汉族人尿白蛋白的评价方法及尿白蛋白排泄率的相关因素研究

【作者】 樊晓红；

【导师】 李学旺；

【作者基本信息】 北京协和医学院 ， 内科肾脏病学， 2011， 博士

【摘要】 背景及目的微量白蛋白尿(MA)是慢性肾脏病(CKD)、心血管疾病(CAD)预后及死亡的独立预测因子,因此在CKD及CAD高危人群中进行MA筛查至关重要。晨尿白蛋白肌酐比值(ACR)因与尿白蛋白排泄率(UAE)相关性好且简便易行,已广泛用于临床及流行病学调查。在诊断白蛋白尿上,UAE已有着国际公认的界值,而ACR的界值各指南不一,在汉族人群更是缺乏设计良好的研究。目前在糖尿病、高血压等疾病以及普通人群中的MA相关因素研究表明早期检测MA对于预防各类疾病的发展和靶器官损害具有重要意义。肥胖,作为MA的相关因素之一,其衡量指标诸多。对于哪种指标能更好地预测MA,国外已有相关研究,但尚无定论；迄今国内尚无肥胖的各评价指标与MA相关关系的研究。另外,尿酸作为MA的相关因素之一,国外已经有许多研究证明尿酸肾损害的风险在女性中高于男性,但在尿酸肾损害作用的性别差异上,国内外的研究结果尚不一致。目前国内外关于MA相关因素的研究中,白蛋白尿的评价方法多为ACR,应用UAE作为评价方法的MA相关因素的研究非常少见。因此本研究拟以8小时过夜尿UAE作为MA的诊断标准,研究不同方法诊断MA的准确性、确定中国汉族人尿ACR诊断MA的界值以及影响UAE的相关因素。方法本研究对象主要来自于在北京平谷区随机抽样获得的并接受“代谢综合征肾损害”流行病学调查的人群；所有受试者均行问卷调查、人体学指标测量、留取首次中段晨尿和8小时过夜尿检测尿白蛋白及肌酐,以及抽取空腹静脉血检测生化指标(如血脂、肾功、空腹血糖、尿酸、高敏C反应蛋白等)。分别用健康人群95%分位数法及以UAE为金标准、应用受试者工作特征曲线(ROC曲线)方法确定MA的诊断界值。分别用多元线性回归及Logistic回归研究UAE相关因素。结果1、一般资料：1056名受试者中,男性494人、女性562人,平均年龄为50.2+12.1岁。2、(1)过夜段尿和晨尿间ACR的相关性(r=0.918,P<0.001)优于尿白蛋白浓度(UAC)(z=23.7,P<0.001)；(2)UAE与晨尿ACR间相关性(r=0.662,P<0.001)优于UAC(z=2.68,P=0.007)；(3)ROC曲线分析显示：对于诊断白蛋白尿,段尿ACR优于段尿UAC (P=0.001)、晨尿ACR(P<0.001)及晨尿UAC(P<0.001),但段尿UAC、晨尿ACR及晨尿UAC间比较无显著差异(P>0.05)。3、ROC曲线确定诊断MA的ACR界限值：(1)下界值：男性为1.95mg/mmol(敏感性97.6%,特异性88.6%),女性为3.62mg/mmol(敏感性83.8%,特异性89.1%),总体受试者为ACR为2.78mg/mmol(敏感性88.7%，特异性为86.0%):(2)上界值:总体受试者为22.59mg/mmol(敏感性100%，特异性98.8%)。(3)诊断一致性检验显示本研究性别特异性诊断界值的敏感性91.3%，特异性88.2%，阳性及阴性似然比为7.96和0.1，阳性及阴性预测值为56.9%和98.4%。4、ROC曲线分析显示:各人体学指标(体重指数、腰围、腰围臀围比、腰围身高比、锥削度指数)对于MA的预测无显著差异。多元线性回归显示:在女性中，经多因素校正后仅体重指数仍与UAE显著相关，单是在男性中上述各人体学指标与UAE均无显著相关。5、Logistic回归显示:男性、血压、血糖、血尿酸以及HsCRP是为女性受试者UAE的独立相关因素(尿酸的OR=7.129，95%Cl=2.123-23.943，P=0.001;HsCRP的OR=2.491，95%CI=l.241-5.001，P=0.010)，而男性中未见血尿酸及HsCRP的明显作用。结论1.对于微量白蛋白尿的诊断，段尿优于次尿，ACR优于UAC;2.汉族人成人以晨尿诊断微量白蛋白尿时，ACR界值存在性别差异，分别为男性为1.95mg/mmol，女性为3.62mg/mmol;3.各人体学指标在预测MA上无显著差异，与UAE相关性有性别差异，在女性中显著相关;4.UAE的独立相关因素有男性、高血压、糖尿病、尿酸及HsCRP升高，其中尿酸及HsCRP对于UAE的影响存在性别差异，女性更为显著。更多还原

【Abstract】 Background and aimsMicroalbuminuria (MA) is an independent risk factor for cardiovascular diseases, renal outcome, as well as all-cause mortality. So it is very important to screen for MA for individuals at increased risk for chronic kidney disease (CKD) and cardiovascular diseae(CVD). Albumin creatinine ratio(ACR) in a first-void morning urine sample, as a simple and comparably accurate way for detecting MA, is widely recommended. There is world-wide acceptted cutoff of urinary albumin excretion(UAE) for defining MA, whereas the cutoff of ACR for MA varies among guidelines and has not yet been well-established in Chinese. Hence, well-designed study is sorely needed for determining the cutoff of ACR for Chinese. In addition, improvement in MA was found to be associated with better outcomes of CAD and CKD, so it is essential to find out the potential risk factors for MA. Obesity, as a risk factor for MA, can be estimated by different anthropometric parameters. But it remains elusive which of them is best for predicting MA. Up to now there is no report about the relationship between anthropometric parameters and MA in China. On the other hand, hyperuricemia, as the relative factor for MA, seemed to result in more CKD in female than male, but the opposite effects on CKD was reported in China, so further research is necessary. Because urinary creatinine in ACR is the confounding factor, UAE is the better method for detecting albuminuria in the research of relative factors for MA, which is rare used because of the inconvenient collection. We used eight-hour overnight UAE as the standard assessment of albuminuria, and assess to investigate the accuracy of urinary albumin measurements for detecting MA, the Chinese gender-specific discriminator value of ACR for MA, and the UAE associated relative factors in Chinese general population.MethodsThe epidemiological study regarding relationship between metabolic syndrome and chronic kidney disease in Pinggu district, Beijing has been performeded in our research. The investigated subpopulation devoid of known renal disease was analyzed. Another 59 participants were student volunteers, aged 20 to 29 years. The participants completed a standardized interview including a medical questionnaire, a physical examination, collection of the first morning urine and 8-hour overnight urine sample, and a series of laboratory tests (e.g. kidney function, lipid, glucose, serum uric acid, hypersensitive C-reactive protein, urinary albumin and creatinine). Eight-hour overnight UAE was referred as the standard assessment for albuminuria. Receiver Operating Characteristic Curve (ROC) analysis was used to determine the ACR cut-off value for microalbuminuria. The multiple linear regression and Logistic regression analyzed the relative factors of UAE. Spss (version 18.0)and Medcalc (version 11.1.1.0) were used for the analysis.Results1、A total of 1056 study participants (494 males and 562 females,50.2±12.1 years of age)were enrolled.2、(1)The ACR of the overnight urine and first-void morning urine had the best correlation(r=0.918, P< 0.001) compared with UAC (z=23.7,P<0.001); (2) ACR in first morning void urine also had the better correlation with 8h-UAE(r=0.662, P=0.000; z=2.68, P=0.007). (3) According to ROC curve analyses, the overnight-urine ACR, as a measure to define albuminuria, was better than ACR and UAC of the morning urine and overnight-urine UAC. There was no significant difference among ACR and UAC of the morning urine.3、(1)The discriminator value of ACR for microalbuminuria by ROC curve analysis was 1.95mg/mmol (sensitivity 97.6% and specificity 88.6%) for men,3.62mg/mmol (sensitivity 83.8% and specificity 89.1%) for women and 2.78mg/mmol (sensitivity 88.7% and specificity 85.9%) for overall. (2)The upper boundary of microalbuminuria by ROC curve analysis was 22.59mg/mmol (sensitivity 100% and specificity 98.8%). (3) According to the inter-rater agreement analysis, the gender-specific discriminator values for ACR derived from the present study had a sensitivity of 91.3% and a specificity of 88.2%. And its positive and negative likelihood ratios were respectively 7.56 and 0.1, and its positive and negative predictive values were respectively 56.9% and 98.4%.4、The AUC estimates for the prediction of abnormal UAE by five Anthropometric indicators performed no significant difference. The multiple linear analysis showed body mass index (BMI) was significantly related to 1gUAE in overall and females, but showed no relationship in males.5、In multiple logistic regression, male, hypertension, diabetes, hyperuricemia and elevated hsCRP were significantly associated with albuminuria. If stratified by gender, the associations of albuminuria with diabetes, hyperuricemia and elevated hsCRP remained significant in females, whereas in males only that of albuminuria with hypertension and diabetes persisted significantly. (serum uric acid OR=8.997,95%CI=2.640-30.668,P<0.001; HsCRP OR=2.570,95%CI=1.258-5.250, P=0.010).Conclusions1. The timed urine and ACR had better accuracy compared with random urine and UAC when detecting MA.2、The genetic-specific ACR discriminator value for determining microalbuminuria in Han Chinese was 1.95mg/mmol for male and 3.62mg/mmol for female.3、Five anthropometric indicators performed no significant difference in predicting abnormal UAE, and had significant relationship wirth UAE especially in females.4、The potential risk factors of UAE included male, hypertension, diabetes, hyperuricemia and elevated HsCRP, but hyperuricemia and elevated HsCRP had the genetic-specific effect on UAE, especially in females.更多还原