A comparison of volume clamp method-based continuous noninvasive cardiac output (CNCO) measurement versus intermittent pulmonary artery thermodilution in postoperative cardiothoracic surgery patients

Julia Y. Wagner,Annmarie Körner,Leonie Schulte-Uentrop,Mathias Kubik,Hermann Reichenspurner,Stefan Kluge,Daniel A. Reuter,Bernd Saugel
DOI: https://doi.org/10.1007/s10877-017-0027-x
IF: 1.977
2017-05-24
Journal of Clinical Monitoring and Computing
Abstract:The CNAP technology (CNSystems Medizintechnik AG, Graz, Austria) allows continuous noninvasive arterial pressure waveform recording based on the volume clamp method and estimation of cardiac output (CO) by pulse contour analysis. We compared CNAP-derived CO measurements (CNCO) with intermittent invasive CO measurements (pulmonary artery catheter; PAC-CO) in postoperative cardiothoracic surgery patients. In 51 intensive care unit patients after cardiothoracic surgery, we measured PAC-CO (criterion standard) and CNCO at three different time points. We conducted two separate comparative analyses: (1) CNCO auto-calibrated to biometric patient data (CNCObio) versus PAC-CO and (2) CNCO calibrated to the first simultaneously measured PAC-CO value (CNCOcal) versus PAC-CO. The agreement between the two methods was statistically assessed by Bland–Altman analysis and the percentage error. In a subgroup of patients, a passive leg raising maneuver was performed for clinical indications and we present the changes in PAC-CO and CNCO in four-quadrant plots (exclusion zone 0.5 L/min) in order to evaluate the trending ability of CNCO. The mean difference between CNCObio and PAC-CO was +0.5 L/min (standard deviation ± 1.3 L/min; 95% limits of agreement −1.9 to +3.0 L/min). The percentage error was 49%. The concordance rate was 100%. For CNCOcal, the mean difference was −0.3 L/min (±0.5 L/min; −1.2 to +0.7 L/min) with a percentage error of 19%. In this clinical study in cardiothoracic surgery patients, CNCOcal showed good agreement when compared with PAC-CO. For CNCObio, we observed a higher percentage error and good trending ability (concordance rate 100%).
anesthesiology
What problem does this paper attempt to address?