We present a case of inadvertent injection of a mixture of rocuronium and morphine into the caudal epidural space in an awake patient.

A 53-yr-old, 80-kg, otherwise healthy man was scheduled to undergo emergency surgery during general anesthesia because of a rectal abscess. For postoperative analgesia, caudal injection was planned before general anesthesia with 13 ml of a mixture of 100 mg lidocaine and 3 mg morphine. After premedication with 3 mg midazolam and 0.5 mg atropine intramuscularly and standard anesthetic monitoring, with the patient in the prone position, the caudal space was easily identified with a 25-gauge needle. After a negative aspiration test result for blood or cerebrospinal fluid, the prepared solution was injected in three divided doses. The patient reported pain and a burning feeling during every injection. The needle was withdrawn 5 mm, and the injection was continued. Nevertheless, he still reported the burning feeling during every injection. Fifteen seconds after the injection was completed, he reported difficulty in breathing and consequently in speaking. First, total spinal anesthesia was considered, and the patient was immediately turned to the supine position. His spontaneous breathing was very weak, and his eyelids were lowered. Assisted ventilation was started via  facemask with 100% oxygen, and 100 mg propofol was administered for possible awareness. At this point, the patient’s blood pressure and heart rate were very stable. It was then realized that a mixture of 100 mg rocuronium and 3 mg morphine had been administered into caudal space because of a preparation error. Five minutes after caudal injection, the patient was intubated easily with an additional 100 mg propofol, and anesthesia was maintained with 1.5% sevoflurane in nitrous oxide and oxygen. The patient’s wife and surgeons were fully informed; it was decided to perform the surgery because it was an emergency case. After obtaining written informed consent from his wife, the patient was placed in the lithotonic position, and surgery was started. After the first 20 min, he had pinpoint pupils. There was nothing else to note preoperatively. The surgery was completed 45 min after caudal injection, and neuromuscular blockade was reversed with 2.5 mg neostigmine and 1 mg atropine intravenously. Two minutes after the reversal, the patient was maintaining an adequate breathing rate and depth, and the endotracheal tube was removed. After an additional 5 min, his sustained head lift was longer than 5 s, and he could obey verbal commands. Ten minutes after reversal, he was completely awake and oriented. His neurologic findings were normal except for pinpoint pupils. There was no motor or sensorial block, and he was completely pain free. He could void easily 5 h later. On the second day, neurologic examination was normal except that his pupil size was still 2 mm. Thirty hours after injection, his pupil size had completely recovered, and he was still pain free. Three days later, he was discharged from the hospital without any complaints. Thereafter, we followed up with the patient by telephone for 1 month. He never experienced any problems.

No data about inadvertent injection of rocuronium or other nondepolarizing neuromuscular blockers (NDNMBs) into the caudal epidural space in awake patients have been documented in the literature. Vecuronium1and cisatracurium2were reported without any neurologic or cardiovascular side effects or other symptoms of local or systemic toxicity. However, because both drugs were administered through a catheter into the lumbar epidural space during general anesthesia, the patients were unable to report problems, and some symptoms were not observed.

Drugs administered inadvertently into the epidural space have resulted in serious morbidity and mortality due to a direct drug or drug-additive neurotoxic, pH, or osmolality effect.3Rocuronium, a steroid NDNMB, has a rapid onset and an intermediate duration. It is an isotonic solution, and its isotonicity is obtained using sodium chloride; a pH of 4 is achieved by adding acetic acid, sodium hydroxide, or both. Its metabolite, 17-desacetyl-rocuronium, has been observed rarely in the plasma or urine of humans.*

It has been demonstrated that NDNMBs introduced into central nervous system are pharmacologically active, and they cause an increase in intracellular calcium concentrations and activation of either nicotinic acetylcholine receptors or glutamate receptors in rat brain, resulting in autonomic dysfunction, weakness, prolonged neuromuscular blockade, neuronal injury and death, and seizures.4,5 

The only side effect of the current inadvertent injection was nonreactive pinpoint pupils perioperatively and postoperatively. A possible explanation of the miosis may be the migration of rocuronium and morphine solution from the epidural space into the cerebrospinal fluid, then to the basal cisternae, and thence through brain tissue around the fourth ventricle. The patient’s lithotonic position might have magnified intracranial spread of drugs. In humans, morphine has a miotic effect, which is generally explained by direct stimulation of preganglionic parasympathetic fibers in the Edinger-Westphal nucleus.6Prolonged time of the miosis (e.g. , 30 h) is appropriate with the analgesic time of the morphine applied neuroaxially.

As mentioned above, contrary to what was seen in the neuromuscular receptors, NDNMBs may activate, rather than inhibit, particular subtypes of nicotinic acetylcholine receptors in the central nervous system.5Therefore, activation of brain acetylcholine receptors by NDNMBs might participate in miosis mechanism. That may be an answer to the question, Why were other side effects of opioids, such as respiratory depression, sedation, and mental status changes, not observed with miosis? It can be postulated that activation of brain acetylcholine receptors prevented these side effects.

The caudal epidural space is rich in venous plexuses. Although the rate of absorption of rocuronium from the caudal space is unknown, the onset time may be estimated to be between that of an intravenous and an intramuscular bolus. Intravenous injection of 1.2 mg/kg rocuronium (approximately the dose of the current injection) provides good intubating conditions in most patients in less than 2 min, and this dose may be expected to provide 67 (38–160) min of clinical relaxation.*It has been demonstrated that the absorption half-life is approximately 6.6 min, plasma concentrations peak at 13 min, and less than 4% of the drug remains to be absorbed from the intramuscular depot 30 min after rocuronium is administered into the deltoid muscle in children.7In our patient, constitution of excellent intubation condition was slower (5 vs. < 2 min) but duration of action was not longer as compared with clinical duration after intravenous administration (50 vs.  67 [38–160] min).

In this case, pain during injection should have warned us about the wrong drug administration. However, sometimes caudal injection may also be painful because of subperiosteal injection. In addition, paresthesia or a feeling of fullness is common during injection, and some anxious patients may feel and report this as pain.

When an inadvertent epidural injection has occurred, some practitioners attempt to dilute the concentration of the drug in the epidural space by flushing with distilled water or saline. Others have used epidural or intravenous corticosteroids to reduce the inflammatory response.3But these attempts were speculative, and they could potentially worsen the situation because of upward spread of drugs. Therefore, we decided only to observe the patient and provided symptomatic and supportive treatment as required.

† Ataturk University, Erzurum, Turkey. mcesur@atauni.edu.tr

Kostopanagiotou G, Mylona M, Massoura L, Siafaka I: Accidental epidural injection of vecuronium. Anesth Analg 2000; 91:1550–1
Vassilakos D, Tsakiliotis S, Veroniki F, Zachariadou G, Giala M: Inadvertent epidural administration of cisatracurium. Eur J Anaesthesiol 2004; 21:663–72
Hew CM, Cyna AM, Simmons SW: Avoiding inadvertent epidural injection of drugs intended for non-epidural use. Anaesth Intensive Care 2003; 31:44–9
Szenohradszky J, Trevor AJ, Bickler P, Caldwell JE, Sharma ML, Rampil IJ, Miller RD: Central nervous system effects of intrathecal muscle relaxants in rats. Anesth Analg 1993; 76:1304–9
Cardone C, Szenohradszky J, Yost S, Bickler PE: Activation of brain acetylcholine receptors by neuromuscular blocking drugs: A possible mechanism of neurotoxicity. Anesthesiology 1994; 80:1155–61
Murray RB, Adler MW, Korczyn AD: The pupillary effects of opioids. Life Sci 1983; 33:495–509
Reynolds LM, Lau M, Brown R, Luks A, Sharma M, Fisher DM: Bioavailability of intramuscular rocuronium in infants and children. Anesthesiology 1997; 87:1096–105