To evaluate the effect of combined kidney and pancreas transplantation on insulin action and glucose metabolism, 15 Type 1 (insulin-dependent) diabetic patients who were undergoing combined kidney-pancreas transplantation were studied before transplantation by means of the euglycaemic hyperinsulinaemic clamp technique combined with 3-3H-glucose infusion and indirect calorimetry. Nine of the original 15 patients were studied again after four months and six after 12 months, successful combined kidney-pancreas transplantation with the same experimental protocol. Nine volunteers formed the group of normal subjects. Combined kidney-pancreas transplantation normalised hepatic glucose production and reduced peripheral insulin resistance in Type 1 diabetic uraemic patients, despite chronic immunosuppressive therapy. To further evaluate the hypothesis that residual insulin resistance was due to chronic steroid therapy. 11 additional subjects with chronic uveitis (six of whom were treated with only prednisone, and five treated only with cyclosporin) underwent the same protocol demonstrating a normal hepatic glucose production. The insulin-stimulated peripheral glucose uptake was reduced in the prednisone-treated group, but normal in cyclosporin-treated subjects. Four additional diabetic patients with a kidney transplant were also studied. They showed a peripheral insulin sensitivity intermediate between diabetic uraemic patients and patients after combined transplant. We conclude that short-term (one year) combined kidney-pancreas transplantation improves glucose metabolism by restoring normal rates of hepatic glucose production and reducing peripheral insulin resistance; chronic steroid therapy is the major determinant of residual reduced insulin action. Both kidney and pancreas substitution play a role in reducing peripheral insulin resistance.