Technical aspects of segmental pancreatic autotransplantation in dogs

Heterotopic segmental pancreatic autotransplantation in dogs is more appropriate than allograft models for the investigation of several problems associated with transplantation. We have defined the anatomic variations of blood supply in the pancreatic tail and designed various modifications of vascular anastomosis to the iliac vessels in order to eliminate technical failures, such as thrombosis, as much as possible. In 187 of 240 dogs (77.9 percent), the pancreatic artery originated from the splenic artery and the pancreatic vein entered the splenic vein (normal anatomy). The main venous variation was direct confluence of the pancreatic and portal veins (12.1 percent), and the main arterial variation was origin of the pancreatic artery from the superior mesenteric artery (10 percent). Ninety-seven animals with normal anatomy qualified for a comparative study of seven methods of segmental pancreatic autotransplantation. Venous anastomoses were always performed in an end-to-side fashion between the splenic and external iliac veins. Arterial anastomosis techniques follow. Group I: interposition of the splenic artery into the external iliac artery; (14 days, failure rate 50 percent), Group II: end-to-end arterial anastomosis of the splenic artery to a long external iliac artery segment with the graft directed caudad, resulting in an acute curve to the vessel loop (8 dogs, failure rate 38 percent); Group III: end-to-end arterial anastomosis to a long external iliac artery with the graft directed cephalad, resulting in a gentle curve to the vessel loop (11 dogs, failure rate 36 percent); Group IV: end-to-end arterial anastomosis to a short external iliac artery stump (20 dogs, failure rate 20 percent); Group V: same as in Group IV with the addition of a distal splenic arteriovenous fistula (12 dogs, failure rate 17 percent); Group VI: end-to-end anastomosis of the splenic artery to the external iliac artery (31 dogs, failure rate 6 percent); Group VII: same as in Group VI but with the addition of an arteriovenous fistula of the distal splenic vessles (1 dog, failure rate 0). The end-to-side technique proved to be straight-forward and reliable. The low failure rate with this method allows metabolic preservation and other aspects of pancreatic transplantation to be studied and the results to be interpreted without the influence of a high complication rate from the operation itself.