Background: In daily practice serum creatinine measurement still often used as an idex of renal function. The sensitivity of this test however is low. A number of prediction equations have been developed to estimate the GFR using serum creatinine and some other variables. Of all those equations the Cockroft-Gault forumla and the MDRD formula have been studied most.
Questions: Which formula estimates GFR most accurately? Which formula can be implemented easily in daily general practice?
Methods: We retrospectively collected data from the Intego database of all patients witha known weight, serum creatinine, age and gender. Only patients older than 20 years were enrolled. The GFR was estimated by using the CG and the MDRD equation. We analysed the prevalence of a decreased kidney function in function of age and sex using the five stages of the American Natinal Kidney Founation guidelines. We foused on differences in estimated GFR between the two formulas.
Results: Required data were known fromm 9969 persons, of whitch 5442 patiens were woman and 4527 were men. The average GFR was 90 ml/min (Sd 33,55) estimated by the CG formula and 80ml/min/1,73m2 (SD 20,32) estimated by the MDRD formula. The two formulas classified 88% of all patients within the same stage. Differences in classification were most pronounced for patients being 80 years and older (only 54% of those patients were classifed within the same stage).
Conclusion: The MDRD formula in general is a more accurate estimation in comparison with the CG formula. The MDRD formula only requires serum creatinine, age and sex, without the need to know patient's weight. We propopose all laboratories to mention the estimated FGR next to te serum creatinine value.