Comparison of endoscopic closure modalities for standardized colonic perforations in a porcine colon model

Background and study aim: Colonic perforation is a serious complication of colonoscopy, with surgical repair usually indicated. The aim was to compare acute strength of various endoscopic colonic closure techniques by assessing air leak pressures in a previously described ex vivo experimental apparatus. Methods: Standardized colonic perforations were created using fresh porcine colon and subsequently closed on a bench. Six techniques included surgical suture (gold standard), QuickClips, T-tags, over-the-scope-clip (OTSC) system, and two types of flexible stapler (Covidien). After closure, each specimen was fixed in the apparatus and pressure was gradually increased until air bubbles were seen. Leak pressure was the primary outcome parameter. Closure using the gold standard (first 15 experiments) resulted in a mean leak pressure of 86.9 mmHg (SD 7). Using a noninferiority design a sample size of 12 specimens for each closure technique was determined. Results: Mean colotomy leak pressures in millimeters of mercury (mmHg) and difference (with 95% confidence intervals [CI]) between each technique and the gold standard were: QuickClips 85.1 (difference -1.8; 95% CI -7.0 to 3.9); T-tags 53.9 (difference -33.0; -39.0 to -27.0); OTSC 90.3 mmHg (difference 3.4; -6.1 to 12.9); 15-mm shaft stapler 98.5 mmHg (difference 9.7; 0.8 to 18.5) and 8-mm shaft stapler 96.6 mmHg (difference 11.6; 1.5 to 21.7). Conclusions: OTSCs, QuickClips, and both flexible staplers produced results comparable to hand-sewn colotomy closure in this ex vivo porcine colonic model. These devices seem to be prime candidates for further evaluation in survival animal studies