#1680 Cardiovascular disease in kidney transplant recipients with post-transplant diabetes mellitus—retrospective analysis from a randomized trial

Abstract

Abstract Background and Aims Kidney transplant recipients are at risk for glucometabolic deterioration. While many individuals recover their glycemic profiles, some develop posttransplant diabetes mellitus (PTDM). Previous reports have shown that PTDM is associated with increased risk for cardiovascular disease (CVD) and mortality. Here we aimed at assessing the association of PTDM with the development of CVD in kidney transplant recipients from a randomized trial. Method CVD with functioning graft (WFG), all-cause death WFG and death-censored graft loss were retrospectively investigated in N = 263 participants from the Insulin Therapy for the Prevention of New Onset Diabetes after Transplantation multicenter randomized-controlled trial (ITP-NODAT, ClinicalTrials.gov NCT03507829). The clinical trial was conducted from November 2012 through May 2018; kidney transplant recipients without previous diabetes diagnosis had received basal insulin therapy or control for afternoon hyperglycemia. For the current analysis, CVD was defined as a composite of cerebrovascular disease (stroke, transient ischemic attack), coronary artery disease (myocardial infarction, revascularization, stenting), hospitalization for heart failure and vascular/peripheral arterial disease interventions. Participants were to perform three OGTTs at 6, 12 and 24 months posttransplant, which we handled as landmarks in our present analysis. Besides PTDM versus NON-PTDM, we stratified according to basal insulin intervention and control. Cox proportional hazard models were presented unadjusted and adjusted for age, sex and pre-transplant CVD. All patients gave informed consent to participate in the trial and we obtained institutional review board approval for the present analysis. Results Cumulative incidence curves for CVD WFG are presented in Fig. 1. The associated risk of PTDM at 6, 12 and 24 months for CVD WFG tended to be higher in PTDM versus NON-PTDM. At the respective timepoints, the control versus basal insulin groups were similar (Fig. 1). Cause-specific hazard ratios for CVD WFG, all-cause death WFG and death-censored graft loss are presented in Table 1. PTDM was associated with CVD WFG in the univariable models. Corresponding Hazard Ratios (with 95% Confidence Intervals) at 6, 12 and 24 months were 2.53 (1.23, 5.18), 3.19 (1.51, 6.78) and 4.28 (1.94, 9.48), respectively. After adjustment for age, sex and pre-transplant history of CVD, corresponding HRs (95% CIs) were 1.37 (0.64, 2.91), 1.37 (0.61, 3.06) and 3.01 (1.26, 7.22), respectively. PTDM was not associated with all-cause death WFG in the univariable models. Corresponding HRs (95% CIs) at 6, 12 and 24 months were 1.97 (0.82, 4.75), 2.44 (0.95, 6.25) and 2.32 (0.89, 6.05), respectively. PTDM was also not associated with death-censored graft loss in the univariable models. Corresponding HRs (95% CIs) at 6, 12 and 24 months were 1.44 (0.62, 3.36), 2.10 (0.89, 4.94) and 1.15 (0.39, 3.37), respectively. PTDM was neither associated with all-cause death WFG or death-censored graft loss in the adjusted models. Conclusion PTDM diagnosed by OGTT at month 24 was significantly associated with cardiovascular disease independent of age, sex or pre-transplant CVD. The association, however, was not significant when PTDM was diagnosed before month 24. It is possible that transient PTDM diagnosis, early CVD events and study discontinuations before month 24 may have contributed to the differing results over time. Timely PTDM diagnosis may open a window of opportunity for interventions, such as in the form of novel glucose-lowering agents, that may reduce cardiovascular risk in kidney transplant recipients.