Training of peroral endoscopic myotomy (POEM) in a survival porcine model

Introduction: Peroral endoscopic myotomy (POEM) has been introduced from a technique of natural orifice transluminal endoscopic surgery (NOTES) as a minimally invasive and safe endoscopic procedure for the therapy of esophageal achalasia. However, due to the relatively low incidence of achalasia in the general population, the potentially serious complications and the technically demanding procedure, proper training has a critical importance. Our study aim was to trial animal model for training as to improve the initial learning curve for POEM.

Methods: We applied a survival porcine model for training in POEM, since this is the most appropriate animal model due to its anatomy being quite similar to that of humans. A standard forward-viewing diagnostic gastroscope (Olympus EXERA II GIF-Q180) was used for the POEM procedure. An oblique cap (MH-588; Olympus) was attached and securely fixed at the tip of the endoscope. Endoscopic CO2 insufflation with a controlled gas feed of 1.2 L/min was applied for reducing the risk of mediastinal emphysema. A high-frequency electrosurgical energy generator (VIO 300D; ERBE) that has a spray coagulation mode with noncontact tissue was used (endocut mode Q, effect 3, and spray coagulation mode, effect 1, 60 W) for mucosal incision, submucosal dissection and myotomy. For final closure of the mucosal entry site, hemostatic clips (EZ-CLIP, HX-110QR; Olympus) were applied.

Results: All 10 POEM procedures were successful and made by two experienced interventional endoscopist (ML and SZA) on 5 separate occasions. 10 female, 50 – 60 kg-s of weight pigs were operated under general anesthesia, with intratracheal intubation in a supine position. The mean length of procedure (LOP) was 76 ± 31 minutes. The mean myotomy length was 12 cm (range, 8 – 18 cm). As our experience progressed, the means of the LOP per centimeter myotomy and variability decreased as follows: first day, 8 ± 6 minutes; second, 6 ± 5 minutes; third, 5 ± 3 minutes; fourth, 4 ± 2 minutes; and fifth, 3 ± 2 minutes. There were 7 pigs with capnoperitonium resolved by Veress needle decompression and another three with bilateral capnothoraces that were associated with hemodynamic instability and death in two. No postoperative bleeding or other complication was detected in the remaining 8 pigs.

Conclusions: POEM training on survival porcine models proved to be useful to improve the initial learning curve and to diminish complications for POEM. Therefore we strongly suggest porcine model training before the performance of POEM in humans. (Thanks for the technical support by Anamed LTD, and ERBE-MED LTD Hungary, and Foundation of Kozma and Damján. Also many thanks for expert opinion supports by Martínek J and Schoefl R.)