Reflex epilepsies constitute 2–6% of all epilepsy cases. They are classified as a syndrome in which epileptic events are triggered by specific motor, sensory, or cognitive stimulation. Reflex seizures usually occur in association with spontaneous seizures. As highlighted by Irmen et al. [
7], the stimulus can be considered a reflex trigger only if it is strongly associated with the seizure. However, the probability that a certain trigger may elicit a seizure is lower than 100%. Although a specific stimulus can be assumed to act via known epileptogenic mechanisms or hyperexcitable networks, its triggering activity may depend on its degree of interference with the epileptogenic area. EE includes patients “who had more than 50% of fits during or within 30 min of eating breakfast, lunch or dinner” [
8]. Eating is a complex phenomenon, mainly involving psychic, sensory, proprioceptive, enteroceptive, and somatosensory stimuli [
2,
3]; hence, the exact pathophysiological mechanism by which eating seizures could be induced remains multifaceted. Case reports seem to stress that the action of eating may not solely act as a trigger and also chewing, swallowing, but the context of eating, the mere sight of food or the content of the meal may play a role (see Table
1). Reflex seizures provoked by eating are rare except for specific countries, e.g. Sri Lanka. Seneviratne et al. [
1] found one or more provoking factors causing reflex seizures in 47 out of 526 patients (8.9%) from the epilepsy clinic in Ratnapura General Hospital. Eating was the commonest provoking factor, being observed in 28 patients (59.6%). In a prospective study on 1287 epileptic patients seen at Peradeniya, 223 (17.3%) were found to have reflex epilepsy and eating was the commonest stimulus (191 patients, 85.7%) [
17]. The very high prevalence of EE in Sri Lanka [
1,
17], could be related to genetic or ethnic factors, as well as to the bulky meals rich in carbohydrates consumed by these patients. The high prevalence of EE in Asia is probably due to rice-made foods and heavy meals in such cases, a different mechanism is likely to be implicated such as gastric distention through a vagal reflex. However, familial clustering of a shared genetic background could not be excluded [
5]. Out of Sri Lanka, EE is very rare and more prevalent among young adults, even if neonatal cases have been described [
9]. EE usually starts in the second-third decade of life, with a male predominance (Table
1). A family history of epilepsy may be present, including siblings suffering from eating seizures as well [
5]. Eating can be reported as the only provoking factor or mixed with others, as eating-induced seizures are usually associated with symptomatic epilepsies [
4]. Seizures can occur in the early phase of eating (generally within the first 5 min) or in the middle or end of meal [
6]. Focal seizures with impaired responsiveness are the commonest seizure type followed by focal seizures without impaired responsiveness and with secondary generalization [
6]. Seizures usually arise in temporal-limbic or extra-temporal perisylvian regions and may progress to secondary generalization. EEG often shows focal spikes and sharp/slow waves in the temporal areas. Some patients show a lesional aetiology at neuroimaging; nearly 30 individuals with a likely genetic cause have been reported (Table
1). However, despite the aetiology, the large majority of the patients show EEG patterns consistent with the diagnosis of temporal lobe epilepsy [
3,
12,
13,
15]. The physiopathological mechanisms of EE are complex. The temporal lobe is interconnected with the amygdala function and the amygdala with its lower seizure threshold and propensity to induce masticatory seizures may probably be a likely target for stimulation by the oral and masticatory factors [
16]. When cases report the mere sight of food or sensory stimuli being able to elicit seizures, the temporal lobe, through the olfactory and gustatory stimuli, may again be involved [
14]. Two critical functional cortical loops are likely to be involved: the interconnections between the gustatory cortex (insular, parietal and frontal region) and the frontal-insular-hippocampal network [
18] or the frontal-opercular area related to oral movements [
19,
20]. In patients with cortical malformations involving these areas, repeated stimuli (such as eating) could induce the hyper-activation of these areas due to the hyperexcitability secondary to the cortical lesion. EE was also described in some rare genetic syndromes [
2,
10,
11,
21]. In some genetic epilepsies (e.g.,
SYNGAP1 encephalopathy,
MECP2 duplication syndrome), the electroclinical profile correlates with the dysfunction of a limited cortical region (especially the frontal-central cortex), despite the genetic alteration affects the entire brain. The emotional or autonomic components of eating, along with gastric distension or stimulation of the mouth or pharynx, may also play an important role in inducing seizures [
22‐
24]. Monotherapy or polytherapy were used; the most effective treatments reported were sodium valproate, carbamazepine, clobazam and phenobarbital [
6]. Monotherapy with carbamazepine likely offers the best seizure control in focal seizures [
1]. Patients with “generalized” reflex seizures usually benefit from epileptic drugs typically used for idiopathic generalized epilepsy, such as valproic acid, lamotrigine, or levetiracetam [
22]. In some cases, clobazam before meals showed to be effective in preventing eating-related seizures [
3,
5]. Some patients need combination therapy. Surgical treatment may be considered when seizures are intractable. In the patient reported by Blauwblomme et al. [
18] selective resection of a small region of the insula resulted effective in seizure control. Gujjar et al. [
25] further reported a patient with refractory EE who was successfully treated with a temporal lobectomy. The prognosis of reflex seizures is variable due to wide clinical heterogeneity and mostly depends on the nature of the underlying seizure disorder [
22]. Both our patients showed additional unprovoked seizures, which is consistent with the findings that unprovoked seizures do occur in EE. In our patients, refractory epilepsy with poor response to multiple antiseizure medications was observed
. Neuroimaging in patient 1 revealed a cortical bilateral dysplasia involving mid-temporal areas which contain bilateral representations of the face, mouth, and throat. They are also responsive to taste, touch, and proprioception from the tongue and mouth. Excitation of these areas is a normal phenomenon caused by eating, but the presence of structural lesions likely leads to a hyperexcitation fostering the epileptogenesis [
26]. The dysfunction of specific cortical regions in the context of a germline genetic alteration might account for the finding of reflex epilepsy in some genetic syndromes, as shown by patient 2. In this case, a relevant contribution to the neurological phenotype is provided by the haploinsufficiency of disease genes involved in the 15q25.3-q26.1 deletion. In particular, de novo loss of function variants in
CHD2 cause an early-onset epileptic encephalopathy and
CHD2 haploinsufficiency in our patient might therefore explain the neurodevelopmental disabilities, dysmorphic features, hypotonia, and complex epileptic phenotype with early onset, mixed, and drug-resistant seizure [
27]. Similarly, de novo loss of function variants in
POLG cause autosomal dominant progressive external ophthalmoplegia (PEO) type 1, a severe neurological condition characterized by progressive weakness of extraocular muscles, proximal or generalized myopathy, and neurological features (e.g., ataxia and parkinsonism) [
28]. Accordingly,
POLG haploinsufficiency might contribute to the neurological and neuromuscular phenotype observed in patient 2. Further research will help clarify the aetiopathogenesis of reflex seizures and shed light on the mechanisms underlying EE, to improve the diagnosis and the management of these rare conditions.
Table 1
Review of clinical, imaging, and genetic aspects of EE patients
| 12 | 13.5 y | 11 M; 1 F | eating alone (75%); eating + anxiety; eating + bathing; eating + spontaneously | normal (7 pts); + 5 pts. (41.6%), focal/bilateral sclerosis or gliosis | na |
von Stülpnagel et al., 2019 [ 2] | 8 | 6.9 y | 4 M; 4 F | biting; eating; chewing; oral sensory stimuli | normal | SYNGAP1 mutations |
Jagtap et al., | 47 | 14.3 ± 9.8 y | 41 M; 6 F | eating; eating rice made food; oral sensory stimuli | + 16 pts. (34%), mainly PC lesions | na |
Yacubian et al., | 3 | 15 y | 3 F | eating (independently of type of food) | normal | probably genetic due to familial clustering, but tested negative |
Sillanpää et al., | 1 | 0 y | F | breast feeding | normal | na |
Patel et al., | 6 | 11.3 ± 2.16 y | 3 M; 3 F | eating; eating rice made food; “thinking of eating” | + 5 pts. (83.3%), perysilvian F lobe and high F lesions | na |
Kokes et al., | 6 | 20.3 y | 4 M; 2 F | chewing; swallowing; oral sensory stimuli | + 4 pts. (66.7%), L hemisphere lesions | na |
de Palma et al., | 1 | 6 y | M | eating; oral and gustatory sensory stimuli (mainly spicy food) | normal | MECP2 duplication |
Roche Martínez et al., 2011 [ 11] | 1 | 16 y | F | eating (independently of type of food) | normal | Rett syndrome but MECP2, CDKL5, FOXG1 tested negative |
Bae et al., | 2 | 39.5 y | 1 M; 1 F | eating (independently of type of food) | normal | na |
d’Orsi et al., | 1 | 25 y | M | chewing; eating; swallowing | + bilateral opercular dysplasia | na |
Loreto et al., | 3 | 22.7 y | 2 M; 1 F | eating; sensory stimuli | + 2 pts. (66.7%), not specific | na |
Mandal et al., | 20 | na | 16 M; 4 F | eating; eating Indian, rice made food/ heavy meals | normal (in 7 pts. tested) | na |
Koul et al., | 78 | na | na | eating; swallowing | na | na |
Senanayake, | 20 | 20 y | na | eating | na | probably genetic due to familial clustering |
Koul et al., | 50 | 23.8 y | na | chewing; eating rice made food; swallowing | na | na |
Loiseau et al., | 2 | 20.5 y | 2 M | chewing (mainly); eating | na | na |
Nagaraja et al., | 13 | 14 y | 8 M; 5 F | chewing; eating Indian, rice made food/heavy meals; drinking | na | na |
Aguglia et al., | 3 | 21.3 y | 2 M; 1 F | chewing; eating (independently of type of food) | na | na |
Ahuja et al., | 3 | 21.7 y | 3 M | eating (only at home in 2 cases; both at home and outside in 1 case) | na | na |
Robertson et al., | 1 | 14 y | M | eating (independently of type of food) | + internal capsule astrocytoma (involvement of the right caudate nucleus) | na |
Cirignotta et al., | 1 | 16 y | F | eating (independently of type of food) | na | na |
Scollo-Lavizzari et al., | 1 | 12 y | M | chewing; eating; swallowing; sensory stimuli | na | na |