Comparing Two Equations for Estimation of Kidney Function (Cockcroft-gault and Glomerular Filtration Rate Assessment in Liver Disease) for Vancomycin Dosing in Adult Liver Transplant Recipients: a Pilot, Randomized Clinical Trial

Abstract

Background: In the setting of impaired liver function, estimation of glomerular filtration rate (GFR) using common creatinine based equations is inaccurate. Recently, Glomerular Filtration Rate Assessment in Liver Disease (GRAIL) model has been introduced for estimating GFR in liver transplantation. This study was conducted to compare vancomycin dose adjustment in liver transplant patients using Cockcroft-Gault (C-G) versus GRAIL method.Methods: In this pilot, randomized clinical trial, adult liver transplant recipients who were candidate to receive intravenous vancomycin were enrolled. GFR and creatinine clearance were estimated using GRAIL model and C-G equation in the intervention and control arms, respectively and vancomycin maintenance doses were adjusted accordingly. At the steady state , peak and trough serum concentrations of vancomycin were collected for pharmacokinetic comparisons.Results: Fifteen patients were enrolled in each arm of the study. Mean daily dose of vancomycin was estimated insignificantly lower for individuals in GRAIL arm than C-G group (1500.00±544.45 versus 1750.00± 389.60mg). The rate of patients who achieved the target vancomycin trough concentration was similar between the two study arms (20%). Compared with C-G group, higher rate of patients in GRAIL arm experienced below-target vancomycin trough concentrations (40% versus 13.3%) and lower rate showed above target trough concentration (40% versus 66.7%), although these differences did not reach statistical significance. Higher rates of patients with at target and above target vancomycin AUC/MIC were seen in the C-G group compared with GRAIL group (40% versus 26.7% and 53.3% versus 26.7%, respectively), while individuals in the GRAIL arm represented significantly higher rate of below target vancomycin AUC/MIC than C-G arm (46.7% versus 6.7%) (P=0.049). No differences in clinical outcomes were observed between the two groups.Conclusion: Using GRAIL model for vancomycin dose selection may result in less percent of patients with at target AUC/MIC exposure compared to C-G method and expose more percent of patients at risk of vancomycin under dosing.