Surgical-site infection (SSI) is the most common complication after stoma reversal, with reported rates up to 40%1. Numerous skin-closure strategies have been proposed to reduce this risk; however, available randomized controlled trials (RCTs) are mostly small, single-center, and limited to pairwise comparisons, precluding a comprehensive comparison of all techniques. Existing systematic reviews and pairwise meta-analyses have similarly focused on selected strategies and have not integrated the full body of randomized evidence2, 3. We therefore performed a systematic review and Bayesian network meta-analysis to compare skin-closure techniques after stoma reversal and establish a comparative hierarchy for SSI prevention. The study was conducted in accordance with the PRISMA-NMA guidelines4 (Supplemental Digital Content Table S1, available at: https: //links. lww. com/JS9/H228). The protocol was prospectively registered in PROSPERO and methodological quality was assessed using the AMSTAR 2 checklist5 (Supplemental Digital Content Table S2, available at: https: //links. lww. com/JS9/H229). MEDLINE and the Cochrane Central Register of Controlled Trials were searched from inception to October 31, 2025 (Supplemental Digital Content Table S3, available at: https: //links. lww. com/JS9/H230). RCTs comparing at least two skin-closure techniques after ileostomy or colostomy reversal in adults were eligible. Observational studies, pediatric populations, and studies not reporting SSI were excluded. Two reviewers independently screened studies, extracted data, and assessed risk of bias using the Cochrane RoB 2 tool6, with disagreements resolved by consensus. The primary outcome was overall SSI within 30 days. A Bayesian random-effects network meta-analysis was performed, estimating odds ratios (ORs) with 95% credible intervals (CrI). Between-study heterogeneity was modeled using a Bayesian random-effects framework, with estimation of the between-study standard deviation (τ) on the log-odds ratio scale. Analyses were conducted using the MetaInsight platform, which applies Bayesian random-effects models implemented through established R packages. Treatment ranking probabilities were summarized using the surface under the cumulative ranking curve (SUCRA). The treatment network was well connected (Fig. 1) and included seven interventions across 27 randomized trials (Supplemental Digital Content Figure S1, available at: https: //links. lww. com/JS9/H227), comprising 2144 patients and 258 SSI events. Twenty-six studies were two-arm trials and one was a multi-arm study (Supplemental Digital Content Table S4, available at: https: //links. lww. com/JS9/H231), allowing 21 possible pairwise comparisons, of which eight were informed by direct evidence. Figure 1.: Network geometry, pooled odds ratios, and ranking of skin-closure techniques after stoma reversal. (A) Network diagram showing all randomized comparisons among closure methods. Each node represents an intervention, and the width of connecting lines and node size are proportional to the number of studies and patients, respectively. (B) Forest plot displaying pooled odds ratios (OR, 95% credible intervals) for surgical-site infection (SSI) compared with primary skin closure (PrimarySC). (C) Cumulative ranking curves summarizing the probability of each intervention being among the best options for preventing SSI, expressed as the surface under the cumulative ranking curve (SUCRA). Primary skin closure with prophylactic negative-pressure wound therapy (Primary SC + pNPWT) ranked highest (SUCRA 80%), followed by gunsight (74%) and purse-string closure (73%). In the primary analysis, purse-string closure significantly reduced SSI risk compared with primary skin closure (OR 0. 20, 95% CrI 0. 08–0. 44). Primary closure combined with prophylactic negative-pressure wound therapy (pNPWT) showed the largest relative reduction in SSI risk (OR 0. 12, 95% CrI 0. 01–1. 09), although credible intervals crossed unity. Primary closure with a subcutaneous drain did not significantly reduce SSI risk (OR 0. 50, 95% CrI 0. 13–1. 95), and gunsight closure showed no clear benefit (OR 0. 14, 95% CrI 0. 01–3. 13). Secondary skin closure was associated with the highest SSI risk (OR 2. 44, 95% CrI 0. 28–22. 9) (Fig. 1, Supplemental Digital Content Table S5, available at: https: //links. lww. com/JS9/H232). SUCRA rankings favored primary closure with pNPWT (≈80%) and purse-string closure (≈73%) as the most effective strategies (Fig. 1). Sensitivity analyses restricted to ileostomy reversal yielded consistent results (Supplemental Digital Content Figure S2, available at: https: //links. lww. com/JS9/H227). Between-study heterogeneity was moderate, no major inconsistency was detected, and assessment of small-study effects was limited by sparse data. Most trials were judged as having some concerns for risk of bias (Supplemental Digital Content Table S6, available at: https: //links. lww. com/JS9/H233). Limitations include heterogeneity among trials, variable SSI definitions, limited assessment of publication bias, and lack of patient-level adjustment. In conclusion, both purse-string closure and primary closure combined with pNPWT are associated with reduced SSI risk after stoma reversal. Purse-string closure represents a simple, low-cost strategy supported by relatively precise effect estimates, whereas pNPWT involves additional expense and resource use, with wide credible intervals reflecting imprecision due to limited sample sizes. Given that many SSI after stoma reversal are superficial and managed conservatively, routine use of pNPWT may not be cost-effective, supporting selective application in patients at higher risk of wound complications.
Meyer et al. (Mon,) studied this question.