Clinical manifestations

Presentation and course

The clinical manifestations of autonomic seizures are quite heterogeneous and vary according to the cortical areas of the central autonomic network from which the epileptic discharge arises and spreads. This article describes the symptomatology of autonomic seizures according to the 2017 ILAE classification (47). A detailed description of spatial and temporal sequence of the autonomic symptoms may have localizing and lateralizing value, and it is important to preliminarily distinguish autonomic seizures from their nonepileptic mimics.

Focal autonomic seizures with cardiovascular symptoms

Ictal sinus tachycardia. Sinus tachycardia with a heart rate of more than 100 beats per minute (bpm) at rest is reported to occur as an ictal phenomenon in 71% of focal-onset seizures.

Ictal sinus tachycardia was reported to be more prevalent in seizures arising from the temporal lobe than in those arising from extratemporal areas (15) and to occur earlier in seizures arising from the hippocampal formation than in those from extrahippocampal zones (24).

The currently available data in the literature concerning the lateralizing value of ictal sinus tachycardia are still inconclusive (15).

Ictal bradyarrhythmia and ictal asystole. Sinus bradycardia (ie, hearth rate lower than 60 bpm) as an ictal autonomic epileptic phenomenon occurs less frequently than sinus tachycardia, with its prevalence ranging from 1.3% to 5.5% of the cases. It may present as mild asymptomatic sinus bradycardia or severe symptomatic bradyarrhythmia (pronounced sinus bradycardia, sinus arrest, atrioventricular block). Prolonged asystole is reportedly in only 0.32% of patients with drug-resistant epilepsy at video-EEG monitoring. Long-term monitoring studies using implantable cardiac loop recorders (61) showed that ictal asystole may not constantly occur in every seizure and may go unnoticed during short-term video-EEG. Ictal asystole may occur early (less than 1 year) or late (more than 1 year) with respect to epilepsy onset and may last less than 30 or more than 30 seconds, with the ictal asystole longer than 30 seconds being less frequent and generally associated with focal seizures evolving to bilateral tonic-clonic seizures. Treatment with sodium channel blockers or antiepileptic drugs have been reported to not increase the frequency and duration of ictal asystole (59).

Most studies indicate that ictal bradycardia is associated with epileptic discharges arising from the temporal lobe, namely from the insula, in the left hemisphere (61; 59). Ictal asystole could be either a direct consequence of the epileptic discharge affecting the limbic area of the central autonomic network or an indirect effect of a vasovagal reflex triggered by the seizure, which may trigger a vasovagal reflex. Seizures with syncopal ictal asystole are shorter than those with nonsyncopal ictal bradycardia and appear to be self-limited (61). The cerebral anoxia caused by syncope related to asystole could be a potential mechanism of seizure termination (49; 61).

In medically refractory patients with ictal asystole and related falls, cardiac pacemaker implantation should be considered as it has been proven to be effective in preventing falls (38; 59).

Focal autonomic seizures with respiratory symptoms

Ictal hyperventilation. The prevalence of hyperventilation as an ictal epileptic autonomic focal seizure is poorly defined. This is due in part to the fact that respiratory parameters are not routinely assessed during video-EEG monitoring (04). Ictal hyperventilation is more frequently overserved with epileptiform discharges arising from mesial areas than from neocortical areas of the temporal lobe (16). This difference may reflect the connectivity between the structures of the mesial temporal lobe, the hippocampus, and autonomic nuclei of the brainstem. Harvey and colleagues reported that ictal hyperventilation also occurs as a symptom of epileptic seizures arising from the frontopolar and the orbitofrontal regions (22).

Ictal oxygen desaturation, ictal hypoventilation, and ictal central apnea. Ictal oxygen desaturations of mild (80% to 90%) to severe degrees (less than 70%) were reported in 70.8% of 79 focal seizures (32).

Ictal oxygen desaturations may be due to either hypoventilation with a parallel increase in end-tidal carbon dioxide (ETCO₂) or central apneas. Ictal oxygen desaturations significantly correlate with right temporal lobe seizure onset, with the most severe desaturations (less than 80% and less than 70%) being observed with long seizure duration and spreading of the seizure to the contralateral lobe (09; 01). Ictal central apneas were reported to precede EEG seizure onset by about 8 seconds in 54.3% of seizures and to be the only clinical manifestation in 16.5% of seizures monitored with EEG polygraphy, including breathing measurements. Of note, some ictal central apneas may go unnoticed by the patients (Lacuey et al 2017).

Autonomic seizures with gastrointestinal symptoms

Epigastric seizures. Epigastric seizures are associated with temporal lobe epilepsy in 73.6% of cases and are tendentially related to mesial temporal lobe epilepsy (MTLE) (Foldvary et al 1997; 23; Maillard et al 2004).

The lateralizing value of epigastric auras was initially stressed in some studies reporting that the epigastric aura was more frequently seen in seizures involving the nondominant temporal lobe (Gupta et al 1983; Taylor and Lochery 1987). However, more recent data suggest that epigastric auras have no lateralizing significance (23).

Ictal vomiting and retching in adults. Ictal vomiting and retching are infrequent visceral clinical seizure manifestations associated with epileptic discharges arising from a complex neuronal network involving medial and lateral parts of the temporal lobe, especially the lateral superior temporal cortex, including the insula and possibly the occipital lobes. This localization was documented by Kramer and colleagues in a study using subdural grid electrodes and by Baumgartner and colleagues in an ictal SPECT study (31; 06). However, Pietrafusa and colleagues reported episodes of ictal vomiting that were strictly related to epileptic discharges limited to mesial temporal structure (44).

Current data in the literature indicate that ictal vomiting and retching do not have any lateralizing value (Baysal-Kirac and Baykan 2015).

Ictal vomiting in children. Ictal vomiting is a more frequent event in children than in adults, and it is associated with idiopathic focal epilepsies of childhood, such as SeLEAS (43) and POLE (41). The mechanism causing vomiting in idiopathic focal epilepsies of childhood has been attributed to an infrasylvian spread of seizure to the nondominant temporal lobe (20). However, other hypotheses have been put forward. In fact, vomiting associated with occipital lobe seizures could be due to a postictal migrainous phenomena triggered by an occipital epileptic discharge (25).

Hypersalivation. Hypersalivation is a rare clinical ictal epileptic manifestation (51) indicating a mesial temporal lobe seizure onset, without lateralizing value (Janszky et al 2007).

Ictal flatulence. Ictal flatulence is a very rare ictal phenomenon. It was reported to occur in 0.6% of patients with focal epilepsies who were monitored at the Interdisciplinary Epilepsy Center of Marburg between 2006 and 2009. Ictal flatulence would indicate a temporal or insular involvement during the seizure, without a lateralizing value (56).

Ictal urge to defecate. Reported in a few case reports, this urge seems to indicate a seizure onset in the nondominant temporal lobe (57).

Autonomic seizures with cutaneous symptoms

Ictal flushing. On rare occasions, transient paroxysms of flushing, hypertension, tachycardia, and increase of catecholamine plasma levels with temporal lobe seizures have been reported (37). Furthermore, flushing is among the first symptoms in SeLEAS (54).

Ictal pallor. Pallor as an ictal symptom is reported to typically occur in SeLEAS (Panayiotopoulos 1999; Specchio et al 2010) and in temporal lobe epilepsy, mainly in patients with seizures with ictal fear and in patients with the so-called “abdominal epilepsy” featured in gastrointestinal symptoms, such as abdominal pain and nausea (18; 11).

Ictal sweating. This symptom has been anecdotally reported in various cases, such as a variant of Shapiro syndrome (Klein et al 2001), in a patient with “abdominal epilepsy” (Mendler et al 1998), and in a male patient with Ma2-positive autoimmune encephalitis associated with testicular teratoma (17). In this case, the sweating was very profuse and strictly involved the left face; the seizure-onset zone was localized to the ipsilateral (left) temporo-parietal or posterior insula.

Ictal piloerection. Ictal piloerection is reported to occur in 0.4% to 1.2% of patients with medically refractory seizures, mainly in patients with temporal lobe epilepsy (Stefan et al 2002; Rocamora et al 2003). Unilateral piloerection is most frequently associated with an ipsilateral seizure-onset zone, whereas bilateral ictal piloerection has no lateralizing value. Epileptic discharges arising from the insula or amygdala seem to underlie ictal piloerection manifestations (33).

Autonomic seizures with sexual and genital manifestations

Seizures are classified as sexual when symptoms have an erotic content and as genital when they involve the genitals without an associated erotic component. According to the 2017 ILAE seizure classification, sexual seizures can be classified as focal emotional seizures; genital seizures as focal sensory seizures; sexual automatisms as focal emotional or focal hyperkinetic seizures; and genital automatisms as focal seizures with automatisms.

The clinical manifestations of sexual seizures range from erotic thoughts and feelings to sexual arousal or orgasm and may be accompanied by viscerosensory and autonomic changes of sexual excitement. Sexual seizures occur more frequently in women than in men. They are reported to be associated with epileptic discharges in the right amygdala without spreading to the orbitofrontal, temporobasal, and insular cortex (12). Orgasmic feelings occur more frequently in seizures arising from the nondominant temporal lobe; however, they do not have absolute lateralizing significance (29; 27).

Seizures involving the genital areas consist of unpleasant, even painful, frightening or emotionally neutral somatosensory sensations in the genitals. They may be accompanied by ictal orgasm. They are associated with epileptic discharges in the parasagittal postcentral gyrus (10).

Sexual automatisms are characterized by hypermotor movements consisting of writhing, thrusting, and rhythmic movements of the pelvis, arms, and legs, sometimes in association with picking and rhythmic manipulation of the groin or genitals, exhibitionism, and masturbation. They are associated with epileptic discharges arising from the frontal lobe (62). However, genital automatisms consist of repeated grabbing, fondling, or scratching of the genitals and can be accompanied by masturbation and exhibitionistic behavior. They occur in 3% to 11.4% of patients with drug-resistant epilepsy, more frequently in men with temporal lobe seizures, and are associated with epileptic discharges in the ipsilateral hemisphere (13).

Autonomic seizures with urinary symptoms

Ictal urinary urge consists of an ictal “desire to void,” and it is a rare symptom that occurs in 0.4% to 3% of patients with temporal lobe epilepsy (28; 03), with the patient either remembering or being amnesic regarding the symptom in the postictal period.

In most patients, the seizure onset could be lateralized to the nondominant temporal lobe (03). The lateralization to the nondominant hemisphere can be explained by a hemisphere-specific representation of central bladder control. An ictal SPECT study performed during ictal urinary urge demonstrated hypoperfusion of the superior temporal gyrus and insular cortex (03). This hypothesis was further supported by PET studies in normal subjects in whom urination was associated with an activation of the right dorsal pontine tegmentum and the right inferior frontal gyrus, whereas during the filled bladder condition, the right frontal operculum or the right anterior insula were activated (08). Thus, epileptic activity within the insular cortex could be responsible for a filled bladder sensation, resulting in ictal urinary urge. Alternatively, ictal urinary urge could be mediated by the spread of epileptic activity to frontal lobe structures, ie, the right inferior frontal gyrus, which are responsible for suprapontine bladder control (05).

Biological basis

Etiology and pathogenesis

Autonomic dysfunctions result from a perturbation of the central autonomic network, a complex network of interconnected cortical, subcortical, and brainstem neuronal centers, which plays a fundamental role in regulating the function of the autonomic nervous system. Based on CNS lesional, neurostimulation, and fMRI studies, the main cortical and subcortical components of the central autonomic network are the hypothalamus, thalamus (in particular, ventral posterior medial and lateral nuclei), mesial prefrontal cortex, and insular cortex. The brainstem/medulla system includes the solitary tract, ambiguous, dorsal vagal, pre-Bötzinger/Bötzinger complex, parabrachial, and Kölliker-Fuse nuclei; the rostral and caudal ventral respiratory group; the serotoninergic raphe; and mesencephalic periaqueductal gray/reticular formation. In physiologic conditions, the cortical and subcortical components of the central autonomic network are activated by external stimulations and initiate the appropriate response via the brainstem/medulla system, with the posterior insula playing a key role in the autonomic integration of cortical and brainstem central autonomic network components (40).

Autonomic seizures occur due to a perturbation of the central autonomic network when an epileptic discharge arises within the anterior cingulate, insular, posterior orbitofrontal and prefrontal cortices; amygdala; or hypothalamus (14; 26), leading to an increase of the sympathetic activity that results in autonomic dysfunction (15). Autonomic seizures frequently occur in the context of temporal lobe epilepsy, namely MTLE of various etiologies.

Differential diagnosis

Confusing conditions

A detailed and accurate reconstruction of the symptoms, their temporal sequence, duration, place, timing, triggers, and stereotypy is the starting key point to distinguish autonomic seizures from paroxysmal nonepileptic events that can mimic them. However, a definite diagnosis cannot be reached in a few cases and require instrumental investigations, such as video-EEG polygraphy or long-term EEG or non-EEG wearable monitoring devices.

Autonomic seizures frequently need to be distinguished from acute cardiorespiratory, gastrointestinal, urogenital, sexual, and psychogenic events and, in cases of sleep-related seizures, some sleep disorders, namely arousal NREM parasomnia.

Ictal sinus tachycardia, bradycardia, ictal asystole, and other peri-ictal arrhythmias should be differentiated from cardiogenic arrhythmias during wakefulness and sleep. Because cardiogenic arrhythmias may coexist in a patient, it may be very difficult to disentangle the difference in the origin of the disorders. In most cases, instrumental investigations are needed to reach a definite differential diagnosis.

As for ictal respiratory manifestations, ictal hyperventilation or shortness of breath during focal emotional seizures should be differentiated from panic attacks and acute hyperventilation of a different origin (05).

Ictal oxygen desaturations, ictal hypoventilation, and ictal central apnea must be differentiated from primary breathing disorders during wakefulness and sleep (central sleep apnea, sleep-related hypoventilation syndrome, etc).

Epigastric aura is frequently misinterpreted in patients with gastric disorders, leading to an important delay in proper diagnosis of epilepsy. The same can be said for ictal vomiting, which may be misdiagnosed as primary gastric disorders (gastritis, pyloric hypertrophy, etc) or metabolic disorders (cyclic vomiting syndrome, migraine).

Ictal urinary urge and urination should be differentiated from these symptoms in common bladder neurologic and non-neurologic disorders (cystitis, bladder cancer), prostatitis, prostatic hypertrophy, and cancer.

Sexual and genital seizures should be differentiated from psychogenic disorders with genital or sexual manifestations and, when occurring during sleep, from sexsomnia episodes, a rare subtype of NREM arousal parasomnia (48). Sexual and genital automatisms need to be differentiated from repetitive self-stimulatory behavior, which may occur in young people. Sexual epileptic automatisms occurring during sleep may need to be distinguished from sleep-related rhythmic movement disorder (35).

Associated and underlying disorders

Sudden unexpected death in epilepsy. According to the current definition, sudden unexpected death in epilepsy (SUDEP) is defined as a sudden, unexpected, nontraumatic, non-drowning death in a person with epilepsy, with an autopsy negative for either an anatomical or toxicological cause of death (52).

SUDEP is the main cause of premature death (10% to 50%) in epileptic patients (60; 53; 46). A SUDEP incidence of 2.34 per 1000 person-years was reported in a Chinese community cohort of 1562 epileptic patients followed for a 5-year period (19).

SUDEP is comprehensively covered elsewhere in MedLink Neurology. This article focuses on the potential relationship between autonomic seizures and SUDEP.

The pathogenesis of SUDEP is multifactorial, complex, and still not completely understood. However, most studies concerning interictal and ictal findings in epilepsy patients who died from SUDEP indicated that the imbalance of the neural autonomic regulatory system of cardiorespiratory functions plays a key role. Cardiac-related mechanisms include reduction in heart rate variability and prolongation of QT interval, which can lead to arrhythmias. Neural mechanisms include impairment of 5-hydroxytryptamine (serotonin) and adenosine neuromodulation (19).

A few observations in patients dying from SUDEP during epilepsy monitoring suggested a postictal breakdown of autonomic control consisting of a severe alteration of the respiratory and cardiac function, leading to a generalized EEG suppression and eventually to terminal cardiorespiratory arrest (02; 50).

Autonomic seizures are frequently seen in resistant temporal lobe epilepsy. In these cases, there have been reports of brainstem atrophy with volume loss in regions involved in autonomic control. Structural damage in these regions is thought to increase the risk of a fatal dysregulation during situations with increased autonomic activity (39).

Based on these notions, focal autonomic seizures with cardiac and respiratory dysautonomic manifestations would be expected to be at high risk of SUDEP. However, there is no evidence of increased SUDEP occurrence in patients with self-limited autonomic seizures.

Diagnostic workup

Video-EEG may be particularly useful in identifying autonomic seizures and distinguishing them from nonepileptic phenomena. Wearable seizure detection devices may be helpful.

Management

Autonomic seizures are treated similarly to other seizures with antiseizure medications. Epilepsy surgery is indicated in drug-resistant cases. Dietary interventions and neurostimulation have a role.