Temporal Lobe Epilepsy

Temporal Lobe Epilepsy

Temporal Lobe Epilepsy J Engel Jr. and N Salamon, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA ã 2015 Elsevier Inc. All rights reserv...

2MB Sizes 0 Downloads 22 Views

Temporal Lobe Epilepsy J Engel Jr. and N Salamon, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA ã 2015 Elsevier Inc. All rights reserved.

Glossary Anosmia Loss of smell. Borborygmi Stomach gurgling. CA2 Defines an area of the hippocampus. Deja entendu The feeling of having heard something before. Deja vecu The feeling of having experienced something before.

Introduction Temporal lobe epilepsy (TLE) is the most common form of epilepsy. The term TLE is generally used to connote epilepsy conditions characterized by focal seizures due to lesions in the temporal lobe or otherwise mediated by temporal lobe structures. The classification system of the International League Against Epilepsy (ILAE), however, does not recognize a discrete syndrome called TLE (Berg et al., 2010). There are, in fact, many forms of TLE. Mesial temporal lobe epilepsy (MTLE), by far the most prevalent, is a specific constellation recognized by the current ILAE classification, but is, itself, composed of a number of different conditions (Engel et al., 2008a). The most common pathological substrate of MTLE is hippocampal sclerosis (Mathern et al., 2008), but any structural lesion in mesial temporal limbic structures can give rise to this condition, and a genetic form of MTLE also exists (Cendes et al., 2008). Epileptic seizures generated by mesial temporal limbic structures have characteristic features that differentiate them from focal seizures originating in the neocortex. The former have been referred to as temporal lobe seizures, but more correctly are called limbic seizures, as seizures originating in the lateral temporal neocortex that do not preferentially project to mesial temporal structures have different manifestations. Epilepsies associated with seizures that begin in the lateral temporal neocortex are referred to as lateral temporal lobe epilepsy (LTLE), and the associated seizure types are variable. If the ictal discharge preferentially propagates to mesial temporal structures, which is the most common situation, the behavioral features are essentially identical to seizures of MTLE. If they remain limited to the neocortex, or propagate to other neocortical areas, however, they take on behavioral characteristics dependent upon the function of the neocortex involved. Finally, lesions in the neocortex outside the temporal lobe, particularly limbic structures such as the orbital frontal and cingulate cortex, as well as the occipital cortex below the level of the calcarine fissure, can preferentially propagate to mesial temporal structures, giving rise limbic seizures that are indistinguishable from those of MTLE. These conditions may be misdiagnosed as TLE when the extratemporal location of the epileptogenic region is not identified, but will not be

Brain Mapping: An Encyclopedic Reference

Deva vu The feeling of having seen something before. Dystonic posturing Increased time causing a sustained unusual posture. Jamais entendu The feeling of never having heard something before. Jamais vecu The feeling of never having experienced something before. Jamais vu The feeling of never having seen something before.

considered further here, per se, although limbic seizures in general will be discussed. Prognosis and treatment of TLE depend upon the nature of the underlying pathophysiology, as some pathological substrates require specific interventions. Precise localization of the epileptogenic abnormality to one temporal lobe, or to a particular part of a temporal lobe, is of clinical value when resective surgical therapy is considered for medically refractory epileptic seizures.

Limbic Seizures Mechanisms The clinical manifestations of limbic seizures are mediated by mesial temporal structures, including the hippocampus, amygdala, perihippocampal cortex, and their projection sites, particularly the insula, hypothalamus, and limbic neocortex (Engel and Williamson, 2008). Areas of ictal involvement can be seen as hypermetabolism on 18F-fluorodeoxyglucose positron emission tomography (FDG-PET), with surrounding hypometabolism, presumably reflecting postictal depression (Figure 1).There is no convincing evidence that the clinical manifestations of limbic seizures generated in the neocortex but mediated by mesial temporal structures are different from those initially generated in mesial temporal structures, unless features of the functional neocortex, such as visual symptoms for the occipital cortex, also occur initially. There is reason to believe, however, that the fundamental neuronal mechanisms responsible for ictal onset in hippocampus, and perhaps other mesial temporal structures, are different from mechanisms of ictal activity originating in the neocortex (Engel et al., 2008b). Mechanisms responsible for the generation of limbic seizures have been studied in animal models of MTLE produced by neurotoxins such as kainic acid and pilocarpine, or repetitive electrical stimulation, all of which cause status epilepticus and subsequent neuronal loss in the hippocampus. The lesions are pathologically similar to those of human hippocampal sclerosis. Loss of principal neurons, particularly in the hilus of the dentate gyrus, and CA fields of the hippocampus proper (sparing CA2 in classical hippocampal sclerosis), as well as loss of certain inhibitory interneurons, particularly in the hilus,

http://dx.doi.org/10.1016/B978-0-12-397025-1.00101-9

853

854

INTRODUCTION TO CLINICAL BRAIN MAPPING | Temporal Lobe Epilepsy

Figure 1 Left: Interictal FDG-PET scan revealing a wide area of hypometabolism in the right frontal and temporal lobes (right is on the right side in these older scans). Right: Scan obtained when FDG was injected just prior to several dyscognitive focal seizures. The most marked zone of hypermetabolism involves the right posterior frontal and anterior temporal lobes, but discrete smaller areas of increased metabolic activity can be seen in the right cingulate cortex, posterior hippocampus, and thalamus. Reproduced from Engel, J., Jr., Kuhl, D. E., Phelps, M. E., Rausch, R., & Nuwer, M. (1983). Local cerebral metabolism during partial seizures. Neurology, 33, 400–413, with permission.

result in axonal sprouting and synaptic reorganization of surviving neurons. This leads to both enhanced excitation and inhibition, causing a predisposition to pathological hypersynchronization, which appears to be the neuronal basis of most hippocampal onset seizures (Engel et al., 2008b). Epileptiform discharges that are confined to one hippocampus are not necessarily associated with detectable signs or symptoms, and inherent protective mechanisms within the hippocampus appear to resist propagation to deep and contralateral structures responsible for impaired consciousness. As a result, ictal discharges often remain limited to unilateral limbic structures and, in this case, can give rise to auras that do not evolve into more severe seizures (auras in isolation). It is generally assumed that bilateral hippocampal involvement is necessary for impaired consciousness and postictal amnesia seen with more severe limbic seizures, but there may be many ways that epileptiform activity could affect structures outside the hippocampus to alter consciousness (Blumenfeld, 2011). Although patients may experience limbic seizures without auras, when auras occur, they typically are more frequent in isolation than with seizures that impair consciousness, presumably as a result of inherent protective mechanisms that prevent spread. This is not usually the case with neocortical seizures. Limbic seizures are the most pharmacoresistant seizure type (Semah et al., 1998), in part due to the fact that potential antiseizure compounds usually are not screened against animal models that mimic this seizure type. There is evidence that repeated limbic seizures can cause progressive functional and structural disturbances, including memory deficits specific for the dominant or nondominant hemisphere, and hippocampal atrophy.

consciousness were referred to as complex partial seizures, and those without impaired consciousness as simple partial seizures (Commission on Classification, 1981). Recently, this distinction has been abandoned as a means of classifying seizure types (Berg et al., 2010), but it is still important to identify impaired consciousness in individual patients, as this has an important impact on associated disabilities. The term dyscognitive focal is now used when referring to focal seizures with impaired consciousness in individual patients (Berg et al., 2010). Although some patients experience dyscognitive focal seizures without auras, most limbic seizures begin with autonomic, psychic, or certain sensory auras, which may or may not progress to impaired consciousness. Auras in isolation are usually more common than auras that proceed to dyscognitive focal seizures. Dyscognitive focal seizures typically begin with an arrest reaction or motionless stare (Engel and Williamson, 2008) and may be bland with no clinical features other than amnesia for the ictal event. More often, automatisms occur. A typical dyscognitive limbic seizure lasts one to two minutes, followed by a postictal period of disorientation lasting minutes to hours. Automatisms may also occur during the postictal period:



• Phenomenology According to the 1981 ILAE International Classification of Epileptic Seizures, ictal events associated with impaired

Autonomic auras: The most common ictal autonomic symptoms consist of abdominal discomfort or nausea, often ascending to the throat (epigastric rising); stomach pain; borborygmi; belching; flatulence; and even vomiting. Other autonomic symptoms include pallor, flushing, sweating, piloerection, pupil dilatation, alterations in heart rate and respiration, and urination. Sexual arousal and orgasm have been reported. Psychic auras: Ictal dysmnesic auras are distorted memory experiences such as feelings of inappropriate familiarity (de´ja` vu, de´ja` entendu, and de´ja` ve´cu) or strangeness (jamais vu, jamais entendu, and jamais ve´cu), flashbacks, forced thinking, and rapid recollection of past events. Dysphasic

INTRODUCTION TO CLINICAL BRAIN MAPPING | Temporal Lobe Epilepsy









auras occur with seizures in the language-dominant hemisphere and present with more complex signs and symptoms than neocortical aphasic auras or negative motor phenomena. Cognitive auras include dream states, distortion of time sense, sensations of unreality, and depersonalization. The most common affective aura is fear, but other symptoms such as anger, depression, and embarrassment occur. Gelastic seizures (ictal inappropriate sudden mirthless laughter) are also often limbic seizures. Limbic sensory auras: Olfactory auras, also referred to as uncinate fits, are classically associated with limbic seizures but are not particularly common. Smells are usually unpleasant, and anosmia occurs occasionally. Gustatory auras are also usually disagreeable. Nonspecific somatosensory auras can occur with limbic seizures, such as nonlateralized tingling sensations of the face or hands. Automatisms: Automatisms are involuntary automatic behaviors that usually occur during dyscognitive limbic seizures but can rarely occur before impairment of consciousness, in which case they are recalled postictally. Automatisms also commonly occur during the postictal period. Automatisms can be spontaneous and stereotyped or reactive. The most common spontaneous automatisms are oroalimentary automatisms, such as lip smacking and chewing. Gestural automatisms include picking at clothes, scratching, dressing and undressing, or rearranging objects; ambulatory automatisms involve walking or running, usually in a particular pattern; verbal automatisms consist of simple phrases; mimetic automatisms are alterations in facial expression; hyperkinetic automatisms, such as chaotic flailing of the limbs, bicycle movements of the legs, screaming, and spitting, can also indicate a neocortical disturbance, usually in the frontal lobe; and rare sexual automatisms include behavior such as pelvic thrusting and masturbation and also can indicate frontal lobe involvement. Reactive automatisms are not stereotyped, but rather are determined by environmental stimuli at the time of the seizure. These may consist of aimless repetition of behaviors that were in progress at the time of seizure onset. Patients may be able to skillfully respond to new situations to avoid injury or respond inappropriately, including violently, to perceived threat. Other motor behaviors: Dystonic posturing of one upper extremity often occurs contralateral to the site of ictal onset. Gestural automatisms, therefore, typically involve the upper extremity ipsilateral to ictal onset. A dystonic disturbance may briefly persist into the postictal period, so that early voluntary movements of one upper extremity after the seizure is over, such as nose wiping, is also typically ipsilateral to the site of ictal onset. Version and contraversion can occur with head and eye deviation toward, or away from, the site of ictal onset. When a dyscognitive seizure evolves into a secondarily generalized tonic–clonic seizure, however, the head and eye deviation is almost invariably contralateral to the site of ictal onset. Postictal period: Although patients may cough or wipe their nose to indicate the transition from the ictal to the postictal period, usually there is no clear demarcation. In addition to amnesia for the ictal event, patients can exhibit receptive or expressive aphasia if ictal onset is in the language-dominant hemisphere. Reactive automatisms are common during the

855

postictal period and patients should not be restrained, in order to avoid violent responses to perceived threat. Disorientation, fatigue, and headache can dissipate in minutes or can last for hours.

Electroencephalography The interictal electroencephalography (EEG) is characterized by unilateral or independent bilateral anterior temporal spikes, best seen with basal electrodes. Temporal intermittent rhythmic delta activity (TIRDA) usually reflects repetitive spike-andwave discharges in deep structures where the spike component is not visible on scalp recording. The EEG typically does not show ictal epileptiform activity during limbic auras, although frequent interictal spikes will disappear. EEG changes do occur with dyscognitive limbic seizures and commonly consist of variable suppression of baseline rhythms evolving into higher amplitude 5–7 Hz rhythmic activity in one basal electrode, which spreads ipsilaterally and contralaterally. In some cases, the initial epileptiform EEG discharge is bilateral, followed by unilateral, rhythmic activity (Figure 2). Initial ictal discharges recorded from the hippocampus with depth electrodes may take the form of low-voltage fast activity but more commonly consist of hypersynchronous discharges, which are not reflected in the scalp electrodes (Figure 2). During this time, the patient may experience an aura, or there may be no awareness of the seizure. If an aura evolves into a dyscognitive seizure, this is associated with the transition of the hypersynchronous discharge to low-voltage fast activity and propagation to the contralateral hemisphere (Figure 2).

Differential Diagnosis Distinction between limbic seizures of mesial temporal origin and those of frontal lobe origin, particularly orbital frontal and anterior cingulate, can be difficult when neuroimaging is negative. Scalp EEG features can be identical. Although patients may fall with mesial temporal onset seizures, sudden loss of tone producing a drop attack usually indicates a frontal lobe onset. Hyperkinetic and sexual automatisms, and urinary incontinence, can indicate frontal lobe seizures. Frontal lobe seizures are typically briefer than those originating in mesial temporal structures, can have briefer postictal disturbances, and are more likely to be nocturnal. Distinction between limbic seizures of mesial temporal lobe origin and those of frontal limbic or neocortical origin has no practical prognostic or therapeutic value unless seizures are pharmacoresistant and resective surgery is an option.

Mesial Temporal Lobe Epilepsy Clinical Description The most common form of MTLE is MTLE with hippocampal sclerosis (Engel et al., 2008a). More than one type of hippocampal sclerosis exists (Wieser et al., 2004). Typically, habitual seizures of MTLE with hippocampal sclerosis begin in late childhood, before puberty, although they can appear at any age. Almost all patients experience dyscognitive limbic seizures and approximately one-quarter do not experience auras.

856

INTRODUCTION TO CLINICAL BRAIN MAPPING | Temporal Lobe Epilepsy

Figure 2 Examples of EEG telemetry-recorded ictal onsets from four patients with temporal lobe seizures. (a) Low-voltage 6–7 Hz rhythmic activity appears at the right sphenoidal electrode (arrow) 5 s before it is seen over the right temporal convexity. (b) After a diffuse burst of muscle and eye movement artifact, low-voltage fast activity is recorded by the right sphenoidal electrode (arrow). This becomes progressively slower, and the amplitude increases; 5 s later, it is seen diffusely over the right hemisphere. (c) Irregular, sharply contoured slow waves demonstrate phase reversal at the right sphenoidal electrode (arrow) and are reflected as low-amplitude delta waves, without phase reversal, over the right hemisphere. (d) In this lateralized but not localized ictal onset, voltage suppression and low-voltage fast activity occur over the right frontotemporal area and are best seen at the right sphenoidal electrode (arrow). This precedes by 3 s the appearance of diffuse 3 Hz spike-and-wave discharges, which are prominent from the right frontotemporal and sphenoidal derivations. After 10 s, this latter activity evolves into high-voltage 7 Hz rhythmic activity, which phase reverses at the right sphenoidal electrode and laterally at the right anterior to midtemporal region. Patterns shown in (a) and (b) (initial focal onsets) and (d) (delayed focal onset) have more reliable localizing significance than the irregular delta pattern shown in (c). Reproduced from Engel, J., Jr., Crandall, P. H., & Rausch, R. (1983). The partial epilepsies. In R. N. Rosenberg (Ed.), The clinical neurosciences, vol. 2 (pp. 1349-1380). New York: Churchill Livingstone, with permission.

Low voltage fast (p348) 0.5 mV

Hypersynchronous (p367)

0

20

40

60

80

sec

100

Figure 3 Single channels from depth electrode recordings of the hippocampus showing two different seizure onset patterns in two different patients with mesial temporal lobe epilepsy and hippocampal sclerosis. Top channel shows a typical low-voltage fast (LVF) ictal onset, and the lower channel shows a typical hypersynchronous (HYP) ictal onset. Reproduced from Engel (2013), with permission.

Seizure frequency ranges from a few a week to a few a year. There is an increased incidence of family history of epilepsy in these patients, and a history for initial precipitating insults during the first few years of life is commonly elicited, usually consisting of one or more prolonged febrile seizures, but other insults, such as trauma or infections, also can be identified. This suggests a genetic predisposition for an injury, occurring during a window of time during early life, to cause the hippocampal cell loss and synaptic reorganization that underlies the

epileptogenic disturbance responsible for generating spontaneous limbic seizures years later. Patients typically are normal neurologically except for material-specific memory deficits that increase with time. Unilateral, or bilateral, hippocampal atrophy and T2 enhancement are seen in most patients on magnetic resonance imaging (MRI) (Figure 4). Interictal unilateral or bilateral temporal lobe hypometabolism on FDG-PET is invariably present when the MRI is abnormal (Figure 4) and may also occur

INTRODUCTION TO CLINICAL BRAIN MAPPING | Temporal Lobe Epilepsy

857

Figure 4 Flair coronal imaging (a) demonstrates high signal intensity of the left hippocampus (white arrow). T1 coronal imaging (b) shows decreased volume of the left hippocampus (white arrow). FDG-PET and MRI coregistration (c) shows area of hypometabolism in the left temporal lobe (white arrows).

Epidemiology “The Postictal Switch” LT

RT

Interictal

Ictal

Postictal

Austin Hospital Melbourne Figure 5 99Tc-HMPAO SPECT in temporal lobe epilepsy. Interictal, ictal, and postictal SPECT images of the temporal lobes in a patient with left temporal lobe epilepsy. The interictal study shows relatively symmetrical temporal lobe perfusion, with a minor decrease in the left anteromesial temporal region (arrow). The ictal study shows marked left temporal hyperperfusion involving the mesial and especially the lateral temporal regions (arrow). The postictal study shows hyperperfusion of the left mesial temporal region (short arrow) with relative hypoperfusion of the lateral temporal cortex (long arrow). This change from the ictal to the postictal state is the ‘postictal switch.’ Reproduced from Newton, M. R., Berkovic, S. F., Austin, M. C., et al. (1995). SPECT in the localisation of extratemporal and temporal seizure foci. Journal of Neurology, Neurosurgery, and Psychiatry, 59, 26–30, with permission.

before hippocampal sclerosis is evident on MRI. Temporal hypoperfusion on interictal single-photon emission computed tomography (SPECT) is not as localizing as hypometabolism on FDG-PET, but conversion of the hypoperfused region to hyperperfusion with ictal SPECT is pathognomonic for a mesial temporal seizure (Figure 5). More recently, structural and functional white matter patterns demonstrated by diffusion tensor imaging and resting functional MRI (rfMRI) are identifying aberrant connectivity patterns and extrahippocampal damage that may be clinically useful. Magnetoencephalography superimposed on MRI and simultaneous EEG with fMRI are also used to help localize the epileptogenic region for surgery. Material-specific temporal lobe dysfunction can often be determined with neuropsychological testing, and the modified intracarotid amobarbital procedure can further demonstrate that mesial temporal structures of one hemisphere are unable to support memory. Certain personality traits have been ascribed to patients with MTLE, but this association remains controversial. Patients with MTLE do, however, have an increased risk of depression, and some workers have considered paranoid schizophrenia to be more common in patients with limbic seizures.

TLE is the most common form of epilepsy, and most patients with this diagnosis have MTLE with hippocampal sclerosis. However, the true prevalence of MTLE with hippocampal sclerosis is unknown because this diagnosis is usually only definitively made when patients with pharmacoresistant seizures are referred for surgical treatment. No data are available to determine the prevalence of MTLE among patients whose seizures are well controlled, and the majority of patients with poorly controlled MTLE are not referred for surgery (Englot et al., 2012). Hippocampal sclerosis, however, is the most common epileptogenic lesion identified on MRI scans of patients referred to epilepsy centers and is the most likely MRI finding to be associated with pharmacoresistance (Semah et al., 1998). Whereas 40% of patients with epilepsy are reported to have dyscognitive focal seizures (Gastaut et al., 1975), not all patients with dyscognitive focal seizures have MTLE.

Etiology It remains undetermined as to whether hippocampal sclerosis is in fact the initial cause of MTLE or merely a result of recurrent epileptic seizures; however, most likely, both situations apply (Mathern et al., 2008). When the initial precipitating injury is a prolonged febrile seizure, it is unknown whether it is the seizure itself that is responsible for the subsequent hippocampal damage or whether the seizure is merely a manifestation of some other process, such as an indolent cerebral infection, which compromises the hippocampus. There is more than one type of hippocampal sclerosis; the classic form spares CA2, while CA2 cell loss is included in others. Cell loss is not homogeneous, but occurs in a patchy distribution that can be identified using statistical parametric mapping of high-resolution MRI images (Ogren et al., 2009; Figure 6). Such studies appear to distinguish between the classical hippocampal sclerosis with CA2 sparing, which is usually unilateral, and the hippocampal sclerosis with more diffuse cell loss, which is likely to have more contralateral hippocampal damage. Autopsy data reveal some degree of bilateral hippocampal sclerosis in 50% of patients, but it is equal on both sides in only 10% (Armstrong and Bruton, 1987). The most common epileptogenic lesions found in MTLE, other than hippocampal sclerosis, are focal cortical dysplasia, dysembryoplastic neuroepithelioma (DNET), heterotopias, cysts, cavernous malformations, other vascular malformations, and cicatrices induced by trauma, infection, or infarction. The term dual pathology refers to hippocampal sclerosis in association with another lesion, most commonly a

INTRODUCTION TO CLINICAL BRAIN MAPPING | Temporal Lobe Epilepsy

Middle

LVF

Superior aspect

HYP

0.14

0.12

0.10

P value

Inferior aspect

Lateral aspect

(b)

0.06

0.04

Inferior aspect 0.02

medial

0.00

Inferior aspect

Posterior

Lateral

Contralateral LVF

Superior aspect

Superior aspect

Anterior

(a)

Ipsilateral HYP

0.08

858

(c)

Figure 6 (a) Three-dimensional contour map depicting location of the dentate gyrus (DG), hippocampal subfields (CA1-4), and subicular cortex (S) within anterior, middle, and posterior regions of the hippocampal formation based on anatomical landmarks and atlas. (b) Color-coded averaged contour maps indicating areas of significant atrophy (white and red) in the hippocampus ipsilateral and contralateral (c) to seizure onset between patients with hypersynchronous (HYP) and low-voltage fast (LVF) depth EEG ictal onset. Modified from Ogren, J. A., Bragin, A., Wilson, C. L., et al. (2009). Three-dimensional hippocampal atrophy maps distinguish two common temporal lobe seizure-onset patterns. Epilepsia, 50, 1361–1370, with permission.

malformation of cortical development. In this situation, it is not clear whether the hippocampal sclerosis could be a result of the seizures initially caused by the other epileptogenic lesion. A familial form of MTLE is well described (Cendes et al., 2008). Most patients have seizures that are controlled by medication; however, some become pharmacoresistant, develop hippocampal sclerosis, and can be successfully treated with surgery. It is conceivable, therefore, that at least some patients with MTLE and hippocampal sclerosis have a sporadic form of the familial disorder. Given that patients with MTLE and hippocampal sclerosis often have a family history of epilepsy, some cases may fall within the spectrum of epilepsies currently referred to as genetic epilepsies with febrile seizures plus (GEFS þ) (Scheffer and Berkovic, 2008). The pathological substrate of MTLE is not confined to mesial temporal structures. There is evidence suggesting that some degree of bilateral hippocampal involvement is necessary for clinical seizures to occur (Soper et al., 1978), and cortical thickness maps performed on patients with classical MTLE and hippocampal sclerosis reveal large bilateral areas of neocortical atrophy that may also contribute to the occurrence of epileptic seizures, as well as their clinical manifestations, and interictal behavioral disturbances (Lin et al., 2007) (Figure 7).

Differential Diagnosis A family history of epilepsy and initial precipitating injury tends to distinguish MTLE with HS from other forms of MTLE, and MRI is important in identifying and characterizing the epileptogenic lesion (Figure 4). Otherwise, the clinical characteristics, including EEG, treatment, and prognosis, do not depend on the nature of the lesion. Limbic seizures beginning early in childhood usually are not due to hippocampal sclerosis. It is important to distinguish MTLE presenting in childhood from benign epilepsy with centrotemporal spikes (BECTS), a genetic condition with an excellent prognosis (Fejerman, 2008). This differential diagnosis is usually easily made on the basis of the characteristic focal spike-wave discharges of BECTS on EEG. Differential diagnosis between MTLE with seizures that involve brief lapses of consciousness only and childhood and juvenile absence epilepsies is also usually easily made by the characteristic interictal generalized spike-wave EEG patterns of the latter. Differential diagnosis between MTLE with mesial temporal lesions and those with neocortical lesions that preferentially project to mesial temporal structures, or epilepsies with atypical dyscognitive seizures, is important when seizures are pharmacoresistant and resective surgical treatment is a

INTRODUCTION TO CLINICAL BRAIN MAPPING | Temporal Lobe Epilepsy

explosive disorder, and many other conditions, can be easily confused with MTLE. Inpatient video-EEG monitoring is useful when consciousness is impaired, but if seizures do not involve impaired consciousness, a normal ictal EEG does not rule out MTLE.

Mean cortical thickness in control group 5 4 3 2 Mean thickness (mm)

(a)

Left mesial temporal lobe epilepsy Reduction in gray matter thickness

(b)

Same <10% 10% 20% 30% >30% Thinner

Significance >0.05

0.02 <0.05 P-value

(c)

Right mesial temporal lobe epilepsy Reduction in gray matter thickness Same <10% 10% 20% 30% >30%

(d)

859

Thinner

Treatment Initial treatment of MTLE is antiseizure medication, and there are many available drugs to choose from. Decision about medication is based on side effects, cost, ease of administration, and other considerations, in addition to demonstrated effectiveness on clinical trials. Psychological and social intervention is often necessary because of the embarrassing nature of limbic seizures and the associated interictal behavioral disturbances. If epileptic seizures continue to interfere with work, school, or interpersonal relationships after two appropriate trials of antiseizure medication have failed, patients should be referred to an epilepsy center. Most patients with MTLE are excellent candidates for resective surgery. When the epileptogenic region is well localized to one mesial temporal area, disabling seizures can be abolished in 65–85% of patients (Engel et al., 2012; Wiebe et al., 2001). For patients who are not surgical candidates or are not interested in surgery, vagus nerve stimulation and other alternative treatments are available at epilepsy centers.

Prognosis

Significance >0.05

0.02

(e)

<0.05 P-value

Figure 7 Cortical thickness maps: regional reduction in MTLE groups. The mean cortical thickness for controls (N ¼ 19) is shown on a colorcoded scale in (a). Cortical thickness is measured in millimeters as shown in the color bar in which red colors indicate a thicker cortex and blue colors indicate a thinner cortex. The mean reduction in cortical thickness in LMTLE and RMTLE groups as a percent of the control average in (b and d). Red colors in the bilateral frontal poles; frontal operculum; orbital frontal, lateral temporal, and occipital regions; and the right angular gyrus and primary sensorimotor cortex surrounding the central sulcus denote up to 30% decrease in thickness, on average, compared with corresponding areas in controls. The significance of these changes is shown as a map of P values in (c and e). Reproduced from Lin, J. J., Salamon, N., Lee, A. D., et al. (2007). Reduced neocortical thickness and complexity mapped in mesial temporal lobe epilepsy with hippocampal sclerosis. Cerebral Cortex, 17, 2007–2018, with permission.

consideration. In this case, inpatient video-EEG monitoring, neuroimaging, and neurocognitive testing usually identify the site of the epileptogenic region, but in some patients, repeat video-EEG monitoring with depth or subdural grid electrodes is necessary. Nonepileptic seizures, including psychogenic seizures, migraine (particularly in children), psychotic delusions and hallucinations, panic attacks, fugue states, intermittent

Studies suggest that less than half of patients with MTLE become seizure-free with medical treatment (Stephen et al., 2001), and one study found that figure to be as low as 10% (Semah et al., 1998). The true natural history of MTLE with HS, or any other lesion, however, is unknown. Progression of hippocampal atrophy (Cendes, 2005) and memory deficits (Hermann et al., 2006) have been demonstrated, and it is currently believed that early surgical treatment offers the best opportunity to prevent irreversible adverse psychological and social consequences of recurrent seizures, morbidity, and mortality. The ILAE has concluded that, for most patients, failure of two appropriate trials of antiseizure medication is sufficient to warrant considering a patient’s seizures pharmacoresistant for purposes of referral (Kwan et al., 2010).

Lateral Temporal Lobe Epilepsy LTLE rightly belongs under the heading of TLE, and most epileptogenic lesions in the temporal neocortex generate epileptiform discharges that preferentially propagate to mesial temporal structures. Therefore, the ictal semiology is the same as that of MTLE. Initial ictal features that suggest a lateral versus mesial ictal onset include vertigo, auditory phenomena, and multimodal hallucinations suggestive of lesions in the temporal–parietal–occipital region. Differential diagnosis between lateral and mesial epileptogenic regions is only important when surgical resection is considered. Usually, this is obvious from neuroimaging, but when MRI is normal, video-EEG monitoring with depth electrodes is often necessary.

860

INTRODUCTION TO CLINICAL BRAIN MAPPING | Temporal Lobe Epilepsy

Autosomal-dominant LTLE, also called autosomaldominant epilepsy with auditory features, is a well-described genetic form of LTLE (Kahane et al., 2012; Ottman et al., 1995). Onset typically begins in late adolescence and seizures are characterized by elementary or complex auditory auras, although sensory, autonomic, and psychic auras, as well as aphasia, can also occur. Auras evolve into typical dyscognitive limbic seizures and at times secondarily generalized tonic– clonic seizures. Some patients experience only auditory auras, and seizures are triggered by auditory stimuli in a quarter of patients. The EEG and neuroimaging studies suggest left temporal lobe involvement. The genetic defect is a mutation in the leucine-rich glioma-inactivated 1 (LGI1) gene, leading to malfunction of potassium channels or alpha-amino-3 hydroxy-5methyl-4-isoxazoleproprionic acid (AMPA) receptors in approximately half of cases, and this defect can be detected by genetic testing. Seizures are usually easily controlled with medication.

Acknowledgments Original research reported by the author was supported in part by grants NS-02808, NS-15654, NS-33310, and NS-80181.

See also: INTRODUCTION TO ANATOMY AND PHYSIOLOGY: Amygdala; Functional Connectivity; INTRODUCTION TO CLINICAL BRAIN MAPPING: Epilepsy Therapeutics; Focal Cortical Dysplasia; Functional Surgery: From Lesioning to Deep Brain Stimulation and Beyond; Intraoperative Monitoring; Organic Amnesia; Presurgical Assessment for Epilepsy Surgery; INTRODUCTION TO COGNITIVE NEUROSCIENCE: Memory Attribution and Cognitive Control; Semantic Memory: Cognitive and Neuroanatomical Perspectives; Short-Term Memory; The Neural Underpinnings of Spatial Memory and Navigation; Working Memory; INTRODUCTION TO SYSTEMS: Memory; Working Memory.

References Armstrong, D. D., & Bruton, C. J. (1987). Postscript: What terminology is appropriate for tissue pathology? How does it predict outcome? In J. Engel (Ed.), Surgical treatment of the epilepsies (pp. 541–552). New York: Raven Press. Berg, A. T., Berkovic, S. F., Brodie, M. J., et al. (2010). Revised terminology and concepts for organization of seizures and epilepsies: Report of the ILAE Commission on Classification and Terminology, 2005–2009. Epilepsia, 51, 676–685. Blumenfeld, H. (2011). Epilepsy and the consciousness system: Transient vegetative state? Neurologic Clinics, 29, 801–823. Cendes, F. (2005). Progressive hippocampal and extrahippocampal atrophy in drug resistant epilepsy. Current Opinion in Neurology, 18, 173–177. Cendes, F., Kobayashi, E., Lopes-Cendes, I., Andermann, F., & Andermann, E. (2008). Familial temporal lobe epilepsies. In J. Engel & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook (pp. 2487–2493). (2nd ed.). Philadelphia: Lippincott Williams & Wilkins. Commission on Classification and Terminology of the International League Against Epilepsy, (1981). Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia, 22, 489–501.

Engel, J., Jr., Crandall, P. H., & Rausch, R. (1983). The partial epilepsies. In: R. N. Rosenberg (Ed.), The clinical neurosciences (vol. 2, pp. 1349–1380). New York: Churchill Livingstone. Engel, J., Jr., Dichter, M. A., & Schwartzkroin, P. A. (2008). Basic mechanisms of human epilepsy. In J. Engel & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook (pp. 495–507). (2nd ed.). Philadelphia: Lippincott Williams & Wilkins. Engel, J., Jr., Kuhl, D. E., Phelps, M. E., Rausch, R., & Nuwer, M. (1983). Local cerebral metabolism during partial seizures. Neurology, 33, 400–413. Engel, J., Jr., McDermott, M. P., Wiebe, S., et al. (2012). Early surgical therapy for drugresistant temporal lobe epilepsy: A randomized trial. JAMA, 307, 922–930. Engel, J., Jr., & Williamson, P. D. (2008). Limbic seizures. In J. Engel & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook (pp. 541–552). (2nd ed.). Philadelphia: Lippincott-Raven. Engel, J., Jr., Williamson, P. D., & Wieser, H. G. (2008). Mesial temporal lobe epilepsy with hippocampal sclerosis. In J. Engel & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook (pp. 2479–2486). (2nd ed.). Philadelphia: LippincottRaven. Engel, J. Jr. (2013). Seizures and Epilepsy (pp. 706) (2nd ed.). Oxford: Oxford University Press. Englot, D. J., Ouyang, D., Garcia, P. A., Barbaro, N. M., & Chang, E. F. (2012). Epilepsy surgery trends in the United States, 1990–2008. Neurology, 78, 1200–1206. Fejerman, N. (2008). Benign childhood epilepsy with centrotemporal spikes. In J. Engel & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook (pp. 2369–2377). (2nd ed.). Philadelphia: Lippincott-Raven. Gastaut, H., Gastaut, J. L., Goncalves e Silva, G. E., & Fernandez Sanchez, G. R. (1975). Relative frequency of different types of epilepsy: A study employing the classification of the International League Against Epilepsy. Epilepsia, 16, 457–461. Hermann, B. P., Seidenberg, M., Dow, C., et al. (2006). Cognitive prognosis in chronic temporal lobe epilepsy. Annals of Neurology, 60, 80–87. Kahane, P., Bartolomei, F., & Trottier, S. (2012). Temporal lobe epilepsy syndromes. In M. Burean, A. Delgado-Escueta & C. Dravet et al. (Eds.), Epileptic syndromes in infancy, childhood and adolescence. London: John Libbey. Kwan, P., Arzimanoglou, A., Berg, A. T., et al. (2010). Definition of drug resistant epilepsy: Consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Epilepsia, 51, 1069–1077. Lin, J. J., Salamon, N., Lee, A. D., et al. (2007). Reduced neocortical thickness and complexity mapped in mesial temporal lobe epilepsy with hippocampal sclerosis. Cerebral Cortex, 17, 2007–2018. Mathern, G. W., Wilson, C. L., & Beck, H. (2008). Hippocampal sclerosis. In J. Engel & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook (pp. 121–136). (2nd ed.). Philadelphia: Lippincott-Raven. Newton, M. R., Berkovic, S. F., Austin, M. C., et al. (1995). SPECT in the localisation of extratemporal and temporal seizure foci. Journal of Neurology, Neurosurgery, and Psychiatry, 59, 26–30. Ogren, J. A., Bragin, A., Wilson, C. L., et al. (2009). Three-dimensional hippocampal atrophy maps distinguish two common temporal lobe seizure-onset patterns. Epilepsia, 50, 1361–1370. Ottman, R., Risch, N., Hauser, W. A., et al. (1995). Localization of a gene for partial epilepsy to chromosome 10q. Nature Genetics, 10, 56–60. Scheffer, I. E., & Berkovic, S. F. (2008). Generalized (genetic) epilepsy with febrile seizures plus. In J. Engel & T. A. Pedley (Eds.), Epilepsy: A comprehensive textbook (pp. 2553–2558). (2nd ed.). Philadelphia: Lippincott-Raven. Semah, F., Picot, M.-C., Adam, C., et al. (1998). Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology, 51, 1256–1262. Soper, H. V., Strain, G. M., Babb, T. L., Lieb, J. P., & Crandall, P. H. (1978). Chronic alumina temporal lobe seizures in monkeys. Experimental Neurology, 62, 99–121. Stephen, L. J., Kwan, P., & Brodie, M. J. (2001). Does the cause of localisation-related epilepsy influence the response to antiepileptic drug treatment? Epilepsia, 42, 357–362. Wiebe, S., Blume, W. T., Girvin, J. P., & Eliasziw, M. (2001). A randomized, controlled trial of surgery for temporal lobe epilepsy. New England Journal of Medicine, 345, 311–318. Wieser, H.-G., O¨zkara, C¸., Engel, J.Jr., , et al. (2004). Mesial temporal lobe epilepsy with hippocampal sclerosis: Report of the ILAE Commission on Neurosurgery of Epilepsy. Epilepsia, 45, 695–714.