Background. The two-layer method [University of Wisconson solution (UW)/perfluorochemical plus O2] for pancreas preservation has been demonstrated to be superior to simple UW storage alone in the canine model. For the first time, we applied the two-layer method to clinical whole-pancreas transplantation. Methods. Pancreases were placed in the two-layer method in 10 cases and UW alone in 44 cases before transplant. The mean cold ischemic time was 16.5 hr in the two-layer group versus 18.1 hr in the UW group (P=NS). We compared the condition of graft at the time of reperfusion, and then 3 months posttransplant graft function and complications. Results. At the time of reperfusion, no grafts in the two-layer group were edematous, compared with 10(23.3%) of 43 in the UW group (P=0.18). Seven (70%) of 10 grafts in the two-layer group obtained the best overall quality score, compared with 24(57.1%) of 42 in the UW group (P=0.72). Nine (90%) of 10 recipients in the two-layer group became insulin-independent during hospitalization, compared with 31(70.5%) of 44 in the UW group (P=0.26). Time to insulin independence was no different between the two groups. No pancreas grafts preserved by the two-layer method suffered acute rejection. Conclusions. The two-layer preservation method is feasible in human clinical transplantation. It was at least equivalent and may be superior to UW alone in both morphologic and functional assessment of the transplanted pancreas.