This study was performed to develop an integrated pharmacokinetic-pharmacodynamic model for estimating the contribution of morphine-6-glucuronide (M6G) to morphine-associated antinociception in humans. Healthy volunteers (n = 8) received 10 mg of morphine sulfate as a 5-minute i.v. infusion. A Contact Thermode heat probe was placed on the volar forearm to elicitpain. Thermal threshold, defined as the temperature at which pain was first perceived, was measured at fixed time intervals over 8 hours. Serum concentrations of morphine and M6G were determined by LC/MS. Concentration- and effect-time data were analyzed by stepwise nonlinear least-squares regression. The pharmacodynamic parameter estimates were recovered with a linear effect-compartment model and were used to assess the contribution of M6G to morphine-associated analgesia. The estimates (mean +/- SEM) for morphine total clearance and steady-state volume of distribution were 1.0 +/- 0.07 L/h/kg and 1.6 +/- 0.1 L/kg, respectively. The AUC ratio of M6G to morphine was 0.73 +/- 0.06. The contribution of M6G to analgesia ranged from < 0.1% to 66% and was inversely related to the overall effect elicited by the morphine dose (r2 = 0.776). Differences in gender were observed where the contribution (mean +/- SEM) of M6G to analgesia was 32% +/- 19% in males (n = 3) and 13% +/- 8% in females (n = 5). These results suggest that as the overall effect of morphine increases, the fractional contribution of M6G declines and the contribution of M6G to analgesia may differ between males and females. Alterations in the M6G/morphine system may have clinically significant pharmacodynamic consequences.