A pharmacokinetic model for amiodarone in infants developed from an opportunistic sampling trial and published literature data

Abstract

Amiodarone is a first-line antiarrhythmic for life-threatening ventricular fibrillation or ventricular tachycardia in children, yet little is known about its pharmacokinetics (PK) in this population. We developed a population PK (PopPK) model using samples collected via an opportunistic study design of children receiving amiodarone per standard of care supplemented by amiodarone PK data from the literature. Both study data and literature data were predominantly from infants < 2 years old, so our analysis was restricted to this group. The final combined dataset consisted of 266 plasma drug concentrations in 45 subjects with a median (interquartile range) postnatal age of 40.1 (11.0-120.4) days and weight of 3.9 (3.1-5.1) kg. Since the median sampling time after the first dose was short (study: 95 h; literature: 72 h) relative to the terminal half-life estimated in adult PopPK studies, values of the deep compartment volume and flow were fixed to literature values. A 3-compartment model best described the data and was validated by visual predictive checks and non-parametric bootstrap analysis. The final model included body weight as a covariate on all volumes and on both inter-compartmental and elimination clearances. The empiric Bayesian estimates for clearance (CL), volume of distribution at steady state, and terminal half-life were 0.25 (90% CL 0.14-0.36) L/kg/h, 93 (68-174) L/kg, and 266 (197-477) h, respectively. These studies will provide useful information for future PopPK studies of amiodarone in infants and children that could improve dosage regimens.

Keywords: Amiodarone; Cardiac arrhythmias; Pediatrics; Population pharmacokinetics.

Figures

Fig. 1

Standard goodness-of-fit plots of the final model. Observed versus population (A) and individual (B) predictions for the amiodarone model (note log-log scale). The line of identity is indicated by a solid line, the smooth fit to the data by the dashed line. Conditional weighted residuals (CWRES) versus time after dose (C) and population predictions (D) for the base amiodarone PK model.

Fig. 2

Visual predictive check of the amiodarone population PK model displays observations and prediction intervals for the PTN (A) and Ramusovic (B) datasets. Solid lines and surrounding shaded areas represent, from top to bottom, 5th, 50th and 95th percentile of model predictions, each with their respective +/− 1 SD. Top and bottom dashed lines are the 95th and 5th percentiles for the observed data; middle dashed line is the median. Circles represent the observed data collected across all subjects.

Fig. 3

Observed and predicted amiodarone concentrations determined in 13 patients over 2 years of age

Fig. 4

Mean values (+/− SD) from simulation to illustrate impact of third, deep compartment on expected amiodarone concentration accumulation and decay with chronic dosing (open circles = 3-compartment model with large, deep compartment – final model; filled circles = 2-compartment model)