CHICAGO – Patients with cirrhosis and medium to large esophageal varices were 81% less likely to develop a first esophageal variceal bleed within 5 years if they took rifaximin to manage hepatic encephalopathy, a retrospective study of 238 patients found.
Of the 97 patients who received rifaximin (a nonabsorbable antibiotic with broad-spectrum activity), 94% remained free of variceal hemorrhage at 3 years, 4 years, and 5 years after their sentinel endoscopy, the historic cohort study showed. Among the 141 patients who were not on rifaximin, 75% had not developed a variceal bleed within 3 years, and 58% had no variceal bleeds at years 4 and 5, Dr. Bradley Confer and his associates reported.
The cumulative incidence of esophageal variceal bleed was 11% in 5 years, or 4% for patients on rifaximin and 16% for those not on rifaximin, said Dr. Confer of the Cleveland Clinic.
Those statistically significant differences between groups suggest that rifaximin use decreases the risk of a first esophageal bleed and decreases the time to a first bleed in cirrhotic patients with medium to large varices, but a larger prospective randomized trial would be needed to confirm that, he said at the annual Digestive Disease Week.
All patients were treated with nonselective beta-blocker medications, esophageal band ligation, or a combination of both and were followed to the time of a first esophageal variceal bleed, liver transplant, last follow-up, or death. Beta-blockers were used by 63% of patients, and 87% underwent at least one banding procedure. Patients on rifaximin used the drug for a median of 9 months.
Although five factors were associated with the risk for a first variceal hemorrhage in a univariate analysis, only three remained significant after a multivariate analysis that adjusted for the use of beta-blockers and band ligation: rifaximin use (associated with an 81% decrease in risk), age (a 20% increase in risk with 5 additional years of age), and the patient’s International Normalized Ratio (which can be associated with more than a sixfold increase in bleed risk).
Previous data suggest that cirrhotic patients with medium to large esophageal varices still have a 10%-20% risk of developing a first esophageal variceal bleed even after treatment with either nonselective beta-blockers or band ligation. Many patients do not tolerate nonselective beta-blockers well, and the drugs may negatively affect a subgroup of patients with cirrhosis, he said. Repeating variceal ligation is a costly and invasive management strategy.
The investigators turned their attention to rifaximin after separate data showed that 4 weeks of use were associated with a significant decrease in hepatic venous pressure gradient in patients with alcoholic cirrhosis. They evaluated records on 1,121 patients with cirrhosis who underwent esophagogastroduodenoscopy in 2008-2011 and excluded any with no varices, small varices, a history of esophageal variceal bleeding, or noncirrhotic portal hypertension, or who received a transjugular intrahepatic portosystemic shunt.
The mean age at the time of esophagogastroduodenoscopy was 53 years, and 72% of patients were male. They had an average Model for End-Stage Liver Disease score of 14.3 and an average Child-Pugh score of 8.7. Compared with patients not on rifaximin, those who got the antibiotic were significantly less likely to use nonselective beta-blockers and were sicker, Dr. Confer said, with significantly higher serum creatinine levels, lower serum albumin, and a greater likelihood of hepatic encephalopathy at baseline.
Dr. Confer reported having no financial disclosures.
On Twitter @sherryboschert
This article was updated on June 26, 2014.