ObjectiveTo evaluate perfusion index (PI) as a noninvasive tool to determine effectiveness and onset of epidural anesthesia in dogs. Study designProspective clinical trial. AnimalsA total of 21 adult dogs, aged 6.5 ± 3 years and weighing 34.9 ± 6.4 kg, undergoing a tibial plateau leveling osteotomy. MethodsDogs were premedicated intramuscularly with acepromazine (0.03 mg kg–1) and hydromorphone (0.1 mg kg–1) and anesthetized with intravenous propofol (to effect) and isoflurane in oxygen. A surface transflectance probe was secured to the tail base to monitor PI and a dorsal pedal artery catheter was placed for invasive blood pressure monitoring. A lumbosacral epidural was performed with the dog in sternal recumbency. Dogs were randomly assigned for inclusion of epidural morphine (0.1 mg kg–1) or morphine (0.1 mg kg–1) and lidocaine (4 mg kg–1). PI was recorded following instrumentation of each dog just prior to the epidural (baseline), at 10 minute intervals for 30 minutes, before and after the surgical skin incision and before and after completion of the osteotomy. Physiological variables and end-tidal isoflurane were recorded at the same time points. ResultsThere was no significant difference in PI between the groups at any time point. There was a significant change in end-tidal isoflurane before and after the skin incision in the epidural morphine and epidural morphine–lidocaine groups (p = 0.04, p = 0.05, respectively) and before and after the osteotomy in each group for heart rate (p = 0.001, p = 0.04), diastolic (p = 0.01, p = 0.01) and mean arterial blood pressure (p = 0.03, p = 0.05). Conclusions and clinical relevancePI did not provide an objective means for determining the onset or effectiveness of epidural anesthesia in anesthetized dogs and alternate methods of noninvasive assessment should be investigated.