Association of diuretic use with increased risk for long-term post-transplantation diabetes mellitus in kidney transplant recipients

Background Post-transplantation diabetes mellitus (PTDM) is a major clinical problem in kidney transplant recipients (KTRs). Diuretic-induced hyperglycaemia and diabetes have been described in the general population. We aimed to investigate whether diuretics also increase PTDM risk in KTRs. Methods We included 486 stable outpatient KTRs (with a functioning graft >= 1 year) without diabetes from a prospective cohort study. Participants were classified as diuretic users and non-users based on their medication use verified by medical records. Results At the baseline study, 168 (35%) KTRs used a diuretic (thiazide, n = 74; loop diuretic, n = 76; others, n = 18) and 318 KTRs did not use a diuretic. After 5.2 years [interquartile range (IQR) 4.0-5.9] of follow up, 54 (11%) KTRs developed PTDM. In Cox regression analyses, diuretic use was associated with incident PTDM, independent of age, sex, fasting plasma glucose (FPG) and haemoglobin A(1c) (HbA(1c)) {hazard ratio [HR] 3.28 [95% confidence interval (CI) 1.84-5.83]; P <0.001}. Further adjustment for potential confounders, including lifestyle, family history of cardiovascular disease, use of other medication, kidney function, transplantation-specific parameters, BMI, lipids and blood pressure did not materially change the association. Moreover, in Cox regression analyses, both thiazide and loop diuretics associated with the development of PTDM, independent of age, sex, FPG and HbA(1c) [HR 2.70 (95% CI 1.24-5.29); P = 0.012 and HR 5.08 (95% CI 2.49-10.34); P <0.001), respectively]. Conclusions This study demonstrates that diuretics overall are associated with an increased risk of developing PTDM in KTRs, independent of established risk factors for PTDM development. The association was present for both thiazide and loop diuretics.