Published in:
01-05-2011 | Original Research Article
Association between Glomerular Filtration Rate and Adverse Drug Reactions in Elderly Hospitalized Patients
The Role of the Estimating Equation
Published in: Drugs & Aging | Issue 5/2011
Login to get accessAbstract
Background: Reduced renal function increases the risk of adverse drug reactions (ADRs) to hydrosoluble drugs (hADRs). However, the ability of different equations to calculate estimated glomerular filtration rate (eGFR) or estimated creatinine clearance (eCCr) and thereby predict the risk of developing hADRs has not previously been compared.
Objective: The aim of this study was to investigate which of three different equations for estimating renal function (Cockcroft-Gault [CG], Modification of Diet in Renal Disease [MDRD] and Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI]) was the most effective at predicting incident hADRs.
Methods: This multicentre study had an observational design and included 81 acute-care general (internal) or geriatric medicine wards in academic hospitals throughout Italy. Our series consisted of 10 442 hospitalized patients with a mean ± SD age of 70.2 ± 14.9 years enrolled in the GIFA study. The main outcome measures were incident ADRs during hospital stay. Data on these were collected and classified as hADRs or ADRs to liposoluble drugs (lADRs). Patients were grouped according to their eGFR (mL/min/1.73m2) or eCCr (mL/min): ≥90, 60–89.9, 45–59.9, 30–44.9 or <30.
Results: The multivariable adjusted risk of hADRs progressively increased with decreasing eGFR as determined by estimates of mL/min/1.73 m2 calculated using CKD-EPI (60–89.9: hazard ratio [HR] = 1.07 [95% CI 0.70, 1.72]; 45–59.9: HR = 1.62 [95% CI 1.0,2.69]; 30–44.9: HR = 2.13 [95% CI 1.24, 3.64]; <30: HR = 2.30 [95% CI 1.28, 4.14]) and, to a lesser extent, MDRD (60–89.9: HR = 1.15 [95% CI 0.75, 1.76]; 45–59.9: HR = 1.73 [95% CI 1.09, 2.73]; 30–44.9: HR = 2.14 [95% CI 1.30, 3.53]; <30: HR = 1.99 [95% CI 1.11, 3.57]) equations. The risk of hADRs also increased with lower eCCr, but only at CG eCCr <45mL/min (30–44.9: HR = 1.61 [95% CI 0.96, 2.77]; <30: HR = 1.76 [95% CI 1.0, 3.18]). Neither eGFR nor eCCr were associated with lADRs.
Conclusions: CKD-EPI-based estimates of eGFR outperformed MDRD-based estimates of eGFR and CG-based estimates of eCCr as a predictor of hADRs.