Share this article on:

Pulse Wave Transit Time Measurements of Cardiac Output in Patients Undergoing Partial Hepatectomy: A Comparison of the esCCO System with Thermodilution

Tsutsui, Masato MD, PhD*; Araki, Yoshiyuki MD*; Masui, Kenichi MD, PhD*; Kazama, Tomiei MD*; Sugo, Yoshihiro; Archer, Thomas L. MD, MBA; Manecke, Gerard R. Jr MD

doi: 10.1213/ANE.0b013e3182a44c87
Technology, Computing, and Simulation: Research Report

BACKGROUND: Measuring cardiac output accurately during anesthesia is thought to be helpful for safely controlling hemodynamics. Several minimally invasive methods to measure cardiac output have been developed as alternatives to thermodilution with pulmonary artery catheterization. We evaluated the reliability of a novel pulse wave transit time method of cardiac output assessment to trend with thermodilution cardiac output in patients undergoing partial hepatectomy.

METHODS: Thirty-one patients (ASA physical status II or III) undergoing partial hepatectomy under general anesthesia were evaluated. Cardiac output measurements by pulse wave transit time method and by thermodilution were recorded after induction of anesthesia, after a change in body positioning to 20° head up, after a change to 20° head down, after volume challenge with 10 mL·kg−1 hydroxyethyl starch 6%, during the Pringle maneuver, and immediately after Pringle maneuver release. Trending was assessed using Bland-Altman analysis and concordance analysis.

RESULTS: The direction of change between consecutive pulse wave transit time measurements and the corresponding thermodilution measurements showed a concordance rate of 96.0% (lower 95% confidence interval = 64%), with limits of agreement −1.51 and 1.61 L·min−1.

CONCLUSIONS: The pulse wave transit time method had good concordance but fairly wide limits of agreement with regard to trending in patients with changes in preload and systemic vascular resistance. There are potential inaccuracies when vasopressors are used to treat hypotension associated with decreased systemic vascular resistance. The study limitations are that the cardiac output data were collected in a nonblinded fashion, and an existing intraarterial catheter was used, although the system requires only routine, noninvasive cardiovascular monitors. This is a promising technique that currently has limitations and will require further improvements and clinical assessment.

From the *Department of Anesthesiology, National Defense Medical College, Saitama Vital Sign Sensor Department, Monitoring Technology Center, Nihon Kohden Corporation, Tokyo, Japan; and Department of Anesthesiology, University California, San Diego Medical Center, San Diego, California.

Accepted for publication May 22, 2013.

Funding: Funded by Department of Anesthesiology, National Defense Medical College; Device manufacturer is Nihon Kohden Corporation.

The authors declare no conflicts of interest.

Address correspondence and reprint requests to Masato Tsutsui, MD, PhD, Department of Anesthesiology, National Defense Medical College, 3-2 Namiki, Tokorozawa, Saitama, 359–8513 Japan. Address e-mail to

Measuring cardiac output accurately during anesthesia is thought to be helpful for safely managing hemodynamics. Thermodilution with a pulmonary artery catheter is considered the clinical “gold standard.” Because of the risks of pulmonary artery catheterization,1,2 less invasive devices based on arterial pulse contour analysis have been developed, such as the FloTrac/Vigileo system (Edwards Lifesciences, Irvine, CA), the PiCCO system (Pulsion, Munich, Germany), and the PulseCO system (LiDCO, Cambridge, United Kingdom). However, these less invasive methods require an invasive arterial catheter to obtain the arterial blood pressure waveform.

In 2004, a measurement method for determining continuous cardiac output based on pulse contour analysis of the pulse-oximetry waveform and arterial pulse wave transit time (esCCO system, Nihon Kohden, Tokyo, Japan) was introduced and evaluated in the early postoperative period after cardiac surgery.3 This system requires an initial 3-minute period of stable hemodynamics for calibration against another cardiac output measurement system. It uses no sensors other than routine cardiovascular monitoring equipment (pulse oximetry, electrocardiogram, noninvasive arterial blood pressure monitoring), although in this study an intraarterial catheter for pulse pressure determination was used. Details about this technology are presented in the Appendix.

The aim of this study was to determine the trending reliability of cardiac output measured by the esCCO system (COES) compared with cardiac output measured by intermittent thermodilution (COTD) in patients undergoing partial hepatectomy, with significant changes of preload and systemic vascular resistance caused by head-up tilt, head-down tilt, fluid challenge, and the Pringle maneuver. The Pringle maneuver involves clamping of the hepatoduodenal ligament, interrupting blood flow through the hepatic artery and the portal vein in liver surgery to control bleeding. It was studied here because of its dramatic effects on cardiac preload and systemic vascular resistance.

Back to Top | Article Outline


Patients and Anesthesia

After IRB approval and written informed consent from each patient, 35 patients (ASA physical status II or III) scheduled to undergo partial hepatectomy under general anesthesia with pulmonary artery catheterization were enrolled. Patients with preoperative tricuspid valve regurgitation or persistent arrhythmias were excluded (persistent arrhythmias render the esCCO system inaccurate).

After premedication with IM atropine 0.5 mg and hydroxyzine 50 mg, epidural catheterization was performed at the thoracic T9–10 or T10–11 interspace. Anesthesia was induced and maintained with a target-controlled infusion of propofol,4 supplemented with a nontarget-controlled infusion of remifentanil. After endotracheal intubation facilitated by vecuronium, a radial arterial catheter was placed in the left wrist (20-gauge catheter, Insyte A®, Becton, Dickinson, Franklin Lakes, NJ). A pulmonary artery catheter (744HF75, Edwards Lifesciences, Irvine, CA) was inserted via an introducer (AVA Triple Lumen, Edwards Lifesciences) placed in the right internal jugular vein. Ventilation was controlled to keep the end-tidal PCO2 30 to 40 mm Hg, with respiratory rate 8 to 12 cycles·min−1 and tidal volume 8 to 12 mL·kg−1 with 34% oxygen in air. The inspiratory-to-expiratory ratio was 1:2. No positive end-expiratory pressure was applied. After an initial epidural injection of 1% mepivacaine (6–10 mL), a continuous epidural infusion (4–8 mL·h−1) was administered during the operation. Fluid administration (other than a volume challenge), blood transfusion, and administration of inotrophic and vasodilatory drugs were left to the discretion of the anesthesiologist.

Back to Top | Article Outline

Thermodilution Cardiac Output Measurement

COTD using 10 mL cold isotonic saline injected through the proximal right atrial port of the pulmonary artery catheter was measured using a Vigilance (Edwards Lifesciences) cardiac output computer. Each indicator injection was synchronized to the end-expiratory phase of the respiratory cycle. Of 3 consecutive COTD measurements, a value >15% from the average was discarded as an outlier. The average COTD was calculated using the remaining 2 measurements. Fluid administration via the introducer containing the pulmonary artery catheter was halted during COTD measurements.

Back to Top | Article Outline

Investigation Procedure

Hemodynamic recordings at each time point included heart rate, mean arterial blood pressure (MAP), central venous pressure (CVP), mean pulmonary arterial pressure (mPAP), pulmonary capillary wedge pressure, COTD, and COES. COES was displayed continuously on a laptop computer, using a Nihon Kohden BSM9101 bedside monitor (Nihon Kohden, Tokyo, Japan) with the finger probe (TL-201T), in which the esCCO system program software (esCCOMon, version 01-04) was installed. The 20-gauge radial arterial catheter was connected to a commercially available fluid-filled transducer to obtain pulse pressure (TruWave Disposable Pressure Transducer™, Edwards Lifesciences). The arterial waveform was observed to have acceptable morphology before pulse pressure measurements. After preparation of the esCCO system, hemodynamic measurements were manually recorded at the following time points: after induction of anesthesia, 1 minute after a change in body position to 20° head-up tilt, 1 minute after a change to 20° head-down tilt, 1 minute after intravascular volume loading with 10 mL·kg−1 hydroxyethyl starch 6% (HES 70/0.55, Fresenius Kabi, Stans, Switzerland) >20 minutes, 1 minute after application of the first Pringle maneuver, and 1 minute after the release of the Pringle maneuver. After each tilting and before the next measurement, a 1-minute adaptation period was provided to allow hemodynamics to stabilize.

After induction of anesthesia, the esCCO system was calibrated to COTD. At each subsequent hemodynamic measurement time point, COES was recorded before and after the 3 thermodilution measurements. The values reported for COES are the averages of the values obtained immediately before and after the 3 COTD measurements. The anesthesiologist recording COES and COTD measurements was not blinded to either measurement. Measurements were not performed during or immediately after bolus catecholamine administration or during periods of acute rapid changes in hemodynamics. If the MAP was <40 mm Hg at 1 minute after head-up tilt, or if the mPAP was >30 mm Hg after volume challenge, the investigation was discontinued for reasons of patient safety. Data from these patients were excluded.

Back to Top | Article Outline

Data Analysis and Statistics

A power analysis determined that a sample size of 31 patients was required to ensure that a mean difference (bias) of 0.5 L·min−1 would be detected at the significance level of 0.05 with a statistical power of 90%, assuming a standard deviation (SD) of differences to be 0.75 L·min−1 (based on mean cardiac output of 5.0 L·min−1 and clinically acceptable error 2 SD·mean−1 ± 30%).5 Statistical analysis was performed using GraphPad PRISM (version 4.0, GraphPad software, San Diego, CA). Hemodynamic variables at each time point were compared with the cardiac output measurement after induction by analysis of variance for repeated measurements, which has the statistical assumptions of normal distribution, homogeneity of variance, and sphericity. Tukey multiple comparison test was done as the post hoc analysis. A value of P < 0.05 was considered statistically significant.

The ability of the esCCO system to track changes in COTD was assessed by plotting ΔCOES against ΔCOTD on a 4-quadrant plot (Fig. 1A) with regression analysis, and a Bland-Altman plot (Fig. 1B), where the change in the cardiac output measurement (ΔCOES or ΔCOTD) was the change in that measurement between consecutive time points. We calculated the SD of difference about the mean to show limits of agreement in the pooled data. However, there is an error of repeated measurement in the original Bland-Altman plot, which is on the basis of the within-subject variance estimated by the random effects model. To correct for this error, we determined 2 different variances, that for repeated differences between the 2 methods on the same subject and that for the differences between the averages of the 2 methods across subjects as recommended by Bland and Altman in 2007.7 The first variance, that within subjects, can be estimated using 1-way analysis of variance, using the difference in matched pairs. This variance is the residual mean square. The other component of the variance, for differences between the average difference across subjects, can also be found from this result, using the difference between the mean squares for subjects and the residual mean square. Then, this value is divided by a value that depends on the numbers of observations on each subject. There is the assumption of repeated measurement. A concordance analysis with lower 95% confidence limits was performed to determine the directional agreement between the devices in detecting changes in cardiac output between time points.7–10 The concordance rate was calculated as the percentage of the total number of changes that were in the same direction. To exclude random statistical effects, this rate was determined for all consecutive COTD values that differed by at least 0.5 L·min−1.10

Figure 1

Figure 1

Back to Top | Article Outline


All patients in the study underwent partial hepatectomy, and none suffered cirrhosis. Of the 35 patients, 3 were excluded because the MAP decreased to <40 mm Hg at 1 minute after head-up tilt, and 1 was excluded because low voltage on the electrocardiogram (ECG) precluded accurate assessment of the R wave. Hence 31 patients and 155 data pairs for COTD and COES were obtained. Significant arrhythmias or massive bleeding did not occur. β-Blockers were used preoperatively in 2 patients. Of the 31 patients studied, 2 patients received low-dose dopamine (3 µg·kg−1·min−1), and 7 patients received phenylephrine boluses (0.1 mg) for hypotension at release of the Pringle maneuver. Moreover, 3 patients received nicorandil 6 mg·h−1, a potassium channel activator with arterial and venous vasodilatory properties. Initial epidural injection was performed between 1 minute after head-down tilt and the beginning of surgery. In all cases, surgery started after volume challenge.

Table 1 shows the demographic data of the 31 patients. Hemodynamic variables are presented in Table 2. Individual COTD measurements ranged from 1.50 to 10.10 L·min−1, and individual COES measurements ranged from 1.42 to 10.47 L·min−1. The 4-quadrant trend plot and Bland-Altman plot between ΔCOTD and ΔCOES are shown in Figure 1. A significant trending effect is demonstrated, with an r-value of 0.85 with lower 95% confidence interval = 0.73 (P < 0.0001). The slope of the regression line is 0.68 with 2-sided confidence interval, 0.63 to 0.79. The direction of change in the cardiac output readings revealed a concordance rate of 96.0%, with lower 95% confidence interval at 64%. Table 3 shows the concordance rate between each time point calculated from ΔCOTD to ΔCOES. Relatively low concordance (83.3%) was shown at Pringle maneuver release, although concordance rates were >95% at all other time points. The trending Bland-Altman plot had a bias of 0.04 L·min−1 and limits of agreement of −1.51 to 1.61 L·min−1. Inspection of the plot suggests that agreement in the magnitude of changes between the 2 systems was better in the cases of small changes than in large changes (the plot splays out at high cardiac output change values). Figure 2 shows the difference between paired measurements of COTD and COES plotted against systemic vascular resistance. There was an extremely low relationship (r = 0.13) and narrow 95% confidence interval, 0.24 to 0.46.

Table 1

Table 1

Table 2

Table 2

Table 3

Table 3

Figure 2

Figure 2

Back to Top | Article Outline


The most important finding of our study is that the esCCO system, when using an arterial line source for pulse pressure, reliably trends directional changes in cardiac output. Regression analysis revealed high correlation between ΔCOTD and ΔCOES (r = 0.85). However, the slope of the regression line (0.68) is less than that of the line of identity. This indicates that the esCCO system may underestimate the magnitude of changes in cardiac output. Biancofiore et al.11 recommended the 4-quadrant plot of serial changes in cardiac output and concordance analysis for evaluation of new methods of cardiac output measurement. This was done in this study, as was a Bland-Altman plot comparing changes. The esCCO system correctly assessed the direction of serial changes in cardiac output, with a concordance rate of 96.0% with COTD. However, wide limits of agreement on Bland-Altman analysis indicate that trending of the magnitude of change is less accurate, particularly when large changes in cardiac output occur. Previous reports have suggested the threshold of concordance should be within 90% to 95% for assuming good trending ability when exclusion criteria of 0.5 to 1.0 L·min−1 are applied,7–9 and these data meet that criterion.

Head-up tilt and head-down tilt position changes and fluid administration using hydroxyethyl starch are often used as first-step maneuvers to treat hemodynamic instability.11–15 Most of the hemodynamic changes associated with the various maneuvers in this study have been described.11–13 With fluid administration, systemic vascular resistance decreased, whereas MAP, CVP, mPAP, pulmonary capillary wedge pressure, and COTD all increased. During the Pringle maneuver, COTD and CVP decreased and systemic vascular resistance increased, as previously noted.16,17 Release of the Pringle maneuver was associated with a statistically nonsignificant decrease in MAP. In 9 cases, however, arterial blood pressure after release of the Pringle maneuver was supported with phenylephrine or dopamine.

There was a low concordance rate (83.3%) at change from Pringle maneuver to release, although there were good concordance rates at changes between all other time points. This demonstrates that ΔCOES may not have clinically good trending ability when vasopressors are used to treat hypotension associated with decreased systemic vascular resistance. However, Figure 2 failed to show a correlation between systemic vascular resistance and bias. We believe drug-induced acute changes of vascular conditions might influence the relationship of pulse wave transit time and ΔCOES because pulse wave transit time is influenced both by changes of cardiac contractility and arterial compliance.18 The Pringle maneuver and its release caused hemodynamic instability as a consequence of changes of preload and systemic vascular resistance. Sudden drug-induced acute changes of systemic vascular resistance (e.g., administration of vasoconstrictor or vasodilator) are likely not physiologically identical to changes in systemic vascular resistance, resulting from changes in loading conditions and might lead to a relatively low relationship (0.85 with lower 95% confidence interval = 0.73). More investigation needs to be done in this area for clarification.

There are some real and potential limitations of the esCCO method. First, this system needs calibration with another cardiac output measurement system at the start of continuous measurement. Second, for calibration, it needs at least 3 minutes of hemodynamic stability. Third, this system has not been evaluated in situations that would likely render the system inaccurate: patients with arrhythmias such as atrial fibrillation and ventricular premature contractions, use of either a cardiac pacemaker or intraaortic balloon counterpulsation, and/or suboptimal signal for ECG or pulse oximetry (e.g., hypothermia, cold environment). Continued improvements in the system, as well as studies of its clinical utility, are needed. Perhaps, in the future, a version of this system that does not require external calibration, using patient demographic data and other factors to calculate a calibration constant (similar in concept to the Edwards FloTrac), may be developed. The potential utility of the system in its current form might be in cases in which a more invasive method of cardiac output measurement was used (e.g., pulmonary artery catheter) and then discontinued. Perhaps, this system could be calibrated to the invasive method before its discontinuation and then used subsequently for hemodynamic management and goal-directed fluid therapy. It is not yet known, however, if or how often this system would require recalibration. Future research should be done to determine the time course and frequency with which the esCCO system would require recalibration.

There are limitations in this study. First, low-risk patients scheduled for only partial hepatectomy were enrolled, and patients with severe hypotension were excluded. Further investigations will be required to assess this technique in higher-risk and hemodynamically unstable patients. Second, the investigator recording COTD and COES was not blinded to the results, potentially leading to bias. Third, only patients under general anesthesia were studied. Fourth, an existing arterial line was used, instead of a noninvasive arterial blood pressure device, for determining pulse pressure. The most recent esCCO software (version 01-04), which was not available at the time of this investigation, allows for use of either an arterial line or noninvasive arterial blood pressure measurement. Further study is required to determine whether using noninvasive arterial blood pressure measurement will result in accuracy comparable with that found in this investigation.

This study showed that during partial hepatectomy the esCCO system provides good trending ability (96% concordance rate) when compared with the thermodilution technique in patients under general anesthesia with preload and systemic vascular resistance changes. The current system may be clinically useful as a substitute for the thermodilution method except perhaps when vasopressors are used to treat hypotension associated with decreased systemic vascular resistance or when the cardiac output is particularly low. With future engineering and technical improvements, use of the arterial pulse wave and pulse transit time offers the possibility of measuring cardiac output accurately, quickly, and noninvasively.

Back to Top | Article Outline


In the esCCO method, the pulse pressure (PP), an index of arterial pulsatility, and pulse wave transit time, an index of vascular resistance, compliance, and cardiac contractility, are used to calculate stroke volume (SV).4,19 PP is proportional to SV, with K, a “resistance/compliance” coefficient, further defining the relationship. The components of pulse wave transit time are illustrated in Figure 3. In the esCCO algorithm, pulse wave transit time is calculated as the time from the electrocardiogram (ECG) R-wave peak to the rise point of the pulse oximeter wave. The rise point of the pulse wave was defined as the point at which the differentiated pulse wave reached 30% of its peak amplitude.18

Figure 3

Figure 3

CO is derived as follows:

K is derived from pulse wave transit time, heart rate (HR), and cardiac output (CO) at the start of measurement in each patient. Accordingly, a CO value derived by another CO measurement system is required at the start of esCCO measurement as a calibration. In this study, COTD was used for this purpose. SV is derived as follows:

Back to Top | Article Outline

Data Processing and Exclusion Criteria

Pulse wave transit time is calculated by averaging 64 consecutive data of heart beats. Average pulse wave transit time and HR are based on data retrieved within every 1-second interval. Data with a large variability in pulse wave transit time (>20 milliseconds) or pulse amplitude deviating from median values (>30%) during calculation are excluded. In addition, calculation process of pulse wave transit time is automatically inhibited when >25% of the 64 heart beats data are excluded in the following conditions:

  1. Either ECG or pulse-oximetry pulse wave signal is not obtained.
  2. Either R wave on ECG or the start point of the ascending portion of pulse-oximetry wave is not clearly identified.

COES is then calculated using these calculated pulse wave transit time and HR. Data of inadequate conditions are also excluded automatically. Pulse wave transit time, HR, and PP are averaged for at least 3 minutes to calculate K and β.

Back to Top | Article Outline


Name: Masato Tsutsui, MD, PhD.

Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.

Attestation: Masato Tsutsui has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Name: Yoshiyuki Araki, MD.

Contribution: This author helped conduct the study.

Attestation: Yoshiyuki Araki approved the final manuscript.

Name: Kenichi Masui, MD, PhD.

Contribution: This author helped analyze the data.

Attestation: Kenichi Masui has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Tomiei Kazama, MD.

Contribution: This author helped design and conduct the study and write the manuscript.

Attestation: Tomiei Kazama approved the final manuscript.

Name: Yoshihiro Sugo.

Contribution: This author helped write the manuscript.

Attestation: Yoshihiro Sugo approved the final manuscript.

Name: Thomas L. Archer, MD, MBA.

Contribution: This author helped write the manuscript.

Attestation: Thomas L. Archer approved the final manuscript.

Name: Gerard R. Manecke Jr, MD.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Gerard R. Manecke Jr has seen the original study data and approved the final manuscript.

This manuscript was handled by: Dwayne R. Westenskow, PhD.

Back to Top | Article Outline


We specially thank Dr. H. Ishihara, Department of Anesthesiology, Hirosaki University, Japan, for providing theory of the esCCO system.

Back to Top | Article Outline


1. Ochiai R, Takeda J, Hosaka H, Sugo Y, Tanaka R, Soma T. The relationship between modified pulse wave transit time and cardiovascular changes in isoflurane anesthetized dogs. J Clin Monit Comput. 1999;15:493–501
2. Sirivella S, Gielchinsky I, Parsonnet V. Management of catheter-induced pulmonary artery perforation: a rare complication in cardiovascular operations. Ann Thorac Surg. 2001;72:2056–9
3. Bender JS, Smith-Meek MA, Jones CE. Routine pulmonary artery catheterization does not reduce morbidity and mortality of elective vascular surgery: results of a prospective, randomized trial. Ann Surg. 1997;226:229–36
4. 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
5. Marsh B, White M, Morton N, Kenny GN. Pharmacokinetic model driven infusion of propofol in children. Br J Anaesth. 1991;67:41–8
6. Critchley LA, Critchley JA. A meta-analysis of studies using bias and precision statistics to compare cardiac output measurement techniques. J Clin Monit Comput. 1999;15:85–91
7. Bland JM, Altman DG. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat. 2007;17:571–82
8. Perrino AC Jr, Harris SN, Luther MA. Intraoperative determination of cardiac output using multiplane transesophageal echocardiography: a comparison to thermodilution. Anesthesiology. 1998;89:350–7
9. Jansen JR, Schreuder JJ, Mulier JP, Smith NT, Settels JJ, Wesseling KH. A comparison of cardiac output derived from the arterial pressure wave against thermodilution in cardiac surgery patients. Br J Anaesth. 2001;87:212–22
10. Osthaus WA, Huber D, Beck C, Winterhalter M, Boethig D, Wessel A, Sümpelmann R. Comparison of electrical velocimetry and transpulmonary thermodilution for measuring cardiac output in piglets. Paediatr Anaesth. 2007;17:749–55
11. Biancofiore G, Critchley LA, Lee A, Bindi L, Bisà M, Esposito M, Meacci L, Mozzo R, DeSimone P, Urbani L, Filipponi F. Evaluation of an uncalibrated arterial pulse contour cardiac output monitoring system in cirrhotic patients undergoing liver surgery. Br J Anaesth. 2009;102:47–54
12. Reuter DA, Felbinger TW, Schmidt C, Moerstedt K, Kilger E, Lamm P, Goetz AE. Trendelenburg positioning after cardiac surgery: effects on intrathoracic blood volume index and cardiac performance. Eur J Anaesthesiol. 2003;20:17–20
13. Felbinger TW, Reuter DA, Eltzschig HK, Bayerlein J, Goetz AE. Cardiac index measurements during rapid preload changes: a comparison of pulmonary artery thermodilution with arterial pulse contour analysis. J Clin Anesth. 2005;17:241–8
14. Berry NM, Rickards CA, Newman DG. Acute cardiovascular adaptation to 10 consecutive episodes of head-up tilt. Aviat Space Environ Med. 2006;77:494–9
15. Chan GS, Middleton PM, Celler BG, Wang L, Lovell NH. Change in pulse transit time and pre-ejection period during head-up tilt-induced progressive central hypovolaemia. J Clin Monit Comput. 2007;21:283–93
16. Senn A, Button D, Zollinger A, Hofer CK. Assessment of cardiac output changes using a modified FloTrac/Vigileo algorithm in cardiac surgery patients. Crit Care. 2009;13:R32
17. Howdieshell TR, Wood M, Swayne M, Duvall R, Mooney S, Stark N. Effects of intraluminal and extracorporeal inferior vena caval bypass on canine hemodynamics. Crit Care Med. 1996;24:631–4
18. Haberstroh J, Ahrens M, Munzar T, Waninger J, Salm R, Matern U, Pauly E, von Specht BU. Effects of the Pringle maneuver on hemodynamics during laparoscopic liver resection in the pig. Eur Surg Res. 1996;28:8–13
19. Sugo Y UT, 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
© 2013 International Anesthesia Research Society