慢性肾病患者肾小球滤过率评估方程诊断价值Meta分析

Diagnostic value of evaluating glomerular filtration rate equations in patients with chronic kidney disease:a Meta-analysis

目的采用Meta分析方法评价eGFR方程对评估慢性肾病肾小球滤过率的诊断价值。方法计算机检索The Cochrane Libray(2015年第1期)MEDLINE、PubMed、EMbase、CBM、CNKI、万方数据,查找有关慢性肾病患者eGFR方程诊断价值的研究,检索时限均为建库至2016年6月。两位研究者按纳入与排除标准独立筛选文献、提取资料和评价质量后,采用RevMan 5.2软件、Meta-Disc软件进行Meta分析。结果最终纳入22个研究观察对象,共纳入6348例CKD患者。Meta分析结果显示:不同eGFR方程与参考方法 r GFR的测定结果具有较好的相关性,CysC-GFR方程GFR均值比参考方法低[MD=-2.59,95%CI(-5.22,0.04),P=0.05]。CKD-EPI(Cr-GFR)方程GFR均值与对照组差异无统计学意义[MD=-0.49,95%CI(-1.22,2.20),P=0.57]。MDRD方程GFR均值与参考方法差异无统计学意义[MD=--3.68,95%CI(-9.87,2.51),P=0.24]。C-G方程GFR均值比参考方法偏高[MD=2.20,95%CI(0.47,3.93),P=0.01]。CKD-EPI(Cr-CysC)方程GFR均值比参考方法偏低[MD=-3.95,95%CI(-5.79,-2.10),P<0.0001]。从MDRD方程亚组分析看,GFR60-89ml/min、GFR30-59ml/min,CKD分期GFR实际评估效应最好,GFR>90ml/min,出现高值区低估现象。结论 eGFR方程与参考方法测定rGFR具有良好的相关性,eGFR方程CKDEPI(Cr-GFR)、MDRD与参考方法 GFR平均水平无差异,但是不同CKD分期eGFR评估水平有偏倚。鉴于eGFR方程评估参数不统一,异质性较大,尚需进一步验证。

Objective To evaluate eGFR evaluation equation of the diagnostic value of glomerular filtration rate in chronic kidney disease by using the Meta analysis method. Methods We comprehensibly searched databases including The Cochrane Library（Issue 1,2015）,MEDLINE,PubMed,EMbase,CBM,CNKI,Wan Fang Data from inception to June 2016,collected studies about the diagnostic value of evaluating glomerular filtration rate equations in Patients with chronic kidney disease. Two researchers independently screened the literature,extracted data,and evaluated the quality of the research by inclusion and exclusion criteria. And then meta analysis was carried out by using RevMan 5.2 software and Meta-Disc software. Results A total of 22 studies were selected finally,6348 cases of CKD patients were enrolled in this study. Meta analysis results show that,the different eGFR estimation equation and the reference method r GFR the determination results have good correlation,GFR value of Cys C-GFR equation is lower than that of reference method [MD=-2.59,95%CI（-5.22,0.04）,P=0.05]. GFR value has no significant difference between the CKD-EPI（Cr-GFR） and the reference method [MD =-0.49,95% CI（-1.22,2.20）,P =0.57]. GFR value has no significant difference between the MDRD equation and the reference method [MD=--3.68,95%CI（-9.87,2.51）,P=0.24]. GFR value of C-G equation is higher than that of reference method [MD=2.20,95%CI（0.47,3.93）,P=0.01]. GFR value of CKD-EPI（CrCys C） equation is lower than that of reference method[MD=-3.95,95%CI（-5.79,-2.10）,P〈0.0001]. From the analysis of MDRD equation subgroup we can know that,GFR60-89ml/min,GFR30-59ml/min,CKD staging GFR the actual assessment of the effect is the best,GFR〉90ml/min,has a high value area of underestimated phenomenon. Conclusion The eGFR equation and the reference method for the determination of rGFR have good correlation. The average level of GFR has no difference between eGFR estimating equation CKD-EPI（Cr-GFR）,MDRD and the reference method,however,the level of eGFR has a bias in different CKD stages. Given the not unified evaluation parameters and large heterogeneous of the eGFR equation,we still need to further verify.