OBJECTIVE To assess the effect of halothane (H), isoflurane (I) or sevoflurane (S) on the bispectral index (BIS), and the effect of the addition of meperidine in dogs subjected to ovariohysterectomy. METHODS Prospective, randomized, blinded, clinical trial. METHODS Forty-eight female mixed-breed dogs, with weights varying from 10 to 25 kg. METHODS All dogs were premedicated with acepromazine (A) (0.1 mg kg(-1) IM) or A and meperidine (M) (3 mg kg(-1) IM) and they were divided into six groups of eight animals (AH, AMH, AI, AMI, AS, and AMS). Fifteen minutes after premedication they were anesthetized with propofol (5 mg kg(-1) IV) and then orotracheally intubated. Anesthesia was maintained with halothane, isoflurane or sevoflurane, respectively. The BIS, variables were recorded at 15 minutes after administering pre-anesthetic medication (T0); 10 minutes of anesthesia maintenance (T1); right ovarian pedicle ligation (T2); muscle suturing (T3); skin suture (T4) and 10 minutes after terminating the inhalant anesthetic (T5), respectively. RESULTS BIS values were decreased at all times when compared to the baseline values in all groups (p<0.05). In the comparative assessment between groups, the values obtained at T0 and T1 were similar for all groups. At T2, the values in AMH were lower than those obtained in AI, AMI and AS (p<0.05). At the same time significantly higher values were found for AI when compared to AMS (p<0.01). There was a correlation between the bispectral index and the expired anesthetic fraction in all groups. CONCLUSIONS Within groups given the same inhalant anesthetic the bispectral index was a good indicator for the degree of hypnosis in dogs, indicating a good correlation with the amount of anesthetic and the nociceptive stimulation. BIS was a less reliable indicator of relative anesthetic depth when comparing equipotent end-tidal concentrations between the three inhalants.