Although particle embolization is a common treatment of hypervascular neuroendocrine metastases of the liver, this treatment is not considered an option for the originating pancreatic neuroendocrine tumor(s) due to theoretical concern for pancreatitis. The purpose of this study was to explore the safety of pancreatic embolization in a porcine model. Approval for this study was obtained from the Institutional Animal Care and Use Committee. A total of 6 domestic swine weighing between 90 and 110 pounds were obtained. A single nonsurvival animal underwent infusion of methylene blue dye into the dorsal pancreatic artery using a microcatheter followed by particle embolization to stasis using 100-300 μm Embosphere particles to confirm technical feasibility. Immediate laparotomy confirmed blue staining of pancreatic parenchyma without evidence of non-target embolization. Five survival animals underwent baseline serum lipase level assay, and particle embolization of the dorsal pancreatic artery was performed under general anesthesia. Animals were observed postoperatively for one week with monitoring of behavior, activity, food intake, and observation for signs of pancreatitis. After one week, repeat serum lipase was obtained, the animals were euthanized, and their pancreata excised for gross pathologic analysis. Technical success with embolization was achieved in all study animals. In the five survival animals that were monitored after embolization, behavior, activity levels and food intake returned to baseline by postoperative day #1. All animals survived to 7 days without signs or symptoms of significant pain, systemic infection or decline. There was no significant change in serum lipase levels comparing pre-procedure to day 7, with average serum lipase levels of 56.8 U/L and 40.2 U/L, respectively. At necropsy, regions of lobular expansion and discoloration were identified in all pancreata that are compatible with inflammation. These findings suggest that particle embolization of the dorsal pancreatic artery in swine is technically feasible and results in adequate ischemia to cause inflammation of target pancreatic tissues, but without biochemical signs of pancreatitis, systemic sequelae, or mortality after one week of observation. Based on this early data, further evaluation of particle embolization of the pancreas in animal models may be justified to further clarify safety and physiologic sequelae.
Read full abstract