Purse-string approximation is superior to primary skin closure following stoma reversal: a systematic review and meta-analysis

• Published in: Techniques in coloproctology Vol. 17; no. 4; pp. 345 - 351
• Main Authors: McCartan, D. P., Burke, J. P., Walsh, S. R., Coffey, J. C.
• Format: Journal Article
• Language: English
• Published: Milan Springer Milan 01-08-2013; Springer Nature B.V
• Subjects: Abdominal Surgery; Colorectal Surgery; Female; Follow-Up Studies; Gastroenterology; Humans; Ileostomy - adverse effects; Ileostomy - methods; Length of Stay; Male; Medicine; Medicine & Public Health; Pain, Postoperative - physiopathology; Pouchitis - diagnosis; Pouchitis - surgery; Proctology; Randomized Controlled Trials as Topic; Reconstructive Surgical Procedures - methods; Reoperation - methods; Retrospective Studies; Review; Risk Assessment; Skin; Surgery; Surgical Stomas; Surgical Wound Infection - etiology; Surgical Wound Infection - physiopathology; Surgical Wound Infection - surgery; Suture Techniques; Sutures; Treatment Outcome; Wound Healing - physiology; Reversal; Surgical site infection; Ileostomy; Stoma
• ISSN: 1123-6337; 1128-045X
• DOI: 10.1007/s10151-012-0970-y

Background
The incidence of surgical site infection (SSI) following stoma reversal can reach 40 %. A recent variation on primary linear closure (PLC) is purse-string approximation (PSA), where the skin is approximated via a purse-string suture but not closed. The optimal technique remains to be determined. The objective of this review was to compare outcomes with PLC versus PSA for skin closure following stoma reversal.
Methods
A literature search of Embase and Medline was performed to identify studies comparing PLC with PSA published between 1966 and 2012. Reviews of each study were conducted and data extracted. Random-effects methods were used to combine data, and between-study heterogeneity was assessed.
Results
Six out of 47 identified studies met the inclusion criteria: 2 randomized controlled trials and 4 case controlled series. For the primary outcome of SSI rate, 233 patients in the PLC and 170 patients in the PSA group were available for comparison. PSA resulted in a reduced rate of SSI (2.4 % PSA vs. 29.6 % PLC; OR 0.083, 95 % CI = 0.03–0.21,
p
 < 0.001). No differences were noted in length of hospital stay. Three studies assessed self-reported cosmetic results at a minimum of 7 months post-operatively. Patients who underwent PSA reported greater satisfaction with the cosmetic outcome (Standard mean difference = 0.47 on ten-point scale, 95 % CI 0.15–0.79,
p
 = 0.005).
Conclusions
Purse-string approximation of stoma wounds is associated with an 80 % reduction in SSI with no negative effect on length of hospital stay or long-term cosmetic outcome.