Gotoh Yoshimitsu, Uemura Osamu, Ishikura Kenji, Sakai Tomoyuki, Hamasaki Yuko, Araki Yoshinori, Hamda Riku, Honda Masataka
Department of Pediatric Nephrology, Japanese Red Cross Nagoya Daini Hospital, 2-9 Myoken-cho Showa-ku, Nagoya, Aichi, 466-8650, Japan.
Department of Clinical Medicine, Japanese Red Cross Toyota College of Nursing, 12-33 Nanamagari, Hakusan-cho, Toyota, Aichi, 471-8565, Japan.
Clin Exp Nephrol. 2018 Aug;22(4):931-937. doi: 10.1007/s10157-018-1529-7. Epub 2018 Jan 25.
The gold standard for evaluation of kidney function is renal inulin clearance (Cin). However, the methodology for Cin is complicated and difficult, especially for younger children and/or patients with bladder dysfunction. Therefore, we developed a simple and easier method for obtaining the estimated glomerular filtration rate (eGFR) using equations and values for several biomarkers, i.e., serum creatinine (Cr), serum cystatin C (cystC), serum beta-2 microglobulin (βMG), and creatinine clearance (Ccr). The purpose of the present study was to validate these equations with a new data set.
To validate each equation, we used data of 140 patients with CKD with clinical need for Cin, using the measured GFR (mGFR). We compared the results for each eGFR equation with the mGFR using mean error (ME), root mean square error (RMSE), P, and Bland-Altman analysis.
The ME of Cr, cystC, βMG, and Ccr based on eGFR was 15.8 ± 13.0, 17.2 ± 16.5, 15.4 ± 14.3, and 10.6 ± 13.0 ml/min/1.73 m, respectively. The RMSE was 29.5, 23.8, 20.9, and 16.7, respectively. The P was 79.4, 71.1, 69.5, and 92.9%, respectively. The Bland-Altman bias analysis showed values of 4.0 ± 18.6, 5.3 ± 16.8, 12.7 ± 17.0, and 2.5 ± 17.2 ml/min/1.73 m, respectively, for these parameters.
The bias of each eGFR equation was not large. Therefore, each eGFR equation could be used.
评估肾功能的金标准是肾菊粉清除率（Cin）。然而，Cin的检测方法复杂且困难，尤其是对于年幼儿童和/或膀胱功能障碍患者。因此，我们开发了一种简单易行的方法，利用几种生物标志物（即血清肌酐（Cr）、血清胱抑素C（cystC）、血清β2微球蛋白（βMG）和肌酐清除率（Ccr））的方程和数值来获取估计肾小球滤过率（eGFR）。本研究的目的是用新的数据集验证这些方程。
为了验证每个方程，我们使用了140例有临床Cin检测需求的慢性肾脏病（CKD）患者的数据，并采用实测肾小球滤过率（mGFR）。我们使用平均误差（ME）、均方根误差（RMSE）、P值和布兰德-奥特曼分析，将每个eGFR方程的结果与mGFR进行比较。
基于eGFR的Cr、cystC、βMG和Ccr的ME分别为15.8±13.0、17.2±16.5、15.4±14.3和10.6±13.0ml/min/1.73m²。RMSE分别为29.5、23.8、20.9和16.7。P值分别为79.4%、71.1%、69.5%和92.9%。布兰德-奥特曼偏差分析显示，这些参数的值分别为4.0±18.6、5.3±16.8、12.7±17.0和2.5±17.2ml/min/1.73m²。
每个eGFR方程的偏差不大。因此，每个eGFR方程都可以使用。
