Closure of gastrotomy in natural orifice transluminal endoscopic surgery: a feasibility study using an ex vivo model comparing endoloop with endoclip.

BACKGROUND: Closure of access site is one of the key issues that will determine the development of natural orifice transluminal endoscopic surgery. Our study was designed to compare the effectiveness of gastrotomy closure using endoloop and endoclip with hand-sewn closure as a control.
METHODS: Gastrotomy was carried out on 24 ex vivo porcine stomachs and the gastrotomies were randomized to be closed with either hand-sewn, endoloop or endoclip techniques. A 2 cm gastrotomy was created with a needle knife and sphincterotome and the defects closed thereafter. We measured the time to closure of gastrotomy and the pressure at which the closure leaked.
RESULTS: Three endoloops were required for all the closures in the endoloop group. In the endoclip group, the mean (range) number of endoclips used was 10 (8-13). There was no difference in the median closure times between endoloop 28 (16-58) min and endoclip 30 (21-40) min; however, time for hand-sewn closure was much shorter (3-5 min). All stomachs were successfully distended with air without leak at the end of the procedure and none experienced fluid leak. The endoclip closure endured a significantly higher median (range) pressure of 72.5 mmHg (15-80 mmHg) before leaking compared to that of an endoloop 25 mmHg (15-37 mmHg) (P < 0.001). The hand-sewn gastrotomy leaked at pressure of 95 mmHg (75-130 mmHg). The majority of air leaks were from the wound site. In the endoclip group, two leaks were noted at the clip bite site.
CONCLUSION: In a bench-top model, endoclips seem to be better for gastrotomy closure because of their potential to endure relatively higher pressure without any prolongation of application time. Animal survival studies are needed to explore technical and wound-closure-related problems arising as a result of use of endoclips and endoloops for gastrotomy closure.