Antibiotics for spontaneous bacterial peritonitis in cirrhotics.

backgroundSpontaneous bacterial peritonitis is mainly a complication of cirrhotic ascites that occurs in the absence of any intra-abdominal, surgically treatable source of infection. Antibiotics have been recommended as the mainstay treatment for spontaneous bacterial peritonitis. However, this recommendation is not based on convincing evidence. It has been proposed that treatment should cover Gram-negative enteric bacteria and Gram-positive cocci, that are responsible for up to 90% of cases.

objectivesTo evaluate the effectiveness and safety of different types and ways of antibiotic therapy for spontaneous bacterial peritonitis in cirrhotic patients. SEARCH STRATEGY: Electronic searches on the Cochrane Library (Issue 3, 2000), the Cochrane Hepato-Biliary Group Trials Register (March 2000), EMBASE (1980-2000), MEDLINE (1966-2000); scanning the references of all identified studies; contacting the first author of each included trial. SELECTION CRITERIA: Randomised trials comparing different types of antibiotics for spontaneous bacterial peritonitis in cirrhotic patients. DATA COLLECTION AND ANALYSIS: Data were independently extracted by two reviewers. Relative risks or weighted mean differences, with their 95% confidence intervals were estimated using 'intention-to-treat' analyses. MAIN

resultsNine trials dealing with 684 patients diagnosed with spontaneous bacterial peritonitis were included. No placebo-controlled trial was found. Each of the included trials compared different antibiotics, and no meta-analysis could be performed. We were unable to establish the optimal dose or duration of antibiotic therapy and found no convincing evidence that cefotaxime is more effective than ampicillin-tobramycin or that oral quinolones should be recommended for patients with less severe manifestations of the disease. REVIEWER'S

conclusionsThis review provides no clear evidence for the treatment of cirrhotic patients with spontaneous bacterial peritonitis. Until large, well-conducted, trials provide adequate evidence, treatment must be based on clinical experience.