Abstract

The objective of this investigation was to characterize the pharmacokinetic/pharmacodynamic correlation of buprenorphine and fentanyl for the antinociceptive effect in rats. Data on the time course of the antinociceptive effect following intravenous administration of buprenorphine or fentanyl was analyzed in conjunction with plasma concentrations by nonlinear mixed-effects analysis. For fentanyl, the pharmacokinetics was described on the basis of a two-compartment pharmacokinetic model. For buprenorphine, a three-compartment pharmacokinetic model best described the concentration time course. To explain time dependencies in pharmacodynamics of buprenorphine and fentanyl, a combined effect compartment/receptor binding model was applied. A log logistic probability distribution model is proposed to account for censored tail-flick latencies. The model converged, yielding precise estimates of the parameters characterizing hysteresis. The results show that onset and offset of the antinociceptive effect of both buprenorphine and fentanyl is mainly determined by biophase distribution. The k_eo was 0.024 min^–1 [95% confidence interval (CI): 0.018–0.030 min^–1] and 0.123 min^–1 (95% CI: 0.095–0.151 min^–1) for buprenorphine and fentanyl, respectively. On the other hand, part of the hysteresis in the buprenorphine pharmacodynamics could be explained by slow receptor association/dissociation kinetics. The k_off was 0.073 min^–1 (95% CI: 0.042–0.104 min^–1) and k_on was 0.023 ml/ng/min (95% CI: 0.013–0.033 ml/ng/min). Fentanyl binds instantaneously to the OP3 receptor because no reasonable values for k_on and k_off were obtained with the dynamical receptor model. In contrast to earlier reports in the literature, the findings of this study show that the rate-limiting step in the onset and offset of buprenorphine's antinociceptive effect is distribution to the brain.