Background. The pulmonary artery catheter (PAC) is an accepted clinical method of measuring cardiac output (CO) despite no prior validation. The ultrasonic cardiac output monitor (USCOM) is a noninvasive alternative to PAC using Doppler ultrasound (CW). We compared PAC and USCOM CO measurements against a gold standard, the aortic flow probe (FP), in sheep at varying outputs. Methods. Ten conscious sheep, with implanted FPs, had measurements of CO by FP, USCOM, and PAC, at rest and during intervention with inotropes and vasopressors. Results. CO measurements by FP, PAC, and USCOM were 4.0 ± 1.2 L/min, 4.8 ± 1.5 L/min, and 4.0 ± 1.4 L/min, respectively, (n = 280, range 1.9 L/min to 11.7 L/min). Percentage bias and precision between FP and PAC, and FP and USCOM was -17 and 47%, and 1 and 36%, respectively. PAC under-measured Dobutamine-induced CO changes by 20% (relative 66%) compared with FP, while USCOM measures varied from FP by 3% (relative 10%). PAC reliably detected -30% but not +40% CO changes, as measured by receiver operating characteristic area under the curve (AUC), while USCOM reliably detected ±5% changes in CO (AUC > 0.70). Conclusions. PAC demonstrated poor accuracy and sensitivity as a measure of CO. USCOM provided equivalent measurements to FP across a sixfold range of outputs, reliably detecting ±5% changes.