Comparison of creatinine clearance estimates in subgroups based on Body Mass Index and albumin

The establishment of accurate equations for glomerular filtration rate (GFR) estimations is still far from the realization. Factors such as age, diabetes, stage of CKD, pregnancy, muscle mass and ethic nation are associated with less reliance upon commonly utilized estimation equations. We aimed to compare the routine use of 24-hour creatinine clearance (CrCl) and GFR estimates calculated by Crockoft-Gault (CG) and modification of diet in renal disease (MDRD) formulas in patients with different levels of renal dysfunction in subgroups, based on Body Mass Index (BMI) and serum albumin (Alb) levels.

Two hundred and seventy-nine non diabetic patients (172 men and 107 women), aged 54±23 years, with BMI 27.3±4.4 were enrolled in the study. All patients presented creatinine 1.8±1.2 (mg/dL) and Alb 3.5±1.3g/dL. The comparison of CrCl versus CG had bias 3.1 while the comparison of CrCl versus MDRD had a bias of 6.6.

Univariate analysis showed that age, sex and BMI were not significant biases related to the CG, MDRD and CrCl. Indeed, the bias related to the CG was significantly lower than that related to MDRD in patients with either low or high serum albumin. Interestingly, the bias associated with CG was 1.3 in non-diabetic patients with Alb ≤3.5 mg /dL suggesting that CG equation could be used interchangeable to CrCl in these patients.

CG gave a better prediction of measured CrCl than MDRD in Mediterranean, non-diabetic, non-hospitalized patients although misclassification of patients with regard to renal impairment stage was not present.