Cardiac output determination after cardiac operation. Lack of correlation between direct measurements and indirect...

Ninety-two mean thermodilution cardiac indices were determined in 25 patients following cardiac operations. Arterial and venous blood gases, hematocrit, body temperature, central venous pressure, left atrial pressures, urine output, heart rate, and mean arterial pressure were simultaneously recorded. Six variables, arterial and venous pH, arterial and venous PCO2, venous PO2, and temperature, showed significant simple correlations with cardiac index, but the degree of correlation was inadequate for use of these variables as reliable indices of cardiac function. When stepwise multiple regression analysis was performed, two variables, venous PO2 and left atrial pressure were associated with the best correlation with cardiac index, such that CI = 0.073 PO2V -- 0.060 LAP + 1.39 (r = 0.60, p less than 0.001). When measured values for venous PO2 and left atrial pressure were substituted into this equation, a "predicted" value for cardiac index could be obtained with only slightly improved reliability. The data indicate that indirect estimation of cardiac output by measurement of the various laboratory parameters described above is not sufficiently reliable for clinical use. The importance of obtaining direct measurements of cardiac output by thermodilution or other means in order to properly evaluate postoperative cardiac function is stressed.