BACKGROUND Morphine-6-glucuronide (M6G) is a metabolite of morphine and a micro-opioid agonist. To quantify the potency and speed of onset-offset of M6G and explore putative sex dependency, the authors studied the pharmacokinetics and pharmacodynamics of M6G in volunteers using a placebo-controlled, randomized, double-blind study design. METHODS Ten men and 10 women received 0.3 mg/kg intravenous M6G and placebo (two thirds of the dose as bolus, one third as a continuous infusion over 1 h) on separate occasions. For 7 h, pain tolerance was measured using gradually increasing transcutaneous electrical stimulation, and blood samples were obtained. A population pharmacokinetic (inhibitory sigmoid Emax)-pharmacodynamic analysis was used to analyze M6G-induced changes in tolerated stimulus intensity. The improvement in model fits by inclusion of covariate sex was tested for significance. P values less than 0.01 were considered significant. Taking into account previous morphine data, a predictive pharmacokinetic-pharmacodynamic model was constructed to determine the contribution of M6G to morphine analgesia. RESULTS M6G concentrations did not differ between men and women. M6G caused analgesia significantly greater than that observed with placebo (P < 0.01). The M6G analgesia data were well described by the pharmacokinetic-pharmacodynamic model. The M6G effect site concentration causing a 25% increase in current (C25) was 275 +/- 135 nm (population estimate +/- SE), the blood effect site equilibration half-life was 6.2 +/- 3.3 h, and the steepness parameter was 0.71 +/- 0.18. Intersubject variability was 167% for C25 and 218% for the effect half-life. None of the model parameters showed sex dependency. CONCLUSIONS A cumulative dose of 0.3 mg/kg M6G, given over 1 h, produces long-term analgesia greater than that observed with placebo, with equal dynamics (potency and speed of onset-offset) in men and women. Possible causes for the great intersubject response variability, such as genetic polymorphism of the micro-opioid receptor and placebo-related phenomena, are discussed. The predictive pharmacokinetic-pharmacodynamic model was applied successfully and was used to estimate M6G analgesia after morphine in patients with normal and impaired renal function.