Pulse contour analysis after normothermic cardiopulmonary bypass in cardiac surgery patients

INTRODUCTION: Monitoring of the cardiac output by continuous arterial pulse contour (CO(PiCCOpulse)) analysis is a clinically validated procedure proved to be an alternative to the pulmonary artery catheter thermodilution cardiac output (CO(PACtherm)) in cardiac surgical patients. There is ongoing debate, however, of whether the CO(PiCCOpulse )is accurate after profound hemodynamic changes. The aim of this study was therefore to compare the CO(PiCCOpulse )after cardiopulmonary bypass (CPB) with a simultaneous measurement of the CO(PACtherm). METHODS: After ethical approval and written informed consent, data of 45 patients were analyzed during this prospective study. During coronary artery bypass graft surgery, the aortic transpulmonary thermodilution cardiac output (CO(PiCCOtherm)) and the CO(PACtherm )were determined in all patients. Prior to surgery, the CO(PiCCOpulse )was calibrated by triple transpulmonary thermodilution measurement of the CO(PiCCOtherm). After termination of CPB, the CO(PiCCOpulse )was documented. Both CO(PACtherm )and CO(PiCCOtherm )were also simultaneously determined and documented. RESULTS: Regression analysis between CO(PACtherm )and CO(PiCCOtherm )prior to CPB showed a correlation coefficient of 0.95 (P < 0.001), and after CPB showed a correlation coefficient of 0.82 (P < 0.001). Bland-Altman analysis showed a mean bias and limits of agreement of 0.0 l/minute and -1.4 to +1.4 l/minute prior to CPB and of 0.3 l/minute and -1.9 to +2.5 l/minute after CPB, respectively. Regression analysis of CO(PiCCOpulse )versus CO(PiCCOtherm )and of CO(PiCCOpulse )versus CO(PACtherm )after CPB showed a correlation coefficient of 0.67 (P < 0.001) and 0.63 (P < 0.001), respectively. Bland-Altman analysis showed a mean bias and limits of agreement of -1.1 l/minute and -1.9 to +4.1 l/minute versus -1.4 l/minute and -4.8 to +2.0 l/minute, respectively. CONCLUSION: We observed an excellent correlation of CO(PiCCOtherm )and CO(PACtherm )measurement prior to CPB. Pulse contour analysis did not yield reliable results with acceptable accuracy and limits of agreement under difficult conditions after weaning from CPB in cardiac surgical patients. The pulse contour analysis thus should be re-calibrated as soon as possible, to prevent false therapeutic consequences.