Raju Dharmapaul L, Grover Vaneeta K, Shoker Ahmed
Division of Internal Medicine, University of Sasatchewan, Saskatoon, SK, Canada.
Clin Transplant. 2005 Apr;19(2):259-68. doi: 10.1111/j.1399-0012.2005.00335.x.
This study aims to compare the performance of endogenous creatinine clearance (CL(cr)) and a number of published mathematical equations to calculate glomerular filtration rate (GFR) in renal transplant patients considering (99m)Tc DTPA isotope scan as the reference method. A total of 152 GFR were performed on 81 renal transplant patients. Accuracy of each method was measured at different percentiles. The bias and precision of all the methods were then compared. A paired t-test was used to compare the performance of each calculation to the respective GFR measured by isotope study performed on the same day. In the total population, all calculated methods correlated significantly with the isotope results. Accuracies within specific ranges of the isotope GFR were limited in all equations (agreement with isotope result </=72% at 30% accuracy range in the total group). Within the limited accuracy, Edwards' equation (K.D. Edwards and H.M. Whyte, Australas Ann Med 1959; vol. 8: p. 218) had the least bias in the total population. Bjornsson (T.D. Bjornsson, Clin Pharmacokinet 1979; vol. 4: p. 200) had the least bias in patients with GFR >/= 50 mL/min and Gates in patients with GFR < 50 mL/min. Salazar (D.E. Salazar and G.B. Corcoran, Am J Med 1988; vol. 84: p. 1053) had the least bias in patients with BMI above 30 kg/m(2) and the Davis equation (G.A. Davis and M.H. Chandler, Am J Health Syst Pharm 1996; vol. 53: p. 1028) in patients with BMI <25 kg/m(2). In all analyses, Nankivell (B.J. Nankivell, S.M. Gruenwald, R.D.M. Allen and J.R. Chapman, Transplantation 1995; vol. 59: p. 1683) overestimated GFR by more than 80% and MDRD 1 and 2 in <10% of the time. The results demonstrate the inherited limitation in the currently available equations to calculate GFR in renal transplant patients.
本研究旨在比较肾移植患者中内源性肌酐清除率（CL(cr)）和一些已发表的数学公式计算肾小球滤过率（GFR）的性能，将（99m）Tc DTPA同位素扫描作为参考方法。对81例肾移植患者共进行了152次GFR检测。在不同百分位数下测量每种方法的准确性。然后比较所有方法的偏差和精密度。采用配对t检验比较每种计算方法与同一天进行的同位素研究测量的各自GFR的性能。在总体人群中，所有计算方法与同位素结果均显著相关。所有公式在同位素GFR特定范围内的准确性有限（在总体组30%的准确性范围内，与同位素结果的一致性≤72%）。在有限的准确性范围内，爱德华兹公式（K.D.爱德华兹和H.M.怀特，《澳大利亚医学杂志》1959年；第8卷：第218页）在总体人群中的偏差最小。比约恩松公式（T.D.比约恩松，《临床药代动力学》1979年；第4卷：第200页）在GFR≥50 mL/min的患者中偏差最小，盖茨公式在GFR<50 mL/min的患者中偏差最小。萨拉查公式（D.E.萨拉查和G.B.科科伦，《美国医学杂志》1988年；第84卷：第1053页）在BMI高于30 kg/m²的患者中偏差最小，戴维斯公式（G.A.戴维斯和M.H.钱德勒，《美国卫生系统药学杂志》1996年；第53卷：第1028页）在BMI<25 kg/m²的患者中偏差最小。在所有分析中，南基韦尔公式（B.J.南基韦尔、S.M.格伦瓦尔德、R.D.M.艾伦和J.R.查普曼，《移植》1995年；第59卷：第1683页）高估GFR超过80%，MDRD 1和2在不到10%的时间内出现这种情况。结果表明，目前可用的计算肾移植患者GFR的公式存在固有局限性。
