检验医学 ›› 2012, Vol. 27 ›› Issue (5): 340-344.

• 生物化学检验论著 • 上一篇    下一篇

实验室肾功能相关检测项目在诊断慢性肾病中的特性分析

王敏,王琦,顾洁,袁维祥,熊立凡   

  1. 1.上海市长宁区同仁医院检验科,上海 200050; 2.上海市长宁区同仁医院肾脏科,上海 200050;3.上海交通大学医学院附属仁济医院检验科,上海 200001
  • 收稿日期:2012-03-13 修回日期:2012-04-15 出版日期:2012-05-30 发布日期:2012-05-10
  • 作者简介:王 敏,女,1973年生,主管技师,主要从事生化检验工作。

Diagnostic Performances of Laboratory kidney Function Tests for Chronic Kidney Disease

  1. 1. Department of Clinical Laboratory,St. Luke′s Hospital, Shanghai 200050,China;2. Department of Nephrology, St. Luke′s Hospital, Shanghai 200050,China;3. Department of Clinical Laboratory, Renji Hospital, Shanghai Jiaotong University School of Medicine,Shanghai 200001,China
  • Received:2012-03-13 Revised:2012-04-15 Online:2012-05-30 Published:2012-05-10

摘要: 目的 分析实验室肾功能检查项目在慢性肾病中的诊断特性。方法 将73例慢性肾病患者以肾小球滤过率估算值(eGFR)分成3组[慢性肾病A组:33例,eGFR>60 mL/(min·1.73 m2);慢性肾病B组:32例,eGFR为30~60 mL/(min·1.73 m2);慢性肾病C组:8例,eGFR<30 mL/(min·1.73 m2)]。分别测定慢性肾病组、慢性肾病高危组(糖尿病或高血压40例)和正常对照组(32名)随机尿微量白蛋白(U-mALb)/肌酐(Cr)比值、随机尿蛋白(UP)/Cr比值、24 h尿微量白蛋白 (24 h U-mALb)、24 h尿蛋白(24 h UP)以及血清尿素(Urea)、Cr、尿酸(UA)、半胱氨酸蛋白酶抑制剂C(Cys C)水平。比较各组检测结果,并以慢性肾病的检验项目为判断金标准[24 h UP的临界值>150 mg或eGFR<60 mL/(min·1.73 m2)],分别计算其他检验项目的诊断性能(诊断灵敏度、特异性、阴性预期值、阳性预期值、阴性似然比、阳性似然比和准确性)。结果 慢性肾病A、B、C 3组之间U-mALb/Cr比值、UP/Cr比值、24 h U-mALb及血清Urea、UA、Cys C检测结果差异均有统计学意义(P<0.05)。慢性肾病A组与慢性肾病高危组比较,除了UA无明显差异外,其余项目检测结果差异均有统计学意义(P<0.05)。慢性肾病高危组和正常对照组之间仅UA、Cys C水平的差异有统计学意义(P<0.05)。U-mAlb/Cr比值、UP/Cr比值、24 h U-mAlb及血清Urea、UA、Cys C对慢性肾病的诊断性能中阴性似然比分别为0.17、0.22、0.14、0.67、0.66、0.51;阳性似然比分别为7.73、78、28.7、34、3.73、5.00;准确性分别为0.87、0.88、0.92、0.66、0.65、0.74。结论 24 h U-mALb在慢性肾病诊断中准确性最高,UP/Cr比值和U-mALb/Cr比值对慢性肾病诊断的准确性最接近24 h U-mALb,故适用于临床筛检慢性肾病。而eGFR主要应用于慢性肾病的分型。

关键词: 微量白蛋白, 尿蛋白, 肌酐, 24 h尿, 随机尿, 诊断性能

Abstract: Objective To analyze the diagnostic performances of laboratory kidney function tests for chronic kidney disease.  Methods A total of 73 patients with chronic kidney disease were classified into 3 groups according to different estimated glomerular filtration rate (eGFR):Group A(33 cases),eGFR:>60 mL/(min·1.73 m2);Group B(32 cases),eGFR: 30-60 mL/(min·1.73 m2);Group C(8 cases),eGFR:<30 mL/(min·1.73 m2). The patients with chronic kidney disease,40 patients with high risk of chronic kidney disease (with diabetes mellitus or hypertension) and 32 healthy subjects were compared with the testing of random urine microalbumin (U-mALb)/creatinine (Cr) ratio, random urine protein (UP)/Cr ratio, 24h-urine microalbumin(24h-UmALb),24h-urine protein(24hUP),serum urea,Cr,uric acid(UA) and cystatin C(CysC). With the gold standard methods [the cut-off value of 24h UP or eGFR was >150 mg or <60 mL/(min·1.73 m2),respectively], the diagnostic sensitivity, specificity, negative predictive value, positive predictive value, negative and positive likelihood ratios and accuracy were calculated. Results There were significant differences in the results of all tests(U-mALb/Cr, UP/Cr, 24hU-mALb, urea, UA and CysC) between the 3 groups of chronic kidney disease(P<0.05). There were also significant differences in the results of all the tests except UA between the Group A of chronic kidney disease and the group with high risk of chronic kidney disease(P<0.05). There only existed significant difference in the results of UA and CysC between the group of healthy subjects and the group with high risk of chronic kidney disease(P<0.05). The diagnostic performances of U-mAlb/Cr, UP/Cr, 24 h U-mAlb, urea, UA and Cys C were as follows:negative likelihood ratios were 0.17,0.22,0.14,0.67,0.66 and 0.51;positive likelihood ratios were 7.73,78,28.7,34,3.73 and 5.00;accuracies were 0.87,0.88,0.92,0.66,0.65 and 0.74. Conclusions The 24hU-mALb has the best accuracy in the diagnosis of chronic kidney disease. UP/Cr and U-mALb/Cr may be used as screening tests for the diagnosis of chronic kidney disease,because their accuracies of diagnostic performance for chronic kidney disease are closest to that of 24 h U-mALb, and eGFR may be mainly used in the classification of chronic kidney disease.

Key words: Microalbumin, Urine protein, Creatinine, 24h-urine, Random urine, Diagnostic performance