Comparison of an advanced minimally invasive cardiac output monitoring with a continuous invasive cardiac output monitoring during lung transplantation

The aim of this study was to compare a continuous non-calibrated left heart cardiac index (CI) measurement by arterial waveform analysis (FloTrac^®/Vigileo^®) with a continuous calibrated right heart CI measurement by pulmonary artery thermodilution (CCOmbo-PAC^®/Vigilance II^®) for hemodynamic monitoring during lung transplantation. CI was measured simultaneously by both techniques in 13 consecutive lung transplants (n = 4 single-lung transplants, n = 9 sequential double-lung transplants) at distinct time points perioperatively. Linear regression analysis and Bland–Altman analysis with percentage error calculation were used for statistical comparison of CI measurements by both techniques. In this study the FloTrac^® system underestimated the CI in comparison with the continuous pulmonary arterial thermodilution (p < 0.000). For all measurement pairs we calculated a bias of −0.55 l/min/m² with limits of agreement between −2.31 and 1.21 l/min/m² and a percentage error of 55 %. The overall correlations before clamping a branch oft the pulmonary artery (percentage error 41 %) and during the clamping periods of a branch oft the pulmonary artery (percentage error 66 %) failed to reached the required percentage error of less than 30 %. We found good agreement of both CI measurements techniques only during the measurement point “15 min after starting the second one-lung ventilation period” (percentage error 30 %). No agreement was found during all other measurement points. This pilot study shows for the first time that the CI of the FloTrac^® system is not comparable with the continuous pulmonary-artery thermodilution during lung transplantation including the time periods without clamping a branch of the pulmonary artery. Arterial waveform and continuous pulmonary artery thermodilution are, therefore, not interchangeable during these complex operations.