▪ With clinical correlation, the high sensitivity and the specificity of IEDs for seizure disorders support the use of IEDs as the electrophysiological signature of an epileptogenic brain.
▪ IEDs represent the macroscopic field created by the summation of potentials from pathologically synchronized bursting neurons.
▪ Spikes, polyspikes, sharp waves, and spike-and-slow-wave complexes, which can be either focal or generalized.
▪ The main types of generalized IED patterns are:
▸ 3-Hz spike and slow wave
▸ Sharp and slow wave
▸ Atypical repetitive spike and slow wave
▸ Multiple spike and slow wave
▸ Paroxysmal fast activity (PFA)
▪ The epileptiform sharp waves and spikes are:
▸ Transient waveforms that may repeat, and arise abruptly out of the EEG background activity.
▸ The waveforms are asymmetric with more than one phase (usually two or three). In contrast, nonepileptiform, sharply contoured transients such as wicket waves are often approximately symmetric.
▪ The epileptiform spikes and sharp waves are often followed by a smoothly contoured slow wave (spike-and-slow-wave complexes), which disrupts the ongoing EEG background rhythm.
▪ The sharp wave or spike is produced by an abrupt change in voltage polarity that occurs over several milliseconds. The duration of an epileptiform sharp wave is between 70 and 200 msec, and the duration of spikes is less than 70 msec, although the distinction is of unclear clinical importance.
▪ The epileptiform discharge should have a physiologic field and not be confined to a single electrode except in newborn.
▪ Besides interictal epileptiform spikes and sharp waves, intermittent rhythmic delta activity over the temporal region (TIRDA) has similar specificity for temporal lobe epilepsy.
▪ There are also localized, periodic patterns of IEDs that are associated with seizures. These periodic patterns of IEDs can be lateralized to one hemisphere, as seen in periodic lateralized epileptiform discharges (PLEDs), or can be bilateral or multifocal (BiPLEDs).
▪Spikes, spike-wave discharges, polyspikes, or sharp waves (interictal epileptiform discharges [IEDs]).
▪ Broad sharp or polyphasic sharp waves (duration >200 msec).
▪ Periodic lateralized epileptiform activity (PLEDs) or bilateral independent periodic lateralized epileptiform activity (BiPLEDs) in an acute/subacute cerebral insult.
▪ Paroxysmal fast activity (PFA)—most commonly in Lennox-Gastaut syndrome.
▪ Hypsarrhythmia in infantile spasm (hemihypsarrhythmia in focal, symptomatic infantile spasm).
▪ Temporal intermittent rhythmic delta activity (TIRDA) in mesial temporal epilepsy.
▪ Occipital intermittent rhythmic delta activity (OIRDA) in generalized epilepsy, especially absence epilepsy.
▪ Frontal intermittent rhythmic delta activity (FIRDA) rarely represents IED.
▪ Continuous, near-continuous, or long trains of localized spikes or rhythmic sharp waves (intrinsic epileptogenicity) in structural abnormalities, especially focal cortical dysplasia (FCD).
▪ Regional polyspikes (especially in extratemporal region) are highly associated with FCD (80%).
▪ Focal low-voltage fast activity or electrodecrement in scalp EEG corresponding to high-frequency oscillation (HFO) in intracranial EEG.
▪ Usually occur with focal slowing, especially polymorphic delta activity (PDA) and focal attenuation.
▪ IEDs can induce brief episodes of impaired cognitive function.
▪ Six to 11 cm2 of synchronously discharging cortex are necessary to be detected by scalp electrodes.
▪ The first EEG will uncover an IED in 30–50% of the patients with epilepsy, and the yield increases to 80–90% by the fourth EEG.
▪ At least 0.5% in healthy young men.
▪ Twelve percent in all age groups and patients with progressive cerebral disorders.
▪ Majority of the reports could not establish the relationship between the AED level and the frequency of focal IEDs, although this issue remains controversial.
▪ Ten percent of patients with epilepsy extratemporal > temporal) show no IEDs.
▪ The yield of a single EEG is increased if:
▸ EEG performed within 1–2 days after the seizure
▸ Monthly seizures > seizure-free for a year:
▸ NREM sleep—increased neuronal synchronization within thalamocortical projection neurons during NREM
▪ IEDs are most focal in REM and least in NREM.
▪ IEDs are most sensitive in NREM and least in REM.
▪ Wakefulness is between REM and NREM.
▪ Usually is not difficult except in two relatively common situations:
▸ Focal IEDs may manifest as secondary bilateral synchronous discharges:
Focal discharges that appear generalized on scalp EEG due to secondary bilateral synchrony (SBS) often can be detected by examining the discharges on an expanded timescale to detect if one hemisphere discharge precedes the other.
▸ Generalized IEDs in primary generalized epilepsy can have fragmentary expression:
The focal fragmentary discharges seen in primary generalized epilepsy are typically maximal over the frontal head region with shifting lateralization and have morphologies similar to the generalized discharge
Most common in JME (pseudolocalization)
▪ Most studies evaluating the sensitivity and specificity of interictal EEG for the diagnosis of epilepsy have a number of methodological flaws:
▸ Epilepsy remains a clinical diagnosis, and some studies may include patients with incorrect diagnoses.
▸ Most studies are retrospective, originating at epilepsy centers and likely representing a significant referral bias toward patients with refractory epilepsy.
▪ Retrospective studies show that IEDs are demonstrated on the initial EEG in 30–50% of patients with the clinical diagnosis of epilepsy.
▪ The sensitivity can often be increased to 80–90% by the use of serial routine EEG up to four EEGs.
▪ In patients with infrequent seizures, the initial and subsequent EEGs are likely to show a lower sensitivity. In a study of patients who had a single seizure, only 12% had IEDs on their first EEG. A subsequent EEG showed IEDs in an additional 14% of these patients, for a cumulative sensitivity of 26% after two EEGs.
▪ There are also patients with refractory epilepsy and infrequent IEDs, e.g., patients with mesial frontal lobe seizure disorders. In a study that examined IEDs in the continuous long-term EEG records of patients with proven epilepsy, 19% of patients had no IEDs after an average of 7 days of prolonged recording.
▪ The specificity of IEDs as a marker for epilepsy depends strongly on the population studied.
▸ Only 69 (0.5%) of 13,658 healthy men who were candidates for aircrew training had IEDs on a routine EEG. Between 5 and 29 years of clinical follow-up was available for 43 of the 69 patients with IEDs, and only 1 patient developed epilepsy.
▸ In contrast, 12% of 521 nonepileptic patients residing in Rochester, MN, had IEDs on routine EEGs performed as part of a neurological evaluation. Seventy-three percent of the patients with IEDs had acute or progressive cerebral disorders. None of these patients had seizures during follow-up.
▪ The results of interictal EEG should be interpreted within the clinical setting, especially in patients with neurological disorders, structural lesions, or previous craniotomy.
▪ In scalp EEG, slow-wave activity was present in 62%. Its distribution was most commonly regional and then multiregional, or generalized respectively. The slow-wave activity was more commonly absent in temporal than extratemporal tumor groups.
▪ Highly correlated with a focal structural lesion, more prominent in acute than chronic processes.
▪ PDA is often surrounded by theta waves and maximally expressed over the lesions.
▪ Superficial lesions cause more restricted field, and deeper lesions cause hemispheric or even bilateral distribution.
▪ A lower voltage of PDA is seen over the area of maximal cerebral involvement, but a higher voltage PDA is noted in the border of lesions.
▪ More severe PDAs (closer to the lesion, more acute, higher association with underlying structural abnormality) consist of the following:
Greater variability (most irregular or least rhythmic)
Slower frequency
Greater persistence
Less reactivity
No superimposed beta activity
Less intermixed activity above 4 Hz
▪ PLEDs define an EEG pattern consisting of sharp waves, spikes (alone or associated with slow waves), or more complex waveforms occurring at periodic intervals.
▪ PLEDs usually occur at the rate of 1–2/sec and are commonly seen in the posterior head region, especially in the parietal areas. They are sometimes associated with EPC.
▪ Usually related to an acute or subacute focal brain lesion involving gray matter.
▪ Chronic PLEDs were also reported in 9% of patients with intractable epilepsy who had structural abnormalities such as cortical dysplasia or severe remote cerebral injury.
▪ PLEDs usually represent acute or subacute cerebral insults. However, in a review of 96 patients with PLEDs, 9 cases of chronic PLEDs (6.2%) were seen in patients with symptomatic focal epilepsy caused by underlying FCD or severe remote cerebral injury.
▪ Acute stroke, tumor, and CNS infection were the most common etiologies. Others included acute hemorrhage, TBI, PRES, familial hemiplegic migraine, and cerebral amyloidosis.
▪ PLEDs were more periodic when they were associated with acute viral encephalitis than with other etiologies.
▪ Seizure activity occurred in 85% of patients with mortality rate of 27%. However, 50% of patients with PLEDs never developed clinical seizures.
▪ Not epileptiform activity and not directly associated with neonatal seizures but rather with underlying structural abnormalities, especially in the deep cerebral white matter, and can be seen in a variety of conditions, including periventricular leukomalacia, hydrocephalus, meningitis, inborn errors of metabolism, stroke, hemorrhages, or HIE.
▪ Common pitfalls in identifying IEDs include misinterpretation of noncerebral potentials, benign cerebral transients and patterns, and artifacts.
▪ The electrodes, the recording equipment, or other electrical devices can produce artifacts. Artifacts can be especially challenging to distinguish in the hospital or intensive-care setting. Artifacts arising from swallowing, eye movements, body movements, sweating, pulse, and an electrocardiogram can be seen. These can be overcome by reviewing simultaneous video recording and annotation by the EEG technologist.
▪ Can be mistaken for IEDs, including:
▸ Wicket waves
▸ Small sharp spikes (SSS)
▸ Rhythmic temporal theta of drowsiness (RTTD)
▸ Subclinical electrographic discharge of adults (SREDA)
▸ 14- and 6-Hz positive bursts
▸ 6-Hz spike and slow wave
▸ Paroxysmal hypnogogic hypersynchrony
▸ Midline theta rhythms
▪ In practice, IEDs often lack some of the characteristic features listed previously, whereas some nonepileptiform discharges can demonstrate many of these characteristics.
▪ It is often useful to review possible IEDs by using different electrode montages, e.g., bipolar, referential, and laplacian recording montages. This often allows a more accurate understanding of the cerebral potential.
▪ The choice of montage can make discharges appear more focal or generalized; the Laplacian montage can make a generalized discharge appear more focal, whereas a referential montage, when the reference electrode is active, may make a focal discharge appear generalized.
▪ Almost always stereotyped for an individual.
▪ Evolving repetitive sharp waves/spikes.
▪ Infrequently, lack of evolution, regular repetitive spikes, desynchronization, or regular rhythmic slowing is noted.
▪ Evolution of frequency, amplitude, topography, and morphology.
▪ May have generalization at the onset such as slowing, attenuation, or high-amplitude sharp waves lasting only a few seconds.
▪ May cause secondary generalization.
▪ Postictal slowing or depression usually occurs in the ictal onset zone.
▪ Postictally increased focal epileptiform activity in the ictal onset zone.
▪ Surface ictal EEG was adequately localized in 72% of cases, more often in temporal than extratemporal epilepsy.
▪ Localized ictal onsets were seen in 57% of seizures and were most common in:
▸ Mesial temporal lobe epilepsy (MTLE)
▸ Lateral frontal lobe epilepsy (LFLE)
▸ Parietal lobe epilepsy (PLE)
▪ Lateralized onset predominates in neocortical temporal lobe epilepsy.
▪ Generalized onset predominates in:
▸ Mesial frontal lobe epilepsy (MFLE)
▸ Occipital lobe epilepsy (OLE)
▪ Approximately two-thirds of seizures were localized, 22% generalized, 4% lateralized, and 6% mislocalized/lateralized.
▪ False localization/lateralization occurred in 28% of occipital and 16% of parietal seizures.
▪ Rhythmic temporal theta at ictal onset was seen exclusively in temporal lobe seizures.
▪ Localized repetitive epileptiform activity was highly predictive of LFLE.
▪ Seizures arising from the lateral convexity and mesial regions of frontal lobe were differentiated by a high incidence of repetitive epileptiform activity at ictal onset in the former and rhythmic theta activity in the latter.
▪ With the exception of MFLE, ictal recordings are very useful in the localization/lateralization of focal seizures. Some patterns are highly accurate in localizing the epileptogenic lobe.
▪ One limitation of ictal EEG is the potential for false localization/lateralization in occipital and parietal lobe epilepsies.
▪ Generalized spike-wave (GSW) discharges are a hallmark of idiopathic generalized epilepsy (IGE). The reticular thalamic nucleus is the pacemaker structure for the rhythmic cortical oscillations in spindle frequency range, which transform into GSW activity in IGE. The nucleus anterior thalami and the zona incerta have an important role in SBS.
▪ GSW discharges were described in patients with shunted hydrocephalus and in hypothalamic lesions.
▪ In the generation of GSW, the cortex is considered to be the decisive factor, while the thalamus is involved secondarily.
▪ The primary role in the synchronized activity of the thalamus and cortex is attributed to the reticular nucleus. Zona incerta contains a GABAergic inhibitory effect on the higher order thalamic nuclei projecting to the neocortex and results in effective, state-dependent gating of thalamocortical information.
▪ A lesion of this system can lead to disinhibition and marked activation of the paroxysmal activity in sleep.
▪ Alterations in normal thalamocortical reciprocal interactions are critical in the generation of the regular GSW discharges characteristic of the idiopathic generalized epilepsies.
▪ Most patients with unilateral thalamic lesion and epilepsy showed bilateral synchronous GSW discharges.
▪ Absence status with bilateral GSW discharges caused by an seizure lesion in the left thalamus was reported.
▪ Children with thalamic lesion should be monitored closely for ESES. Lesions of the inferior-medial-posterior thalamic structures might have a role in the pathogenesis of bilateral SW discharges and ESES by a mechanism of disinhibition, possibly through the GABAergic system of zona incerta and its projections.
▪ HFOs characterized by very fast activity, ranging from 80 to 150 Hz, are noted at the epileptic focus in neocortical epilepsy during subdural EEG recording. Recent findings suggest that HFOs ranging between 100 and 500 Hz might be closely linked to epileptogenesis. Ripples (80–160 Hz) and fast ripples (250–500 Hz) occur frequently during IEDs and may reflect pathological hypersynchronous events. During ictal recordings, HFOs could be identified and occurred mostly in the region of primary epileptogenesis and less frequently in areas of secondary spread. HFOs are an important electrophysiological manifestation of the epileptic tissue and are associated with the spiking region and somewhat with the seizure-onset zone (SOZ). Although ripples and fast ripples share some characteristics, increasing in the SOZ and spiking regions, fast ripples are more specific to the SOZ region than ripples.
▪ The start-stop-start (SSS) phenomenon is defined as a pair of sequential ictal potentials separated by complete or almost complete cessation of seizure activity; the SSS phenomenon was found on subdural recordings in 23% of 98 patients. The two phases were morphologically similar. The first “start” usually had a narrow field. Fifty-eight percent of SSS seizures showed a complete stop. Thirty-five percent of patients and seizures restarted in a different location than the first start. When in different locations, the start, not the restart, is correlated with non-SSS origin. SSS seizures arose in the same region as non-SSS seizures in almost all patients. Subsequently, the SSS phenomenon was also observed in scalp EEG, sphenoidal, and foramen ovale electrodes, and the recognition of the phenomenon may improve the accuracy of seizure localization.
▪ An ictal slow DC shift is a slow and sustained change in EEG voltage resulting from a change in the function or interaction of neurons, glia, or both.
▪ Ictal slow baseline shifts could be recorded with DC amplifiers. They were not seen with conventional EEG systems. When the high-pass filter was opened to 0.01–0.1 Hz, ictal baseline shifts were present in scalp and intracranially recorded seizures and could have localizing value.
▪ Usually scalp-recorded ictal DC shifts are not successfully recorded because movements during clinical seizures cause artifacts. They are highly specific but low in sensitivity. Ictal DC shifts were seen in 14–40% of recorded seizures with sensitivity varying.
▪ Scalp-recorded DC shifts were detected when seizures were clinically intense, while no slow shifts were observed in small seizures. They were restricted to one to two electrodes, very closely related to the onset of low-voltage fast activity and electrodecrement.
▪ Ikeda concluded that (1) ictal DC shifts were observed in 85% of all the recorded seizures in subdural EEG and in 23% of all the scalp EEG recordings, by using LFF of 0.016 Hz for the AC amplifier; (2) ictal DC shifts were mainly surface-negative in polarity; (3) they started 1–10 sec earlier than the conventional ictal EEG onset; (4) ictal DC shifts were seen in a more restricted area (one to two electrodes) compared with the dimension defined by the conventional ictal EEG changes; (5) ictal DC shifts often coincided with the electrodecremental pattern; (6) scalp-recorded ictal slow shifts have high specificity but low sensitivity; (7) a DC amplifier is not necessary to record slow DC shifts; instead, an AC amplifier with long time constant could be used.
▪ Can be used to supplement the standard 10-20 System.
▪ Added sensitivity of additional anterior temporal region surface electrodes for recording IEDs—97% of spikes are detected using additional anterior temporal region surface electrodes, whereas 58% of spikes are detected using only 10-20 System electrodes.
▪ Closely spaced scalp electrodes can improve the yield of spike detection and localization over the standard 10-20 System.
▪ The advantages of using sphenoid electrodes, even implanted under fluoroscopic control, over T1 and T2 electrodes are unclear. Anterior temporal electrodes are able to detect nearly all IEDs recorded by the foramen ovale. Therefore, assuming that the location of the latter is analogous to that of sphenoidal electrodes optimally, the use of sphenoidal electrodes in routine interictal investigations is controversial.
▪ Areas that sphenoidal electrodes may help:
▸ Mesial temporal
▸ Lateral temporal
▸ Orbitofrontal
▪ Epileptiform discharges in temporal lobe epilepsy tend to produce a stereotyped pattern on the scalp, with largest amplitudes at the anterior temporal electrodes, independent of the topographic distribution of corticographic discharges. All these findings can be explained by assuming that a significant proportion of the electrical signal reaches the scalp through high-conductivity holes in the skull, such as the foramen ovale, the optic foramen, or the superior orbital fissure.
▪ As the optic foramen, the superior orbital fissure, and the foramen ovale are located anteriorly, this hypothesis would also explain the higher sensitivity of T1 and T2 electrodes to detect discharges in comparison to A1/A2 or T3/T4 electrodes, which are situated, however, anatomically nearer to the temporal lobe.
▪ Sleep deprivation before routine EEG can increase the yield of recorded IEDs in patients with epilepsy by 30–70%.
▪ Hyperventilation can activate IEDs and ictal EEG discharges in patients with absence seizures and complex partial seizures.
▪ Photic stimulation can activate IEDs—photoparoxysmal response.
▪ Between 70% and 77% of individuals with generalized IEDs activated by photic stimulation have epilepsy.
▪ The IEDs associated with photic stimulation frequently outlast the photic stimulus by seconds and can evolve into a seizure.
▪ The photoparoxysmal response must be distinguished from the photomyogenic response, a noncerebral myogenic and eye movement artifact that usually ceases when the stimulus is stopped.
▪ In patients with nonepileptic seizures, provocative testing and suggestion can frequently elicit the patient’s spell, although some clinicians question the ethical and diagnostic merits of provocative testing in such patients.
▪ Improved sensitivity of prolonged recordings for detecting IEDs and seizures, with a “clinically useful” result achieved in 74% of patients in a large outpatient study.
▪ Of patients with previously normal or nonspecific routine EEGs, clinically useful findings were obtained in 67.5%. Clinically useful findings included recordings of habitual events (normal and abnormal) and detection of subclinical EEG abnormalities.
▪ Ambulatory EEG is a useful tool for determining seizure frequency and response to treatment.
▪ MEG has a similar sensitivity to detect IEDs as scalp EEG.
▪ MEG identified IEDs in one-third of EEG-negative patients, especially in cases of lateral neocortical epilepsies and epilepsies due to FCD.
▪ Surface ictal EEG was adequately localized in 72% of cases, more often in temporal than extratemporal epilepsy.
▪ Localized ictal onsets were seen in 57% of seizures and were most common in:
▸ MTLE
▸ LFLE
▸ Parietal lobe epilepsy
▪ Lateralized onsets predominated in neocortical temporal lobe epilepsy.
▪ Generalized onsets in:
▸ MFLE
▸ Occipital lobe epilepsy
▪ Approximately two-thirds of seizures were localized, 22% generalized, 4% lateralized, and 6% mislocalized/lateralized.
▪ False localization/lateralization occurred in 28% of occipital and 16% of parietal seizures.
▪ Rhythmic temporal theta at ictal onset was seen exclusively in temporal lobe seizures.
▪ Localized repetitive epileptiform activity was highly predictive of LFLE.
▪ Seizures arising from the lateral convexity and mesial regions were differentiated by a high incidence of repetitive epileptiform activity at ictal onset in the former and rhythmic theta activity in the latter.
▪ With the exception of mesial frontal lobe epilepsy, ictal recordings are very useful in the localization/lateralization of focal seizures. Some patterns are highly accurate in localizing the epileptogenic lobe. One limitation of ictal EEG is the potential for false localization/lateralization in occipital and parietal lobe epilepsies.
▪ Higher amplitude of the background activity, especially beta activity on the side of a focal cerebral lesion, is rarely seen in the following conditions: tumor, FCD, abscess, stroke, and arteriovenous malformation.
▪ FCD is often associated with severe focal epilepsy. Intraoperative electrocorticography (ECoG) showed one of the following patterns:
Repetitive electrographic seizures
Repetitive bursting discharges
Continuous or quasicontinuous rhythmic spiking
▪ One or more of these patterns were present in 67% with intractable focal epilepsy associated with FCD, and in only 2.5% with intractable focal epilepsy associated with other types of structural lesions.
▪ These ictal or continuous epileptogenic discharges
▪ (I/CEDs) were usually localized, which contrast with the more widespread interictal ECoG epileptic activity, and tended to correspond with the lesion seen in the MRI.
▪ Complete resection of the FCD displaying I/CEDs correlated with good surgical outcome. Three-fourths of the patients in whom the FCD displaying I/CEDs was entirely excised had favorable surgical outcome. FCDs are highly and intrinsically epileptogenic, and intraoperative ECoG identification of the intrinsically epileptogenic dysplastic cortical tissue is critical to decide the extent of excision.
▪ Trains of continuous or very frequent rhythmic spikes or sharp waves and recurrent electrographic seizures on the scalp EEG were seen in up to 44% of FCD in one series.
▪ Eighty-six percent of patients with FCD also had localized PDA suggesting a structural lesion.
▪ Localized PDA recorded over neocortical lesions is due to underlying white matter abnormalities rather than the lesion itself. Developmental abnormalities affecting gyri are associated with underlying changes in the white matter. Malformations of cortical development must be in the differential diagnosis for localized PDA, especially when associated with epileptiform activity.
▪ Interictal low-voltage beta activity seen in FCD represents intrinsic epileptogenicity.
▪ Intrinsic epileptogenicity in FCD is caused by abnormal synaptic interconnectivity and neurotransmitter changes within the lesion.
▪ FCD has intrinsic epileptogenicity with unique EEG patterns including:
▸Continuous spikes or sharp waves
▸Abrupt runs of high-frequency spikes
▸Rhythmic sharp waves
▸Periodic spike complexes that occur during sleep
▪ The incidences of intrinsic epileptogencity in FCD were 11–20%.
▪ In all patients with FCD, the SOZ was located within the lesion.
▪ Lesional non-SOZ areas might not be directly involved in the seizure origin but might turn into a seizure focus after the removal of primary epileptogenic tissue; therefore, removal of the entire lesion and surrounding interictally active tissue is necessary. Removal of the FCD alone does not lead to a good outcome, suggesting a more widespread epileptogenic network.
▪ A strong relationship was observed between the presence of rhythmic epileptiform discharges (REDs) on the scalp EEGs and the occurrence of continuous epileptiform discharges (CEDs) recorded on ECoG. Eighty percent of patients with REDs had CEDs.
▪ Regional polyspikes, especially in extratemporal region, are highly associated with FCD (80%) and should lead the clinician to perform advanced MRI studies to detect FCD.
▪ FCD is associated with REDs.
▪ Interictal electroencephalograms revealed asymmetric suppression-burst patterns, frequent focal discharges, a nearly continuous burst-suppression pattern over the malformed hemisphere.
▪ As with interictal EEG patterns, an asymmetric ictal pattern correlates with focal or lateralized structural abnormality of the brain.
▪ Consists of a small negative wave, followed by a large-amplitude, positive slow spike. This was followed by a very slow wave, of large amplitude sometimes, which formed a plateau, often associated with monomorphic, sharp theta activity of moderate amplitude. This pattern was observed in patients with partial seizures.
▪ They are seen in patients with the earliest onset of seizures and were associated with the poorest prognosis.
▪ Interrupted by hypoactive phases on the affected side and, on the unaffected side, bursts of large-amplitude, polymorphic polyspikes of the type usually observed in suppression-burst tracings. This pattern was observed at birth or after a few months and coincided with Ohtahara syndrome. It can also be recorded until adult life in patients with epilepsia partialis continua.
▪ Consists of an asymmetrical and large-amplitude, sharp, nonreactive 7- to 12-Hz rhythm, little modified by waking state, apart from the association of slow waves during sleep, which tends to be focused in the abnormal hemisphere. This pattern was recorded in patients with seizures occurring after three months of age and associated with better outcome.
Asymmetric hypsarrhythmia originating on the affected side is also noted.
▪ Focal interictal EEG abnormalities are always consistent with anatomic location of the PNH.
▪ In slow sleep, focal abnormalities frequently change in bilaterally diffuse bursts of polyspikes.
▪ Multifocal interictal EEG abnormalities. Spikes may be asynchronous (mostly independent) and occur in multifocal regions.
▪ In a minority of patients, 3- to 4-Hz spike-wave discharges, mimicking primary generalized epilepsy, are noted.
▪ Ictal EEG abnormalities at the onset and immediately after the end of the seizures were always localized to the brain regions where PNH was located. These findings suggest that epileptic discharges may originate from abnormal circuitries located close to or involving the PNH.
▪ Epilepsy in PNH patients is generated by abnormal anatomic circuitries including the heterotopic nodules and adjacent archicortical and neocortical areas. In the patient with medically intractable epilepsy, the surgical outcome can be very favorable if the abnormal circuitry generating seizures is carefully assessed before and then removed with epilepsy surgery.
▪ Patients with PNH and epilepsy represent a heterogeneous group. Seizures result from complex interactions between PNH and allo- or neocortex. A reduction of GABA-mediated inhibitory activity was demonstrated in both the cortex and the heterotopic gray matter. Abnormalities of cortical architecture, and of cortical neuronal composition and connectivity, may allow the cortex to act as a primary epileptogenic substrate. Patients with nodular heterotopia have a high incidence of cortical abnormalities such as atrophy and polymicrogyria in addition to hippocampal atrophy. Fifty-four percent of patients with PNH have visually detectable cortical abnormalities.
▪ In most patients with bilateral and symmetric PNH, no neurologic deficits or mental retardation are present. However, the lower limits of normal IQ scores and learning disability were noted. Epilepsy is the main clinical symptom in PNH. In patients with bilateral PNH, epilepsy onset is in the second decade of life, preceded by infantile febrile convulsions. GTCS is rare and easily controlled. Focal seizures are observed in all patients and are intractable to medical treatment. Status epilepticus is never observed.
▪ A close relationship exists between heterotopic nodules and cortical regions in bilateral PNH, with an epileptogenic network including both structures. This finding may explain why a limited surgical resection to the temporal lobe fails to stop the seizures in these patients.
▪ Unilateral PNH is frequently located in the posterior paratrigonal region (i.e., a watershed area) of the lateral ventricles and may extend into the white matter to involve adjacent neocortical and archicortical areas. Epilepsy in PNH patients is generated by abnormal anatomic circuitries including the heterotopic nodules and adjacent archicortical and neocortical areas.
▪ Regardless of the different MRI features, the main clinical problem in most PNH patients is the presence of focal drug-resistant epilepsy.
▪ The presence of risk factors for prenatal brain damage, the common location in the paratrigonal region, and the lack of familial cases all suggest that acquired factors, damaging a limited region of the developing brain, may provoke the genesis of unilateral nodules. The selective ablation of a subpopulation of dividing neuroblasts alter the migration and differentiation of subsequently generated neurons, which in turn set the base for the formation of heterotopia.
▪ The clinical picture is related to the amount of heterotopic tissue, the distribution of nodules, and the extension to the overlying cortex. Therefore, the outcome is much more favorable in patients with unilateral PNH, especially single-nodule PNH, compared to patients with PNH associated with periventricular and subcortical nodules extending to the neocortex.
▪ Next to temporal lobe-onset seizures, frontal lobe-onset seizures are the most frequent type of focal seizures.
▪ Frontal seizures can occur at any age and are often caused by structural abnormalities, such as mass lesions, vascular lesions, trauma, or congenital abnormalities, that give rise to the seizures.
▪ Limitations of scalp EEG recording in the frontal lobe:
▸ Inherent risk of sampling error, since only a small portion of the frontal lobe is accessible:
Mesial interhemispheric convexity
Cingulate cortex
Depth of the cerebral sulci
▸ Trend of frontal lobe seizures to undergo rapid seizure spread within and outside the frontal lobe:
Seizure propagation to the temporal lobe via the uncinate fasciculus or the cingulum may result in a constellation of clinical and electrographic features resembling a TLS.
▸ Small epileptogenic lesions located in the depth of frontal lobe, beyond the resolution of scalp recording, which may be reflected as a widespread epileptic disturbance recorded over the fronto-parasagittal or fronto-temporal convexity, thus resembling a large epileptogenic zone, especially ipsilateral fronto-centro-temporal region.
▸ Bifrontal or generalized interictal or ictal epileptic abnormalities are commonly recorded in patients with unilateral FLE:
Mesial parasagittal convexity
Orbitofrontal region
Cingular region
Anterofrontal convexity
▪ Certain factors contribute to difficulty with localization:
▸ Focal onset obscured by the rapid spread to other areas.
▸ Brief duration of seizure.
▸ Little or no postictal change, and the discharge may be obscured by muscle artifact.
▸ Clinical seizure may occur before the ictal discharge is apparent on the EEG.
▸ Ictal EEG discharge may occur without the patient being aware of the seizure (subclinical).
▪ Seen in frontal lobe epilepsy with epileptic foci in either mesial frontal or orbitofrontal regions.
▪ Generalized absences and frontal absences may show similar clinical and EEG features and involve the same neuronal circuits.
▪ The neuronal system primarily involved in these seizures consists of a relatively limited cortico-thalamocortical circuit, including the reticular thalamic nucleus, the thalamocortical relay and the predominantly anterior and mesial frontal cerebral cortex, with the cortex probably acting as the primary driving site.
▪ The anterior cingulate gyrus is involved in self-regulation of fronto-thalamic circuits and may play an important role in both maintenance of arousal and generalized epilepsy in human.
▪ Absence seizures may not be truly generalized but rather involve selective cortical networks as described above.
▪ Can be caused by epileptic discharges arising from several areas of frontal region, including SSMA, orbitofrontal region, and cingulate gyrus.
▪ Compared with absences of childhood absence epilepsy, frontal lobe absences may have subtle repetitive vocalization, rocking movements, mild version, and brief postictal confusion.
▪ The patient may report awareness of motor arrest without loss of consciousness. Staring may evolve into a secondarily GTCS with version of the head and eyes, and focal or bilateral tonic posturing of upper limb(s).
▪ Frontal absences seem to have a more anterior epileptogenic zone than those with bilateral asymmetric tonic seizures. However, the clinical and EEG features can be very close to that of a typical or simple absence seizure.
▪ A variety of IEDs can be seen with frontal seizures:
▸ Spikes
▸ Sharp waves
▸ Spike and wave
▸ Multiple spikes or multiple spikes and waves
▸ Periodic sharp- and slow-wave complexes
▸ Low-voltage fast rhythms
▸ PFA
▪ The interictal discharges can occur as:
▸ Focal discharges
▸ Unilateral discharges over one frontal lobe
▸ Multifocal discharges over one frontal lobe
▸ Lateralized hemispheric discharges
▸ Bifrontal spike or spike-and-wave discharges that are symmetric or asymmetric
▸ Bilaterally synchronous generalized spike and spike-and-wave discharges
▸ SBS with a consistent focal onset in one region
▸ On occasion, no epileptiform activity may be apparent:
SSMA
Orbitofrontal
Ictal EEG activity in patients with SBS
Epilepsia partialis continua
▪ False localization in temporal lobe foci:
▸ Orbitofrontal lesion
▸ Mesial frontal lesion—projection through limbic pathway
▪ Secondarily generalized discharges are a common occurrence with frontal lobe epilepsy. The EEG findings that suggest SBS include:
Focal spikes or sharp waves consistently occurring in one area
Focal spike discharges that precede or initiate more generalized bursts
Persistent lateralized abnormalities such as slowing or an asymmetry over the involved area
▪ Because bilateral synchronous discharges in primary generalized epilepsy are not always perfectly synchronous and symmetric, overinterpretation of SBS should be avoided. Persistent focal abnormality in one area and with consistent initiation of most of the bursts of bilateral synchrony throughout the recording is very important to avoid this error of overinterpretation.
▪ Epileptic focus on the mesial surface of one cerebral hemisphere can project epileptiform activity obliquely to the opposite hemisphere, leading to paradoxical lateralization of the epileptiform activity, which is most likely the result of a horizontal dipole located within the interhemispheric fissure.
▪ Prolonged interictal EEG in mesial frontal epilepsy showed normal findings in 9%, and focal sharp waves were increased or appeared exclusively during sleep in the remaining patients. Interictal sharp waves were over the vertex in 90% and over the ipsilateral frontal-central region in 10%. Vertex sharp waves were asymmetric with greater distribution over the ipsilateral parasagittal region on one side in 40%.
▪ Midline fronto-central IEDs were found in 38–50% of patients with SSMA seizures.
▪ Rhythmical midline theta (RMT) activity not related to drowsiness or mental activation is significantly more common in patients with frontal (48.1%) than with temporal (3.7%) lobe epilepsy.
▪ The localizing value of RMT activity is even more important in those patients with FLE who do not have any IEDs (24%). RMT was observed in the majority of these patients with FLE (62%).
▪ Interictal RMT activity is, thus, common and has a localizing value in patients with FLE, provided that conditions such as drowsiness and mental activation as confounding factors for RMT activity are excluded.
▪ Midline IEDs can sometimes be difficult to differentiate from vertex waves. The clues to differentiate between these two are:
After-going slow waves in IEDs.
Narrower and more persistently asymmetric field in IEDs.
Presence of the same midline IEDs during wakefulness. Paradoxical lateralization of IEDs was seen in 25%.
▪ Other nonepileptiform EEG abnormalities that may be present and may give a clue to the abnormal side are focal or lateralized slowing or an asymmetry of activity over the two hemispheres.
▪ Special electrodes may be helpful in distinguishing temporal from frontal foci (sphenoid, T1/T2, closely spaced additional electrodes).
▪ IEDs in orbitofrontal epilepsy, including (1) SBS; (2) anterior temporal IED; (3) central or frontro-central IED; (4) bifrontal IED caused by volume conduction; (5) contralateral frontal IED; (6) large or blunted bifrontal or fronto-polar sharp waves with or without additional temporal involvement; and (7) various multilobar locations. At times, EEG can be normal.
▪ Ictal discharges can be complex and quite variable, including:
▸ Low-voltage fast activity
▸ Incrementing or recruiting rhythm
▸ Repetitive spikes or spikes and slow waves
▸ Rhythmic slow waves
▸ High-voltage sharp waves
▸ Focal or widespread attenuation
▸ Flattening of the background
▪ At times, the seizure may be “electrically silent,” with no apparent change evident in the EEG.
▪ In contrast to seizures arising from the temporal lobe, spike-and-wave, paroxysmal fast, and poly-frequency discharges with rapid spread are more likely to be seen with extratemporal and, particularly, with frontal seizures.
▪ As with interictal discharges, the ictal discharges may be focal, unilateral in onset, lateralized asymmetric, bilaterally synchronous, or secondarily generalized.
▪ In contrast to temporal lobe seizures, the ictal discharges are frequently more widespread and may become evident later in the course of the clinical seizure.
▪ Precise localization of frontal lobe epilepsy is often difficult and may be misleading, due to:
▸ Limitations of scalp recordings because of the anatomy of the frontal lobe and the network of projection pathways that allow the rapid spread of seizure activity within the frontal lobes or outside the temporal lobes.
▸ Widespread areas of epileptogenicity or multiple foci of seizure generators within the frontal lobe.
▸ Missed epileptic focus with scalp recordings, which may not show the focal onset or may not even show the seizure activity itself because of the inaccessibility of certain areas of the frontal lobe to scalp electrode, including:
Orbitofrontal lobe
Mesial frontal lobe
Cingulum
SSMA
▸ The focus may be too distant from the scalp electrodes.
▸ Small localized foci may not be apparent on scalp recordings because a certain population size of neurons is required for a focus to be propagated to the surface.
▸ The spike discharges that may be present may not represent the primary focus because the lesion could be projected from a buried focus within deep cortical areas.
▪ In general, well-localized frontal foci are the exception. More often, the discharges occur in a regional or lateralized manner or as bilateral discharges with or without a lateralized predominance.
▪ Bisynchronous frontal or generalized interictal and ictal discharges are often seen with seizures originating from:
▸ Orbitofrontal lobe
▸ Cingulate gyrus
▸ SSMA
▪ Frontal lobe seizures are often manifested by widespread seizure discharges with poor localization or lateralization, because frontal lobe seizures commonly propagate to contralateral frontal regions and ipsilateral temporal regions, causing a false localization.
▪ Secondarily generalized discharges are a common occurrence with frontal lobe epilepsy. The EEG findings that suggest secondary bilateral synchrony or secondary generalized discharges include:
▸ Focal spikes or sharp waves consistently occurring in one area.
▸ Focal spike discharges that precede or initiate more generalized bursts.
▸ Persistent lateralized abnormalities such as slowing or an asymmetry over the involved area.
▸ Seizures arising from the mesial frontal lobe are challenging to localize.
▸ Rhythmic alpha, beta, or theta at or adjacent to the midline was observed in 45% of cases
▸ PFA was observed at the onset of seizures arising from the inferior aspect of the SSMA and cingulate gyrus.
▪ Only 25% of mesial frontal seizures showed a localized or lateralized ictal EEG. Ictal EEG showed no EEG change or EEG was obscured by muscle artifact in more than 50%.
▪ Seizures from SSMA most commonly show PFA (33%) or electrodecrement (29%) as the initial ictal pattern, whereas seizures from lateral frontal lobe show repetitive epileptic discharges (36%) or rhythmic delta at onset (26%).
▪ Only 25% of SSMA epilepsy seizures showed correct localization or lateralization, and 75% showed no lateralization. Lateral frontal epilepsy showed correct localization in 60% and correct lateralization in an additional 27%.
▪ Although ictal involvement of SSMA produces bilateral tonic stiffening, ictal onset exclusively in SSMA is rare. The ictal onset zone and epileptogenic zone commonly extend beyond the SSMA to the primary motor cortex, premotor cortex, cingulate gyrus, and mesial parietal lobe.
▪ Intracranial or depth recording may be helpful in patients who may be candidates for surgery for more accurate localization of the seizure focus.
▪ The central midtemporal spikes are characteristically present in children with benign epilepsy of childhood with centro-temporal spikes (BECTS).
▪ Usually, the centro-temporal spike discharges are seen in children between 5 and 10 years of age.
▪ BECTS are characterized by focal twitching or paresthesias involving the hand or face on one side, and during the seizure the patient is unable to speak and exhibits excessive drooling or salivation. The seizures may spread and become generalized. Often they occur at night. The seizures are easily controlled with antiepileptic drugs (AEDs) and resolve after childhood. The children are usually otherwise normal and have no underlying lesion.
▪ The characteristic EEG features consist of slow diphasic spikes with an after-coming slow-wave component.
▪ The spike discharge is usually prominent and is high in amplitude. The spike discharges may vary in shape and amplitude, and at the other end of the spectrum, a small low-voltage spike discharge that is difficult to distinguish from background activity can be seen.
▪ The discharges can occur singly or in brief clusters of two to four in a row.
▪ The spike discharges are maximal over the central midtemporal regions.
▪ If extra electrodes are placed over the lower rolandic area midway between the central and the midtemporal electrodes, the discharge is usually maximal in these electrodes.
▪ The discharge may occur unilaterally, bilaterally, asynchronously, or synchronously with an asymmetric amplitude.
▪ At times the EEG shows what appears to be generalized bursts, but these may merely represent a widespread reflection of the centro-temporal spike discharges.
▪ The discharges can shift from side to side and may vary in location.
▪ The centro-temporal spike often presents as a tangential dipole across the rolandic fissure, with a surface positivity over the frontal region and a surface negativity over the centro-temporal regions.
▪ The frequency of the spike discharges is significantly increased during drowsiness and sleep.
▪ The discharges are not significantly altered by hyperventilation or by photic stimulation.
▪ There is no correlation between the frequency and prominence of the spike discharge and the frequency of clinical seizures. The spike discharge can also be present in patients without seizures.
▪ The EEG background is usually otherwise normal. No anatomic or structural lesions are seen, and the child usually exhibits no abnormalities.
▪ The location of EEG foci can change over time.
▪ It is the morphology of the spike discharge rather than the location that is the distinctive factor in identifying this type of spike discharge in association with the benign epilepsy of childhood.
▪ Ictal discharges of patients with benign epilepsy of childhood have rarely been seen.
▪ When recorded, the ictal discharge consists of low-voltage fast activity, which initially occurs over the centro-temporal region and then evolves into higher amplitude with slower frequencies over the temporal and centro-temporal regions. The seizure discharge may spread ipsilaterally or contralaterally and may secondarily generalize.
▪ Centro-parietal spikes can be seen in childhood between the ages of 2 and 8 years, with a peak at 5 years of age.
▪ Occur with benign seizures of childhood and may have characteristics similar to those of the centro-temporal spikes of childhood.
▪ The benign centro-temporal spikes of childhood may not necessarily be associated with seizures but can also be associated with cerebral palsy or some type of motor dysfunction.
▪ These can also occur in asymptomatic children without epilepsy.
▪ Symptomatic epilepsies arising from the central and parietal regions are associated with focal spikes or sharp-wave discharges over the central and parietal regions.
▪ The centro-parietal spikes that occur in association with symptomatic seizures, in contrast to those seen in benign epilepsy of childhood, often occur as brief or rapid spikes. These are usually more epileptogenic spike and often are associated with some underlying pathology or disturbance of cerebral function affecting the central regions.
▪ As with other focal spikes, these also may have a more widespread reflection to adjacent areas.
▪ Scalp EEG is frequently negative or maybe misleading. In general, IED is an unreliable finding in parietal lobe epilepsy. IEDs are usually widespread, multifocal, and bilateral. Secondary bilateral synchrony is recorded up to 30% of cases.
▪ Furthermore, spread of epileptic discharges from the parietal and occipital lobes to frontal and temporal regions may obscure the seizure origin.
▪ Ictal EEG in PLE is predominantly lateralized.
▪ The maximum ictal activity was over either the central-parietal or the posterior head region in most patients.
▪ Localized parietal seizure onset was noted in only 4 out of 36 patients.
▪ Surface EEG monitoring is often nonlocalizing and unreliable in the parietal lobe.
▪ False localization/lateralization is 16% in PLE. The low sensitivity of extracranial ictal EEG may be related to the predominance of simple partial seizures.
▪ Epileptogenic zone in the frontal, occipital, insular, parietal, and orbitofrontal regions may show falsely localizing IEDs. Closely spaced scalp electrodes can improve the yield of spike detection and localization over the standard 10-20 System.
▪ The most frequent anatomical localization of somatosensory auras (SSAs) was in the upper extremities, followed by lower extremities and then face. Foot involvement was found in about 13%; 48.7% had purely SSAs, whereas evolution of motor seizures occurred in 47.4%.
▪ Tingling was the most common symptom (76%) of SSAs. SSAs are highly correlated with an epileptogenic zone in the central parietal region, particularly if they are well localized in the distal extremity and are associated with a sensory march. It was found to have localizing value in 96% with central parietal epilepsy.
▪ Epileptic focus on the mesial surface of one cerebral hemisphere can project epileptiform activity obliquely to the opposite hemisphere, leading to paradoxical lateralization of the epileptiform activity. “Paradoxical lateralization” is most likely the result of a horizontal dipole located within the interhemispheric fissure.
▪ MEGs only detect dipoles parellel to the surface, and are more sensitive to generators lining the sulci than ones from gyral surfaces. Therefore, MEGs may be helpful in detecting epileptic focus in the mesial aspect of hemispheres.
▪ An prefrontal-occipital (Fp-O) EEG pattern is an age-dependent nonspecific EEG pattern reflecting the maturational process of the brain seen in both idiopathic (Panayiotopoulos syndrome) and symptomatic focal epilepsies.
▪ Highly epileptogenic.
▪ Most patients have different types of clinical seizures, most commonly generalized tonic-clonic or SSMA seizures (unilateral or bilateral).
▪ Over two-thirds of the patients were neurologically impaired.
▪ Most of midline spikes originate from the mesial or paramedian region of the cerebral cortex (supplementary motor seizures or somatosensory seizures arising from the mesial surface of the brain).
▪ They are more common in children than in adults and markedly activated with sleep.
▪ They must be distinguished from normal sleep transients. In the patients with SSMA seizures, routine EEG findings were usually normal, but prolonged EEG showed epileptiform discharges over the vertex.
▪ Typical seizures are either unilateral or bilateral tonic limb involvement with preserved consciousness.
▪ Interictal sharp waves, when present, are seen over the vertex or just adjacent to the midline in the fronto-central region in 50%.
▪ The distinction between epileptic and physiologic vertex sharp waves can be difficult.
▪ Characteristic waveforms indicating epileptiform discharges include:
The presence of prominent slow waves after the initial sharp transients
Morphology of short duration with a subsequent spike-like appearance
The appearance of sharp transients during wakefulness
Morphology of the sharp transients as polyspikes
▪ Distinguishing the midline spikes from vertex waves can be difficult. If the discharges are consistently confined to one or more midline electrodes, the distinction can be made by the following:
Different fields of distribution
The tendency for spikes to be lateralized to one side
Focal slowing in association with the spikes
Occurrence during wakefulness or drowsiness before V-waves are present
Secondary generalization of epileptiform activity
▪ Interictal and ictal scalp EEG findings may be absent, non-lateralizing, or misleading, due to paradoxical lateralization.
▪ None of patients had abnormal scalp recordings in one study, and they underlined that depth recordings were mandatory when surgery was contemplated. SSMA epilepsy cannot be excluded solely on the grounds of a normal EEG.
▪ Midline spikes occur almost exclusively in children, are strongly associated with clinical seizures, and are activated by sleep. Cz is the most frequent spike location. Most patients without sleep activation had CNS disease.
▪ Adults with midline spikes may represent a distinct entity with a worse prognosis. Midline spikes showed strong correlations with clinical seizures; 91% had epileptic seizures of diverse types. Over two-thirds of the patients were neurologically impaired.
▪ Midline spikes most probably originate from discharging lesions of the mesial or paramedian region of the cerebral cortex. Epilepsy with midline spikes is not necessarily benign.
▪ If bilateral or widespread spikes or spike-and-wave discharges are present, it may be difficult to tell whether there is a midline focus or the side or site of origin of the discharges. Simultaneous video EEG monitoring, as well as intracranial monitoring, may be helpful in documenting the seizures, indicating the site of origin and distinguishing the seizures from nonepileptic seizures.
▪ The ictal patterns consist of trains of spikes, sharp waves, spike-and-wave discharges, or rhythmic, or activity occurring focally in the midline. At times there may be an initial attenuation of the background followed by low-amplitude fast activity or spike-and-wave discharges. There may be spread to adjacent parasagittal regions or a more widespread and bisynchronous reflection of the ictal discharges. On the other hand, the ictal EEG may show no change or may be obscured by muscle artifact.
▪ Occipital spikes can be seen at different ages and with young children without seizures but with some type of visual impairment.
▪ Benign epilepsy of childhood with occipital spikes.
▪ Sturge-Weber syndrome.
▪ Epilepsy with bilateral occipital calcification.
▪ Late infantile neuronal ceroid lipofuscinosis and other symptomatic lesions.
▪ Occipital or parietal regions in most patients with congenital blindness and retinopathy during early infancy.
▪ Prevalence of 75% in retrolental fibroplasia age
▪ 3–14 years and 35% in all causes of blindness.
▪ Amplitude 50–250 μV.
▪ can be in isolation or in burst
▪ Activated by sleep.
▪ Low amplitude at 10 months; typical features at 2.5 years; after-going slow waves in mid childhood; disappear by the end of adolescence.
▪ Functional deafferentation of the visual cortex gives rise to an increase in its irritability (denervation hypersensitivity).
▪ Not an epileptiform activity.
▪ Seen more commonly in patients with mental retardation and epilepsy.
▪ Disappear during childhood or adolescence.
▪ Onset is between 3 and 15 years.
▪ Elementary visual hallucination, such as scotomata, flashing lights, and amaurosis.
▪ Complex visual hallucinations.
▪ Eye deviation.
▪ Forced eyelid closure and blinking.
▪ Ictal blindness.
▪ Ictal headache/postictal headache.
▪ Shows the following:
▸ High-amplitude spike and spike-and-wave discharges over the occipital and adjacent head regions in a unilateral, bilaterally independent, or bilaterally synchronous manner.
▸ The discharges can occur singly or in rhythmic trains of 1–3 Hz and are usually maximal over the occipital head regions.
▸ They may also spread to the adjacent posterior temporal and parietal regions.
▸ The discharges are attenuated with eye opening and reoccur with eye closure.
▸ The epileptiform activity is increased during NREM sleep and decreased during REM sleep.
▪ Consist of low-voltage fast activity or fast rhythms or fast spikes or both over one or both occipital regions, which can spread more widely.
▸ There are also subclinical discharges that can occur during sleep. The EEG is otherwise normal. Usually, there is no demonstrable lesion, and the child is otherwise normal and outgrows the seizures and the spike discharges.
▸ The EEG pattern of occipital paroxysmal discharges suppressed by eye opening is not specific to benign epilepsy of childhood. It can also be seen with variants of the entity, including benign nocturnal occipital epilepsy, and in patients who have had basilar migraine associated with seizures.
▪ One of the most common childhood seizure disorders.
▪ Characterized by prolonged, predominantly autonomic symptoms with EEG that shows shifting and/or multiple foci, often with occipital predominance.
▪ Three-quarters of patients have their first seizure between the ages of 3 and 6 years with peak at 5 years.
▪ Seizures in PS occur predominantly in sleep.
▪ Vomiting is the most common symptom.
▪ Versive seizure is seen in 60%, and progression to generalized convulsions is quite frequent.
▪ Headache may be described and is concurrent with other autonomic symptoms.
▪ Most patients will have between two and five seizures.
▪ Approximately one-third had partial status epilepticus.
▪ Normal background with high-amplitude sharp- and slow-wave complexes, similar in morphology to those seen in benign childhood epilepsy with centro-temporal spikes.
▪ There is great variability in location. Occipital localization is the most common, but all other brain regions may be involved.
▪ Frequently shift in location, this possibly being age related.
▪ Brief generalized discharges are occasionally encountered.
▪ Sharp waves or sharp- and slow-wave complexes may repeat themselves regularly and propagate, especially to frontal regions (clone-like). Seen in 17%.
▪ EEG abnormalities in PS are accentuated by sleep.
▪ Not expected to be photosensitive.
▪ Elimination of central vision and fixation.
▪ Variants of the EEG that are uncommon, but compatible, with the diagnosis include mild background abnormalities and small or inconspicuous spikes.
▪ Ten percent of patients with PS may have a normal awake EEG, but abnormalities are nearly always seen in sleep.
▪ EEG or a series of EEGs. Consistently normal EEGs are exceptional.
▪ None of the interictal EEG abnormalities in PS appear to determine prognosis.
▪ EEG foci in most patients with PS are frequently shifting location, multiplying, and propagating diffusely with age rather than persistently localizing in the occipital region.
▪ The occipital EEG spikes appeared initially and then shifted to the Fp region or appeared at the same time as Fp spikes, forming an Fp-O EEG pattern resulting in secondary occipito-fronto-polar synchrony. This phenomenon is an age-dependent nonspecific EEG pattern reflecting the maturational process of the brain.
▪ Clinical features, including tachycardia, irregular breathing, emesis, or coughing, start long after the ictal EEG onset. This onset is characterized by rhythmic theta or delta activity mixed with small spikes in unilateral posterior head region.
▪ MEG in most patients showed an epileptic focus in parieto-occipital sulcus (61.5%) or calcarine sulcus (30.8%). Despite Fp-O synchronization of spike discharges in EEG, no frontal focus was found.
▪ Epileptiform activity and seizures arising from the occipital region or adjacent posterior head regions.
▪ Depression of background activity on the side of the facial nevus and intracranial calcifications.
▪ The EEG findings consist of posterior spike-and-wave discharges, which are attenuated with eye opening in the waking state, and posterior “fast” spikes and polyspikes during the sleep state.
▪ As the disease progresses, the seizures become more frequent, and the EEG shows more diffuse spike and wave discharges and posterior dominant polyspike discharges. The ictal pattern consists of a diffuse recruiting rhythm that is maximal over the posterior head regions.
▪ Spike discharges are present over the occipital regions in the earlier stages of the disease. Later, these become more widespread.
▪ The most specific and diagnostic findings are photic-induced spikes over the occipital head regions at low rates of flash stimuli.
▪ Various symptomatic conditions and structural lesions involving the posterior head regions can also be associated with occipital seizures and epileptiform abnormalities, including:
▸ Head trauma
▸ Birth injury
▸ Porencephaly
▸ Congenital defects
▸ Cortical dysplasia
▸ Tumors
▸ Inflammatory processes
▸ Vascular lesions or malformations
▸ Mitochondrial encephalopathies
▸ Hematoma
▸ Tuberous sclerosis
▪ Focal spikes.
▪ Sharp waves.
▪ Spike-and-wave discharges.
▪ Polyspike-and-wave discharges.
▪ The discharges can occur focally over the occipital head regions or with a more widespread reflection over the posterior quadrant. Epileptiform discharges can also be seen over the temporal or centro-temporal region, or as wandering foci from one region to the other. The EEG may show other abnormalities, including slowing, asymmetry, or attenuation of activity over one or both occipital regions, in addition to the epileptiform abnormalities.
▪ Repetitive spikes, sharp waves, and spike-and-wave or rhythmic fast activity.
▪ The ictal discharges from the occipital region have a lesser tendency to evolve into polymorphic seizure.
▪ Ictal discharges that occur focally over the occipital regions are usually associated with elementary visual symptoms or nystagmus. Occipital seizures can also spread to the temporal, parietal, and frontal lobes, to produce secondarily complex partial seizures or focal motor or sensory seizures.
▪ The ictal discharges that occur in a widespread manner over the posterior head regions with spread to the temporal and other regions may make it difficult to distinguish occipital from temporal or parietal lobe epilepsy.
▪ There may also be a problem in detecting occipital discharges on scalp recordings, particularly if the seizure focus originates from the meso-occipital region at a distance from scalp leads. Rapid spread of the discharges may also make it difficult to identify the origin from the occipital lobe. Muscle artifact can contaminate the recording, adding to the difficulty in detecting the onset of the seizure discharges.
▪ The most common ictal EEG onset in occipital lobe seizures is regional involving the posterior temporal occipital region.
▪ Ictal onsets restricted to the occipital lobe were seen in only 17%.
▪ The most common pattern of spread involved the ipsilateral mesial temporal structures.
▸ Ictal onset was localized to the occipital lobe in 30%, temporal and occipito-temporal lobes in 27%, and was more diffuse in 43%.
▪ False localization and lateralization occurred in 28% of occipital seizures.
▪ The homologous area in the contralateral hemisphere, especially in the frontal and occipital regions, may be associated with similar slow waves and sharp waves, usually of lower amplitude. These findings may be caused by compression, edema, ischemia in the opposite hemisphere, transmission through commissural fibers, or volume conduction.
▪ Removal of visible lesions seen in MRI alone may not lead to a favorable outcome, suggesting a more widespread epileptogenic network or multiple FCD in these patients. Subdural EEG covering adjacent cortex or remote cortex, especially mesial temporal, is necessary for a complete resection of the epileptogenic zone.
▪ Patients with MTS have >90% of IEDs in the anterior temporal region. Therefore, frequent posterior or extratemporal sharp waves decrease the certainty of the diagnosis of MTS.
▪ Rarely, IEDs can be seen in lateral temporal and frontal regions due to spreading of the discharges to the orbitofrontal or temporal neocortex via the limbic system.
▪ In MTLE, concordance of abnormal MRI and IED is associated with good surgical outcome and is a better indicator than concordance of abnormal MRI and ictal EEG activity but non-lateralizing IED.
▪ Although bilateral IEDs decrease the chance of favorable postoperative outcome, patients with bitemporal IEDs but MRI hippocampal abnormalities concordant with the ictal-onset region still can have a good to excellent surgical outcome.
▪ Automatisms with preserved responsiveness were observed exclusively during seizures arising from the right nondominant temporal lobe. They were not observed during seizures arising from the left dominant temporal region.
▪ IEDs in MTLE show negative spikes and sharp waves in the sphenoidal, T1/T2, and F7/F8 electrodes. These IEDs are an expression of epileptiform activity in the parahippocampal cortex.
▪ Spikes in the hippocampus are usually not seen in scalp electrodes.
▪ Most IEDs are accompanied by a widespread positivity over the contralateral central-parietal region or vertex.
▪ Bilateral IEDs occur in one-third of patients, often during NREM sleep. IEDs recorded during wakefulness and REM sleep are more often lateralized and closely associated with the area of seizure onset.
▪ Seizure recording may not be necessary if serial routine EEGs are consistently concordant with MRI-identified unilateral hippocampal atrophy.
▪ In scalp EEG, slow-wave activity was present in 62%. Its distribution was most commonly regional and then multiregional, or generalized. The slow-wave activity was more commonly absent in temporal than extratemporal tumor groups.
▪ IEDs occurred in 32 of the 37 patients and were present in all of those with extratemporal epilepsy. They were most often multiregional (n = 15), or regional (n = 12), or multiregional and generalized (n = 5). They were usually found over the lobe with the tumor, but in three patients, they were predominantly contralateral.
▪ Another study of ganglioglioma in temporal lobe revealed scalp IEDs ipsilateral temporal in 71% of patients, bitemporal in 19%, and generalized in 19%.
▪ An interictal mirror focus was found in 26.9% of patients with temporal lobe epilepsy.
▪ IEDs recorded during wakefulness and REM sleep are more often lateralized and closely associated with the area of seizure onset.
▪ “Hypomotor” seizure is a signature of temporal lobe epilepsy.
▪ A bipolar antero-posterior montage using the Standard 10-20 electrode placement would incompletely record in only about 40–55% of anterior temporal spikes.
▪ Additional electrodes can be used to supplement the standard 10-20 electrode placement. Ninety-seven percent of spikes are detected using additional anterior temporal region surface electrodes (T1 and T2), whereas 58% of the spikes are detected using only standard 10-20 electrode placement Anterior temporal electrodes (T1 and T2) are able to detect nearly all IEDs recorded by the foramen ovale electrode. Therefore, assuming that the location of the foramen ovale electrode is analogous to that of sphenoidal electrodes optimally implanted under fluoroscope, the use of sphenoidal electrodes in routine interictal investigations appears not to be justified.
▪ TIRDA has the same clinical significance as anterior temporal spikes/sharp waves, a characterisctic EEG finding seen in MTLE and hippocampal sclerosis.
▪ EPC may occur in any location but is most often associated with spike discharges over the central regions. Often the discharges consist of low-amplitude rapid spikes.
▪ The spike discharges can occur as continuous repetitive spikes, in bursts or clusters, or in an intermittent fashion.
▪ Back-averaging from the clonic or myoclonic jerks may help to demonstrate the presence of an antecedent spike discharge. At times no spike discharges may be visible on surface recordings, and only a focal slow-wave abnormality or an asymmetry of activity may be present over the involved region.
▪ Corticography or intracranial recordings may be necessary to demonstrate the presence of the epileptiform discharges that are not apparent with scalp recordings.
▪ Rasmussen syndrome is the most common cause of EPC. The EEG abnormalities vary widely depending on the stage of the disease with lateralized abnormalities early in the course and bilateral abnormalities in later stages. PDA occurred in all 49 patients. This was unilateral in 19%, bilateral but with unilateral predominance in 68%, and symmetrical in both hemispheres in 13%. In 32 patients in whom seizures were recorded, a localized onset was found in only 16%.
▪ An early and striking EEG feature in all cases was the presence of focal PDA, mainly over the central and temporal regions.
▪ Other EEG features were early ictal and interictal multifocal epileptiform activity over a single hemisphere, presence of subclinical ictal EEG activity, and progressive unilateral suppression of background activity.
▪ These abnormal EEG findings correspond to typical clinical features, and an MRI indicating progressive disease is rarely observed in other conditions causing symptomatic focal epilepsy besides Rasmussen encephalitis.
▪ The scalp EEG in EPC is nonspecific and is determined by the underlying pathology. EEG can vary from normal, focal slowing with or without spikes or spike-wave complexes, especially over the central or centro-parietal regions. Focal epileptiform discharges may consist of spike, spike-and-wave, or polyspike-and-wave discharges. Focal periodic slow transients and PLEDs were reported. Bilateral, but with unilateral predominance, bursts of delta waves are the rule.
▪ Ictal SPECT and interictal PET are useful tools in presurgical workup for the localization of the epileptogenic focus in patients with epilepsia partialis continua with no definite ictal EEG localization, especially in the early phase.
▪ SPECT with 99mTc-HMPAO may be the only imaging study to suggest Rasmussen encephalitis and to localize an abnormality in a patient with worsening clinical course and normal MRI and CSF examination. Ictal SPECT shows focal hyperperfusion while EEG fails to show epileptic changes.
▪ Semiology of an insular seizure includes sensation of laryngeal constriction, paresthesias affecting large cutaneous territories, dysarthric speech, focal motor convulsive symptoms, dysgeusia and contralateral somatosensory phenomena.
▪ Generally, interictal or ictal epileptiform discharges originating in the insular cortex are unlikely to be detected by scalp-EEG recordings, unless these discharges propagate to lateral neocortical regions.
▪ Normal.
▪ Discontinuous.
▪ Focal or multifocal sharp waves.
▪ Théta pointu alternant pattern.
▪ Seen during waking and sleep and up to 12 days after the seizures are stopped.
▪ Nonspecific EEG pattern seen in status epilepticus caused by a variety of conditions such as HIE, hypocalcemia, meningitis, SAH, and benign idiopathic neonatal seizure.
▪ Described as bursts of an unreactive dominant theta activity intermixed with sharp waves with frequent interhemispheric asynchrony.
▪ The théta pointu alternant pattern is associated with good prognosis.
▪ The patterns suggesting poor prognosis, such as a paroxysmal, inactive, or burst-suppression, have never been reported in BNFC.
▪ Theta activity mixed with sharp waves alternating with relative background suppression.
▪ Was normal (10%) and discontinuous; showed focal or multifocal sharp waves or “théta pointu alternant” pattern. The “théta pointu alternant” pattern seen in half the cases may be seen during waking and sleep and up to 12 days after the seizures are stopped.
▪ Age of onset was under 1 year and ranged from 3 to 20 months.
▪ Seizures occurred in clusters, 1–10 times a day for 1–3 days, possibly recurring 1–8 weeks later.
▪ The duration of seizures ranged from 30 to 217 sec.
▪ Seizures occurred either during wakefulness or during sleep.
▪ They were characterized by motion arrest, decreased responsiveness, staring or blank eyes mostly with automatisms, and mild convulsive movements. Convulsive movements consist of eye deviation or head rotation, mild clonic movements involving the face, eyelids, or limbs, and increased limb tone.
▪ Benign infantile seizures are characterized by:
▸ Familial or nonfamilial occurrence.
▸ Normal development prior to onset.
▸ Onset mostly during the first year of life.
▸ No underlying disorders nor neurological abnormalities.
▸ Complex partial seizures or secondarily generalized seizures often occurring in clusters.
▸ Normal interictal EEG.
▸ Ictal EEG most often showing temporal focus.
▸ Excellent response to treatment.
▸ Normal developmental outcome.
▸ Ictal EEGs and clinical seizure are not much different from those of infants with refractory seizures.
▪ Characterized by generalized minor seizures (i.e., atonic-astatic, myoclonic seizures and atypical absences).
▪ Focal sharp slow waves and spikes (SHW) as observed in rolandic epilepsy (RE), but with exceptionally pronounced activation during sleep.
▪ All patients have at least atonic and nocturnal rolandic seizures.
▪ ABPE broadly overlaps with RE, electrical status epilepticus during sleep, and Landau-Kleffner syndrome.
▪ Regarding the epilepsy, the prognosis is excellent; mental deficit, however, seems to be frequent.
▪ ABPE needs to be differentiated from Lennox-Gastaut syndrome and myoclonic astatic epilepsy.
▪ Interictal EEG during wake shows characteristic features of BECTS in all cases, at least transiently, as well as generalized 3-Hz spike-wave discharges.
▪ Epileptiform activity can also be seen in parietal, temporal, occipital, and frontal regions. EEG during sleep is similar to ESES.
▪ The spike may have a phase reversal in the centro-temporal or parietal regions but less commonly in the frontal or the vertex areas.
▪ A more posterior predominance is often observed in the youngest subjects.
▪ The most striking finding of the centro-temporal spikes is their significant increase in frequency during light NREM sleep.
▪ When the frequency of centro-temporal spikes decreases abruptly during sleep, an underlying structural abnormality needs to be excluded.
▪ During the recording with referential montage, the centro-temporal spike shows a horizontal dipole configuration with a negative pole over the centro-temporal region and a positive pole over the frontal region. The clinical relevance of a dipole in BCTS has become a widely debated issue. MEG demonstrates that the spikes were generated by a single tangential dipole source located in the precentral gyrus, closer to hand SII than to SI cortex, with the positive pole directed frontally and the negative pole directed centro-temporally.
▪ Characteristic spikes over the rolandic area are regarded as neurobiological markers of BECTS. However, rolandic (centro-temporal) spikes have been reported in normal children without clinical seizures or neurologic manifestations. They are seen in 1.2–3.5% of normal healthy children in the community and 6–34% of siblings of patients affected by BECTS. The risk of epilepsy is higher if rolandic spikes remain unilateral during sleep, rolandic spikes continue during REM sleep, and there are GSW discharges.
▪ The frequency of rolandic spikes in children with ADHD (3–5.6%) is significantly higher than expected from epidemiologic studies.
▪ Benign focal epileptiform discharges, mostly rolandic spikes, were seen in 9% of childhood migraine.
▪ Temporal-parietal spikes can occur with BECTS with or without epilepsy.
▪ Malignant rolandic-sylvian epilepsy (MRSE) differs from BECTS and LKS in its refractoriness to medication, clusters of seizures, change in semiology, and secondarily generalized seizures. After careful observation over at least 5 years, surgery is considered to control refractory seizures.
▪ The clinical and EEG features which point to symptomatic focal epilepsy are:
Presence of subtle neuropsychological deficit or oromotor apraxia
Seizures starting with symptoms supporting an onset outside the opercular region
Presence of atypical absences or focal hypomotor seizures
Background EEG abnormalities
Presence of unusual fast activity
Morphological modification of the centro-temporal spikes during sleep
Enhancement of slow waves following the spike/recurrence of spikes in trains
Intermittent slow-wave focus
Frontalization of the spikes
Diffuse discharges of slow-spike-slow-wave complexes
Continuous spikes and waves during slow sleep (CSWS)
Polymorphisms of the ictal discharge
Severe postictal depression
▪ Back-averaging revealed that a biphasic spike in the contralateral rolandic region precedes an EMG burst by 6–22 msec, depending on whether proximal or distal muscles of the arm are involved. This time lag characterizes “cortical myoclonus” passing rostrocaudally through the brain stem, spinal cord, and then activating muscle.
▪ The causes of the initial convulsions in HHE syndrome include meningitis, subdural effusions, stroke, and trauma, although in many patients, no cause is found.
▪ Rhythmic bilateral slow waves, with higher amplitude on the hemisphere contralateral to the clinical seizure, are seen in the initial phase of HHE syndrome.
▪ The ictal EEG is characterized by rhythmic bilateral slow waves, with higher amplitude on the hemisphere contralateral to the clinical seizure.
▪ In the early phase of HHE, pseudorhythmical spike-and-wave discharges, contralateral to the clinical seizures, periodically interrupted by a 1- to 2-sec electrical flattening can also be noted.
▪ Polygraphic recordings do not demonstrate any consistent relation between muscle jerks and EEG discharges.
▪ Abnormal EEGs were found in 43-75% of children with autism.
▪ Forty-six percent had clinical seizures.
▪ Nearly all children with seizures had epileptiform activity.
▪ Almost 20% of those with spike discharges did not have clinical seizures.
▪ Slow-wave abnormalities were more frequent in individuals with autism.
▪ Most epileptic discharges were localized spikes; some had multiple spike foci and, only on rare occasions, generalized spikes.
▪ Seventy-five percent of the epileptic discharge foci were in the frontal region, 2.1% in the temporal region, 14.1% in the centro-parietal region, and 6.4% in the occipital region.
▪ Fifty-five percent of the frontal spikes were at midline, approximately equal at Fz and Cz. The dipole of midline spikes was in the deep midline frontal region. These results suggest that frontal dysfunctions are important in the mechanism and symptoms of autism.
Figure 9-1.
Théta Pointe Alternant Pattern; Benign Familial Neonatal Convulsion (BFNC). An EEG of a younger twin. He developed similar types of seizures on the same day as his older twin. EEG showed bursts of théta pointu alternant pattern.
An interictal EEG in BFNC was normal and discontinuous, and showed focal or multifocal sharp waves or “théta pointu alternant” pattern. The théta pointu alternant” pattern may be seen during waking and sleep and up to 12 days after the seizures are stopped. The théta pointu alternant pattern is a nonspecific EEG pattern seen in status epilepticus caused by a variety of conditions such as HIE, hypocalcemia, meningitis, SAH, and benign idiopathic neonatal seizures. It is described as bursts of an unreactive dominant theta activity intermixed with sharp waves with frequent interhemispheric asynchrony. Patterns suggesting poor prognosis, such as a paroxysmal, inactive, or burst-suppression, have never been reported in BNFC. The théta pointu alternant pattern is associated with good prognosis.1
Figure 9-2.
Théta Pointu Alternant Pattern; Benign Neonatal Convulsion (Non-Familial) or “Fifth Day Fits”. A 6-day-old boy born full term without complication. He developed the first seizure 6 days after birth. His seizure is described as either unilateral or bilateral clonic jerking, as well as apnea. Postictally, he cried for 30–45 sec and returned back to normal. He was otherwise normal. Neurological examination was normal. Workup including cranial CT and metabolic tests were negative. There was no family history of epilepsy or seizure. The seizures lasted for approximately 36 hours. Interictal EEG shows bursts of unreactive theta activity mixed with sharp waves alternating with relative background suppression.
An interictal EEG in benign non-familial neonatal convulsion was normal (10%) and discontinuous, and showed focal or multifocal sharp waves or “théta pointu alternant” pattern. The “théta pointu alternant” pattern seen in half the cases may be seen during waking and sleep and up to 12 days after the seizures have stopped. The théta pointu alternant pattern is described as bursts of an unreactive dominant 4- to 7-Hz theta activity intermixed with sharp waves with frequent interhemispheric asynchrony and shifting predominance between the two hemispheres. It is a nonspecific EEG pattern seen in status epilepticus caused by a variety of conditions such as HIE, hypocalcemia, meningitis, SAH, and benign idiopathic neonatal seizure. The patterns suggesting poor prognosis, such as a paroxysmal, inactive, or burst-suppression, have never been reported in BNFC. The théta pointu alternant pattern is associated with a good prognosis.1
Figure 9-3.
Benign Infantile Seizures (Benign Partial Epilepsy in Infancy). A 4-week-old boy who was born full term without complication. He started having recurrent apneas at 3 days of age. He was treated with phenobarbital and did well until 1 week prior to this EEG recording, when he started having his typical seizures described as eye deviation to the left with left arm and leg clonic jerking followed by apnea and cyanosis. Subsequently, clonic seizures also occurred on the right side with or without generalized tonic-clonic seizures. He had clusters of seizures up to 10/day for 4 days and had recurrent clusters of seizures at 3 and 9 months of age. Extensive metabolic workup was normal. The patient has been seizure-free for 13 months with low-dose levetriacetam. His development has been normal. No family history of seizures was noted. DWI MRI during active seizures shows cytotoxic edema in the right frontal region (open arrow), which subsequently disappeared in the repeated MRI.
Ictal EEG onset (arrow) is described as diffuse background attenuation with superimposed low-voltage spikes intermixed with rhythmic alpha activity in the right hemisphere, maximal in Cz, C4, and T4. Three seconds later, a spike is noted in the T4 (double arrows).
Figure 9-4.
Benign Infantile Seizures (Benign Partial Epilepsy in Infancy). (Continued) Twenty-two seconds after the ictal EEG onset, the EEG shows a train of spikes in the Cz corresponding to left-leg movement. Three seconds later, a train of spikes are noted in the central areas, greater in the C4, corresponding to both arms stiffening. Ictal EEG then evolves into rhythmic sharply-contoured theta activity in the same areas corresponding to clonic jerking of the left arm and leg with eye deviation to the left side.
The age of onset was under 1 year and ranged from 3 to 20 months. Seizures occurred in clusters, 1–10 times a day for 1–3 days, sometimes recurring 1–8 weeks later. The duration of seizures ranged from 30 to 217 sec. Seizures could occur during wakefulness or during sleep. They were characterized by motion arrest, decreased responsiveness, staring or blank eyes mostly with automatisms, and mild convulsive movements. Convulsive movements consist of eye deviation or head rotation, mild clonic movements involving the face, eyelids, or limbs, and increased limb tone.
Benign infantile seizures are characterized by (1) familial or nonfamilial occurrence; (2) normal development prior to onset; (3) onset mostly during the first year of life; (4) no underlying disorders nor neurological abnormalities; (5) complex partial seizures or secondarily generalized seizures often occurring in clusters; (6) normal interictal EEG; (7) ictal EEG most often showing a temporal focus; (8) excellent response to treatment; and (9) normal developmental outcome. Ictal EEGs and clinical seizures are not much different from those of infants with refractory seizures.2–6
Figure 9-5.
Benign Infantile Seizures (Benign Partial Epilepsy in Infancy). A previously healthy 9-week-old boy developed clusters of 5–10 seizures per day for 3 days. Seizures began with twitching of the left upper and lower extremities and spread to the left side of the face and then to the right upper and lower extremities, lasting for 2–5 minutes. The seizures finally stopped after multiple AEDs, including lorazepam, topiramate, phenytoin, and levetriacetam were tried. Extensive workup for metabolic and infectious diseases was unremarkable. MRI reveals cytotoxic edema, probably secondary to seizures in bilateral frontal regions, greater on the left. At a 7-month clinic visit, the patient had normal developmental milestones and had no seizures. A repeat MRI showed disappearance of the previous cytotoxic edema. EEG was normal.
The age of onset was under 1 year and ranged from 3 to 20 months. Seizures occurred in clusters, 1–10 times a day for 1–3 days, sometimes recurring every 1–8 weeks. The duration of seizures ranged from 30 to 217 sec. Seizures occurred either during wakefulness or during sleep. They were characterized by motion arrest, decreased responsiveness, staring or blank eyes mostly with automatisms, and mild convulsive movements. Convulsive movements consist of eye deviation or head rotation, mild clonic movements involving the face, eyelids, or limbs, and increased limb tone.
Benign infantile seizures are characterized by (1) familial or nonfamilial occurrence; (2) normal development prior to onset; (3) onset mostly during the first year of life; (4) no underlying disorders nor neurological abnormalities; (5) complex partial seizures or secondarily generalized seizures often occurring in clusters; (6) normal interictal EEG; (7) ictal EEG most often showing a temporal focus; (8) excellent response to treatment; and (9) normal developmental outcome. Ictal EEGs and clinical seizures are not much different from those of infants with refractory seizures.2–6
Figure 9-6.
Benign Infantile Seizures (Benign Partial Epilepsy in Infancy). (Continued) An ictal EEG shows seizure onset starting at O1, spreading to Cz and C3 when he started having hemiconvulsions on the right side. Later in the seizure, the ictal EEG activity occurs in both hemispheres when the patient develops multifocal clonic jerking. The seizure lasts for approximately 3 minutes. The patient cries afterward with only very brief postictal lethargy.
Figure 9-7.
Atypical Benign Partial Epilepsy of Childhood; Pseudo-Lennox Syndrome. A 7-year-old boy with mild developmental delay before his first seizure at 3 years of age. He had multiple types of seizures, including drop attack, GTCS, hemiconvulsion especially involving the orofacial region, and nonconvulsive status epilepticus (absence). His past EEGs showed different types of epileptiform activity such as generalized slow spike-wave or focal epileptiform activity in various locations, including multifocal, occipital, centro-temporal, or central vertex regions. In the past 2 years, EEG showed a constant finding of centro-temporal spikes with continuous discharges during slow sleep. His seizures have been relatively well controlled, but he developed a moderate degree of language and cognitive impairment.
Atypical benign partial epilepsy (ABPE) or pseudo-Lennox syndrome (PLS) is characterized by generalized minor seizures (i.e., atonic-astatic, myoclonic seizures and atypical absences) and focal sharp slow waves and spikes (SHW) as observed in rolandic epilepsy (RE), but with exceptionally pronounced activation during sleep. All patients have at least atonic and nocturnal rolandic seizures. ABPE broadly overlaps with RE, electrical status epilepticus during sleep, and Landau-Kleffner syndrome. Regarding the epilepsy, the prognosis is excellent; mental deficit, however, seems to be frequent. ABPE needs to be differentiated from Lennox-Gastaut syndrome and myoclonic astatic epilepsy. Interictal EEG during wake shows characteristic features of BECTS in all cases, at least transiently, as well as generalized 3-Hz spike-wave discharges. Epileptiform activity can also be seen in parietal, temporal, occipital, and frontal regions. EEG during sleep is similar to ESES.6–9
Figure 9-8.
Lennox-Gastaut Syndrome; Remote Stroke and Hydrocephalus. A 9-year-old boy with a history of congenital heart disease status post surgical correction, hydrocephalus with VP shunt, remote left middle cerebral artery stroke, mental retardation, and medically intractable epilepsy. He had multiple types of seizures, consisting of complex partial, generalized tonic-clonic and atypical absence seizures. He was hospitalized for a mental status change. Cranial CT shows remote ischemic infarction of the left middle cerebral artery distribution with left lateral ventricular dilatation and VP shunt insertion. Compared to the previous CT, there has been no change. EEG demonstrates continuous generalized 2-Hz spike-and-wave discharges with anterior and right predominance.
Figure 9-9.
Lennox-Gastaut Syndrome; Focal Myoclonic Seizure. EEG of an 8-year-old with cryptogenic Lennox-Gastaut syndrome. The patient developed a myoclonic jerk on the right arm (arrow) accompanied by a very brief run of rapid spikes in the left parietal and central vertex regions. This finding is supportive of the diagnosis of a focal seizure with left centro-parietal focus.
Figure 9-10.
Panayiotopoulos Syndrome. A 5-year-old girl with nocturnal GTCS associated with clicking noises from her mouth. She went back to sleep and then woke up with a throbbing headache. EEG shows clone-like repetitive occipital spike-wave discharges. Brain MRI was normal. The patient has normal development and been seizure free for over 2 years.
Panayiotopoulos syndrome (PS) is one of the most common childhood seizure disorders. It is characterized by prolonged, predominantly autonomic symptoms with EEG that shows shifting and/or multiple foci, often with occipital predominance. Three-quarters of patients have their first seizure between the ages of 3 and 6 years with a peak at 5 years. Seizures in PS occur predominantly in sleep. Vomiting is the most common symptom. Versive seizure is seen in 60%, and progression to generalized convulsions is quite frequent. Headache may be described and is concurrent with other autonomic symptoms. Most patients will have between two and five seizures. Approximately one-third had partial status epilepticus.
The interictal EEG of PS shows a normal background with high-amplitude sharp- and slow-wave complexes. These are similar in morphology to those seen in benign childhood epilepsy with centro-temporal spikes. However, in PS, there is great variability in location. Occipital localization is the most common, but all other brain regions may be involved. Moreover, the complexes frequently shift in location, this possibly being age related. Brief generalized discharges are occasionally encountered. The sharp waves or sharp- and slow-wave complexes may repeat themselves regularly and propagate, especially to frontal regions. The term “clone-like” has been used to describe this appearance. EEG abnormalities in PS are accentuated by sleep. Patients are not expected to be photosensitive. Variants of the EEG that are uncommon, but compatible, with the diagnosis include mild background abnormalities and small or inconspicuous spikes. EEG Similar patterns to the ones seen in PS occasionally occur randomly in other children. Ten percent of patients with PS may have a normal awake EEG, but abnormalities are nearly always seen in sleep EEG or a series of EEGs. Consistently normal EEGs are exceptional. None of the interictal EEG abnormalities in PS appear to determine prognosis.6,10,11
EEG foci in most patients with PS frequently shift locations, multiply, and propagate diffusely with age rather than persistently localizing in the occipital region. The occipital EEG spikes appeared initially and then shifted to the Fp region or appeared at the same time as Fp spikes, forming an Fp-O EEG pattern resulting in secondary occipitofrontopolar synchrony. This phenomenon is an age-dependent nonspecific EEG pattern reflecting the maturational process of the brain.12
MEG in most patients showed epileptic focus in the parieto-occipital sulcus (61.5%) or calcarine sulcus (30.8%). Despite Fp-O synchronization of spike discharges in the EEG, no frontal focus was found.13
Figure 9-11.
Panayiotopoulos Syndrome. A 6-year-old girl with fever and URI. She had headache and vomiting. After the third episode of vomiting, she woke up and had staring episode, which was followed immediately by clonic jerking of the left side of her body lasting for 10 minutes. This was followed by headache and lethargy. She was fine the next morning when she woke up. EEG performed 2 days later revealed right occipital spike-and-slow-wave discharges. Cranial MRI was unremarkable. She did well with oxcarbazepine until 1 year later, when she developed three episodes of seeing “colors” without loss of consciousness.
Interictal EEG shows a normal background with high-amplitude sharp- and slow-wave complexes, which are similar in morphology to those seen in benign childhood epilepsy with centro-temporal spikes. However, in PS, there is great variability in their location. Occipital localization is the most common (70%), but sharp waves may appear anywhere and often shift from one location to another. They are age related. Brief generalized discharges may occur. The sharp waves or sharp- and slow-wave complexes may repeat themselves regularly and propagate, especially to frontal regions. They are termed “clone-like,” which are seen in 19%. EEG abnormalities are activated by sleep and elimination of central vision and fixation.6,10,11
Figure 9-12.
Panayiotopoulos Syndrome. A 6½-year-old girl with a single febrile GTCS at 5 years of age who developed an episode of waking up with vomiting. She was continuously chewing, and her tongue and lips were very swollen. She was alert at the time but was unable to talk. She was noted to have left facial weakness with her tongue deviating to the left; otherwise, her neurological exam was normal. She subsequently had two more episodes of vomiting followed by left arm and head jerking with no LOC or postictal confusion. EEG shows multifocal epileptiform activity but maximally over the left occipital region.
Interictal EEG shows a normal background (slow background in 20%) with high-amplitude sharp- and slow-wave complexes, which are similar in morphology to those seen in benign childhood epilepsy with centro-temporal spikes. However, in PS, there is great variability in their location. Occipital localization is the most common (70%), but sharp waves may appear anywhere and often often shift from one location to another. They are age related. Brief generalized discharges may occur. The sharp waves or sharp- and slow-wave complexes may repeat themselves regularly and propagate, especially to frontal regions. They are termed “clone-like,” which are seen in 19% of PS cases. EEG abnormalities are activated by sleep and elimination of central vision and fixation.6,10,11
Figure 9-13.
Benign Epilepsy with Centro-Temporal Spikes (BCTS). A 12-year-old boy with recurrent episodes of nocturnal seizures described as right facial numbness followed immediately by mouth twitching, speech arrest, and drooling without loss of consciousness. Background EEG activity was normal during wakefulness. EEG during drowsiness and sleep showed frequent bilateral synchronous/independent biphasic spikes followed by slow waves in the centro-temporal regions.
During the recording with a bipolar montage, the spike may have a phase reversal in the centro-temporal or parietal regions but less commonly in the frontal or the vertex areas. A more posterior predominance is often observed in the youngest subjects. The most striking finding of the centro-temporal spikes is their significant increase in frequency during light NREM sleep. When the frequency of centro-temporal spikes decreases abruptly during sleep, an underlying structural abnormality needs to be excluded.14
Figure 9-14.
Benign Epilepsy with Centro-Temporal Spikes (BECTS); Horizontal Dipole. During the recording with referential montage, the centro-temporal spike shows a horizontal dipole configuration with a negative pole over the centro-temporal region and a positive pole over the frontal region.15,16 The clinical relevance of a dipole in BECTS has become a widely debated issue. Magnetoencephalogram (MEG) demonstrates that the spikes were generated by a single tangential dipolar source located in the precentral gyrus, closer to hand SII than to SI cortex, with the positive pole directed frontally and the negative pole directed centro-temporally.17,18
Figure 9-15.
Contralateral Parietal-Midtemporal Spikes; Symptomatic Focal Epilepsy Due to Hemorrhagic Infarction Caused by Streptococcal Infaction.
A 6-year-old girl with focal epilepsy with secondarily generalized tonic-clonic seizures caused by streptococcal infection. Her MRI/MRA is compatible with the diagnosis of hemorrhagic infarction in the left temporal-occipital region. EEG shows consistently slower frequency and less reactivity of the alpha rhythm in the left hemisphere and polymorphic delta slowing and sharp waves (open arrow) in the left posterior temporal region. In addition, there are trains of sharp waves in the right parietal-midtemporal region with horizontal dipole (double arrows), activated by drowsiness and sleep. The background activity of the right hemisphere, otherwise, is unremarkable. There is a very strong family history of nocturnal seizures. Despite active epileptiform activity in the right rolandic region in the subsequent EEGs, the patient has done well without clinical seizures.
This EEG represents two types of abnormalities caused by both left temporal hemorrhagic infarction and the genetic trait of benign epilepsy with centro-temporal spikes (BECTS).
Temporal-parietal spikes can occur in BECTS with or without epilepsy.
Figure 9-16.
Malformation of Cortical Development with Cerebral Atrophy; Congenital CMV Infection. A 12-year-old girl with spastic right hemiparesis, developmental delay, and medically intractable epilepsy with epilepsia partialis continua (EPC) as a main type of seizure associated with a history of congenital CMV infection. CT and MRI showed an extensive malformation of cortical development in the left cerebral hemisphere, consisting of diffuse cerebral atrophy, widespread polymicrogyria (double arrows), and gyral calcifications (arrow). Interictal EEG demonstrates occasional bursts of bilateral synchronous frontal spikes with left-sided predominance (white arrow head) and mild asymmetry with amplitude lower in the left hemisphere.
Patients with CMV infection with polymicrogyria suffer injury between approximately 18 and 24 weeks, whereas those with lissencephaly are injured before 16 or 18 weeks.19
Ictal SPECT is a useful tool in presurgical workup for the localization of the epileptogenic focus in patients with EPC with no definite ictal EEG localization.20,21
Figure 9-17.
Epilepsia Partialis Continua (EPC); Ictal SPECT During Focal Clonic Seizure. (Same patient as in Figure 9-16) Ictal SPECT injection (*) was performed approximately 20 sec from the onset of right focal clonic seizure. Although EEG during the injection demonstrates no definite ictal activity except lambda asymmetry (arrow head), the ictal SPECT shows definite hyperperfusion in the left frontal parietal region (arrow).
EEG in the early phase of a focal motor seizure can be normal or only show subtle abnormality; therefore, ictal SPECT is invaluable in localizing the epileptic focus. Ictal SPECT is a useful tool in presurgical workup for the localization of the epileptogenic focus in patients with epilepsia partialis continua with no definite ictal EEG localization.20,21
Figure 9-18.
Ictal EEG Activity During Focal Clonic Seizure. (Same seizure as in Figure 9-16 and 9-17) EEG findings during the same episode of focal clonic seizure can be variable. The above EEG pages show more prominent rhythmic theta activity in the left parietal temporal regions compared to the EEG activity seen in Figure 9-17.
Figure 9-19.
Rasmussen’s Encephalitis; Epilepsia Partialis Continua. An 18-year-old girl with EPC caused by Rasmussen syndrome. Axial FLAIR and coronal T2 MRIs show left cerebral hemiatrophy with increased signal intensity in the left fronto-parietal region (open arrows). Sagittal T1 MRI shows focal gyral enlargement in the left fronto-parietal region (open arrows).
Ictal single photon emission computed tomography (SPECT) during EPC demonstrates hyperperfusion in the left fronto-parietal region (black arrow) corresponding to the lesion seen in the MRI.
A reduced uptake of HMPAO (hypoperfusion) in the left temporal region in the interictal SPECT despite normal MRI was noted 6 months before the abnormalities, including abnormal neurological signs and left hemispheric atrophy in the MRI, were noted in the patient with EPC.22 Burke et al. found that SPECT with 99mTc-HMPAO was the only imaging study to suggest Rasmussen encephalitis and to localize an abnormality in a patient with a worsening clinical course and normal MRI and CSF examination.23 High concordance among clinical, EEG, CT, and SPECT studies in localization of epileptogenic foci were noted.24 Ictal SPECT showed focal hyperperfusion while EEG failed to show epileptic changes.25
Figure 9-20.
Epilepsia Partialis Continua (EPC); Rasmussen’s Syndrome. An 18-year-old girl with EPC secondary to Rasmussen syndrome. Axial MRI with FLAIR shows left cerebral atrophy with increased signal intensity in the left frontal parietal region (white arrow). Ictal SPECT during EPC demonstrates hyperperfusion in the left parietal region corresponding to the lesion seen in the MRI (black arrow). Ictal EEG during continuous jerking of her right hand shows nearly continuous sharp waves and spikes in the left frontal-temporal region time-locked with right hand jerking (*). Intermixed polymorphic delta activity is also noted over the left hemisphere.
Rasmussen syndrome is the most common cause of EPC. The EEG abnormalities vary widely depending on the stage of the disease with lateralized abnormalities early in the course and bilateral abnormalities in later stages. Polymorphic delta activity (PDA) occurred in all 49 patients studied at the Montreal Neurology Institute. PDA was unilateral in 19%, bilateral but with unilateral predominance in 68%, and symmetrical in 13%. In 32 patients in whom seizures were recorded, a localized onset was found in only 16%.26,27
An early and striking EEG feature in all cases was the presence of focal PDA, mainly over the central and temporal regions. Other EEG features were early ictal and interictal multifocal epileptiform activity over a single hemisphere, presence of subclinical ictal EEG activity, and progressive unilateral suppression of background activity. These abnormal EEG findings correspond to typical clinical features and an MRI indicating progressive disease is rarely observed in other conditions causing symptomatic focal epilepsy.28
Figure 9-21.
Rasmussen Encephalitis; Periodic Lateralized Epileptiform Discharges (PLEDs). (Same patient as in Figure 9-19) An 18-year-old girl with Rasmussen encephalitis and epilepsia partialis continua (EPC). MRI shows a focal cerebral atrophy with increased signal intensity in the right fronto-parietal region (double arrows). Ictal SPECT demonstrates hyperperfusion in the same area as the abnormal MRI (open arrow).
The scalp EEG in EPC is nonspecific and is determined by the underlying pathology. EEG can vary from normal to focal slowing with or without spikes or spike-wave complexes, especially over the central or centro-parietal regions. Focal epileptiform discharges may consist of spike, spike-and-wave, or polyspike-and-wave discharges.29 Focal periodic slow transients and PLEDs have been reported.30,31 Bilateral, but with unilateral predominant bursts of delta waves are the rule.26
Figure 9-22.
HHE (Hemiconvulsions, Hemiplegia, Epilepsy) Syndrome. A 3-year-old boy with high fever, persistent left hemiclonic seizure, and lethargy. T2-weighted MRI shows diffusely increased signal intensity over the entire right hemisphere, maximal in the mesial temporal region. Ictal EEG during the left hemiclonic seizure demonstrates bilateral high-voltage rhythmic, slow waves, intermixed with spikes and polyspikes with amplitudes higher in the right hemisphere. Note the preservation of physiologic sleep spindles in the left frontal region (arrow). Note very low-voltage EEG activity in the C3-P3 channel caused by a salt bridge from excessive smearing of the gel.
Rhythmic bilateral slow waves, with higher amplitude on the hemisphere contralateral to the clinical seizure, are seen in the initial phase of HHE syndrome.32
Figure 9-23.
Hemiconvulsion-Hemiplegia Epilepsy (HHE) Syndrome. (Same patient as in Figure 9-22) EEG shows spike/polyspike-wave complexes time-locked with contralateral hemiclonic seizures of arm and face. Note muscle artifact, maximum in the left temporal region during the left facial twitching (open arrow). Axial and coronal T2 WI MRI shows increased signal intensity in the entire right hemisphere.
The ictal EEG is characterized by rhythmic bilateral slow waves, with higher amplitude on the hemisphere contralateral to the clinical seizure. The spike-wave complexes are periodically interrupted by 1–2 sec of background attenuation.32
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
