The effects of systemically administered medetomidine, an alpha(2)-adrenoceptor agonist, and atipamezole, an alpha(2)-adrenoceptor antagonist, on mustard oil-induced central hyperalgesia were determined in unanesthetized rats. The mechanical threshold for eliciting a hindlimb flexion reflex (a nocifensive response) was determined with a series of calibrated monofilaments. Under control conditions mustard oil produced a significant decrease of the hindlimb withdrawal threshold for mechanical stimuli applied to a distal site in the hindlimb, whereas the corresponding threshold in the (untreated) contralateral side was not changed. Medetomidine administered 12 min prior to mustard oil treatment produced a significant dose-dependent (3-30 mu g/kg s.c.) attenuation of the mustard oil-induced threshold decrease whereas the withdrawal threshold of the contralateral (untreated) hindlimb was not changed at these low doses. The antinociceptive effect of medetomidine (10 mu g/kg) administered 12 min prior to the mustard oil treatment was not significantly stronger than the effect of medetomidine administered immediately after the mustard oil treatment. Atipamezole at a high (1000 mu g/kg) or a low (10 mu g/kg) dose did not influence the mustard oil-induced threshold decrease, whereas at an intermediate dose (100 mu g/kg) atipamezole alone had a significant antinociceptive effect on mustard oil-induced hyperalgesia. The results indicate that medetomidine produces a selective attenuation of central hyperalgesia at doses which are sub-antinociceptive in intact rats. A pre-emptive treatment with medetomidine did not produce stronger antinociception than medetomidine treatment after the development of hyperalgesia. An alpha-adrenoceptor antagonist, atipamezole, attenuated central hyperalgesia in a non-monotonic fashion.