随意尿和晨尿白蛋白／肌酐比值在社区中老年糖尿病高危人群白蛋白尿筛查中的应用评价

The ratio of urine albumin to creatinine used in screening for albuminuria among the middle-aged and elderly at high-risk for diabetes in an urban community with casual and morning urine specimens

目的比较随意尿和晨尿中自蛋白与肌酐比值（ACR）在社区糖尿病高危人群白蛋白尿筛查中的应用价值。方法选取参加糖尿病高危人群筛查的社区人群共443名（男234例，女209例），平均年龄（59±8）岁。收集筛查当日的随意尿和晨尿以及随后2次晨尿，每次间隔2周以上，测定尿ACR。以3次晨尿中2次或2次以上尿ACR≥30mg／g为白蛋白尿诊断标准。结果根据白蛋白尿诊断标准，白蛋白尿患病率为6．3％（28／443）。筛查当日的随意尿ACR水平高于晨尿（7．1mg,／g与3．2mg／g，P〈0．01）；随意尿筛查白蛋白尿的阳性率高于晨尿（14．9％与5．9％，P〈0．05）；敏感度和阴性预测值（NPV）与晨尿相似（96．4％与92．9％；99．7％与99．5％，均P〉0．05），特异度和阳性预测值（PPV）低于晨尿（90．6％与100．0％；40．9％与100．0％，均P〈0，05）。根据受试者工作特征（ROC）曲线，随意尿ACR筛查白蛋白尿的适宜切点为40．0mg／g，其敏感度96．4％，特异度95．2％，PPV57．5％，NPV99．8％。与晨尿和随意尿白蛋白均阴性者相比，晨尿白蛋白阴性而随意尿阳性者具有更多的心血管病和慢性肾脏病危险因素。结论社区中老年糖尿病高危人群筛查白蛋白尿以晨尿为佳，但作为筛查手段，随意尿也有其优势，建议将尿ACR40．0mg／g作为该人群随意尿筛查白蛋白尿的适宜切点。

Objective To compare the value of clinical application of the ratio of urine albumin to creatinine （ACR） in the community-dwelled middle-aged and elderly at high-risk for diabetes in screening for albuminuria with casual and morning urine specimens. Methods Totally, 443 middle-aged and elderly people （234 men and 209 women） at high-risk for diabetes, with an average age of （59 ± 8 ） years, were recruited from Changfeng Community of Shanghai in screening for albuminuria with casual and morning urine specimens collected on the same day, as well as another two morning urine specimens collected at an interval of two weeks, for determination of urine ACR. Albuminuria was defined as ACR equal to or more than 30 mg （albumin）/g （creatinnine） in two or more of the three urine specimens collected. Results Overall prevalence of albuminuria was 6. 3% （28/443） in the middle-aged and elderly according to diagnostic standard mentioned above, significantly higher in casual urine specimens than in morning urine specimens （ 14. 9% vs. 5.9% , P 〈0. 05）. Urine ACR of casual urine specimens was significantly higher than that of morning urine （7. 1 mg/g vs. 3.2 mg/g, P 〈0. 001 ）. There was no significant difference in sensitivity and positive predictive value （PPV） between casual and morning urine specimens in screening for albuminuria （96. 4% vs. 92. 9%, and 99.7% vs. 99. 5%, respectively, P 〉0.05）. Specificity and negative predictive value （NPV） of casual urine specimens was significantly lower than those of morning urine specimens （90. 6% vs. 100.0%, and 40. 9% vs. 100.0%, respectively, P 〈0. 05）. The optimal cut-off value of ACR by casual urine specimens was 40. 0 mg/g in screening for albuminuria, based on receiver operating characteristic （ROC） curve, with sensitivity of 96. 4%, specificity 95.2%, PPV 57.5% and NPV 99. 8%. The middle-aged and elderly with negative albuminuria in morning urine but positive in casual urine had more risk factors for cardiovascular and chronic kidney diseases than those with negative albuminuria in both morning and casual urine specimens. Conclusions Morning urine specimen is better than casual urine for albuminuria screening in the middle-aged and elderly people with risk factors for diabetes in urban community, but casual urine used in screening for albuminuria is very convenient for collection. It is suggested that urine ACR of 40. 0 mg/g be used as an optimal cut-off value for casual urine in albuminuria screening among them.