Estimation of Urinary Creatinine Excretion and Prediction of Renal Function in Morbidly Obese Patients: New Tools from Body Composition Analysis

Background/Aims: In obese subjects the accuracy of prediction of renal function is quite low. The aim of this study was to obtain a more accurate estimate of urinary creatinine excretion (UCr), creatinine clearance (CCr), and GFR from body cell mass (BCM). Methods: Seventy-three adult morbidly obese patients (BMI 35.2-64.5 kg/m2) were examined. BCM was calculated from body impedance analysis. CCr was measured (mCCr) and was predicted from BCM and antropometric data (MR-BCMCCr), with Cockcroft and Gault (C&GCCr) and Salazar and Corcoran (S&CCCr) formulas. GFR was predicted from BCM (BCM GFR) and with MDRD and CKD-EPI formulas. Results: Multiple regression (MR) indicated a strict linear correlation between UCr, BCM and anthropometric data. UCr predicted from MR equation (MR-BCMUCr) was very similar to measured UCr. MR-BCMCCr (168±46 mL/min) and mCCr (167±51 mL/min) were also similar, while significant differences were found between mCCr, C&GCCr and S&CCCr. The correlation and the agreement between MR-BCMCCr and mCCr were closer and prediction error was lower than the other formulas. BCM GFR (125±32 mL/min) had close correlations and agreements with MDRD GFR and CKD EPI formulas. Conclusions: In morbidly obese patients the measurement of BCM meliorates the prediction of UCr and CCr, and allows the prediction of GFR