Migraine, epileptic seizures and psychogenic non-epileptic seizures: Observations in Indian patients in a clinic-based study
The present study focuses on the incidence of epileptic seizures in 1000 patients (200 children and 800 adults) with migraine. Very few patients with migraine had history of epileptic seizures. No patient had migraine-induced seizures and none had seizure-induced migraine like headaches. Occurrence of psychogenic non-epileptic attacks during migraine headaches had not been highlighted in the past. In the present study, about a quarter of patients with migraine, especially adolescent and adult females, had history of psychogenic non-epileptic seizures during attacks of acute migraine. This appears to be the first report on these features in Indian subjects with migraine.
Keywords: Epilepsy, migraine, psychogenic non-epileptic seizures
Migraine and epilepsy are common disorders, and hence, might coexist in the same individual. However, at times, the distinction between these two disorders may appear blurred. This was hinted at by William Gowers over a century ago and migraine was considered to lie in the "borderland" of epilepsy.  The inter-relationship of migraine and epilepsy had been addressed to in recent years by Anderman  and Welch and Douglas Lewis.  However, the nature of the association between these two disorders remains to be determined. The reported frequency of migraine in an epileptic population varies from 8.4 to 23%, whereas the frequency of epilepsy in a migraine population ranges from 1 to 17%.  Recently, Crompton and Berkovic detailed on the clinical and molecular features of the phenomena that mimic epileptic seizures.  Rather uncommonly, migraine may precipitate epileptic seizure (migralepsy) and epileptic seizure may induce post-ictal migraine like headaches responsive to triptans. Parisi et al.  and Parisi  hypothesized on the neuro-physio-pathologic mechanisms linking epilepsy and headache. The International Headache Society proposed a rather strict definition of migraine-induced seizure  (the term of migralepsy has not been used), but this has been criticized. 
The occurrence of psychogenic non-epileptic seizures (PNES) during the course of a migraine attack has not been addressed to adequately in the available literature but had been noted by many practicing clinicians.
The present study focuses on our experience on the incidence of epileptic seizures in a clinic-based population of migraine patients. Furthermore, the occurrence of PNES during acute migraine attacks had also been noted and would be highlighted.
The study was conducted between September 2003 and December 2005 at the Neurology Out-patients' department of a large general hospital in the city of Kolkata in the eastern part of India. A total of consecutive 200 children (aged 7-15 years) and 800 adults (age > 15 years) with migraine (as defined in The International Classification of Headache Dosorders (ICHD II)  were recruited primarily to study pain location at the onset of and during established headaches and the results had been reported earlier. ,, The study questionnaire also included questions relating to other aspects of migraine, particularly triggers  and the occurrence of epileptic seizures in the patient and in their first degree relatives. Occurrence of any form of altered sensorium or motor phenomenon during the course of any migraine attacks was also recorded. In the case of children, the parents were interviewed in detail about these. Subjects with self or family history of epileptic seizures as well as those experiencing spells of altered sensorium during migraine headaches were subsequently interviewed in detail by the corresponding author. Sleep deprived (at least 8 hours) interictal electroencephalography (EEGs), with activation procedures like hyperventilation and photic stimulation, were performed in all such cases.
Diagnosis of epileptic seizure was made purely on clinical grounds as revealed from patients, parents and eye witnesses.
Diagnosis of PNES occurring during the course of migraine headaches was also made essentially on clinical grounds including detailed eye witnesses accounts (specially in relation to duration of spells, sudden or gradual fall, eye closure or opening, presence or absence of stereotyped recognized epileptiform motor phenomenon, frothing, self-injury and incontinence) and great care was taken to exclude fainting spells (warning, pallor, sweating, cold extremities and no motor phenomenon). In 62 subjects, mobile phone video images of patients in the state of unresponsiveness during attacks of migraine could be studied. Interictal EEGs, as detailed above, done in all such subjects revealed no abnormalities. Eight hours video EEG recordings were performed in 38 adult female subjects during episodes of acute migraine attacks, who gave history of spells of unresponsiveness in the past. This included six subjects in whom EEG records were available during the stage of unresponsiveness with migraine headache ("ictal record"). None revealed any abnormal epileptiform activity.
The pediatric migraine group included 200 subjects: 118 males and 82 females (M:F 1:0.69) and age at recruitment varied from 7 to 15 years (mean 11.8 years). Migraine types (based on >50% of attacks) were: 1.1 = 197; 1.2.1 = 3.
The adult migraine group included 800 subjects: 144 males and 656 females (M:F 1:4.56) and age at recruitment varied from 16 to 42 years (mean 26 years). Migraine types (based on >50% of attacks) were: 1.1 = 668; 1.2.1 = 18; 1.6.1 = 114.
The seizure-related observations are shown in [Table 1]. No subjects ever received any treatment for their seizures and none had any seizure in the preceding 2 years. Interictal EEGs in all subjects having had seizures in the past were normal. History of epileptic seizures was uncommon both amongst children and adults with migraine.
Occurrence of psychogenic non-epileptic seizures during migraine attacks
Only those subjects experiencing two or more spells of PNES (on clinical grounds) in the past 2 years were included. All had normal sleep-deprived interictal EEGs with activation procedures. Video EEGs performed in 38 such subjects during attacks of acute migraine revealed no abnormalities.
In children, migraine-related PNES occurred only in 11 subjects (5.5%). All such subjects were females (13.4%) and above 11 years of age. All attacks were of "swoon" type  and the duration varied from 10 to 20 minutes. The spells were characterized by gradual fall, lying motionless and unresponsiveness with eyes closed, teeth clenched and often with clenched fists. The highest number of spells in one individual (a 14-year-old girl) had been 7 in 2 years.
In adults, PNES occurred only in two male subjects (out of 144) but in 152 females (out of 656, i.e., 23.2%) during two or more migraine attacks. These were more common in female subjects above 18 years of age (upper age: 39 years) and one woman aged 23 years experienced at least 20 spells of apparent loss of consciousness in 2 years. The duration of such spells varied from 15 minutes to about 45 minutes. Such spells almost always occurred during bad spells of migraine headaches (≥5 on a pain scale of 7). Attacks were always of the swoon type - no sudden fall, trauma or tongue biting and never associated with any post-ictal confusion or sleep.
The incidence of epilepsy in subjects with migraine, as found in this study, appears quite low. This is much lower than the observations made in the west.  Also, rather surprisingly, this incidence is much lower than the population incidence of epilepsy in India, which is around 0.8%.  There are, however, no Indian studies focusing on the incidence of epileptic seizures amongst migraneurs for comparison. The present study, perhaps for the first time, highlights the occurrence of PNES during acute migraine attacks. PNES occurred in about a quarter of adult female subjects with migraine headaches. The higher frequency of occurrence of PNES in females is not surprising and certainly the frequency increases from adolescent age. All PNES attacks recorded in the present study had been of the swoon type,  and none occurred with any form of motor phenomena. This somewhat is in contrast to our experience with de novo occurrence of psychogenic seizures in Indian females, where many (about a quarter) experienced some form of motor phenomenon. Also, the duration of migraine-related PNES seems to be less than the duration of de novo PNES, which quite often exceeds 1 hour. There are no reports in the literature in relation to occurrence of PNES during migraine attacks. We had been very careful in distinguishing these spells in migraine from fainting spells precipitated by pain. We do not commonly encounter fainting spells during migraine attacks in our practice.
The relationship between migraine and epilepsy is an intriguing one. Is it just a chance co-occurrence or based on the proposed pathophysiologic links? , This question can only be answered through large population-based studies and these need to be done in different geographic locations and amongst people of different ethnicity, as genetic factors are at play in both the conditions. Also intriguing is the high incidence of PNES found in this study during migraine attacks in Indians, a finding not reported earlier. Are Indian migraneures psychodynamically different from their Western counterparts? Again, there is a need for larger population-based studies.
The present study has the following limitations.