Eight RCTs (n=647) were included in meta-analyses. Follow-up duration varied from four to 60 weeks. Six RCTs were judged as high quality using the Jadad score.
Compared with controls, oral prophylactic antibiotics were associated with a significant reduction in all-cause mortality (RR 0.65, 95% CI 0.48 to 0.88, NNT 11; eight RCTs), spontaneous bacterial peritonitis (RR 0.49, 95% CI 0.35 to 0.69, NNT=8; eight RCTs) and overall infections (RR 0.32, 95% CI 0.21 to 0.51, NNT=6; eight RCTs).
Subgroup analysis showed that oral prophylactic antibiotics were associated with a significant reduction in short-term mortality at three-month follow-up (RR 0.28, 95 % CI 0.12 to 0.68, NNT=6; three RCTs).
There were no significant differences between the two groups in: rate of long-term mortality for at least six months follow-up; mortality of patients with primary prophylaxis of spontaneous bacterial peritonitis; and incidence of spontaneous bacterial peritonitis in patients with primary prophylaxis.
Sensitivity analyses of high-quality trials did not materially affect the results. No significant heterogeneity was observed in the outcomes. No evidence of publication bias was found according to visual scanning of funnel plots (funnel plots were not presented in the report).
The incidence of gastrointestinal bleeding, renal failure and hepatic failure did not significantly differ between treatment and control groups (data not reported).