首页|基于列线图预测糖尿病性视网膜病变并发糖尿病肾病的风险

基于列线图预测糖尿病性视网膜病变并发糖尿病肾病的风险

扫码查看
目的 构建预测糖尿病视网膜病变(DR)患者发生糖尿病肾病(DN)的列线图模型,为DR患者的DN筛查提供工具,并评价此模型的区分度与一致性.方法 回顾性临床研究.纳入自2020年1月至2022年12月在我院眼科和内分泌科住院就诊的糖尿病患者140例,根据是否并发DN分为非DN组与DN组,收集患者临床资料并比较两组间差异;通过Logistic回归模型分析DR患者并发DN的危险因素,构建列线图预测模型,并进行区分度与一致性检验.结果 140例DR患者发生DN者88例.与非DN组相比,DN组患者糖尿病病程、体质指数、收缩压、舒张压、血清尿素氮、血清肌酐、低密度脂蛋白、总胆固醇、尿酸、糖尿病性黄斑水肿比例、NPDR与PDR比例水平均升高(均P<0.05),肾小球滤过率水平降低(P<0.05).二元Logistics回归分析发现,收缩压(OR=1.05,P<0.05)、体质指数(OR=1.20,P<0.05)、总胆固醇(OR=1.77,P<0.05)、尿酸(OR=1.01,P<0.05)、血清尿素氮(OR=1.59,P<0.05)、糖尿病性黄斑水肿比例(OR=11.66,P<0.05),是DR患者发生DN的独立危险因素;以收缩压、体质指数、总胆固醇、尿酸、血清尿素氮、糖尿病性黄斑水肿比例六项因素构建预测DR患者并发DN的列线图模型,ROC曲线显示AUC 0.920(0.878~0.962),证明区分度较好.校准曲线显示斜率接近1.结论 基于收缩压、体质指数、总胆固醇、尿酸、血清尿素氮、糖尿病性黄斑水肿比例六项因素构建的列线图模型对DN具有诊断价值,为临床上指导诊断预测DN发挥作用.
Predicting the risk of diabetic retinopathy complicated by diabetic nephropathy based onnomogram
Objective To construct a nomogram model for predicting diabetic nephropathy(DN)in patients with diabetic retinopathy(DR)to provide a tool for DN screening in DR patients,and to evaluate the discrimination and consis-tency of this model.Methods A retrospective study.140 diabetic patients who were hospitalized in the department of oph-thalmology and endocrinology of our hospital from January 2020 to December 2022 were included.They were divided into non-DN group and DN group according to whether they were complicated by DN.The clinical data of the patients were col-lected and the differences between the two groups were compared.The logistic regression model was used to analyze the risk factors of DR patients complicated with DN,a nomogram prediction model was constructed,and the discrimination and consistency tests were performed.Results Among 140 DR patients,88 cases developed DN.Compared with the non-DN group,patients in the DN group had diabetes duration,body mass index,systolic blood pressure,diastolic blood pres-sure,serum urea nitrogen,serum creatinine,low-density lipoprotein,total cholesterol,uric acid,proportion of diabetic macular edema,and the proportion of non-proliferative DR(NPDR)and the proliferative DR(PDR)were increased(all P<0.05),glomerular filtration rate decreased(P<0.05).Binary logistic regression analysis found that systolic blood pressure(OR=1.05,P<0.05),body mass index(OR=1.20,P<0.05),total cholesterol(OR=1.77,P<0.05),uric acid(OR=1.01,P<0.05),serum urea nitrogen(OR=1.59,P<0.05),and diabetic macular edema ratio(OR=11.66,P<0.05)were independent risk factors for DN in patients with DR.The systolic blood pressure,body mass index,total six factors in-cluding cholesterol,uric acid,serum urea nitrogen,and diabetic macular edema ratio were used to construct a nomogram model for predicting DN in DR patients.The receiver operating curve(ROC)showed the area under the curve AUC was0.920(0.878~0.962),which proved that the discrimination was good.The calibration curve showed a slope close to 1.Conclusions The nomogram model constructed based on six factors including systolic blood pressure,body mass index,total cholesterol,uric acid,serum urea nitrogen,and diabetic macular edema ratio has diagnostic value for DN and plays a role in guiding the clinical diagnosis and prediction of DN.

Diabetic retinopathyDiabetic nephropathyDisk factorsMacular edemaNomogram

管一鸣、桂衍超、桂思语、季之遇、林思宇、蒋正轩

展开 >

230000 合肥,安徽医科大学第二附属医院眼科

糖尿病性视网膜病变 糖尿病肾病 危险因素 黄斑水肿 列线图

2024

临床眼科杂志
安徽医科大学,安徽眼科研究所

临床眼科杂志

CSTPCD
影响因子:0.791
ISSN:1006-8422
年,卷(期):2024.32(3)