RESUMEN
BACKGROUND: Postoperative pancreatic fistula (POPF) is a major complication in pancreatic surgery and can cause considerable postoperative morbidity. Advanced surgical-technical approaches to prevent POPF did not yield a substantial improvement. To investigate innovative treatments, experimental animal models of distal pancreatic resection and pancreaticoduodenectomy are of fundamental importance. After a failed attempt to replicate a previously described rat model for pancreatic fistula induction, we proceeded to distal pancreatic resection with splenectomy to provoke pancreatic leakage and generate a suitable animal model. METHODS: Distal pancreatic resection with splenectomy was performed in 40 rats. The rats were sacrificed on postoperative day (POD) 1, 2, 4, 6, 8, or 10, and the abdominal cavity was explored. Ascites probes were collected pre- and postoperatively for the detection of pancreas amylase and lipase. Tissue samples from the naïve pancreas (POD 0) and the postoperatively harvested remnant were evaluated histologically. The extent of necrosis was determined, and samples were examined for neutrophil infiltration. TUNEL staining served for the verification of necrosis in distinct cases. Immunohistochemistry of Ki67, von Willebrand factor, and CD68 was performed to evaluate proliferation, blood-vessel sprouting, and macrophage invasion. RESULTS: The rats showed no clinical symptoms or severe complications in the postoperative course up to 10 days. Abdominal exploration revealed adhesions in the upper abdomen, but no intra-abdominal fluid accumulations were found. Signs of inflammation and tissue damage were evident at the pancreatic resection margin on histological examination whereas the naïve pancreatic tissue was widely unaffected. Statistically significant differences were seen between the preoperative and postoperative extent of necrosis, the presence of neutrophil infiltrate, and levels of ascitic amylase and lipase. Immunohistochemical staining on Ki67, von Willebrand factor, and CD68 did not reveal any workable results on nonstatistical examination, and it was therefore not considered for further analyses. CONCLUSION: Creating a functional animal model of pancreatic fistula that reflects the clinical and pathophysiological impact of pancreatic leakage in humans has not been achieved. Our approach of left pancreatic resection recapitulated inflammation and tissue damage, early events in the development of fistulas, and it could be suitable for the experimental testing of novel targeting methods.