Dexmedetomidine, a highly selective alpha(2)-adrenoceptor agonist, is used in combination with local anesthetics for sedation and analgesia. We tested the hypothesis that dexmedetomidine used for sedation alters the convulsive potency of racemic bupivacaine and levobupivacaine in awake, spontaneously breathing rats. In the first experiments, male Sprague-Dawley rats were randomly divided into six groups: bupivacaine with no dexmedetomidine (bupivacaine control; BC), bupivacaine with small-dose dexmedetomidine (BS), bupivacaine with large-dose dexmedetomidine (BL), levobupivacaine with no dexmedetomidine (levobupivacaine control; LC), levobupivacaine with small-dose dexmedetomidine (LS), and levobupivacaine with large-dose dexmedetomidine (LL) (n = 10 for each group). Continuous infusion of dexmedetomidine (Groups BC and LC, 0 microg x kg(-1) x h(-1); Groups BS and LS, 3.6 microg x kg(-1) x h(-1); and Groups BL and LL, 10.8 microg x kg(-1) x h(-1)) was started after bolus injection (Groups BC and LC, 0 microg/kg; Groups BS and LS, 0.5 microg/kg; and Groups BL and LL, 1.5 microg/kg). Fifteen minutes after the start of the dexmedetomidine infusion, continuous infusion of bupivacaine (Groups BC, BS, and BL) or levobupivacaine (Groups LC, LS, and LL) at 1 mg x kg(-1) x min(-1) was started and continued until tonic/clonic convulsions occurred. Dexmedetomidine achieved significantly different sedation levels both in Groups BC, BS, and BL and in Groups LC, LS, and LL (P < 0.05). Convulsive doses of bupivacaine and levobupivacaine were significantly larger in Groups BL and LL than in Groups BC and LC, respectively (P < 0.01 for both). Concentrations of bupivacaine and levobupivacaine in plasma and in brain at the onset of convulsions were also larger in Groups BL and LL than in Groups BC and LC (P < 0.01 for both). In the second experiment, yohimbine (1 mg/kg) administered 10 min before and 5 min after the start of dexmedetomidine infusion completely reversed the sedative effect of dexmedetomidine (bolus 1.5 microg/kg, followed by 10.8 microg x kg(-1) x h(-1)). Convulsive doses and plasma and brain concentrations of bupivacaine and levobupivacaine at the onset of convulsions in rats receiving yohimbine and dexmedetomidine were significantly smaller than in those receiving only dexmedetomidine (P < 0.05 for all) and were similar to those without dexmedetomidine or yohimbine. We conclude that dexmedetomidine used for sedation decreases the convulsive potency of both bupivacaine and levobupivacaine in rats. Alpha(2)-adrenoceptor agonism may be involved in this anticonvulsant potency.