SAN FRANCISCO, CA—Proton pump inhibitors (PPIs) are safe and effective for reducing the rate of recurrent bleeding among patients with cirrhosis and acute variceal bleeding, according to a single-institution case-control study reported at the The Liver Meeting® 2015.

“PPIs could be safely used in patients with liver cirrhosis by careful selection of candidates,” concluded lead study author Soo Young Park, MD, Department of Internal Medicine, Kyungpook National University School of Medicine, Daegu, Korea, and colleagues.

The efficacy and safety of long-term PPIs in patients with acute variceal bleeding has not before been evaluated, Dr. Park noted. To evaluate safety and efficacy among patients with liver cirrhosis and acute gastroesophageal bleeding, the researchers conducted a prospective case-control study. Between 2008 and 2013, 255 patients (PPI group: 132 patients; control group: 123 patients) were enrolled and “prospectively evaluated for 5-day in-hospital failure, 6-week failure, and recurrent bleeding afterwards,” Dr. Park reported. “Five day in-hospital treatment failures were determined by Baveno V criteria and rebleeding within 6 weeks were prospectively evaluated.”

Baseline cirrhosis etiologies in the PPI and control groups differed (P=0.034), with more alcohol-related cirrhosis among controls (41% vs. 53%) and more “other” (non-hepatitis B virus, hepatitis C virus, or alcohol) etiologies noted among patients in the PPI study group (14% vs. 4.9%). Otherwise, the baseline patient characteristics did not differ significantly between the PPI and control groups, Dr. Park noted.

“There were no significant differences in 5-day treatment failure rate (6.1% vs. 6.5%; P=1.000) and 5-day mortality (5.3% vs. 4.9%; P=1.000) in PPI group and control group,” Dr. Park reported.

Patients in the PPI group showed a lower 6-week rebleeding rate than control-group patients (12.1 % vs. 22.8%; P = 0.031).

“Six-week mortality was similar between two groups (7.6% vs. 6.5% respectively; P=0.810),” Dr. Park reported. “The PPI group showed better event-free survival compared to the control group (P=0.019).”

In multivariate analysis, Child-Pugh score emerged as a predictor for 5-day treatment failure (hazard ratio [HR] 1.648; 95% CI: 1.139, 2.384; P=0.008), and both Child-Pugh score and PPI therapy emerged as independent predictors of 6-week treatment failure (HR 1.328; 95% CI: 1.096, 1.610; P=0.004 and HR 0.465; 95% CI: 0.233, 0.931; P=0.031, respectively).