Single-centre study of 628 adult, primary kidney transplant recipients showing no unfavourable effect of new-onset diabetes after transplant

Aims/hypothesis: To better understand the implications of new-onset diabetes after transplant (NODAT), we used our prospectively followed cohort of 628 adult primary kidney transplant recipients to determine the prognostic impact of pretransplant diabetes and NODAT.

Methods: The study cohort consisted of all participants in four randomised immunosuppression trials performed at our centre since May 2000. For each cause-specific hazard analysed, Cox stepwise regression was used to determine a multivariable model of significant baseline predictors; the multivariable influence of having pretransplant diabetes and NODAT (t) (the latter defined as a zero-one, time-dependent covariate) was subsequently tested. Similar analyses of estimated glomerular filtration rate (eGFR) at 36 and 60 months post transplant were performed using stepwise linear regression. Finally, a repeated measures analysis of mean HbA_1c as a function of diabetes category (pretransplant diabetes vs NODAT) and randomised trial (first to fourth) was performed.

Results: Median follow-up was 56 months post transplant. Patients with pretransplant diabetes comprised 23.4% (147/628), and 22.5% (108/481) of the remaining patients developed NODAT. Pretransplant diabetes had no prognostic influence on first biopsy-proven acute rejection and death-censored graft failure hazard rates, nor on eGFR, but was associated with significantly higher rates of death with a functioning graft (DWFG) (p = 0.003), DWFG due to a cardiovascular event (p = 0.005) and infection that required hospitalisation (p = 0.03). NODAT (t) had no unfavourable impact on any of these hazard rates nor on eGFR, with actuarial freedom from DWFG remaining at over 90% among patients in pre- and post-NODAT states at 72 months post transplant/NODAT. Mean HbA_1c for patients in the first to fourth randomised trials, averaged across diabetes category, decreased by trial (7.28%, 6.92%, 6.87% and 6.64% [56.1, 52.1, 51.6 and 49.1 mmol/mol], respectively; p = 0.02).

Conclusions/interpretation: Less-than-expected post-NODAT risk for graft loss and death may exist in the current climate of tighter glucose monitoring post transplant.