Clinical risk factors and prediction modeling of post-transplantation diabetes mellitus in kidney recipients / 中华器官移植杂志

ABSTRACT

Objective:To explore the clinical risk factors of post-transplantation diabetes mellitus (PTDM) and establish a risk prediction model in kidney recipients.Methods:The clinical data and postoperative bedside measurements of blood glucose (BG) were retrospectively reviewed for 305 renal transplant recipients at First Affiliated Hospital of Zhejiang University School of Medicine from October 2018 to August 2019.According to whether or not PTDM occurred, they were assigned into two groups of PTDM (n=34) and non-PTDM (n=271). Risk factors were screened through single/multi-factor Logistic regression and PTDM prediction model was established.Results:The incidence rate of PTDM was 11.15%(34/305). Logistic regression analysis indicated that deceased donor, age ≥40 years, female, pre-hemoglobin A1c (Pre-HbA1c) and postoperative bedside BG value ≥11.1 mmol/L were the correlated factors for the occurrence of PTDM.Among them, female ( OR=3.13, 95% CI: 1.28-7.61), Pre-HbA1c ( OR=2.05, 95% CI: 1.12-3.74) and BG ≥11.1 mmol/L at 4pm Day 2/3 post-operation ( OR=19.08, 95% CI: 4.34-83.87) were risk factors for the occurrence of PTDM, The area under the model curve was 0.86 (95% CI: 0.79-0.93) with a Jordan index of 0.65, a sensitivity of 82.8% and a specificity of 82.3%. Conclusions:Female, Pre-HbA1c and fasting BG at 4 pm Day 2/3 post-operation ≥ 11.1 mmol/L are risk factors for the occurrence of PTDM.The prediction model has a decent predictive value.It is conducive to early clinical interventions and lowering the incidence rate of PTDM.