INVITED REVIEW
Mesial Temporal Lobe Epilepsy
William O. Tatum IV
Summary: Temporal lobe epilepsy (TLE) is the most common form of adult
localization-related epilepsy. Hippocampal onset accounts for at least 80% of
all temporal lobe seizures. The electroencephalogram (EEG) of mesial TLE
contains interictal features often associated with anterior temporal epilepti-
form discharges with a maximal voltage over the basal temporal electrodes.
Localized ictal patterns on scalp EEGs characteristically reveal unilateral 5- to
9-Hz rhythmic ictal theta or alpha epileptiform activity maximal in the
anterior temporal scalp electrodes. Invasive-scalp EEG comparisons have
yielded direct information about mesial temporal sources and their corre-
sponding electrical fields. Refinement of macroscopic spatial and the
temporal resolution suggest that a more precise seizure localization may
exist beyond 1- to 35-Hz frequencies observed in routine scalp recording.
Defining the focal areas of ictogenesis within the medial temporal lobe
demonstrates a rich connection to a broad network that goes beyond the
medial structures and even the temporal lobe itself. Advanced electrophys-
iologic application in TLE may further our understanding of ictogenesis to
perfect surgical treatment and to elucidate the neurophysiologic corollaries of
epileptogensis itself.
Key Words: mesial temporal lobe epilepsy, hippocampus, EEG, neurophys-
iology, nocal seizures, epilepsy surgery.
(J Clin Neurophysiol 2012;29: 356–365)
T he temporal lobe is the most epileptogenic region of the human
brain. Temporal lobe epilepsy is a group of disorders that pre-
dominately involves dysregulation of hippocampal function caused
by neuronal hyperexcitability (Schwartzkroin, 1986). Therefore,
mesial TLE (mTLE) is perhaps the best-characterized electroclinical
syndrome of all the epilepsies. The inherent potential for neuronal
tissue of temporal lobe origin to be predisposed to focal seizures is
based on the unique anatomic-functional electrophysiologic net-
works that involve the amygdalohippocampal complex and enter-
orhinal cortex. Anatomically, the hippocampal formation (HF) is
a three-layered allocortex with a lamellar organization that comprises
the dentate gyrus, hippocampus, and subiculum. The hippocampal
complex is divided into the head, body, and tail. It ranges in length
from 4 to 4.5 cm. It is divided into subfields of the Cornu Ammonis
(CA1–CA4), which is surrounded by the dentate gyrus and contains
rich anatomic and functional connections through the parahippocam-
pal gyrus to other temporal and extratemporal cortical association
areas (Coan et al., 2009). Pyramidal cells in the subiculum and
hippocampus possess the capability for neural plasticity through
long-term potentiation and are a crucial element for growth and
development (Cooke and Bliss, 2006). The medial temporal lobe
From the Department of Neurology, Mayo Clinic, Jacksonville, Florida, U.S.A.
Dr. Tatum receives research funding from the Mayo Clinic.
Address correspondence and reprint requests to William O. Tatum IV, DO,
Department of Neurology, Mayo Clinic, 4500 San Pablo Road, Jacksonville,
Florida 32224, U.S.A.; e-mail: tatum.william@mayo.edu.
Copyright Ó 2012 by the American Clinical Neurophysiology Society
ISSN: 0736-0258/12/2905-0356
356 Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012
is crucial for transferring short-term to long-term memory in the
association cortices. It is principally involved with episodic (infor-
mation tied to a time and place) and declarative memory (explicit
memory for facts) for experienced events and for spatial navigation
(Engel, 2003). Patients with extensive damage to the mesial temporal
lobes may become amnestic with an inability to form and retain new
memories (Squire, 2009).
Hippocampal sclerosis (HS) is the most common disease
encountered in epilepsy surgery series and may be readily detected
by brain magnetic resonance imaging (MRI) as mesial temporal
sclerosis. Sclerosis of the hippocampus progresses over time as both
a consequence and a cause of seizures (Theodore et al., 1999). A
normal-appearing hippocampus occurs in 30% to 40% (de Lanerolle
et al., 2003). Hippocampal sclerosis is a combination of atrophy and
astrogliosis of the amygdala, hippocampus, parahippocampal gyrus,
and the enterorhinal cortex and is frequently bilateral (Thom et al.,
2009). If isolated subfield involvement from HS occurs, mesial tem-
poral sclerosis may be absent on neuroimaging results. Other causes
of mTLE that are less frequently seen include gliomas, angiomas,
caveromas, or traumatic or infectious lesions with a dual pathology.
These are seen in 5% to 30%. In drug-resistant mTLE, molecular
mechanisms such as low brain gamma amino butyric acid levels
(Petroff et al., 1996), changes in glutamate levels, and neuronal
glutamate transporters are other mechanisms that are disordered
(Crino et al., 2002). Genetic factors are also seen in mTLE. The famil-
ial forms may or may not have autosomal dominance and are generally
not associated with high-frequency oscillation, although they may be
associated with high-frequency oscillation without seizures in first
degree relatives (Cendes et al., 1998). Other genetic markers, including
SCN1B, seem to share a phenotype with mTLE (Scheffer et al., 2007).
The genetic polymorphism in the ABCB1 gene acts as a multiple drug-
resistance gene (Siddiqui et al., 2003) in patients with mTLE.
CLINICAL FEATURES
Common clinical features of mTLE are well described and are
reproducible in the majority of patients. Febrile seizures are
common, though other causes including head trauma, perinatal
injury, congenital brain malformation, central nervous system
infection, and brain tumors may be evident. Up to two-thirds of
patients with mTLE may have had a febrile seizure before
developing focal seizures (French et al., 1993). When prolonged,
febrile seizures may induce hippocampal edema that progresses to
HS (Scott et al., 2003; Van Landingham et al., 1998). The onset of
mTLE characteristically occurs at the end of the first or second
decade. Focal seizures with and without impaired consciousness
are the most common seizure types in mTLE. Status epilepticus
and prolonged seizures are infrequently encountered.
The ictal semiology may exhibit a wide range of heterogene-
ity. Auras frequently appear independently when seizures begin early
in life (Villanueva and Serratosa, 2005) and .80% of patients report
Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012
an aura with experiential and viscerosensory symptoms. Psychic
phenomenon commonly include anxiety, deja vu, and fear, and in
addition viscerosensory auras with a nausea, “butterflies,” or rising
indescribable sensation from the epigastrium commonly occur
(French et al., 1993; Thompson et al., 2000). Focal seizures with
impaired consciousness (aka complex partial seizures) in mTLE are
the primary seizure type. Motor manifestations are more common in
children before 6 years of age (Fogarasi et al., 2002). The validation
of the clinical semiology of mTLE focal seizures has been best
defined through epilepsy surgery that results in seizure freedom
verifying the localization (Engel, 1993). Focal seizures are often
hypomotor or automotor with limited motor movement and a fixed
motionless stare. Additionally, oral and manual automatisms with
impaired consciousness for 30 to 60 seconds frequently accompany
the symptoms. Pupillary dilatation, hyperventilation, piloerection,
and tachycardia are common autonomic features. Dystonic posturing
contralateral to the hemisphere of seizure origin with ipsilateral
automatisms during the seizure are reliable lateralizing signs; how-
ever, propagation of the ictal activity to symptomatic regions may
occur and limit the usefulness of seizure semiology as a sole means
of localization (So, 2006). Ictal speech, vomiting, and intermittent
“responsiveness” suggest nondominant lateralization and postictal
aphasia in the dominant temporal lobe. Tonic or clonic jerking and
postictal Todd paresis lateralize to the contralateral hemisphere
implicating propagation outside the mesial temporal lobe to involve
the ipsilateral motor neocortex. Seizure semiology is often nonspe-
cific in reflecting disease; however, HS may be more associated with
dystonic posturing. Focal seizures evolving to convulsions (aka sec-
ondarily generalized) are relatively infrequent and are usually con-
trolled with antiepileptic drugs (AEDs) (French et al., 1993).
Electrophysiology of Mesial Temporal
Lobe Epilepsy
The EEG representation in human mTLE is based on the
limited sampling of the brain involving pyramidal cell neurons
located at the gyral crests within the basal–lateral neocortex. Tem-
poral lobe epilepsy is a common cause of EEGs containing interictal
epileptiform discharges (IEDs) (Table 1). Interictal epileptiform dis-
charges and focal seizures arise from large numbers of neurons
bursting in a synchronous fashion disrupting operation of normal
neuronal activity (Kibler and Durand, 2011). Waveforms that
TABLE 1. Clinical Features That Lead to a Greater Detection
of IEDs on Routine Scalp EEG
Temporal lobe epilepsy
Earlier age of seizure onset
Children
Specific epilepsy syndrome (LGS, BCECTS, CAE, West/Landau–Kleffner
syndrome)
Size, depth, and orientation of SOZ
Higher seizure frequency
Sleep
,24 Hours after a seizure
AEDs without IED suppression
Number/type of EEG electrodes
Multiple/prolonged EEG recording
AEDs, antiepileptic drugs; BCECTS, benign childhood epilepsy with centrotem-
poral spikes; CAE, childhood absence epilepsy; LGS, Lennox–Gastaut syndrome.
Adapted from Pillai and Sperling (2006).
Copyright Ó 2012 by the American Clinical Neurophysiology Society
Mesial Temporal Lobe Epilepsy
become apparent are then influenced by other intracortical and inter-
cortical networks. Scalp interictal EEG is represented by synchro-
nous oscillations of large pools of neurons collectively summating
postsynaptic potentials. For detection, a sufficient synchrony must
exist, and the cortical region must reside in close proximity to
a recording electrode. Repetitive IEDs, rhythmic frequencies with
sinusoidal waveforms, and complex waveforms or fields in epilepsy
arise depending on the location of the generator and the networks
involved in propagation of the ictal activity. Sustained rhythmic po-
tentials that occur during a burst of epileptiform activity vary in fre-
quency, morphology, amplitude, duration, and electrocerebral field.
At the neuronal level, the basis for focal interictal epileptiform
abnormalities involves the genesis of a single set of paroxysmal
depolarizing shifts. Abnormal neurons produce repetitive action
potentials at rates exceeding normal maximal rates, which results in
prolonged depolarization from sudden shifts in the cell membrane
terminated with an after-going hyperpolarization. When an electro-
physiologic threshold of summated apical dendrites in an involved
neuronal population is reached, the scalp or invasive EEG (iEEG)
detection of a paroxysmal depolarizing shift is translated into visible
IEDs (i.e., spike-or sharp-and-slow-wave). When repetitive neighbor-
ing paroxysmal depolarizing shifts are synchronous and become
continuous, focal electrographic seizures are evident (Buzsaki and
Draguhn, 2004). Intrinsic currents from ion channels of radially
oriented dendrites of pyramidal cells project to the surface for EEG
source localization.
SCALP INTERICTAL ELECTROENCEPHALOGRAM
Electroencephalograms are often associated with anterior
temporal spikes or sharp waves with voltage that is typically
maximal in the anterior temporal regions in .90% of patients with
mTLE (Fig. 1) (Williamson et al., 1993). Midtemporal epileptiform
discharges may occasionally occur in mTLE, but consistent midtem-
poral EDs should increase the possibility of a larger, or extramesial
temporal generator. Approximately one-third of patients have bilat-
eral temporal IEDs, which become apparent with long-term EEG
monitoring (Chung et al., 1991). Mesial TLE may also be associated
with slow waves that have localizing value. Temporal intermittent
rhythmic delta activity is seen in a minority of patients (Fig. 2A) but
is associated with TLE in up to 80% (Geyer et al., 1999). Postictal
slowing frequently correlates with the side of seizure onset, though it
may arise after seizure propagation and is less specific. The location
of the scalp field maximum of frontotemporal IEDs has also been
identified by iEEG (Ebersole and Wade, 1991). Two different pop-
ulations have been defined in TLE. Type 1 possesses an inferior
temporal electographic maximum with vertex positivity. Type 2 is
generated by a lateral temporal maximum. Dipole modeling demon-
strates either an elevated or vertical orientation for IEDs generated in
the basal temporal lobe or a horizontal or radial orientation for those
generated in the lateral neocortex (Ebersole, 1991). With time, there
is fluctuation in the primary spike topography (Fig. 3). Therefore,
characterizing spike voltage fields is a dynamic process when at-
tempting to accurately represent the source. Furthermore, spikes
originating in the basal temporal neocortex (from SP1/SP2, T1/T2,
FT9/FT10 electrodes) may propagate to other regions, including the
lateral temporal neocortex and temporal pole. Projection of the sur-
face negativity anterior to the frontotemporal (detected at F9/F10) or
to the frontopolar derivations (detected at FP1/FP2) depends on the
orientation and projection of the dipole from its source. Scalp-based
anterior temporal IEDs have also been found to be generated by the
357
W. O. Tatum Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012

FIG. 1. A, Scalp electroencephalogram (EEG) demonstrating an anterior-mesial interictal epileptiform discharge (IED) in
temporal lobe epilepsy (TLE) with an anterior field. Note the arrows depicting the field. B, SISCOM (subtraction ictal SPECT
coregistered on MRI) during presurgical evaluation of a patient before left amygdalaohippocampectomy with anterior tip
resection. The disease was nonspecific gliosis. The patient is seizure-free for 6 months. The case and SISCOM images appeared on
www.epilepsy.com7/2011.

lateral temporal neocortex during simultaneous scalp-subdural
recording (Sperling and Engel, 1986). This has been reproduced
when evaluated by simultaneous magnetoencephalography and scalp
recording (Sutherling and Barth, 1989).
The use of additional electrodes may help improve localiza-
tion, especially subtemporal electrodes such as F9/10, T9/10, and P9/
10 (Binnie et al., 1989; Ebersole and Wade, 1991; Kobayashi et al.,
2000). Sphenoidal electrodes may be useful in improving the source
localizing precision of IEDs in mTLE (Fig. 4). Electroencephalo-
graphic source localization has been found to deviate by as much
as 2 cm when sphenoidal electrodes were not used, reflecting the
need for basal temporal neocortical representation in scalp EEG
(Hamaneh et al., 2011). Because of their close proximity to the
foramen ovale adjacent to the mesiobasal temporal lobe, SP electro-
des often demonstrate higher amplitude with better signal-to-noise
ratios. Interictal epileptiform discharges originating in the amygda-
lohippocampal complex are isolated from scalp recording, because
of their curvilinear electrophysiologic field in the anterior–posterior
direction and small region of involvement. “Mesial” temporal IEDs
on scalp EEG truly reflect basal temporal propagation from larger
areas including enterorhinal and fusiform basal cortices. Scalp EEG
typically detects only a small fraction of the IEDs originating in the
mesial temporal structures (Fernandez Torre et al., 1999). Spike
frequency has therefore been believed to be a biomarker for patients
with mTLE and predicts a worse surgical outcome by some by virtue
of reflecting greater extrahippocampal activity (Krendl et al., 2008).
Interictal epileptiform discharges that occur during wakefulness and
rapid eye movement sleep are less frequent, though they may pro-
vide better localizing information and are more often concordant
with the seizure onset zone (Sammaritano et al., 1991).
SCALP ICTAL ELECTROENCEPHALOGRAM
Seizures represent complex waveforms on EEG with sus-
tained rhythmic discharges that evolve in frequency, morphology,
and electrocerebral field. Focal unilateral anterior temporal regular
and rhythmic temporal theta or alpha activity (typically 5–9 Hz) is
the hallmark of mTLE, when it occurs after the appropriate clinical
semiology and seems stable over the lifespan (Fig. 5). Risinger
reported this to occur in up to 94% of patients with mTLE and
correctly lateralized the seizure onset in 95% of the patients
(Risinger et al., 1989). Less commonly, a homologous or similar
parasagittal positivity that is maximal at the vertex may also be seen.
Alternatively, a combination of ictal rhythm localizations may be
seen depending on the precise direction and conduction of the dipole.
When this mixed pattern occurs, it is less specific for mTLE and may
occur during the seizure in patients without hippocampal TLE
(Gil-Nagel and Risinger, 1997). Focal seizures that propagated from
extratemporal or neocortical structures to the mesial temporal lobe
structures may explain the reduced specificity seen in patients with-
out HFA representing pseudo-TLE. Still, patients without HFA and

358 Copyright Ó 2012 by the American Clinical Neurophysiology Society

Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012 Mesial Temporal Lobe Epilepsy

FIG. 2. A, A comparison of basal temporal electrodes in a patient with mesial temporal lobe epilepsy (mTLE). Note the highest
amplitude in the sphenoidal (arrow) and the left true temporal electrode in the Pz reference in the lower portion of the tracing.
SP1/2 ¼ sphenoidal electrodes. B, Four panels of “source brains” with fitted dipole models demonstrating the variability without
and with sphenoidal electrode use for source localization of interictal spikes in patients with temporal lobe epilepsy (TLE).
Borrowed from Hamaneh et al. (2010) with permission.

a normal brain MRI that have seizures associated with a rhythmic
ictal theta discharge at seizure onset may still predict a favorable
response to mesial temporal lobe resection without iEEG (Tatum
et al., 2008). Focal seizures without impaired consciousness (aka
simple partial seizures) do not have the necessary recruitment of
the surrounding tissue to permit detection by the scalp in 70% of
the patients (Kapur et al., 1995; Sperling and O’Connor, 1990).
Event detection can be performed only after large volumes of the
neocortex have been recruited from a small original generator and
have spread to superficial sites of the brain that are able to be
detected at the scalp. Frequency characteristics during focal seizures
suggest mTLE if the seizure onset occurs within 30 seconds before
the clinical seizure onset when there is rhythmic ictal theta onset that
is $5 Hz (Risinger et al., 1989). However, using routine montages,
IEDs may not appear on the scalp if there is ,10 cm2 of the cortex
involved (Tao et al., 2005). Furthermore, the use of the 10 to 20
system of electrode placement may limit ictal recordings. A greater
number of electrodes in the form of high-density EEG may be able to
overcome propagation patterns that may lead to erroneous source
localization (Lantz et al., 2003). Simultaneous intracranial and sur-
face ictal EEG recordings (64 total channels) obtained from a com-
bination of invasive electrodes (subdural strips and intracerebral
depths) and scalp electrodes, demonstrated a difference in the types
of TLE (Pacia and Ebersole, 1997). When the ictal rhythm demon-
strates hippocampal involvement alone, no scalp EEG rhythms will
be detected until the recruitment of the inferolateral temporal
neocortex occurs. Subsequently, a regular 5- to 9-Hz subtemporal
and temporal rhythmic ictal theta pattern can be noted on scalp EEG
recordings. However, when the ictal discharge is confined to the
mesiobasal temporal cortex, a vertex dominant rhythm can be noted
in association with the vertical orientation of the dipole. Finding
widespread bilateral background attenuation, which corresponds to
a 20- to 40-Hz discharge on iEEG, and a pattern composed of irreg-
ular 2- to 4-Hz focal temporal delta ictal activity on scalp EEG at
onset are more likely to be associated with seizures of neocortical
origin. Scalp EEG patterns of temporal lobe seizures are not a direct
reflection of the ictal activity at seizure onset. Rather, they reflect
the differences in seizure development, propagation, and synchrony
of the direct cortical discharges (Pacia and Ebersole, 1997). The best
surgical outcomes are derived when 100% unilateral temporal IEDs
are combined with scalp ictal EEGs that remain regionalized with-
out contralateral propagation (Table 2) (Schulz et al., 2000). Tran-
sient epileptic amnesia is an ictal–postictal episode with clinical
manifestations that involves dysfunction of the hippocampus
bilaterally (Fig. 5B). Transient epileptic amnesia may not be
clearly demonstrable with scalp electrodes in some patients (Pal-
mini et al., 1992). Other cases of transient epileptic amnesia seem
more heterogeneous with recruitment of the mesial structures by
the parahippocampal neocortex and mesial temporal postictal
slowing identified by depth electrode recording (Walsh et al.,
2011). Ictal rhythms with a predominant basal source component
have hippocampal onset seizures, while those with anterior temporal

Copyright Ó 2012 by the American Clinical Neurophysiology Society 359

W. O. Tatum Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012

FIG. 3. Figure depicting inverse modeling in electroencephalogram (EEG) source localization (ESL). The scalp-recorded EEG field is in
the upper left-hand corner. The midleft graphic depicts the corresponding incremental field. The lower left represents the independent
component analysis field maps and the changing mean global contributions over time. Dipole and distributed ESL for single versus
averaged interictal epileptiform discharges (IEDs) in mesial temporal lobe epilepsy (mTLE) are modeled. Rotating sources are represented
in green and moving sources in blue. The 3 rows represent the onset, recruitment, and propagation at the beginning, midupswing, and
peak of the IED electronegativity. Standardized low-resolution electromagnetic tomography (sLORETA) constrained to cortex using
rotating sources is mapped with a boundary element method (BEM). Borrowed with permission from Plummer et al. (2010).

360 Copyright Ó 2012 by the American Clinical Neurophysiology Society

Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012 Mesial Temporal Lobe Epilepsy

FIG. 4. Localization with slow waveforms. In (A) temporal intermittent delta activity is present bilaterally in a patient with temporal
lobe epilepsy (TLE). B, Postictal slowing after a left temporal focal seizure in a patient with left surgically proven mesial temporal lobe
epilepsy (mTLE).

source maximum have enterorhinal onset and those with lateral
source maximum have neocortical onset (Assaf and Ebersole,
1997). Regional postictal slowing may appear in up to 70% of
patients with TLE (Fig. 2B) and may be a useful localizing sign
(Bowman et al., 2010).
INVASIVE ELECTROENCEPHALOGRAM
In some instances, the seizure onset zone in mTLE cannot be
appreciated, localized by scalp recordings alone, or it bears
discordant information preventing localization. Scalp EEG alone
has a limited capability to solve the inverse problem of source

FIG. 5. Typical scalp ictal electroencephalogram (EEG) with the characteristic onset of unilateral 7-Hz rhythmic temporal theta
before clinical seizure onset of a focal seizure that evolves from an aura of “deja vu” to a focal seizure with dyscognitive features.
The patient has been seizure-free off antiepileptic drugs (AEDs) for .2 years.

Copyright Ó 2012 by the American Clinical Neurophysiology Society 361

W. O. Tatum Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012

(Weinand et al., 2001). Other ictal features included a loss of inter-

TABLE 2. Interictal EEG Features and Ictal Scalp EEG
ictal spike or sharp wave discharges, which may challenge recogni-
Propagation Patterns and Percentage That Are Seizure-Free
tion of the pattern of ictal transformation. Propagation patterns after
After Resective Surgery for mTLE at 1 Year
seizure onset are usually slower and of higher amplitude than ictal
Propagation and scalp EEG Seizure-free outcome onset frequencies potentially limiting scalp reflection in patients with
(n ¼ 58; 347 seizures) (n ¼ 49) at 1 year mTLE (Fig. 6). Mesial–mesial and mesial–lateral temporal projec-
100% Unilateral temporal IEDs 84.6% tions are the localized regions of seizure spread after onset while
,100% Unilateral temporal IEDs 52.2% (P ¼ 0.015) dorsolateral frontal, orbitofrontal, lateral temporal, parietal, and
Regionalized EEG without 82.8% occipital onsets usually propagate to the ipsilateral temporal lobe
contralateral propagation (Jenssen et al., 2011). A nonlateralizing ictal EEG pattern often
Regionalized EEG with 45.5% (P ¼ 0.007) results in a poor postsurgical outcome in nonlesional TLE (Schulz
contralateral propagation et al., 2000). Similar to scalp recording, iEEG has used many forms
100% Unilateral temporal 88.9% of invasive electrodes (i.e., foramen ovale) that approximate the
IEDs 1 regionalized EEG mesial temporal lobes when bilateral seizure scalp ictal onset in
1 of 2 Variables 73.7%
TLE requires further clarification (Fig. 7). Long interhemispheric
,100% Unilateral temporal 33.3% (P ¼ 0.007)
propagation times are associated with good surgical outcomes (King
IEDs 1 regionalized EEG
and Spencer, 1995). Infrequently false lateralization on scalp EEG
with contralateral propagation
occurs. Bitemporal independent ictal onsets are noted in some
Adapted from Schulz et al. (2000).
patients with TLE. In a significant number of these patients, depth
records may be able to isolate a single active hippocampus for suc-
cessful surgical resection suggesting typical and atypical scalp EEG
propagation patterns from a single focus (Henry et al., 1999). Rarely,
identification. Using depth electrodes placed in the hippocampus, some temporal lobe seizures may involve rapid propagation with
IEDs were found to be asynchronous with ipsilateral temporal lobe generalized or diffuse bilateral onset reflecting large cortical volume
IEDs on scalp-sphenoidal recording (Tao et al., 2005). Furthermore, recruitment by the time of detection and reflecting superficial sites of
iEEG-scalp EEG comparison demonstrated the ability of iEEG to neocortical origin.
detect 25% more discharges and also discharges that have a broader The conventional range of scalp EEG analysis involves
distribution than scalp EEG (Gloor, 1991). Progressive rhythmic frequencies of 1 to 35 Hz. Atypical frequencies include both ultrafast
EEG activity was characteristic of spontaneous focal seizures in and ultraslow frequencies (Fig. 8) beyond the resolution of scalp
humans (Geiger and Harner, 1978). Focal hypersynchronous dis- EEG that may yield more specific localizing information with
charges were the most common patterns seen in approximately implications in epileptogenesis. Recently, depth electrodes and both
50% of patients and were an accurate localizing indication of cortical visual and spectral analyses have demonstrated interictal and ictal
irritability (Geiger and Harner, 1978). When ictal EEG precedes focal high-frequency oscillations in localization-related epilepsy
clinical seizure onset by .10 seconds, a greater accuracy in antici- (Ramachandran nair et al., 2008; Worrell et al., 2008). Physiologic
pating seizure freedom after amygdalohippocampectomy is expected frequencies may occur around 100 to 200 Hz. This bandwidth of

FIG. 6. A, Depth electrodes demonstrating subclinical seizures in the depth (red arrow) without surface detection (lower
channels). B, Subclinical status epilepticus with seizure durations of up to 20 minutes. The patient was without any observable
clinical signs with memory improvement after left amygdalohippocampectomy.

362 Copyright Ó 2012 by the American Clinical Neurophysiology Society

Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012 Mesial Temporal Lobe Epilepsy

FIG. 7. Electroencephalogram (EEG) recorded with foramen ovale electrodes and surface-scalp electrodes. Note the early
onset demonstrated in the left foramen ovale electrode as a high-frequency discharge that becomes noted on the scalp EEG
18 seconds after the onset. Courtesy of Stefan Rampp, MD, Department of Neurology, Epilepsy Center, University Hospital
Erlangen, Schwabachanlage 10, 91054 Erlangen, Germany.

FIG. 8. Intracranial
electroencephalogram (EEG)
demonstrating the sites of maximal
ictal baseline shift at seizure onset with
an “open” high-pass filter. The DC
shift is shown by the circle during
a right temporal lobe seizure (arrow).
T ¼ temporal; O ¼ occipital; letters ¼
electrodes with nonoperational
channels TC2D-F. Note the circle with
the lowest amplitudes that occur at
seizure onset. Courtesy of Stefan
Rampp, MD, Department of
Neurology, Epilepsy Center,
University Hospital Erlangen,
Schwabachanlage 10, 91054
Erlangen, Germany.

Copyright Ó 2012 by the American Clinical Neurophysiology Society 363

W. O. Tatum Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012
frequencies is generated in the hippocampus and enterorhinal cortex
and may reflect normal cognitive processing and the transfer of
information underlying memory tasks (Chrobak and Buzsaki,
1998). Ripples (80–250 Hz) and fast ripples (250–500 Hz) are fre-
quencies beyond the recording limits of routine scalp recording.
When they are regionalized or widespread, they may appear as an
electrodecremental response on scalp EEG. Ultrahigh frequencies
(250–500 Hz) are abnormal and seem to be generated by hypersyn-
chronous pyramidal cell bursts in the epileptogenic mesial temporal
lobe (Bragin et al., 1999). Discrete high-frequency oscillations occur
mainly in the seizure onset zone and not within areas of propagation
in both the mesial temporal and neocortical temporal lobes on
iEEG (Fisher et al., 1992; Jacobs et al., 2009) and interictal periods
(Worrell et al., 2004). These high-frequency oscillations may be
present during the immediate preictal period at the region of seizure
onset in mTLE and rarely in the regions of propagated or nonlocal-
ized seizures (Jacobs et al., 2009).
IMPLICATIONS FOR TREATMENT
More than 60% of patients with focal seizures will achieve
seizure freedom from AED (Kwan and Brodie, 2000). It is the failure
of the first AED to effectively control the seizures that is the most
powerful predictor of drug-resistance (Kwan and Brodie, 2000), with
,10% of patients becoming seizure-free after 2 AEDs fail to control
the seizures (Kwan et al., 2010). When strict unilateral temporal
IEDs are evident on scalp EEG and ipsilateral hippocampal forma-
tion atrophy is present on MRI, a high concordance with seizure
onset is found leading some to question the need for ictal recordings
in these patients (Cendes et al., 2000). This raises the question of
whether patients should have epilepsy surgery discussed during ini-
tial treatment, when the MRI demonstrates mesial temporal sclerosis
and the EEG reveals IEDs in the corresponding area.
Observational studies suggest that when seizure freedom does
occur with the introduction of new AEDs, it may last up to $12
months, but only in a limited number of refractory patients (Luciano
and Shorvon, 2007). Glutamatergic preictal discharges have emerged
in slice preparations before ictal transition (Huberfeld et al., 2011).
Newer generation AEDs have thus far had similar inefficacy when
drug resistance is established. Perhaps, seizure prediction will facil-
itate early, more specific, AED treatment in the future that is more
effective in limiting the preictal to ictal transformation.
When AEDs are ineffective resective epilepsy surgery is
considered. Temporal lobectomy performed for drug-resistant focal
seizures represents two-thirds of the surgical neurosurgical proce-
dures. Hippocampal sclerosis has the worst prognosis for medical
control with only 11% to 42% of patients becoming seizure-free
(Semah et al., 1998; Stephen et al., 2001). Surgical outcome approaches
$70% and is the prototype for surgically remediable epilepsy syn-
dromes (Engel, 1996). Among many uncontrolled studies, a single
1-year randomized controlled trial performed in Canada, demon-
strated superiority of surgical treatment in TLE over continued med-
ical treatment (58% vs. 3% seizure-free) once drug resistance is
established (Wiebe et al., 2001). A meta-analysis of surgical out-
come further demonstrated that 2 years after surgery 55% of refrac-
tory patients are seizure-free and 68% are free from complex partial
seizures (Schmidt and Stavem, 2009). Surgical excision is curative
in a significant number of patients with drug-resistant mTLE, espe-
cially when a structural lesion is present. Lesioning sections of the
HF have demonstrated the ability to block synchronous propagation
of epileptiform abnormalities (Derchansky et al., 2006). In a rodent
364
model, even a selective transaction of the CA3 region alone is
capable of significantly decreasing or eliminating transhippocampal
propagation of epileptiform activity (Kibler and Durand, 2011). Neu-
rostimulation is a form of therapy that may be considered when
surgery is not an option. Despite the surge of new AEDs, the greater
availability of resective epilepsy surgery, and the resurfacing of the
ketogenic diet, neurostimulation maintains a unique role in the treat-
ment of refractory mTLE. Vagus nerve stimulator is not a curative
device with ,5% that become seizure-free. A mean reduction in
seizure frequency of 25% to 28% at 3 months, improves to approx-
imately 40% by year 1 (Handforth et al., 1998). Other forms of
intracranial neurostimulation include thalamic deep brain stimulation
and the responsive neurostimulator. The responsive neurostimulator/
latter uses depth or subdural leads placed at 1 to 2 predetermined
seizure foci. Neurostimulators are designed to “learn” an individual’s
abnormal electrocorticographic activity and respond by delivering
preprogrammed electrical stimulation to the regions of the brain
implanted to reduce seizure frequency (Morrell, 2011).
CONCLUSIONS
The inherent potential for neuronal tissue of the mesial
temporal lobe to be predisposed to focal seizures is based on the
unique anatomic-functional electrophysiologic networks that involve
the amygdalohippocampal complex and enterorhinal cortex. Scalp
EEG possesses several significant limitations; however, it is the
foundation for localization in mTLE for surgical therapy. Invasive
EEG may become necessary during presurgical evaluations to
lateralize or localize the seizure’s focus or function in one temporal
lobe. Because clinical neurophysiology expands to incorporate ultra-
high frequencies (250–500 Hz) and functional techniques coupled
with EEG, greater spatial-temporal resolution will provide more
powerful localizing diagnostic and therapeutic options for patients
with drug-resistant mTLE.
REFERENCES
Assaf BA, Ebersole JS. Continuous source imaging of scalp ictal rhythms in
temporal lobe epilepsy. Epilepsia 1997;38:1114–1123.
Binnie CD, Marston D, Polkey CE, Amin D. Distribution of temporal spikes in
relation to the sphenoidal electrode. Electroencephalogr Clin Neurophysiol
1989;73:403–409.
Bowman R, Gibson G, Padgett M. Particle tracking stereomicroscopy in optical
tweezers: control of trap shape. Opt Express 2010;18:11785–11790.
Bragin A, Engel J Jr, Wilson CL, et al. High-frequency oscillations in human
brain. Hippocampus 1999;9:137–142.
Buzsaki G, Draguhn A. Neuronal oscillations in cortical networks. Science
2004;304:1926–1929.
Cendes F, Lopes-Cendes I, Andermann E, Andermann F. Familial temporal lobe
epilepsy: a clinically heterogeneous syndrome. Neurology 1998;50:554–557.
Cendes F, Li LM, Watson C, et al. Is ictal recording mandatory in temporal lobe
epilepsy? Not when the interictal electroencephalogram and hippocampal
atrophy coincide. Arch Neurol 2000;57:497–500.
Chrobak JJ, Buzsaki G. Operational dynamics in the hippocampal–entorhinal axis.
Neurosci Biobehav Rev 1998;22:303–310.
Chung MY, Walczak TS, Lewis DV, et al. Temporal lobectomy and independent
bitemporal interictal activity: what degree of lateralization is sufficient? Epi-
lepsia 1991;32:195–201.
Coan AC, Appenzeller S, Bonilha L, et al. Seizure frequency and lateralization
affect progression of atrophy in temporal lobe epilepsy. Neurology
2009;73:834–842.
Cooke SF, Bliss TV. Plasticity in the human central nervous system. Brain
2006;129:1659–1673.
Crino PB, Jin H, Shumate MD, et al. Increased expression of the neuronal gluta-
mate transporter (EAAT3/EAAC1) in hippocampal and neocortical epilepsy.
Epilepsia 2002;43:211–218.
de Lanerolle NC, Kim JH, Williamson A, et al. A retrospective analysis of hip-
pocampal pathology in human temporal lobe epilepsy: evidence for distinc-
tive patient subcategories. Epilepsia 2003;44:677–687.
Copyright Ó 2012 by the American Clinical Neurophysiology Society
Journal of Clinical Neurophysiology
Volume 29, Number 5, October 2012
Derchansky M, Rokni D, Rick JT, et al. Bidirectional multisite seizure propagation
in the intact isolated hippocampus: the multifocality of the seizure “focus”.
Neurobiol Dis 2006;23:312–328.
Ebersole JS. EEG dipole modeling in complex partial epilepsy. Brain Topogr
1991;4:113–123.
Ebersole JS, Wade PB. Spike voltage topography identifies two types of fronto-
temporal epileptic foci. Neurology 1991;41:1425–1433.
Engel J Jr. Update on surgical treatment of the epilepsies. Summary of the Second
International Palm Desert Conference on the Surgical Treatment of the Epi-
lepsies (1992). Neurology 1993;43:1612–1617.
Engel J Jr. Surgery for seizures. N Engl J Med 1996;334:647–652.
Engel J Jr. A greater role for surgical treatment of epilepsy: why and when?
Epilepsy Curr 2003;3:37–40.
Fernandez Torre JL, Alarcon G, Binnie CD, et al. Generation of scalp discharges
in temporal lobe epilepsy as suggested by intraoperative electrocorticographic
recordings. J Neurol Neurosurg Psychiatry 1999;67:51–58.
Fisher RS, Webber WR, Lesser RP, et al. High-frequency EEG activity at the start
of seizures. J Clin Neurophysiol 1992;9:441–448.
Fogarasi A, Jokeit H, Faveret E, et al. The effect of age on seizure semiology in
childhood temporal lobe epilepsy. Epilepsia 2002;43:638–643.
French JA, Williamson PD, Thadani VM, et al. Characteristics of medial temporal
lobe epilepsy: I. Results of history and physical examination. Ann Neurol
1993;34:774–780.
Geiger LR, Harner RN. EEG patterns at the time of focal seizure onset. Arch
Neurol 1978;35:276–286.
Geyer JD, Bilir E, Faught RE, et al. Significance of interictal temporal lobe delta
activity for localization of the primary epileptogenic region. Neurology
1999;52:202–205.
Gil-Nagel A, Risinger MW. Ictal semiology in hippocampal versus extrahippo-
campal temporal lobe epilepsy. Brain 1997;120:183–192.
Gloor P. Preoperative electroencephalographic investigation in temporal lobe epi-
lepsy: extracranial and intracranial recordings. Can J Neurol Sci 1991;18
(4 suppl):554–558.
Hamaneh MB, Limotai C, Luders HO. Sphenoidal electrodes significantly
change the results of source localization of interictal spikes for a large
percentage of patients with temporal lobe epilepsy. J Clin Neurophysiol
2011;28:373–379.
Handforth A, DeGiorgio CM, Schachter SC, et al. Vagus nerve stimulation ther-
apy for partial-onset seizures: a randomized active-control trial. Neurology
1998;51:48–55.
Henry TR, Ross DA, Schuh LA, Drury I. Indications and outcome of ictal record-
ing with intracerebral and subdural electrodes in refractory complex partial
seizures. J Clin Neurophysiol 1999;16:426–438.
Huberfeld G, Menendez de la Prida L, Pallud J, et al. Glutamatergic pre-ictal
discharges emerge at the transition to seizure in human epilepsy. Nat Neuro-
sci 2011;14:627–634.
Jacobs J, Zelmann R, Jirsch J, et al. High frequency oscillations (80–500 Hz) in the
preictal period in patients with focal seizures. Epilepsia 2009;50:1780–1792.
Jenssen S, Roberts CM, Gracely EJ, et al. Focal seizure propagation in the intra-
cranial EEG. Epilepsy Res 2011;93:25–32.
Kapur J, Pillai A, Henry TR. Psychogenic elaboration of simple partial seizures.
Epilepsia 1995;36:1126–1130.
Kibler AB, Durand DM. Orthogonal wave propagation of epileptiform activity in
the planar mouse hippocampus in vitro. Epilepsia 2011;52:1590–1600.
King D, Spencer S. Invasive electroencephalography in mesial temporal lobe
epilepsy. J Clin Neurophysiol 1995;12:32–45.
Kobayashi K, James CJ, Yoshinaga H, et al. The electroencephalogram
through a software microscope: non-invasive localization and visualization
of epileptic seizure activity from inside the brain. Clin Neurophysiol
2000;111:134–149.
Krendl R, Lurger S, Baumgartner C. Absolute spike frequency predicts surgical
outcome in TLE with unilateral hippocampal atrophy. Neurology
2008;71:413–418.
Kwan P, Arzimanoglou A, Berg AT, et al. Definition of drug resistant epilepsy:
consensus proposal by the ad hoc Task Force of the ILAE Commission on
Therapeutic Strategies. Epilepsia 2010;51:1069–1077.
Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med
2000;342:314–319.
Lantz G, Spinelli L, Seeck M, et al. Propagation of interictal epileptiform activity
can lead to erroneous source localizations: a 128-channel EEG mapping
study. J Clin Neurophysiol 2003;20:311–319.
Luciano AL, Shorvon SD. Results of treatment changes in patients with apparently
drug-resistant chronic epilepsy. Ann Neurol 2007;62:375–381.
Morrell MJ. Responsive cortical stimulation for the treatment of medically intrac-
table partial epilepsy. Neurology 2011;77:1295–1304.
Pacia SV, Ebersole JS. Intracranial EEG substrates of scalp ictal patterns from
temporal lobe foci. Epilepsia 1997;38:642–654.
Palmini AL, Gloor P, Jones-Gotman M. Pure amnestic seizures in temporal lobe
epilepsy. Definition, clinical symptomatology and functional anatomical con-
siderations. Brain 1992;115:749–769.
Copyright Ó 2012 by the American Clinical Neurophysiology Society
Mesial Temporal Lobe Epilepsy
Petroff OA, Rothman DL, Behar KL, Mattson RH. Low brain GABA level is
associated with poor seizure control. Ann Neurol 1996;40:908–911.
Plummer C, Wagner M, Fuchs M, et al. Dipole versus distributed source locali-
zation for single versus averaged spikes in focal epilepsy. J Clin Neurophy-
siol 2010;27:141–162.
Pillai J, Sperling MR. Interictal EEG in the diagnosis of epilepsy. Epilepsia
2006;47(suppl 1):14–22.
Ramachandran Nair R, Ochi A, Imai K, et al. Epileptic spasms in older pediatric
patients: MEG and ictal high-frequency oscillations suggest focal-onset seiz-
ures in a subset of epileptic spasms. Epilepsy Res 2008;78:216–224.
Risinger MW, Engel J Jr, Van Ness PC, et al. Ictal localization of temporal lobe
seizures with scalp/sphenoidal recordings. Neurology 1989;39:1288–1293.
Sammaritano M, Gigli GL, Gotman J. Interictal spiking during wakefulness and
sleep and the localization of foci in temporal lobe epilepsy. Neurology
1991;41:290–297.
Scheffer IE, Harkin LA, Grinton BE, et al. Temporal lobe epilepsy and
GEFS1 phenotypes associated with SCN1B mutations. Brain 2007;130:
100–109.
Schmidt D, Stavem K. Long-term seizure outcome of surgery versus no surgery
for drug-resistant partial epilepsy: a review of controlled studies. Epilepsia
2009;50:1301–1309.
Schulz R, Luders HO, Hoppe M, et al. Interictal EEG and ictal scalp EEG prop-
agation are highly predictive of surgical outcome in mesial temporal lobe
epilepsy. Epilepsia 2000;41:564–570.
Schwartzkroin PA. Hippocampal slices in experimental and human epilepsy. Adv
Neurol 1986;44:991–1010.
Scott RC, King MD, Gadian DG, et al. Hippocampal abnormalities after
prolonged febrile convulsion: a longitudinal MRI study. Brain 2003;126:
2551–2557.
Semah F, Picot MC, Adam C, et al. Is the underlying cause of epilepsy a major
prognostic factor for recurrence? Neurology 1998;51:1256–1262.
Siddiqui A, Kerb R, Weale ME, et al. Association of multidrug resistance in
epilepsy with a polymorphism in the drug-transporter gene ABCB1. N Engl
J Med 2003;348:1442–1448.
So EL. Value and limitations of seizure semiology in localizing seizure onset.
J Clin Neurophysiol 2006;23:353–357.
Sperling MR, Engel J Jr. Sphenoidal electrodes. J Clin Neurophysiol 1986;3:67–73.
Sperling MR, O’Connor MJ. Auras and subclinical seizures: characteristics and
prognostic significance. Ann Neurol 1990;28:320–328.
Squire LR. The legacy of patient H.M. for neuroscience. Neuron 2009;61:6–9.
Stephen LJ, Kwan P, Brodie MJ. Does the cause of localisation-related epilepsy
influence the response to antiepileptic drug treatment? Epilepsia
2001;42:357–362.
Sutherling WW, Barth DS. Neocortical propagation in temporal lobe spike foci on
magnetoencephalography and electroencephalography. Ann Neurol
1989;25:373–381.
Tao JX, Ray A, Hawes-Ebersole S, Ebersole JS. Intracranial EEG substrates of
scalp EEG interictal spikes. Epilepsia 2005;46:669–676.
Tatum WO, Benbadis SR, Hussain A, et al. Ictal EEG remains the prominent
predictor of seizure-free outcome after temporal lobectomy in epileptic
patients with normal brain MRI. Seizure 2008;17:631–636.
Theodore WH, Bhatia S, Hatta J, et al. Hippocampal atrophy, epilepsy
duration, and febrile seizures in patients with partial seizures. Neurology
1999;52:132–136.
Thom M, Martinian L, Catarino C, et al. Bilateral reorganization of the dentate
gyrus in hippocampal sclerosis: a postmortem study. Neurology
2009;73:1033–1040.
Thompson SA, Duncan JS, Smith SJ. Partial seizures presenting as panic attacks.
BMJ 2000;321:1002–1003.
Van Landingham KE, Heinz ER, Cavazos JE, Lewis DV. Magnetic resonance
imaging evidence of hippocampal injury after prolonged focal febrile con-
vulsions. Ann Neurol 1998;43:413–426.
Villanueva V, Serratosa JM. Temporal lobe epilepsy: clinical semiology and age at
onset. Epileptic Disord 2005;7:83–90.
Walsh RD, Wharen RE Jr, Tatum WO. Complex transient epileptic amnesia.
Epilepsy Behav 2011;20:410–413.
Weinand ME, Kester MM, Labiner DM, Ahern GL. Time from ictal subdural EEG
seizure onset to clinical seizure onset: prognostic value for selecting temporal
lobectomy candidates. Neurol Res 2001;23:599–604.
Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of
surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311–318.
Williamson PD, French JA, Thadani VM, et al. Characteristics of medial temporal
lobe epilepsy: II. Interictal and ictal scalp electroencephalography, neuropsy-
chological testing, neuroimaging, surgical results, and pathology. Ann Neurol
1993;34:781–787.
Worrell GA, Parish L, Cranstoun SD, et al. High-frequency oscillations and sei-
zure generation in neocortical epilepsy. Brain 2004;127:1496–1506.
Worrell GA, Gardner AB, Stead SM, et al. High-frequency oscillations in human
temporal lobe: simultaneous microwire and clinical macroelectrode record-
ings. Brain 2008;131:928–937.
365