Prediction of Glomerular Filtration Rate From Body Cell Mass and Plasma Creatinine

Abstract

The gold standards for the measurement of glomerular filtration rate (GFR) are inulin clearance and radioisotopic methods. However, creatinine clearance is the most used test to evaluate GFR in clinical practice. Its adequacy is questionable, since its repeatability is quite poor, mainly due to errors in the collection of urine. The aim of this study was to evaluate a new method to predict GFR from the body cell mass (BCM) and plasma creatinine (Pcr), avoiding urine collection. The values of BCM were obtained in 275 adult renal patients with different renal function, ranging from normality to advanced renal failure. The relationship of GFR (clearance of (99m)Tc-DTPA) with BCM and Pcr was calculated in the first 85 patients. A highly significant linear correlation was found between GFR and the ratio BCM/Pcr. Thereafter, GFR was predicted from BCM and Pcr (BCM GFR) with formulas derived from the relationships found between GFR and the ratio BCM/Pcr. For comparison, GFR was predicted also according to other prediction formulas: Cockcroft and Gault (CG GFR), and the simplified MDRD formula (MDRD GFR). BCM GFR gave a more precise estimate of GFR than CG GFR and MDRD GFR. In fact, BCM GFR had the best correlation and agreement with true GFR ((99m)Tc-DTPA). Furthermore, CG GFR and MDRD GFR markedly overestimated true GFR. Finally, the error of prediction of BCM GFR was definitely lower than that of the two other estimates of GFR. GFR can be predicted from BCM and plasma creatinine. This method, which is very simple and accurate, seems suitable to establish the adequate dosage of drugs cleared by the kidneys.