The simultaneous measurement of wedged hepatic vein pressure (WHVP) and portal vein pressure (PVP) was performed in 156 cirrhotic patients. In the 110 alcoholic cirrhotic patients (97 micronodular and 13 macronodular cirrhosis), WHVP and PVP were closely related (25.8 +/- 6.3 vs. 25.9 +/- 6.3 mm Hg; p = not statistically significant). The difference between the two parameters was greater than 4 mm Hg in only six patients. In the 46 patients with nonalcoholic cirrhosis (41 macronodular and 1 micronodular cirrhosis; 4 primary biliary cirrhosis), PVP was significantly higher than was WHVP (25.8 +/- 6.2 vs. 21.7 +/- 6.8 mm Hg; p less than 0.001); in 20 patients, PVP exceeded WHVP by more than 4 mm Hg, and the mean difference was 7.5 mm Hg. There was no correlation between the porto-hepatic gradient and total hepatic blood flow measured by the indocyanine green single injection method or the portal fraction of total hepatic blood flow measured by indicator dilution curves. It is concluded that: (i) measurement of WHVP in alcoholic cirrhosis provides a reliable estimate of the severity of the portal hypertension, and (ii) hemodynamic evaluation of nonalcoholic cirrhosis should include PVP measurement in order to avoid underestimation of the porto-hepatic gradient.