Purse-string approximation is the closure modality of choice following ileostomy reversal: a meta-analysis and systematic review

Incidence of surgical site infection (SSI) following ileostomy reversal can reach 41%. Variations on primary linear closure (PLC) include delayed primary closure, antibiotic implants and more recently pursestring approximation (PSA), where the skin is approximated via purse-string but not closed. The optimal approach remains to be determined. This meta-analysis compares outcomes with primary closure versus pursestring approximation for ileostomy reversal. Embase and Medline were searched without language restriction to identify trials comparing PLC to PSA for closure of ileostomy wounds. Pooled odds ratios were calculated for SSI, and mean differences for the secondary continuous outcomes of length of hospital stay, operative duration and cosmetic outcomewere calculated using the fixed-effects and random-effects models for meta-analysis. Five out of 40 identified articles met the inclusion criteria. For the primary outcome of SSI rate, 159 patients in the PLC and 102 patients in the PSA groups were available for comparison. PSA was associated with a marked reduction in SSI (5 vs. 26%, Odds ratio: 0.192; 95% CI:0.050.79). No differences were noted in operative duration or in length of hospital stay. Long term cosmetic results were assessed in three studies at a minimum of 1 year post surgery with no differences in self reported cosmesis between the two methods (mean difference on 10-point visual analogue scale: 0.24; p = 0.302). Purse-string approximation of ileostomy wounds is associated with reduced surgical site infection without detrimental impact on cosmetic appearance. This data supports the adoption of purse-string approximation as the closure modality of choice Conflict of interest: The authors have no conflicts of interest or disclosures to declare regarding this work.