Population pharmacokinetics of phenobarbitone administered as oral rescue therapy in children with refractory status epilepticus

Background: Parenteral phenobarbitone, used in refractory status epilepticus, was unavailable in South Africafrom 2005 to 2006. Some centres supplemented status epilepticus management with a bolus of nasogastric phenobarbitone for children approaching refractory status. We assessed the efficacy, safety and phenobarbitone pharmacokinetics of this practice.

Methods: Patients admitted in status epilepticus were enrolled unless impaired gastric absorption or systemically compromised. Phenobarbitone (20mg/kg) via nasogastric tube was given after benzodiazepine boluses and a phenytoin infusion. Drug concentrations were taken post loading with routine blood sampling. A population approach using a 1?compartment model with first order absorption was conducted using NONMEM VI with FOCE and La Placian transformation, a transit compartment absorption model, an exponential interindividual and a proportional residual variability model. The influence of patients’ characteristics and concomitant therapy were screened for significance (P = 0.01) in nested, multiplicative models. Model stability was assessed by the non-parametric bootstrap (N=200).

Results: The following model best described the log-transformed data:

 CL/F (mL·h-1 70 kg-1)  =  0.133 x (WT/70)0.75 x (1+βCL x EXP(-PNA * ln(2)/tCL)  x DPH

V/F (L·70 kg-1)  =  1.56 (WT/70)1 x (1+βvol x EXP(-PNA * ln(2)/tvol) x DPH

The model predicted CL/F to be slow and immature (βCL = -9.24, tCL = 441 months) in the absence of auto-induction of phenobarbitone clearance. V/F was predicted to reach maturity rapidly (βvol = -0.368,tvol = 0.3 hours). Prior phenytoin infusion (N = 11/16) increased phenobarbitone V/F (DPH = 1.22) and reduced CL/F (DPH = 0.72). Absorption was rapid (Ka = 1.33 h-1) and the population predicted phenobarbitone concentration-time plot showed that the minimum effective concentration (45 μmol/L) was reached within 1 hour after loading.

Conclusions:. Phenobarbitone concentrations were predicted to reach the minimum effective concentration (45 μmol/L) within 1 hour of dosing. Phenobarbitone CL/F was slow in the absence of auto-induction and repeated loading can result in toxicity. We recommend monitoring concentrations 1 hour after loading.