Developing an animal model of massive ulcer bleeding for assessing endoscopic hemostatic devices

Background: Existing animal models of ulcerative bleeding are not suitable for endoscopic investigation. We describe a new porcine model of massive arterial bleeding in the stomach and its use for assessing a novel endoscopic suturing device. Methods: Two animal models were investigated. In model 1, the short gastric artery (mean diameter 2 mm) was divided near its gastric end. A mucosal defect was created near the greater curve and the divided artery was brought into the lumen of the stomach through a submucosal tunnel. An inflatable plastic cuff was placed around the base of the artery. Cuff deflation led to massive bleeding. In model 2, the short gastric artery was carefully exposed along a segment of 2 cm on the side facing the stomach. It was then anchored to a small gastrostomy made at the posterior wall near the vessel. At endoscopy an ulcer-like lesion could be seen with a pulsatile vessel at the base and brisk bleeding could be started by cutting a hole in the artery using endoscissors. The pigs were heparinized by an intravenous bolus of 110-300 units per kilogram, in both models. A prototype suturing device, the Eagle Claw, was inserted using a gastroscope and the curved needle was driven around the bleeding artery. Extracorporeal knotting or intracorporeal ligation was done endoscopically. Results: Pulsatile arterial bleeding was successfully created in four pigs using model 1, and in another four pigs using model 2. Model 2 was more reproducible and less time-consuming to create. Endoscopic suturing controlled arterial bleeding in five out of eight pigs with a single stitch and in another three pigs with an additional stitch. Conclusion: This animal model provides reproducible massive hemorrhage suitable for endoscopic studies. Control of gastric bleeding from large arteries by endoscopic suturing is possible.