Estimating the benefits of solitary pancreas transplantation in nonuremic patients with type 1 diabetes mellitus: A theoretical analysis

Background The goal of early pancreas transplantation in type 1 diabetes mellitus is to achieve euglycemia and thereby prevent the renal, retinal, and vascular complications of this disease. The purpose of this analysis was to examine the conditions and assumptions that would make early solitary pancreas a viable therapeutic option. Methods. A Markov model was constructed to compare outcomes for patients with type 1 diabetes mellitus and early overt nephropathy assigned to either standard insulin therapy or solitary pancreas transplantation. Probabilities for development of end stage renal disease (ESRD), blindness, mortality, and direct health care costs were taken from the literature. Utility scores for the relevant health states were determined by the standard gamble method on 16 type 1 diabetic patients suitable for pancreas transplantation. Results. Assuming a baseline graft life expectancy for the pancreas of 10 years, early pancreas transplantation could provide 0.42 more life years and 2.2 more quality adjusted life years (discounted at 3%) to patients above standard insulin therapy. The model was sensitive to estimates of pancreas graft life expectancy (<8 years being inadequate to extend patient life), timing of pancreas transplantation with respect to rate of progression to ESRD, and drug nephrotoxicity rates. The incremental costs (charges) for early pancreas transplantation over standard therapy were estimated to be modestly high (about $56,600/quality adjusted life year for the baseline case). Pancreas transplant costs were also a very sensitive parameter in the cost utility analysis. Conclusions. The success of early solitary pancreas transplantation may well be at the stage to consider a trial in selected type 1 diabetic patients at risk for renal and retinal disease.