CKD-EPI方程计算肾小球滤过率的误差分析与评估  被引量:5

Error analysis and evaluation in the calculation of GFR by CKD-EPI equation

在线阅读下载全文

作  者:赵宇[1] 陈奎[1] 张晓良[1] 

机构地区:[1]东南大学医学院,南京210009

出  处:《肾脏病与透析肾移植杂志》2015年第5期407-411,458,共6页Chinese Journal of Nephrology,Dialysis & Transplantation

基  金:国家自然科学基金(81370826);江苏省十二五医学重点人才项目(RC2011124);江苏省临床医学研究中心项目(BL2014080);中央高校基本科研业务费专项基金(KYLX15_180)

摘  要:目的:比较美国慢性肾脏病流行病合作组(CKD-EPI)三种计算肾小球滤过率(GFR)方法(CKDEPISCr、CKD-EPICys C与CKD-EPISCr/Cys C),估计其各自临床应用时的计算误差。方法:收集78例CKD患者同一时间段的血清肌酐(SCr)、血清胱抑素C(Cys C)和99mTc-DTPA肾动态显像资料,用三种CKD-EPI方程估算GFR(e GFR)与99mTc-DTPA肾动态显像检测的GFR参考值(r GFR)的误差ΔSCr,ΔCys C与ΔSCr/Cys C,作P-P图、Pearson相关分析与Bland-Altman散点图进行数据预处理,采用t检验与χ2检验分别计算误差的总体期望、方差的置信区间,由3σ原理得到误差范围。结果:正态性分析结果显示:误差服从正态分布。相关性与一致性分析显示:三种e GFR与r GFR显著相关(r=0.832、0.838、0.917),ΔSCr、ΔCys C与ΔSCr/Cys C分别有93.59%、96.15%、94.87%散点位于其各自的95%一致性界限内。t检验与χ2检验分析表明:ΔSCr,ΔCys C与ΔSCr/Cys C期望的95%CI分别为(-5.62,1.22)、(3,18,10.50)、(-6.42,-1.59),方差的95%的置信区间分别为(171.66,324.18)、(196.90,371.85)、(85.55,161.57),由3σ原理得到ΔSCr∈(-59.63,55.23),ΔCys C∈(-54.67,68.36),ΔSCr/Cys C∈(-44.55,36.54)。结论:CKDEPISCr/Cys C方程较CKD-EPISCr方程、CKD-EPICys C方程计算误差范围小,更适于临床GFR的估计。Objective: To compare the three equations of CKD-EPISCr,CKD-EPICys Cand CKD-EPISCr / Cys Cand evaluate the error of each equation in the clinical practice. Methodology: Seventy eight patients with CKD were enrolled into this study. The data of serum Cystatin C,creatinine and r GFR coming from99mTc-DTPA renal dynamic imaging were collected. Three e GFR( e GFRSCr,e GFRCys Cand e GFRSCr / Cys C) were calculated by the equation of CKD-EPISCr、CKD-EPICys C and CKD-EPISCr / Cys Crespectively. Errors( ΔSCr,ΔCys Cand ΔSCr / Cys C) were compared with r GFR. Proportion-proportion plot,Pearson correlation analysis,Bland-Altman scatter diagram,Student's t test and Chi-square test were performed for statistical analysis. Results: The data of ΔSCr,ΔCys Cand ΔSCr / Cys Cwere shown by Gaussian distribution. Pearson correlation analysis and Bland-Altman scatter diagram suggested that the three e GFRs were correlated with r GFR significantly with the correlation coefficients of 0. 832,0. 838 and 0. 917 respectively. The percentage of points of ΔSCr,ΔCys Cand ΔSCr / Cys Clocated in each 95% conformity boundary was 93. 59%,96. 15% and 94. 87% respectively. Student's t test and chi-square test suggested that three 95% confidence intervals of the errors' expectations were(-5. 62,1. 22),( 3,18,10. 50) and(-6. 42,-1. 59),and three 95% confidence intervals of the errors' variances were( 171. 66,324. 18),( 196. 90,371. 85) and( 85. 55,161. 57). Further analysis by the 3σ principle showed that ΔSCr∈(-59. 63,55. 23),ΔCys C∈(-54. 67,68. 36) and ΔSCr / Cys C∈(-44. 55,36. 54) when each expectation and variance were critical values. Conclusion: Error range of CKD-EPISCr / Cys C equation is minimum among the three CKD-EPI equations. CKD-EPISCr / Cys Cequation tends to be better than the CKD-EPICys C equation and the CKD-EPISCrequation in calculating GFR in the clinical practice.

关 键 词:肾小球滤过率 CKD-EPI方程 误差估计 

分 类 号:R692[医药卫生—泌尿科学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象