Service de Nutrition-Diabétologie, Hôpital Haut-Lévêque, avenue de Magellan, 33600 Pessac, France.
Diabetes Metab. 2011 Nov;37(5):359-66. doi: 10.1016/j.diabet.2011.05.002. Epub 2011 Jun 15.
Diabetes is the leading cause of chronic kidney disease (CKD), which makes estimation of renal function crucial. Serum creatinine is not an ideal marker of glomerular filtration rate (GFR), which also depends on digestive absorption, and the production of creatinine in muscle and its tubular secretion. Formulas have been devised to estimate GFR from serum creatinine but, given the wide range of GFR, proteinuria, body mass index and specific influence of glycaemia on GFR, the uncertainty of these estimations is a particular concern for patients with diabetes. The most popular recommended formulas are the simple Cockcroft-Gault equation, which is inaccurate and biased, as it calculates clearance of creatinine in proportion to body weight, and the MDRD equation, which is more accurate, but systematically underestimates normal and high GFR, being established by a statistical analysis of results from renal-insufficient patients. This underestimation explains why the MDRD equation is repeatedly found to give a poor estimation of GFR in patients with recently diagnosed diabetes and is a poor tool for reflecting GFR decline when started from normal, as well as the source of unexpected results when applied to epidemiological studies with a 60mL/min/1.73m(2) threshold as the definition of CKD. The more recent creatinine-based formula, the Mayo Clinic Quadratic (MCQ) equation, and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) improve such underestimation, as both were derived from populations that included subjects with normal renal function. Determination of cystatin C is also promising, but needs standardisation.
糖尿病是慢性肾脏病（CKD）的主要病因，这使得肾功能评估至关重要。血清肌酐不是肾小球滤过率（GFR）的理想标志物，GFR 还取决于消化吸收以及肌肉中肌酐的产生和肾小管分泌。已经设计了一些公式来根据血清肌酐估算 GFR，但鉴于 GFR、蛋白尿、体重指数和血糖对 GFR 的具体影响范围广泛，这些估算值的不确定性是糖尿病患者特别关注的问题。最受欢迎的推荐公式是简单的 Cockcroft-Gault 方程，该方程不准确且存在偏差，因为它按体重比例计算肌酐清除率，而 MDRD 方程则更准确，但系统地低估了正常和高 GFR，该方程是通过对肾功能不全患者的结果进行统计分析建立的。这种低估解释了为什么 MDRD 方程在新近诊断为糖尿病的患者中多次被发现对 GFR 的估算不准确，以及为什么它不能很好地反映从正常开始的 GFR 下降，也解释了为什么当应用于以 60mL/min/1.73m²为 CKD 定义的阈值的流行病学研究时会出现意外结果。基于肌酐的更近期公式，即 Mayo 诊所二次方程（MCQ）和慢性肾脏病流行病学合作组（CKD-EPI）公式，改善了这种低估，因为这两个公式都是从包括肾功能正常的受试者的人群中推导出来的。半胱氨酸蛋白酶抑制剂 C 的测定也很有前途，但需要标准化。
