Cockcroft-Gault formula is biased by body weight in diabetic patients with renal impairment
Introduction
The evaluation of renal function is crucial in patients with chronic kidney disease [1]. Glomerular filtration rate (GFR) is traditionally considered the best overall index of renal function in health and disease [2]. According to recent guidelines [3], [4], the determination of GFR allows the diagnosis and the stratification of chronic renal failure (CRF), which has practical implications. Complications should be evaluated and treated in patients with moderate renal failure (GFR <60 mL/[min · 1.73 m2]), whereas patients with severe renal failure (GFR <30 mL/[min · 1.73 m2]) must be referred to a nephrologist with a view to dialysis. Delayed referral and treatment are associated with poor prognosis [5].
Serum creatinine concentration is widely used as an indirect marker of GFR, but it is influenced by muscle mass and diet [6]. It should not be used as the sole index of kidney function [3]. GFR can be directly measured by infusion of external substances such as inulin or 51Cr-EDTA [7], but these methods are expensive and time consuming. The use of prediction equations to estimate GFR from serum creatinine and other variables (age, sex, race, and body size) is therefore recommended by the National Kidney Foundation [3]. The proposed equations are the Cockcroft-Gault (CG) formula [8] and the modification of diet in renal disease (MDRD) study equation [2]. However, these equations have not been validated in conditions of extreme body size or severe malnutrition. Their predictive values have not been compared, and GFR may need to be measured directly in such cases [3].
Diabetes is the leading cause of end-stage renal disease in the Western world [9]. Mainly because of the high prevalence and better life expectancy of type 2 diabetic patients [10], the proportion of patients with both diabetes and end-stage renal disease is rising dramatically in developed countries [11]. The use of the CG formula to assess renal function is recommended by the American Diabetes Association [12], but this equation includes weight, which varies widely among diabetic patients; both obesity and malnutrition are more frequent in hemodialyzed diabetic patients [13]. The more recent MDRD equation does not include body weight, but it has not been validated in diabetic kidney disease [2]. Its superiority over the CG has been mentioned in some [14], but not all [15], recent reports. None of these studies included diabetic patients with kidney damage.
In 122 diabetic patients with kidney damage, we compared CG- and MDRD-estimated GFR to values measured by a reference method (51Cr-EDTA). We studied the correlation between both estimates and isotopic measurement of GFR, and we performed a Bland and Altman procedure [16]. To determine the influence of body weight on the estimations, we repeated the comparison and the correlation studies after the patients were categorized as normal (body mass index [BMI] <25 kg/m2), overweight (BMI ≥25 to <30 kg/m2), or obese (BMI ≥30 kg/m2), and we calculated the CG based on the ideal body weight of the subjects (as defined by body weight for BMI = 22 kg/m2). Comparisons and correlations were also studied separately in types 1 and 2 diabetic patients.
Section snippets
Patients
One hundred twenty-two adult diabetic patients attending our clinical unit were studied. The patients were included if they were renal insufficient, indicated by an isotopic GFR below 60 mL/(min · 1.73 m2) (n = 87) or if they had at least some kidney damage as defined by an isotopic GFR below 90 mL/(min · 1.73 m2) and microalbuminuria of more than 30 mg/24 hours (n = 35). Patients with nephrotic proteinuria (>3 g/24 hours) or anasarca were excluded. No patient was treated by dialysis at the
Results
Both sexes (67 men, 55 women) and types of diabetes (35 type 1, 87 type 2) were represented. Mean hemoglobin A1c was 8.4% ± 1.6%. A wide range of age (30-83 years; mean ± SD, 65.6 ± 11.3 years), BMI (15.6-48.9 kg/m2; mean ± SD, 27.5 ± 4.8 kg/m2), and serum creatinine levels (67-371 μmol/L; mean ± SD, 152 ± 71 μmol/L) were represented. Mean proteinuria was 626 ± 947 mg/24 hours.
Mean isotopic GFR was 44.6 ± 21.1 mL/(min · 1.73 m2). Mean CG overestimated GFR (51.4 ± 23.1 mL/[min · 1.73 m2], P <
Discussion
Our comparison between the CG and MDRD estimations of GFR in diabetic patients with renal impairment is clearly in favor of the MDRD formula. The mean MDRD estimates were close to isotopically measured GFR, with higher correlation coefficients, and no bias because of BMI. These results are in line with those of the original MDRD study [2] and the more recent report from Hallan et al [14]. These studies however included a marginal proportion of diabetic patients (6% in the MDRD, not reported in
Acknowledgment
We thank Dr S Jarman for revision of the English manuscript.
References (27)
- et al.
Comparison between inulin and 51Cr-labelled edetic acid for the measurement of glomerular filtration rate
Lancet
(1968) - et al.
End-stage renal failure in type 2 diabetes—a medical catastrophe of worldwide dimension
Am J Kidney Dis
(1999) - et al.
Malnutrition in hemodialysis diabetic patients: evaluation and prognostic influence
Kidney Int
(2002) - et al.
Validation of the modification of diet in renal disease formula for estimating GFR with special emphasis on calibration of the serum creatinine assay
Am J Kidney Dis
(2004) - et al.
Timing of nephrology referral: influence on mortality and morbidity
Am J Kidney Dis
(2000) - et al.
Predicting creatinine clearance and renal drug clearance in obese patients from estimated fat-free body mass
Am J Med
(1988) Obesity epidemic
Lancet
(2004)Comparative physiology of the kidney
- et al.
A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group
Ann Intern Med
(1999) - et al.
National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification and stratification
Ann Intern Med
(2003)