Radial Artery Applanation Tonometry

Radial Artery Applanation Tonometry for Continuous Noninvasive Cardiac Output Measurement: A Comparison With Intermittent Pulmonary Artery Thermodilution in Patients After Cardiothoracic Surgery

Julia Y. Wagner, MD1; Harun Sarwari1; Gerhard Schön, MSc2; Mathias Kubik, MD3,4; Stefan Kluge, MD3; Herrmann Reichenspurner, MD4; Daniel A. Reuter, MD1; Bernd Saugel, MD1

Objectives: Radial artery applanation tonometry allows completely a pulmonary artery catheter (PAC-CO) with regard to accuracy, noninvasive continuous cardiac output estimation. The aim of the precision of agreement, and trending ability. present study was to compare cardiac output measurements Design: A prospective method comparison study. obtained with applanation tonometry (AT-CO) using the T-Line Setting: The study was conducted in a cardiosurgical ICU of a system (Tensys Medical, San Diego, CA) with cardiac output German university hospital. measured by intermittent pulmonary artery thermodilution using Patients: We performed cardiac output measurements in 50 patients after cardiothoracic surgery. Interventions: None. 1Department of Anesthesiology, Center of Anesthesiology and Intensive Care Medicine, University Medical Center Hamburg-Eppendorf, Ham- Measurements and Main Results: Three independent sets of burg, Germany. three consecutive thermodilution measurements (i.e., PAC- 2Department of Medical Biometry and Epidemiology, University Medical CO) each were performed per patient, and AT-CO was mea- Center Hamburg-Eppendorf, Hamburg, Germany. sured simultaneously. The average of the three thermodilution 3 Department of Intensive Care Medicine, University Medical Center Ham- cardiac output measurements was compared with the aver- burg-Eppendorf, Hamburg, Germany. age of the corresponding three AT-CO values resulting in 150 4Department of Cardiovascular Surgery, University Heart Center Ham- burg, Hamburg, Germany. paired cardiac output measurements. In 13 patients, cardiac Dr. Wagner and Sarwari contributed equally to this work. output–modifying maneuvers performed for clinical reasons Tensys Medical (San Diego, CA) provided the technical equipment for additionally allowed to evaluate trending ability. For statistical the study. analysis, we used Bland-Altman analysis, the percentage error, Dr. Wagner received grant support from Tensys Medical, San Diego, four-quadrant plot, and concordance analysis. Mean PAC-CO CA, and received support for travel from CNSystems Medizintechnik was 4.7 ± 1.2 L/min and mean AT-CO was 4.9 ± 1.1 L/min. The AG, Graz, Austria (refunds of travel expenses), and from Tensys Medi- cal, San Diego, CA (refunds of travel expenses). Dr. Kluge served as a mean of differences was –0.2 L/min with 95% limits of agree- board member for Astellas, Gambro, Gilead, MSD, NovaLung, Novartis, ment of –1.8 to + 1.4 L/min. The percentage error was 34%. and Pfizer and lectured for Astellas, Biotest, Gambro, Gilead, MSD, The concordance rate was 95%. Novalung, Pfizer, Sedana, and Thermo Fisher Scientific. Dr. Reuter served as a board member for Pulsion Medical Systems (Medical Advi- Conclusions: Continuous cardiac output measurement using the sory Board); consulted for Masimo Corp; and lectured for Edwards, noninvasive applanation tonometry technology is basically fea- Fresenius, and B. Braun. Dr. Saugel received support for travel from sible in ICU patients after cardiothoracic surgery. The applanation Tensys Medical (San Diego, CA) (refunds of travel expenses), served as a board member, lectured, consulted for Pulsion Medical Systems SE tonometry technology provides cardiac output values with reason- (Feldkirchen, Germany) (member of the Medical Advisory Board); and able accuracy and precision of agreement compared with inter- received support for travel from CNSystems Medizintechnik AG (Graz, mittent pulmonary artery thermodilution measurements in a clinical Austria) (refunds of travel expenses); and received grant support from Tensys Medical (San Diego, CA) (unrestricted research grants). The study setting and is able to reliably track cardiac output changes remaining authors have disclosed that they do not have any potential induced by cardiac output–modifying maneuvers. (Crit Care Med conflicts of interest. 2015; XX:00–00) For information regarding this article, E-mail: [email protected] Key Words: cardiac output; critical care; hemodynamic monitoring; Copyright © 2015 by the Society of Critical Care Medicine and Wolters pulmonary artery catheter; T-Line Kluwer Health, Inc. All Rights Reserved. DOI: 10.1097/CCM.0000000000000979

ardiac output (CO) is an essential variable when it operating room to the cardiosurgical ICU. Independent of comes to adequately monitoring and optimizing a the study, all medical decisions regarding analgosedation and Cpatient’s hemodynamics in the ICU and during inter- hemodynamic management during the postoperative period mediate-risk and high-risk surgical procedures (1, 2). were at the discretion of the treating ICU physician. This In the 1970s, the pulmonary artery catheter (PAC) devel- included the need for IV opioids for analgesia and propofol oped by Harold Jeremy Swan and William Ganz became avail- for sedation and the timing of extubation. The patient’s hemo- able for the invasive CO measurement by thermodilution (3, dynamic state and fluid responsiveness were evaluated (e.g., 4). Despite its decline approximately 30 years later triggered by using passive leg raising test) and optimized with IV fluids and studies demonstrating an unfavorable risk-benefit profile (5, catecholamines. Each diagnostic and therapeutic intervention 6), the PAC still proves to be of outstanding importance in dis- was based on the ICU physicians’ medical decisions indepen- tinct clinical situations (7) and is still considered as the clinical dent from the study measurements. criterion standard method for measuring CO (8). After arrival in the ICU, patients were connected to the AT During the years after the decrease in routine PAC use, a num- study device (T-Line 400 monitor). A bracelet was placed on ber of alternatives for invasive, less-invasive, and noninvasive CO the patient’s wrist equipped with an electromechanical system measurements have been introduced and evaluated (9–14). and a sensor that moves over the radial artery in order to find One of the novel technologies that provides continuous the optimal applanation pressure as described before (21, 22). completely noninvasive CO monitoring is the radial artery When the optimal signal is available, the T-Line monitor pro- applanation tonometry (AT). One commercially available vides a continuous arterial waveform. In order to compensate device using the AT technology for the assessment of hemo- for height differences between the study limb and the patient’s dynamic variables is the T-Line system (Tensys Medical, San heart level, the level function of the T-Line system was applied. Diego, CA). The AT technology’s ability to measure blood pres- By using the level function one levels the T-Line electronically sure continuously and noninvasively has already been demon- by entering on the T-Line monitor’s touchscreen the vertical strated (15–20). Furthermore, our group recently described height of the study limb above or below the patient’s heart level. and evaluated an algorithm for CO determination based on AT After entering the patient’s biometric data (height, weight, in a proof-of-concept analysis using impeccable arterial wave- age, and gender), the AT system’s monitor displays CO values forms of ICU patients selected from a database (12). derived from pulse contour analysis using a proprietary auto- In the present study, we investigate the accuracy, precision calibrating algorithm. of agreement, and trending ability of the AT technology for The study measurements were performed at three differ- the measurement of CO in comparison with pulmonary artery ent time points during the postoperative period. The intervals thermodilution as the clinical criterion standard method in between the time points differed depending on the clini- postoperative cardiothoracic surgery patients. cal routine events. At each time point, one pulmonary artery thermodilution measurement using the PAC consisted of five injections of 10-mL ice-cold sodium chloride. The five cor- MATERIALS AND METHODS responding AT CO measurements (AT-CO) were recorded Patients, Inclusion/Exclusion Criteria, and Study simultaneously. In order to avoid an impact due to discrep- Measurements ancies between the single pulmonary artery thermodilution Our study was conducted in a cardiosurgical ICU of a Ger- measurements, the average of the three of five closest thermo- man university hospital (University Medical Center Ham- dilution CO measurements (PAC-CO) was used for compari- burg-Eppendorf, Hamburg, Germany). The ethics committee son with the average of the corresponding three AT-CO values. (Ethikkommission der Ärztekammer Hamburg) approved the Immediately after the patient monitor displayed the thermo- study. We obtained written informed consent from all patients dilution-derived CO value, the corresponding CO value mea- prior to study enrollment. sured with AT was recorded using screen shots. The study included patients aged at least 18 years, who In a subgroup of the study patients, the ICU physician in underwent nonemergency cardiothoracic surgery, and had charge decided that a passive leg raising maneuver or a volume a PAC inserted for medical reasons according to the standard challenge was indicated in order to assess volume responsive- clinical protocol of the Department of Anesthesiology unrelated ness (23, 24) or another CO-influencing clinical intervention to the present study. Patients were not eligible for study inclu- took place (propofol bolus, a change in the settings of the sion if AT study measurements at the patient’s wrist were not patient’s external pacemaker) during our study measurements. possible because of anatomical abnormalities of the study limb. In these cases, we performed one set of study measurements The PAC (INTRATHERMODIN 4lumen [7F, 110 cm]; intra- prior to the clinical intervention (time point 1 or 2) and one special catheters GmbH, Rehlingen-Siersburg, Germany) was shortly after the intervention (time point 2 or 3). inserted preoperatively after routine anesthetic induction, and all PAC-obtained values were displayed on the vital signs moni- Statistical Analysis tor (Infinity Delta; Dräger AG, Lübeck, Germany). For the patient characteristics, we calculated either absolute The study measurements were performed within the and relative numbers (percentages) or the median with the first postoperative hours after the patient’s transfer from the interquartile range (i.e., 25–75% percentile range).

For PAC-CO and AT-CO, the mean CO values ± sd were Table 1. Patient and Clinical Characteristics separately calculated. We computed the mean of the differ- Variable Data for 50 Patients ences (= bias) between PAC-CO and AT-CO, the sd, and the 95% limits of agreement (= bias ± 1.96 × sd). The Bland- Age, yr 74 (66–78) Altman method accounting for repeated measurements was performed (25). Furthermore, we computed the percentage Sex, male, n (%) 38 (76) error as described by Critchley and Critchley (26). Height, m 1.75 (1.66–1.76) For the assessment of the AT technology’s trending ability, we Weight, kg 84 (74–94) created a four-quadrant plot and performed concordance analy- 2 sis. Furthermore, we calculated a variance components model Body mass index, kg/m 28.0 (25.2–32.1) (mixed model) in order to know the amount of variation that Type of surgery can be explained by the different sources of variation: 1) varia- Off-pump coronary artery bypass, 48 (96) tion due to patient, 2) variation due to different measurement n (%) time points, and 3) residual variation (i.e., variation due to the Off-pump coronary artery bypass plus 1 (2) three single measurement values used to calculate the mean CO carotid artery surgery, n (%) value at each of the three measurement time points per patient). All statistical analyses were performed using Excel Lung transplantation, n (%) 1 (2) (Microsoft, Redmond, WA), IBM SPSS Statistics 21 (SPSS, Clinical characteristics on the day of Chicago, IL), and the statistical software package R Version study inclusion 3.1.2 (R Core Team [2014]; R: A language and environment for Simplified Acute Physiology Score, 32 (26–35) statistical computing; R Foundation for Statistical Computing, points Vienna, Austria; URL: http://www.R-project.org). Therapeutic Intervention Scoring 22 (18–22) System, points RESULTS Mechanical ventilation, n (%) 46 (92) Patients Noradrenaline therapy, n (%) 21 (42) We enrolled 60 patients in the study. In seven patients, CO Noradrenaline dose, μg/min 4 (2–6) measurements in the postoperative period were not possible due to technical problems with the PAC (three patients) or Mean arterial pressure, mm Hg 74 (66–79) AT (four patients). Furthermore, we secondarily excluded Systolic arterial pressure, mm Hg 116 (102–123) three patients without sinus rhythm from the study. Finally, 50 Diastolic arterial pressure, mm Hg 54 (49–61) patients were available for the statistical analysis. Fourty-eight patients had undergone off-pump coronary artery bypass sur- Central venous pressure, mm Hg 9 (8–11) gery, one patient additionally received carotid artery surgery, Mean pulmonary artery pressure, 24 (19–28) and one patient underwent lung transplantation. mm Hg The patients’ demographic and clinical characteristics are Systolic pulmonary artery pressure, 33 (29–37) presented in Table 1. mm Hg Diastolic pulmonary artery pressure, 16 (12–22) CO Measurements mm Hg A total of 150 matched CO data points were available for the Pulmonary artery occlusion pressure, 13 (12–17) final statistical analysis. mm Hg Thirteen of the 50 study patients were available for evaluating the AT technology’s ability to track changes in CO because Systemic vascular resistance, 1,073 (838–1,400) dyn·s/cm5 a CO-influencing clinical intervention took place during our Data are presented as the median and interquartile ranges (25–75% study measurements, and the CO values measured by the percentile) or as absolute frequencies with percentages. criterion standard method (PAC) subsequently increased or decreased. In eight patients, a passive leg raising maneuver was performed, three patients received a propofol bolus injection By calculating the variance components model (mixed during the period of our study measurements, one patient model) in order to know the amount of variation that can be received a fluid challenge (500 mL fluid bolus), and in one explained by the different sources of variation, we received patient the settings of the external pacer were changed. the following results for the variable PAC-CO: 1.27 for the The mean PAC-CO was 4.7 ± 1.2 L/min and the mean variation due to patient, 0.08 for the variation due to different AT-CO was 4.9 ± 1.1 L/min. The Bland-Altman analysis measurement time points, and 0.09 for residual variation. The revealed a bias of –0.2 L/min with an sd of ± 0.8 L/min and corresponding percentage distribution was 88%, 6%, and 6%, upper and lower limits of agreement of –1.8 and + 1.4 L/min, respectively. Thus, approximately 88% of the total variance respectively (Fig. 1). The percentage error was 34%. of the variable PAC-CO can be attributed to the individual

Figure 2. The ability of the radial artery applanation tonometry to track Figure 1. Bland-Altman plot comparing cardiac output (CO) changes in cardiac output (CO) is shown using the four-quadrant plot. The measurements obtained noninvasively (radial artery applanation concordance analysis resulted in a concordance of 95%. An exclusion tonometry–derived CO [AT-CO]) and invasively (pulmonary artery zone of 0.5 L/min was applied. AT-CO = radial artery applanation catheter–derived CO [PAC-CO]) in 50 patients. The continuous horizontal tonometry–derived CO, PAC-CO = pulmonary artery catheter–derived CO. line shows the mean of the differences (= bias) between the two methods, and the dashed horizontal lines show the upper and lower 95% limits of agreement (= bias ± 1.96 × sd). The percentage error was 34%. The percentage error of 34% was slightly higher than the generally accepted cutoff value for acceptable precision of agreement of 28.3% (frequently rounded up to 30%) proposed patient, 6% to the measurement time point, and around 6% by Critchley and Critchley (26). However, as nicely demon- are variation of the three single thermodilution measurements strated by Peyton and Chong (33), it is questionable to uncon- used to calculate the average CO value for each measurement ditionally and generally apply this cutoff value (that requires time point. For the variable AT-CO, the results were as follows: a precision of method for both the reference technology and 1.03 for the variation due to patient, 0.16 for the variation due the studied technology of at least 20%) to clinical CO method to different measurement time points, and 0.10 for residual comparison studies. In this context, definite recommendations variation. The corresponding percentage distribution was for the definition of clinical acceptable agreement between 80%, 13%, and 8%, respectively. two CO monitoring technologies are still a matter of debate The ability of the AT technology to track changes in CO (29). Peyton and Chong (33) demonstrated that less-invasive is shown in the four-quadrant plot (Fig. 2). The concordance technologies—including pulse contour analysis based on an analysis resulted in a concordance rate of 95%, indicating invasively assessed arterial pressure wave signal—usually fail acceptable trending ability (27). to fulfill the percentage error threshold of 30%. What they are proposing instead is a percentage error in agreement with ther- DISCUSSION modilution of ± 45% as a more realistic achievable precision of Our study shows for the first time that the AT technology mea- agreement in a clinical setting (33). sures CO with reasonable accuracy and precision of agreement In order to evaluate a novel noninvasive CO measurement compared with intermittent pulmonary artery thermodilution technology’s accuracy and precision of agreement, the com- in postoperative cardiothoracic surgery patients in a clinical parison with an invasive clinical standard criterion method is study setting. Assessing the AT technology’s performance to necessary (34). However, with regard to an adequate clinical track changes in CO, we observed a concordance rate of 95% risk-benefit ratio, a PAC, for example, is reserved to high-risk with the PAC-derived CO values, indicating an acceptable surgical or critically ill patients. As a consequence, a novel non- trending ability according to Critchley et al (27). invasive CO measurement technology like the AT must be eval- Considering that there is a clear trend toward less-invasive uated in a patient population that often does not represent its and noninvasive hemodynamic monitoring (28, 29) and that ideal target patient group (34). However, the goal is not neces- continuous monitoring has proven to be superior to intermit- sarily to replace invasive CO monitoring by using noninvasive tent monitoring in different groups of patients (30–32), the CO monitoring devices (e.g., the AT technology). Instead, the findings of our study are promising. objective should be to find clinical settings in which patients

4 www.ccmjournal.org XXX 2015 • Volume XX • Number XXX Copyright © 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved. Clinical Investigation might benefit from continuous noninvasive CO monitoring CO measurement method like the pulmonary artery ther- (e.g., intermediate-risk surgery, high-dependency unit) (34). modilution used in our study implies a higher possibility of Conversely, in clinical situations in which an invasive catheter’s a delay in displaying such CO changes. Furthermore, it has removal is indicated, but continuous CO monitoring would been argued that in certain clinical scenarios, the tracking of still be valuable, a noninvasive technology might provide a rea- CO changes might even be of more relevance than the CO sonable alternative. So when thinking about the clinical use of monitor’s ability to provide CO values with high accuracy and a noninvasive CO technology, it is important to differentiate precision of agreement (37). Of course, in critically ill patients between the patient groups studied in clinical validation stud- treated in an ICU and during high-risk surgical procedures, ies depending on the presence of an invasive catheter and the artery cannulation is usually performed not only for continu- real final target patient group. ous blood pressure measurements but also for arterial blood Another completely noninvasive method for measuring CO gas analyses. Furthermore, the presence of an arterial catheter is the volume clamp method (35) using an inflatable finger cuff allows minimally invasive radial artery waveform-based pulse (36). Validation studies evaluating this technology in compari- contour analysis (38, 39). However, these technologies still son with pulmonary artery or transpulmonary thermodilution bear the arterial catheter–associated risks, and their measure- revealed similar results compared with the AT technology’s CO ment performance is controversially discussed (10, 40). Such measurement performance in our study (13, 36). considerations again show clearly how important it is to select During study measurements, we noticed two limitations of appropriate application areas for a noninvasive continuous the AT technology that need to be mentioned. First, it takes a CO measurement technology. certain time, up to several minutes, from the application of the device until the sensor has detected the optimal applanation CONCLUSIONS pressure and displays an arterial waveform and thus CO values. Our study demonstrates that CO measurements using the Nonetheless, bearing in mind that the placement of an inva- noninvasive AT technology are basically feasible in postopera- sive arterial catheter might be challenging in certain patient tive cardiothoracic surgery patients in an ICU setting. The AT groups and therefore time consuming in daily clinical practice technology provides CO values with reasonable accuracy and as well, this can be seen as a relative disadvantage of the nonin- precision of agreement compared with intermittent pulmo- vasive AT technology. Second, the AT technology is very sensi- nary artery thermodilution measurements in a clinical study tive to movement artifacts. For example, slight movements of a setting and is able to reliably track CO changes induced by CO- patient during the postoperative recovery period or a medical modifying maneuvers. professional bumping against the study limb lead to a period of “motion recovery” during which CO values are not avail- REFERENCES able. This limitation emphasizes the current importance of 1. Gurgel ST, do Nascimento P Jr: Maintaining tissue perfusion in high- adequately choosing the optimal target patient group and also risk surgical patients: A systematic review of randomized clinical trials. for the further improvement of the technology’s measurement Anesth Analg 2011; 112:1384–1391 performance, for example, by reducing the bracelet size (34). 2. Hamilton MA, Cecconi M, Rhodes A: A systematic review and meta- analysis on the use of preemptive hemodynamic intervention to Our proof-of-concept analysis evaluating the AT technol- improve postoperative outcomes in moderate and high-risk surgical ogy’s CO algorithm for the first time (12) revealed a higher patients. Anesth Analg 2011; 112:1392–1402 accuracy and precision of agreement with a low percentage 3. Ganz W, Donoso R, Marcus HS, et al: A new technique for measurement of cardiac output by thermodilution in man. Am J Cardiol 1971; error of 23% compared with the results of the present study. 27:392–396 This finding is explained by the analysis of only selected 4. Swan HJ, Ganz W, Forrester J, et al: Catheterization of the heart in impeccable waveforms of an ICU patient database in con- man with use of a flow-directed balloon-tipped catheter. N Engl J Med trast to CO measurements obtained during clinical routine in 1970; 283:447–451 5. Connors AF Jr, Speroff T, Dawson NV, et al: The effectiveness of right an unselected group of postoperative cardiothoracic surgery heart catheterization in the initial care of critically ill patients. SUP- patients. In this prospective study, we were now able to dem- PORT Investigators. JAMA 1996; 276:889–897 onstrate the feasibility of the AT technology’s CO measure- 6. Sandham JD, Hull RD, Brant RF, et al; Canadian Critical Care Clinical ments in a clinical setting. In addition, we were also able to Trials Group: A randomized, controlled trial of the use of pulmonary- artery catheters in high-risk surgical patients. N Engl J Med 2003; show the AT technology’s ability to reliably track changes in 348:5–14 CO by performing CO-modifying maneuvers. As one limi- 7. Richard C, Monnet X, Teboul JL: Pulmonary artery catheter monitoring tation of our study, we have to mention that CO-modifying in 2011. Curr Opin Crit Care 2011; 17:296–302 interventions in order to evaluate the AT technology’s trend- 8. Rajaram SS, Desai NK, Kalra A, et al: Pulmonary artery catheters for adult patients in intensive care. Cochrane Database Syst Rev 2013; ing ability were performed in only a limited number of the 2:CD003408 study patients and thus in a probably heterogeneous patient 9. Funk DJ, Moretti EW, Gan TJ: Minimally invasive cardiac output moni- group. Nevertheless, the fact that the AT technology reliably toring in the perioperative setting. Anesth Analg 2009; 108:887–897 tracked changes in CO is a very important finding because 10. Marik PE: Noninvasive cardiac output monitors: A state-of the-art it might be a particular benefit of a continuous noninvasive review. J Cardiothorac Vasc Anesth 2013; 27:121–134 11. Appel PL, Kram HB, Mackabee J, et al: Comparison of measurements CO measurement technology to display changes in CO to the of cardiac output by bioimpedance and thermodilution in severely ill treating physician at the earliest possible stage. An intermittent surgical patients. Crit Care Med 1986; 14:933–935

12. Saugel B, Meidert AS, Langwieser N, et al: An autocalibrating algo- 26. Critchley LA, Critchley JA: A meta-analysis of studies using bias and rithm for non-invasive cardiac output determination based on the precision statistics to compare cardiac output measurement tech- analysis of an arterial pressure waveform recorded with radial artery niques. J Clin Monit Comput 1999; 15:85–91 applanation tonometry: A proof of concept pilot analysis. J Clin Monit 27. Critchley LA, Lee A, Ho AM: A critical review of the ability of con- Comput 2014; 28:357–362 tinuous cardiac output monitors to measure trends in cardiac output. 13. Bubenek-Turconi SI, Craciun M, Miclea I, et al: Noninvasive continu- Anesth Analg 2010; 111:1180–1192 ous cardiac output by the Nexfin before and after preload-modifying 28. Vincent JL, Rhodes A, Perel A, et al: Clinical review: Update on hemo- maneuvers: A comparison with intermittent thermodilution cardiac dynamic monitoring—A consensus of 16. Crit Care 2011; 15:229 output. Anesth Analg 2013; 117:366–372 29. Saugel B, Reuter DA: Are we ready for the age of non-invasive hae- 14. de Waal EE, Wappler F, Buhre WF: Cardiac output monitoring. Curr modynamic monitoring? Br J Anaesth 2014; 113:340–343 Opin Anaesthesiol 2009; 22:71–77 30. Nowak RM, Sen A, Garcia AJ, et al: Noninvasive continuous or inter- 15. Janelle GM, Gravenstein N: An accuracy evaluation of the T-Line mittent blood pressure and heart rate patient monitoring in the ED. Tensymeter (continuous noninvasive blood pressure management Am J Emerg Med 2011; 29:782–789 device) versus conventional invasive radial artery monitoring in surgi- 31. Wagner JY, Prantner JS, Meidert AS, et al: Noninvasive continuous cal patients. Anesth Analg 2006; 102:484–490 versus intermittent arterial pressure monitoring: Evaluation of the vas- 16. Szmuk P, Pivalizza E, Warters RD, et al: An evaluation of the T-Line cular unloading technique (CNAP device) in the emergency depart- Tensymeter continuous noninvasive blood pressure device during ment. Scand J Trauma Resusc Emerg Med 2014; 22:8 induced hypotension. Anaesthesia 2008; 63:307–312 32. Benes J, Simanova A, Tovarnicka T, et al: Continuous non-invasive 17. Saugel B, Fassio F, Hapfelmeier A, et al: The T-Line TL-200 system monitoring improves blood pressure stability in upright position: Ran- for continuous non-invasive blood pressure measurement in medical domized controlled trial. J Clin Monit Comput 2015; 29:11–17 intensive care unit patients. Intensive Care Med 2012; 38:1471–1477 33. Peyton PJ, Chong SW: Minimally invasive measurement of cardiac 18. Meidert AS, Huber W, Hapfelmeier A, et al: Evaluation of the radial output during surgery and critical care: A meta-analysis of accuracy artery applanation tonometry technology for continuous noninvasive and precision. Anesthesiology 2010; 113:1220–1235 blood pressure monitoring compared with central aortic blood pres- 34. Wagner JY, Saugel B: When should we adopt continuous noninvasive sure measurements in patients with multiple organ dysfunction syn- hemodynamic monitoring technologies into clinical routine? J Clin drome. J Crit Care 2013; 28:908–912 Monit Comput 2015; 29:1–3 19. Saugel B, Meidert AS, Hapfelmeier A, et al: Non-invasive continuous 35. Penáz J, Voigt A, Teichmann W: Contribution to the continuous arterial pressure measurement based on radial artery tonometry in the indirect blood pressure measurement. Z Gesamte Inn Med 1976; intensive care unit: A method comparison study using the T-Line TL- 31:1030–1033 200pro device. Br J Anaesth 2013; 111:185–190 36. Bogert LW, Wesseling KH, Schraa O, et al: Pulse contour cardiac 20. Meidert AS, Huber W, Müller JN, et al: Radial artery applanation output derived from non-invasive arterial pressure in cardiovascular tonometry for continuous non-invasive arterial pressure monitoring disease. Anaesthesia 2010; 65:1119–1125 in intensive care unit patients: Comparison with invasively assessed 37. Bein B, Renner J, Scholz J, et al: Comparing different methods of car- radial arterial pressure. Br J Anaesth 2014; 112:521–528 diac output determination: A call for consensus. Eur J Anaesthesiol 21. Dueck R, Goedje O, Clopton P: Noninvasive continuous beat-to-beat 2006; 23:710 radial artery pressure via TL-200 applanation tonometry. J Clin Monit 38. Smetkin AA, Hussain A, Kuzkov VV, et al: Validation of cardiac output Comput 2012; 26:75–83 monitoring based on uncalibrated pulse contour analysis vs transpul- 22. Saugel B, Dueck R, Wagner JY: Measurement of blood pressure. monary thermodilution during off-pump coronary artery bypass graft- Best Pract Res Clin Anaesthesiol 2014; 28:309–322 ing. Br J Anaesth 2014; 112:1024–1031 23. Monnet X, Rienzo M, Osman D, et al: Passive leg raising predicts fluid 39. Pratt B, Roteliuk L, Hatib F, et al: Calculating arterial pressure-based responsiveness in the critically ill. Crit Care Med 2006; 34:1402–1407 cardiac output using a novel measurement and analysis method. 24. Cecconi M, Parsons AK, Rhodes A: What is a fluid challenge? Curr Biomed Instrum Technol 2007; 41:403–411 Opin Crit Care 2011; 17:290–295 40. Monnet X, Vaquer S, Anguel N, et al: Comparison of pulse contour 25. Bland JM, Altman DG: Agreement between methods of measure- analysis by Pulsioflex and Vigileo to measure and track changes ment with multiple observations per individual. J Biopharm Stat 2007; of cardiac output in critically ill patients. Br J Anaesth 2014; 17:571–582 114:235–243