Objective:To compare semicontinuous cardiac output (CCO) with bolus cardiac output (BCO), in the immediate postoperative period after valvular surgery, under hypothermic cardiopulmonary bypass with two CCO pulmonary artery catheters, based on the pulsed warm thermodilution technique, i.e., Opti-Q from Abbott or IntelliCath from Baxter-Edwards (Abbott and Baxter groups, respectively).Design:Prospective study.Setting:University hospital.Patients:Forty-four adult patients scheduled for mitral and/or aortic valve surgery were randomized into two groups. Tricuspid or pulmonary valvulopathy diagnosed by echocardiography was excluded.Interventions:Cardiac output was measured every 20 mins during the 3 postoperative hrs. BCO was the mean of three boluses (10 mL) of an ice-cold saline solution injected within 3 secs. CCO was the mean of two CCO values obtained in normal mode immediately before and after BCO measurements.Measurements and Main Results:Two groups of 22 patients underwent 198 pairs of cardiac output measurements. The mean difference or bias was calculated as the difference between BCO and CCO, and precision was the SD of the mean bias. The limits of agreement were defined as bias ± 2 SD. A two-sample Wilcoxon's test was used for comparison of bias and precision in sinus and non-sinus rhythm, and stable and unstable mean arterial pressure in each group and between the two pulmonary artery catheters. The coefficient of correlation was also calculated. Bias ± precision was 0.066 ± 0.526 L/min, r2= .83, for the Abbott group, and 0.015 ± 0.490 L/min, r2= .85 (not significant), for the Baxter group. There was no significant difference within and between groups for bias and precision in sinus and non-sinus rhythm, nor in stable and unstable mean arterial pressure.Conclusions:The study, during the immediate postoperative period in valvular surgery under hypothermic cardiopulmonary bypass, showed a satisfactory correlation between CCO and BCO with the two systems.