Dosing of key renally cleared drugs in the elderly - Time to be wary of the eGFR

Background: The Cockroft-Gault and Modification of Diet in Renal Disease (MDRD) approaches are commonly used to estimate renal function. An optimised version of the Cockroft-Gault equation also exists. The MDRD approach occasionally generates renal function estimates that are vastly different from the Cockroft-Gault approach. Due to its ready availability, prescribers are beginning to use the MDRD approach for dosing of renally cleared drugs, despite contrary recommendations. Aim: To examine differences in renal function estimates between the MDRD against the Cockroft-Gault and optimised Cockroft-Gault approaches for a group of elderly hospitalised patients. To examine factors driving these differences and if these differences cause clinically significant differences in the dosing of enoxaparin and gentamicin. Method: Renal function for 1067 inpatients was calculated. Patients were grouped into quartiles based on the proportional difference between the MDRD against the Cockroft-Gault and optimised Cockroft-Gault approaches, respectively. Clinically significant dosing was defined as 2 mg/kg instead of 1 mg/kg for enoxaparin, and ≥ 75% difference for gentamicin dosing. Results: As the proportional difference in renal function estimates between the MDRD and the Cockroft-Gault approach increased, patients were progressively older and lighter (p < 0.01 ). As the difference increased compared to the optimised Cockroft-Gault version, patients were progressively older, shorter, lighter and female (p < 0.01). Enoxaparin and gentamicin doses from the MDRD approach led to significantly greater doses in 8% and 23% of patients respectively compared to Cockroft-Gault, and 15% and 37% of patients respectively compared to the optimised Cockroft-Gault approach. This occurred predominantly in very elderly patients (> 80 years). Conclusion: The use of MDRD to calculate drug doses in an elderly population leads to greatly increased doses being prescribed compared to the currently recommended approach of using Cockroft-Gault. This is largely in older patients, who are most at risk of experiencing adverse effects from increased doses. The appropriateness of MDRD for drug dosing in an elderly hospitalised population needs to be assessed urgently, against the possibility that it greatly overestimates renal function in these patients.