BACKGROUND: Hepatocyte transplantation using three-dimensional matrices is under investigation as an alternative therapy for several liver diseases. For sufficient transplantation results hepatotrophic stimulation is necessary. We investigated the stimulatory effect of cotransplanted pancreatic islets in different ratios. METHODS: Lewis rats were used as donors and recipients. A portocaval shunt (group A) or sham operation (groups B-G) was performed 1 week before hepatocyte transplantation. Four polyvinyl-alcohol matrices each containing 1.25 x 10(7) hepatocytes (groups A and B) or 1.25 x 10(7) hepatocytes and 125 (C), 250 (D), 500 (E), or 750 (F) islets were implanted between small bowel mesenteric leaves. In group G, medium soaked matrices were implanted. One month after implantation, specimens were harvested and investigated using albumin-RNA in situ hybridization, and insulin, glucagon, and bromodesoxy uridine immunohistochemistry. The hepatocyte area was assessed using image analysis. RESULTS: Hepatocyte area and proliferation ratio increased depending on the number of cotransplanted islets with a peak at 40 islets per 1 million hepatocytes (group E). Cotransplantation of islets in higher concentrations did not further increase hepatocyte area or proliferation ratio. Hepatocytes in all groups expressed albumin RNA at normal transcription levels as compared to standard liver sections. Islets displayed insulin and glucagon in physiological distribution. DISCUSSION: Three-dimensional matrices provide a sufficient environment for transplanted hepatocytes and islets. The hepatotrophic effect of cotransplanted islets is comparable to portocaval shunting and has a saturation limit at 40 islets per 1 million hepatocytes. For further application of islet cotransplantation, this ratio seems to be preferable.
