Relationship between portal pressure, esophageal varices, and variceal bleeding on the basis of the stage and cause of cirrhosis.

BACKGROUND AND METHODS

Hepatic venous pressure gradient, esophageal varices, and variceal bleeding were investigated in 957 patients with cirrhosis. The causes (alcoholic/virus) and stage (Child-Pugh's classification) of cirrhosis were also taken into account.

RESULTS

The prevalence of variceal bleeding was 35% in patients with large varices and 17% in those with small varices (P < 0.05). It was higher in patients with alcoholic cirrhosis (41% and 19%, respectively) than in those with viral cirrhosis (22% and 10%, respectively). In patients with alcoholic cirrhosis the hepatic venous pressure gradient was higher in Child A and B patients with small or large varices than in those with no varices; these differences were not found in Child C patients and in patients with viral cirrhosis. In all subgroups the pressure gradient was higher in Child C patients than in Child A patients. There was no significant difference in the hepatic venous pressure gradient between patients with varices and previous variceal bleeding and those with no bleeding whatever the stage of cirrhosis.

CONCLUSIONS

This study shows that the hepatic venous pressure gradient is associated with the stage and causes of cirrhosis and the presence of varices. These factors should be taken into account in studies evaluating the hepatic venous pressure gradient in heterogeneous groups of patients.