C-peptide levels do not correlate with pancreas allograft failure: Multicenter retrospective analysis and discussion of the new OPT definition of pancreas allograft failure

The OPTN Pancreas Transplantation Committee performed a multicenter retrospective study to determine if undetectable serum C-peptide levels correspond to center-reported pancreas graft failures. C-peptide data from seven participating centers (n = 415 graft failures for transplants performed from 2002 to 2012) were analyzed pretransplant, at graft failure, and at return to insulin. One hundred forty-nine C-peptide values were submitted at pretransplant, 94 at return to insulin, and 233 at graft failure. There were 77 transplants with two available values (at pretransplant and at graft failure). For recipients in the study with pretransplant C-peptide <0.75 ng/mL who had a posttransplant C-peptide value available (n = 61), graft failure was declared at varying levels of C-peptide. High C-peptide values at graft failure were not explained by nonfasting testing or by individual center bias. Transplant centers declare pancreas graft failure at varying levels of C-peptide and do not consistently report C-peptide data. Until February 28, 2018, OPTN did not require reporting of posttransplant C-peptide levels and it appears that C-peptide levels are not consistently used for evaluating graft function. C-peptide levels should not be used as the sole criterion for the definition of pancreas graft failure.