Glucose metabolism after renal transplantation

Abstract

Objective: We determined prevalence, risk factors, phenotype, and pathophysiological mechanism of new-onset diabetes after transplantation (NODAT) to generate strategies for optimal pharmacological management of hyperglycemia in NODAT patients.

Research design and methods: Retrospective cohort study comparing demographics, laboratory data, and oral glucose tolerance test (OGTT)-derived metabolic parameters from kidney transplant recipients versus subjects not receiving transplants.

Results: Among 1,064 stable kidney transplant recipients (≥ 6 months posttransplantation), 113 (11%) had a history of NODAT and 132 (12%) had pretransplant diabetes. In the remaining patients, randomly assigned OGTTs showed a high prevalence of abnormal glucose metabolism (11% diabetes; 32% impaired fasting glucose, impaired glucose tolerance, or both), predominantly in older patients who received tacrolimus as the primary immunosuppressant. Compared with 1,357 nontransplant subjects, stable kidney transplant recipients had lower basal glucose, higher glycated hemoglobin, lower insulin secretion, and greater insulin sensitivity in each of the three subgroups, defined by OGTT 2-h glucose (<140, 140-199, ≥ 200 mg/dL). These findings were reinforced in linear spline interpolation models of insulin secretion and sensitivity (all P < 0.001) and in another regression model in which the estimated oral glucose insulin sensitivity index was substantially higher (by 79-112 mL/min m(2)) for transplant versus nontransplant subjects despite adjustments for age, sex, and BMI (all P < 0.001).

Conclusions: Glucose metabolism differs substantially between kidney transplant recipients and nontransplant controls. Because impaired insulin secretion appears to be the predominant pathophysiological feature after renal transplantation, early therapeutic interventions that preserve, maintain, or improve β-cell function are potentially beneficial in this population.

Figure 1

Glucose metabolism after renal transplantation. OGTT-derived measures are shown from subjects at 3 months posttransplant (TIP study patients = early kidney transplant recipients [KTRs]) and ≥6 months posttransplant (TAHG study patients = stable KTRs) in comparison with the general population (VGH patients = non-KTRs). A: Insulin secretion by 2-h glucose (left) and insulin sensitivity by 2-h glucose (right). Model shows ordinary least-squares regression analysis of 2-h glucose (independent variable) against (left) insulin AUC and (right) OGIS (dependent variables). Within each of the three patient groups (TIP, TAHG, non-KTRs), 2-h glucose was modeled using a linear spline function with a single knot at 140 mg/dL. P for interaction testing slope difference between stable KTRs and non-KTRs was <0.001 (left) and <0.001 (right). P for interaction testing slope difference between early KTRs and non-KTRs was 0.016 (left) and <0.001 (right). B: Insulin sensitivity by β-cell function (left) in patients with 2-h glucose <140 mg/dL and (right) in patients with 2-h glucose ≥140 mg/dL. Model shows ordinary least-squares regression of IGI (independent variable) against OGIS (dependent variable). Results were centered at the mean OGIS within each category of 2-h glucose. P for interaction testing slope difference between stable KTRs and non-KTRs was 0.21 (left) and 0.05 (right). P for interaction testing slope difference between early KTRs and non-KTRs was 0.06 (right). A and B: Model results for the subgroup of TIP study patients with 2-h glucose <140 mg/dL were hidden because of insufficient sample size. x, Non-KTRs; ○, stable KTRs; ▲, early KTRs.