Epikpsia, 39(Suppl. 4):S49-S61. 1998 Lippincon-Raven Publishers, Philadelphia 0 International League Against

Mesial Frontal Epilepsy

Norman K. So

Oregon Comprehensive Epilepsy Program, Legacy Portland Hospitals, Portland, Oregon, U.S.A.

Summary: The mesiofrontal cortex comprises a number of occur. The task of localization of the epileptogenic zone can be distinct anatomic and functional areas. Structural lesions and challenging, whether EEG or imaging methods are used. Suc- cortical dysgenesis are recognized causes of mesial frontal epi- cessful localization can lead to a rewarding outcome after epi- lepsy, but a specific gene defect may also be important, as seen lepsy surgery, particularly in those with an imaged lesion. in some forms of familial epilepsy. The predomi- Key Words: Mesial frontal epilepsy-cingulate - nant manifestations, which are not necessarily strictly --Hypermotor correlated with a specific ictal onset zone, are absence, hyper- seizure-Postural tonic seizure-. motor, and postural tonic . Other also

FUNCTIONAL ANATOMY convolution. Traditional cytoarchitectonics have further subdivided this anterior frontal region. The frontal pole The frontal lobe is the largest lobe in the brain, ac- refers to the anterior most portion of the frontal lobe, but counting for one-third to one-half of total brain volume there is little consensus on how far back this extends. and weight. On the medial surface, the most important One or more curved.sulci are seen anterior and inferior to landmark is the cingulate (Fig. 1). This runs as an the , called the superior and inferior ros- inverted “C” following the contour of the corpus callo- tral sulci. The most inferior band of medial frontal cortex sum. Starting at the region opposite the lamina terminalis is contiguous with the orbital surface of the frontal lobe below the rostrum of the is the subcal- and the various , and can be regarded as part losal area. This merges with the anteriormost extent of of the . the cingulate gyrus, the long band of cortex lying be- The functional areas on the medial frontal surface that tween the callosal and cingulate sulci. The cingulate gy- have been best studied and defined are the anterior cin- NS curves posteriorly beyond the confines of the frontal lobe with its posterior portion in the , and gulate cortex and the supplementary motor area (SMA). continues around the splenium of the corpus callosum to The anterior comprises cytoarchitec- become continuous with the . As tonic area 24, and functionally includes area 32, which is the cingulate sulcus is followed posteriorly, it turns up- immediately dorsal, and 25 (1,2). It has wards into the marginal ramus and reaches the superior the pattern of frontal agranular cortex with prominent convexity. The indents the superior surface layer V . Afferent connections are predominant of the brain anterior to the marginal ramus, thus defining from the anterior, dorsomedial, midline, and intralaminar the pre- and postcentral parts of the . thalamic nuclei, and efferent projections are widespread The paracentral lobule is limited anteriorly by the para- to the primary , dorsal and ventral striatum, central sulcus, a short ascending sulcus arising from the the , , and periaqueductal gray in mid- cingulate sulcus. , brain tegmentum. Stimulation studies evoke autonomic, The situated in front of the para- visceromotor, or affective sensations, and simple to com- central lobule and beyond the cingulate gyrus is exten- plex movement patterns (3). Motor functions found in sive and is marked by many irregular gyrations. It can the midportion of cingulate gym probably overlap with also be considered the medial surface of the first frontal those of the supplementary motor area. Lesions of the anterior cingulate cortex produce affective and motor neglect, clinically apparent as apathy and akinetic mut- Address correspondence and reprint requests to Dr. N. K. So at Neu- rological Clinic, 1040 NW 22nd Street, Suite 420, Portland, OR 97210, ism, and can lead to relief of chronic pain and obsessive U.S.A. compulsions (1). The posterior cingulate cortex has dif-

s49 S50 N. K. SO

A

FIG. 1. A: Medial surface of the . B: Cyto- architectonic areas on the medial surface of the human cerebral hemisphere after Brodmann (1909). (Reproduced with permis- sion from Gray’s Anatomy 36th ed., Williams PL and Warwick R, Churchill Livingstone, 1980.)

ferent cytoarchitectonics and connectivity and appears to limit of this functional area is variable and may overlap participate more in visuospatial and memory functions. with and extend into the cingulate gym. Superiorly, it (4). can extend to the convexity of the superior frontal gyms. The supplementary motor (and sensory) area, or sup- Afferent connections derive from thalamic nuclei ven- plementary sensorimotor area (SSMA), is best viewed as tralis anterior and lateralis and from the dorsomedial a functional rather than a strictly anatomically defined nucleus. Projections are found to the putamen, thalamus, region. It is situated around and contiguous with the pri- pontine nuclei, and the spinal cord (5,9). There are also mary motor area for the lower limbs that occupies the extensive reciprocal connections to many cortical areas. precentral part of the paracentral lobule, and extends Cortical stimulation produces proximal, segmental, uni- about 4-5 cm along the medial frontal gym. It includes lateral or bilateral, tonic postural movements as well as a all or most of the Brodmann medial area 6, and possibly range of somatic sensations. Lesions of the SMA can parts of medial area 8 (5,6). A somatotopic organization cause acute mutism, particularly in the speech dominant has now been established in the human (7,8) as in the hemisphere, and contralateral or bilateral weakness that primate brain, in confirmation of Penfield’s model frequently resolve (see below). (Fig. 2). Most anteriorly is a , The remaining anterior mesial frontal cortex com- probably in the medial portion of area 8. Motor repre- prises Brodmann’s areas 9 to 12 and is considered to be sentations for the upper limbs precede those for the lower a part of the . There are extensive affer- limbs. The supplementary sensory areas are most poste- ent and reciprocal connections to the dorsomedial tha- riorly situated in the medioparietal cortex. The inferior lamic nucleus and efferent projections to other cortical MESIAL FRONTAL EPILEPSY S51

series), followed by gliosis (33%, 33%), and cortical dys- genesis (7.5%, 15.5%). The frequency of tumors in fron- tal lobe epilepsy treated surgically is probably a good deal higher than the figures above. In a group of 18 patients with SMA epilepsy treated surgically, seven had a tumor, five had evidence of cortical dysgenesis, and three exhibited changes of encephalomalacia (14). Head trauma much more commonly damages the anterior polar or orbitofrontal regions of the frontal lobe rather than its medial surfaces. As cerebral imaging techniques ad- vance, small areas of cortical dysgenesis are increasingly MOTOR recognized in patients with partial epilepsy (15). Until the recent definition of familial forms of tempo- ral and frontal lobe with autosomal dominant patterns of inheritance, it has been assumed that partial epilepsies persisting into adulthood are usually symp- tomatic in nature. Although well recognized, the benign partial epilepsies of childhood are conceived of as idio- pathic localization-related unique to childhood, which run a benign course with frequent re- mission later in life. A benign of childhood had been proposed (16), akin to benign child- SEN SO RV \( hood epilepsy with centrotemporal spikes. More re- FIG. 2. Somatotopic’organization of the primary and supplemen- cently, children with seizure characteristics similar to tary motor and sensory areas according to Penfield and Jasper (47). (Reproduced from ref. 47 with permission.) those of SMA epilepsy (17), or of suspected medial fron- tal origin (18) have been described. The affected children exhibited a relatively benign clinical course, and a posi- areas, the basal ganglia, hypothalamus, and periaqueduc- tive family history of epilepsy was common. They very tal gray in the midbrain tegmentum. likely represent the early presentation of autosomal dom- All areas of the mesial frontal cortex have extensive inant frontal lobe epilepsy (19,20). In affected families, connections with the opposite hemisphere via callosal clinical seizure semiology is stereotyped, with features to commissural fibers and with other cortical areas via ar- be discussed below, that are strongly suggestive of sei- cuate fibers. The SMA have relatively dense callosal zures of mesial frontal origin. Scalp ictal EEGs have connections compared to the . been difficult to localize or are obscured by artifacts. Single shock stimulation studies (10) evoked shorter Interictal PET and SPECT studies confiied regional transcallosal latencies on the order of 10 ms between hypometabolism or hypoperfusion, while ictal SPECT homologous SMA sites compared with longer latencies showed regional hyperperfusion involving mesial frontal of 30-40 ms between prefrontal sites. cortex, with varying degrees of involvement of fronto- polar and parasaggital regions (21). Onset is usually be- PATHOGENESIS fore the age of 20, and nocturnal seizures predominate. The gene for autosomal dominant nocturnal frontal lobe Knowledge of the pathologic substrate of frontal lobe epilepsy in a large Australian kindred has been reported epilepsy is strongly biased toward material obtained at to be mapped to chromosome 20q, with the gene defect epilepsy surgery and toward imaged lesions. Of 100 pa- found to be a missense mutation at codon 248 in the gene tients with frontal lobe epilepsy treated surgically at the for the 1x4-subunit of the neuronal nicotinic acetylcholine Hdpital St. Anne, Paris (1 l), only 7% had a , receptor (22). However, not all families with this syn- and 15% each were attributed to , head drome share the same genetic defect. trauma, and diverse causes, 18% to neonatal anoxia, while 37% were of unknown origins. In the nontumoral CLINICAL MANIFESTATIONS Montreal series of 40 patients with ‘‘pure” frontal epi- lepsy who became seizure-free after surgery (12), and in There are several ways in which a set of seizure phe- a later compilation of up to 187 surgical specimens taken nomena can be recognized to correlate with a site of from the frontal lobe at the time of epilepsy surgery (13), seizure onset, and with the mesiofrontal cortex in par- the most common pathologic diagnosis was that of me- ticular. One “gold standard’’ has been to analyze seizure ningocerebral cicatrix (50%, 33% respectively in the two semiology in patients who have become completely sei- S52 N. K. SO zure free after a relatively restricted cortical resection mation of the epileptogenic zone, one can test whether (12,23). Another is to use localizational information of certain groupings or sequences of ictal symptoms and ictal EEG onset obtained by intracranial electrodes (14, signs correlate with sites of seizure onset. One such 24-28). When localization is based on scalp EEG record- method is cluster analysis, initially employed for seizures ings (29), the information obtained, as will be discussed, of origin (36,37), and, more recently, in is far too variable to be trusted. A third method is to use frontal lobe epilepsy (3,38). One problem in this type of the location of abnormal structural or functional imaging, analysis, is that a predetermined set of symptoms and provided that the abnormality is discrete and not diffuse signs have to be recognized, hopefully appropriate to the (30,31). The latter two approaches are subject to quali- seizures under study. Second, the analysis will be unsuc- fications to avoid mistakes. The accuracy of the ictal cessful in recognizing a correlation unless a sufficient EEG localization, particularly when obtained with intra- number of examples from different sites are available in cranial electrodes, is influenced by the strategy of elec- the sample. trode placement, and skill of EEG interpretation. Imaged With these provisos, it appears that mesial frontal epi- lesions are correlated with the site of the epileptogenic lepsy most commonly finds expression in one or more of zone by force of experience, despite the lesion being in three clinical seizure types: absence, hypermotor, and itself electrically inert, except for those in cortical dys- postural tonic seizures. Other seizure types, such as ver- genesis. Small, solitary discrete lesions such as a low- sive or psychomotor seizures, are not precluded but oc- grade glial or ganglionic tumor, or an arteriovenous mal- cur less frequently. All seizure types can progress to formation (AVM), are most reliably correlated with the secondarily generalized tonic-clonic seizures. Further- ictal epileptogenic zone. When the abnormality is exten- more, the different seizure types may sometimes blend sive and involves several lobes or distorts normal into each other or may coexist in the same patient. These anatomy, it becomes difficult to identify a lobe or region observations support the notion that more than one sei- of seizure onset without independent EEG or clinical zure type can arise from a given brain region. It is un- data. Furthermore, the underlying pathology may be certain whether this is because of our still rudimentary multifocal or more diffuse than is apparent even with the ability to precisely localize the epileptogenic zone and best imaging technique, as often occurs in cortical dys- correlate it with its functional anatomy or because sei- genesis or in some individuals with cavernous hemangi- zure expression is indeed a function of one of a number oma. of potential pathways of propagation from the site of Some seizures of frontal origin appear clinically inert seizure origin. This point of view is somewhat different when the ictal discharge remains confined to the frontal from the anatomically constrained model proposed by lobe proper (32,33). On the other hand, typical ictal be- the Commission on Classification and Terminology (39), haviors can be observed even when the ictal discharge is in which each brain region is correlated with its own strictly confined to the frontal lobe in short seizures seizure type. Rather, it is proposed that a number of (26,34). There are purists who argue that it is only proper seizure types may be found in frontal lobe epilepsy, as in to talk of seizure semiology as pertaining to a certain any other localization-relatedepilepsy, based on the pref- region or lobe, when the seizure discharge, as defined by erential activation of one or more systems for seizure intracranial EEG, remain strictly within that stated struc- expression, a sentiment shared by others (35,40). ture. Although logically correct, this approach has sev- eral practical limitations because ictal propagation is Absence seizures . more often the rule than not. Even if different parts of the Absence seizures may be the least common seizure brain were recruited by the seizure discharge, it does not type of frontal lobe origin. Seizures are characterized by preclude one part or system alone as the primary gen- speech and behavioral arrest, staring, reduced to com- erator of the abnormal symptoms and signs, since not all plete loss of , (sometimes with minor head parts of the brain necessarily “speak with equal loud- and eye turning), and simple automatisms. There do not ness” during a seizure, and some can remain “silent.” appear to be a preceding . Absences can be as short This logic also fails to take into account that a set of as 10 s, typical of absence seizures in the idiopathic symptoms and signs, even if a result of propagation, generalized epilepsies, but may continue for 30 s or more provided that they are organized and not random or in- and may have a longer postictal period before complete finitely variable, can still lead us back to a site or a small recovery. Seizures may progress further into generalized range of sites of seizure origin. In fact, it should be , often with complex or asymmetric motor recognized that in general there are several preferred features. Frontal absences have been described with ictal patterns of seizure propagation from each of the main onset sites in the medial frontal gyrus (28), the anterior epileptogenic zones (35). cingulate gyrus (24), and orbitofrontal cortex (40,41). Finally, with the knowledge gleaned by the post fact0 Frontal absences have not been seen in patients with analysis of seizure semiology based on anatomic infor- seizures from the SMA, and seem to originate from the MESIAL FRONTAL EPILEPSY s53 more anterior half of the mesial frontal cortex. One ies, but the epileptogenic zone can also involve or extend would hypothesize the areas maximally involved in the to the adjacent paracentral lobule, the cingulate ictal discharge to have dense callosal connections, or gyrus, and the first frontal convolution dorsally projections to the thalamus, to account for the abrupt (14,27,30,31,33,46-49).An aura is present in about half alteration of awareness and bilateral ictal EEG findings. of the patients, usually somatosensory in nature, al- though rarely is it localized as in seizures from the pa- Hypermotor seizures rietal lobe. A whole body segment or more is involved in It must be admitted there is as yet no uniform accep- the aura, unilaterally or bilaterally. Sensations other than tance of this term proposed by the Cleveland Clinic tingling or numbness are also described, such as tight- group. Complex motor seizures would also be appropri- ness, “floating away,” or a sensation of movement. ate, except that the term as used originally by Bancaud When lateralized, the somatosensory aura is contralateral and colleagues referred to secondarily generalized sei- to the hemisphere of seizure onset. The initial motor sign zures preceded by head and eye deviation, and unilateral isusually that of bilateral but asymmetric tonic posturing posturing of the limbs (28). This seizure type is above all of the extremities or if initially unilateral, the changes characterized by motor agitation charged with an emo- rapidly become bilateral. There may be an associated tional quality (31). A variable number of patients expe- facial grimace or, as more appropriately described by rience an aura at onset. The range of symptoms is very Ajmone Marsan and Ralston (35), bilateral facial con- wide: viscerosensory symptoms such as chest, abdomi- traction producing a wide-eyed grin. The arms elevate nal, or throat sensations, breathlessness, and palpitations; and abduct but do not clear the head in every case. When olfactory sensations probably via early seizure spread they do, and one flexes while the other extends, one from orbitofrontal cortex to the of the observes the classic fencing posture. The legs are in- temporal lobe (33), ill-localized somatosensory feelings volved and assume postures of flexion or extension. Vo- of tingling, numbness, tightness, and heat; nonspecific calization occurs in about a third of cases, with a forced lightheaded and cephalic sensations; experiential visual repetitive or affective quality. Truncal, pelvic, and bipe- or auditory illusions and hallucinations; and emotional dal movements develop after the initial tonic phase in feelings including fear. Motor signs consist of early, about a quarter of instances. Seizures often terminate at complex large-amplitude movements that have been de- this stage, and hence seizures are often relatively short, scribed variously as repetitive, stereotyped, persevera- in the range of 20-30 s. Focal clonic activity, if seen, tive, sexual (pelvic thrusting), bimanual-bipedal (thrash- develops later, usually in the course of secondary gen- ing, bicycling), often bizarre and charged with an emo- eralization, and can provide a reliable lateralizing sign. tional, aggressive, or distressing quality. The behaviors The same applies to head version, if it occurs at all. are often accompanied by forced laughter, crying, whin- Manual automatisms are infrequent, fragmentary, and ing, and vocalizations or verbalizations, which can some- late in occurrence. Awareness and memory are retained times be obscene. Urinary incontinence may occur. Con- in very brief seizures and become progressively impaired sciousness is variably impaired, usually in relation to the as the seizure intensifies. Nocturnal predominance of the intensity and duration of the seizure. Although many seizures is again common. seizures are relatively brief (10-20 s), progression to a secondary is possible. Many patients have nocturnal seizures, sometimes exclusively. Hypermotor DIAGNOSTIC CONSIDERATIONS seizures are described to originate from the medial fron- tal gyrus (26,42), the anterior cingulate cortex (1,28), and Scalp EEG orbitofrontal and frontal polar regions (26,30,31,41,43, Findings are somewhat variable reflecting the large 44). Studies of intracranial ictal EEG during frontal lobe anatomic temtory and the medial interhemispheric loca- seizures with bimanual-bipedal automatisms showed tion of electrical generators. The medial frontal cortex is that lateral and mediofrontal structures were the most well recognized as a site for the generation of “second- frequently involved, with spread to temporal structures ary bilateral synchrony” (50). The reverse does not hold only in a minority (45). When bimanual-bipedal automa- true, because bilateral discharges can arise not only from tisms occur during complex partial seizures of temporal frontal parasagittal sites but also from medial parietal lobe origin they were frequently associated with spread and temporal locations. ‘It must be remembered that sec- of the ictal discharge to orbitofrontal and lateral frontal ondary bilateral synchrony is a hypothetical mechanism regions. and that the term encompasses a variety of EEG phe- Postural tonic seizures nomena (50). Interictal sharp waves from medial inter- These seizures typically arise from the posterior half hemispheric or high parasagittal generators are often not of the medial frontal gyrus, corresponding to the SMA bilateral generalized discharges but are much more likely when functionally defined by electrical stimulation stud- to represent midline discharges (Fig. 3). Transverse mon- s54 N. K. SO

FP1-- FP1--

FIG. 3. Midline interictal sharp waves with Cz maximum in a patient with supplementary motor area epilepsy (patient 9, Table 1). tages and referential mapping of the potential field, in- Epileptogenic zones at the frontal pole involving me- cluding midline electrodes, usually make this clear. A dial or orbitofrontal surfaces may exhibit localized or parasagittal to midline potential localization is quite typi- bilaterally symmetric or asymmetric frontal interictal cal for patients with SMA epilepsy (27,49) and is seen in sharp waves (30,41,44,51). Sometimes only slow waves a majority of patients. NREM sleep often activates these are present. Detection may be aided by the placement of discharges. Midline rhythmic slow waves, more com- additional electrodes according to the 10-10 system or at monly in the 8 than in the ti range, are also not uncom- supraorbital and infraorbital sites (44,52). Nasoethmoi- mon. Ictal EEG recordings in SMA epilepsy are often dal electrodes have also had some success in localizing obscured by muscle artifact. However, a burst of vertex discharges from the orbitofrontal cortex (53,54), but suf- maximal sharp waves, or diffuse EEG attenuation can fer from the disadvantages of requiring a special ENT often be seen at onset before EMG artifacts make inter- procedure and of the lack of tolerability for long-term pretation impossible (49). In longer seizures, rhythmic monitoring. However, the orbitofrontal cortex, anterior ictal activity may be glimpsed. Postictal depression and medial frontal gyrus, and cingulate gyrus are epilepto- EEG slowing are functions of seizure intensity, clearly genic zones notorious for falsely localized interictal dis- evident after secondarily generalized convulsions, but charges, commonly over one or both temporal regions may be entirely absent after brief postural tonic seizures (26,31,41). Ictal scalp EEG recordings of seizures arising in SMA epilepsy. Both interictal and ictal EEG findings from these regions appear to be highly variable, and are often make lateralization difficult if not impossible. often nonlocalizing. Sometimes the voltage potential field is clearly asym- The scalp EEG findings of epileptic foci in the inter- metric. The side of higher voltage more commonly than mediate portion of the mediofrontal gyrus may exhibit not is on the side of the epileptogenic zone, but the features of the type seen in SMA epilepsy or of those reverse has been seen. The scalp topographic field is seen in orbitofrontal epilepsy. A clear picture is lacking. probably influenced by the degree to which the dorsal One category of seizures and its EEG characteristics de- convexity of the first frontal convolution is recruited in serves separate mention. It is sometimes seen in patients electrical generation. When it is recruited, voltage maxi- who appear to have frontal absence seizures. The ictal mum is on the side of the generator. A deeper dipole and interictal EEGs in some patients are striking for generator in the medial interhemispheric surface, on the regular, rhythmic 2- to 3-Hz, generalized spike-and- other hand, may “paradoxically” project the voltage wave, or multiple spike-and-wave complexes, maximal field to the contralateral hemisphere. over the frontal regions, symmetrically or with voltage MESIAL FRONTAL EPILEPSY s55 predominance to one side (Fig. 4). Clues to secondary for intracerebral hemorrhage. By contrast, subdural strips bilateral synchrony are sometimes found in independent are associated with low risk, but it can be difficult to or leading spikes, persistent amplitude asymmetry, or ensure they are placed where they were intended to go. localized slow waves. These frontal absences have been Subdural grids are most helpful when detailed functional shown to arise from the medial frontal region and from and EEG information are needed for resections antici- the cingulate gyrus (24,3135). Techniques for the analy- pated to be close to essential cortical areas. sis for small time differences (56,57) may help to reveal A few points on the ictal EEG, as recorded by intra- a leading generator when applied to the scalp EEG. Other cranial electrodes, will be mentioned. At onset, the ictal patients with frontal absence seizures may have more discharge frequency is often extremely fast (20-100 Hz) localized ictal EEG findings or no scalp EEG changes. and of low amplitude (58,59). It can be mistaken for noise and artifact from cable movement and from the Intracranial EEG environment. An injudicious use of high-frequency fil- Intracranial EEG is frequently called for in the presur- ters will obliterate this signature. Ictal propagation to gical evaluation of mesial frontal epilepsy except when a contralateral homologous structures (24,60) and to the discrete structural abnormality is present that is entirely dorsal convexity of the is rapid consistent with other data. Intracranial EEG is used to (61). When symmetric depth or subdural electrodes have localize the epileptogenic zone among a number of pos- been placed, onsets can appear disconcertingly bilateral sibilities, to lateralize the site of seizure onset even when and simultaneous (1 0,24,60). High temporal resolution approximate lobar location is known, and to integrate analysis for small time differences can help to reveal a functional and EEG information to guide the plan of leading site but is possible only with fast analogue-to- resection. Depth and subdural grid and strip electrodes digital sampling well beyond the conventional 200-Hz each have their advantages and disadvantages, and their sampling rate of most commercial systems. Frontal ab- appropriate selection should be made with the particular sences characterized by generalized spike-and-wave situation and questions in mind for the individual case complexes on the scalp EEG unfortunately tend to ap- (49). For exploration of the medial frontal region, ste- pear identical when explored by intracranial electrodes. reotactically implanted depth electrodes provide the In these cases, an interictal “hot spot” may provide the greatest anatomic accuracy but also have the highest risk only clue to the location of the epileptogenic zone.

FIG. 4. lctal scalp EEG showing bilateral mul- tiple spike-and-wave complexes of a frontal ab- sence seizure in a patient with an anterior cin- gulate lesion (patient 11, Table 1).

Epilepsia, Vol. 39, Suppl. 4. 1998 S56 N. K. SO

Brain imaging localized interictal hypometabolism of 10 to 25% in vari- MRI can now reliably detect all tumors and AVMs ous subregions. There have also been reports of reduced and an increasing proportion of cortical dysgenetic le- benzodiazepine receptor binding on flumazenil-labeled sions (Fig. 5). With the continuing improvements in MR PET images of patients with frontal lobe epilepsy (66). imaging, it is difficult to estimate the sensitivity rate in Ictal or immediate postictal SPECT imaging has been lesion detection from case series just a few years old. successful in several reports of patients with orbitofron- Series from surgical centers are likely to be biased in tal, medial frontal, or SMA epilepsy (30,67). The diffi- favor of structural lesions because their presence helps in culty in timely injection of the radioisotopic ligand in the surgical selection process. Recent MRI improve- seizures that are often abrupt in onset and short in dura- ments include extremely high-resolution, 1-mm thick- tion must be considerable. Not all centers have been able ness contiguous slices, whole-brain SPGR acquisition, to replicate such a high rate of success. and FLAIR (fluid attenuated inversion recovery) se- The role of MR spectroscopy in frontal lobe epilepsy quences (62). Surface rendering to study gyral patterns remains to be defined. In one report of eight patients with can detect subtle areas of cortical dysgenesis (15), al- frontal lobe epilepsy (68), phosphorous MRS revealed though this technique has been mainly applied to study interictal alkalosis in the epileptogenic lobe in all pa- of the cortical convexity rather than the fissures. tients, and decreased phosphomonoester levels in 7. Earlier PET studies had occasionally found interictal glucose hypometabolism in patients with suspected fron- OUTCOMES AND OPERATIVE TREATMENT tal lobe epilepsy. More recent studies reported improved sensitivity of detection by quantitative techniques of Given the relatively small number of patients with analysis using regional templates normalized to controls confirmed mesiofrontal epilepsy and that the confirmed (63-65). Quantitative studies were able to demonstrate cases were almost all selected for surgical treatment, it is

FIG. 5. Preoperative MRI demonstrating an area of focal cortical dysplasia in the right anterior medial frontal gyms (A), and post- operative appearance of the resection (6)(patient 1, Table 1). MESIAL FRONTAL EPILEPSY s57 difficult to make broad generalizations on the course of zure-free or had only rare attacks (73). In 89 patients condition. Etiologic factors are probably more important with some of the features of cingulate epilepsy and who than the electroclinical localization in determining long- underwent a frontal resection, 39 (44%) obtained a suc- term outcome. Just as familial cessful result with less than one seizure a year (24). (69,70) may present a relatively “benign” picture com- Failures were most obvious in the 21 patients who only pared with that of mesial temporal sclerosis or temporal had excision of the convexity of the frontal lobe with lobe lesions, so it appears that autosomal dominant noc- only one obtaining a good result. Of another 31 patients turnal frontal lobe epilepsy may also show a more favor- at the HBpital St. Anne (11) who had medial frontal able course (18-20). There appears to be considerable cortisectomies, the greatest improvement occurred in the variability in severity, but at least a third of patients group that had a posterior resection (80%),with the least achieved satisfactory control with (19). success after mid-frontal(l8%), and intermediate results The clinical, EEG, PET and SPECT features of autoso- after anterior frontal resections (50%). mal nocturnal frontal epilepsy strongly suggest a mesio- At the Oregon Comprehensive Epilepsy Program, 32 frontal origin, either from the SMA or from more ante- patients underwent a frontal resective procedure (31 op rior medial frontal regions. Only intracranial EEG local- erated locally, with one patient operated elsewhere after ization has yet to be attained. presurgical evaluation) between 1982 and 1997. Eleven, A number of reports highlighted the associated psy- of the resections were primarily mesial frontal. Localiza- chopathology in frontal lobe epilepsy. Of 100 patients tion of a mesial frontal epileptogenic zone was obtained who underwent a frontal resection at the HBpital St. by subdural strips or grids or both in seven patients and Anne (1 I), 38% had problems of personality and behav- was inferred from an imaged lesion in the other four. ior, with 10% rated as severe. In 36 patients with cingu- Altogether, seven patients had an imaged abnormality: late epilepsy who achieved a good surgical outcome, three tumors, one each with cortical dysgenesis, AVM, psychopathology was noted in 72%, with 55% rated as scar from cerebral abscess, and lesion of indeterminate being in an “almost permanent‘major psychotic” state pathology. The resection was predominantly anterior (24). Bursts of aggression, hostility, and agitation were (anterior half of the medial frontal gyms, with or without notable. Other psychopathic abnormalities associated removal of cingulate or orbitofrontal cortex) in eight pa- with cingulate epilepsy include obsessive-compulsive tients (Fig. 5), and predominantly posterior (posterior behaviors and sexual preoccupation (1.71). Many pa- half of the medial frontal and sometimes cingulate gyri), tients with hypermotor seizures or postural tonic seizures including the SMA, in two (Fig. 6). The adjacent first of medial frontal onset, have been misdiagnosed as hav- frontal convolution was usually also removed in the pro- ing pseudoseizures (26,72). Eight of Williamson’s 10 cess. One patient had a cingulate lesionectomy. Six of patients were believed to be hysterical before epilepsy the 11 patients (55%) obtained a good outcome, with five was diagnosed. Unlike the patients with cingulate epi- patients entirely seizure-free at l-year follow-up. An- lepsy, those with SMA epilepsy do not appear to have a high rate of psychopathology. Pharmacologic treatment follows the same principles as for other localization-related epilepsies. There is noth- ing in accumulated experience to suggest that frontal lobe epilepsies or the different seizure types of mesial frontal origin show a selective response to one or more AEDs. Except in patients with a well-defined discrete struc- tural lesion distant from the rolandic cortex, intracranial EEG investigation is a necessary prelude to the surgical treatment of mesial frontal epilepsy. Localization of es- sential functions is a critical part of the surgical planning process. Frontal cprticectomies vary greatly in size and location and may be combined with partial section of the corpus callosum, undercutting of subcortical fiber tracts, and multiple subpial transection. The overall seizure out- comes after medial frontal resections are probably com- parable to those after extratemporal neocortical resec- tions a whole. In the large series of 283 patients with as FIG. 6. Postoperative MRI demonstrating resection’of the right nontumoral frontal lobe epilepsy who underwent surgery supplementary motor area with seizure-free outcome (patient 9, at the Montreal Neurological Institute, 36% became sei- Table 1).

Epilcpsin, Vol. 39, Suppl. 4. 1998 S58 N. K. SO other patient only had nondisabling focal motor seizures often anticipates a less than perfect outcome. In addition postoperatively (Table 1). to the completeness of resection, other factors reported to There are several more series of patients with surgi- influence seizure outcome after operation on the frontal cally treated mesial frontal epilepsy. All five patients in lobe (1 1) include the presence of a localized lesion which Williamson's 1985 report who underwent a frontal re- improved surgical success, although too-extensive cere- section for hypermotor seizures arising from mesial fron- bral damage from head trauma diminished the results. tal or orbitofrontal regions became seizure-free or had Frontal resections that included subcortical fibers of the only rare breakthrough seizures. Half of a group of eight corona radiata were felt to have a better seizure outcome patients who underwent limited resection of the mesial than when these same fiber tracts were preserved (1 1). part of one frontal lobe became seizure free or almost Some patients who had poorly lateralized EEG findings seizure-free (34): This is mirrored by another 12 patients and who were nevertheless suspected to have medial who underwent mesial frontal resections, half of whom frontal epilepsy have been treated with corpus callos- were seizure-free or had only rare seizures (74). In the otomy, but the results have been quite variable and dif- Cleveland Clinic series (60) of 20 patients who under- ficult to interpret (39). went corticectomies for SMA epilepsy guided by the Surgical complications and deficits have thus far not results of subdural grid investigation, 10 (50%) were proved to be a limiting factor for operations on the me- seizure-free (four) or had rare seizures (three), or had diofrontal region except at its posterior limit, where re- only residual nondisabling simple partial seizures (three). moval of primary motor areas would lead to unaccept- Four of eight patients from the Yale series who under- able loss of function. Detailed consideration must be went surgery for SMA epilepsy became seizure-free given to vascular anatomy, because interruption of ve- (75). All five patients reported by Mihara et al. (76) were nous drainage after removal of the superior frontal gyrus either seizure-free or had only rare seizures. Taking the can lead to edema and hemorrhagic complications (77). combined experience of these more recent reports, it can Removal of the medial frontal gyrus anterior to the SMA be seen that some 58% of patients operated on for mesial appear to produce no complications. SMA resection is frontal epilepsy have become seizure-free or almost sei- now well recognized to often produce a temporary state zure-free (Table 2). of profound contralateral weakness with mutism (49,60, Because the SMA is adjacent to the primary motor 78,79). The weakness has been described to be accom- areas for the lower extremity and because the epilepto- panied by hypotonia, normal tone, or hypertonia, with genic zone can extend to the rolandic cortex'on the con- normal to increased reflexes. There i's also a superim- vexity, incomplete resection of the epileptogenic cortex posed akinesia so that movements are diminished bilat-

TABLE 1. Postsurgical results

Age Age at at Side of Outcome Pt onset surgery AlUa Seizure type Imaging Invasive EEG Etiology/pathology surgery class, 1 yr Anterior mesial frontal resections 1 4 16 Abdominal,fear Hypermotor Dysplastic gyrus, R (Depths and strips) R Cortical dysplasia RIb ant mesial front mesial front 2 9 13 None GTCS with vmive Calcified Am,L (Strips) L mesial front AVM L IVb signs to R inferior front 3 16 17 None Absences, GTCS Calcified lesion. L (Strips) R mesial front Astrocytoma LIb ant mesial front 4 7 31 Cephalic. fear Partial tonic, Negative (Depths and strips) L Unknown L IIIb hypermotor mesial frontal 5 3 10 None Hypermotor Negative (Strips) L mesial & lat Nondiagnostic L Ia front . 6 25 31 None Absence, R inferior mesial (Strips, no mesial cover) Oligodendroglioma R Ia psychomotor front R opercular front 7 15 27 None Absence, R ant mesial front, Not done Gliosis R Expired psychomotor encephalomalacia (drowning) 8 7 12 None Crying, laughing, Negative (Strips) L mesiofront Gliosis L ma psychomotor Posterior mesial frontal resections 9 7 40 Tingling L leg Postural tonic Negative (Grid and strips) R Neuronal ectopia R Ia and buttocks mesial front 10 5 42 Tingling Lleg Postural tonic Calcified lesion Not done Oligodendroglioma R IIIa, R post-mesial nondisabling, front and SPS only cingulate Lesionectomy 11 8 29 See dots Absences, GTCS L cingulate lesion Not done Gliosis, L XI indeterminate

GTCS, generalized tonic-clonic seizure, outcome classes I to IV according to ; ant, anterior; lat, lateral; front, frontal.

Epilepsia, Vol. 39, Suppl. 4, 1998 MESIAL FRONTAL EPILEPSY s59

TABLE 2. Surgical outcome in mesial frontal epilepsy electrical stimulation. Electroencephalogr Clin Neurophysiol 1994;91:179-93. Seizure-free 9. Wise SP. Corticospinal efferents of the supplementary sensorimo- Study Year n or almost tor area in relation to the primary motor cortex. In: Liiders HO, ed. Williamson et al. 1985 5 5 Supplementary sensorimotor area. Philadelphia: Lippincott- Veilleux et al. 1992 8 4 Raven, 1996:5747. [Advances in , Vol. 701. Van Ness et al. 1993 20 10 10. Buser P, Bancaud J, Chauvel P. Callosal transfer between mesial Mihara et al. 1996 5 5 frontal areas in frontal lobe epilepsies. In: Chauvel P, Delgado- Smith and King 1996 12 6 Escueta AV, et al., eds. Frontal lobe seizures and epilepsies. New Spencer et al. 1996 8 4 York: Raven Press, 1992:589404. [Advances in neurology, Vol. OCEP 1997 11 6 57.1 Total 69 40 (58%) 11. Talairach J, Bancaud J, Bonis A, et al. Surgical therapy for frontal epilepsies. In: Chauvel P, Delgado-Escueta-AV, et al: ids. Frontal lobe seizures and epilepsies. New York Raven Press, 1992:707- erally. Muteness is associated with intact comprehension 32. [Advances in neurology, Vol. 57.1 12. Rasmussen T. Characteristics of a pure culture of frontal lobe and, as later recalled by one patient, there was simply no epilepsy. Epilepsia 1983;24482-93. desire to talk. This temporary muteness and other lesser 13. Robitaille Y, Rasmussen T, Dubeau F, Tampieri D, Kemball K. degrees of language dysfunction have been observed Histopathology of nonneoplastic lesions in frontal lobe epilepsy. both with operations on the speech dominant and non- Review of 180 cases with recent MRI and PET correlations. In: Chauvel P, Delgado-Escueta AV, et al. eds. Frontal lobe seizures dominant hemisphere, more common with the former. and epilepsies. New York: Raven Press, 1992:499-513. [Advances Traction injury to the contralateral SMA during opera- in neurology, Vol. 57.1 tions on the nondominant hemisphere cannot be ex- 14. So NK. Supplementary motor area epilepsy: the clinical syndrome. cluded. Another acute neurologic abnormality found on In: Wolf P, ed. Epileptic seizures and syndromes. London: John Libbey & Co., 1994:299-317. examination is that of forced grasping in the contralateral 15. Sisodiya SM, Stevens JM, Fish DR, Free SL, Shorvon SD. The hand (47). These abnormal signs usually subside within demonstration of gyral abnormalities in patients with cryptogenic a few days or 1 to 2 weeks after surgery. Persistent partial epilepsy using three-dimensional MRI. Arch Neurol 1996; disabling deficits when the primary‘motor areas have not 53:28-34. 16. Beaumanoir A, Nahory A. Les tpilepsies btnignes partielles: 11 been violated and when vascular complications are ex- cas d’6pilepsie partielle frontale a tvolution favorable. Rev EEC cluded have been uncommon (49,60). A slight clumsi- Neurophysiol 1983;13:207-11. ness, particularly in bimanual coordination, may be 17. Connolly MB, Langill L, Wong PKH, Farrell K. Seizures involv- found on careful testing. There is insufficient informa- ing the supplementary sensorimotor area in children: a video-EEG analysis. Epilepsia 1995;36 1025-1032. tion available on the postoperative outcome with regard 18. Vigevano F, Fusco L. Hypnic tonic postural seizures in healthy to neuropsychological function. Psychiatric improve- children provide evidence for a partial epileptic syndrome of fron- ment has been reported in some patients, whereas others tal lobe origin. Epilepsia 1993;39:110-19. exhibited the same premorbid traits (24). 19. Scheffer IE, Bhatia KP, Lopes-Cendes I, et al. Autosomal domi- nant frontal lobe epilepsy, a distinctive clinical disorder. Brain 1995;118:61-73. REFERENCES 20. Oldani A, Zucconi M, Ferini-Strambi L, Bizzozero D, Smime S. Autosomal dominant nocturnal frontal lobe epilepsy: electroclini- 1. Devinsky 0, Morrell MJ, Vogt BA. Contributions of anterior cin- cal picture. Epilepsia 1996;37:96&76. gulate cortex to behavior [Review]. Brain 1995;118:279-306. 21. Hayman M, Scheffer IE, Chinvarun Y, Berlangieri SU, Berkovic 2. Vogt BA. Structural organization of cingulate cortex: areas, neu- SF. Autosomal dominant nocturnal frontal lobe epilepsy: demon- rons, and somatodendritic receptors. In: Vogt BA and Gabriel M, stration of focal frontal onset and intrafamilial variation. Neurol- eds. Neurobiology of cingulate cortex and limbic thalamus: a com- ogy 1997;49:969-75. prehensive handbook. Boston: Birkhauser, 1993: 19-70. 3. Talairach J, Bancaud J, Geier S, et al. The cingulate gyrus and 22. Steinlein OK, Mulley JC, Propping P, et al. A missense mutation human behavior. Electroencephalogr Clin Neurophysiol 1973;34 in the neuronal nicotinic acetylcholine receptor alpha 4 subunit is 45-2. associated with autosomal dominant nocturnal frontal lobe epi- 4. Vogt BA, Finch DM, Olson CR. Functional heterogeneity in cin- lepsy. Nature Genetics 1995;11:201-3. gulate cortex: the anterior executive and posterior evaluative re- 23. Quesney LF, Constain M, Fish DR, Rasmussen T. The clinical gions [Review]. 1992;2:43543. differentiation of seizures arising in the parasagittal and anterolat- 5. Rizzolatti G, Luppino G, Matelli M. The classic supplementary erodorsal frontal convexities. Arch Neurol 1990;47:677-9. motor area is formed by two independent areas. In: Luders HO, ed. 24. Mazars G. Criteria for identifying cingulate epilepsies. Epilepsia Supplementary sensorimotor area. Philadelphia: Lippincott- 1970;11:41-7. Raven, 199645-56. [Advances in neurology, Vol. 70.1 25. Geier S, Bancaud J, Talairach J, et al. The seizures of frontal lobe 6. Zilles K, Schlaug G, Geyer S, et al. Anatomy and transmitter epilepsy, a study of clinical manifestations. Neurology 1977;27: receptors of the supplementary motor areas in the human and non- 951-8. human primate brain. In: Liiders HO, ed. Supplementary sensori- 26. motor area. Philadelphia: Lippincon-Raven, 19962943. [Ad- Williamson PD, Spencer DD, Spencer SS, Novelly RA, Mattson vances in neurology, Vol. 70.1 RH. Complex partial seizures of frontal lobe origin. Ann Neurol 1985;18:497-504. 7. Fried I, Katz A, McCarthy G, et al. Functional organization of human supplementary motor cortex studied by electrical stimula- 27. Morris HH, Dinner DS, Liiders H, Wyllie E, Kramer R. Supple- tion. J Neurosci 1991;11:3656-66. mentary motor seizures: clinical and electroencephalographicfind- 8. Lim SH, Dinner DS, Pillay PK, et al. Functional anatomy of the ings. Neurology 1988;38: 1075-82. human supplementary sensorimotor area: results of extraoperative 28. Bancaud J, Talairach J. Clinical semiology of frontal lobe seizures. S60 N. K. SO

In: Chauvel P, Delgado-Escueta AV, et al. eds. Frontal lobe sei- drum BS, eds. Recent advances in epilepsy, 3rd ed. New York zures and epilepsies. New York Raven Press, 19923-58. [Ad- Churchill Livingstone, 198681-1 10. vances in neurology, Vol. 57.1 53. Lehtinen LOJ, Bergstrom L. Naso-ethmoidal electrode for record- 29. Waterman K, Purves SJ, Kosaka B, Strauss E, Wada JA. An epi- ing the electrical activity of the inferior surface of the frontal lobe. leptic syndrome caused by mesial frontal lobe seizure foci. Neu- Electroencephalogr Clin Neurophysiol 197@29:303-5. rology 1987;37:577-82. 54. Quesney LF, Katsarkas A, Gloor P, Andermann F. Contributions 30. Harvey AS, Hopkins JJ, Bowe JM, et al. Frontal lobe epilepsy: of naso-ethmoidal electrode recording in the electrographic explo- clinical seizure characteristics and localization with ictal 99mTc- ration of frontal and temporal lobe epilepsy. In: Dam M, Gram L, HMPAO SPECT. Neurology 1993;43:1966-80. Penry JK, eds. Advances in epileptology: XIIth Epilepsy Intema- 3 1. Manford M, Fish DR, Shorvon SD. An analysis of clinical seizure tional Symposium. New York Raven Press, 1981:293-304. patterns and their localizing value in frontal and lobe 55. Ralston BL. Cingulate epilepsy and secondary bilateral synchrony. epilepsies. Brain 1996;119:17-40. Electroencephalogr Clin Neurophysiol 1961;13591-8. 32. Quesney LF, Krieger C, Leimer C, Gloor P, Olivier A. Frontal lobe 56, J. Interhemispheric relations during bilateral spike-and- epilepsy: clinical and electrographic presentation. In: Porter RJ, et wave activity. Epilepsia 1981;22:45346. al., eds. Advances in epileptology: XVth Epilepsy International Symposium. New York Raven Press, 1984503-8. 57. Sharbrough FW,Cascino GD. Preoperative digital electroencepha- 33. Munari C, Bancaud J. Electroclinical symptomatology of partial lography in frontal lobe epilepsy. In: Jasper HH, Riggio S, Gold- seizures of orbital frontal origin. In: Chauvel P, Delgado-Escueta man-Rakic PS, eds. Epilepsy and the functional anatomy of the frontal lobe. AV, et al., eds. Frontal lobe seizures and epilepsies. New York New York Raven Press, 1995:167-78. [Advances in Raven Press, 1992:257-265. [Advances in neurology, Vol. 57.1 neurology, Vol. 66.1 34. Veilleux F, Sht-Hilaire J, Giard N, et al. Seizures of the human 58. Allen PJ, Fish DR, Smith SJM. Very high-frequency rhythmic medial frontal lobe. In: Chauvel P, Delgado-Escueta AV, et al., activity during SEEG suppression in frontal lobe epilepsy. Elec- eds. Frontal lobe seizures and epilepsies. New York Raven Press, troencephalogr Clin Neurophysiol 1992;82: 155-9. 1992:245-55. [Advances in neurology, Vol. 57.1 59. Fisher RS, Webber WRS, Lesser RP, Arroyo S, Uematsu S. High- frequency EEG activity at the start of seizures. J Clin Neurophysiol 35. Aimone Marsan C. Ralston BL. The eoileoticI. seizure: its func- tional morphology and diagnostic significance. Springfield, IL: 1992;9:441-8. Charles C. Thomas, 1957. 60. Van Ness P, So NK, Bleasel A, et al. Epilepsy operation outcomes 36. Wieser HG. Electroclinical features of the psychomotor seizure. with supplementary motor seizures [Abstract]. Ann Neurol 1993; London: Butterworths, 1983. 34262. 37. Kotagal P, Liiders H, Williams G, Nichols T, McPherson J. Psy- 61. Baumgartner C, Flint R, Tuxhom I, et al. Supplementary motor chomotor seizures of temporal lobe onset: analysis of symptom area seizures: propagation pathways as studied with invasive re- clusters and sequences. Epilepsy Rks 1995;2049-67. cordings. Neurology 1996;46508-14. 38. Salanova V, Moms HH, Van Ness P, et al. Frontal lobe seizures: 62. Bergin PS, Fish DR, Shorvon SD, et al. Magnetic resonance im- electroclinical syndromes. Epilepsia 1995;36: 16-24. aging in partial epilepsy: additional abnormalities shown with the 39. Commission on Classification and Terminology for the Interna- fluid attenuated inversion recovery (FLAIR) sequence. J Neurol tional League Against Epilepsy. Proposal for revised classification Neurosurg Psychiatry 1995;58:439-43. of epilepsies and epileptic syndromes. Epilepsia 1989;30:389-99. 63. Swartz BE, Khonsari A, Brown C, et al. Improved sensitivity of 40. Williamson PD. Frontal lobe seizures, problems of diagnosis and ‘*FDG-positron.emissiontomography scans in frontal and “frontal classification. In: Chauvel P, Delgado-Escueta AV, et al., eds. plus” epilepsy. Epilepsia 1995;36388-95. Frontal lobe seizures and epilepsies. New York Raven Press, 64. Schaug G, Antke C, Holthausen H, et al. Ictal motor signs and 3992:289-309. [Advances in neurology, Vol. 57.1 interictal regional cerebral hypometabolism. Neurology 1997;49 41. Ludwig B, Ajmone Marsan C, Van Buren J. Cerebral seizures of 341-50. probable orbitofrontal origin. Epilepsia 1975;16:141-58. 65. Da Silva EA, Chugani DC, Muzik 0, Chugani HT. Identification 42. Geier S, Bancaud J, Talairach J, et al. Automatisms during frontal of frontal lobe epileptic foci in children using positron emission lobe epileptic seizures. Brain 1976;99:447-58. tomography. Epilepsia 1997;38:1198-1208. 43. Rougier A, Loiseau P. Orbital frontal epilepsy: a case report. J 66. Savic I, Thorell JO, Roland P. [“C] Flumazenil positron emission Neurol Neurosurg Psychiatry 1988;51: 146-7. tomography visualizes frontal epileptogenic regions. Epilepsia 44. Chang CN, Ojemann LM, Ojemann GA, Lettich E. Seizures of 1995;36 1225-32. fronto-orbital origin: a proven case. Epilepsia 1991;32:487-91. 67. Laich E, Kuzniecky R, Mountz J, et al. Supplementary sensorimo- 45. Swartz BE. Electrophysiology of bimanual-bipedal automatisms. . tor area epilepsy. Seizure localization, cortical propagation and Epilepsia 1994;35:26474. subcortical activation pathways using ictal SPECT. Brain 1977; 46. Penfield W, Welch K. The supplementary motor area of the cere- 120:8554. bral cortex, a clinical and experimental study. Arch Neurol Psy- 68. Garcia PA, Laxer KD, van der Grond J, et al. Phosphorus magnetic chiatry 1951;66289-3 17. resonance spectroscopic imaging in patients with frontal lobe epi- 47. Penfield W, Jasper H. Epilepsy and the functional anatomy of the lepsy. Ann Neurol 1994;35:217-21. . Boston: Little, Brown and Co. 1954. 69. Berkovic SF, McIntosh A, Howell RA, et al. Familial temporal 48. Chauvel P, Trottier S, Vignal JP,Bancaud J. Somatomotor seizures lobe epilepsy: a common disorder identified in twins. Ann Neurol of frontal lobe origin. In: Chauvel P, Delgado-Escueta AV, et al., 1996;4O:227-35. editors. Frontal lobe seizures and epilepsies. New York Raven Press, 1992: 185-232. [Advances in neurology, Vol. 57.1 70. Smith WB, So NK, Thompson K. Familial temporal epilepsy [Ab- stract]. Epilepsia 1996;37(suppl 5):34. 49. Bleasel A, Comair Y, Liiders HO. Surgical ablations of the mesial frontal lobe in humans. In: Liiders HO, ed. Supplementary senso- 7 1. Levin B, Duchowny M. Childhood obsessive-compulsivedisorder rimotor area. Philadelphia: Lippincott-Raven, 1996:217-35. [Ad- and cingulate epilepsy. Biol Psychiatry 1991;301049-55. vances in neurology, Vol. 70.1 72. Kanner AM, Moms HH, Liiders H, et al. Supplementary motor 50. Tiikel K, Jasper H. The electroencephalogram in parasagittal le- seizures mimicking pseudoseizures: some clinical differences. sions. Electroencephalogr Clin Neurophysiol 1952;4:481-94. Neurology 199040: 1404-7. 51. Quesney LF, Constain M, Rasmussen T, Stefan H, Olivier A. How 73. Rasmussen T. Tailoring of cortical excisions for frontal lobe epi- large are frontal lobe epileptogenic zones? EEG, ECoG, and SEEG lepsy. Can J Neurol Sci 1991;18:60&10. evidence. In: Chauval P, Delgado-Escueta AV, et al., eds. Frontal 74. Smith JR, King DW. Surgical strategies for patients with supple- lobe seizures and epilepsies. New York Raven Press, 1992311- mentary sensorimotor area epilepsy. In: Liiders HO, editor. Supple- 23. [Advances in neurology, Vol. 57.1 mentary sensorimotor area. Philadelphia: Lippincott-Raven, 52. Quesney LF. Seizures of frontal lobe origin. In: Pedley TA, Mel- 1996415-27. [Advances in neurology, Vol. 70.1 MESIAL FRONTAL EPILEPSY S61

75. Spencer DD, Schumacher J. Surgical management of patients with Surgical treatment of the epilepsies. 2nd ed. New York Raven intractable supplementary motor area seizures. In: Liiders HO, ed. PKSS, 1993:489-500. Supplementary sensorimotor area. Philadelphia: Lippincott- 78. Laplane D, Talairach J, Meininger V, Bancaud J, Boucharine A. Raven, 1996:445-50. [Advances in neurology, Vol. 70.1 Motor consequences of motor area ablations in man. J Neurol Sci 76. Mihari T, Tottori T, Inoue Y, Seino M. Surgical strategies for l977;3 1:2949. patients with supplementary sensorimotor area epilepsy. In: Liiders 79. Rostomily R, Berger M, Ojemann G, et al. Postoperative deficits HO, ed. Supplementary sensorimotor area. Philadelphia: Lippin- and functional recovery following removal of tumors involving the con-Raven, 1996405-14. [Advances in neurology, Vol. 70.1 dominant hemisphere supplementary motor area. J Neurosurg 77. Olivier A, Awad IA. Extratemporal resections. In: Engel J Jr, ed. 1991 ;75:62-8.