Analysis of risk factors and establishment of prediction model for post transplantation diabetes mellitus in renal transplant recipients / 器官移植

ABSTRACT

Objective To analyze the risk factors for the occurrence of post transplantation diabetes mellitus (PTDM) in renal transplant recipients, establish a prediction model for PTDM and evaluate its prediction value. Methods Clinical data of 915 renal transplant recipients were retrospectively analyzed. According to the occurrence of PTDM, all recipients were divided into the PTDM group (n=78) and non-PTDM group (n=837). The main indexes of recipients were collected. The risk factors for the occurrence of PTDM in renal transplant recipients were analyzed by univariate and multivariate analysis. The prediction model for PTDM was established and its prediction value was evaluated. Results Family history of diabetes mellitus, body mass index (BMI), preoperative 2 h postprandial blood glucose and preoperative glycosylated hemoglobin were the independent risk factors for the occurrence of PTDM in renal transplant recipients. The prediction model for PTDM was logit (P)=2.199×family history of diabetes (yes=1, no=0)+0.109×BMI+0.151×2 h postprandial blood glucose (mmol/L)+0.508×glycosylated hemoglobin (%)-9.123. The results of receiver operating characteristic (ROC) curve showed that the area under the curve (AUC) of these 4 predictors combined for predicting PTDM in renal transplant recipients was 0.830 [95% confidence interval (CI) 0.786-0.873], the cut-off value was 0.0608, the sensitivity was 0.821, the specificity was 0.700, and the Youden index was 0.521 (P < 0.05). Conclusions Family history of diabetes mellitus, BMI, preoperative 2 h postprandial blood glucose and preoperative glycosylated hemoglobin are the independent risk factors for the occurrence of PTDM in renal transplant recipients. The prediction model for PTDM combined with4 predictors yield relatively high prediction value for PTDM.