尿白蛋白与肌酐比值升高对老年2型糖尿病患者随访新发心力衰竭的预测价值分析  

Analysis of the predictive value of increased urinary albumin creatinine ratio for new-onset heart failure in elderly patients with type 2 diabetes

作　　者：刘爱英 杨静 冯利娜 Liu Aiying;Yang Jing;Feng Lina(Department of Internal Medicine,Zhangjiakou Maternal and Child Health Hospital,Zhangjiakou 075001,China)

机构地区：[1]张家口市妇幼保健院普通内科,075001

出　　处：《中国心血管杂志》2023年第4期361-366,共6页Chinese Journal of Cardiovascular Medicine

基　　金：张家口市软科学研究项目(1321170I)。

摘　　要：目的探索尿白蛋白与肌酐比值(uACR)升高对老年2型糖尿病(T2DM)患者随访新发心力衰竭(NOHF)的预测价值,并分析将uACR纳入现有模型是否可以提高预测性能。方法回顾性队列研究。连续纳入2015年1月至2019年2月在张家口市妇幼保健院住院治疗的632例老年T2DM患者。检测基线的尿白蛋白和血肌酐水平计算uACR,并根据uACR分为三组,包括正常蛋白尿组(<3 mg/mmol,310例)、微量白蛋白尿组(3~30 mg/mmol,215例)和大量白蛋白尿组(>30 mg/mmol,107例)。取随访数据截至2022年12月。比较三组的基线资料,采用生存分析比较三组随访4年的NOHF风险。采用Cox风险比例回归模型和限制性次样条曲线评估uACR与NOHF的关系,采用受试者工作特征曲线下面积(AUC)、净重分类改进(NRI)和综合判别改进(IDI)比较将uACR纳入现有模型前、后的预测NOHF的性能。结果共纳入632例患者,平均年龄为(66.4±7.2)岁,男性421例(66.6%)。三组患者的年龄、体质指数、糖尿病病程、合并高血压、高敏C反应蛋白、空腹血糖、餐后2 h血糖、糖化血红蛋白、血肌酐、uACR、肾小球滤过率、左室整体纵向应变、左室质量指数等差异均有统计学意义(均为P<0.05)。平均随访时间为(43.0±5.9)个月,有35例(5.5%)NOHF,三组分别有8例(2.6%)、13例(6.0%)和14例(13.1%)。生存分析曲线显示,三组的NOHF风险差异有统计学意义(log-rankχ^(2)=15.121,P=0.001)。调整协变量后,与正常蛋白尿组比较,微量白蛋白尿组(HR=2.324,95%CI:1.117~4.835,P=0.024)和大量白蛋白尿组(HR=5.116,95%CI:1.403~18.655,P=0.013)的NOHF风险均增高。限制性三次样条曲线结果显示,随着uACR的增加,NOHF风险逐渐增加(P趋势<0.05)。加入uACR后,ROC曲线显示模型预测NOHF的准确性明显提高(AUC分别为0.692和0.785,P<0.001)。uACR可显著改善患者预后的分类,NRI为0.343(95%CI:0.101~0.585,P=0.006),IDI为0.032(95%CI:0.013~0.051,P=0.001)。结论uACR是预测老年T2DObjective To explore the predictive value of increased urinary albumin creatinine ratio(uACR)for new-onset heart failure(NOHF)in elderly patients with type 2 diabetes mellitus(T2DM),and to analyze whether incorporating uACR into existing models can improve the predictive performance.Methods This was a retrospective cohort study.A total of 632 elderly T2DM patients who were hospitalized in Zhangjiakou Maternal and Child Health Hospital from January 2015 to February 2019 were included consecutively.The baseline urinary albumin and serum creatinine levels were measured to calculate uACR,and were divided into three groups according to uACR:normal albuminuria group(<3 mg/mmol,310 cases),microalbuminuria group(3-30 mg/mmol,215 cases),and macroalbuminuria group(>30 mg/mmol,107 cases).Follow up data was collected up to December 2022.The baseline data of three groups were compared and the survival data was analyzed with Kaplan-meier survival analyses to compare the risk of NOHF among the three groups after a 4-year follow-up.The Cox risk proportional regression model and restricted sub spline curve were used to evaluate the relationship between uACR and NOHF.The predictive performance of incorporating uACR into existing models was compared using the area under the receiver operating characteristic curve(AUC),net weight classification improvement(NRI),and comprehensive discriminant improvement(IDI).Results A total of 632 patients were enrolled,with an average age of(66.4±7.2)years,and 421(66.6%)were males.There were significantly differences in age,body mass index,course of diabetes,rates of hypertension,high-sensitive C-reactive protein,fasting blood glucose,2h postprandial blood glucose,glycosylated hemoglobin,serum creatinine,uACR,glomerular filtration rate,left ventricular longitudinal strain,left ventricular mass index among the three groups(all P<0.05).The average follow-up time was(43.0±5.9)months,with 35 cases(5.5%)developed NOHF.There were 8 cases(2.6%),13 cases(6.0%),and 14 cases(13.1%)in the three groups,respe

关 键 词：老年人 糖尿病 尿白蛋白与肌酐比值 新发心力衰竭 预测价值 

分 类 号：R587.2[医药卫生—内分泌] R541.6[医药卫生—内科学]