Transplantation for type I diabetes: comparison of vascularized whole-organ pancreas with isolated pancreatic islets

Abstract

Objective: We sought to compare the efficacy, risks, and costs of whole-organ pancreas transplantation (WOP) with the costs of isolated islet transplantation (IIT) in the treatment of patients with type I diabetes mellitus.

Summary background data: A striking improvement has taken place in the results of IIT with regard to attaining normoglycemia and insulin independence of type I diabetic recipients. Theoretically, this minimally invasive therapy should replace WOP because its risks and expense should be less. To date, however, no systematic comparisons of these 2 options have been reported.

Methods: We conducted a retrospective analysis of a consecutive series of WOP and IIT performed at the University of Pennsylvania between September 2001 and February 2004. We compared a variety of parameters, including patient and graft survival, degree and duration of glucose homeostasis, procedural and immunosuppressive complications, and resources utilization.

Results: Both WOP and IIT proved highly successful at establishing insulin independence in type I diabetic patients. Whole-organ pancreas recipients experienced longer lengths of stay, more readmissions, and more complications, but they exhibited a more durable state of normoglycemia with greater insulin reserves. Achieving insulin independence by IIT proved surprisingly more expensive, despite shorter initial hospital and readmission stays.

Conclusion: Despite recent improvement in the success of IIT, WOP provides a more reliable and durable restoration of normoglycemia. Although IIT was associated with less procedure-related morbidity and shorter hospital stays, we unexpectedly found IIT to be more costly than WOP. This was largely due to IIT requiring islets from multiple donors to gain insulin independence. Because donor pancreata that are unsuitable for WOP can often be used successfully for IIT, we suggest that as IIT evolves, it should continue to be evaluated as a complementary alternative to rather than as a replacement for the better-established method of WOP.

FIGURE 1. A Kaplan-Meier graft survival (a) and insulin independence (b) comparing WOP recipients and IIT recipients. One islet recipient was excluded from graft survival analysis because he or she was withdrawn from the study before completing the treatment protocol. The death in the pancreas transplant group is counted as an early graft lost. Graft loss in islet recipients was defined as a persistent C-peptide level <0.5 ng/mL and a return to pretransplant levels of insulin requirement. For the insulin-free survival analysis, 2 islet recipients were excluded from the curve, ie, the patient mentioned above who withdrew from the study after a single partially effective infusion and the patient who received 3 infusions but did not gain insulin independence.

FIGURE 2. Length of hospital stay for IIT recipients and WOP recipients (a). The islet patients had markedly fewer inpatient days than the pancreas recipients, both for their transplants and for their readmissions. This occurred despite the fact that most islet recipients underwent 2 or more infusions. Only 1 islet recipient required care in an intensive care unit, whereas the pancreas transplant recipients were routinely admitted to the intensive care unit immediately after transplantation. Figure 2b compares the direct cost of WOP with that of IIT. The unadjusted WOP cost includes all kidney-related charges for SPK recipients. The adjusted WOP cost subtracts all kidney-related charges to provide a more meaningful comparison with patients receiving islets but not kidney graft. Intent to transplant factors for islets include the organ and processing costs of the 23 pancreata that were processed with intent to clinically transplant but did not meet laboratory release criteria.