Recent animal studies suggest that nutritional repletion may improve function of liver allografts, and the authors have found that intraportal glucose infusion in pigs produces rapid and substantial hepatic glycogenation. A controlled prospective randomized study in 32 patients was done to determine glycogen content and degradation in human livers during transplantation, and the effect of intraportal glucose-insulin infusions during the donor operation on these variables and on outcome of transplantation. Peripheral blood glucose concentrations were "clamped" at 14 mmol/L during the glucose-insulin infusion. Liver biopsies were taken at various stages of the procedure. Liver glycogen decreased 2.0 +/- 1.2 g/100 g dry weight liver (mean +/- standard error of the mean) in controls, but increased 6.8 +/- 1.8 g/100 g dry weight in glucose-infused donors. In both groups there was glycogen degradation during periods of cold preservation, anoxic rewarming, and after reperfusion with portal blood. Degradation rates were greater in the glucose-infused group than in controls in all three periods (p less than 0.05). Despite wide variation in postoperative aspartate aminotransferase (AST) levels among recipients in both groups, the difference in peak postoperative AST levels approached significance (p = 0.06). In addition, peak AST levels were closely correlated to anoxic rewarming time in both groups, but the slope of the relationship was much lower (3834 versus 734, p less than 0.01) in the glucose-infused group. Thus at anoxic rewarming times over 90 minutes, glycogenation was protective of liver function. Peak postoperative AST was significantly correlated to glycogen degradation in the cold preservation and rewarming periods in the glucose-infused group only. Intraoperative glucose infusions in humans can reglycogenate the liver, increase glycogen degradation, and improve certain outcome measures in liver transplantation.