A dog model using an implanted system for protracted hepatic arterial chemotherapy.

A model for hepatic arterial chemotherapy studies using large dogs and an implantable infusion pump has been developed. Using this technique near complete perfusion (greater than 90%) of the liver can be achieved in vivo as determined by hepatic arterial perfusion scintigraphy with technitium 99m macroaggregated albumin. The system is reliable and has been in use for a total of 1353 days (mean of 104 days, range 52-239) in 13 dogs. Pump implantation causes no apparent acute liver damage based on pre- and post-operative alkaline phosphatase and serum glutamic-pyruvic transaminase determinations and does not affect the general mobility or behavior of the animals. Careful placement of the catheter and attention to the physicochemical properties of the solutions loaded are factors contributing to the success of the model. The model permits comprehensive preclinical pharmacokinetic and toxicologic studies of new or preexistent chemotherapeutic agents in the same device that will be used for later administration in human subjects. By providing the means to examine and develop new treatment modalities, it enables the design of even more potent cytotoxic therapy directed into the tumor vascular bed.

Enhanced endothelialization of Dacron grafts by external vein wrapping.

The efficacy of external vein wrapping of vascular grafts in promoting the development of a luminal endothelial surface was assessed in 12 dogs who underwent thoracoabdominal bypasses with 26-29 cm X 6-mm i.d. double-velour knitted Dacron grafts. In group I (n = 6) 6-cm segments of the grafts were wrapped with autologous jugular vein with its endothelial surface applied against the outside of the graft. In group II (n = 6) the wrap procedure was performed using abdominal wall fascia. The degree and character of graft incorporation was quantitated in all prostheses at 28 days postimplantation. Group I vein wrap prostheses demonstrated uniform endothelial surface coverage in the vein wrap area (means = 88%) that was significantly greater (P less than 0.04) than coverage in adjacent unwrapped segments (means = 48%). Group II fascial wrap grafts demonstrated variable endothelial coverage in the wrapped area (means = 48%) that was consistently less than in adjacent unwrapped portions (means = 62%), and significantly less than vein wrap sections of group I animals (P less than 0.04). No significant differences existed in endothelial coverage of unwrapped regions of group I and II grafts. This investigation documented that wrapping knitted Dacron grafts with vein enhanced endothelialization of their luminal surface.

Transplantation of the pancreas into a retroperitoneal jejunal loop.

A new technic of pancreatic transplantation was developed and assessed in sixty-nine dog experiments. The body and tail of the pancreas were transplanted into the side of the Roux-en-Y retroperitoneal limb of jejunum. A dual venous anastomosis of the splenic vein to the vena cava was utilized to avoid venous congestion. Mean survival with normoglycemia was 32.4 days in thirty dogs with autografts and 24.6 days in twenty-seven dogs with allografts. Two dogs with autotransplants remain alive at 106 and 128 days, and the longest normoglycemic survival achieved with an allograft was 85 days. With experience the incidence of pancreatitis and abscess formation decreased. The problem of venous thrombosis was eliminated and the patency of the pancreatic duct was maintained.