Severe postictal laryngospasm as a potential mechanism for sudden unexpected death in epilepsy: a near-miss in an EMU


Address correspondence to Jinny Tavee, MD, Department of Neurology S90, Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195, U.S.A. E-mail:


A 42-year-old man with refractory epilepsy experienced a 1-min generalized tonic–clonic seizure followed by persistent inspiratory stridor and cyanosis while being monitored in our epilepsy monitoring unit (EMU). Although his cardiac parameters remained stable throughout the event, the patient’s respiratory status rapidly declined, despite the urgent administration of oxygen via bag-valve-mask. He was subsequently intubated by the emergency code team, who noted severe laryngospasm while trying to insert the endotracheal tube. The patient was successfully resuscitated. This monitored case demonstrates that postictal laryngospasm may represent another potential cause of sudden unexpected death in epilepsy (SUDEP).

SUDEP is characterized by sudden death in epilepsy patients that occurs in the absence of trauma, drowning, status epilepticus, or an identifiable anatomic cause at autopsy (Nashef, 1997). Certain risk factors for SUDEP have been identified, and there is a high association with generalized tonic–clonic (GTC) seizures; however, the pathophysiology of SUDEP is still unknown. Suspected mechanisms include cardiac arrhythmia, seizure-related apnea, postictal respiratory arrest, and primary cessation of brain activity. A small number of witnessed cases of SUDEP have been reported in the literature, with only a handful that were reported to have occurred in the setting of continuous electroencephalographic (EEG) monitoring. We report a patient who experienced severe postictal laryngospasm that resulted in a near-SUDEP while undergoing EEG and video monitoring.

Case Report

A 42-year-old right-handed man with a history of medically refractory epilepsy was admitted to the EMU at the Cleveland Clinic for presurgical evaluation with continuous EEG and video monitoring. Perinatal history and early developmental milestones were normal until the patient was 2 years old, at which time he developed viral encephalitis. He was subsequently noted to be cognitively delayed with major learning disabilities. He had his first seizure at the age of 6 years. His typical seizures lasted several minutes and were described as an indescribable aura followed by impaired awareness with or without automatisms and rare secondary generalization. Seizure frequency had been progressively increasing over the last year, and at the time of admission was up to four per month. He had failed to achieve seizure control despite trials of five major anticonvulsants, and was taking lamotrigine 600 mg daily, which was discontinued on admission. General medical and neurologic examinations were normal, except for posttraumatic blindness of his right eye, which had been injured during a seizure.

At the time of admission, magnetic resonance imaging (MRI) of the brain showed an atrophic left hippocampus and arachnoid cyst in the choroidal fissure of the left temporal lobe. During 48 h of recording, the patient had one aura and five partial seizures, two of which progressed to secondary generalization. Interictal EEG study revealed frequent sharp waves at the left sphenoidal electrode and somewhat less frequent sharp waves at TP9, T7, and T9. Ictal EEG study revealed paroxysmal fast activity over the left hemisphere, with maximal amplitude in the temporal leads. Seizure number six occurred during morning rounds with two of the authors present (HM and JT). The patient awoke from sleep with his right face and arm twitching. This was followed by extension of the right arm, flexion of the left arm, head version to the right, and secondary generalization; the seizure duration was 82 s (Figs. 1 and 2).

Figure 1.

 EEG recording of seizure followed by near-SUDEP event.

Figure 2.

 Clinical onset: right head version.

Following termination of the seizure, the patient developed loud inspiratory stridor and marked cyanosis (Fig. 3). Attempts to improve ventilation by repositioning the patient and administering oxygen via emergency bag-valve-mask did not improve his respiratory status. The patient maintained normal blood pressures throughout, with no evidence of cardiac arrhythmias. The emergency code team was called and the patient was promptly intubated in the EMU. The anesthesiologist consultant noted laryngospasm and frothy secretions from his airways at the time of intubation. The patient was then transferred to the intensive care unit where a chest x-ray showed no evidence of pulmonary edema. The attending pulmonologist stated the cause for the respiratory emergency was laryngospasm that was likely triggered by aspiration.

Figure 3.

 Onset respiratory stridor.

In the intensive care unit, the patient was given a loading dose of phenytoin intravenously with a postload level of 15.8 μg/ml. Postictally, EEG and video monitoring demonstrated continuous slowing, until 6 h after the seizure when the patient’s electroencephalogram showed regular bursts of paroxysmal fast activity with continuous generalized slowing consistent with nonconvulsive status epilepticus. Overnight, he was successfully treated with valproate, propofol, benzodiazepines, and phenytoin, and he was awake and alert the following morning. He recovered to baseline status over the next few days and was discharged home on valproate.


Although SUDEP accounts for up to 17% of deaths in patients with epilepsy, the terminal event is usually unwitnessed (So et al., 2000). As described in a number of retrospective studies, which included witnessed cases of SUDEP, most events occurred at home and were preceded by generalized tonic–clonic seizures (Earnest et al., 1992; Nashef et al., 1998; Langhan et al., 2000; Lear-Kaul et al., 2005). Including our case, there are now three cases of SUDEP and two of near-SUDEP that have been reported in the setting of continuous EEG monitoring (Table 1) (Bird et al., 1996; Lee, 1998; So et al., 2000; McLean & Wimalaratna, 2007). All of the events except for one case of SUDEP occurred in an EMU during video-EEG monitoring. The exceptional case occurred at home while the patient was undergoing outpatient ambulatory monitoring; the patient died on the morning she was to follow up for review of the EEG study (McLean & Wimalaratna, 2007). Like our patient, all of the cases had a history of intractable, longstanding epilepsy, with a duration of at least 20 years. All events were immediately preceded by a GTC seizure, with four patients demonstrating focal EEG onset and secondary generalization. All but one patient had subtherapeutic anti-epileptic drug (AED) levels that were either iatrogenic (held for monitoring) or possibly the result of noncompliance.

Table 1.   Case reports of patients with SUDEP or near SUDEP while undergoing EEG monitoring
Author Age onset/ sexEpilepsySeizure length preceding SUDEPAED/levelEEGSUDEP mechanism
  1. AED, anti-epileptic drug; CPR, cardiopulmonary resuscitation; EEG, electroencephalography; ECG, electrocardiography; GTC, generalized tonic–clonic; SUDEP, sudden unexpected death in epilepsy.

Tavee (2008) [this article] (near SUDEP)42MChildhoodGTC 82 sLamotrigine/ subtherapeutic (held)Focal onset: left temporal Postictal diffuse slowingPostictal laryngospasm; successful CPR
So et al. (2000) (near SUDEP)20FInfancyGTC 56 sFelbamate Valproate
Nonlateralizing onset with bifrontal slow waves Postictal diffuse slowing 20 s, gradually followed by generalized suppressionPostictal central apnea induced bradycardiaand cardiac arrest; successful CPR
Bird et al. (1996)47MInfancyGTC 150 sPhenytoin Gabapentin
Carbamazepine Subtherapeutic (held)
Focal onset: right temporal
Burst suppression with spindling spike discharges for 16 s, then flat in right hemisphere followed by left hemisphere 8 s later
Cessation of brain activity; continuing pulse artifact for 120s implying noncardiac cause of death
Lee (1998)41F19 yearsGTC 70 sPhenobarbital
Subtherapeutic (held)
Focal onset (location not reported) Postictal slowing with left temporal sharp waves for 40 s, followed by marked suppressionCessation of brain activity; EKG with initial bradycardia that recovered, then progressively slowed and ceased 18 min after seizure
McLean and Wimalaratna (2007)50FChildhoodGTC 56 sLamotrigine
Onset unknown
Progressive spike wave d/c, then continuous polyspikes for 52 s with abrupt termination and flat EEG
Cessation of brain activity; no EKG or respiratory monitoring

In the three cases of SUDEP, one of which was intracranially EEG-monitored, a generalized convulsive seizure was followed by marked EEG suppression and then electrocerebral silence. Because the postmortem examination in all three patients demonstrated no significant evidence of pulmonary or cardiac abnormalities, the cause of death was postulated by the authors to be primary inhibition of brain activity. However, none of these cases were observed directly during the terminal event, and cardiac monitoring was performed in only one case, which was reported by Lee (1998). The authors of the SUDEP case that was intracranially monitored noted pulse artifact in the electroencephalogram at the onset of the electrocerebral silence, implying that the heart was still beating at the time of the event (Bird et al., 1996) In the other near-SUDEP case reported by So et al. (2000), persistent apnea following a generalized tonic–clonic seizure was witnessed. No electrocardiographic (ECG) changes were seen initially, but a progressive bradycardia ensued after 10 s, and after a minute the heart stopped completely. Cardiopulmonary resuscitation was successful and the authors of this article suspected that postictal apnea was the primary etiology.

In our case, the development of severe and persistent laryngospasm toward the end of the seizure would likely have led to the patient’s death had he not received urgent intubation. Although central apnea and severe pulmonary edema are certainly potential respiratory causes of SUDEP, our patient clearly had harsh, stridorous breathing and was found to have a narrowed airway on placement of the endotracheal tube, consistent with laryngospasm. Witnesses in one series reported that 12 of 15 patients had experienced difficulty breathing before death, whereas another study described a witnessed case of SUDEP as exhibiting “stentorous breathing” following a nocturnal GTC seizure (Nashef et al., 1998; Langhan et al., 2000). Ictal respiratory compromise has also been reported in the setting of complex partial seizures, in which six patients were found to have transient reductions in oxygen saturation, ranging from 77–91% within 1–2 min of seizure onset (Blum et al., 2000).

Our case is in agreement with the mechanism proposed by So et al., in that the terminal seizure resulted in respiratory distress (apnea in their case; stridor in ours), which is one of multiple probable mechanisms of SUDEP. Although the shared characteristics of these patients cannot be extrapolated to describe SUDEP risk factors for a larger epilepsy population, given the limited number of cases, these findings are supportive of what previous studies have identified as potential risk factors for SUDEP (Walczak et al., 2001; Nashef et al., 2007).


We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Disclosure of conflicts of interest: None of the authors has any conflict of interest to disclose.