Abstract
AIMS: The main purpose of this paper is to describe the relationship between serum concentrations of glibenclamide and its main metabolites and the effects on blood glucose levels, the clinically most relevant parameter to assess in diabetes. METHODS: Serum concentrations and blood glucose lowering effects (expressed as percent blood glucose reduction vs placebo) of glibenclamide (Gb) and its active metabolites, 4-trans-hydroxy-(M1) and 3-cis-hydroxy-glibenclamide (M2), were analysed in eight healthy subjects participating in a placebo-controlled, randomized, single-blind crossover study, using intravenous administration of each compound as well as oral administration of Gb. RESULTS: Plots of % blood glucose reduction vs log serum concentration demonstrated counter-clockwise hysteresis for parent drug and its metabolites. An effect compartment was linked to appropriate pharmacokinetic models and pharmacokinetic and pharmacodynamic modelling was used to fit the pharmacokinetics of Gb by both routes and the metabolites for each individual. Based on the individual concentration-time profiles a PK/PD-model was applied to all effect data simultaneously. An increase in the steady-state serum concentration when the effect is 50% of maximal, CEss50, was found in the sequence M1 (23 ng ml-1), M2 (37 ng ml-1) and Gb (108 ng ml-1). Corresponding interindividual variabilities expressed as CV% were 25%, 47% and 26%. The elimination rate constants from the effect site (kEO) were estimated and increased in the order M1 (0.178 h-1, CV 13%), M2 (0.479 h-1, CV 8.5%) and Gb (1.59 h-1, CV 36%). Corresponding equilibration half-lives for the effect site (kEO-HL) were 3.9 h, 1.4 h and 0.44 h. Estimated Emax-values obtained for M1, M2 and Gb were 40% (CV 30%), 27% (CV 56%) and 56% (CV 14%), respectively. CONCLUSIONS: It is concluded that the two major metabolites of Gb are hypoglycaemic in man, that they may have higher activity at low concentrations and that they may have a longer effect duration than the parent drug.