As mentioned in the section 'Introduction', ever since Lugaresi et al. described the episodes of dystonic and dyskinetic movements occurring out of NREM sleep (NPD), whether they were arousal disorders (parasomnias) or epileptic seizures has been controversial. In favor of the former were: first, the intriguing lack of clear EEG discharges between and during attacks; second, the occasional coexistence of the most striking event (NPD) with phenomenologies whose semiology could be considered more akin to parasomnias;1-4'15'49'90 third, the frequent positive family history of various kinds of parasomnias; and fourth, the normal neurological examinations and neuroimaging studies. In favor of the epileptic hypothesis were: first, the occurrence at times of similar diurnal attacks, or of generalized convulsions; second, the often favorable response to CBZ; third, the generally greater severity of the bizarre repertoire of "motor activities," their stereotypy and clustering of short episodes in one night, their occurrence during earlier stages of NREM sleep; and fourth, often some recollection of episodes in the morning. Although prolonged video-scalp EEG monitoring can offer some diagnostic clues, often it may not be possible to reach a diagnosis of epilepsy or of parasomnia. Even most of those who suggested a specific locus for the ictogenesis of NPD collectively agreed that scalp derived EEGs could not reliably be of assistance either inter- or intra-ictally17'28-30'32'42'43'45'76'91-94 Some investigators resorted to functional neuroimaging (SPECT, PET scans) since routine MRIs, as the EEGs, were of no help unless there was a demonstrable and occasional lesion.32,43,44, 68,73,93,94,99 But the results of functional neuroimaging studies also failed to reveal the true epileptogenic zone in the brain since SPECT and PET scans pointed to quite disparate loci, and it is often difficult to know if what they show is the primary zone or is the result of the ictal spread from another region.
Further analysis of available literature thus indicated, on the one hand, a striking disparity of opinions and hypotheses regarding the epileptogenic zone of NPD and, on the other hand, a general paucity of persuasive data supporting them. It is fair to state that there has been scarce electroencephalographic support for any of the suggested zones in the brain that might be the primary locus for the epileptogenesis of NPDs. Papers reporting that a few subjects displayed some EEG abnormalities, as reviewed in the literature, are not persuasive. Sometimes "epileptiform features" were mentioned, but these were not better qualified; these most often were without localizing value, and some actually were normal EEG features related to different stages of NREM sleep, such as K-complexes, for example6,11,16,18,29-32,34,36,37,40,41,46,50,51,55,83,90,93
The most potent boost to drive the pendulum toward the epileptic hypothesis came when neurosurgeons, using intracerebral techniques, recorded from mesial frontal cortex or from the inferior surface of the frontal pole, ictal EEG seizures that could not be detected via scalp recordings. Importantly, some of these electrical discharges were accompanied by symptomatologies that had features similar to those occurring in subjects with NPD attacks.24-27,32,42,43,48,49,95,96,98
Therefore, the issue arose about considering invasive monitoring as the procedure more likely to resolve such diagnostic dilemmas. However, an invasive video-EEG polygraphy is hardly justifiable in patients with sleep-related disorders that do not impinge seriously on their quality of life, like the PA disorders such as night terror, somnambulism or panic attack. But at times the major events, dubbed as NPDs, if not medically controllable, can severely disrupt a patients well-being, and cause cognitive and behavioral problems. These patients should become candidates for intracranial techniques, in conjunction with functional neuroimaging findings. These are now the only available means to reveal the presence of an occult lesion or of an epileptogenic cortical area amenable to surgical ablation.
Such was the situation that arose in the child presented here
His spells of complex, bizarre motor phenomenology arising in early stages of NREM were the hallmark for a tentative diagnosis of the so-called NPD but with otherwise non-contributive video scalp-derived EEG or neuroimaging. When these episodes became pharmacologically uncontrolled and he showed evidence of cognitive decline, it was decided to proceed with an invasive video-EEG polygraphy. The problem then was what areas of his brain should be monitored in the hope of discovering the cortical source of the NPD. The perusal of the literature had shown that there are quite discordant opinions and very few data regarding the epileptogenic locus for such seizures. A majority of the authors had favored the frontal lobe as the generator of NPD attacks, but without specifying in which region of the lobe it might be located.31,34,37,41,45,46,68, 75,99,100,101 Some did focus on specific cortices, mentioning the SMA as the possible site for the epileptogenesis of NPDs but without confirmations by either interictal or ictal electrographic data.17,28-30,40,91 Others instead favored at times the temporal lobe, also on only a circumstantial basis.11,28,46,76,86 A few, using invasive techniques, chose the cingulate gyri as the best candidates for the origin of the NPD,42,93-95 while pre-rolan-dic lesions were thought by others to be its epileptogenic source.32,76 Recently, a ste-reo-encephalographic study on patients with nocturnal hypermotor seizures suggested that these might be generated within the antero-superior insular cortex.43
Thus these few reports using invasive techniques, in spite of unequivocally confirming the epileptic nature of sleep-related hypermotor paroxysms, again failed to agree about the brain's zone responsible for their ictogenesis. Another small number of invasive studies were performed, but these dealt with rather peripheral issues. As mentioned, two of them were directed to establish if the minor events of these nocturnal episodes were also epileptic in nature and these studies reached opposite conclusions.48,49 Others were done to show the outcome of surgical ablations in patients in whom the stereo-EEG was said to provide localizing information. However, 20 of 21 such subjects had focal CD likely to guide where best to locate the recording electrodes.85
What did the invasive study in this child show?
As stated in the results and illustrated in the figures, invasive monitoring of this patient confirmed the epileptic nature of the nocturnal dyskinetic episode (NPD); (2) established that its epileptogenic locus resided in a very limited area of the orbital cortex, corresponding, in this child, to an occult small dysplastic lesion; (3) the ictal discharge in the orbital cortex spread after several seconds (4-7) to the SMA without invading any of the monitored areas in the frontal, parietal, or temporal lobes; (4) the clinical phenomenologies became apparent only when both anterior and posterior SMA were invaded by the discharge; (5) the epileptic phenotypes usually arising from the SMA did not resemble those events captured by the video in this child; and (6) the electrical stimulations induced an ictal discharge only when applied to the two electrodes from where the spontaneous electrical seizures were recorded, and never when other areas of frontal, parietal, or temporal lobes were tested. The ablation of the small lesion, minimally including the surrounding cortex brought full seizure control, now for over 6 years, on no medication for 4 years.
There have been a few previous papers regarding seizures said to arise from fronto-orbital cortex with electrical-clinical correlation. To my knowledge the first was from Tharp who wrote about an "orbito-frontal seizure syndrome" in three children. Clinically there were prominent autonomic symptoms together, at times, with "disorganized motility.' ' While interictally the scalp EEC contained spike discharges, unfortunately an electrical-clinical correlation was vitiated because no electrical or clinical seizures were recorded when a "depth electrode" was inserted into the orbital cortex of one child. Also, the surgical ablation of the structural lesion present in the child largely crossed the limits of the orbital cortex.103 An earlier paper102 mentioned seizures of "probable orbito-frontal origin" but this was based only on finding interictal discharges without obtaining a clinical and electrical seizure. Another paper described some patients with episodic somnambulism who responded to antiepileptic drugs, suggesting an epileptic origin. However, there was again no clinico-electrical correlation.10 Williamson et al.,24 in describing clinical criteria to distinguish "frontal-pole seizures" from the CPS of temporal lobe origin, mentioned that in those originating from the frontal pole the patients frequently displayed motor automatisms and often violent and bizarre behaviors, akin to those exhibited by my patient. They also state that these attacks may be misdiag-nosed as "hysterical" episodes because inter- and intraictal scalp EEC were often misleading. However, clear clinical electrical correlation was again lacking. A somewhat similar phenomenon occurred during the extensive investigations by Munari and Bancaud.103 They found that in patients with uncontrollable epilepsy of frontal lobe origin, when an electrical seizure remained in the orbital cortex without spreading, there were no clinical symptoms. However, they also state that a majority of discharges originating in the orbital cortex spread to other frontal and extra-frontal regions. When discharges reached the SMA cortex for a few seconds there were again no clinical manifestations. However, when SMA or cingulated gyrus were invaded, some patients would engage in "complex gestural automatisms." From such observations Munari wrote "the ictal discharges starting in the orbital cortex" without spreading elsewhere are "asymptomatic." This is what was found in my patient. Namely, the ictal discharges recorded from the more proximal electrodes placed on the orbital cortex were clinically asymptomatic until they invaded both the anterior and the posterior portions of the SMA. At about this time, the bizarre and complex motor phenomenology began (see figures).
However, as mentioned before, the phenotypes of seizures originating from the SMA are generally notably different from the very complex hypermotor seizures occurring in this syndrome. The SMA discharges also tend to invade other frontal lobe areas.29,83,90,91 It is therefore tempting to hypothesize that the discharge originating from orbital cortex, then invading SMA, will not spread from SMA to other frontal areas but rather spread downward to nuclei of the striatum, as has been shown to occur in kinesiogenic dyski-nesias.53 Rich efferent and afferent connections of the SMA with various basal ganglia nuclei have been shown in several studies.104-106 Obviously, such downward spread of ictal discharges from orbital cortex to striatum remains hypothetical.
Was this article helpful?