Parameters measured before (T1), during (T2), and after (T3) the ARM are reported in Table 3. CO values were normally distributed. As shown, TDco dropped by 31% (95% CI, 26–35; P < 0.001) between T1 and T2 and increased by 51% (95% CI, 42–63; P < 0.001) between T2 and T3. ESco dropped by 6% (95% CI, 2–11; P = 0.048) between T1 and T2 and increased by 8% (95% CI, 3–14; P = 0.016) between T2 and T3. ECco dropped by 16% (95% CI, 11–22; P < 0.0001) between T1 and T2 and increased by 20% (95% CI, 10–31; P < 0.0001) between T2 and T3. MAP dropped by 13% (95% CI, 6–21; P = 0.006) between T1 and T2 and increased by 21% (95% CI, 10–32; P = 0.002).
We analyzed means of 3 CO measurements at each time, resulting in 81 ESco and 77 ECco values being collected. Bland-Altman plots are displayed in Figure 1. Bias between both tested methods and TDco were normally distributed. Between esCCO and TDco, the mean bias was +0.7 L/min with limits of agreement of −2.1 L/min and +3.5 L/min. Between ECOM and TDco, the mean bias was +0.2 L/min with limits of agreement of −2.0 L/min and +2.4 L/min.
Fifty-four pairs of ΔESco/ΔTDco as well as 48 pairs of ΔECco/ΔTDco were collected. After exclusion of any values with mean ΔCO <0.75 L/min, 32 pairs of ΔESco/ΔTDco data and 33 pairs of ΔECco/ΔTDco data were ultimately analyzed. Results are presented in Figure 2. The CCC between ΔMAP and ΔTDco was 0.69 (95% CI, 0.54–0.80) with a Pearson correlation of 0.76 and a bias correction factor of 0.91. The CCC between ΔECco and ΔTDco was significantly superior to the CCC between ΔESco and ΔTDco (P = 0.0065) and was equivalent to the CCC between ΔMAP and ΔTDco (P = 0.16). The 4-quadrant plot analysis showed a concordance rate of 81% (95% CI, 74–88) between ΔESco and ΔTDco and 100% between ΔECco and ΔTDco.
Fifty-four pairs of ΔESco/ΔTDco and 48 pairs of ΔECco/ΔTDco were analyzed after exclusion of any values with mean ΔCO <0.5 L/min. Results are presented in Figure 3. The polar plot analysis for ΔESco showed a bias of −29° with radial limits of agreement of −67° to +8° and a polar concordance rate of 41% (95% CI, 34–48). The polar plot analysis for ΔECco showed a bias of −15° with radial limits of agreement of –44° to + 14° and a polar concordance rate of 85% (95% CI, 79–90).
This study is the first to evaluate the accuracy and tracking abilities of esCCO and ECOM when compared to TD using PAC during an ARM in postoperative cardiac surgery patients. Our results demonstrate that both monitors have a consistent negative polar bias greater than the ±5° suggested.29 In other words, they both underestimate the successive decrease and increase during an ARM when compared to TDco. The large polar limits of agreement values observed with esCCO (–67° to +8°) are notably worse than the acceptable polar limits of agreement values of ECOM (within the ±30° of angular bias). This result is further exemplified by: (1) the significantly larger CCC of ΔECco compared to ΔESco and (2) an ECOM concordance rate acceptable for the polar plot (85%) and the 4-quadrant plot (100%). However, in this particular setting with no significant variation of SVR, absence of CCC differences between ECOM and MAP indicates that MAP may be as accurate as ECOM for tracking changes in CO. It is also noteworthy that neither device is clinically interchangeable with TD due to their own large clinical limits of agreement.
The intervention (ARM) had a significant impact on TDco: the lower limit of the 95% CI for all ΔTDco was 26%, a value significantly different from the LSC of TDco (7%).21 Despite its known precision error,37,38 we chose PAC bolus TD due to its ability to track CO.38 Significant changes in CO allowed us to include a high proportion of paired of ΔCO values to calculate the concordance rate. Indeed, a large central zone for the 4-quadrant plot analysis and opposite paired of ΔCO values for the polar plot analysis are excluded for the concordance rate, therefore decreasing the power of the analysis. Interestingly, the 4-quadrant plot view showed a concordance rate of 81% for esCCO despite a low CCC (ρc = 0.42) and a poor polar concordance rate (41%). The large difference in the concordance rate can be explained partly by the wider acceptance of ΔCO values for the 4-quadrant plot analysis. From an angular description, as many plots range between 30° and 45° (circle plots near the horizontal line, Fig. 2) from the identity line that is a diagonal, these plots are accepted in the 4-quadrant plot analysis and rejected in the polar plot analysis (wherein identity line is horizontal with predefined limits of agreement of 30°). Therefore, the threshold ratio used to calculate the concordance rate markedly differs. Therefore, we also chose to use CCC which can explore both the precision and accuracy between the tested devices and the reference method. A good agreement between 2 methods requires a CCC >0.9.30 Hence, the combination of 3 “concordance” statistical analyses highlights the potential limits of one and demonstrates the need for additional 2 analyses. Of note, the CCCs of ΔMAP versus ΔTDco and ΔECco versus ΔTDco were not statistically different, questioning the idea that either of the CO monitors evaluated in this particular setting are more clinically useful that arterial blood pressure monitoring for following changes in CO.
The negative angular bias for both methods is intriguing, but has been previously shown with esCCO8,9 and is due to either underestimation by the 2 devices or overestimating by TDco of the true changes in CO during an ARM. As TDco measures the right-sided CO, it may be possible that right-sided CO is transiently more sensitive to acute preload decreases and acute afterload increases (such as occurs during an ARM) when compared to a measure of left-sided CO.39 The 2-minute period of stabilization after each PEEP modification may have not lasted long enough to allow the equilibration of left- and right-sided CO.36 However, the precision of each technique was below 20%, therefore acceptable (Table 2), and not different regardless of the time plot, demonstrating that variation of CO at each time plot was not significant. Another factor affecting measurements is that PEEP is known to occasionally cause significant tricuspid regurgitation.14 This is due to an increased portion of right outflow returning to the right atrium, in which case right-sided TD underestimates pulmonary valve outflow.40 Interestingly, this physiological condition should have caused a positive angular bias for esCCO and ECOM measurements. A smart approach to elucidate these potential limitations would have been to add a left-sided CO measurement, such as echocardiography, which could have measured the left outflow tract and also evaluated tricuspid regurgitation. When questions of this quantity and diversity arise, validating tested devices with a true gold standard CO measurement like transit time ultrasonic flowprobes29,41 in animal studies or cardiac surgery (where great vessels are accessible) should be considered.
These results agree with the findings of previous comparisons between these devices and TD that found large limits of agreement depicted by percent errors between 44% and 80% for esCCO8,42–44 and between 40% and 55% for ECOM.45–49 The trending ability of these techniques have been studied much less, with one previous study14 showing a concordance rate of 96% for esCCO with an exclusion zone of <0.5 L/min (although it has been suggested to be <0.75 L/min).31,34 Another difference between that study and ours is that they used an invasive calibration method, which may less clinically relevant, as this monitor is designed to be used in a completely noninvasive manner. Additionally, one of the authors had significant conflicts of interest.9 In addition, limited other studies report similar findings. For example, after cardiac surgery Ishihara et al.50 found an acceptable angular bias of −1.6° for esCCO using an invasive calibration method (TDco). They, however, found wide radial limits of agreement (±53°), similar to our study (±37°), indicating the poor trending ability of the esCCO system. Regarding the ECOM system, 5 studies have examined the trending availability.46–49,51 Our results contrast with Fellahi et al.47 and Møller-Sørensen,48 which both demonstrated a poor 4-quadrant trending ability (recalculated using displayed graphs in both studies)47,51 in patients after different functional changes of positioning and a fluid challenge. These different ways of changing CO, with the potential motion of bioimpedance sensors48,51 and a different reference method,47,51 could partly explain these results. Using a polar plot analysis, 2 other studies showed that 83% and 87% of ΔECco measurements were within the clinically acceptable limit of 0.5 L/min, although they reached opposite conclusions on the clinical usefulness of these values.46,49 The absence of an exclusion zone and different limits of good trending result in a difficult comparison with our results. Finally, all these different methods of analyzing trending availability (linear regression, 4-quadrant plot, and polar plot with radial limits of agreement or with a fixed ΔCO threshold) could be a source of confusion when comparing different articles concerning one device, therefore calling for a standardization of CO trending analysis.
Several factors of our experiment could explain the poor results of esCCO. First, the small sample size of this study could not represent a representative cardiac population, but our power calculation was based on a predefined polar concordance rate value of 0.8, a threshold that we consider clinically acceptable. Of note, the lower limit of an acceptable polar concordance rate is discussed by the authors who originally proposed this trending analysis.32,34 Second, we feel the need to stress that we only tested 1 approach to modify CO. For example, it was shown here that MAP significantly tracked CO during the various interventions, as SVR was only minimally affected by the ARM. Additional studies are required that address clinical situations where SVR changes, such as testing vasoconstrictors, fluid challenges, or a longer experiment duration.
Additionally, this study was conducted after cardiac surgery where inflammation and vasoactive drugs could affect the quality of the recorded peripheral signal.52 However, a recent multicenter study did not find any influence of cardiopulmonary bypass on esCCO values when compared to other ICU patients.7 Additionally, the esCCO pulse oximeter was placed on the same side as the arterial line, which may have risked decreasing the quality of the peripheral signal.9 Another potential limitation is the effect of ARM-induced sympathetic tone on the plethysmographic SQI. This is not likely a major issue, as SVR was relatively unchanged during the entire experiment (Table 2). During an ARM, the increase in CVP may have lead to “venous pulsation,”53 which can affect the plethysmographic waveform as well as the pulse index, both of which are used in the estimation of PWTT. Indeed, pulse index significantly decreased during the ARM in our study. All of these considerations further demonstrate why the esCCO is inappropriate for tracking the hemodynamic effects of an ARM. However, the SQI of both tested devices were acceptable, and the SQI of esCCO during the ARM (T2) was not significantly different from T1 and T3. Only in 1 time point of 1 patient was the SQI of ECOM too low to display a CO value. One may argue that the breadth of esCCO monitoring should not include perioperative cardiac surgery, but there is strong precedent for testing this device in such conditions42,43,54 to allow for ethical PAC insertion. Notably, the rate of complications with central venous access in this study was surprisingly high, contrasting with previous results from the same team that demonstrated no complications.35,55
When attempting to define the clinical applicability of novel CO monitoring devices with the intention of improving the overall perioperative process, it is important to evaluate the devices in the context of known hemodynamic perturbations. We have demonstrated that in postoperative cardiac surgery patients, absolute CI values of esCCO and ECOM are not interchangeable with TD during the known hemodynamic changes of an ARM. EsCCO failed to adequately track CO changes during this intervention, whereas ECOM is as accurate as invasive arterial pressure, although at a higher cost and lower availability.
The authors thank Nihon Kohden Corporation and Sylvain Thuaudet, MD (I.S.T. Cardiology, Saint Contest, France), for kindly providing all the facilities necessary for hemodynamic monitoring with the esCCO™ and ECOM™ devices.
1. Hamilton MA, Cecconi M, Rhodes A. A systematic review and meta-analysis on the use of preemptive hemodynamic intervention to improve postoperative outcomes in moderate and high-risk surgical patients. Anesth Analg. 2011;112:1392–402
2. Pearse RM, Harrison DA, MacDonald N, Gillies MA, Blunt M, Ackland G, Grocott MP, Ahern A, Griggs K, Scott R, Hinds C, Rowan KOPTIMISE Study Group. OPTIMISE Study Group. . Effect of a perioperative, cardiac output-guided hemodynamic therapy algorithm on outcomes following major gastrointestinal surgery: a randomized clinical trial and systematic review. JAMA. 2014;311:2181–90
3. Peyton PJ, Chong SW. Minimally invasive measurement of cardiac output during surgery and critical care: a meta-analysis of accuracy and precision. Anesthesiology. 2010;113:1220–35
4. Schwann NM, Hillel Z, Hoeft A, Barash P, Möhnle P, Miao Y, Mangano DT. Lack of effectiveness of the pulmonary artery catheter in cardiac surgery. Anesth Analg. 2011;113:994–1002
5. Chikhani M, Moppett IK. Minimally invasive cardiac output monitoring: what evidence do we need? Br J Anaesth. 2011;106:451–3
6. Ishihara H, Okawa H, Tanabe K, Tsubo T, Sugo Y, Akiyama T, Takeda S. A new non-invasive continuous cardiac output trend solely utilizing routine cardiovascular monitors. J Clin Monit Comput. 2004;18:313–20
7. Yamada T, Tsutsui M, Sugo Y, Sato T, Akazawa T, Sato N, Yamashita K, Ishihara H, Takeda J. Multicenter study verifying a method of noninvasive continuous cardiac output measurement using pulse wave transit time: a comparison with intermittent bolus thermodilution cardiac output. Anesth Analg. 2012;115:82–7
8. Bataille B, Bertuit M, Mora M, Mazerolles M, Cocquet P, Masson B, Moussot PE, Ginot J, Silva S, Larché J. Comparison of esCCO and transthoracic echocardiography for non-invasive measurement of cardiac output intensive care. Br J Anaesth. 2012;109:879–86
9. Tsutsui M, Araki Y, Masui K, Kazama T, Sugo Y, Archer TL, Manecke GR Jr. Pulse wave transit time measurements of cardiac output in patients undergoing partial hepatectomy: a comparison of the esCCO system with thermodilution. Anesth Analg. 2013;117:1307–12
10. Claxton BA, Morgan P, McKeague H, Mulpur A, Berridge J. Alveolar recruitment strategy improves arterial oxygenation after cardiopulmonary bypass. Anaesthesia. 2003;58:111–6
11. Duggan M, Kavanagh BP. Pulmonary atelectasis: a pathogenic perioperative entity. Anesthesiology. 2005;102:838–54
12. Futier E, Constantin JM, Paugam-Burtz C, Pascal J, Eurin M, Neuschwander A, Marret E, Beaussier M, Gutton C, Lefrant JY, Allaouchiche B, Verzilli D, Leone M, De Jong A, Bazin JE, Pereira B, Jaber SIMPROVE Study Group. . A trial of intraoperative low-tidal-volume ventilation in abdominal surgery. N Engl J Med. 2013;369:428–37
13. Magder S, Lagonidis D, Erice F. The use of respiratory variations in right atrial pressure to predict the cardiac output response to PEEP. J Crit Care. 2001;16:108–14
14. Jardin F. PEEP, tricuspid regurgitation, and cardiac output. Intensive Care Med. 1997;23:806–7
15. Knottnerus A, Tugwell P. STROBE–a checklist to Strengthen the Reporting of Observational Studies in Epidemiology. J Clin Epidemiol. 2008;61:323
16. Ramsay MA, Savege TM, Simpson BR, Goodwin R. Controlled sedation with alphaxalone-alphadolone. Br Med J. 1974;2:656–9
17. Sugo Y, Ukawa T, Takeda S, Ishihara H, Kazama T, Takeda J. A novel continuous cardiac output monitor based on pulse wave transit time. Conf Proc IEEE Eng Med Biol Soc. 2010;2010:2853–6
18. Sugo Y, Sakai T, Terao M, Ukawa T, Ochiai R. The comparison of a novel continuous cardiac output monitor based on pulse wave transit time and echo Doppler during exercise. Conf Proc IEEE Eng Med Biol Soc. 2012;2012:236–9
19. Fellahi JL, Fischer MO, Rebet O, Massetti M, Gérard JL, Hanouz JL. A comparison of endotracheal bioimpedance cardiography and transpulmonary thermodilution in cardiac surgery patients. J Cardiothorac Vasc Anesth. 2012;26:217–22
20. Cecconi M, Dawson D, Grounds RM, Rhodes A. Lithium dilution cardiac output measurement in the critically ill patient: determination of precision of the technique. Intensive Care Med. 2009;35:498–504
21. Cecconi M, Rhodes A, Poloniecki J, Della Rocca G, Grounds RM. Bench-to-bedside review: the importance of the precision of the reference technique in method comparison studies–with specific reference to the measurement of cardiac output. Crit Care. 2009;13:201
22. Bubenek-Turconi SI, Craciun M, Miclea I, Perel A. Noninvasive continuous cardiac output by the Nexfin before and after preload-modifying maneuvers: a comparison with intermittent thermodilution cardiac output. Anesth Analg. 2013;117:366–72
23. Cummings SR, Black D. Should perimenopausal women be screened for osteoporosis? Ann Intern Med. 1986;104:817–23
24. Glüer CC. Monitoring skeletal changes by radiological techniques. J Bone Miner Res. 1999;14:1952–62
25. Monge García MI, Romero MG, Cano AG, Rhodes A, Grounds RM, Cecconi M. Impact of arterial load on the agreement between pulse pressure analysis and esophageal Doppler. Crit Care. 2013;17:R113
26. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307–10
27. Bland JM, Altman DG. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat. 2007;17:571–82
28. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics. 1989;45:255–68
29. Critchley LA, Lee A, Ho AM. A critical review of the ability of continuous cardiac output monitors to measure trends in cardiac output. Anesth Analg. 2010;111:1180–92
30. McBride GB A proposal for strengh-of-agreement criteria for Lin’s concordance coreelation coefficient. 2005 National Institute of Water & Atmospheric Research Ltd., Client Report HAM2005-062
31. Perrino AC Jr, O’Connor T, Luther M. Transtracheal Doppler cardiac output monitoring: comparison to thermodilution during noncardiac surgery. Anesth Analg. 1994;78:1060–6
32. Critchley LA, Yang XX, Lee A. Assessment of trending ability of cardiac output monitors by polar plot methodology. J Cardiothorac Vasc Anesth. 2011;25:536–46
33. Desebbe O, Henaine R, Keller G, Koffel C, Garcia H, Rosamel P, Obadia JF, Bastien O, Lehot JJ, Haftek M, Critchley LA. Ability of the third-generation FloTrac/Vigileo software to track changes in cardiac output in cardiac surgery patients: a polar plot approach. J Cardiothorac Vasc Anesth. 2013;27:1122–7
34. Critchley LACannesson M, Pearse R. Evaluation of a cardiac output monitor. In: Perioperative Hemodynamic Monitoring and Goal Directed Therapy: From Theory to Practice. 2014 Cambridge Cambridge University Press:120–31
35. Desebbe O, Boucau C, Farhat F, Bastien O, Lehot JJ, Cannesson M. The ability of pleth variability index to predict the hemodynamic effects of positive end-expiratory pressure in mechanically ventilated patients under general anesthesia. Anesth Analg. 2010;110:792–8
36. Squara P, Rotcajg D, Denjean D, Estagnasie P, Brusset A. Comparison of monitoring performance of Bioreactance vs. pulse contour during lung recruitment maneuvers. Crit Care. 2009;13:R125
37. Yang XX, Critchley LA, Joynt GM. Determination of the precision error of the pulmonary artery thermodilution catheter using an in vitro continuous flow test rig. Anesth Analg. 2011;112:70–7
38. Yang XX, Critchley LA, Rowlands DK, Fang Z, Huang L. Systematic error of cardiac output measured by bolus thermodilution with a pulmonary artery catheter compared with that measured by an aortic flow probe in a pig model. J Cardiothorac Vasc Anesth. 2013;27:1133–9
39. Luecke T, Pelosi P. Clinical review: positive end-expiratory pressure and cardiac output. Crit Care. 2005;9:607–21
40. Balik M, Pachl J, Hendl J, Martin B, Jan P, Jan H. Effect of the degree of tricuspid regurgitation on cardiac output measurements by thermodilution. Intensive Care Med. 2002;28:1117–21
41. Dean DA, Jia CX, Cabreriza SE, D’Alessandro DA, Dickstein ML, Sardo MJ, Chalik N, Spotnitz HM. Validation study of a new transit time ultrasonic flow probe for continuous great vessel measurements. ASAIO J. 1996;42:M671–6
42. Ball TR, Tricinella AP, Kimbrough BA, Luna S, Gloyna DF, Villamaria FJ, Culp WC Jr. Accuracy of noninvasive estimated continuous cardiac output (esCCO) compared to thermodilution cardiac output: a pilot study in cardiac patients. J Cardiothorac Vasc Anesth. 2013;27:1128–32
43. Fischer MO, Balaire X, Le Mauff de Kergal C, Boisselier C, Gérard JL, Hanouz JL, Fellahi JL. The diagnostic accuracy of estimated continuous cardiac output compared with transthoracic echocardiography. Can J Anaesth. 2014;61:19–26
44. Sinha AC, Singh PM, Grewal N, Aman M, Dubowitz G. Comparison between continuous non-invasive estimated cardiac output by pulse wave transit time and thermodilution method. Ann Card Anaesth. 2014;17:273–7
45. Ball TR, Culp BC, Patel V, Gloyna DF, Ciceri DP, Culp WC Jr. Comparison of the endotracheal cardiac output monitor to thermodilution in cardiac surgery patients. J Cardiothorac Vasc Anesth. 2010;24:762–6
46. Maus TM, Reber B, Banks DA, Berry A, Guerrero E, Manecke GR. Cardiac output determination from endotracheally measured impedance cardiography: clinical evaluation of endotracheal cardiac output monitor. J Cardiothorac Vasc Anesth. 2011;25:770–5
47. Fellahi JL, Fischer MO, Rebet O, Massetti M, Gérard JL, Hanouz JL. A comparison of endotracheal bioimpedance cardiography and transpulmonary thermodilution in cardiac surgery patients. J Cardiothorac Vasc Anesth. 2012;26:217–22
48. Møller-Sørensen H, Hansen KL, Østergaard M, Andersen LW, Møller K. Lack of agreement and trending ability of the endotracheal cardiac output monitor compared with thermodilution. Acta Anaesthesiol Scand. 2012;56:433–40
49. Maass SW, Roekaerts PM, Lancé MD. Cardiac output measurement by bioimpedance and noninvasive pulse contour analysis compared with the continuous pulmonary artery thermodilution technique. J Cardiothorac Vasc Anesth. 2014;28:534–9
50. Ishihara H, Sugo Y, Tsutsui M, Yamada T, Sato T, Akazawa T, Sato N, Yamashita K, Takeda J. The ability of a new continuous cardiac output monitor to measure trends in cardiac output following implementation of a patient information calibration and an automated exclusion algorithm. J Clin Monit Comput. 2012;26:465–71
51. Fellahi JL, Fischer MO, Dalbera A, Massetti M, Gérard JL, Hanouz JL. Can endotracheal bioimpedance cardiography assess hemodynamic response to passive leg raising following cardiac surgery? Ann Intensive Care. 2012;2:26
52. Pezawas T, Rajek A, Skolka M, Schneider B, Plöchl W. Perspectives for core and skin surface temperature guided extubation in patients after normothermic cardiopulmonary bypass. Intensive Care Med. 2004;30:1676–80
53. Shelley KH, Tamai D, Jablonka D, Gesquiere M, Stout RG, Silverman DG. The effect of venous pulsation on the forehead pulse oximeter wave form as a possible source of error in Spo2
calculation. Anesth Analg. 2005;100:743–7
54. Wacharasint P, Kunakorn P, Pankongsap P, Preechanukul R. Clinical validation of pulse contour and pulse wave transit time-based continuous cardiac output analyses in Thai patients undergoing cardiac surgery. J Med Assoc Thai. 2014;97(Suppl 1):S55–60
55. Desgranges FP, Desebbe O, Ghazouani A, Gilbert K, Keller G, Chiari P, Robin J, Bastien O, Lehot JJ, Cannesson M. Influence of the site of measurement on the ability of plethysmographic variability index to predict fluid responsiveness. Br J Anaesth. 2011;107:329–35