Diabetes Mellitus Following Pancreas Transplantation

O312 Aim: Diabetes mellitus following pancreas transplantation is commonly attributed to graft loss, but may occur despite an otherwise well-functioning pancreas allograft. We sought to define the incidence and possible causes of diabetes mellitus after pancreas transplantation. Methods: We retrospectively studied all patients (n=88) undergoing pancreas transplantation at our institution between January 2001 and December 2002. Diabetes mellitus was defined as: 1) the need for pharmacotherapy to achieve a non-diabetic fasting serum glucose (<126 mg/dl) and 2) evidence of graft function, i.e. an increase in serum C-peptide from pretransplant levels and a normal allograft ultrasound. Data were analyzed using Student’s t-test and the Chi-square test and are expressed as median (25%-75% inter-quartile range). Results: Of the 88 patients receiving pancreas transplants, 81 had Type 1 diabetes mellitus and 7 had Type 2 diabetes mellitus. Forty-seven patients (53%) received a pancreas after kidney transplant, 29 (33%) received a pancreas transplant alone and 12 (14%) received a simultaneous kidney-pancreas transplant. Median follow-up was 25 months (18-30 months). Actual 1-year patient and graft survival were 97% and 90%, respectively. During the entire study period, 4 grafts were lost to immediate post-operative thrombosis, 2 to late thrombosis, and 5 to patient death with a functioning graft. Of the remaining 77 patients, 57 (74%) remained normoglycemic, 6 (8%) were hyperglycemic for less than 1 month and 14 (18%) required pharmacotherapy longer than 1 month despite evidence of graft function. Hyperglycemia occurred in 23% (16/70) of Type 1 diabetics and 57% (4/7, p=0.0489) of Type 2 diabetics. Other factors associated with hyperglycemia included: high pre-transplant insulin dose, high BMI and increased rejection episodes (p<0.0001, p=0.0005 and p=0.0002, respectively). With respect to insulin dose, hyperglycemia developed in all patients (6/6) requiring >100 units of insulin/day pretransplant, in 83% (10/12) requiring >75 units/day and in only 15% (10/65) of those requiring ≤75 units/day (p< 0.0001). Despite the development of hyperglycemia, glucose control was significantly improved after transplant. The median pretransplant hemoglobin A1c for patients developing hyperglycemia was 8.0 (6.7-8.7) compared to 6.6 (5.5-7.4) 12 months following transplant (p=0.0294). There were no significant differences between the groups with regard to age, gender, pre-transplant hemoglobin A1c, steroid doses or tacrolimus concentrations at 1 month and 1 year. Conclusions: The development of diabetes mellitus in patients with a functioning pancreas allograft is due to high pre-transplant insulin requirements, increased body mass index and rejection episodes. However, despite the persistent hyperglycemia, glycemic control is significantly improved after pancreas transplantation.