THERE are indications that sevoflurane may be epileptogenic in some circumstances. 1–7There have been several reports of epileptiform electroencephalographic activity in adults and children and occasional reports of grand mal convulsions in children. However, there has been only one report of a grand mal–type seizure occurring in an adult, and the frequency of such events and the predisposing circumstances, if any, are undefined. We present the case of a 19-yr-old man with no seizure history who, during two consecutive emergences from general anesthesia with sevoflurane, experienced tonic–clonic motor activity consistent with grand mal convulsions.
A 19-yr-old man presented for left-sided radical orchiectomy. History was significant for methamphetamine use by inhalation. He was taking no medications and asserted no drug use in the preceding three weeks. There was no personal or family history of seizures or other neurologic disorders. Hemoglobin, electrolytes, and plasma glucose values were normal.
Midazolam, 1 mg, was administered before induction. Preoxygenation was performed with use of a semiclosed anesthesia circuit and a tightly applied mask. Induction of anesthesia was accomplished uneventfully by introducing 8% sevoflurane, in 50% nitrous oxide, with oxygen at a fresh gas flow of 10 l/min. Anesthesia was maintained with 50% nitrous oxide and 2% sevoflurane (dial setting) by mask. The patient breathed spontaneously. Respiratory rate was initially 28 breaths/min, with an end-tidal carbon dioxide concentration of 22–24 mmHg. Fentanyl, 100 μg, was administered in 25-μg increments. Respiratory rate decreased to 14 breaths/min and end-tidal carbon dioxide increased to 31–33 mm Hg. Blood loss was 400 ml. Lactated Ringer’s solution, 1 l, and 500 cc hetastarch were administered. The patient remained hemodynamically stable, with a blood pressure of 100–115/60–74 mmHg and a pulse rate of 85–100 beats/min throughout the 90-min procedure.
After wound closure, sevoflurane and nitrous oxide were discontinued. Oxygen administration continued at 8 l/min. Approximately 21/2 min later, rhythmic jerking movements began in the legs and spread quickly to the remainder of the body. The movements were accompanied by an arched back and a stiff neck and jaw. Effective respiration ceased. Oxygen saturation measured by pulse oximetry (Spo2) decreased to the mid 80s (%), and the capnograph showed no expired carbon dioxide. To facilitate ventilation via mask, sevoflurane was reintroduced at a dial setting of 8%. The tonic–clonic activity ceased within 30 s, and effective manual ventilation was achieved immediately. Spo2returned to normal rapidly, and the total duration of Spo2less than 90% was less than 2 min. The patient resumed spontaneous ventilation immediately. Approximately 3 min thereafter, sevoflurane administration was discontinued again. Within 2 min, an essentially identical occurrence of tonic–clonic activity ensued. Midazolam, 1 mg, and propofol, 30 mg, were administered, and the apparent seizure activity abated promptly. Mask ventilation with oxygen was reinstituted, and spontaneous ventilation again resumed quickly. Diazepam, 5 mg, was administered. During the next 10 min, the patient awakened but was disoriented.
Neurologic consultation was obtained, and diphenylhydantoin was administered. Serum toxicology screening yielded negative results for stimulant-type drugs. Computed tomography and magnetic resonance imaging scans showed a lesion in the left posterior cortex that was suspicious for ganglioneuroma (a benign tumor of mixed neuronal and glial origin). Encephalomalacia, secondary to trauma and cysticercosis, were offered as less likely diagnostic possibilities. There was no evidence of micrometastases. No further seizure activity was observed, and the patient’s sensorium returned to normal within 4 h. The patient later reported that he had experienced blunt trauma to the head approximately 2 yr previously. The radiologist’s opinion was that a traumatic lesion dating from that time would have been associated with more atrophy than was evident from the scans. Electroencephalography 24 h later was interpreted as normal during wakefulness, during sleep, and after hyperventilation. The testicular mass was reported subsequently to be a malignant germ cell tumor.
Epileptiform electroencephalographic activity and seizure-like motor activity have been reported in association with the administration of sevoflurane. The latter has occurred principally in children during induction with relatively high concentrations of sevoflurane. The majority of reports have described epileptiform electroencephalographic activity, mostly polyspike and polyspike-and-wave patterns, with minimal or no associated motor manifestations. Epileptiform electroencephalographic activity has been observed in healthy adults, 6,7a nonepileptic child, 5and two children with histories of epilepsy. 3There have been a limited number of reports of frank motor convulsions in association with sevoflurane. In a report by Haga et al. , 2“convulsions” were reported to have occurred during induction in 6% of pediatric patients who received either 4% or 6% sevoflurane. Adachi et al. 1reported seizure-like motor activity, also during induction, in a 9-yr-old girl. There has been only one case report of a grand mal convulsion occurring in an adult patient in association with administration of sevoflurane. 4In that case, an apparent grand mal convulsion also occurred during emergence from anesthesia. The neurologically normal adult patient had received a general anesthetic that included thiopental and vecuronium for induction, followed by a maintenance regimen similar to that used in the current patient, i.e. , nitrous oxide, sevoflurane, and fentanyl (50 μg). Neostigmine and atropine were administered to reverse neuromuscular blockade, and, soon thereafter, a grand mal convulsion occurred. Subsequent electroencephalography was normal. However, the patient did not undergo radiologic evaluation, and, therefore, the presence of small, potentially predisposing cortical lesions, such as that discovered in the present patient, cannot be excluded.
Potentially predisposing factors were sought in our patient. Computed tomography revealed a preexisting cortical lesion, which could have served as the nidus of seizure initiation. No other predisposing factors were identified. The patient’s history of illicit drug use was remote. No other anesthetic agents with well-established proconvulsant or central nervous system–irritant properties were employed. Although various narcotics have been suspected of having proconvulsant properties, fentanyl, in a dose of 100 μg, has never been implicated. The patient was not significantly hypocarbic during the procedure.
It is not possible to be certain whether the motor activity observed in our patient was a true convulsion, intense myoclonic activity, or something similar to the opisthotonic reactions that have been associated with propofol. 8In this situation, there was some suggestion of a Jacksonian progression and an apparent postictal state early in the recovery period, although interpretation of the latter may have been confounded by the administration of 5 mg diazepam. For the clinician managing one of these occurrences, the potential patient hazards associated with all three events are similar, and the necessity for readiness to intervene rapidly to prevent patient injury is equivalent.
This patient had no apparent neurologic sequelae. The same has been true of all of other reported patients in whom either abnormal electroencephalographic or motor activity has been observed. Accordingly, it seems inappropriate to recommend significant restraint on the use of sevoflurane. Events of this nature are rare. In addition, the current case included a confluence of circumstances that are probably relatively uncommon in adult anesthesia, i.e. , the use of an inhalation induction in an adult patient with a preexisting intracerebral lesion who received an anesthetic that included a minimal amount of agents with potential seizure suppressant effects (i.e. , no induction agent and only 1 mg midazolam). Nonetheless, our experience indicates that clinicians should be aware of the possibility of the occurrence of seizure-like activity at emergence from sevoflurane anesthesia. They should be particularly attentive to that possibility in patients whose preexisting state includes some condition that potentially might predispose to the occurrence of seizures, including intraparenchymal, especially cortical, cerebral lesions (scarring, primary tumors, metastases) or pharmacologic agents that lower seizure threshold.