doi:10.1093/brain/awh694
Brain (2006), 129, 346–351
Left hippocampal pathology is associated with
atypical language lateralization in patients with
focal epilepsy
Bernd Weber,1 Jörg Wellmer,1 Markus Reuber,4 Florian Mormann,1 Susanne Weis,1 Horst Urbach,2
Jürgen Ruhlmann,3 Christian E. Elger1 and Guillén Fernández5
Departments of 1Epileptology and 2Radiology, University of Bonn and 3Department of Diagnostic and Therapeutic
Neuroradiology, Medical Center Bonn, Bonn, Germany, 4Academic Neurology Unit, Division of Genomic Medicine,
University of Sheffield, Royal Hallamshire Hospital, Sheffield, UK and 5F.C. Donders Center for Cognitive Neuroimaging
and Department of Neurology, Radboud University Nijmegen, Nijmegen, The Netherlands
Correspondence to: Dr Bernd Weber, Department of Epileptology, Sigmund Freud Street 25, 53105 Bonn, Germany
E-mail: [email protected]
It is well recognized that the incidence of atypical language lateralization is increased in patients with focal
epilepsy. The hypothesis that shifts in language dominance are particularly likely when epileptic lesions are
located in close vicinity to the so-called language-eloquent areas rather than in more remote brain regions such
as the hippocampus has been challenged by recent studies. This study was undertaken to assess the effect of
lesions in different parts of the left hemisphere, lesions present during language acquisition, on language
lateralization. We investigated 84 adult patients with drug-resistant focal epilepsy with structural lesions
and 45 healthy control subjects with an established functional MRI language paradigm. Out of the 84 patients
43 had left hippocampal sclerosis, 13 a left frontal lobe lesion and 28 a left temporal-lateral lesion. All these
lesions were likely to have been present during the first years of life during language acquisition. To assess the
lateralization of cerebral language representation globally as well as regionally, we calculated lateralization
indices derived from activations in four regions of interest (i.e. global, inferior frontal, temporo-parietal and
remaining prefrontal). Patients with left hippocampal sclerosis showed less left lateralized language representations than all other groups of subjects (P < 0.005). This effect was independent of the factor of region,
indicating that language lateralization was generally affected by a left hippocampal sclerosis. Patients with left
frontal lobe or temporal-lateral lesions displayed the same left lateralization of language-related activations as
the control subjects. Thus, the hippocampus seems to play an important role in the establishment of language
dominance. Possible underlying mechanisms are discussed.
Keywords: fMRI; language; epilepsy; hippocampus
Abbreviations: AVM = arteriovenous malformation; DNT = dysembryoblastic neuroepithelial tumour; FCD = focal cortical
dysplasia; FFE = fast field echo; fMRI = functional MRI; ROIs = regions of interest; TSE = turbo spin echo; TSE sequence = sagittal
T1-weighted 3D gradient echo sequence
Received September 1, 2005. Revised October 10, 2005. Accepted October 24, 2005. Advance Access publication December 5, 2005
Introduction
In 94% of healthy subjects language functions are strongly
lateralized to the left (Springer et al., 1999; Knecht et al.,
2000), but early damage to the left hemisphere can result
in atypical, i.e. bilateral or right-lateralized, representation
of language (Rasmussen and Milner, 1977; Staudt et al.,
2002). The mechanism underlying the higher incidence of
atypical language dominance associated with left-hemispheric
pathologies is not very well understood. Furthermore, the
#
question of whether patients with lesions near classical language-related areas, i.e. Broca’s or Wernicke’s area, display a
higher degree of atypical dominance than patients with lesions
spatially separated from these areas is under debate (Knecht,
2004; Liegeois et al., 2004).
Initial evidence suggests that left temporal lobe pathology
is more often associated with atypical language lateralization than left frontal pathology (Liegeois et al., 2004).
The Author (2005). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved. For Permissions, please email: [email protected]
Language lateralization in focal epilepsy
This suggestion was derived from the localization of abnormalities within the left hemisphere detected by structural
magnetic resonance imaging in ten children with intractable
epilepsy. Five of the patients had lesions in the left inferior
frontal region, while the other five patients had lesions in the
left temporal lobe. None of the patients with inferior frontal
lesions displayed language-related activation in the right
hemisphere, while four out of the five patients with left
temporal lobe lesions showed atypical language dominance
leading to the hypothesis that lesions near classical languagerelated areas do not necessarily lead to a transhemispheric
shift of language dominance but rather to a shift within the
hemisphere to neighbouring regions. A different study based
on 100 patients with temporal lobe epilepsy related to hippocampal sclerosis suggested that atypical language representation can not only be caused by temporal-lateral lesions
adjacent to the so-called eloquent cortex (i.e. Wernicke’s
area), but also by damage to temporomedial structures
(Janszky et al., 2003). While all 17 patients with right hippocampal lesions showed typical left-sided language dominance,
20 out of the 83 patients with left-sided lesions exhibited
atypical language dominance. In that study language lateralization was determined by the intracarotid amobarbital procedure (Wada and Rasmussen, 1960). Although results of the
Wada-test largely depend on the suppression of frontal functions, they cannot assess the lateralization of temporo-parietal
and frontal language-related areas independently (Lehericy
et al., 2000). In addition, the study by Janszky and co-workers
examined patients with hippocampal sclerosis only and thus
it could not reveal how its finding was specifically related to
hippocampal lesions (Janszky et al., 2003).
Hence, two questions remain incompletely answered.
Namely, are lesions close to classic language-related areas
or are hippocampus lesions more likely to lead to a shift in
language dominance and are frontal and temporo-parietal
regions similarly involved in the displacement of language
representations? To answer these questions we compared a
group of 45 healthy control subjects and three groups of
patients with epileptogenic lesions present during language
acquisition in different brain regions of the left hemisphere
(hippocampus, lateral temporal cortex, frontal lobe). Cortical
representation of language was assessed by functional
MRI (fMRI) with a behavioural paradigm with a semanticperceptual contrast.
Subjects and methods
Subjects
A total of 84 patients with drug-resistant epilepsy from our presurgical epilepsy programme and 45 healthy subjects were included
in the study. All were native German speakers and had normal or
fully corrected vision. Table 1 shows the lesion for each patient and
Table 2 gives an overview of the demographic and clinical data. All
patients in the hippocampal sclerosis group had either their first
seizure or a complex febrile convulsion before the age of 2 years.
The lesions in the other groups were either developmental (e.g. focal
cortical dysplasias) or acquired perinatally (e.g. asphyxia during
Brain (2006), 129, 346–351
347
Table 1 Lesion patients with frontal lobe or temporallateral lesions
Patient age
Sex
Patients with left frontal lobe lesions
1
54
M
2
43
M
3
36
F
4
37
F
5
38
M
6
22
M
7
46
M
8
30
M
9
35
F
10
19
M
11
20
M
12
25
F
13
12
F
Patients with left temporal-lateral lesions
1
35
F
2
21
M
3
36
M
4
32
M
5
31
F
6
45
M
7
39
F
8
26
M
9
28
M
10
65
F
11
40
F
12
37
M
13
39
M
14
56
M
15
50
F
16
35
M
17
49
M
18
51
F
19
25
M
20
46
M
21
39
M
22
22
F
23
66
F
24
39
M
25
33
M
26
11
M
27
23
F
28
50
M
Lesion
Cavernoma
Asphyxia during birth
Hamartoma
FCDIlb
FCDIlb
Leucomalacia
Asphyxia during birth
DNT
FCDIlb
FCDIlb
DNT
FCDIlb
DNT
Cavernoma
FCDIlb
FCDIlb
Ganglioglioma
Atrophy
Ganglioglioma
DNT
DNT
Cavernoma
Cavernoma
DNT
Ganglioglioma
Ganglioglioma
Ganglioglioma
FCDIlb
Cortical dysplasia
AVM
Asphyxia during birth
Cortical dysplasia
Ganglioglioma
FCDIlb
Ganglioglioma
Cavernoma
Cavernoma
Cavernoma
Ganglioglioma
FCDIlb
Cortical dysplasia
birth). Hence, all patients had neocortical lesions or hippocampus
sclerosis or functional deficits of the left hippocampus leading to
sclerosis during the time of language acquisition. Lesions were
diagnosed by an expert neuroradiologist (H.U.). Written informed
consent was obtained according to the Medical Ethics Committee of
the University of Bonn in accordance with the Declaration of
Helsinki (1991).
Structural MRI
During presurgical evaluation, every patient underwent a standardized high-resolution MR-evaluation. MRI was performed on a 1.5 T
scanner (Gyroscan NT-Intera, Philips Medical Systems, Best,
Netherlands). The protocol included sagittal T1-weighted 3D gradient echo (FFE) sequence (slice thickness: 1 mm, repetition time
348
Brain (2006), 129, 346–351
B. Weber et al.
Table 2 Demographic data of patients and controls
Sex (male/female)
Handedness (EHI-score)
Age (years)
Hippocampal sclerosis
Temporal-lateral lesion
Frontal lobe lesion
Control subjects
16/27
67.73 6 62.81
40.40 6 10.04
18/10
63.92 6 60.35
37.65 6 12.80
8/5
97.69 6 4.39
31.92 6 12.20*
19/26
80.34 6 46.46
29.10 6 5.62**
*P < 0.05; **P < 0.005 in comparison to the hippocampal sclerosis group (Mann-Whitney U-test).
(TR): 12 ms, echo time (TE): 3.6 ms), axial T2-weighted TSE
sequence (slice thickness: 5 mm; interslice gap 1.0 mm; TR: 4849
ms; TE: 100 ms), axial FLAIR TSE sequence (slice thickness: 5 mm;
interslice gap: 1 mm; inversion time 2000 ms; TR: 6000 ms; TE:
120 ms), coronal FLAIR TSE sequence (slice thickness: 3 mm; TI:
2000 ms; TR: 6000 ms; TE: 120 ms), coronal T2-weighted TSE
sequence (slice thickness: 2 mm; TR: 4263 ms; TE: 100 ms) and
coronal T1-weighted inversion recovery sequence (slice thickness:
5 mm; interslice gap: 1 mm; TI: 400 ms; TR: 3223 ms; TE: 14 ms). In
patients with temporal lobe epilepsy, sequences were angulated perpendicularly or parallel to the longitudinal axis of the hippocampus.
Behavioural procedure
In the scanner, a series of item pairs, either word or consonant
string pairs, were presented back-projected onto a translucent screen,
which subjects viewed by way of a mirror. Both constituents of
each pair were simultaneously presented for 4 s interleaved with
the presentation of a central fixation cross for 0.2 s, above and
below a central fixation cross. A semantic condition (synonymjudgement task) alternated with a perceptual condition (lettermatching task) every 25 s so that six item pairs were presented
for each of 30 half-cycles. Hence, the behavioural experiment
took 12.5 min in total.
The verbal stimuli were 180 common German nouns (5–11
letters), forming 45 pairs of words with identical or highly similar
meanings (synonyms) and 45 pairs of semantically unrelated words.
The consonant strings were developed pseudorandomly to represent
45 pairs of two identical strings and 45 pairs in which one letter
was different between the two constituents. Strings were matched
with words with regard to the number of letters. Subjects were
required to push a button with the index finger of their right
hand whenever they identified a pair of two synonyms or identical
letter strings.
fMRI data acquisition
Sixteen axial slices with a matrix size of 64 · 64 and a field of view of
220 · 220 mm were collected at 1.5 T (Symphony, Siemens, Erlangen,
Germany). We acquired 248 T2*-weighted, gradient echo EPIscans including eight initial scans that were discarded to achieve
steady-state magnetization with the following parameters: slice
thickness, 6.0 mm; interslice gap, 0.6 mm; TR, 3.125 s; TE,
50 ms. In addition, we acquired a sagittal T1-weighted 3D-FLASH
sequence with 120 slices (matrix size, 256 · 256; field of view,
230 · 230 mm; slice thickness, 1.5 mm (no interslice gap); TR,
11 ms; TE: 4 ms).
fMRI data analysis
MR images were processed using SPM99 (www.fil.ion.ucl.ac.uk/
spm/) and the following steps were performed: (i) Realignment
with 3D motion correction. (ii) Normalization onto the MNI atlas
(Montreal Neurological Institute). (iii) Spatial smoothing with a
7 mm Gaussian kernel (full width at half maximum). (iv) Modelling
of the expected haemodynamic responses (box-car regressor in a
general linear model, GLM). This regressor was convolved with a
canonical haemodynamic response function (hrf ) to represent brain
physiology. (v) Temporal filtering of the acquired time-series to
reduce high- and low-frequency noise attributable to scanner
drifts and physiological noise. (vi) Calculation of parameter estimates for each condition covariate from the least mean squares fit of
the model to the data. (vii) Definition of the pre-experimentally
planned effects of interest (synonym-judgement > letter-matching
and letter-matching > synonym-judgement) and generation of contrast images for each subject and each effect of interest. To estimate
global and regional lateralization indices, we performed an automated quantification of global and regional activation clusters as
described previously (Fernández et al., 2001, 2003). We selected
three regional regions of interest (ROIs): inferior frontal region,
remaining prefrontal region and a temporo-parietal region. Overlay
masks were generated with the same paradigm, but in an independent sample of 12 healthy control subjects (Fernández et al., 2001)
from the averaged statistical parameter map with a region-growing
algorithm starting from the local activation maximum. Adding
mirror images so that a corresponding negative matched each positive X-coordinate generated the symmetric masks used. For the
global laterality index, we masked the whole supratentorial brain,
excluding three sagittal midline planes to minimize errors due to
normalization discrepancies. For the calculation of the laterality
index, we had to objectively adjust individual thresholds for the
identification of activated pixels because of intersubject variability
in general activation levels. This approach has been used in previous
studies of language lateralization (Fernández et al., 2001, 2003) and
showed high test–retest reliability as well as validity as compared
with results obtained by the intracarotid amobarbital procedure
(Wada and Rasmussen, 1960). Individual thresholding was achieved
by first calculating a mean maximum t-value defined as the mean of
those 5% of voxels showing the highest level of activation in each
volume of interest (VOI) of both hemispheres. The threshold for
inclusion in the calculation of the laterality index was then set at
50% of this mean maximum t-value. Finally, the sum of t-values
of voxels with values above this threshold was entered into the formula used to produce weighted laterality indices (Fernández et al.,
2001, 2003):
P
P
XL X
PV R ,
LI ¼ P V
X
þ
X
V L
V R
where V = set of activated voxels; XL = t-value of left hemispheric
voxels; XR = t-value of right hemispheric voxels.
Since laterality indices were not normally distributed, we applied
the Mann–Whitney U-test when comparing the four groups of
subjects.
Language lateralization in focal epilepsy
Results
Structural lesions
Table 1 gives an overview of the structural lesions in the
different patient groups. All patients with left hippocampal
sclerosis had their first seizure within the first year of life.
Seven of those patients had a history of complex febrile
convulsions. Of the patients with frontal lesions, five patients
had cortical dysplasias, three had dysembryoblastic neuroepithelial tumours (DNT), two had lesions due to perinatal
asphyxia, one a cavernoma, one a hamartoma and one had
leucomalacia. The group of patients with temporal-lateral
lesions consisted of seven patients with a ganglioglioma, six
with a cavernoma, eight with cortical dysplasias, three with
a DNT, two with lesions due to perinatal asphyxia, one with
atrophy of the temporal lobe without sclerosis of the hippocampus and one with an arteriovenous malformation (AVM).
All of these lesions were either developmental or acquired
before the age of 2 years.
Behavioural performance
Reaction times and the number of correct trials are listed in
Table 3. Patients and control subjects were able to solve the
semantic task clearly above chance level (t = 16.96, P < 0.001).
A MANOVA revealed significant differences in reaction
times and performance (number of correct responses) for
both the semantic and the perceptual control task between
the control group and the patient groups (max P < 0.001;
F = 10.273). However, we did not find any significant difference between the patient groups (min P = 0.153).
Brain (2006), 129, 346–351
349
only found in the hippocampal sclerosis group. Even after
excluding these five extreme cases, the hippocampal group
showed a significantly larger variance of the laterality indices
than all other groups (inferior frontal: Z = 1.630, P < 0.05;
temporo-parietal: Z = 1.446, P < 0.05; remaining prefrontal:
Z = 1.389, P < 0.05; global: Z = 1.389, P = 0.052). The other
patient groups and the control group did not differ significantly in their laterality indices in any of the regions of interest
(min P = 0.212).
A two-way ANOVA on lateralization indices with the
within-subject factor region (inferior frontal region, remaining prefrontal region, temporo-parietal region, hemisphere)
and the across-subject factor of group (hippocampal sclerosis,
temporal-lateral lesion, frontal lobe lesion and control group)
did not reveal a significant interaction between the factors
region and group (F = 0.494, n.s.), indicating that frontal as
well as temporo-parietal activations related to the semantic
task were uniformly less left lateralized in the hippocampal
sclerosis group than in the other groups.
Influences of clinical and demographic
factors on laterality indices
A stepwise regression analysis was performed to determine the
influence of clinical and demographic factors on the laterality
Influences of lesion site on lateralization
of language-related cortical activity
The mean laterality indices of patients with hippocampal
sclerosis were significantly lower in all four regions of interest
(i.e. less left lateralized) than of the patients with frontal
or temporal-lateral lesions or the control group (Fig. 1).
The mean global laterality index in the hippocampal sclerosis
group was 0.36, and ranged in the other groups between
0.61 and 0.81, displaying a decrease in the mean laterality
index of at least 0.25. Moreover, patients with a fully rightlateralized activation pattern (i.e. laterality index < 0.7) were
Fig. 1 Mean laterality indices and their standard deviations in the
ROI. Indices scaled from 1.0 (complete right lateralization) to
+1.0 (complete left lateralization). *P < 0.05; **P < 0.005;
***P < 0.001 in comparison with the hippocampal sclerosis group
(Mann–Whitney U-test).
Table 3 Behavioural data in the semantic and perceptual control task
Semantic task
RT 6 SD (ms)
No. of correct responses 6 SD
Perceptual control task
RT 6 SD
No. of correct responses 6 SD
Hippocampal sclerosis
Temporal-lateral lesion
Frontal lobe lesion
Control subjects
2211 6 464
33.62 6 11.61
2143 6 401
38.21 6 8.01
2046 6 349
39.10 6 5.98
1602 6 368**
43.13 6 3.55**
2697 6 444
34.32 6 11.61
2939 6 372
37.61 6 6.72
2796 6 330
32.40 6 10.76
2450 6 367*
42.38 6 3.40**
Mean reaction times (RT) and mean number of errors of the respective groups. The three groups of patients did not differ with respect to
either behavioural measure. The control group was significantly faster and made fewer errors in comparison with the three groups of
patients. *P < 0.01; **P < 0.001 in comparison with patient groups (ANOVA).
350
Brain (2006), 129, 346–351
indices in each region of interest. Only handedness displayed
an influence on the laterality index in all four ROIs (standard
coefficient: inferior frontal, 0.180; P < 0.05; temporo-parietal,
0.280; P < 0.005; remaining prefrontal, 0.314; P < 0.001;
global, 0.252; P < 0.005). However, this effect cannot explain
the differences in laterality indices between groups, because
the groups did not differ significantly with respect to their
handedness-scores (F = 1.546, P = 0.206, see Table 2). Moreover, the gender did not show any effect on the laterality
indices in either of the ROIs (P = 0.311; 0.583; 0.224 and
0.382), nor age (P = 0.281; 0.337; 0.181 and 0.180).
Discussion
Our results show that lesions of the left hippocampus are
associated more frequently with atypical language dominance
in inferior frontal as well as temporo-parietal language areas
than neocortical lesions in left frontal or left temporal lobes.
While our findings are in line with previous research (Janszky
et al., 2003; Liegeois et al., 2004) they provide new insights,
because they show that the hippocampus but not the temporal
lobe in general is the critical structure for language lateralization and that lateralization of both frontal and temporal
aspects of cortical language representations are affected by
left hippocampal lesions.
Liegeois and co-workers showed in a study of 10 children
that interhemispheric shifts in language dominance are not
so much associated with lesions found in close proximity to
classical language-related areas but in the temporal lobe
(Liegeois et al., 2004). However, the small sample size
makes it difficult to generalize these findings, especially
since the lesions outside language-related areas were spatially
distributed and not due to a single type of pathology. The
large group of adult patients with well-described lesions
examined here enables us to confirm and extend the results
of previous studies in that it identifies left hippocampal
lesions as particularly relevant to atypical language dominance. This observation is in keeping with a study by Janszky
and co-workers who showed in patients with left hippocampal
sclerosis that epileptic activity, i.e. the quantity of seizures and
count of epileptic discharges in the EEG, corresponded to a
higher degree of atypical language dominance as measured by
the Wada-test (Janszky et al., 2003). An important constraint
of the Wada-test is, however, that it can only examine the
function of a whole hemisphere, which makes it impossible to
discern the localization of different language-related areas,
whose laterality may be affected in different ways by pathological processes in the dominant hemisphere. Our fMRI
study confirms that hippocampal sclerosis is associated
with an increase in the displacement of all language-related
areas to the right hemisphere and demonstrates that this
displacement is not seen with temporal-lateral or frontal
lesions, which were not included in the previous study.
Medial temporal lobe epilepsy due to hippocampal
sclerosis is the most common diagnosis in epilepsy surgery
B. Weber et al.
programmes. Although patients with hippocampal sclerosis
can become seizure-free with antiepileptic drug therapy,
75% of patients become medically intractable during the
course of their disease (Spencer, 2002; Wieser and ILAE
Commission on Neurosurgery of Epilepsy, 2004). This high
degree of intractability indicates that this is a particularly
severe focal epilepsy syndrome which may be more closely
related to disturbances of cortical function than other forms
of focal epilepsy. Interictal discharges have been shown to
be correlated to neuropsychological deficits (Aldenkamp
and Arends, 2004). To our knowledge a direct comparison
of the amount of interictal epileptiform activity associated
with different cortical lesions has not been performed. However, several studies revealed that medial temporal lobe
epilepsy shows a higher frequency of interictal discharges
than other focal lesions (Hamer et al., 1999; Stuve et al.,
2001; Pfander et al., 2002). Since epileptic activity may lead
to a stronger involvement of the otherwise non-dominant
hemisphere in language functions (Regard et al., 1994), this
may explain the higher degree of atypical language dominance
in left medial temporal lobe epilepsy patients compared with
other focal left-hemispheric epilepsies.
The hippocampus represents higher-order associative cortex and thus it is integrated into different cognitive systems
by multiple reciprocal connections (Squire and Zola-Morgan,
1991; Eichenbaum et al., 1996). This means that functional
disturbances originating in the hippocampus can spread readily throughout the ipsilateral hemisphere as shown e.g. by
glucose-hypometabolism outside of the temporal lobe in
patients with hippocampal lesions (Hammers et al., 2001).
This may also explain why patients with hippocampal
sclerosis show not only memory deficits but also a variety
of other cognitive impairments not directly linked to the
temporal lobe (Elger et al., 2004; Wieser and ILAE
Commission on Neurosurgery of Epilepsy, 2004). Moreover,
deficits in these cognitive functions (especially those involving
the frontal lobe), tend to improve after successful epilepsy
surgery suggesting a functional disturbance of the whole
hemisphere due to epileptic activity originating in the hippocampus (Elger et al., 2004). Hence, our findings may be
explained by a more intense and widespread effect of a hippocampal than of a neocortical focus. It could be a consequence of such remote effects that right hemispheric
homologues of eloquent language areas take over certain language function in patients with left hippocampal sclerosis
more than in those with left neocortical lesion.
However, there may be another explanation for our results
(Knecht, 2004). To date, it remains unclear to what extent
medial temporal lobe structures are involved in language
acquisition. Some studies have found that hippocampal
integrity is a prerequisite for the acquisition of language
(DeLong and Heinz, 1997; Gross et al., 1998), and others
have found evidence for a semantic memory system independent of the medial temporal lobe (Vargha-Khadem et al.,
1997). A model describing the acquisition of a mental lexicon
during language development has proposed a contribution of
Language lateralization in focal epilepsy
the hippocampus in this mnemonic task (Ullman, 2004).
Under normal conditions the left hippocampus may interact
with neocortical language areas in the left hemisphere during
language development (Dehaene-Lambertz et al., 2002; Opitz
and Friederici, 2003). Hence, one could hypothesize that in
case of left hippocampal damage prior to or during language
acquisition the right hippocampus would play a more prominent role. As a consequence of this transhemispheric shift in
hippocampal contributions, right homologues of typically
left-hemispheric language areas may get involved in language
operations relevant for a semantic task as used here. This
involvement may be true not only for the acquisition of
lexical semantics but also for the development and learning
of grammatical rules which have been shown to rely on
intrahemispheric fronto–temporal interactions (DehaeneLambertz et al., 2002).
In conclusion, our data provide evidence for the critical
role of the left hippocampus in the determination of
language dominance. Further prospective studies are required
to determine whether the hippocampus is really important
for the acquisition of language thereby determining language
dominance or whether the atypical language dominance
observed in this study was caused by ongoing epileptic
activity.
References
Aldenkamp AP, Arends J. Effects of epileptiform EEG discharges on cognitive
function: is the concept of “transient cognitive impairment” still valid?
Epilepsy Behav 2004; 5: S25–34.
Dehaene-Lambertz G, Dehaene S, Hertz-Pannier L. Functional neuroimaging
of speech perception in infants. Science 2002; 298: 2013–5.
DeLong GR, Heinz ER. The clinical syndrome of early-life bilateral
hippocampal sclerosis. Ann Neurol 1997; 42: 11–7.
Eichenbaum H, Schoenbaum G, Young B, Bunsey M. Functional organization
of the hippocampal memory system. Proc Natl Acad Sci USA 1996; 93:
13500–7.
Elger CE, Helmstaedter C, Kurthen M. Chronic epilepsy and cognition.
Lancet Neurol 2004; 3: 663–72.
Fernández G, De Greiff A, Von Oertzen J. Reuber M, Lun S, Klaver P, et al.
Language mapping in less than 15 minutes: real-time functional MRI
during routine clinical investigation. Neuroimage 2001; 14: 585–94
Fernández G, Specht K, Weis S, Tendolkar I, Reuber M, Fell J, et al. Intrasubject reproducibility of presurgical language lateralization and mapping
using fMRI. Neurology 2003; 60: 969–75.
Gross DW, Li LM, Andermann F. Catastrophic deterioration and
hippocampal atrophy after childhood status epilepticus. Ann Neurol
1998; 43: 687.
Brain (2006), 129, 346–351
351
Hamer HM, Najm I, Mohamed A, Wyllie E. Interictal epileptiform
discharges in temporal lobe epilepsy due to hippocampal sclerosis versus
medial temporal lobe tumors. Epilepsia 1999; 40: 1261–8.
Hammers A, Koepp M, Labbe C, Brooks DJ, Thom M, Cunningham VJ, et al.
Neocortical abnormalities of [C-11]-flumazenil PET in mesial temporal
lobe epilepsy. Neurology 2001; 56: 897–906.
Janszky J, Jokeit H, Heinemann D, Schulz R, Woermann FG, Ebner A.
Epileptic activity influences the speech organization in medial temporal
lobe epilepsy. Brain 2003; 126: 2043–51.
Knecht S. Does language lateralization depend on the hippocampus? Brain
2004; 127: 1217–8.
Knecht S, Deppe M, Drager B, Bobe L, Lohmann H, Ringelstein E, et al.
Language lateralization in healthy right-handers. Brain 2000; 123: 74–81.
Lehericy S, Cohen L, Bazin B, Samson S, Giacomini E, Rougetet R, et al.
Functional MR evaluation of temporal and frontal language dominance
compared with the Wada test. Neurology 2000; 54: 1625–1633.
Liegeois F, Connelly A, Cross JH, Boyd SG, Gadian DG, Vargha-Khadem F,
et al. Language reorganization in children with early-onset lesions of the left
hemisphere: an fMRI study. Brain 2004; 127: 1229–36.
Opitz B, Friederici AD. Interactions of the hippocampal system and the
prefrontal cortex in learning language-like rules. Neuroimage 2003; 19:
1730–7.
Pfander M, Arnold S, Henkel A, Weil S, Werhahn KJ, Eisensehr I, et al.
Clinical features and EEG findings differentiating mesial from neocortical
temporal lobe epilepsy. Epileptic Disord 2002; 4: 189–95.
Rasmussen T, Milner B. The role of early left-brain injury in determining
lateralization of cerebral speech functions. Ann NY Acad Sci 1977; 299:
355–69.
Regard M, Cook ND, Wieser HG, Landis T. The dynamics of cerebraldominance during unilateral limbic seizures. Brain 1994; 117: 91–104.
Spencer SS. When should temporal-lobe epilepsy be treated surgically? Lancet
Neurol 2002; 1: 375–82.
Springer JA, Binder JR, Hammeke TA, Swanson SJ, Frost JA, Bellgowan PS,
et al. Language dominance in neurologically normal and epilepsy
subjects—a functional MRI study. Brain 1999; 122: 2033–45.
Squire LR, Zola-Morgan S. The medial temporal lobe memory system. Science
1991; 253: 1380–6.
Staudt M, Lidzba K, Grodd W, Wildgruber D, Erb M, Krageloh-Mann I.
Right-hemispheric organization of language following early left-sided brain
lesions: functional MRI topography. Neuroimage 2002; 16: 954–67.
Stuve O, Dodrill CB, Holmes MD, Miller JW. The absence of interictal spikes
with documented seizures suggests extratemporal epilepsy. Epilepsia 2001;
42: 778–81.
Ullman MT. Contributions of memory circuits to language: the declarative/procedural model. Cognition 2004; 92: 231–70.
Vargha-Khadem F, Gadian DG, Watkins KE, Connelly A, Van Paesschen W,
Mishkin M. Differential effects of early hippocampal pathology on episodic
and semantic memory. Science 1997; 277: 376–80.
Wada JA, Rasmussen T. Intracarotid injection of sodium amytal for
the lateralization of cerebral speech dominance. J Neurosurg 1960; 17:
266–82.
Wieser HG, ILAE Commission on Neurosurgery of Epilepsy. Mesial temporal
lobe epilepsy with hippocampal sclerosis. Epilepsia 2004; 45: 695–714.