Secondary Logo

Journal Logo

The Influence of Acute Pulmonary Hypertension on Cardiac Output Measurements

Calibrated Pulse Contour Analysis, Transpulmonary and Pulmonary Artery Thermodilution Against a Modified Fick Method in an Animal Model

Kutter, Annette P. N. Dr. Med. Vet.*; Mosing, Martina Dr. Med. Vet.*; Hartnack, Sonja Dr. Med. Vet.; Raszplewicz, Joanna Med. Vet.*; Renggli, Martina Med. Vet.*; Mauch, Jacqueline Y. Dr. Med.; Hofer, Christoph K. Prof. Dr. Med.§

doi: 10.1213/ANE.0000000000000655
Technology, Computing, and Simulation: Research Report
Free

BACKGROUND: In critically ill patients with significant pulmonary hypertension (PH), close perioperative cardiovascular monitoring is mandatory, considering the increased morbidity and mortality in this patient group. Although the pulmonary artery catheter is still the standard for the diagnosis of PH, its use to monitor cardiac output (CO) in patients with PH is decreasing as a result of increased morbidity and possible influence of tricuspid regurgitation on the measurements. However, continuous CO measurement methods have never been evaluated under PH regarding their agreement and trending ability. In this study, we evaluated the influence of acute PH and different CO states on transpulmonary thermodilution (TPTD) and calibrated pulse contour analysis (PiCCO; both assessed with PiCCO plus™), intermittent pulmonary artery thermodilution (PATD), and continuous thermodilution (CCO) compared with a modified Fick method (FICK) in an animal model.

METHODS: Nine healthy pigs were studied under anesthesia. PH of 25 and 40 mm Hg (by administration of the thromboxane analog U46619), CO decreases, and CO increases were induced to test the different CO measurement techniques over a broad range of hemodynamic situations. Before each step, a new baseline data set was collected. CO values were compared using Bland-Altman analysis; trending abilities were assessed via concordance and polar plot analysis. The influence of pulmonary pressure on CO measurements was analyzed using linear mixed models.

RESULTS: A mean bias of -0.26 L/min with prediction intervals of −0.88 to 1.4 L/min was measured between TPTD and FICK. Their concordance rate was 100% (94%–100% confidence interval), and the mean polar angle −3° with radial limits of agreement of ±28° indicated good trending abilities. PATD compared with FICK also showed good trending ability. Comparisons of PiCCO and CCO versus FICK revealed low agreement and poor trending results with concordance rates of 84% (71%–93%) and 88% (74%–95%), mean polar angles from −17° and −19°, and radial limits of agreement of ±45° and 40°. Pulmonary pressures influenced only the difference between FICK and PiCCO, as assessed by linear mixed models.

CONCLUSIONS: TPTD compared with FICK was able to track all changes induced during the study period, including those by PH. It yielded better agreement than PATD both compared with FICK. PiCCO and CCO were not mapping all changes correctly, and when used clinically in unstable patients, regular controls with intermittent techniques are required. Acute pharmacologically induced PH did influence the difference between FICK and PiCCO.

Published ahead of print March 4, 2015

From the *Section of Anesthesiology, Equine Department and Section of Epidemiology, Vetsuisse Faculty of the University of Zurich, Zurich, Switzerland; Department of Anesthesiology, Kantonsspital Luzern, Luzern, Switzerland; and §Institute of Anesthesiology and Intensive Care Medicine, Triemli City Hospital, Zurich, Switzerland.

Accepted for publication December 9, 2014.

Published ahead of print March 4, 2015

Funding: This study was supported by Axon Lab Switzerland and funded by a grant from the Forschungskredit of the University of Zurich.

Conflict of Interest: See Disclosures at the end of the article.

Reprints will not be available from the authors.

Address correspondence to Annette P. N. Kutter, DVM, Section of Anesthesiology, Equine Department, Vetsuisse Faculty of the University of Zurich, Winterthurerstrasse 260, 8057 Zurich, Switzerland. Address e-mail to akutter@vetclinics.uzh.ch.

Patients with pulmonary hypertension (PH) have an increased morbidity and mortality in the perioperative setting of both cardiac and noncardiac surgery.1–3 Although close monitoring of cardiovascular function is mandatory in patients with PH,4 the perioperative use of the pulmonary artery catheter has been questioned as a result of an increased risk of arrhythmias and vessel ruptures in this patient group.5,6 Furthermore, pulmonary artery thermodilution (PATD) for cardiac output (CO) measurement may be distorted by tricuspid regurgitation.7 The severity of PH does not predict mortality, but measuring CO repeatedly and assessing the function of the right heart are important prognostic factors and measure of therapeutic success.4

Transpulmonary thermodilution (TPTD) may be a valuable alternative to PATD for CO measurements in the setting of PH and may overcome some problems related to PATD in PH. Using TPTD, the PATD catheter is replaced typically by a femoral or long radial artery thermodilution catheter.8 A cold fluid bolus is injected in the jugular vein or the right atrium by means of a central venous catheter. The temperature change after the administration of cold fluid is not detected in the pulmonary artery but in the central systemic circulation after the fluid bolus has passed both sides of the heart and the lung. Therefore, this CO measurement method is called “transpulmonary.” It is used in the PiCCO plus™ monitoring system (Pulsion Medical Systems, Munich, Germany) to calibrate the continuous pulse contour cardiac analysis (PiCCO). Thus, this system can be used at the bedside to assess CO.

However, TPTD and PiCCO have never been compared with a reference technique under increased pulmonary pressure conditions, and their accuracy and trending ability in this setting are not known. Differences between methods can be induced through bias, different variability, or different susceptibility to extraneous factors,9 such as changes of systemic or pulmonary arterial pressures. For clinical CO monitoring, however, the accurate mapping of changes over time may be more important than the absolute CO value per se, to detect hemodynamic instability or to assess results of therapeutic measures.10,11

Continuous thermodilution (CCO) measures CO with a modified pulmonary artery catheter equipped with a thermal coil, which is positioned in the right ventricle. The coil generates heat pulses and by repeatedly doing so the CO computer can construct a thermodilution washout curve and estimate CO automatically. The method has been evaluated in many studies with good agreement for slow hemodynamic changes, but during fast changes, a time delay from 5 to 20 minutes to track changes has been reported.12

The aim of this study was to evaluate the influence of acute PH and different CO states on CO measurements by TPTD, PiCCO, PATD, and CCO compared with a modified Fick method (FICK) in an animal model. The hypothesis was that changes in lung perfusion would influence TPTD and PiCCO less than PATD and continuous CO assessment by thermodilution (CCO) using the pulmonary artery catheter.

Back to Top | Article Outline

METHODS

General

This study was approved by the Cantonal Veterinary Office of Zurich (176/2011). Nine healthy landrace pigs aged 62 ± 1 day (mean ± SD) and weighing 25.7 ± 1.9 kg were studied. Although the sample size of 10 pigs was chosen on the basis of previous studies investigating agreement between CO methods in animal models,13,14 only 9 pigs were included in the current study because of missing FICK CO data in 1 pig.

The pigs were premedicated with midazolam 1 mg/kg (Dormicum; Roche Pharma, Reinach, Switzerland) and ketamine 15 mg/kg (Narketan; Vetoquinol, Ittigen, Switzerland) intramuscularly. Anesthesia was induced with propofol (Fresenius Kabi AG, Oberdorf, Switzerland) and maintained with midazolam 0.5 mg/kg/h, propofol 4 mg/kg/h, fentanyl 20 μg/kg/h (Sintenyl, Sintetica SA, Mendrisio, Switzerland), and pancuronium 0.2 mg/kg/h (Pavulon; Essex Chemie AG, Luzern, Switzerland). Ringer’s lactate solution at a rate of 3 mL/kg/h (Ringer-Laktat; Fresenius Kabi) was infused during the experiment, and temperature was kept constant at 38.5 ± 0.3°C by means of a forced-air warming blanket (Bair Hugger; Carbamed AG, Liebefeld, Switzerland).

After endotracheal intubation, the pigs were placed in dorsal recumbency, and mechanical ventilation (S/5 Advance Anesthesia Machine, Datex-Ohmeda Inc., Madison, WI) was started using a volume-controlled mode with the following settings: tidal volume of 6 mL/kg, positive end-expiratory pressure level of 7 cm H2O, inspiratory-to-expiratory ratio of 1:1, and FIO2 of 0.5. Respiratory rate was adjusted to keep the end-tidal partial pressure of CO2 at 40 ± 3 mm Hg.

Back to Top | Article Outline

Instrumentation

A 20-G catheter was placed in the carotid artery for blood gas sampling. A 7.5-Fr CCO pulmonary artery catheter (Swan-Ganz TD Catheter®, Edwards Lifesciences AG, Horw, Switzerland) was placed via an 8.5-Fr introducer in the right internal jugular vein using pressure guidance. A 4-Fr (22 cm) thermistor-tipped femoral artery catheter (Pulsiocath PVPK2014L22-A; Pulsion Medical Systems, Munich, Germany) was introduced into the right femoral artery. An additional catheter was placed in the right atrium via the right jugular vein for administration of U46619. All catheters were inserted by means of a surgical cut-down. Standard 3-lead electrocardiography, pulse oximetry, blood temperature, right atrial, pulmonary, and peripheral arterial pressures were displayed using a multiparameter monitor (GE BL850; Anandic Medical Systems AG, Feuerthalen, Switzerland). Bispectral index (Bispectral Index Monitor, Model A-2000; Aspect Medical System, Inc., Newton, MA) was applied for monitoring and adjustment of depth of anesthesia in pigs, as previously described.15 Expired CO2 analysis was performed with a mainstream CO2 infrared analyzer (NICO2; Respironics Inc., Murrysville, PA). Mixed venous and arterial blood samples were taken before each CO measurement, and blood gases as well as hemoglobin were immediately measured with a co-oximeter (IL GEM 4000; Axon Lab, Baden, Switzerland).

Back to Top | Article Outline

CO Measurements

To assess CO with a modified FICK, the CO2 production per minute was calculated from the area under the curve of the capnogram during inspiration and expiration multiplied with the respiratory rate.16 The CO2 production was divided by a respiratory quotient of 0.8 to receive O2 consumption. This value was divided by the arteriovenous O2 content difference to receive FICK CO values. Fick measurement was chosen as the reference method to exclude a potential influence of PH or tricuspid regurgitation that is reported for thermodilution methods.

Each intermittent thermodilution measurement consisted of 4 fluid boluses of 10 mL ice-cold 5% dextrose manually injected by the same operator into the proximal port of the pulmonary artery catheter located in the right atrium. Fluid bolus temperature was detected by 2 serial in-line sensors of the 2 CO computers (PATD: Vigilance I, Edwards Lifesciences AG; and TPTD: PiCCO plus, software version 6.0, Pulsion Medical Systems). The temperature changes in the pulmonary artery and in the distal abdominal aorta were recorded simultaneously with the 2 thermistor-tipped catheters, and CO was calculated based on the Stewart-Hamilton equation by the corresponding CO computer. All thermodilution curves were controlled visually for regular shape. The mean of 3 thermodilution measurements within a variation of 10% was used to calculate PATDmean and the 3 corresponding TPTD to calculate TPTDmean, which were used for statistical analysis.

The PiCCO plus monitor automatically calibrates the pulse contour measurement PiCCO after every new set of TPTD bolus measurements. This calibration was allowed at each baseline measurement. During the step measurement, the automatic recalibration was not allowed. To prevent this calibration, the TPTD bolus measurements obtained during the steps were deleted manually in the monitors’ memory. This enabled us to analyze the trending ability over each intervention step (PH25, PH40, COup, and COdown) during a 1-hour period.

To avoid influence of the cold injectate, continuous PiCCO and CCO values were recorded directly before a new set of bolus measurements was performed. We measured pulmonary artery wedge pressures after CO measurements by occluding the pulmonary artery with the balloon on the tip of the pulmonary artery catheter. If the artery could not be occluded during the step PH25 or PH40 because of the dilation of the vessel, no repositioning of the catheter was attempted to avoid interference with concurrent dead-space measurements performed for another study in the same pigs.14 We report the global ejection fraction (GEF) as it is calculated by the PiCCO plus monitor: GEF (mL) = stroke volume/global end-diastolic volume (GEDV)/4. GEDV is assessed from the TPTD curve: CO × (mean transit time − exponential downslope time). GEDV and GEF assessments have been described in detail previously.17

Back to Top | Article Outline

Study Protocol

To assess the influence of changes of pulmonary and systemic pressures and the performance of different CO measurement methods, 4 individual steps were defined (Fig. 1). Two levels of PH, that is, PH25 (mean pulmonary artery pressure of 25 mm Hg) and PH40 (mean pulmonary artery pressure of 40 mm Hg) were induced by infusing 2 μg/kg/min U46619, a thromboxane analog, into the right atrium. The rate was increased or decreased to reach a stable target pressure of either a mean pulmonary artery pressure of 25 or 40 mm Hg. To increase pulmonary pressures by increasing lung perfusion, CO was increased by 50% from baseline (COup) by administration of 30 mL/kg Ringer’s lactate solution and dobutamine (Dobutrex; Teva Pharma AG, Basel, Switzerland) at an initial dose of 5 μg/kg/min. The dose of dobutamine was adapted to reach the target CO value. CO was decreased by 40% from baseline (COdown) by sodium nitroglycerine infusion 30 μg/kg/min (Perlinganit; UCB Pharma AG, Bulle, Switzerland) and esmolol 500 μg/kg/min (OrPha Swiss GmbH, Switzerland). All steps were performed in random order in an individual pig. Baseline data were obtained before each protocol step. CO measurements were started after reaching stable conditions for 5 minutes of the predefined pressure or CO values. At least 30 minutes were allowed between steps, and the next step was initiated when cardiovascular variables were within 5% of the baseline values. At the completion of the experiment, the pigs were killed by IV administration of pentobarbital (Esconarkon, Streuli AG, Uznach, Switzerland).

Figure 1

Figure 1

Back to Top | Article Outline

Statistical Analysis

Cardiovascular changes were analyzed by paired t tests. Comparison of absolute CO values measured by the different techniques was performed by linear mixed effects models and by Bland-Altman analysis. To adjust for the small sample size, the t statistic for 9 pigs was calculated:

. To calculate the prediction intervals (limits of agreement), 2.431 was multiplied with the SD of the mean bias. The commonly used value of 1.96 can be used in studies with n ≥ 60. Mean percentage error was calculated as 100 × 2.431 × SD of the mean bias/mean CO of both methods to make the results comparable with former studies and between steps.18

For the comparison of trending capabilities of the different techniques, changes of CO (Δ CO) were calculated for each set of measurements, and concordance and polar plots were done as described by Critchley et al.,10,11 setting the exclusion zone at CO changes of ≤10% (polar plot) and ≤15% (concordance). To calculate the confidence interval (CI) of the concordance rate, generalized mixed effects models with pig as random effect and an intercept were fitted. The exponential of the β-coefficient of the intercept (± 2 × SE) corresponds to the odds (95% CI of odds). The odds, being equal to π/1 − π, were solved for π (and the corresponding lower and upper limits).

Linear mixed effects models also were used to assess whether a potential association between mean arterial or mean pulmonary artery blood pressure and the measured CO values was influenced by the method (interaction effect between method and mean arterial or mean pulmonary artery blood pressure). Using the triplicate measurements, we applied linear mixed effects models to test the difference of the variability of PATD and TPTD. Model selection (e.g., deciding which of the explanatory variables—mean arterial blood pressure, mean pulmonary artery blood pressure, effect of CO method, or an interaction term—should be included in the final model) was based on Akaike information criteria, which is a goodness-of-fit criterion model that allows a qualitative assessment of the variables with lower values indicating a better model fit.19 Data management was performed with Excel for Macintosh (Office X, Microsoft, Redmond, WA). Graphs were performed with Prism 6.0f (GraphPad Software, San Diego, CA) and SigmaPlot 10.0 (Systat Software, San Jose, CA). Linear mixed models and Bland-Altman calculations with repeated measures were performed with R20 and the packages nlme21 and MethComp.22

Back to Top | Article Outline

RESULTS

Hemodynamic data during the study period are displayed in Table 1. All techniques (FICK, PATD, CCO, TPTD, and PiCCO) detected the significant CO reduction (all P < 0.039) and CO increase (all P < 0.016) during step COdown and COup, respectively. Drug-induced PH of 25 mm Hg did not result in significant changes in CO assessed by any of the used methods (all P > 0.134), whereas PH 40 mm Hg induced a significant decrease in CO in all methods (all P < 0.025) but CCO (P = 0.236).

Table 1

Table 1

Linear mixed model and Bland-Altman results (Fig. 2, A–D) and percentage errors calculated for each step are presented in Table 2. The TPTD bolus method showed good trending ability compared with FICK and PATD with a concordance rate of 100% with 95% CI of 94% to 100%. The concordance rate of PATD against FICK was 96% (87%–100%). Both continuous methods showed limited trending abilities with concordance rates <90% (Table 3 and Fig. 3, A–D).

Table 2

Table 2

Table 3

Table 3

Figure 2

Figure 2

Figure 3

Figure 3

Mean pulmonary artery blood pressure did influence the difference between FICK and PiCCO measurements as assessed by linear mixed effects models. The differences between FICK and PATD, TPTD and CCO, respectively, were not influenced by mean pulmonary artery blood pressure. Both mean arterial and mean pulmonary artery blood pressure were influenced by the steps (interaction effect between mean arterial and mean pulmonary artery blood pressure and steps). Based on Akaike information criteria, there was no evidence that the triplicate measurements with PATD and TPTD differed in their variability.

Back to Top | Article Outline

DISCUSSION

In this study, the influence of 2 degrees of acute PH and CO decreases and increases on the accuracy and the trending ability of 4 CO methods compared with FICK were assessed in 9 pigs. Both intermittent methods TPTD and PATD yielded good trending ability compared with FICK. The bolus method TPTD showed better agreement than PATD with FICK CO measurements. Both continuous CO measurement methods, PiCCO and CCO, showed poor accuracy and limited trending ability compared with FICK. No influence of pulmonary artery pressure on the difference between all thermodilution methods PATD, TPTD, and CCO was revealed by linear mixed models. Only the difference between FICK and PiCCO was influenced by pulmonary artery pressure.

The results of this study regarding TPTD are in agreement with a variety of other studies in which authors evaluated the performance of TPTD. In these previous studies authors assessed CO by TPTD against PATD under different clinical conditions, but none of them involved PH. On the basis of the present favorable results, it can be argued that TPTD can reliably be used for CO measurements as an alternative to PATD avoiding a pulmonary artery catheter. Although adequate accuracy of CO measurement by thermodilution techniques in patients with PH23 or animal models of tricuspid regurgitation have been reported,24 unfortunately it has to be emphasized that thermodilution may under- or overestimate CO because of inadequate mixing or loss of indicator when PH and tricuspid regurgitation develop.7,25 Therefore, FICK has still to be considered the standard of CO measurement under these specific clinical conditions.4 The current data suggest better agreement and trending ability of TPTD than those of PATD when compared with FICK. Considering technical aspects of these techniques, PATD (measurement in the right heart) may be more influenced than TPTD (measurement of the total vascular bed from right atrium to the left ventricular outflow) by changing pulmonary pressures.

All pulse contour methods have been shown to be susceptible to changes in peripheral vascular resistance.13,26–28 The calibrated methods have the advantage that they can be recalibrated to better reflect any acute changes of vascular tone. The present study protocol involved major induction of hemodynamic changes and vasoactive drug application. As a result, inferior agreement of PiCCO and FICK measurements was observed, despite recalibration before any next hemodynamic change was performed. Still, other authors observed a comparable PiCCO performance without the influence of PH.11 The trending ability of PiCCO against FICK with concordance rates between 80% and 90% and polar plot results (radial limits of agreement of 45°) confirm these results. However, the applied linear mixed models did reveal an influence of mean pulmonary artery pressure on the difference between PiCCO and FICK measurements, opposed to no influence on all thermodilution measurement. This finding is unexpected and cannot be explained with the other findings of the current study. The device may still be used in its continuous mode when increased pulmonary artery pressures are present, considering the requirement of frequent recalibration to guarantee acceptable CO measurement results under unstable hemodynamic conditions.13,27

The PiCCO system combining TPTD and the pulse contour method PiCCO allows the additional assessment of different volumetric and functional hemodynamic variables. These variables may help with the early detection of hemodynamic instability and to initiate prompt and appropriate hemodynamic therapy. During PH, the right ventricle is dilated and can eject less flow through the lung vessels to the left ventricle. This seems to be correctly reflected by TPTD because the left ventricle cannot deliver more flow than it receives from the right side. The concept “pulmonary flow equals systemic flow” applies in this setting, even if acute increases in mean pulmonary artery pressure may influence the cardiac work of the ventricles differently. If a patient with PH is monitored with TPTD and PiCCO, no direct measurements of pulmonary artery pressures are possible. Because the severity of PH does not predict mortality,4 it may be more important to correctly monitor cardiac function than measure exact pulmonary artery pressures with a pulmonary artery catheter. Echocardiography can be used to intermittently estimate pressures and assess the function of both ventricles.

In the present study, comparing CCO with FICK resulted in low agreement and a limited trending ability. In this acute setting, CCO updates were apparently too slow to reliably track changes.12 To detect changes appropriately, the changes induced during and after the different hemodynamic steps would have to be stable for at least 10 to 20 minutes before each comparative hemodynamic assessment. High CCO values measured may be the result of a catheter-related problem. Clearly, the positions of the pulmonary artery catheter proximal and distal openings were verified using the detection of the typical pressure waveforms and the related changes during insertion. Still, the length of the heating wire might have been too long for these pigs that had a weight of 26 kg. However, linear mixed models indicate that PH does not influence the difference between CCO and FICK, and CCO measurements can probably be used in PH patients to monitor CO at the bedside considering a delay of up to 10 minutes12 and the typical limitations of thermodilution methods under PH.

Major limitations of the present study are primarily related to the animal model and the measurement techniques used during PH. Pharmacologically induced hypertension in animals with normal pulmonary vessels was used as a model for a variance of pathophysiologically different diseases causing PH.29 Typically, approximately 50% of the possible increase of pulmonary vascular resistance in patients with PH is mediated through reversible vasoconstriction,4 although individual variations of vascular involvement might be observed. Moreover, our model is not able to mimic vascular remodeling, thrombosis, or congestion from the left atrium. It can only reflect pulmonary arterial hypertension with low left atrial pressure, that is, the so-called “pre-capillary PH.”29 Therefore, no conclusion regarding CO measurement performance in patients with different classes of PH having vascular remodeling or left ventricular impairment as result of a failing right heart can be made.

Two bolus methods were assessed by analyzing the temperature changes induced by the same bolus of ice-cold fluid, a technique that should reduce the influence of temporal changes of CO and therefore decrease type I errors between the methods. Recalibration of PiCCO before each subsequent measurement period may also have reduced the error compared with longer calibration-free periods. Another source of error may be the IV administration of 30 mL/kg Ringer’s lactate solution to increase CO in a randomized order. Moreover, the comparison of a less-invasive method with a reference method requiring a pulmonary artery catheter possibly increased tricuspid regurgitation along the catheter, which may lead to a further loss of indicator in the thermodilution methods. To avoid this, TPTD may be compared with a CO measurement method such as Doppler flow measurement. However, the placement of a flowprobe around the pulmonary artery requires a thoracotomy, which was avoided in these animals because of concurrent studies.

The Fick measurement was chosen as the reference method to exclude a potential influence of PH or tricuspid regurgitation. We did not measure oxygen consumption during the study and therefore had to use a FICK. We divided the measured CO2 production by a respiratory quotient of 0.8 to receive O2 consumption. This estimated respiratory quotient may have introduced a source of error that cannot be quantified in the current study. Also, the presence of tricuspid regurgitation or other valvular abnormalities was not quantified by transesophageal echocardiography.

Our data suggest good trending ability of TPTD and PATD compared with FICK. TPTD yielded better results of agreement than PATD with FICK CO measurements. The trending abilities of the PiCCO and CCO were limited and when used clinically in unstable patients, regular controls with intermittent techniques are required. Acute pharmacologically induced PH did influence the difference between FICK and PiCCO.

Back to Top | Article Outline

DISCLOSURES

Name: Annette P. N. Kutter, Dr. Med. Vet.

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

Attestation: Annette P. N. Kutter approved the final manuscript, attests to the integrity of the original data and the analysis reported in this manuscript, and is the archival author.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Martina Mosing, Dr. Med. Vet.

Contribution: This author helped design the study, conduct the study, and collect the data.

Attestation: Martina Mosing approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Sonja Hartnack, Dr. Med. Vet.

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

Attestation: Sonja Hartnack analyzed the data.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Joanna Raszplewicz, Med. Vet.

Contribution: This author helped conduct the study and collect the data.

Attestation: Joanna Raszplewicz approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Martina Renggli, Med. Vet.

Contribution: This author helped conduct the study and collect the data.

Attestation: Martina Renggli approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Jacqueline Y. Mauch, Dr. Med.

Contribution: This author helped conduct the study.

Attestation: Jacqueline Y. Mauch approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Christoph K. Hofer, Prof. Dr. Med.

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

Attestation: Christoph K. Hofer approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Conflicts of Interest: Christoph K. Hofer received lecturing fees from Edwards Lifesciences and Pulsion Medical Systems.

This manuscript was handled by: Maxime Cannesson, MD, PhD.

Back to Top | Article Outline

REFERENCES

1. Reich DL, Bodian CA, Krol M, Kuroda M, Osinski T, Thys DM. Intraoperative hemodynamic predictors of mortality, stroke, and myocardial infarction after coronary artery bypass surgery. Anesth Analg. 1999;89:814–22
2. Kaw R, Pasupuleti V, Deshpande A, Hamieh T, Walker E, Minai OA. Pulmonary hypertension: an important predictor of outcomes in patients undergoing non-cardiac surgery. Respir Med. 2011;105:619–24
3. Lai HC, Lai HC, Wang KY, Lee WL, Ting CT, Liu TJ. Severe pulmonary hypertension complicates postoperative outcome of non-cardiac surgery. Br J Anaesth. 2007;99:184–90
4. Chemla D, Castelain V, Hervé P, Lecarpentier Y, Brimioulle S. Haemodynamic evaluation of pulmonary hypertension. Eur Respir J. 2002;20:1314–31
5. Harvey S, Harrison DA, Singer M, Ashcroft J, Jones CM, Elbourne D, Brampton W, Williams D, Young D, Rowan KPAC-Man Study Collaboration. . Assessment of the clinical effectiveness of pulmonary artery catheters in management of patients in intensive care (PAC-Man): a randomised controlled trial. Lancet. 2005;366:472–7
6. Hoeper MM, Lee SH, Voswinckel R, Palazzini M, Jais X, Marinelli A, Barst RJ, Ghofrani HA, Jing ZC, Opitz C, Seyfarth HJ, Halank M, McLaughlin V, Oudiz RJ, Ewert R, Wilkens H, Kluge S, Bremer HC, Baroke E, Rubin LJ. Complications of right heart catheterization procedures in patients with pulmonary hypertension in experienced centers. J Am Coll Cardiol. 2006;48:2546–52
7. Heerdt PM, Blessios GA, Beach ML, Hogue CW. Flow dependency of error in thermodilution measurement of cardiac output during acute tricuspid regurgitation. J Cardiothorac Vasc Anesth. 2001;15:183–7
8. Orme RM, Pigott DW, Mihm FG. Measurement of cardiac output by transpulmonary arterial thermodilution using a long radial artery catheter. A comparison with intermittent pulmonary artery thermodilution. Anaesthesia. 2004;59:590–4
9. Hartnack S. Issues and pitfalls in method comparison studies. Vet Anaesth Analg. 2014;41:227–32
10. 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
11. 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
12. Poli de Figueiredo LF, Malbouisson LM, Varicoda EY, Carmona MJ, Auler JO Jr, Rocha e Silva M. Thermal filament continuous thermodilution cardiac output delayed response limits its value during acute hemodynamic instability. J Traum. 1999;47:288–93
13. Piehl MD, Manning JE, McCurdy SL, Rhue TS, Kocis KC, Cairns CB, Cairns BA. Pulse contour cardiac output analysis in a piglet model of severe hemorrhagic shock. Crit Care Med. 2008;36:1189–95
14. Axiak Flammer SM, Critchley LA, Weber A, Pirbodaghi T, Brinks H, Vandenberghe S. Reliability of lithium dilution cardiac output in anaesthetized sheep. Br J Anaesth. 2013;111:833–9
15. Martín-Cancho MF, Carrasco-Jiménez MS, Lima JR, Ezquerra LJ, Crisóstomo V, Usón-Gargallo J. Assessment of the relationship of bispectral index values, hemodynamic changes, and recovery times associated with sevoflurane or propofol anesthesia in pigs. Am J Vet Res. 2004;65:409–16
16. Mosing M, Kutter APN, Iff S, Raszplewicz J, Mauch J, Bohm SH, Tusman G. The effects of cardiac output and pulmonary arterial hypertension on volumetric capnography derived-variables during normoxia and hypoxia. J Clin Monit Comput. 2015;29:187–96
17. Combes A, Berneau JB, Luyt CE, Trouillet JL. Estimation of left ventricular systolic function by single transpulmonary thermodilution. Intensive Care Med. 2004;30:1377–83
18. 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
19. Akaike H. A new look at the statistical model identification. IEEE Trans Autom Control. 1974;19:716–23
20. R Core Team. R: A Language and Environment for Statistical Computing. 2014 Vienna, Austria R Foundation for Statistical Computing http://www.R-project.org/. Accessed February 16, 2015
21. Pinheiro J, Bates D, DebRoy S, Sarkar DR Core Team (2015). R Core Team (2015). nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-119. http://CRAN.R-project.org/package=nlme. Accessed February 16, 2015
22. Carstensen B, Gurrin L, Ekstrom C, Figurski M MethComp: functions for analysis of agreement in method comparison studies. 2013 http://CRAN.R-project.org/package=MethComp. Accessed February 16, 2015
23. Hoeper MM, Maier R, Tongers J, Niedermeyer J, Hohlfeld JM, Hamm M, Fabel H. Determination of cardiac output by the Fick method, thermodilution, and acetylene rebreathing in pulmonary hypertension. Am J Respir Crit Care Med. 1999;160:535–41
24. Buffington CW, Nystrom EU. Neither the accuracy nor the precision of thermal dilution cardiac output measurements is altered by acute tricuspid regurgitation in pigs. Anesth Analg. 2004;98:884–90
25. Nishikawa T, Dohi S. Errors in the measurement of cardiac output by thermodilution. Can J Anaesth. 1993;40:142–53
26. Hadian M, Kim HK, Severyn DA, Pinsky MR. Cross-comparison of cardiac output trending accuracy of LiDCO, PiCCO, FloTrac and pulmonary artery catheters. Crit Care. 2010;14:R212
27. Bein B, Meybohm P, Cavus E, Renner J, Tonner PH, Steinfath M, Scholz J, Doerges V. The reliability of pulse contour-derived cardiac output during hemorrhage and after vasopressor administration. Anesth Analg. 2007;105:107–13
28. Bektas RN, Kutter AP, Hartnack S, Jud RS, Schnyder M, Matos JM, Bettschart-Wolfensberger R. Evaluation of a minimally invasive non-calibrated pulse contour cardiac output monitor (FloTrac/Vigileo) in anaesthetized dogs. Vet Anaesth Analg. 2012;39:464–71
29. Hoeper MM, Bogaard HJ, Condliffe R, Frantz R, Khanna D, Kurzyna M, Langleben D, Manes A, Satoh T, Torres F, Wilkins MR, Badesch DB. Definitions and diagnosis of pulmonary hypertension. J Am Coll Cardiol. 2013;62:D42–50
© 2015 International Anesthesia Research Society