Antibiotic prophylaxis for the prevention of bacterial infections in cirrhotic patients with gastrointestinal bleeding: A meta-analysis

cirrhotic patients with gastrointestinal bleeding, antibiotic prophylaxis d ecreases the incidence of infections but most randomized trials have not sh own an increase in survival. The aim of this meta-analysis was to assess th e efficacy of antibiotic prophylaxis in the prevention of infections and it s effect on survival rate in cirrhotic patients with gastrointestinal bleed ing. Four end points were assessed: infection, bacteremia and/or spontaneou s bacterial peritonitis (SBP), incidence of SEP, and death. For each end po int, heterogeneity and treatment efficacy were assessed by Der Simonian and Peto methods. Five trials including 534 patients, 264 treated with antibio tic prophylaxis for 4 to 10 days and 270 without, were identified. Mean fol low-up was 12 days. Antibiotic prophylaxis significantly increased the mean percentage of patients free of infection (32% mean improvement rate, 95% c onfidence interval [CI]: 22-42, P < .001), bacteremia and/or SEP (19% mean improvement rate, 95% CI: 11-26, P < .001), and SEP (7% mean improvement ra te, 95% CI: 2.1-12.6, P = .006). Antibiotic prophylaxis also significantly increased the mean survival rate (9.1% mean improvement rate, 95 % CI: 2.9- 15.3, P = .004), without significant heterogeneity. In cirrhotic patients w ith gastrointestinal bleeding, short-term antibiotic prophylaxis significan tly increases the mean percentage of patients free of infection and signifi cantly increases short-term survival rate.