Portal Pressure, Presence of Gastroesophageal Varices and Variceal Bleeding

This study was performed to examine the relationships between portal pressure measurements and the presence of esophagogastric varices, the size of varices and the occurrence of hemorrhage from varices in 93 patients with alcoholic cirrhosis, using standardized measurements of portal pressure by hepatic vein catheterization. The mean hepatic vein pressure gradient (HVPG) was significantly higher in 49 patients who had bled from varices than in 44 cirrhotic patients who had not (20.4 +/- 5.1 vs. 16.0 +/- 5.2; p less than 0.001). None of the 49 patients who had bled from varices had an HVPG less than 12 mm Hg. Among the 87 patients who had been examined by endoscopy for varices, all 72 with varices had an HVPG greater than 12 mm Hg. Six of 15 cirrhotic patients without varices had HVPG less than 12 mm Hg. The mean HVPG in the 15 patients without varices (15.1 +/- 6.8 mm Hg) was lower than the 72 patients with varices (19.3 +/- 4.8 mm Hg; p less than 0.01). Of the 72 patients with varices, 40 had large varices, 28 had small varices, and in four patients variceal size could not be assessed adequately. The mean HVPG was similar in the patients with large or small varices (19.8 +/- 4.8 vs. 18.3 +/- 5.0 mm Hg; p greater than 0.10). There was a positive relationship between the presence of large varices and the occurrence of bleeding from varices.(ABSTRACT TRUNCATED AT 250 WORDS)