OBJECTIVE To determine whether infection with Helicobacter pylori is a risk factor for portosystemic encephalopathy in patients with acute, moderate or severe alcoholic hepatitis. DESIGN Prospective, multicenter cohort study. SETTING Eight Veterans Affairs Hospitals. PATIENTS A cohort of 273 male patients enrolled in a Department of Veterans Affairs Cooperative Study performed to evaluate the efficacy of oxandrolone in combination with nutritional supplementation in moderate or severe alcoholic hepatitis. MEASUREMENTS Admission serum IgG antibody titers against H. pylori by a specific and sensitive ELISA, demographic characteristics of patients, degree of protein calorie malnutrition, presence of ascites, bilirubin level, and known risk factors for hepatic encephalopathy (gastrointestinal bleeding, azotemia, hepatorenal syndrome, infection, and severity of disease); outcome was the presence of portosystemic encephalopathy. RESULTS Of 188 patients with decompensated alcoholic hepatitis available for analysis, 117 (62.2%) had encephalopathy. Ninety-two (78.6%) of these were infected with H. pylori, compared with 62% of patients without encephalopathy (p = 0.013). In a step-wise regression model, H. pylori was an independent risk factor (relative risk: 2.4, 95% CI: 1.2-4.8) adjusting for ascites and protein-calorie malnutrition. CONCLUSIONS Patients with acute, moderate or severe alcoholic hepatitis have a high H. pylori infection rate (as determined by serology), and those infected are at higher risk for portosystemic encephalopathy.