Background and study aims: Various techniques using surgical and natural orifice transluminal endoscopic surgery (NOTES) have been evaluated to create a gastrojejunal bypass. The aim of the current study was to determine the safety, feasibility, and efficacy of a new technique using a pure endoscopic approach and tissue-apposing stent placement for gastrojejunal anastomosis (GJA).

Materials and methods: This was a prospective, experimental study on six live pigs weighing 20 - 45 kg. Endoscopies were performed using a double-channel gastroscope, and included the creation of a GJA using a tissue-apposing, fully covered, self-expanding metallic stent. Antibiotic therapy was continued for 7 days after the procedure, and food was gradually reintroduced from Day 3. Changes in weight following the procedure were compared with a control group of age-matched animals. Anastomosis functionality was confirmed by endoscopy at 3 weeks (before the animals were euthanized), and during histopathological analysis. The primary outcomes were morbidity and mortality at 3 weeks. Secondary outcomes were technical feasibility, procedure time, and patency of the GJA.

Results: The procedures were performed successfully in all animals. The mean procedure time was 26 ± 6.7 minutes (range 15 - 32 minutes). One case of stent migration occurred during the procedure; the stent was successfully replaced using the same procedure. All animals were alive after 3 weeks. The mean weight gain during follow-up was 0.85 ± 2.56 kg (range - 2 to + 2 kg) compared with 5.2 ± 1.6 kg (range 3 - 7 kg) in control animals (P = 0.007). At necropsy, the stents were still in place in all animals, without evidence of peritonitis. Histopathology confirmed permeable anastomoses with continuity of the mucosa and mucosa muscle layers.

Conclusions: GJA with a tissue-apposing stent is safe, feasible, and reproducible without anastomotic leakage in a porcine model using a pure endoscopic approach and standard endoscopic equipment.