Management of adverse effects after inadvertent intrathecal injection of morphine in a dog undergoing thoracotomy

A 33-kg adult Labrador Retriever dog presented for investigation of pneumothorax. A computed tomography scan confirmed the diagnosis, and right lateral thoracotomy and lung lobectomy were performed under general anaesthesia. During surgery, anaesthesia was maintained with a propofol constant-rate intravenous (IV) infusion (premedication with methadone and dexmedetomidine) and IV atracurium was given to facilitate surgery. A lumbosacral epidural injection of preservative-free morphine (0.3 mg kg^-1 diluted to 6.6 mL with sterile saline) was performed. Using a spinal needle, the 'hanging drop' method of detecting the epidural space was attempted but the saline drop was not aspirated, and cerebrospinal fluid entered the needle hub, indicating inadvertent dural puncture. The needle was partially withdrawn and a lack of resistance to injection confirmed before injection of morphine. Following injection, myoclonus of the tail was observed throughout surgery, continuing after atracurium administration. Following surgery, the propofol infusion was reduced from 0.3 to 0.1 mg kg^-1 minute^-1 to encourage spontaneous ventilation after confirmation of neuromuscular function. Spontaneous ventilation, swallow and palpebral reflexes did not return. The following treatments were unsuccessful in eliciting spontaneous ventilation: allowing end-tidal CO₂ to increase [up to 60 mmHg (8 kPa)], reducing inspired oxygen fraction to 0.4, injection of neostigmine and glycopyrrolate, and stopping propofol delivery. Three doses of butorphanol (each 0.5 mg kg^-1 IV) were given 30 minutes apart to antagonize the effects of morphine as it was suspected that inadvertent intrathecal injection had been performed. Butorphanol administration was associated with return of spontaneous ventilation, palpebral and swallow reflexes, and full recovery from anaesthesia. Continuous rate infusions of butorphanol and dexmedetomidine were used for 12 hours to treat any residual respiratory depression and provide sedation to manage continued myoclonus. Myoclonus resolved within 24 hours. The dog was discharged 3 days later and suffered no ongoing adverse effects.