RESUMO
PURPOSE: To identify the risk factors of calcineurin inhibitor (CNI)-associated new-onset diabetes mellitus (NODM) in chronic kidney disease (CKD) treatment. METHODS: We retrospectively screened patients treated with CNIs in our hospital from January 2015 to December 2018. The inclusion criteria were as follows: a clear diagnosis of CKD and patients receiving CNI treatment. We compared patients with and without CNI-associated NODM. RESULTS: Ninety-eight of the 336 assessed patients met the inclusion criteria, 15 (15.3% [15/98]) of whom developed CNI-associated NODM. Multiple logistic regression analysis revealed that baseline glycosylated hemoglobin (OR=4.141; 1.024-16.743; p=0.046) and CNI trough concentration (1 year) (OR=1.028; 1.009-1.047, p=0.004) were independent risk factors for NODM. In contrast, glucocorticoid type (prednisone) (OR=0.075; 0.011-0.526, p=0.009) was identified as an independent protective factor for NODM. Using a receiver operating characteristic curve, a cutoff cyclosporin A trough concentration of 102.1 ng/mL was identified as a predictive factor of NODM. Univariate logistic regression showed that the incidence of diabetes was significantly higher in patients with baseline glycosylated hemoglobin in non-diabetic range but higher than 5.65% (10.2% vs. 29.2%, p=0.038). One NODM patient (6.7% [1/15]) recovered at 12.7 months after the onset of diabetes mellitus. CONCLUSIONS: We recommend that more attention be paid to patients with baseline glycosylated hemoglobin in non-diabetic range but higher than 5.65% during CKD treatment with CNIs. High trough concentrations of cyclosporin A, particularly those >102.1 ng/mL, contribute to NODM. CNI-associated NODM may be reversible in the treatment of CKD.