Meta-analysis of relaparotomy for secondary peritonitis.

Background: Planned relaparotomy and relaparotomy on demand are two frequently employed surgical treatment strategies for patients with abdominal sepsis. Methods: The available literature was evaluated to compare the efficacy of both surgical treatment strategies. A systematic search for studies comparing planned and on-demand relaparotomy strategies in adult patients with secondary peritonitis was employed. Studies were reviewed independently for design features, inclusion and exclusion criteria, and outcomes. The primary outcome measure was in-hospital mortality. Results: No randomized studies were found; eight observational studies with a total of 1266 patients (planned relaparotomy, 286; relaparotomy on demand, 980) met the inclusion criteria and were included in the meta-analysis. These eight studies were heterogeneous on clinical and statistical grounds (chi(2) = 40.7, d.f. = 7, P < 0.001). Using a random-effects approach, the combined odds ratio for in-hospital mortality was 0.70 (95 per cent confidence interval 0.27 to 1.80) in favour of the on-demand strategy. Conclusion: The combined results of observational studies show a statistically non-significant reduction in mortality for the on-demand relaparotomy strategy compared with the planned relaparotomy strategy when corrected for heterogeneity in a random-effects model. Owing to the non-randomized nature of the studies, the limited number of patients per study, and the heterogeneity between studies, the overall evidence generated by the eight studies was inconclusive.