Comparison of creatinine clearance estimates with routine measured clearance in adult Jordanians

ABSTRACT

To determine the correlation between endogenous creatinine clearance, the Modification of Diet in Renal Disease and Cockcroft-Gault equations, and to examine agreement between these methods. Creatinine clearance measured by 24-hour urine creatinine in patients with chronic kidney disease was compared with estimated clearance using the Cockcroft-Gault and Modification of Diet in Renal Disease equations. Correlation, regression analysis, limits of agreement and concordance analysis were used. There was a positive correlation between both the Cockroft-Gault [r=0.931, p<0.001] and the Modification of Diet in Renal Disease [r=0.741, p<0.001] equations with creatinine clearance; the former being statistically superior [P=0.0073] comparing the correlation coefficients. Using Bland-Altman method, there was better agreement with the Cockcroft-Gault formula where the limits of agreement were [-10.1 to + 11.2] compared with [-22.8 to + 18.5] for the Modification of Diet in Renal Disease. Concordance analysis showed a concordance correlation coefficient of 0.938 [95% CI, 0.89-0.97] for Cockroft-Gault and 0.713 [95% CI, 0.55-0.82] for Modification of Diet in Renal Disease. Pearson p coefficient [precision] and bias correlation factor Cb [accuracy] of Cockcroft-Gault were 0.939, 0.998 and of Modification of Diet in Renal Disease they were 0.713 and 0.947, respectively. In Jordanian patients, the Cockcroft-Gault equation had a superior performance in the estimate of creatinine clearance compared to the Modification of Diet in Renal Disease equation. Further studies to evaluate these formulae with larger number of patients and using a gold standard method are needed. In the meantime, we suggest using the Cockcroft-Gault equation in patients with a moderate-to-severe chronic renal failure