Abstracts

Rationale: A whole blood assay is preferred for drug monitoring in children and infants as the volume that can be collected is often limited. For most drugs the drug concentration in red blood cells (RBC) is the same as in plasma. This is, however, not the case for topiramate (TPM) as it exhibits saturable binding to carbonic anhydrases in RBCs resulting in a nonlinear relationship at low, but clinically relevant, plasma concentrations. The primary objectives of this study were to compare the pharmacokinetics (PK) of IV TPM in plasma and whole blood and characterize the relationship between whole blood and plasma concentrations in dogs with naturally-occurring epilepsy. Canine epilepsy closely resembles human epilepsy and is being utilized by our research group to study approved and investigational antiepileptic drugs. Using these dogs, we're able to further explore the nonlinear relationship between plasma and whole blood TPM concentrations that have been reported in the literature. Methods: The experiment was carried out with four dogs with natural-occurring epilepsy, two of whom were on phenobarbital, a known enzyme inducer, housed at the University of Minnesota Veterinary Medical Center. Each dog was administered a single injection of 10 mg/kg IV TPM (10 mg/ml in 10% Captisol®) over 5 minutes. Blood samples were obtained pre-dose and at specified time points over 9 hrs post-injection. A 250 μL sample of whole blood was aliquoted, the remaining blood was centrifuged for plasma analysis. Plasma and whole blood TPM concentrations were measured using a validated LC/MS method. Whole blood TPM concentration-time profiles were analyzed using noncompartmental and compartmental methods. Binding of TPM to RBC carbonic anhydrases was explored using a saturable binding model. Results: Using noncompartmental analysis, we found in whole blood the range for TPM clearance was 0.04-0.63 L/hr/kg and half-life was 1-5 hrs. In plasma, clearance ranged from 0.11-0.9 L/hr/kg and half-life from 0.5-3.7 hrs. A two-compartment model best fit the whole blood TPM concentration data with distribution- and elimination-phase half-lives of 2-6 min and 1-12 hours, respectively. The relationship between whole blood and plasma was nonlinear with whole blood TPM concentrations 1.3 (at plasma concentrations >15 ug/mL) to 3.7 (at plasma concentrations <