A Study of Epileptic Discharges in Early Infancy.Purpose:The aim of this study was to clarify the characteristics of epileptic discharges in early infancy. A retrospective electroclinical study was conducted on high-risk infants, and the relationship between their postconceptional ages and the chronological and/or topographical appearance of epileptic discharges was evaluated.Subjects and Methods:Forty-two infants born with severe perinatal complications at Okayama University Hospital from 1997 to 2001 were included in the study. EEGs were recorded at least twice for each infant, with at least one EEG recorded before and one EEG recorded after the corrected age of 6 months. The subjects were divided into two groups, according to the gestational age at birth. Group A contained those born at<34 weeks of gestation, and group B contained those born at>34 weeks of gestation.Results:Twenty-nine subjects were born at<34 weeks of gestational age and belonged to group A. The mean gestational age was 28.4 weeks. Twenty-three of them had extremely low birth weights of<1,000 g. The remaining 13 subjects belonged to group B, with birth weights ranging from 1,564 to 3,850 g (mean: 2,744.1 g). Infants in group B had various perinatal complications, including severe asphyxia, meningocele, and hemimegalencephaly. Nineteen of the 42 subjects (45.2%) had epileptic discharges during the follow-up period. Twelve of these 19 subjects evolved to West syndrome (63.2%), 4 had localization-related epilepsy (including 2 infants with hemimegalencephaly; 21.0%), and 3 showed transient EEG abnormalities (15.8%). The other 23 cases without epileptic discharges had no clinical seizures except 2 infants who had simple febrile convulsions. The initial epileptic discharges on EEGs were observed at postconceptional ages ranging from 36.1 to 97.7 weeks. Sixteen of the 19 cases had initial epileptic discharges before 65 weeks of postconceptional age. The localization of epileptic discharges was also investigated in 17 of 19 subjects whose first EEG findings were hypsarrhythmia. In 5 of 8 subjects (62.5%) with initial epileptic discharge before 50 weeks of postconceptional age, the epileptic discharges were observed in the frontal regions. In 7 of 9 subjects (77.8%) with initial epileptic discharge was after 50 weeks of postconceptional age, the epileptic discharges were located in the occipital regions. These topographical characteristics of the epileptic discharges were observed not only in infants who later developed West syndrome but also in infants who later developed localization-related epilepsy. However, 2 subjects with hemimegalencephaly had independent multifocal epileptic discharges at an extremely early stage of less than 40 weeks of postconceptional age. Of 12 subjects who later developed West syndrome, 10 subjects belonged to group A, and the remaining 2 subjects belonged to group B with hypoxic-ischemic encephalopathy due to severe asphyxia. Changes in epileptic discharges were also evaluated in all subjects with West syndrome. Focal cortical discharges appeared at ages ranging from 41 to 64 weeks postconception (mean: 52.0 weeks). These focal cortical discharges then changed to hypsarrhythmia at ages ranging from 51 to 82 weeks postconception (average: 63.4). These changes occurred irrespective of gestational age at birth and the age at onset of epileptic discharges.Conclusions:In the present study, the distribution of epileptic discharges in early infancy was related to postconceptional age. Evolution to hypsarrhythmia was also related to the postconceptional age of infants who developed West syndrome. Localization of Epileptiform Discharges Detected with Continuous EEG/ fMRI Recording.Purpose:A combination of EEG and functional magnetic resonance imaging (MRI) (EEG/ fMRI) has become a promising alternative for identifying brain areas involved in epileptic activation with high spatial resolution. We recently established a technique of continuous EEG/ fMRI recording called thestepping-stone samplingmethod (SSS method)[Anami K, et al. Stepping stone sampling for retrieving artifact-free EEG during functional magnetic resonance imaging.NeuroImage(in press)], which enables blood oxygen level–dependent (BOLD) signal sampling of the whole processes of epileptiform discharges in anfMRI run. This method allows event-relatedfMRI style analysis and has a potential capability of describing a BOLD signal-change curve relevant to an epileptic discharge. The purpose of this study was to delineate the areas of BOLD activation related to epileptiform discharges, by using the continuous EEG/ fMRI recording, and to compare thefMRI results with clinical EEG findings.Methods:Six patients with partial epilepsy (five frontal lobe epilepsy, one temporal lobe epilepsy) were studied. In each patient, two or three runs offMRI were performed on a 1.5-Tesla MRI scanner (Siemens Vision Plus) in the interictal state. 420 EPI volume scans (21-slice volume, 64× 64 matrix, 60-mm thickness, 0.6-mm gap, TR: 2,000 ms) were acquired continuously over a period of 14 min per run. Nine-channel scalp EEG was simultaneously recorded in the scanner. After artifact correction, EEG data were estimated retrospectively to identify epileptic discharges and the correspondingfMRI volume scan numbers. ThefMRI data were analyzed with SPM99. After realignment and smoothing, a regressor was made based on these volume scan numbers for statistical analysis by using a general linear model. The resulting SPM{ t}was thresholded at p<0.005 (uncorrected).Results:Four patients showed a focal spike-related BOLD activation concordant with the EEG focus. In three of them, an additional BOLD activation was detected in the cortex distant from the EEG focus. One patient showed extensive BOLD activation in bilateral cerebral hemispheres related to polyspikes arising from bilateral frontal regions. The remaining one patient showed no BOLD activation with few sampled spikes.Conclusions:We demonstrated interictal epileptic activation in partial epilepsy by using EEG/ fMRI by the SSS method. In addition to the EEG foci, activation was found in other sites of the brain. These results suggest that the present method may detect secondary activation, besides primary activation related to epileptiform discharges. Determination of Language Dominance with Synthetic Aperture Magnetometry: Comparison with Wada Test and Degree of Handedness.Purpose:Language dominance is an essential piece of information for the surgical treatment of the intractable epilepsy. Recently, synthetic aperture magnetometry (SAM), a spatial filtering for the analysis of magnetoencephalography (MEG) based on adaptive beamformer, was developed. This method enables supersensitive detection of the electrical activity inside the brain, as if an electrode were inserted (SAM virtual sensor). This method also enables delineation of the activated volume of the brain as a topographic image by statistically discriminating the event-related synchronization (ERS) or desynchronization (ERD; SAM statistical method). In this study, we visualized the spatiotemporal process of cortical activation by language task by using the SAM statistical method and compared the results with those of Wada test and handedness to estimate if this noninvasive method could substitute for the Wada test.Methods:Sixteen neurosurgical patients participated in this study. MEG study for patients in this institute is approved as an Advanced Medical Treatment and regulated by the Ministry of Health, Labor, and Welfare. Handedness was judged by the Edinburgh inventory test. One hundred 3-character Japanese hiragana semantic words were presented serially at the center of the liquid crystal monitor for 3 s, with an interval of 3 s. The subjects were instructed to read them only once without phonation, immediately after each word was presented. MEG signal data of 5,000-ms duration with a 2,500-ms prestimulus interval were collected for each of the 100 trials. Volumes of interest (VOIs) for SAM analysis were set at 5-mm intervals covering the entire brain. Differences of the signal power for each VOI between the active state (0–1,000 ms after stimulus) and the control state (1,000–0 ms before stimulus) were analyzed statistically for each frequency band (alpha to high gamma) and displayed on the individual magnetic resonance (MR) images as the statistical significance of ERD or ERS. By this method, language dominance was judged according to the laterality of significant ERD or ERS at the inferior or middle frontal cortex, which is assumed as the motor language area, and categorized in three types: type 1, ERD in 13- to 25-Hz or 25- to 50-Hz frequency bands; type 2, ERD in the adjacent frequency bands (3–8 Hz, 8–13 Hz, or 50–100 Hz); and type 3, obscure dominance. The results were compared with those of the Wada test.Results:Focal distributions of ERD and ERS were detected in the language-related areas with temporal specificity: bilateral calcarine sulcus, dorsolateral region of middle frontal gyrus, opercular and triangular region of inferior frontal gyrus, temporooccipital base, angular gyrus, and posterior region of middle or superior temporal gyrus. Laterality of ERD categorized as type 1 was consistent with the language dominance determined by Wada test in eight right-handed and three left-handed patients. Type 2 ERD was observed in four right-handed patients and was consistent with the Wada test. Only one in 16 was categorized as type 3. Laterality of ERD or ERS in other regions such as the occipital, parietal, and temporal lobes was not always consistent with the results of the Wada test.Discussion:With the SAM statistical method, we demonstrated regional activation of language-related cortices with spatiotemporal separation. Our previous study in normal subjects indicated that type 1 ERD in dorsolateral frontal cortex was most consistent compared with other frequency bands and among language-related cortices. Similarly, in the patients, type 1 and type 2 ERD together account for 93% (15 of 16) and were consistent with the Wada test without exception.Conclusions:Our noninvasive MEG-SAM method might be a substitute for the semiinvasive Wada test in determining language dominance. Ictal Scalp EEG in Mesial Temporal Lobe Epilepsy with Unitemporal Versus Bitemporal Interictal Epileptiform Discharges.Purpose:To assess the lateralizing value and propagation patterns of ictal scalp EEG in mesial temporal lobe epilepsy patients with unitemporal (UIEDs) and bitemporal independent interictal epileptiform discharges (BIEDs) and mesial temporal sclerosis or lesions.Methods:We retrospectively analyzed ictal scalp EEG in 287 seizures of 48 patients who had undergone anterior temporal lobectomy. Patients were classified into a UIED group (n= 32) and a BIED (n= 16) group (unitemporal,>90% IEDs over one temporal lobe). Seizures and individual patients were lateralized on the basis of their associated ictal scalp EEG patterns according to a predefined lateralization protocol. An ictal pattern was considered correctly lateralized if it corresponded to the side of surgery. We investigated ictal scalp EEG propagation patterns and calculated a spike ratio by dividing the number of IEDs ipsilateral to the side of surgery by the total number of IEDs recorded independently on both temporal lobes.Results:In UIED patients, seizures were lateralized in 75.1% by the pattern at onset (PAO) and in an additional 19.4% by a later significant pattern (LSP). In BIED patients, seizures were lateralized in 71.0% by the PAO and in an additional 9.3% by an LSP. Ictal patterns were correctly lateralized significantly more often in UIED patients (93.8%) compared with BIED patients (62.5%). Correct lateralization occurred in 93.5% of seizures in the UIED group, but only in 57.0% of seizures in the BIED group. Scalp ictal EEGs, when lateralized at onset or later, were significantly more likely to predict the correct side for surgery in patients with UIED (98.9%) than in patients with BIED (71.0%). Switch of lateralization and bilateral independent onset were seen more frequently in the BIED group, whereas the maintenance of ictal discharges in the unilateral temporal lobe or hemisphere through the whole tracing occurred more frequently in the UIED group. Spike ratio was significantly higher in seizures that were correctly lateralized as compared with seizures incorrectly lateralized or nonlateralized.Conclusions:Interpretation of ictal scalp EEG by both PAO and LSP has more lateralizing value. UIEDs predict ipsilateral lateralization of ictal scalp EEG and the surgery side. Interpretation of ictal scalp EEG must be considered more cautiously in patients with BIEDs. BIEDs suggest bitemporal excitability or epileptogenesis, reflecting bilateral susceptibility to seizure generation or rapid seizure propagation to the contralateral temporal lobe. Nursing for Patients Not Complying with Long-Term Intracranial EEG Monitoring.Purpose:Long-term intracranial EEG monitoring (LTIEEG) is one of the important procedures of the epilepsy surgery program. However, some patients became reluctant to continue this procedure, even though they have given informed consent beforehand. We examined retrospectively the states of noncompliance of these patients.Methods:Among 82 patients older than 18 years who underwent LTIEEG (usually 2 weeks) during the past 5 years, 15 patients (18%; 10 men and five women; ages ranged from 20 to 48 years; mean, 29 years; frontal lobe epilepsy, two; occipital lobe epilepsy, three; temporal lobe epilepsy, 10) became restless or were reluctant to continue monitoring. The states of noncompliance were divided into (I) irritability (such as agitating, restless, or aggressive behavior); (II) mood alteration (such as depressive, unstable, or apathetic emotion); (III) somatization (headache, nausea, or other various somatic complaints); and (IV) confusion (confused behavior or paranoid–hallucinatory states). The time of appearance was divided into (a) initial, (b) middle, and (c) late stage. The triggering factors were retrospectively investigated. Demographics including psychiatric history, intellectual status, and previous behavior traits were examined.Results:The states of noncompliance were in I, eight subjects; II, six; III, six; and IV, five. The time of appearance of the first noncompliance was in a, nine subjects; b, four; and c, two. Seven patients had triggering factors, comprising increased/intensified seizures in six and infection in one; and five of them showed confusion. Irritability was often observed in patients with personality disorder (five patients) or a history of mood disorder (one patient) and appeared from the initial stage in them. The other states appeared in various stages. IQs of nine patients were<70, but the pattern of noncompliance did not depend on the intellectual status. Although individual and intensive nursing was provided, five patients required tranquilizers, and nine necessitated attendance of family members, even if briefly.Conclusions:Although LTIEEG is a demanding procedure, some patients, especially those with personality disorders, became irritable from the beginning, even after giving consent. Mood alteration and somatization disorders appeared in various stages of monitoring. Confusion could result from increased or intensified seizures. These observations suggest that nurses should be ready for various modes of noncompliance and should plan countermeasures before monitoring. EEG Abnormalities in Children with First Unprovoked Seizure.Purpose:To study EEG abnormalities in children with a first unprovoked seizure.Methods:In a prospective cohort study, EEG findings were studied from 63 children with a first unprovoked afebrile seizure seen at a private Epilepsy Center for Children in Mumbai, India, over the period of June 2001 to May 2002. Eligible candidates were children aged between 2 months and 10 years. Children with status, cluster of seizures within 24 h, and those with prior neonatal seizures, febrile seizures, posttraumatic seizures, or other provoked seizures that now had a first unprovoked seizure also were included. A seizure was considered unprovoked when no identifiable proximate insult could be found. Children with typical absence seizures, myoclonic seizures, and infantile spasms as well as those with first generalized tonic–clonic (GTC) seizure but with prior absence or myoclonic seizures were excluded. Seizures were classified as GT, GTC, GC, atypical absence, atonic (sudden fall), simple partial, complex partial, partial with secondary generalization, unconsciousness, or unclassifiable. An electroencephalogram (EEG) was performed in all. More than 95% of EEGs were interictal, obtained 48 h after the seizures. An attempt was made to obtain both awake and asleep recordings in all. In younger and neurologically impaired children, only sedated sleep recordings could be obtained. All EEGs were interpreted by the first author. EEGs were classified as normal or abnormal. Specific epileptiform abnormalities such as focal spikes, multifocal spikes, centrotemporal spikes, generalized spike-and-wave complexes, and background abnormalities, such as background asymmetry and slowing, were coded separately.Results:Characteristics of study group: The study group included 63 children, 32 (50.79%) boys and 31 (49.20%) girls. The mean age at the time of first seizure was 6.14 years. Fifteen (23.80%) of the children were younger than 3 years; 30 (47.61%) were between 3 and 10 years; and 18 (28.57%) were older than 10 years at the time of their first seizure. Twenty-eight (44.4%) of the children had a generalized first seizure, and 19 (30.15%) had a partial seizure with or without secondary generalization. EEG characteristics: EEGs were available in all, and 35 (55.5%) had an abnormal EEG. EEG abnormalities included focal spikes (n= 20), multifocal spikes (n= 7), generalized spike-and-wave discharges (n= 1), continuous spikes and waves during sleep (CSWS; n= 1), and background abnormalities (n= 11; asymmetry, six; slowing, five). Five children with slowing also had focal spikes. EEG abnormalities were seen in an equal percentage of cases with GTC (59.09%) and partial seizures (58.82%). Of four children with seizure clusters, EEGs were normal in three (75%). Although abnormalities were more common in children older than 3 years (older than 3 years, 58.3%; younger than 3 years, 4 6.6%); the incidence of abnormal EEGs in younger children was much higher than that described in the literature. Records, including both awake and sleep tracings, were available in 27 (42.8%) cases. Only sleep tracings were available in 27 (42.8%) and only awake in nine (14.2%). Seventeen (62%) of the 27 patients with awake and asleep tracings had abnormal EEGs, whereas 12 (44.4%) sleep records and six (66.6%) of awake records were abnormal. Background abnormalities (n= 11, 31.4%) were common and better seen when awake recording was available along with sleep tracing.Conclusions:Children with even a single unprovoked seizure have a high incidence of EEG abnormalities, and the awake recording is valuable to bring out background abnormalities. Electrical Status During Slow-Wave Sleep (ESES): A Review of Fifteen Pediatric Cases from India.Purpose:To review the clinical and electrographic features of electrical status during slow-wave sleep (ESES) by using short-term EEG monitoringMethods:The study was done by using short-term (3 h) EEG recording with polygraphic variables of awake and sleep states. ESES is an EEG-defined syndrome, and stringent EEG criteria for inclusion were followed by using spike-and-wave index occupying≥85% of recorded sleep, less frequent or ceasing to be continuous in the awake state, and findings present on more than two occasions. Each case was assessed for the type, frequency of seizures, neuropsychological testing, neuroimaging with computed tomography (CT) or magnetic resonance imaging (MRI), and followed up closely for a period of 6 months to 6 years with repeated EEG at 1- to 3-month intervals.Results:Of 2,080 patients with pediatric epilepsy older than 10 years, typical ESES (n= 15) form the basis of study with an incidence of 0.72%. Excluded from study were seven cases; atypical ESES was observed in six cases, three with asymmetrical ESES and three with a spike–wave index of 50–60% of the sleep record; and a single case with transient, drug-induced ESES. Age at onset was between 7 months and 6 years, with a preponderance of male patients (10 of 15); eight had normal development before seizure onset (idiopathic), whereas seven were symptomatic. History of febrile seizures was present in two, and one had a family history of epilepsy. Language dysfunction was noted (n= 4), two with delayed speech development but normal motor milestones, one with autistic features, and one had total aphasia for 22 days before ESES, but not the Landau–Kleffner syndrome. Cases were categorized as per the seizure semiology; group 1 with infrequent nocturnal seizures (n= 2), group 2 with absence and motor seizures (n= 7), and group 3 with myoatonic falls (n= 6). None had tonic seizures. Regardless of seizure frequency, detection of ESES coincided with neuroregression in all. Before ESES, nine of 15 cases had seizures, and EEGs showed multifocal spikes predominantly in the frontotemporal region with secondary bilateral synchrony and later progressed to ESES syndrome. The remaining six cases showed ESES at the time of presentation. None had remission of ESES, even after 6 years of follow-up, the oldest case being age 13.5 years. Remission of seizures was observed in four cases, even though the EEGs persisted in showing ESES.Conclusions:ESES is associated with deterioration of neurologic function, irrespective of seizure control, and if strict criteria are used, one can eliminate other conditions, with the EEG showing a pattern simulating ESES. ESES is usually self-limiting and normally remits by adolescence. However, a longer follow-up is required to establish this entity. Diagnosis of ESES can be made on a brief sleep recording, but an overnight sleep recording would be essential if discharges are<85% and at follow-up to review the progress. ESES is a definitely identifiable, specific, sleep-related syndrome that may be overlooked if a sleep recording during routine EEG is not carried out. The neurologic dysfunction and regression is uniformly reported in all series and provides a clue to the emergence of this syndrome. A dilemma exists with regard to morphology and evolution of the spike–wave discharges, the percentage of SW index, day-to-day variations in the same person, and the lack of definitive treatment to suppress these discharges. [ABSTRACT FROM AUTHOR]