Estimation of Renal Function in Subjects With Normal Serum Creatinine Levels: Influence of Age and Body Mass Index Jacobien C. Verhave, MD, Pierre Fesler, MD, Jean Ribstein, MD, Guilhem du Cailar, MD, and Albert Mimran, MD Background: The Cockcroft-Gault (CG) and simplified Modification of Diet in Renal Disease (MDRD) formulas are the most widely used estimates of renal function. The influence of age and body mass index (BMI) on the performance of these equations was analyzed in 850 subjects with serum creatinine levels less than 1.5 mg/dL (<133 mol/L). Methods: Glomerular filtration rate (GFR) was measured as urinary clearance of continuously infused technetium Tc 99m–labeled diethylene triaminopentaacetic acid. Performance was assessed as bias, precision, and accuracy. Results: In the total population, the CG and MDRD calculations based on enzymatic measurement of serum creatinine (which is constantly less than that obtained by using the alkaline picrate [Jaffé] method) significantly underestimated GFR by 4.9 and 12.4 mL/min/1.73 m 2 (0.08 and 0.21 mL/s/1.73 m 2 ), respec- tively. In patients 65 years and older, underestimation by means of the CG formula was enhanced, whereas that by means of the MDRD formula was blunted, compared with the group younger than 65 years (11.3 versus 3.7 mL/min/1.73 m 2 [0.19 versus 0.06 mL/s/1.73 m 2 ] for CG and 3.7 versus 14.0 mL/min/1.73 m 2 [0.06 versus 0.23 mL/s/1.73 m 2 ] for MDRD). GFR was underestimated to a large extent by means of the MDRD equation irrespective of BMI. Conversely, the underestimation by means of the CG formula found in lean people (13.0 mL/min/1.73 m 2 [0.22 mL/s/1.73 m 2 ]) was blunted in overweight people (BMI, 25 to 30 kg/m 2 ) and reversed to overestimation (10.1 mL/min/1.73 m 2 [0.17 mL/s/1.73 m 2 ]) in obese subjects (BMI > 30 kg/m 2 ). Conclusion: As suggested by estimations obtained using enzymatic serum creatinine measurement, the MDRD equation may be the estimation of choice in elderly patients, whereas the CG estimate is preferable in subjects younger than 65 years. Nevertheless, when obesity is present, no reliable estimation can be obtained by using the CG or MDRD formula. Am J Kidney Dis 46:233-241. © 2005 by the National Kidney Foundation, Inc. INDEX WORDS: Glomerular filtration rate (GFR); renal function estimates; Cockcroft-Gault; Modification of Diet in Renal Disease; aging; obesity. T HE NUMBER OF patients with end-stage renal disease is increasing worldwide. 1,2 Defined by using the sex-specific 95th percen- tiles of serum creatinine of a reference popula- tion (cutoff values, 1.4 mg/dL [124 mol/L] in women, 1.5 mg/dL [133 mol/L] in men), the prevalence of impaired renal function was 8% to 9% in the Framingham Heart Study. 3 In the Third National Health and Nutrition Examination Sur- vey cohort, a prevalence of impaired renal func- tion of 11% to 13% (defined as a glomerular filtration rate [GFR] estimate 60 mL/min/1.73 m 2 [1.00 mL/s/1.73 m 2 ] by means of the Modi- fication of Diet in Renal Disease [MDRD] for- mula) was observed. 4,5 Of note, recent studies have shown that even modest impairment in renal function may be an independent predictor of cardiovascular morbidity and mortality. 6-8 Urinary clearance of intravenously infused inulin or labeled tracers with timed urine collec- tions is considered the gold standard for determi- nation of GFR. When rapid assessment of renal function is required (ie, for dosing of drugs with renal excretion in elderly patients) and in large clinical trials, several estimates of renal function using formulas based on serum creatinine level were proposed, including the Cockcroft-Gault (CG) and MDRD equations. The CG formula was designed in 1976 as an estimate of creatinine clearance calculated from 24-hour urine collec- tion and within a range of 11 to 186 mL/min (0.18 to 3.10 mL/s) in 236 hospitalized male patients aged 18 to 92 years. 9 The MDRD equa- tion was developed in 1,628 patients with chronic kidney disease with reference to GFR (renal clearance of subcutaneously-injected iodine 125– iothalamate) and within a range of 10 to 130 mL/min/1.73 m 2 (0.17 to 2.17 mL/s/1.73 m 2 ). 10 From the Department of Internal Medicine, CHU Montpel- lier, France. Received January 3, 2005; accepted in revised form May 2, 2005. Originally published online as doi:10.1053/j.ajkd.2005.05.011 on June 24, 2005. Address reprint requests to Albert Mimran, MD, Depart- ment of Internal Medicine, Hôpital Lapeyronie, 34295 Mont- pellier Cedex 5, France. E-mail: amimran@wanadoo.fr © 2005 by the National Kidney Foundation, Inc. 0272-6386/05/4602-0006$30.00/0 doi:10.1053/j.ajkd.2005.05.011 American Journal of Kidney Diseases, Vol 46, No 2 (August), 2005: pp 233-241 233