ObjectiveTo compare the effects of constant rate infusions (CRI) of fentanyl or dexmedetomidine, combined with lidocaine and ketamine, on cardiovascular response during surgery, sevoflurane requirement and postoperative pain in dogs undergoing mastectomy. Study designProspective, randomized, blinded, clinical trial. AnimalsA total of 29 female dogs with mammary tumors. MethodsPremedication consisted of intramuscular acepromazine and morphine. General anesthesia was induced with intravenous propofol and maintained with sevoflurane. Dogs were randomized to be administered intravenous DLK [dexmedetomidine 1 μg kg–1 loading dose (LD) and 1 μg kg–1 hour–1; lidocaine 2 mg kg–1 LD and 3 mg kg–1 hour–1; ketamine 1 mg kg–1 LD and 0.6 mg kg–1 hour–1; n = 14] or FLK (fentanyl 5 μg kg–1 LD and 9 μg kg–1 hour–1; same doses of lidocaine and ketamine; n = 15) during anesthesia. Cardiorespiratory variables and end-tidal sevoflurane (Fe′Sevo) were recorded during surgery. The number of dogs administered ephedrine to treat arterial hypotension [mean arterial pressure (MAP) < 60 mmHg] was recorded. Meloxicam was administered to both groups. Postoperative pain and rescue analgesia requirement were assessed for 24 hours using the short form of the Glasgow Composite Measure Pain Scale. Data were compared using a mixed effects model or a Mann–Whitney test. ResultsMore dogs required ephedrine in FLK than in DLK (67% versus 7%). Heart rate was not significantly different between groups, whereas lower values of MAP (p ≤ 0.01) and Fe′Sevo (p = 0.018) were observed in FLK than in DLK. Rescue analgesia was administered to 2/15 dogs in FLK and 0/14 dogs in DLK. Conclusions and clinical relevanceBased on the cardiovascular response during surgery, intraoperative infusions of FLK and DLK provided adequate antinociception. Infusion of DLK provided greater stability of blood pressure. Both protocols resulted in minimal need for additional analgesia within 24 hours postoperatively.