Introduction

Reflex epilepsy (RE) or reflex seizures refers to epileptic syndromes characterized by focal or generalized seizures elicited by a specific stimulus or activity. These stimuli can be simple (visual, auditive, proprioceptive, or tactile) or complex (while eating, talking, tooth-brushing, bathing, etc.)¹,². RE estimated prevalence represents 4-7% of all epilepsies and 21% of idiopathic generalized type, being photosensitive epilepsy the most common³.

Eating epilepsy (EE) is a rare form of RE with a prevalence of 1:1000-2000 of all patients with epilepsy, although it ranges higher in Asia³-⁶. EE is more common in males and usually presents as focal seizures with impaired awareness which can occur before, during, and/or after eating³,⁴,⁷.

In this critical review, we aimed to answer the following questions: which are the sociodemographic, semiological, and radiological characteristics of EE in the general population?

Background

Eating is a complex mechanism that includes olfactory, taste, somatosensory, and other interoceptive inputs, which activate brain structures such as the insular cortex, frontal operculum, orbitofrontal cortex, and the amygdala⁸. The hypothalamus and other components of the autonomic nervous system regulate central and peripheral homeostasis of digestion and metabolism⁹. These structures are of particular interest in the study of hunger, satiety, obesity, as well as eating and body image disorders such as anorexia and bulimia. Moreover, all of the aforementioned structures are well recognized as trigger zones of EE in various studies⁸,¹⁰.

Although the pathophysiology of EE is not clear, according to case reports and small case series in the literature, plausible mechanisms for its appearance have been proposed. A genetic involvement is shown in subjects with mutations in genes SYNGAP1², MECP2¹¹; as well as familial EE⁵, associated with Rett¹², Cri du Chat,¹³,¹⁴ and congenital or acquired bilateral opercular syndromes¹⁵,¹⁶. Ethnic origin and environmental factors such as chemical composition of food, culinary habits, feeding behavior, and emotional and psychological involvement may be important in EE etiology⁷,¹⁷. EE can present with other epileptic syndromes, structural pathologies (malformations, tumors), and/or brain injury (hypoxic brain damage, gliosis, encephalomalacia, and meningoencephalitis)⁴,⁷,¹⁷. Therefore, it has been proposed that EE may be a long-term manifestation of an initial precipitating event in the past⁷,¹⁷.

Clinical presentation of EE is diverse. Seizures usually present as focal onset with impaired awareness and less commonly with generalized onset, including atonic, myoclonic, or even autonomic types³,⁴,⁷,¹⁷. EE occurs more frequently in the context of temporal or frontal lobe epilepsies⁴ and it usually is symptomatic or associated with imaging abnormalities in these locations (Tables 1 and 2).

In patients with EE, seizures can be caused by diverse and heterogeneous stimuli; nevertheless, all of them are related to the feeding process, either just before, during, or after it. These triggers may involve visual and olfactory stimuli (sight and smell), digestive and autonomic functions (salivating, chewing, swallowing, gastric distention, and gastric acid secretion), proprioceptive stimuli, thinking about food, bulky meals rich in carbohydrates, and gastroesophageal reflux²,¹²,¹⁷-²⁰. Moreover, association of EE to vasovagal syncope attacks has been reported, suggesting a vagal mechanism⁷.

Electroencephalographic (EEG) findings can be normal or present focal or diffuse epileptiform abnormalities³. Magnetic resonance imaging (MRI) can also be normal or present structural abnormalities specifically in temporolimbic or suprasylvian areas, particularly in the insular and opercular cortex³,⁴,⁷,¹². A clinical neurotopographic correlation can be made with EEG finding and/or imaging findings.

Although identification and avoidance of stimulus in some patients could aid in seizure control, in EE, this is not always possible, except in very specific associated situations³. Some patients respond and benefit from antiepileptic drugs. Although management of choice has not yet been established, intake of clobazam before meals may be effective as add-on therapy in the management of EE³,⁴. Kokes et al. suggest that seizures that originate from the left temporal region may be more resistant to antiepileptic management⁷. Some patients with therapy resistance may benefit with surgical treatment²¹ or vagus nerve stimulation²², especially those with imaging abnormalities²¹.

Methods

This article is based on unsystematic research in Google Scholar and PubMed for original manuscripts about "Eating epilepsy," "Eating reflex seizures," "Eating seizures," "Reflex eating epilepsy," and "Reflex eating seizures;" followed by a discretionary selection of publications. We included observational studies (such as case reports and case series) published between 1967 and 2020. Editorial notes, literature systematic reviews and clinical images were excluded from the study.

Results

In this review, we analyzed 18 case series and 23 case reports (Tables 1 and 2) and we included the case of a 40-year-old male patient with EE evaluated in our epilepsy clinic (Table 2).

Discussion

Due to the inherent limitations of the case report and case series, result heterogeneity is obvious, with the consequent weakness of the conclusions. Despite these limitations from the data not reported or specified, the following information is relevant. We found that out of 237 patients, 126 were male and 61 were female (the sex of 50 patients was not specified by the author) with a resulting 2:1 ratio, highlighting the predominance among men. The reason why there is a clear predominance in males is unknown; we wonder if it has to do with the diagnosis approach, or if women are underdiagnosed or if there is a more complex pathophysiological explanation.

As for the type of seizures, the most frequently reported were the focal type with or without impaired awareness (Tables 1 and 2).

In both the case series and the case reports, the most common and constant trigger was by definition the act of eating itself²,⁴,¹¹,¹³-¹⁹,²¹-²⁴,²⁶-³¹,³³-³⁸,⁴²-⁴⁶ but in some cases other stimuli were detailed such as chewing²,⁷,¹⁴,¹⁸,²⁴,²⁵,²⁷,³⁴,³⁶, swallowing⁷,²⁵,²⁷,³⁶, drinking⁷,²⁴, seeing¹⁸,³⁷,³⁹,⁴⁵, smelling¹¹,³⁷,³⁹, and even thinking³¹,³⁷ or talking³⁷ about food. Orofacial stimuli², taking food to the mouth⁷, hunger alone³⁷, eating snacks²⁴, and tasting food¹¹ were also described.

In the case report series, other specific triggers to EE related to the type of food have also been described, including bulky meals, conventional Indian meals²⁴, wheat-based diet, and on a rice-based diet³¹,³². Only 4 patients had a preference for a specific type of food (spicy food¹¹, bread¹⁷, strawberry syrup⁴², and minced meat⁴⁵). Furthermore, just a few authors specified the patient's diet, so it remains unclear if EE has a relation to specific diet type.

MRI and CT findings were variable. Some abnormalities were encephalomalacia, sclerosis, pachygyria, polymicrogyria, dysplasia, glioma, ventriculomegaly, meningioma, cavernoma, astrocytoma, and other lesions²,⁴,⁷,¹⁴-¹⁶,²¹,²²,²⁶,²⁹,³¹,³²,³⁴,³⁶,³⁷,⁴³. Four authors reported positron emission tomography; three presented hypometabolism⁷,³³,⁴⁴, and one was normal⁴⁶. Five authors reported single-photon emission computed tomography that showed perfusion changes²⁶,²⁷,³¹,³³,³⁹. Two authors reported functional MRI showing activation of both insula and dorsolateral frontal and parietal cortex⁴² and the other showing left temporal, bilateral parietal, paracentral lobule, cingulate, medial frontal gyrus, and lingual and medial occipital gyrus abnormalities³⁹. EEG findings were highly variable and different in each patient (Tables 1 and 2). Two authors detailed stereoelectroencephalography, finding affectation of the insula, hippocampus, and temporal cortex and/or amigdala⁴²,48]. One patient underwent electrocorticography with operculum activity³⁷. Surgery was performed on 10 patients, consisting of left-temporal lobectomy and amygdalohippocampectomy²¹, right-temporo-occipital resection, left-frontal resection²², lesionectomy³¹, subtotal resection of low grade astrocytoma³⁴, subtotal resection of left-frontal operculum (grade IV glioblastoma)³⁷, anterior insula resection⁴², right-anterior-temporal lobectomy including mesial structures and partial inferior insulectomy⁴⁶, and non-specified in two patients. Only three patients underwent vagus nerve stimulation²².

Furthermore, the presence or absence of pharmacoresistance was specified in 131 patients (55% of the reviewed patients), of which 45 (41%) presented pharmacoresistance although sex, age, or type of seizure predominance was not clear.

Conclusion

EE is not the most frequent type of RE and because it can coexist with other epileptic syndromes, we speculate that EE may be under-diagnosed, so RE should be investigated in all patients with known epilepsy. It is our belief that semiology and specific activities during the eating process should be specified in order to identify the possible physiological-etiological mechanism, which until the present day is unknown. Finally, EE can be a therapeutic challenge since each patient presents variability in age of onset, type of stimuli and seizure, clinical course, findings in complementary studies, and response to management.

Supplementary data

Supplementary data are available at Revista Mexicana de Neurociencia online (www.revmexneurociencia.com). These data are provided by the corresponding author and published online for the benefit of the reader. The contents of supplementary data are the sole responsibility of the authors.