Abstract

The aim of this study was to develop a model for the relationship between concentration and anticonvulsant effect of phenobarbital in the rat. The method of quantitating the anticonvulsant response on the basis of plasma pentylenetetrazol threshold concentrations, was optimized by application of a longitudinal study design. The pharmacokinetics of phenobarbital were studied in a separate set of rats (elimination half-life 11 +/- 2 hr). The anticonvulsant effect was measured 1 hr after i.v. administration. Concentration-effect relationships were assessed for serum (total and free), cerebrospinal fluid and brain concentrations on the basis of the sigmoid maximum effect model. The EC50 for total serum was 76 +/- 9 mg/l and for free serum, cerebrospinal fluid and brain 44 +/- 5, 43 +/- 7 and 50 +/- 8 mg/l and mg/kg, respectively (means +/- S.E.). The maximum effect was about 120 mg/l of pentylenetrazol. The respective EC50 ratios corresponded closely with the concentration ratios between the different compartments as determined by orthogonal regression analysis. For the greater part of the concentration range the concentration-effect relationship also could be fitted accurately to a simple linear model. Such relationships were not perturbed by the occurrence of seizures as these did not alter the distribution profile of phenobarbital. The pharmacokinetics and pharmacodynamics of p-hydroxyphenobarbital, the main metabolite of phenobarbital, also were investigated. This compound is eliminated very rapidly and has neither anticonvulsant activity of its own nor influences the anticonvulsant effect of phenobarbital at serum concentrations up to 30 mg/l. The described experimental strategies allow the study of factors influencing the concentration-anticonvulsant effect relationship of phenobarbital.