Evaluation of population pharmacokinetics in therapeutic trials. IV. Application to postmarketing surveillance.

The feasibility of incorporating blood sampling for population pharmacokinetic analysis into postmarketing surveillance was evaluated. Demographic and drug disposition data, consisting of two blood samples collected at a random time during two different dose intervals, was prospectively collected for 94 psychiatric inpatients (mean age, 48 +/- 13 years) receiving alprazolam. Mixed-effect modeling was used to estimate population pharmacokinetic parameters. The mean alprazolam clearance, volume of distribution, and absorption rate constant were 0.05 L/hr/kg, 0.7 L/kg, and 1.1 hr-1, respectively. Clearance was increased by 59% in women, decreased by 26% in patients with multiple organ disease, and decreased by 23% in patients older than 60 years of age. These estimates are similar to those determined from rigorous premarketing clinical trials. Interindividual variability in alprazolam clearance was relatively small (40%) after adjustment for significant patient covariates. Population pharmacokinetic analysis represents a reasonable approach to assessment of pharmacokinetic variability in a large, heterogenous patient population.