Performance of glomerular filtration rate estimation equations in patients with type 2 diabetes in yaounde

Dayawa Da Agoons
Internal Medicine and Specialties, University of Yaounde I
July, 2013


In clinical practice, glomerular filtration rate (GFR) is often estimated by equations such as the Modification of Diet in Renal Disease (MDRD), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) and Cockcroft-Gault (CG) equations. These equations however have not been validated in Sub-Saharan Africa. The aim of our study was to evaluate the performance of these equations to estimate GFR in patients with type 2 diabetes.

This cross-sectional study included 103 participants (51 patients with type 2 diabetes and 52 healthy non-diabetic controls) aged 55.4 ± 8.5 years; 64 (62%) were men. Creatinine clearance (CrCl) was measured in two consecutive 24-hour urine samples, and the result adjusted for body surface area using the Dubois formula. Venous samples were collected for the measurement of serum creatinine using the kinetic Jaffe method. Measured serum creatinine was calibrated to standardized levels using the following equation: Standardized Scr = 0.95 x Scr – 0.10. GFR was estimated according to the re-expressed MDRD, CKD-EPI and CG equations. Bland-Altman analysis was used to examine the agreement between measured and estimated GFR.

In diabetic individuals, CrCl was 72.7 ± 45.7 ml/min/1.73m2, MDRD GFR 72.5 ± 25.7 ml/min/1.73m2, CKD-EPI GFR 74.5 ± 27.4 ml/min/1.73m2 and CG GFR 67.3 ± 22.9 ml/min/1.73m2 (p = 0.298). In healthy volunteers, the corresponding values were 65.0 ± 42.2, 84.4 ± 14.6, 88.0 ± 15.1 and 78.3 ± 15.8 ml/min/1.73m2 (p < 0.0001). In diabetic patients, the MDRD (r = 0.58), CKD-EPI (r = 0.55) and CG (r = 0.61) correlated well with CrCl (all p < 0.05). In non-diabetic controls, the correlations ranged from 0.18 to 0.62. Bias was 0.3 for MDRD, -1.7 for CKD-EPI and 5.4 for CG in diabetic patients. All three equations overestimated GFR in control subjects. Precision and accuracy were sub-optimal in both groups. The sensitivity for the diagnosis of impaired renal function was 89.3%, 96.4% and 85.7% for MDRD, CKD-EPI and CG respectively.

The MDRD, CKD-EPI and CG equations performed poorly in normoglycemic individuals and type 2 diabetic patients. The MDRD equation however seems to perform better than the other two equations in patients with type 2 diabetes. This suggests that equation development or adjustment might be required in Sub-Saharan African diabetic populations.