糖尿病肾脏病进展风险因素和预测模型构建

Construction of Risk Factors and Predictive Models for the Progression of Diabetic Kidney Disease

目的 探究2型糖尿病患者合并糖尿病肾脏病微量白蛋白尿期进展至大量白蛋白尿期的危险因素,开发和验证辅助临床预测糖尿病肾脏病进展的预测模型。方法 依据纳入标准和排除标准选取2018年12月1日—2021年12月1日于辽宁中医药大学附属医院内分泌科、肾内科、内分泌康复科住院并诊断为糖尿病肾脏病微量白蛋白尿期的患者病历资料834例,按7∶3的比例随机分为训练集584例、验证集250例,收集患者基本资料、人体测量指标、2型糖尿病病程、合并病、中医标证、实验室指标、复查指标,并参照复诊结果依据分期标准将训练集分为微量白蛋白尿组442例和大量白蛋白尿142例组。应用SPSS 26.0进行统计分析,通过多因素Logistic回归分析建立预测方程,绘制ROC曲线,外部诊断性验证,评价糖尿病肾脏病预测方程的临床有效性。结果 糖尿病肾脏病微量白蛋白尿期进展至大量白蛋白尿期的独立危险因素为性别(男)、高血压病病史、糖尿病视网膜病变病史、湿证、湿热证、糖化血红蛋白(HbA1c)(≥7%)、低密度脂蛋白胆固醇(LDL-C)[≥2.6 mmol·L^(-1)(无冠心病)/1.8 mmol·L^(-1)(合并冠心病)]、血清白蛋白(ALB)(<35 g·L^(-1))、钙离子(Ca2+)、游离三碘甲状腺原氨酸(FT3)、胱抑素C(CysC)。预测方程ROC曲线下面积为0.920(95%CI:0.895~0.946),外部验证提示预测方程一致性良好。结论 11个独立危险因素构成的预测方程可用于预测2型糖尿病合并糖尿病肾脏病微量白蛋白尿期进展至大量白蛋白尿期的风险。

Objective To explore the risk factors of progression from microalbuminuria to macroalbuminuria in patients with type 2 diabetes mellitus complicated with diabetic kidney disease(DKD),and to develop and validate a predictive model to assist clinical prediction of the progression of DKD.Methods According to the inclusion and exclusion criteria,the medical records of 834 patients who were hospitalized and diagnosed with DKD with microalbuminuria in the department of endocrinology,department of nephrology and department of endocrine rehabilitation of the affiliated hospital of Liaoning university of traditional Chinese medicine from December 1,2018 to December 1,2021 were selected. According to the ratio of 7∶3,they were randomly divided into 584 cases in the training set and 250 cases in the validation set. The basic information of patients,anthropometric indicators,type 2 diabetes course,comorbidities,TCM standards,laboratory indicators,and re-examination indicators were collected. The training set was divided into 442 cases of microalbuminuria group and 142 cases of macroalbuminuria group according to the staging standard. SASS 26.0 was used for statistical analysis,multivariate Logistic regression analysis was used to establish the prediction equation,ROC curve was drawn,external diagnostic verification was performed,and the clinical validity of the prediction equation for DKD was evaluated. Results The independent risk factors for the progression from microalbuminuria stage to macroalbuminuria stage of DKD were gender(male),history of hypertension,history of diabetic retinopathy,wet syndrome,damp-heat syndrome,HbA1c(≥ 7%),LDL-C[ ≥ 2.6 mmol·L^(-1)(without coronary heart disease)/1.8 mmol·L^(-1)(with coronary heart disease)],ALB(<35 g·L^(-1)),Ca2+,FT3,CysC. The area under the ROC curve of the prediction equation was 0.920(95%CI:0.895-0.946),and the external validation indicated that the prediction equation was consistent. Conclusion The prediction equation composed of 11 independent risk factors can be used to predict the risk of progression from microalbuminuria to macroalbuminuria in type 2diabetes mellitus complicated with DKD.