Background: beta-Cell dysfunction and insulin resistance combine to cause new-onset diabetes after transplantation. The product of these two parameters, quantitatively measured as disposition index (DI), is a mathematical constant in normoglycemia and declines in advance of impending hyperglycemia. The aim of this study was to derive a simple surrogate for the DI to expose predysglycemic abnormalities posttransplantation.
Methods: First-phase insulin secretion and sensitivity were determined by mathematical minimal model analysis of 58 frequently sampled, intravenous glucose tolerance tests in 58 non-diabetic renal transplant recipients and correlated against surrogate indexes based on fasting blood samples. Products of insulin secretion/resistance indexes were correlated against calculated DI, regression analysis performed for hyperbolic compatibility, autocorrelation studies conducted, and surrogates tested in various subgroups of renal transplant recipients to ensure robustness in a heterogeneous group.
Results: The best correlation was achieved with "HOMA(sec) (first-phase insulin secretion)xMcAuley's index (insulin resistance)" (r=0.594, P<0.001). Regression analysis was consistent with a mathematical hyperbola (ln HOMA(sec) vs. ln McAuley's index, r=-0.639 [95% confidence interval, -1.772 to -0.950]), statistical autocorrelation was excluded (in a subset of 20 patients with repeat metabolic investigations), and the surrogate remained valid in different subgroups of transplant recipients.
Conclusions: Our surrogate "HOMA(sec)xMcAuley's index," requiring only fasting glucose, insulin, and triglycerides, is a simple and noninvasive surrogate for the DI. Its predictive utility for identifying impending hyperglycemia posttransplantation should be investigated further to ascertain whether its experimental nature can translate to clinical validity.