Antibiotics for prophylaxis of spontaneous bacterial peritonitis: systematic review & Bayesian network meta-analysis

Abstract

Background: Various antibiotic regimens are used for primary and secondary prevention of spontaneous bacterial peritonitis (SBP). A systematic review and network meta-analysis to compare various antibiotics regimens for primary and secondary prevention of SBP were done.

Methods: We did a comprehensive literature search using various databases (i.e. MEDLINE via Ovid and PubMed, Embase, Cochrane Central Register of Controlled Trials and others) from inception to 26th October 2019 using various keywords. Only randomised studies which evaluated the role of antibiotics in adult cirrhotic patients with ascites for primary or secondary prophylaxis of SBP were included. The primary outcome was occurrence/recurrence of SBP episode and other outcomes assessed were extra-peritoneal infections and reduction in mortality. We did random-effects network meta-analysis using a Bayesian approach, and calculated odds ratios (ORs) and 95% credible intervals (CrI); agents were ranked using rank probabilities.

Results: We found total 1701 records in our systematic database search and out of these 17 randomised trials were found eligible for network meta-analysis. For primary prevention of SBP, the odds ratio (95% CrI) for norfloxacin daily was 0.061 (0.0060, 0.33) and for rifaximin daily was 0.037 (0.00085, 0.87) and norfloxacin and rifaximin alternate month was 0.027 (0.00061, 0.61) when compared to placebo or no comparator. For the secondary prevention of SBP, rifaximin daily had odds of 0.022 (0.00011, 0.73).

Conclusion: Rifaximin is useful for both primary and secondary prevention of SBP whereas norfloxacin daily and alternate norfloxacin and rifaximin are useful for primary prophylaxis.

Keywords: Ascites; Cirrhosis; Cotrimoxazole; Fluoroquinolone; Network meta-analysis; Norfloxacin; Portal hypertension; Prophylaxis; Rifaximin; Spontaneous bacterial peritonitis.