Increased Risk of Supratherapeutic Vancomycin Exposure Among Lvad Recipients

Abstract

Introduction Hemodynamic derangements due to heart failure are associated with alterations in pharmacokinetics. Although use of mechanical circulatory support mitigates such derangements, little evidence is available regarding pharmacokinetics in patients with LVADs. This is particularly concerning for medications with narrow therapeutic indices for which changes in drug disposition could lead to increased toxicity or a lack of efficacy. A previous pharmacokinetic analysis of vancomycin in LVAD patients observed a reduced volume of distribution and clearance compared with estimates based on population kinetics. Hypothesis The use of a hospital-wide vancomycin dosing nomogram would result in supratherapeutic vancomycin levels in patients with LVADs. Methods A retrospective cohort study was conducted in patients with LVADs who received vancomycin along with a pharmacist dosing consult. Internal medicine patients without heart failure were the control group (1:2) since the dosing nomogram was validated in this population. Patients aged 18-75 years were included if they received nomogram-dosed vancomycin with an appropriately drawn steady-state trough. Patients with unstable renal function, ESRD, acute decompensation, cardiac surgery within the preceding 5 days, or weight >110 kg were excluded. The primary outcome was attainment of goal trough (10-25 mg/L). Secondary outcomes included mean trough level and change in renal function. Results A total of 28 LVAD patients (53% Heartmate II, 43% HVAD, 4% Heartmate 3) were included. They were similar in age, BMI, CrCl, and number of concomitant nephrotoxic medications when compared to the control group (n=56). However, they were more often men (82% vs. 50%), had a greater mean weight (86 kg vs. 77 kg) and baseline SCr (1.12 mg/dL vs 0.91 mg/dL), and were more likely to receive loop diuretics (54% vs. 18%). The most common indication was skin and soft tissue infections in both groups. No difference was observed in goal trough achievement (64% of LVAD patients vs. 71% control patients, p=0.50). However, mean trough was higher in LVAD patients (23.4 mg/L vs. 17.7 mg/L, p=0.017). Furthermore, the distribution of trough levels differed (p=0.025), with troughs in LVAD patients more likely to be >25 mg/L (32% vs. 14%) and less likely to be Conclusions The use of vancomycin in LVAD patients may result in higher trough levels when compared to a control group of medicine patients receiving equivalent doses. Increased monitoring frequency or conservative dosing may result in fewer supratherapeutic levels in these patients, which has implications in preventing nephrotoxicity in this high-risk population.