The 4 variable MDRD equation was developed based on a US sample of 1628 participants in which glomerular filtration rate was measured as the renal clearance of I-iothalamate and a prediction equation was formulated using serum creatinine, age, sex, and race. For black race, the answer is multiplied by 1.2.
My question is, how well does the MDRD equation predict GFR in different global populations?
Two studies performed in sub-Saharan Africa (Ghana N=944 and South Africa N=100) comparing the MDRD and Cockcroft-Gault (CG) estimates of glomerular filtration rate (GFR) against measured creatinine clearance (24 hour urine creatinine and clearance of chromium-51-EDTA) showed that the MDRD equation performed better without using the ethnicity factor of 1.2. In a small Saudi study (N=32), GFR estimated by MDRD revealed the strongest correlation with the measured inulin clearance (r= 0.976, P= 0.0000). The correlation between eGFR and Clearance of inulin in a Japanese population (N=248) was better with the 0.881xMDRD equation than with the 1.0xMDRD study equation.
Clearly the prediction of GFR by MDRD varies by global region and most likely this variability correlates with genetic/environmental factors contributing to body muscle mass.