To investigate whether measuring cardiac output and its course after cardiac surgery by a new analysis technique of radial artery pressure waves, without need for calibration (FloTrac/Vigileo [FV]; Edwards Lifesciences, Irvine, CA), conforms to the standard bolus thermodilution method via a pulmonary artery catheter (PAC). Prospective study. Intensive care unit of university hospital. Twenty patients for up to 24 hours after cardiac surgery. Simultaneous and triplicate PAC thermodilution and FV cardiac output measurements at 1 and 3 hours after surgery and the following morning. Fifty-six simultaneous measurement sets were obtained. Mean cardiac output (PAC) ranged between 2.8 and 10.3 L/min and for the FV method between 3.3 and 8.8 L/min. The coefficient of variation for pooled measurements was 7.3% for the PAC and 3.0% for the FV method. For pooled data, the r2 was 0.55 (p < 0.001), with a bias of -0.14, precision of 1.00 L/min, and 95% limits of agreement between -2.14 and 1.87 L/min in a Bland-Altman plot. Also, the FV method tended to overestimate cardiac output when <7 L/min and increased with time, whereas mean arterial pressure increased and PAC cardiac output did not change. Changes in cardiac output correlated (r2 = 0.52, p < 0.001). The FV arterial pressure waveform analysis method is a clinically applicable method for cardiac output assessment without calibration, after cardiac surgery. It performs well at low cardiac outputs but remains sensitive to changes in vascular tone.