Figures
Abstract
Background
We determined reliability of cardiac output (CO) measured by pulse wave transit time cardiac output system (esCCO system; COesCCO) vs transthoracic echocardiography (COTTE) in mechanically ventilated patients in the early phase of septic shock. A secondary objective was to assess ability of esCCO to detect change in CO after fluid infusion.
Methods
Mechanically ventilated patients admitted to the ICU, aged >18 years, in sinus rhythm, in the early phase of septic shock were prospectively included. We performed fluid infusion of 500ml of crystalloid solution over 20 minutes and recorded CO by EsCCO and TTE immediately before (T0) and 5 minutes after (T1) fluid administration. Patients were divided into 2 groups (responders and non-responders) according to a threshold of 15% increase in COTTE in response to volume expansion.
Results
In total, 25 patients were included, average 64±15 years, 15 (60%) were men. Average SAPSII and SOFA scores were 55±21.3 and 13±2, respectively. ICU mortality was 36%. Mean cardiac output at T0 was 5.8±1.35 L/min by esCCO and 5.27±1.17 L/min by COTTE. At T1, respective values were 6.63 ± 1.57 L/min for esCCO and 6.10±1.29 L/min for COTTE. Overall, 12 patients were classified as responders, 13 as non-responders by the reference method. A threshold of 11% increase in COesCCO was found to discriminate responders from non-responders with a sensitivity of 83% (95% CI, 0.52-0.98) and a specificity of 77% (95% CI, 0.46-0.95).
Citation: Feissel M, Aho LS, Georgiev S, Tapponnier R, Badie J, Bruyère R, et al. (2015) Pulse Wave Transit Time Measurements of Cardiac Output in Septic Shock Patients: A Comparison of the Estimated Continuous Cardiac Output System with Transthoracic Echocardiography. PLoS ONE 10(6): e0130489. https://doi.org/10.1371/journal.pone.0130489
Editor: Philip Alexander Efron, University of Florida, UNITED STATES
Received: February 19, 2015; Accepted: May 19, 2015; Published: June 30, 2015
Copyright: © 2015 Feissel et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Adequate tissue perfusion and oxygen delivery is the primary goal of therapeutic management in patients with circulatory failure. Cardiac output (CO) is a major determinant of oxygen delivery and is a function of the interaction between the cardiac pump and venous return. Volume expansion is key in order to improve CO[1] and reduce mortality[2]. Furthermore, hemodynamic monitoring plays an important role in the management of these patients, since excessive volume loading may exacerbate the commonly associated respiratory failure[3], limit oxygen diffusion to the tissues by interstitial fluid accumulation[4] and contribute to increased mortality[5].
The optimal monitoring system depends on the individual patient, the current or potential condition for which monitoring is required, and the equipment and expertise available on site[6]. Thermodilution has become the gold standard for the measurement of CO[7], but is now less frequently performed due to its invasive nature.
Many technologies have been developed for minimally invasive monitoring of CO, such as the pulse contour method, based on the relationship between arterial blood pressure and stroke volume. Arterial waveform-based CO is also widely used, but requires arterial puncture[8–9].
One commercially available device is a non-invasive continuous cardiac output measurement method (esCCO, Nihon Kohden, Tokyo Japan) that has recently been validated. It is based on pulse contour analysis combined with pulse wave transit time[10–11]. This method requires only the measurement of peripheral pulse oximetry, and non-invasive blood pressure and ECG-monitoring[12].
Echocardiography has been validated for CO measurement in intensive care patients, particularly in septic shock where the causes of circulatory failure are multiple and often overlap[13–14]. However, echocardiography is not amenable to continuous CO monitoring, and also requires a minimum of training for the operator to be proficient.
In this context, the primary aim of this study was to determine the reliability of CO measured by the esCCO system (COesCCO) as compared with CO measured by transthoracic echocardiography (COTTE) in a homogeneous population of mechanically ventilated patients in the early phase of septic shock. A secondary aim was to evaluate the ability of esCCO to detect a change in CO after fluid infusion.
Methods
Patient population
We prospectively screened for eligibility all mechanically ventilated patients admitted to the ICU, older than 18 years, in sinus rhythm, and in the early phase of septic shock (defined as persistent hypotension, despite adequate vascular filling of 20 mL/kg, and need for norepinephrine). We only included patients who required additional fluids because of persistent hypotension (systolic blood pressure < 90 mm Hg), oliguria (urine output < 0.5 mL/kg per hour), or metabolic acidosis (base deficit ≥ 5 or lactates > 2.5 mmol/L). Sedation and analgesia were provided by continuous infusion of midazolam and sufentanil titrated for adaptation without spontaneous breathing (Ramsay score of 4–5).
We excluded patients with an increase in filling pressures defined by transmitral Doppler echocardiography criteria (i.e. E/E’ >15; E/EDT > 5.2)[15] as well as patients in whom the quality of transthoracic echocardiography images was insufficient; patients with intra-aortic balloon pump; patients with severe aortic stenosis; and patients undergoing pacing.
Fluid challenge was performed with a view to achieving mean arterial pressure of 65 mm Hg or higher, and maintaining adequate tissue perfusion. All patients were mechanically ventilated in a volume-controlled mode, with a tidal volume of 6–8 mL/kg of predicted body weight.
The institutional review board (Comité de Protection des Personnes Est I, Dijon) approved the protocol, and considered it to constitute routine clinical practice. The need for informed consent was waived, but all patients or their relatives were given clear information about the study, and their non-opposition was obtained.
Measurements
EsCCO was displayed continuously on a laptop computer, using a Nihon Kohden BSM9101 bedside monitor (Nihon Kohden, Tokyo, Japan), in which the esCCO system program software (esCCOMon, version 01–04) was installed. The pulse wave transit time is the measure of the interval between the ECG R wave and the pulse plethysmograph upstroke[16]. Details of CO calculation have previously been published elsewhere[11].
All transthoracic echocardiography (TTE) measurement was performed using a CX 50 device (Philips Medical Systems, Andover, MA, USA). Measurements were performed by an investigator (MF, JB) who was blinded to the measurements determined by esCCO. The Doppler echocardiography estimated CO (COTTE) was derived from the Doppler estimated Stroke Volume (SV) using the velocity-time integral (VTI) of flow through the left ventricular outflow tract (LVOT), the diameter of the LVOT, and heart rate recorded during the imaging study[14]. The VTI value was averaged over five consecutive measurements. The method of CO calculation has previously been described[17].
Intervention
In all patients included in the study, we performed fluid infusion of 500 mL of crystalloid solution over a 20-minute period and recorded COesCCO and COTTE before (T0) and 5 minutes after (T1) administration of the fluids. Patients were divided into 2 groups, namely, responders and non-responders, according to the increase in COTTE in response to volume expansion. In accordance with previous studies[18–19], we used an increase of COTTE of 15% or more after fluid infusion to differentiate responders from non-responders. The ventilatory settings and the rate of administration of vasoactive drugs remained unchanged throughout the intervention.
Data recorded
We recorded the following variables for all patients: socio-demographic data, Simplified Acute Physiologic Score (SAPS II)[20] (the worst value within 24 hours after admission), Sepsis-related Organ Failure Assessment (SOFA) score[21] (calculated within the first 24 hours after onset of septic shock, with the most abnormal value for each of the 6 organ systems recorded), origin of sepsis, and outcome.
Statistical analysis
All data are presented as mean ± SD for quantitative variables and number (percentage) for qualitative variables.
Inter-observer variability was calculated from 3 measures by each observer in each of 10 randomly selected patients, and was expressed as the mean percent error (i.e. the difference between the 2 observed values divided by the mean of the 2 observed values). Intra-observer variability was calculated from 3 COTTE measures by the same observer from the same 10 randomly selected patients,
Agreement between COesCCO and COTTE was assessed in three steps:
Firstly, the relationship between the two methods was estimated using simple linear regression. Robust variance estimator and bootstrapped variance estimator were also used.
Secondly, Pearson’s correlation coefficient between COesCCO and COTTE was calculated. Agreement was estimated using the 95% limit of-agreement method developed by Bland and Altman[22], in which the difference between X and Y is plotted against their means, with the 95% confidence intervals. The direction and magnitude of the spread of values around the zero line provides an approximation of any potential systematic bias and random error, respectively. Heteroscedasticity can also be detected where the difference is found to be a function of the mean.
Finally, we computed Lin’s concordance coefficient correlation ρc [23]. Lin’s coefficient combines measures of precision and accuracy to determine whether the observed data deviate significantly from the line of perfect concordance (i.e. the 45° line through the origin). No adjustment for location was performed, since the influence of this variable was not statistically significant.
The limits of agreement for the replicate measurements (pooling of before and after measures) were assessed using a mixed model (estimation of the variance components) as described by Carstensen et al [24].
The diagnostic accuracy of COesCCO change was expressed as the area under the corresponding receiver operating characteristic curve (ROC). The optimal threshold value was then selected using this ROC curve. Sensitivity (Se), specificity (Sp) and likelihood ratios (LR) were estimated. Confidence intervals of proportions (Se, Sp) were assessed using the binomial exact method.
All statistical analyses were performed using STATA software version 12.0 (Statacorp, College Station, TX, USA).
Results
Study population
From a total of 30 consecutive potentially eligible septic shock patients admitted from June 2013 to November 2013, 5 patients were excluded from the study for calibration problems (3 because the VTIao was not correctly aligned and 2 because pulse wave transit time was not obtained). A total of 25 patients were included in the final analysis. Average age was 64 ± 15 years, most were men (Table 1). Average SAPS II and SOFA scores were 55 ± 21.3 and 13 ± 2, respectively. The most common origin of sepsis was community-acquired pneumonia (60%). ICU mortality was 36%.
Comparison between COesCCO and COTTE
Mean cardiac output at T0 was 5.8±1.35 L/min by esCCO and 5.27±1.17 L/min as assessed by COTTE. At T1, the respective values were 6.63 ± 1.57 L/min for esCCO and 6.10±1.29 L/min for COTTE.
Fig 1A and 1B show the linear correlations between COesCCO and COTTE at T0 and T1, respectively r2 = 0.71 p < 0.0001 and r2 = 0.81 p < 0.0001.
(A) Immediately before administration of 500 mL of crystalloid solution over a 20-minute period (T0). (B) Immediately after administration of 500 mL of crystalloid solution over a 20-minute period (T1).
Fig 2 shows the linear correlation between COesCCO and COTTE change (from T0 to T1), r2 = 0.70 p < 0.0001.
The blue line represents the line of perfect concordance; the brown line is the reduced major axis, representing the line of best fit.
The bias and limits of agreement (95% confidence interval) between COesCCO and COTTE at T0 and T1 were respectively -0.60 L/min (-2.05 to 0.85 L/min) and -0.54 L/min (-1.92 to 0.84 l/min.)
The average bias of agreement (95% confidence interval) between T0 and T1 measured by COesCCO and COTTE was 0.2 L/min (-2 to + 2.4 L/min) (Fig 3).
Overall, 12 patients were classified as responders, and 13 patients were classified as non-responders by the reference echocardiography method. There was no significant difference in terms of norepinephrine dose between responders and non-responders at baseline (0.8±0.3 vs 0.7±0.4 gamma/kg/min respectively).
A threshold value of 11% increase in COesCCO was found to discriminate responders from non-responders with a sensitivity of 83% (95% CI, 0.52–0.98) and a specificity of 77% (95% CI, 0.46–0.95). The area under the ROC curve was 0.84 (95% CI, 0.69–0.99). The positive LR was 3.6 and the negative LR was 0.21 (Fig 4).
Intra-observer variability with COTTE was 7±4% and the inter-observer variability (MF and JB) was 7±5%.
Discussion
The main finding of this study is that esCCO can detect rapid changes in CO in adult patients in the early phase of septic shock. The changes in CO determined by the non-invasive esCCO device were correlated with CO as measured by the reference method, namely transthoracic echocardiography (COTTE) (r² = 0.70) after rapid fluid infusion. We also found that a threshold value of 11% increase in COesCCO after fluid infusion adequately discriminated responders from non-responders with high sensitivity (83%) and specificity (77%).
Recent studies compared CO measured by esCCO (COesCCO) and by thermodidution [11, 25]. The authors reported a good correlation between the two methods, with a small bias (0.04 to 0.13 L/min) and limits of agreement -1.15 and 1.61 L/min. Comparison of esCCO vs CO measured by transthoracic echocardiography (COTTE) has yielded discordant results. Mansencal et al[26] demonstrated the best correlation (R2 = 0.92) in a patients hospitalized in coronary care unit, with limits of agreement from -0.60 to +0.68 L/min. Bataille et al[27] also reported a good level of correlation between COesCCO and COTTE (R2 = 0.61) in a patients hospitalized in the Intensive Care Unit (ICU), but with a bias of -1.6 L/min and limits of agreement from -4.7 to + 1.5 L/min.
Appropriate early management of patients with septic shock is vital in view of the high mortality that persists in this patient population[2, 28–29]. The objectives of therapy in hemodynamic terms over the first 6 hours were recently underlined in an international consensus conference[30], namely to obtain central venous pressure of 8–12 mmHg, mean arterial pressure ≥ 65 mmHg, urinary output ≥ 0.5 mL/kg/h and superior vena cava oxygenation saturation (Scvo2) or mixed venous oxygen saturation (Svo2) of 70% or 65%, respectively. Measurement of central venous pressure or mixed venous oxygen saturation requires insertion of a central catheter (either by the jugular or subclavian route), which may not always be possible in critical care patients, particularly during the initial acute management. Similarly, the presence of a central line does not allow evaluation of cardiac output or response to vascular filling, once hemodynamic objectives have been achieved[9]. For these reasons, several indices have been developed in recent years to predict the response to fluid challenge, both static and dynamic measures that are based on the variations in pulsed pressure during mechanical ventilation. These indices are much more reliable, but require insertion of an arterial catheter, and measures should be performed in strict conformity with the conditions in which they were validated [18, 31]. In addition to these restrictions on the conditions for valid use, the presence of diastolic dysfunction can also invalidate these predictive indices [32]. Conversely, diastolic dysfunction does not affect the evaluation of cardiac output by Doppler, or the estimation of filling pressures.
Current trends in the management of critical care patients tend towards increasing use of non-invasive methods. For many years, there has been a focus on validating non-invasive techniques both to better understand pathophysiological mechanisms, and to improve management of hemodynamic instability[17, 33–34].
According to Vincent et al[6], a good hemodynamic monitoring system should provide accurate and reproducible measurements of relevant variables, provide interpretable data, be easy to use, readily available, and operator-independent, have a rapid response time, cause no harm, be cost-effective and provide information that can be used to guide therapy. The esCCO technique appears to fulfil the majority of these criteria, although the paucity of literature data precludes drawing any firm conclusions.
In our study, the response to fluid challenge was assessed using VTIao, which has previously been shown to be a valid indicator in septic shock patients[34]. VTIao is easy to measure, reproducible, and widely used in clinical practice. TTE is accepted as a reliable method for CO measurement in clinical practice and has been validated in comparison with thermodilution by several studies[13–14, 35–36].
In their study published in 2012, Bataille et al[27] reported a correlation coefficient of 0.61 between COTTE and COesCCO (p <0.001). The Bland and Altman analysis corrected for repeated measures showed a bias of -1.6 L/min, with limits of agreement from -4.7 to + 1.5 L/min, and a percentage error (2 SD/mean CO) of 49%. The correlation for CO changes was significant (R = 0.63, p< 0.001). In our study, the corresponding results were slightly better in terms of bias and limits of agreement, respectively -0.60 L/min (-2.05 to 0.85 L/min) at T0 and -0.54 L/min (-1.92 to 0.84 L/min.) at T1. The average bias of agreement (95% CI) between T0 and T1 measured by COesCCO and COTTE was 0.2 L/min (-2 to + 2.4 L/min). Several potential criticisms of the study by Bataille et al have been put forward[37], including the assertion that the authors did not standardize the timing of CO measurement, the clinical conditions and the reasons why the operator chose to measure CO. The study was also reproached for not having given any information about the physician who performed TTE or any information about patient characteristics. In designing our own study, we took these remarks into account to avoid similar pitfalls, and included a homogeneous population of patients, with a detailed protocol developed by a team with long experience in this area[17, 33–34], aided by an expert statistician. In the context of coronary care, Mansencal et al[26] observed similar results to those of our study, with a correlation coefficient between COTTE and COesCCO of 0.92 (p <0.001) in their study. Furthermore, they reported excellent agreement between CO as determined by esCCO and echocardiography, with a 95% confidence interval of -0.60 to +0.60 L/min.
In a meta-analysis of studies comparing techniques of cardiac output measurement using bias and precision statistics, Critchley et al[38] concluded that acceptance of a new technique should rely on limits of agreement of up to 30%. Other studies have been performed using other thermodilution techniques, such as intermittent bolus thermodilution CO measurement [11, 25], pulse contour CO measurements using the FloTrac system (Edwards Lifesciences)[39] or the PiCCO pulse contour system (Pulsion Medical systems)[40], with reportedly good correlation and good accuracy between these techniques[41].
In routine practice, the most important element is the system’s ability to record variations occurring in line with the patient’s hemodynamic condition, or during vascular filling. In our study, the average bias of agreement (95%CI) between T0 and T1 (after fluid infusion) as measured by COesCCO and COTTE was 0.2 L/min (-2 to + 2.4 L/min), in line with the findings of Bataille et al[27] and Mansencal et al[26].
Norepinephrine is a powerful vasoconstrictor that is frequently used to restore arterial pressure in circulatory failure. However, it is known to increase cardiac preload and cardiac index and reduce cardiac preload dependency, as recently shown by Monnet et al[31]. In our study, all patients were receiving norepinephrine at inclusion, although there was no significant difference between responders and non-responders in the dose. Nonetheless, we cannot exclude the possibility that high doses of norepinephrine in some patients may have masked cardiac preload dependency.
We believe that our study findings yield several arguments in support of the use of esCCO. Firstly, esCCO is a simple, non-invasive technique to evaluate hemodynamic status. Secondly, training ICU physicians in the use and interpretation of esCCO is quick, easy, and inexpensive. Thirdly, removing the need for central and arterial catheter placement represents a considerable time gain and avoids an additional puncture site that could be a potential port of entry for bacteria. Lastly, optimizing hemodynamic status as early as possible could help avoid progression to multiorgan failure, thereby reducing morbidity and mortality in patients with severe sepsis and septic shock.
However, some limitations of this work should be acknowledged. Firstly, this is a pilot study with a relatively small sample size, conducted in septic patients. This may explain the poor lower range of the CIs for sensitivity and specificity for distinguishing between responders and non-responders. Furthermore, although the use of esCCO has been validated in other contexts, this is the first study to examine the use of this technique in septic shock patients in the ICU, and therefore, no external validation in a larger population could be performed. Secondly, 2 patients were excluded because no esCCO value could be displayed. This is likely because it is difficult to obtain a reliable signal in patients who remain unstable with cold extremities. Thirdly, the patients must be have regular cardiac rhythm, which may limit the applicability of the device. However, our findings offer a proof of concept and suggest that further research would be appropriate.
In conclusion, this study shows a strong correlation between the two methods of measuring CO, namely esCCO and echocardiography, and the changes in CO following fluid infusion in ICU patients, with clinically acceptable limits of agreement. Therefore, we purport that the esCCO monitor can be recommended for critically ill patients.
Acknowledgments
The authors thank all the paramedical of Belfort ICU hospital for help during the implementation of the study. The authors thank Fiona Ecarnot (EA3920, University Hospital Besancon, France) for translation and editorial assistance.
Author Contributions
Conceived and designed the experiments: MF RB JPQ. Performed the experiments: MF SG RT JB RB JPQ. Analyzed the data: MF SA RB JPQ. Contributed reagents/materials/analysis tools: MF SA SG RT JB RB JPQ. Wrote the paper: MF SA SG RT JB RB JPQ.
References
- 1. Hollenberg SM, Ahrens TS, Annane D, Astiz ME, Chalfin DB, Dasta JF, et al. Practice parameters for hemodynamic support of sepsis in adult patients: 2004 update. Crit Care Med. 2004;32(9):1928–48. Epub 2004/09/03. doi: 00003246-200409000-00018 [pii]. pmid:15343024.
- 2. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–77. Epub 2002/01/17. pmid:11794169.
- 3. Krausz MM, Perel A, Eimerl D, Cotev S. Cardiopulmonary effects of volume loading in patients in septic shock. Ann Surg. 1977;185(4):429–34. Epub 1977/04/01. pmid:843138; PubMed Central PMCID: PMC1396139.
- 4. Wang P, Zhou M, Rana MW, Ba ZF, Chaudry IH. Differential alterations in microvascular perfusion in various organs during early and late sepsis. Am J Physiol. 1992;263(1 Pt 1):G38–43. Epub 1992/07/01. pmid:1636714.
- 5. Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA. Fluid resuscitation in septic shock: a positive fluid balance and elevated central venous pressure are associated with increased mortality. Crit Care Med. 2011;39(2):259–65. Epub 2010/10/27. pmid:20975548.
- 6. Vincent JL, Rhodes A, Perel A, Martin GS, Della Rocca G, Vallet B, et al. Clinical review: Update on hemodynamic monitoring—a consensus of 16. Crit Care. 2011;15(4):229. Epub 2011/09/03. cc10291 [pii] pmid:21884645; PubMed Central PMCID: PMC3387592.
- 7. Shah MR, Hasselblad V, Stevenson LW, Binanay C, O'Connor CM, Sopko G, et al. Impact of the pulmonary artery catheter in critically ill patients: meta-analysis of randomized clinical trials. JAMA. 2005;294(13):1664–70. Epub 2005/10/06. 294/13/1664 [pii] pmid:16204666.
- 8. Michard F, Boussat S, Chemla D, Anguel N, Mercat A, Lecarpentier Y, et al. Relation between respiratory changes in arterial pulse pressure and fluid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care Med. 2000;162(1):134–8. Epub 2000/07/21. pmid:10903232.
- 9. Osman D, Ridel C, Ray P, Monnet X, Anguel N, Richard C, et al. Cardiac filling pressures are not appropriate to predict hemodynamic response to volume challenge. Crit Care Med. 2007;35(1):64–8. Epub 2006/11/03. pmid:17080001.
- 10. Ishihara H, Sugo Y, Tsutsui M, Yamada T, Sato T, Akazawa T, et al. 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(6):465–71. Epub 2012/08/03. pmid:22854918; PubMed Central PMCID: PMC3494869.
- 11. Yamada T, Tsutsui M, Sugo Y, Sato T, Akazawa T, Sato N, et al. 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(1):82–7. Epub 2012/04/03. ANE.0b013e31824e2b6c [pii] pmid:22467885.
- 12. Sugo Y, Ukawa T, Takeda S. A non-invasive continuous cardiac output measurement method utilizing ECG and spO2 pulse wave. The Japanese Journal of Medical Instrumentation. 2005;75(2):63–9.
- 13. Lewis JF, Kuo LC, Nelson JG, Limacher MC, Quinones MA. Pulsed Doppler echocardiographic determination of stroke volume and cardiac output: clinical validation of two new methods using the apical window. Circulation. 1984;70(3):425–31. Epub 1984/09/01. pmid:6744546.
- 14. McLean AS, Needham A, Stewart D, Parkin R. Estimation of cardiac output by noninvasive echocardiographic techniques in the critically ill subject. Anaesth Intensive Care. 1997;25(3):250–4. Epub 1997/06/01. pmid:9209605.
- 15. Feissel M, Maizel J, Robles G, Badie J, Faller JP, Slama M. Clinical relevance of echocardiography in acute severe dyspnea. J Am Soc Echocardiogr. 2009;22(10):1159–64. Epub 2009/10/06. S0894-7317(09)00624-5 [pii] pmid:19801305.
- 16. Sharwood-Smith G, Bruce J, Drummond G. Assessment of pulse transit time to indicate cardiovascular changes during obstetric spinal anaesthesia. Br J Anaesth. 2006;96(1):100–5. Epub 2005/11/01. aei266 [pii] pmid:16257996.
- 17. Feissel M, Teboul JL, Merlani P, Badie J, Faller JP, Bendjelid K. Plethysmographic dynamic indices predict fluid responsiveness in septic ventilated patients. Intensive Care Med. 2007;33(6):993–9. Epub 2007/03/30. pmid:17393139.
- 18. Bendjelid K, Romand JA. Fluid responsiveness in mechanically ventilated patients: a review of indices used in intensive care. Intensive Care Med. 2003;29(3):352–60. Epub 2003/01/22. pmid:12536268.
- 19. Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest. 2002;121(6):2000–8. Epub 2002/06/18. pmid:12065368.
- 20. Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA. 1993;270(24):2957–63. Epub 1993/12/22. pmid:8254858.
- 21. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22(7):707–10. Epub 1996/07/01. pmid:8844239.
- 22. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307–10. Epub 1986/02/08. doi: S0140-6736(86)90837-8 [pii]. pmid:2868172.
- 23. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics. 1989;45(1):255–68. Epub 1989/03/01. pmid:2720055.
- 24. Carstensen B, Simpson J, Gurrin LC. Statistical models for assessing agreement in method comparison studies with replicate measurements. Int J Biostat. 2008;4(1):Article 16. Epub 2008/01/01. pmid:22462118.
- 25. Tsutsui M, Araki Y, Masui K, Kazama T, Sugo Y, Archer TL, et al. 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(6):1307–12. Epub 2013/11/22. [pii]. pmid:24257379.
- 26. Mansencal N, Delobelle J, Balagny P, Badie J, Ihaddaden M, Arslan M, et al. Usefulness of a noninvasive cardiac output measurement using pulse wave transit time in coronary care unit. Int J Cardiol. 2013;168(4):4411–2. Epub 2013/05/30. S0167-5273(13)00941-8 [pii] pmid:23714596.
- 27. Bataille B, Bertuit M, Mora M, Mazerolles M, Cocquet P, Masson B, et al. Comparison of esCCO and transthoracic echocardiography for non-invasive measurement of cardiac output intensive care. Br J Anaesth. 2012;109(6):879–86. Epub 2012/08/22. aes298 [pii] pmid:22907340.
- 28. Quenot JP, Binquet C, Kara F, Martinet O, Ganster F, Navellou JC, et al. The epidemiology of septic shock in French intensive care units: the prospective multicenter cohort EPISS study. Crit Care. 2013;17(2):R65. Epub 2013/04/09. cc12598 [pii] pmid:23561510; PubMed Central PMCID: PMC4056892.
- 29. Waechter J, Kumar A, Lapinsky SE, Marshall J, Dodek P, Arabi Y, et al. Interaction between fluids and vasoactive agents on mortality in septic shock: a multicenter, observational study*. Crit Care Med. 2014;42(10):2158–68. Epub 2014/07/30. pmid:25072761.
- 30. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580–637. Epub 2013/01/29. [pii]. pmid:23353941.
- 31. Monnet X, Dres M, Ferre A, Le Teuff G, Jozwiak M, Bleibtreu A, et al. Prediction of fluid responsiveness by a continuous non-invasive assessment of arterial pressure in critically ill patients: comparison with four other dynamic indices. Br J Anaesth. 2012;109(3):330–8. Epub 2012/06/28. aes182 [pii] pmid:22735299.
- 32. Shim JK, Song JW, Song Y, Kim JH, Kang HM, Kwak YL. Pulse pressure variation is not a valid predictor of fluid responsiveness in patients with elevated left ventricular filling pressure. J Crit Care. 2014;29(6):987–91. Epub 2014/09/14. S0883-9441(14)00280-9 [pii] pmid:25216949.
- 33. Feissel M, Kalakhy R, Banwarth P, Badie J, Pavon A, Faller JP, et al. Plethysmographic variation index predicts fluid responsiveness in ventilated patients in the early phase of septic shock in the emergency department: a pilot study. J Crit Care. 2013;28(5):634–9. Epub 2013/05/21. S0883-9441(13)00072-5 [pii] pmid:23683565.
- 34. Feissel M, Michard F, Mangin I, Ruyer O, Faller JP, Teboul JL. Respiratory changes in aortic blood velocity as an indicator of fluid responsiveness in ventilated patients with septic shock. Chest. 2001;119(3):867–73. Epub 2001/03/13. pmid:11243970.
- 35. Jardin F, Valtier B, Beauchet A, Dubourg O, Bourdarias JP. Invasive monitoring combined with two-dimensional echocardiographic study in septic shock. Intensive Care Med. 1994;20(8):550–4. Epub 1994/11/01. pmid:7706566.
- 36. Marcelino P, Germano N, Marum S, Fernandes AP, Ribeiro P, Lopes MG. Haemodynamic parameters obtained by transthoracic echocardiography and Swan-Ganz catheter: a comparative study in liver transplant patients. Acta Med Port. 2006;19(3):197–205. Epub 2007/01/20. pmid:17234080.
- 37. Faraoni D, Barvais L. Correlation between esCCO and transthoracic echocardiography in critically ill patients. Br J Anaesth. 2013;110(1):139–40. Epub 2012/12/14. aes437 [pii] pmid:23236108.
- 38. 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(2):85–91. Epub 2003/02/13. pmid:12578081.
- 39. Della Rocca G, Costa MG, Chiarandini P, Bertossi G, Lugano M, Pompei L, et al. Arterial pulse cardiac output agreement with thermodilution in patients in hyperdynamic conditions. J Cardiothorac Vasc Anesth. 2008;22(5):681–7. Epub 2008/10/17. S1053-0770(08)00047-5 [pii] pmid:18922423.
- 40. Godje O, Hoke K, Goetz AE, Felbinger TW, Reuter DA, Reichart B, et al. Reliability of a new algorithm for continuous cardiac output determination by pulse-contour analysis during hemodynamic instability. Crit Care Med. 2002;30(1):52–8. Epub 2002/03/21. pmid:11902287.
- 41. Ishihara H, Okawa H, Tanabe K, Tsubo T, Sugo Y, Akiyama T, et al. A new non-invasive continuous cardiac output trend solely utilizing routine cardiovascular monitors. J Clin Monit Comput. 2004;18(5–6):313–20. Epub 2005/06/17. pmid:15957621.