Cardiac output can be estimated non-invasively by electrical cardiometry with the ICON® monitor (Osypka Medical GmbH, Berlin, Germany). Conflicting results have been reported regarding the cardiac output measurement performance of electrical cardiometry. In this prospective method comparison study, we compared cardiac output measured using electrical cardiometry (EC-CO; test method) with cardiac output measured using intermittent pulmonary artery thermodilution (PATD-CO; reference method) in patients after coronary artery bypass graft (CABG) surgery. We calculated the mean of the differences with 95%-limits of agreement (95%-LOA) and their corresponding 95%-confidence intervals (95%-CI) using Bland-Altman analysis and calculated the percentage error. We also analyzed trending using four-quadrant plot analysis. We analyzed 157 paired cardiac output measurements of 41 patients. Mean ± standard deviation PATD-CO was 5.1 ± 1.3L/min and mean EC-CO was 5.3 ± 1.3L/min. The mean of the differences ± SD between PATD-CO and EC-CO was -0.2 (95%-CI -0.5 to 0.2) ± 1.2L/min with a lower 95%-LOA of -2.6 (95%-CI -3.1 to -2.0) L/min and an upper 95%-LOA of 2.3 (95%-CI 1.6 to 2.9)L/min. The percentage error was 47% (95%-CI, 37 to 56%). The concordance rate for cardiac output changes was 48%. In this study, the agreement between EC-CO and PATD-CO was not clinically acceptable in patients after CABG surgery. The trending ability of EC-CO was poor.
Read full abstract