This study compared the accuracy of a formula calculating cardiac output utilizing a patient size modification of a pulse pressure formula of Starr with that measured with a standard thermal dilution technique. During a six-month period 111 patients in the intensive care unit (ICU) on the cardiothoracic and vascular surgical services had comparison of their cardiac output by these two methods. The basic formula of Starr for stroke volume was converted to a stroke volume index by dividing by 1.7 and the empirically derived average body surface area in meters square. The stroke volume index was multiplied by the body surface area (BSA) of the patient to determine the patient's stroke volume in ml. Thus the modified stroke volume formula was 100-0.6 age-0.6 Diastolic Pressure + 0.5 Pulse Pressure x Patients BSA (m2)" over 1.7. Cardiac output was calculated by multiplying the stroke volume by heart rate. Nearly 60% of the patients had less than a 5% difference between the two methods, and over 90% had less than a 10% variance. In this particular population the highest variation was 18%. Thus, using only a carefully measured sphygmomanometer blood pressure, stroke volume and cardiac output can be determined with sufficient accuracy for clinical use.