Estimated Glomerular Filtration Rate in Determining Aminoglycoside Dosing

Abstract

insufficiency Chronic kidney disease (CKD) is often associated with altered pharmacokinetics of many drugs. In hospitalised patients, noncompliance with dosing guidelines for CKD patients is common, ranging from 19% to 67% [1]. Thus, patients with CKD are at risk for dosing errors and drug toxicity. Aminoglycosides such as gentamicin, tobramycin and amikacin are bactericidal antibiotics indicated for severe gram negative infections. Aminoglycoside clearance is dependent on renal function and they are also nephrotoxic. Dose adjustments are required in patients with CKD. On the other hand, under-dosing is a valid concern as rapid achievement of target concentrations is crucial for antibacterial efficacy. It is clear that an accurate assessment of renal function is required to optimise the initial dose of aminoglycosides while minimising toxicity. The gold standard of assessing glomerular filtration rate (GFR) by radionuclide methods is impractical for guiding drug dosing. The Cockroft-Gault (CG) formula estimates creatinine clearance as a surrogate for GFR. Most guidelines and product information recommend the CG formula for estimating aminoglycoside dosing. The CG formula requires the patient weight, which may be adjusted for ideal body weight, lean body weight or body surface area to account for obese patients [2]. However, these modifications have been inconsistently applied. Recently, improved equations for estimated GFR (eGFR) have surfaced, with the addition of the Modification of Diet in Renal Disease (MDRD) and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formulas [3,4]. The MDRD formula is further modified as the isotope dilution mass spectroscopy (IDMS) traceable 4-variable equation following standardisation of creatinine assays [5]. These eGFR formulas were developed from population studies and are primarily used to detect and stage CKD. However, they represent an alternative to the CG formula for guiding drug dosing. Their main benefit is that eGFRs are readily available from automatic reporting with serum creatinines. However, their role in drug dosing is not clear. Some studies suggest that the CG, MDRD and CKD-EPI formulas are not interchangeable and may result in different renal function estimates and antimicrobial dosing. In a study of 180 patients (mean age 85 years), Gill et al. reported that the CG and 4-variable MDRD equations produced discordant eGFRs in over 60% of patients, with only one third of patients sharing the same CKD stage [6]. Use of MDRD over CG would have resulted in a 20% discordant dose recommendation for amantadine. The study by Golik et al. (mean age 64 years) also noted that the CG and 4-variable MDRD equations produced a discordance rate of 22-36% for dosing four non-aminoglycoside antibiotics [7]. Their study cohort was non-ICU patients with stable CKD (GFR <90 ml/min/1.73 m