New-onset diabetes after transplantation (NODAT) is a growing concern in transplantation. All modifiable risk factors are not yet identified. We assessed the relationship between baseline clinical and biochemical parameters and NODAT. Eight-hundred and fifty-seven in-Caucasian renal transplant recipients were included. Charts were individually reviewed. The follow-up was 5.3 years (ranges: 0.25-20.8; 5613 patient-years). The incidence of NODAT was 15.0%, 18.4% and 22.0% at 10, 15 and 20 years following transplantation. Age, body mass index (BMI), glucose (all P < 0.0001) and triglycerides [hazard ratio (HR) per 1 mmol/l: 1.44 [1.17-1.77], P = 0.0006] were potent risk factors whereas steroid withdrawal (HR: 0.69 [0.47-1.01], P = 0.0601) reduced the risk. As compared to cyclosporine, sirolimus (HR: 3.26 [1.63-6.49], P = 0.0008) and tacrolimus (HR: 3.04 [2.02-4.59], P < 0.0001) were risk factors for NODAT. The risk of NODAT was comparable for sirolimus (HR: 2.35 [1.06-5.19], P = 0.0350) and tacrolimus (HR: 2.34 [1.46-3.75], P = 0.0004) after adjustments on age, BMI, glucose and steroid withdrawal; however, unlike sirolimus, tacrolimus remained significant after adjustment on triglycerides. The risk of NODAT appeared similar, but its pathophysiology seemed different in sirolimus- and tacrolimus-treated patients; this observation needs confirmation. However, main independent risk factors were age, BMI, initial glucose and triglycerides.