A Population Pharmacokinetic Model of Pentobarbital for Children with Status Epilepticus and Severe Traumatic Brain Injury

• Published in: Clinical pharmacokinetics Vol. 62; no. 7; pp. 1011 - 1022
• Main Authors: Ketharanathan, Naomi, Lili, Anastasia, de Vries, Julia M. Penning, Wildschut, Enno D., de Hoog, Matthijs, Koch, Birgit C. P., de Winter, Brenda C. M.
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.07.2023
• Subjects: Anti-Bacterial Agents - pharmacokinetics; Brain Injuries, Traumatic - drug therapy; Child; Creatinine; Critical Illness; Humans; Infant; Internal Medicine; Medicine; Medicine & Public Health; Original; Original Research Article; Pentobarbital; Pharmacology/Toxicology; Pharmacotherapy; Prospective Studies; Retrospective Studies; Status Epilepticus - drug therapy
• ISSN: 0312-5963; 1179-1926
• DOI: 10.1007/s40262-023-01249-z

Background Pentobarbital pharmacokinetics (PK) remain elusive and the therapeutic windows narrow. Administration is frequent in critically ill children with refractory status epilepticus (SE) and severe traumatic brain injury (sTBI). Objectives To investigate pentobarbital PK in SE and sTBI patients admitted to the paediatric intensive care unit (PICU) with population-based PK (PopPK) modelling and dosing simulations. Methods Develop a PopPK model with non-linear mixed-effects modelling (NONMEM ® ) with retrospective data ( n = 36; median age 1.3 years; median weight 10 kg; 178 blood samples) treated with continuous intravenous pentobarbital. An independent dataset was used for external validation ( n = 9). Dosing simulations with the validated model evaluated dosing regimens. Results A one-compartment PK model with allometrically scaled weight on clearance (CL; 0.75) and volume of distribution ( V d ; 1) captured data well. Typical CL and V d values were 3.59 L/70 kg/h and 142 L/70 kg, respectively. Elevated creatinine and C-reactive protein (CRP) levels significantly correlated to decreased CL, explaining 84% of inter-patient variability, and were incorporated in the final model. External validation using stratified visual predictive checks showed good results. Simulations demonstrated patients with elevated serum creatinine and CRP failed to achieve steady state yet progressed to toxic levels with current dosing regimens. Conclusions The one-compartment PK model of intravenous pentobarbital described data well whereby serum creatinine and CRP significantly correlated with pentobarbital CL. Dosing simulations formulated adjusted dosing advice in patients with elevated creatinine and/or CRP. Prospective PK studies with pharmacodynamic endpoints, are imperative to optimise pentobarbital dosing in terms of safety and clinical efficacy in critically ill children.