Direct puncture of the small bowel under endoscopic guidance (direct EPJ) is possible in patients whose stomach has been removed or whose small bowel cannot be punctured by other methods. From January 1990 to June 1992 a total of 39 patients underwent successful direct EPJ at our institution. The indications were malnutrition after partial or total gastrectomy (n = 19), insufficient anastomosis or a stenosis after esophageal resection and esophagojejunostomy (n = 13), esophageal perforation (n = 3), fistulas (n = 2), or severe trauma (n = 2). The tubes were inserted at the bedside under local anesthesia using the string pull-through technique. The procedure was attempted in five other patients but it was technically impossible to insert the tubes in these patients. Postoperative enteral feeding was possible in all 39 patients whose direct EPJ was successful. Complications included tube dysfunction due to plugging and fracture in five patients, pressure-induced enteric ulcers in two, and local infections in three patients. The ulcers and infections were managed conservatively. We conclude that direct EPJ is a safe, effective alternative to surgical catheter-jejunostomy.

Successful transplantation of encapsulated islets (bioartificial pancreas) would circumvent problems of islet availability and rejection in the treatment of insulin-dependent diabetes with biological organ replacement. Alginates are widely used as a hydrogel matrix or membrane for immunoprotected transplantation. A major problem in the use of diffusion-based devices is the biocompatibility of the material used. The foreign body reaction after implantation of empty microcapsules into different compartments in rats, dogs and pigs is evaluated in this article. However, biocompatibility of the bioartificial pancreas has three different aspects: reaction of the entrapped islet to the encapsulation technique and material; reaction of the recipient against the incorporated device ( = foreign body reaction); and finally the reaction of the recipient against the encapsulated islet ( = immunology of bioartificial pancreas). It is obvious from different experiments that even if foreign body reactions (reactions against material) are almost abolished the recipient may react against material released from the encapsulated islet. In conclusion, transplantation of encapsulated islets induces various morphological reactions (i.e. inflammation and fibrosis) as a result of a variety of donor and recipient related factors. Therefore, the use of an adequate animal model that reflects the human situation is essential for progress in the development of a bioartificial pancreas.

A major reason for the failure of clinical islet transplantations may be a limited islet mass. The aim of this study was to determine the critical islet mass necessary for normalization of glucose metabolism in a porcine model. Diabetes was induced by total pancreatectomy. The splenic lobe of the pancreas was intraductally distended with UW-solution containing 2.67-3.33 mg/ml collagenase, and the distended pancreas was digested in a continuous digestion filtration device. The islets were purified on a isoosmotic Ficoll-sodium-diatrizoate gradient. The survival period of the diabetic recipients in group 2 and 3 receiving, respectively, a low (2.14+/-0.39 microL/kg body weight) and a high (4.99+/-0.83 microL/kg body weight) islet mass was significantly prolonged compared to that of diabetic recipients in group 1 receiving no islet transplantation. However, the survival period of the recipients in group 2 was not significantly different to that in group 3. Three recipients of an islet mass of >5 microl/kg body weight became normoglycemic (fasting blood glucose <100 mg/dl) for more than two months. Furthermore, the glucose and insulin release reactions to the glucose challenge were comparable to that before pancreatectomy. Contrarily, another five diabetic recipients of an islet mass of <4 microL/kg body weight became a fasting blood glucose level of <200 mg/dl. The glucose and insulin release reactions to the glucose challenge were improved only, but not normalized compared to that before pancreatectomy. The data presented in this study demonstrate that metabolic normalization in pancreatectomized diabetic minipigs can be established by autotransplantation of an islet mass of >5 microl/kg body weight.

We conclude that autotransplantation with a sufficient islet mass can induce normoglycemia with a normal insulin secretion response to glucose stimulation in pancreatectomized diabetic pigs and that allotransplantation can be successfully achieved when 15-deoxyspergualin and antithymocyte globulin are combined with the triple-drug immunosuppression described above. However, this immunosuppressive protocol results in a high rate of infectious complications.

Preoperative radiochemotherapy followed by surgery is able to achieve clear resection margins in more than 70% of patients with locally advanced rectal cancer and may improve the rate of sphincter-conserving surgery.