The electroencephalogram (EEG) is extraordinarily useful in diagnosing epilepsy and differentiating subtypes, but it is less valuable in the diagnosis of migraine. EEG findings such as posterior slowing and response to hyperventilation, as well as other features such as 14- and 6-Hz positive spikes, are not specific for migraine. Epileptiform discharges and focal slowing occur in the EEGs of patients with and without migraine. EEGs recorded during a migraine with aura, unlike those recorded during a clinical seizure, are usually normal. Focal slowing sometimes occurs during migraine auras, although this is not a consistent finding. Although photic driving beyond a stimulation rate of 20 Hz (the so-called harmonic response) was thought to be characteristic of migraine, it can be seen in children without a history of migraine and is not very specific.81
The Quality Standards Subcommittee of the American Academy of Neurology has reviewed the usefulness of the EEG in headache. They found that no study has consistently demonstrated that the EEG improves diagnostic accuracy for the patient who experiences headache. The EEG has not been convincingly shown to identify headache subtypes, nor has it been shown to be an effective screening tool for structural causes of headache. They concluded that the EEG is not useful in the routine evaluation of patients with headache. This does not exclude the use of EEG to evaluate headache patients with associated symptoms suggesting a seizure disorder, such as atypical migrainous aura or episodic loss of consciousness. Assuming head imaging capabilities are readily available, EEG is not recommended to exclude a structural cause for headache.82
The incidences of spikes and paroxysmal rhythmic events (PREs) in 10-hour overnight EEGs of normal adult volunteers (n = 135) were studied at 11 sites with a computer-assisted ambulatory EEG monitoring system with automatic spike and PRE detection. Spikes were evident in the overnight EEG of one subject (0.7%), and PREs were apparent in the overnight EEG of the same subject (0.7%). Among 24 other subjects with a history of migraine, a family history of epilepsy, or both, the overnight EEGs of 12.5% of those with migraine and 13.3% of those with a family history of epilepsy showed spikes, significantly more than for the others.83
The EEG and 24-hour closed-circuit television EEG recording can help differentiate migraine aura and epileptic aura. These procedures can also facilitate the diagnosis of comorbid epilepsy and migraine, as well as the migralepsy syndrome. Marks and Ehrenberg73 studied patients with migralepsy, using multiple 24-hour video-EEG telemetry recordings. In two patients, the entire migraine-epilepsy sequence was captured, showing changes during the clinical migraine aura that were atypical for electrographic epilepsy. During migraine aura, bursts of spike activity may resemble the ictal EEG during an epileptic seizure. In most reported cases, however, the EEG does not show the usual temporal evolution with progressive increases and declines in the frequency and amplitude of rhythmic, repetitive epileptiform activity typical of ictal EEGs in epilepsy. In addition, the EEG during migraine aura may show “waxing and waning” patterns, separated by completely normal EEG activity, despite the persistence of clinical symptoms.
Manzoni et al.84 and Terzano et al.85,86 coined the term intercalated seizures to denote epileptic seizures occurring between the migrainous aura and the headache phase of migraine. They found that of 450 patients with migraine, 16 (3.6%) also had seizures. The two conditions appeared to be coincidental in 4 of the 16 patients. In another 5 patients, the two types of attacks were quite distinct, but often an epileptic seizure was followed by a migraine attack and vice versa. The remaining 7 patients had intercalated seizures. All had a family history of migraine, and 2 had relatives with epilepsy. They all had visual seizures consisting of highly stylized contours of plain figures or single or multicolored spots that often rotated. The seizures lasted for 1 to 2 minutes and came out of a scintillating scotoma, slowly developing in the visual field and evolving into unilateral or bilateral hemianopia.
DeRomanis et al.87,88 studied patients who had brief ictal visual hallucinations of colored dots or discs and interictal occipital paroxysms on EEG. EEG during a seizure showed that they had occipital epilepsy and not migraine with aura.50
Striking EEG patterns have been described in specific subtypes of migraine.89 The brain regions most often involved in the published EEG samples in basilar-type migraine include the posterior temporal, parietal, and occipital regions. The posterior electrographic localization may not pertain to other forms of migraine.90 Paroxysmal lateralizing epileptiform discharges (PLEDs) or PLED-like activity has been associated with hemiplegic migraine, prolonged migraine aura, or incipient migrainous infarction. Those patients with PLED-like activity did not have any of the usual entities associated with PLEDs, such as stroke, brain abscess, glioblastoma, or viral encephalitis, and their PLEDs usually resolved within 24 hours. Certain migralepsy patients had clinical seizures when PLEDs were present on their EEGs.6
Reviewed and revised April 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.
© 2014 Epilepsy.com. All rights reserved.