP, Carvalho P, Charan NB, Crowley JJ. Con- controlled study. Ann Intern Med 1993;119:963-8. trolled trial of oral prednisone in outpatients with acute COPD exacerba- 20. McEvoy CE, Ensrud KE, Bender E, et al. Association between corti- tion. Am J Respir Crit Care Med 1996;154:407-12. costeroid use and vertebral fractures in older men with chronic obstructive 5. Emerman CL, Connors AF, Lukens TW, May ME, Effron D. A ran- pulmonary disease. Am J Respir Crit Care Med 1998;157:704-9. domized controlled trial of methylprednisolone in the emergency treat- 21. Lederle FA, Pluhar RE, Joseph AM, Niewoehner DE. Tapering of ment of acute exacerbations of COPD. Chest 1989;95:563-7. corticosteroid therapy following exacerbation of asthma: a randomized, 6. Callahan CM, Dittus RS, Katz BP. Oral corticosteroid therapy for pa- double-blind, placebo-controlled trial. Arch Intern Med 1987;147:2201- tients with stable chronic obstructive pulmonary disease: a meta-analysis. 3. Ann Intern Med 1991;114:216-23. 22. O’Driscoll BR, Kalra S, Wilson M, Pickering CAC, Carroll KB, 7. McEvoy CE, Niewoehner DE. Adverse effects of corticosteroid therapy Woodcock AA. Double-blind trial of steroid tapering in acute asthma. for COPD: a critical review. Chest 1997;111:732-43. Lancet 1993;341:324-7.

Volume 340 Number 25 · 1947