Relief of obstructed defecation syndrome after stapled transanal rectal resection (STARR): a meta-analysis.

OBJECTIVES: Stapled transanal rectal resection (STARR) is a promising new treatment for obstructed defecation syndrome (ODS) associated with rectal intussusception and/or rectocele. The aim of this work was to assess the efficacy of STARR to treat ODS. METHODS: Outcome data after STARR for ODS were pooled according to the used constipation score. As different types of constipation scores were reported, and standardized effect sizes were calculated before performing a meta-analysis. RESULTS: Twenty-six publications were identified with a median follow-up of 12 months (range: 3-42). In total 1298 patients were included. Six different scoring systems were used. In total 43 estimates of the effect STARR were analyzed. All studies showed a significant improvement in ODS yielding a combined standardized effect size of 3.8 (95% CI : 3.2-4.5). Although a very high degree of heterogeneity between effect sizes has been observed (I2 = 93.3%), suggesting an overestimation of this improvement. This is partially due to the use of various instruments, but largely originating from (unmeasured) study characteristics. CONCLUSIONS: The consistent finding of a decrease in the various ODS-scores confirms that STARR can reduce ODS but the effect is overestimated. This meta-analysis clearly highlights some methodological shortcomings in published data. Heterogeneity in ODS scoring implies the need for standard effect size calculation to compare published results, and underlines the urgent need for a more uniform and accurate data reporting.