Discrepancies between the Cockcroft–Gault and Chronic Kidney Disease Epidemiology (CKD-EPI) Equations: Implications for Refining Drug Dosage Adjustment Strategies

Abstract

The dosages of many medications require adjustment for renal function. There is debate regarding which equation, the Chronic Kidney Disease Epidemiology (CKD-EPI) equation vs. the Cockcroft-Gault (CG) equation, should be recommended to estimate glomerular filtration rate. We used a mathematical simulation to determine how patient characteristics influence discrepancies between equations and analyzed clinical data to demonstrate the frequency of such discrepancies in clinical practice. In the simulation, the modifiable variables were sex, age, serum creatinine, and weight. We considered estimated glomerular filtration rate results in mL/min, deindexed for body surface area, because absolute excretory function (rather than per 1.73m2 body surface area) determines the rate of filtration of a drug at a given plasma concentration. An absolute and relative difference of maximum (±) 10mL/min and 10%, respectively, were considered concordant. Clinical data for patients aged over 60years (n=9091) were available from one hospital and 25 private laboratories. In the simulation, differences between the two equations were found to be influenced by each variable but age and weight had the biggest effect. Clinical sample data demonstrated concordance between CKD-EPI and CG results in 4080 patients (45%). The majority of discordant results reflected a CG result lower than the CKD-EPI equation. With aging, the CG result became progressively lower than the CKD-EPI result. When weight increased, the opposite occurred. The choice of equation for excretory function adjustment of drug dosage will have different implications for patients of different ages and body habitus. The optimum equation for drug dosage adjustment should be defined with consideration of individual patient characteristics.