Beruflich Dokumente
Kultur Dokumente
Michael Trimble and Bettina Schmitz have assembled a multinational team of experts to review
the most recent ndings which explore the interface between epilepsy and behaviour disorders.
They begin by looking at the classications available and examine how adequate they are for
dening the subtleties of behavioural changes in patients with neurological disorders. Coverage
is broad-ranging, from related cognitive problems and the biological underpinnings, to clinical
aspects, including pseudoseizures and treatment issues.
There has been a great deal of research in this area over recent years, but limited published
reviews. This timely book covers the practical implications of ongoing research, and oers both
a diagnostic and a management perspective. It will be essential reading for all professionals
engaged in the treatment of epileptic patients.
Michael Trimble is Professor of Behavioural Neurology at the Institute of Neurology in
London.
Bettina Schmitz is based in the Epilepsy Research Group of the Deparment of Neurology,
Humboldt University, Berlin.
The Neuropsychiatry of
Epilepsy
Edited by
Michael Trimble
and
Bettina Schmitz
caxniioci uxiviisir\ iiiss
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo
Cambridge University Press
The Edinburgh Building, Cambridge cn: :iu, United Kingdom
First published in print format
isbn-13 978-0-521-81374-7 hardback
isbn-13 978-0-521-00516-6 paperback
isbn-13 978-0-511-06480-7 eBook (NetLibrary)
Cambridge University Press 2002
Every effort has been made in preparing this book to provide accurate and up-to-date
information which is in accord with accepted standards and practice at the time of
publication. Nevertheless, the authors, editors and publishers can make no warranties
that the information contained herein is totally free from error, not least because
clinical standards are constantly changing through research and regulation. The authors,
editors and publisher therefore disclaim all liability for direct or consequential
damages resulting from the use of material contained in this book. Readers are strongly
advised to pay careful attention to information provided by the manufacturer of any
drugs or equipment that they plan to use.
2002
Information on this title: www.cambridge.org/9780521813747
This book is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
isbn-10 0-511-06480-2 eBook (NetLibrary)
isbn-10 0-521-81374-3 hardback
isbn-10 0-521-00516-7 paperback
Cambridge University Press has no responsibility for the persistence or accuracy of
uiis for external or third-party internet websites referred to in this book, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
isnx-:,
isnx-:c
isnx-:,
isnx-:c
isnx-:,
isnx-:c ;
o
:
;
,
;
Contents
List of contributors page viii
Part I Background
1 Introduction 3
M.R. Trimble and B. Schmitz
2 Neuropsychiatric disorders in epilepsy epidemiology and classication 5
E.S. Krishnamoorthy
3 Limbic connectivity: anatomical substrates of behavioural disturbances
in epilepsy 18
J. Engel Jr, C. Wilson and F. Lopez-Rodriquez
Part II Clinical aspects
4 The psychiatry of idiopathic generalized epilepsy 41
Dieter Janz
5 Epilepsy and learning disorders 62
Cesare Maria Cornaggia and Giuseppe Gobbi
6 Subtle cognitive and behavioural eects of epilepsy 70
Frank M.C. Besag
7 Aggression and epilepsy 81
L. Tebartz van Elst
8 Epilepsy and suicide: a neuropsychiatric analysis 107
Dietrich Blumer
9 Postictal psychoses, revisited 117
Kousuke Kanemoto
v
Part III Cognitive aspects
10 Dementia and epilepsy 135
Stephen W. Brown
11 The risk of cognitive decline in patients with refractory temporal
lobe epilepsy 152
Hennric Jokeit and Alois Ebner
12 Behavioural and neuropsychological aspects of frontal lobe epilepsy 164
Christoph Helmstaedter
Part IV Nonepileptic attacks
13 Epilepsy, dissociation and nonepileptic seizures 189
Richard J. Brown
14 Psychobiology of psychogenic pseudoseizures 210
J. Chris Sackellares and D. Kalogjera-Sackellares
15 Epilepsy and panic disorder 226
Howard A. Ring and Nuri Gene-Cos
Part V Treatment complications
16 The eects of antiepileptic drugs on behaviour 241
Bettina Schmitz
17 Antiepileptic drug treatment and epileptic seizures eects on
cognitive function 256
Albert P. Aldenkamp
18 Psychiatric eects of surgery for temporal lobe epilepsy 266
Steffi Koch-Stoecker
19 Vagus nerve stimulation and mood 283
Christian E. Elger and Christian Hoppe
Part VI Treatment
20 On the use of psychotropic drugs in patients with seizure disorder 299
M.R. Trimble and Anke Hensiek
vi Contents
21 The role of psychotherapy in the treatment of epilepsies 313
Martin Schndienst
22 Choosing measures to assess quality of life (QOL) in epilepsy 323
Caroline E. Selai, Katja Elstner and M.R. Trimble
Index 343
vii Contents
Albert P. Aldenkamp
Secretary General
Instituut voor Epilepsiebestrijding
Meer en Bosch
Postbus 21
2100 AA Heemstede
The Netherlands
Frank M.C. Besag
Learning Disability Service
Bedfordshire & Luton Community NHS
Trust
Twinwoods Health Resource Centre
Milton Road
Clapham
Bedfordshire MK41 6AT, UK
Dietrich Blumer
University of Tennessee
College of Medicine
Department of Psychiatry
135 North Pauline
Memphis TN38105, USA
Stephen W. Brown
Cornwall Healthcare Trust
Unit 10 Bodmin Business Park
Harleigh Road
Bodmin
Cornwall PL31 1AH, UK
Richard J. Brown
National Hospitals for Neurology and
Neurosurgery
Institute of Neurology
Queen Square
London WC1N 3BG, UK
Cesare Maria Cornaggia
Clinical Psychiatry
University of Milano Bicocca
San Gerardo Hospital
Via Donizetti 106
Mona, Italy
Alois Ebner
Epilepsy Centre Bethel
Epilepsy Surgery Program
Maraweg 21
D33617 Bielefeld, Germany
Christian E. Elger
Department of Epileptology
University of Bonn
Sigmund-Freud-Strasse 25
Bonn, FRG-53105
Katja Elstner
Raymond Way Neuropsychiatry Unit
Institute of Neurology
Queen Square
London WC1N 3BG, UK
viii
Contributors
Jerome Engel
Chief Division of Epilepsy and Clinical
Neurophysiology
Department of Neurology
Reed Neurological Research Centre
UCLA School of Medicine
710 Westwood Plaza
Los Angeles
CA 90095-1769, USA
Nuri Gene-Cos
St. Bartholomews & The Royal London
School of Medicine
London E1 1BB, UK
Giuseppe Gobbi
Department of Neuropsychiatry
Maggiore Hospital
Bologna, Italy
Christoph Helmstaedter
Universittsklinik fur Epileptologie
Sigmund-Freud-Str. 25
Universitt Bonn
53105 Bonn, Germany
Anke Hensiek
Department of Neurology
Addenbrookes Hospital
P.O. Box 165
Hills Road
Cambridge CB2 2HE, UK
Christian Hoppe
Department of Epileptology
University of Bonn
Sigmund-Freud-Strasse 25
FRG-53105 Bonn
Germany
Dieter Janz
Burgunder Strasse 8
14129 Berlin, Germany
Hennric Jokeit
Schweizerisches Epilepsie-Zentrum
Neuropsychologie
Bleulerstrasse 60
CH-8008 Zurich
Switzerland
Dalma Kalogjera-Sackellares
Department of Neuroscience
University of Florida
Florida, USA
Kousuke Kanemoto
Department of Neuropsychiatry
Aichi Medical University
21 Yazako-karimata, Nagakute
480-1195, Aichi, Japan
SteKoch-Stoecker
Klinik Mara I
Maraweg 21
D33617 Bielefeld, Germany
E.S. Krishnamoorthy
National Hospitals for Neurology and
Neurosurgery
Institute of Neurology
Queen Square
London WC1N 3BG, UK
Howard A. Ring
Academic Department of Psychological
Medicine
Third Floor, Alexandra Wing
Royal London Hospital
Whitechapel
London E1 1BB, UK
ix List of contributors
F. Lopez-Rodriquez
Department of Psychiatry and
Biobehavioural Sciences
UCLA School of Medicine
710 Westwood Plaza
Los Angeles
CA 90095-1769
USA
J. Chris Sackellares
College of Medicine Department of
Neurosciences
PO Box 100244
Gainesville
Florida 32610-0244, USA
Bettina Schmitz
Neurologische Klinik und Poliklinik
Charit, Virchow Klinikum
Humboldt-Universitt
Augustenburger Platz 1
13353 Berlin, Germany
Martin Schndienst
Epilepsie-Zentrum Bethel
Maraweg 21
33617 Bielefeld, Germany
Caroline E. Selai
126 Scott Ellis Gardens
London
NW8 9HG, UK
Ludger Tebartz van Elst
Department of Psychiatry and Psychotherapy
Albert-Ludwigs-University
Haupstr.5
79104 Freiburg, Germany
Michael R. Trimble
National Hospitals for Neurology and
Neurosurgery
Institute of Neurology
Queen Square
London
WC1N 3BG, UK
C. Wilson
Department of Neurology
UCLA School of Medicine
710 Westwood Plaza
Los Angeles
CA 90095-1769, USA
x List of contributors
Part I
Background
1
Introduction
M.R. Trimble
1
and B. Schmitz
2
1
Institute of Neurology, London, UK
2
Humboldt University, Berlin, Germany
In 1997, Dr Pete Engel, President of the International League Against Epilepsy,
invited us to establish a commission on psychobiology, and this book represents one
of the achievements of the commissions work. The task of the commission was to
explore the interface between epilepsy and behaviour disorders, from a biological
and social point of view. To these ends a number of subcommissions were estab-
lished, and their task was to explore the existing knowledge base of the discipline, to
suggest research strategies for interventions and to educate both patients and carers
about aspects of epilepsy which a number of people consider to have been neglected.
The present book is divided into several parts, covering a spectrum of clinical
topics which have been of concern to the commission. Some old chestnuts, for
example the interictal psychoses of epilepsy, have not been allocated specic chap-
ters, and a number of other areas, particularly relating to learning disability, cog-
nitive decline, dissociative attacks and vagus nerve stimulation have been included.
It is hoped that by expanding upon the literature on some of these less-well-dis-
cussed aspects of psychobiology in epilepsy further interest will be stimulated,
leading to both intellectual discussion and research endeavours.
We start our text with an introduction to the classication of psychiatric disor-
ders in epilepsy. The point is made that existing classications used in psychiatry
such as the DSMIV are quite inadequate when it comes to dealing with the sub-
tleties of the behaviour changes of patients with neurological disorders.
We then discuss the biological underpinnings of some behaviour problems, in
terms of exploring the limbic system and related structures that are aected by the
process of epilepsy and which are also related to behavioural disorders.
The part on clinical aspects explores in particular the problems of learning dis-
ability, and introduces the important area of state-dependent learning disabilities:
patients with cognitive decits that can be profoundly reversed by appropriate
treatment strategies. Other important areas covered include the ever-controversial
topic of aggression, the importance of suicide, and the group of psychoses that
occur postictally.
3
The next part looks further at cognitive problems in patients with epilepsy,
examining whether the concept of dementia is relevant, discussing the question as
to whether or not there is cognitive decline in patients with various types of epi-
lepsy over time, and also examining the issue of frontal lobe epilepsies. The latter
have been well dened from the seizure point of view in recent years, but the behav-
ioural and cognitive associations have yet to be claried.
We make no apology for including a section on nonepileptic seizures. The fact
that many patients who are diagnosed as having epilepsy do not have epilepsy, but
have some form of nonepileptic attack disorder (pseudoseizure) is now well recog-
nized. This problem has been around for centuries, and such eminent neurologists
as Charcot have spent some considerable time attempting to dierentiate between
nonepileptic and epileptic seizures. However, this often still proves dicult. We still
have inadequate information as to the mechanisms for the development of nonepi-
leptic attack disorder, and these, and the possible biological associations, are taken
up in this section.
The nal sections deal with treatments and their side eects. Of importance in
this section are the references to surgery, not only temporal lobe resection, but also
more recent advances such as vagus nerve stimulation. The benecial and negative
psychiatric consequences of these treatments are at the present time being actively
explored, and some early work is presented here. However, in the context of
psychobiology, our treatment strategies must go beyond medication and surgical
interventions, and we include a discussion of psychodynamic principles in rela-
tionship to the management of epilepsy, and also a chapter on quality of life.
We, the editors, hope that the book will enliven the debate about the links
between epilepsy and behaviour, an area which is often not well discussed, partly
because of some worry that any association between psychiatry and epilepsy may
stigmatize patients with epilepsy even more than they already are. However, the
problems that we have identied are a reality not only in the clinic, but also for
patients and carers themselves. It is dicult to dene treatment and management
strategies if problems are ignored, and so our intention is to enliven this area with
these up-to-date reviews on behavioural and cognitive problems in epilepsy, and
their social and biological underpinnings.
4 M.R. Trimble and B. Schmitz
2
Neuropsychiatric disorders in epilepsy
epidemiology and classification
E.S. Krishnamoorthy
Institute of Neurology, London, UK
Introduction
The association between epilepsy and psychiatric disorders has a long and che-
quered history. For centuries seizures were considered to be a form of demonic pos-
session. Beginning late in the nineteenth century, considerable attention has been
directed towards cataloguing, describing and understanding disorders at the inter-
face between epilepsy and psychiatry, particularly by European neurologists and
psychiatrists. However, it is only in the past few decades that any attention has been
paid to the epidemiology of these disorders. Similarly, aside from some early
attempts by European physicians, there have been no eorts to develop an opera-
tional classication of psychiatric disorders in epilepsy (Schmitz and Trimble, 1992
for a review).
The paucity of epidemiological research at this interface, and the failure to
develop an operational international system of classication, is in stark contrast
with developments both in epilepsy per se, and in mental health research. The epi-
demiology of epilepsy has been well studied in many countries and considerable
data (both descriptive and analytical) are now available. Indeed, epilepsy has been
subject to the gamut of epidemiological research including cross-sectional, case-
control and cohort studies (Hauser, 1998). Similarly, operational international
systems of classifying epilepsy and its syndromes have been developed both by the
Commission on Classication and Terminology of the International League
Against Epilepsy (1989), and the World Health Organization (1967), and are used
by epileptologists around the world.
Impressive developments have also taken place in the eld of mental health epi-
demiology. Eorts by the World Health Organizations Division of Mental Health,
and other pioneering organizations around the world, have led to a signicant
understanding of the epidemiology of psychiatric disorders. This has led to the
development of universally accepted classicatory systems in psychiatry, such as the
Diagnostic and Statistical Manual now in its fourth edition (DSMIV; American
5
Psychiatric Association, 1994), and the mental disorders component of the
International Classication of Diseases, now in its tenth edition (ICD10; World
Health Organization, 1992).
The commonly held conviction among epileptologists and neuropsychiatrists is
that psychiatric comorbidity is not only common in epilepsy, but that distinct and
unique forms of psychopathology are prevalent (Krishnamoorthy 2000, 2001). In
the past three decades attention has been directed towards discrete forms of
psychopathology in epilepsy such as the temporal lobe personality, interictal and
postictal psychosis, and interictal dysphoric disorder (Bear and Fedio, 1977;
Blumer 1995, 2000). This combined with the observation of similarities in behavi-
our during seizures and in psychopathological states has strengthened the notion
of an anity between epilepsy and psychiatric disorder. Yet, the evidence that
psychiatric disorders are overrepresented in epilepsy is far from convincing, with
conicting results in dierent studies.
In this chapter the epidemiology of psychiatric comorbidity in adult, non-learn-
ing-disabled patients with epilepsy will be reviewed. There is a considerable litera-
ture on children and the learning disabled that is being addressed elsewhere in this
book (Chapters 5 and 6). Some ideas on how psychiatric disorders in epilepsy may
be classied, and the work of the subcommission on classication of the
International League Against Epilepsy the Commission on Epilepsy and
Psychobiology in this regard, will also be discussed.
Epidemiology
A majority of studies in this area has been hospital- and institution-based. While
the contribution of these studies to the current understanding of psychopathology
in epilepsy has been invaluable, the strong selection bias in these studies does make
the extrapolation of their ndings to the majority of patients with epilepsy, who
live in the community, dicult.
There have been some population-based studies of psychiatric comorbidity that
are summarized here. Most studies have been cross-sectional and some have com-
pared cases with controls. By and large, save one or two exceptions, these studies
have generated crude estimates of prevalence, rather than more specic epidemi-
ological indices.
Population-based studies of psychiatric comorbidity in epilepsy
One of the earliest investigations to be carried out was that of Pond and Bidwell
(1959/60), who surveyed patients from 14 doctors surgeries in the south-east of
England. They found that 29% of 245 patients had psychological disorders of su-
cient severity to seek treatment, i.e. conspicuous morbidity. The main criticism lev-
elled against this study is its use of a social worker rather than a trained mental
6 E.S. Krishnamoorthy
health professional, and a lack of standardized techniques to assess patients with
epilepsy for psychiatric comorbidity. The strength of this study, however, lies in its
recognizing, four decades ago, the importance of an epidemiological approach.
Gudmundsson (1966) personally surveyed 987 patients with epilepsy living in
Iceland and reported that 512 (52%) had personality changes of various kinds. Of
these 271 (27.5%) were described as ixoid, 73 (7.4%) as ixothymic and 168
(17.0%) as neurotic. More men were ixoid and more women neurotic. While
Gudmundsson, unlike Pond, personally examined every subject, the clinical termi-
nology and classication used have few parallels today, and no attempts were made
to reduce bias. However, the high proportion of subjects with behavioural changes
in this community-based population is striking and worthy of note.
Edeh and Toone (1987) conducted a survey in doctors surgeries in south
London. They interviewed 88 adult patients with epilepsy drawn from doctors sur-
geries in the area, using the Clinical Interview Schedule, and reported that 48%
emerged as psychiatric cases. They also found that while patients with temporal
lobe epilepsy (TLE) and focal non-TLE did not dier in terms of psychiatric mor-
bidity, both groups were signicantly more impaired than patients with primary
generalized epilepsy. The techniques of ascertainment used in this study are com-
mendable. Subjects with epilepsy underwent both CT scans and EEG tests, in con-
rmation of their diagnosis. The study also used a validated instrument for
common mental disorder, the CIS-R (Lewis et al., 1992). In criticism, however, it
must be said that the study failed to examine matched population-based controls,
psychopathology specic to epilepsy was not examined, and while cases with
psychosis were identied, no validated diagnostic instrument for psychosis was
administered, the CIS-R being a validated instrument for common mental disor-
der alone.
Cockerell et al. (1996) conducted a nation-wide survey in the UK of acute
psychological disorders (APD) in patients with epilepsy using the British
Neurological Surveillance Unit. Sixty-four incident cases were ascertained over a
period of one year. Thirty-one were considered to have APD due to ictal or post-
ictal activity and 33 were interictal. In 30% of cases the APD was reported by the
referring physician to be secondary to the prescription of an antiepileptic drug
(AED). The drugs most commonly implicated were carbamazepine, lamotrigine
and vigabatrin. The broad psychiatric categories diagnosed included delirium
(25%), schizophreniform (31%), aective (30%), delusional (5%) and other dis-
orders (9%). The ndings of this study are of interest as it gives us crude incidence
gures of acute psychiatric disorder in epilepsy and highlights the importance of
antiepileptic drugs in precipitating comorbid psychiatric illness in epilepsy.
However, as the study used a reporting system, rather than a population-based
cohort, the results cannot be used to generate population-based incidence gures,
or be generalized.
7 Classification
Jalava and Sillanpaa (1996) examined a prospective population-based cohort
(mean follow-up of 35 years) of patients with epilepsy since childhood, for co-
morbid somatic, psychosomatic and psychiatric disorders. In comparison with
random controls, patients with epilepsy had a fourfold risk of psychiatric disorders
or combinations of somatic, psychosomatic and/or psychiatric disorders. Thus
patients with childhood-onset epilepsy demonstrated a higher risk for psychiatric
or psychosomatic disorders and this appeared to be related to epilepsy and not AED
administration.
This is perhaps the only cohort study of psychiatric comorbidity in epilepsy and
the ndings have great relevance. The results clearly indicate that subjects with epi-
lepsy are at higher risk of developing comorbid psychiatric illness, when compared
with population-based controls, and indicate the need for greater provision for
psychiatric treatment in primary care settings for epilepsy. However, as individual
cases were not ascertained in any systematic way, it is possible that the ndings do
not represent the true extent of comorbidity, with subtle nevertheless disabling
forms of psychopathology, or those not requiring medical attention or admission,
being missed. This is of relevance, as subtle forms of psychopathology that often do
not meet conventional diagnostic criteria may be overrepresented in epilepsy.
Bredkjaer et al. (1998) conducted a record-linkage study in Denmark between a
sample of people with epilepsy from the National Patient Register and from the
Danish Psychiatric Register. They found that the incidence of nonorganic nonaec-
tive psychoses including personality disorders that were broadly within the schizo-
phrenia spectrum was signicantly increased for both men and women with
epilepsy, even after excluding all people diagnosed as suering from a learning dis-
ability or substance misuse. The standardized incidence ratio was signicantly
increased for the entire schizophrenia spectrum (P10
8
), nonaective psychosis
(P10
8
) and schizophrenia alone (P0.0001).
In the absence of long-term prospective data, this study based on national regis-
ters provides evidence that disorders in the schizophrenia spectrum are clearly
overrepresented in epilepsy. The study enabled the calculation of more sophisti-
cated epidemiological indices, such as standardized incidence ratio, that have not
been estimated in previous studies. However, the methodological limitations of
reliance upon a case-register, i.e. the lack of standardization of ascertainment
methods, both for epilepsy and psychoses, and the exclusion of more subtle cases,
or those not requiring admission, do apply here.
Stefansson et al. (1998) conducted a case-control study comparing the preva-
lence of nonorganic psychiatric disorders among patients with epilepsy and con-
trols with other somatic diseases, both groups being of normal intelligence. The
two groups were drawn from a disability register of the State Social Security
Institute in Iceland. In this way, 241 index cases meeting inclusion criteria were
8 E.S. Krishnamoorthy
identied, and the ratio between subject (epilepsy) and control (somatic illness)
cases was 1:2. Psychiatric diagnosis was present among 35% of cases as compared
with 30% of controls, the dierence not being statistically signicant. Psychiatric
disorders were, however, signicantly more common in men with epilepsy, but not
in women, the dierence being due to a signicantly higher rate of psychosis, par-
ticularly schizophrenia or paranoid states, among men.
Some large hospital-based studies
Currie et al. (1971) surveyed 666 patients recorded to have features of temporal
lobe epilepsy in the hospital diagnostic index and the records of the neurology,
neurosurgery and EEG departments. They found 375 (56%) to be normal, 127
(19%) to be anxious, 71 (11%) to be depressed, 47 (7%) to be aggressive, 41 (6%)
to be obsessive and 38 (6%) to have a severe disturbance of aect.
Smith et al. (1986) studied 622 patients in the USA in a nation-wide cooperative
study spanning 10 Veterans Administration Medical centres, using a battery of
neuropsychological testing procedures. The majority of patients was not on anti-
convulsant drugs at the time of initial testing, and the few who were had subthera-
peutic levels on measurement. They found that patients with epilepsy scored
signicantly and consistently below the level of the 74 control subjects on all but
three behavioural measures. Dierences reaching statistical signicance were
found on tests of motor function (nger tapping, pegboard, colour naming),
cognitive-attention (digit symbol, discrimination reaction time, word uency) and
subtests of the Prole of Mood States (tension, depression, vigour and confusion).
These they felt provided a prole of behavioural characteristics of unmedicated
patients with epilepsy.
Gureje (1991) evaluated 204 unselected patients with epilepsy attending a neuro-
logical clinic using the Clinical Interview Schedule (Goldberg, 1972); 37% emerged
as psychiatric cases. Of these 53% had a neurosis, 29% had a psychosis and 7% were
diagnosed to have a personality disorder.
Mendez et al. (1993) conducted a retrospective investigation of neurology clinic
attenders. They found that interictal schizophrenic disorders occurred in 149
(9.25%) of 1611 patients with epilepsy as compared to only 23 (1.06%) of 2167
patients with migraine. They went on in the latter part of the study to compare 62
epilepsy and schizophrenia patients with 62 patients who had epilepsy alone on 6
seizure variables, and 62 patients with schizophrenia alone on 10 psychosis variables.
The epilepsy and schizophrenia group was found to have a later age of onset of
epilepsy with more complex partial seizures, more patients with auras and fewer
patients with generalized epilepsy. Except for increased suicidal behaviour, patients
with epilepsy did not dier from controls on psychosis variables; however, psy-
chotic symptoms often emerged with increased seizure activity. They felt that the
9 Classification
data supported a distinct association of schizophrenic disorders with epilepsy, par-
ticularly with seizures emanating from the temporal limbic system.
Manchanda et al. (1996) studied 300 consecutive patients refractive to treat-
ment, admitted for evaluation of their candidature for epilepsy surgery over a
6-year period. Of these, 231 had a temporal lobe focus, 43 had a nontemporal lobe
focus and 26 had generalized and multifocal seizure onset; 142 (47.3%) emerged as
psychiatric cases based on DSMIIIR criteria. A principal Axis I diagnosis was
made in 88 (29.3%). Anxiety disorders (10.7%) and schizophrenia (4.3%) were the
most common Axis I diagnoses. Dependent and avoidant personality traits were
frequent (18%) although patients rarely fullled criteria for a personality disorder.
Are psychiatric disorders commoner in epilepsy?
This question needs to be addressed from a public health perspective. Were psychi-
atric disorders to be commoner in patients with epilepsy, specic mental health
resources would need to be created in the community for this patient group. On
the other hand were there no excess in psychiatric comorbidity, when patients with
epilepsy were compared with other illness groups, matched for age, sex and disabil-
ity, and normal controls, such resources would not be required. Here we shall
examine the evidence, to see if depression and psychosis are commoner in epilepsy.
A majority of studies has shown depression to be common in epilepsy. Many of
these have employed the Minnesota Multiphasic Personality Inventory (MMPI).
Whitman et al. (1984) used a MMPI sequential diagnostic system (Goldberg, 1972)
to reanalyse 87 published proles of patients with epilepsy, other neurological dis-
orders and chronic physical illnesses, encompassing a total of 2786 patients. This
included 10 studies of epilepsy encompassing a total of 809 subjects. They found
that patients with epilepsy were at higher risk of psychopathology than normal
controls. However, no dierence was found between people with epilepsy and those
with other chronic disorders, or between people with TLE and those with general-
ized epilepsy. A similar investigation was also reported by Dodrill and Batzel
(1986), who found that patients with epilepsy demonstrated more psychopathol-
ogy than normal controls and patients with other neurological disorders, but that
there were no dierences in rates of psychopathology between TLE and other forms
of epilepsy.
Investigations using other instruments such as the Present State Examination
have also shown a higher prevalence of depression in epilepsy (Standage and
Fenton, 1975). However, other investigations have failed to demonstrate an
increased prevalence of depression in epilepsy. For a review of these studies and a
discussion of the phenomenology of depression in epilepsy see Lambert and
Robertson (1999).
10 E.S. Krishnamoorthy
Of all the dierent psychiatric disorders in epilepsy, it is psychosis for which
there is considerable evidence of overrepresentation. The prevalence of psychosis
in epilepsy is reported to be in the order of 4% (see Manchanda et al., 1996 for
example), sometimes rising as high as 10%. Psychotic disorders are 10 times more
common in epilepsy than in the general population, and this is borne out in well-
designed population-based cohort and case-control studies, reviewed here. For a
detailed review of studies of psychosis and of the nature and phenomenology of the
epileptic psychosis, see Trimble (1991).
Another reason for the diculty in answering this question is the selection bias
in most studies mentioned above. In this chapter we have deliberately concentrated
on population-based studies. Reviewing these (Table 2.1) it is apparent that there is
a considerable degree of psychiatric comorbidity in epilepsy, even in well-designed
cohort studies (see Jalava and Sillanpaa, 1996 for example). However, while the evi-
dence for a higher prevalence of psychotic disorders stands out, both in cohort
studies (Bredjkaer et al., 1998) and nested case-control studies (Stefannson et al.,
1998), the evidence for other psychiatric disorders, while present (see Jalava and
Sillanpaa, 1996 for example), is contradictory and not as compelling.
The classification of psychiatric disorders in epilepsy
The classication of psychiatric disorders in epilepsy has always been controversial.
There are two main schools of thought. The rst is that the existing systems of clas-
sication in psychiatry, the current being the ICD10 and DSMIV, in its fourth
edition, have made adequate provision for organic conditions like epilepsy, and
further subsystems of classication would only add to their complexity. The second,
most often voiced by neuropsychiatrists with an interest in epilepsy, is that the exist-
ing systems of classication are hopelessly inadequate as far as neurological disor-
ders in general and epilepsy specically are concerned (Krishnamoorthy, in press).
One recurrent theme in reviewing the literature about psychiatric disorders in
epilepsy is that the failure to identify an excess of psychopathology is due more to
the instruments used (generic to mental disorder and not specic to mental disor-
ders in epilepsy), rather than a true nding. It was this that led Bear and Fedio
(1977) to develop their own instrument, and conduct a study of psychopathology
in patients with temporal lobe epilepsy. The traits that they looked for were those
identied by Gastaut, and later Geschwind, who described the constellation of per-
sonality traits that characterize patients with temporal lobe epilepsy, including
hypergraphia, hyposexuality, religiosity and emotional viscosity.
The study by Bear and Fedio (1977) showed that while the MMPI failed to iden-
tify dierences between patients with TLE and other patient groups, the dierences
were all too apparent when the responses to the instrument they developed were
11 Classification
12 E.S. Krishnamoorthy
Table 2.1. Important epidemiological studies of neuropsychiatric comorbidity in epilepsy
Investigators
Year (country) Results Comments
1960 Pond and Bidwell 29% of 245 patients had Study in 14 doctors surgeries
(UK) signicant morbidity Conducted by psychiatric social
worker
Instruments not standardized
1966 Gudmundsson 52% of 987 patients had personality Personal survey by expert
(Iceland) changes Instruments and diagnosis not
standardized
1987 Edeh and Toone 48% of 88 patients emerged as Primary care-based
(UK) cases Sophisticated case ascertainment
Standard instruments but not
epilepsy-specic
1996 Cockerell et al. 64 incident cases of acute Nation-wide survey
(UK) psychological disorder on AED Relied on reporting system
institution Crude data on incidence cannot be
generalized
1996 Jalava and Sillanpaa Patients with epilepsy fourfold Prospective cohort study with 35-
(Finland) risk of somatic, psychosomatic and/ year follow-up (only cohort study to
or psychiatric disorder in date)
combination compared with Results clearly indicate that subjects
population-based controls with epilepsy are at higher risk of
Results related to epilepsy and not developing comorbid psychiatric
antiepileptic drug administration illness
1998 Bredjkaer et al. Incidence of schizophrenia spectrum Record linkage study between a
(Denmark) psychoses signicantly increased for sample of people with epilepsy from
both men and women with epilepsy the National Patient Register and the
Standardized incidence ratio for the Danish Psychiatric Register
entire schizophrenia spectrum Enabled the calculation of
(P10
8
), nonaective psychosis sophisticated epidemiological indices
(P10
8
) and schizophrenia alone not estimated in previous studies
(P0.0001)
1998 Stefansson et al. Psychiatric diagnosis in 35% of 241 Patients with epilepsy, and controls
(Iceland) epilepsy cases as compared with 30% with other somatic diseases, both
of controls, the dierence not being groups being of normal intelligence
statistically signicant drawn from a disability register of
Signicantly higher rate of the State Social Security Institute
schizophrenia among men
analysed. A number of studies have been conducted since then with conicting
results (for reviews Shetty and Trimble, 1997; Trimble, 1991).
More recently, Blumer (1995, 2000) has drawn attention to the mood disorder
that is seen in patients with refractory epilepsy, particularly TLE that may occur in
conjunction with these personality traits. This interictal dysphoric disorder (IDD)
of epilepsy is described as being polymorphic, and characterized by a constellation
of eight symptoms, typically of short duration and occurring in dierent permu-
tations and combinations at dierent times.
Blumer argues that the personality traits seen in these patients, such as a serious
demeanour, deliberate speech, an ethical and spiritual orientation, especially when
subtle, can be positive attributes. However, when the IDD coexists, there may be a
paroxysmal venting of angry aects not normally characteristic of the person, fol-
lowed by a sense of remorse. This he contends can be the source of signicant dis-
ability.
A modied version of the BearFedio scale, the Neurobehavioural Inventory
(NBI; Blumer, 1995) is reportedly sensitive to this disorder, helping to identify
some of its typical features. However, as the psychometric properties of this instru-
ment have not been tested, and population-based studies have not been carried out,
the validity of this diagnosis has not been established.
Other dierences have also been reported to characterize psychopathology in
epilepsy and to dierentiate it from psychopathology in general. The interictal
psychosis of epilepsy is reported in many studies to be characterized by the preser-
vation of aect, religiosity and paranoid ideation, rather than the undierentiated,
or hebephrenic picture seen in schizophrenia. Slater rst observed this in his land-
mark study at the National Hospital, Queen Square (Slater and Beard, 1963).
Subsequently, many others have commented on these ndings (Trimble, 1991) and
indeed, they have to some extent been borne out in the population-based studies
reviewed here (Stefannson et al., 1998).
Further, while psychiatric classications such as the ICD and DSM tend to
subsume epilepsy under the organic umbrella, thus limiting the ability of such clas-
sication to be versatile, and clinically relevant, current classicatory systems in epi-
lepsy pay little or no attention to psychopathology. Besides, there is no place in
existing systems of classication for the psychiatric disorders that are reported to be
specic to epilepsy. A distinct classicatory system enables clearer phenomenologi-
cal descriptions of these disorders. It also lends itself more easily to empirical testing.
A convincing argument can therefore be made for a distinct classicatory system,
and this is currently the subject of a discussion document of the ILAE Commission
on Epilepsy and Psychobiology (Krishnamoorthy et al., in preparation).
As reviewed here, the neuropsychiatric disorders specic to epilepsy comprise
the gamut of neuropsychiatry. Included are the so-called organic mental disorders,
13 Classification
such as postictal confusional states and complex partial status with psychopatho-
logical manifestations; personality changes (the GastautGeschwind syndrome of
temporal lobe epilepsy, and the labile personality of juvenile myoclonic epilepsy);
a spectrum of psychoses with varying intensity, features and manifestations
depending on the temporal relationship with seizure(s) (Trimble, 1991); and a
spectrum of neuroses with predominantly aective features (Blumer 1995, 2000).
These disorders are also inexorably linked to their relationship with the seizure(s)
per se (preictal, postictal, interictal and perhaps periictal); their relationship to the
EEG (for example forced normalization of Landolt and alternative psychosis of
Tellenbach; Krishnamoorthy and Trimble, 1999); and their relationship to antiepi-
leptic drug (AED) therapy (the AED-induced neuropsychiatric disorders; Trimble,
1998). The inclusion of nonepileptic attack disorder (NEAD) in such a classica-
tion is rather more controversial, as there is a growing understanding that NEAD
is, in a number of people, the manifestation of a much wider spectrum of psycho-
pathology than that specic to epilepsy (Brown and Trimble, 2000).
Any classicatory system, to be viable, will need to take all these factors into con-
sideration. Further, it is important to acknowledge that patients with epilepsy could
like anyone, especially patients with chronic medical conditions, have comorbid
psychiatric disorders that match existing descriptions in the ICD10 and DSMIV.
It would serve little purpose to try and reclassify these disorders when associated
with epilepsy. The judgement about whether to record the illness in a given patient
as a comorbid disorder or as an epilepsy-specic disorder would be best left to the
clinician dealing with that individual case. It also goes without saying that such a
classicatory system should link closely with the ILAE Classication of Epilepsies
and Epileptic Syndromes.
While there is little doubt that classications grounded in aetiology and
pathophysiology are an ideal that we must aspire for in the long term, our under-
standing of causation and its mechanisms in psychiatry, even the neuropsychiatry
of epilepsy, is fairly rudimentary, and much ground needs to be covered before we
can move with any certainty towards an aetiological model. Aetiologically based
systems of classication also require specialized knowledge and access to suppor-
tive investigative techniques. Both of these are unavailable in a number of settings,
particularly in the developing world. Thus, at least at present, classicatory systems
that aim to be culture-free and acceptable across the board, would do well to adopt
a descriptive approach, based on a good history and clear clinical descriptions. Such
descriptive approaches mirror good clinical practice around the world, and make
few demands in terms of specialist expertise or investigation.
An ideal classication would be one that is simple, user-friendly, grounded in
good history taking and concise clinical descriptions, links with existing systems of
classication both in epilepsy and psychiatry, and one which is applicable across
14 E.S. Krishnamoorthy
the board. Such a classicatory system will be helped by its application in good
prospective population-based research leading eventually to its being operational
and valid in the years to come.
Undoubtedly, as our understanding of aetiology and pathophysiology in epi-
lepsy and neuropsychiatry improves, one would expect a radical alteration in such
descriptive classicatory systems, as we are beginning to see in other neurological
and indeed neuropsychiatric disorders. Until then, however, it would perhaps be
wise for us to continue in the classical descriptive tradition of good clinical medi-
cine.
Conclusions
Psychiatric disorders are common in epilepsy, and encompass the spectrum of
conditions, from those that are a direct consequence of epileptogenic activity, to
others that are simply comorbid.
There is considerable evidence from epidemiological research to suggest that the
psychoses are greatly overrepresented in epilepsy; the evidence for an overrepre-
sentation of other psychiatric disorders is less compelling.
While hospital-based data indicate the presence of epilepsy-specic psychopa-
thology, this has never been examined in the epidemiological setting. Further,
instruments such as the NBI that are supposedly sensitive to epilepsy-specic
psychopathology have not been validated in this setting.
Existing classicatory systems, both in epilepsy and in mental health, are inade-
quate, with regard to neuropsychiatric disorders in epilepsy. Eorts to build con-
sensus in this regard, and to develop a user-friendly classicatory system are
currently being made by the ILAE Commission on Epilepsy and Psychobiology.
To be relevant and applicable, a classicatory system must not be demanding in
terms of specialist knowledge or investigation. A descriptive approach that
mirrors good clinical practice is therefore recommended. Such a classicatory
system must also link in with existing systems in epilepsy and mental health.
Systematic population-based research using reliable methods of ascertainment,
and controls matched for age, sex, disability and ethnicity, based on the ILAE
classication of neuropsychiatric disorders in epilepsy, with long-term follow-up
of such cohorts, must be conducted in the future.
REFERENCES
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders
(Fourth Revision) DSMIV. Washington, DC: American Psychiatric Association.
15 Classification
Bear, D.M. and Fedio, P. (1977). Quantitative analysis of interictal behaviour in temporal lobe
epilepsy. Arch Neurol, 34, 45467.
Blumer, D. (1995). Personality disorders in epilepsy. In Neuropsychiatry of Personality Disorders,
ed. J.J. Ratey, pp. 23063. Boston: Blackwell Science.
Blumer, D. (2000). Dysphoric disorders and paroxysmal eects: recognition and treatment of
epilepsy-related psychiatric disorders. Harv Rev Psychiatry, 8, 817.
Bredkjaer, S.R., Mortensen, P.B. and Parnas, J. (1998). Epilepsy and non-organic non-aective
psychosis: National Epidemiologic Study. Br J Psychiatry, 172, 2358.
Brown, R.J. and Trimble, M.R. (2000). Dissociative psychopathology, non-epileptic seizures and
neurology (Editorial). J Neurol Neurosurg Psychiatry, 69, 28591.
Cockerell, O.C., Moriarty, J., Trimble, M.R., Sander, J.W.A.S. and Shorvon, S.D. (1996). Acute
psychological disorders in patients with epilepsy: a nation-wide study. Epilepsy Res, 25, 11931.
Commission on Classication and Terminology of the ILAE (1989). Proposal for revised classi-
cation of epilepsies and epileptic syndromes. Epilepsia, 30, 38999.
Currie, S., Heatheld, K.W.G., Henson, R.A. and Scott, D.F. (1971). Clinical course and progno-
sis of temporal lobe epilepsy a survey of 666 patients. Brain, 94, 17390.
Dodrill, C.B. and Batzell, L.W. (1986). Interictal behavioural features of patients with epilepsy.
Epilepsia, 27 (Suppl. 2), S64S76.
Edeh, J. and Toone, B. (1987). Relationship between interictal psychopathology and the type of
epilepsy. Br J Psychiatry, 151, 95101.
Goldberg, L.R. (1972). Man versus mean: the exploitation of group proles for the construction
of diagnostic classication systems. J Abnorm Psychol, 79, 12131.
Gudmundsson, G. (1966). Epilepsy in Iceland a clinical and epidemiological investigation. Acta
Neurol Scand, Suppl 25, 43, 6490.
Gureje, O. (1991). Interictal psychopathology in epilepsy prevalence and pattern in a Nigerian
clinic. Br J Psychiatry, 158, 7005.
Hauser, A.W. (1998). Incidence and prevalence. In Epilepsy A Comprehensive Textbook, ed. J.R.
Engel and T.A. Pedley TA. New York: Lippincott-Raven.
Jalava, M. and Sillanpaa, M. (1996). Concurrent illnesses in adults with childhood-onset epi-
lepsy: a population-based 35-year follow-up study. Epilepsia, 37, 115563.
Krishnamoorthy, E.S. and Trimble, M.R. (1999). Forced normalization clinical and therapeu-
tic relevance. Epilepsia, 40 (Suppl. 10), S5764.
Krishnamoorthy, E.S. (2000). An approach to classifying neuropsychiatric disorders in epilepsy.
Editorial. Epilepsy Behav, 1, 3737.
Krishnamoorthy, E.S. (2001). Psychiatric issues in epilepsy. Curr Opinion Neurol, 14, 21724.
Lambert, M. and Robertson, M.M. (1999). Depression in epilepsy. Etiology, phenomenology,
and treatment. Epilepsia, 40 (Suppl. 10), S2147.
Lewis, G., Pelosi, A.J., Araya, R. and Dunn, G. (1992). Measuring psychiatric disorder in the com-
munity: a standardised assessment for use by lay interviewers. Psychol Med, 22, 46586.
Manchanda, R., Schaefer, B., McLachlan, R. et al. (1996). Psychiatric disorders in candidates for
surgery for epilepsy. J Neurol Neurosurg Psychiatry, 61, 829.
Mendez, M.F., Grau, R., Doss, R.C. and Taylor, J.L. (1993). Schizophrenia in epilepsy: seizure and
psychosis variables. Neurology, 43, 10737.
16 E.S. Krishnamoorthy
Pond, D.A. and Bidwell, B.H. (1959/60). A survey of epilepsy in fourteen general practices II.
Social and psychological aspects. Epilepsia, 1, 28599.
Schmitz, B. and Trimble, M.R. (1992). Epileptic equivalents in psychiatry: some 19th century
views. Acta Neurol Scand, 86 (Suppl. 140), 1226.
Shetty, T. and Trimble, M.R. (1997). The Bear Fedio Inventory: twenty years on. J Epilepsy, 10,
25462.
Slater, E. and Beard, A.W. (1963). The schizophrenia-like psychoses of epilepsy. Br J Psychiatry,
109, 95150.
Smith, D.B., Craft, B.R., Collins, J., Mattson, R.H., Cramer, J.A. and the VA Co-operative Study
Group 118 (1986). Behavioural characteristics of epilepsy patients compared with normal
controls. Epilepsia, 27, 7608.
Standage, K.F. and Fenton, G.W. (1975). Psychiatric symptom proles of patients with epilepsy:
a controlled investigation. Psychol Med, 5, 15260.
Stefansson, S.B., Olafsson, E. and Hauser, A.W. (1998). Psychiatric morbidity in epilepsy: a case
controlled study of adults receiving disability benets. J Neurol Neurosurg Psychiatry, 64,
23841.
Trimble, M.R. (1991). The Psychoses of Epilepsy. New York: Raven Press.
Trimble, M.R. (1998). New antiepileptic drugs and psychopathology. Neuropsychobiology, 38,
14951.
Whitman, S., Hermann, B.P. and Gordon, A.C. (1984). Psychopathology in epilepsy: how great
is the risk? Biol Psychiatry, 19, 21336.
World Health Organisation (1992). The International Classication of Mental and Behavioural
Disorders, Tenth Edition (ICD10). Geneva: WHO.
17 Classification
3
Limbic connectivity: anatomical substrates of
behavioural disturbances in epilepsy
J. Engel Jr, C. Wilson and F. Lopez-Rodriguez
UCLA School of Medicine, Los Angeles, USA
Introduction
Behavioural disturbances associated with epilepsy are due, in part, to a number of
important psychological and social factors; however, it is useful to acknowledge
that there are neurobiological factors as well. Not only does identication of
organic bases for behavioural aberrations help reduce the stigma they invariably
engender, but also it provides a rational scientic approach to prevent or reverse a
signicant cause of disability experienced by persons with epilepsy.
Cognitive and psychiatric disturbances associated with epilepsy may be due to
the same underlying pathological process that causes the epileptic condition, and
there is evidence that behavioural impairment may be related to the pathophysio-
logical nature of the underlying lesion, its eect on development and its location
(Engel et al., 1986; Engel and Shewmon, 1991). With respect to location, mesial
temporal and other limbic epileptogenic lesions are most likely to be associated
with behavioural disturbances (Engel et al., 1986; Engel and Shewmon, 1991).
Temporal lobe epilepsy is the most common form of human epilepsy, and hippo-
campal sclerosis is the most common human epileptogenic lesion (Engel, 1998).
Furthermore, mesial temporal lobe epilepsy with hippocampal sclerosis may be the
most refractory to antiepileptic drugs (Engel, 1998). Consequently, limbic system
dysfunction is undoubtedly the most important cause of behavioural disturbances
associated with epilepsy.
Epileptogenicity itself is known to cause enduring changes in brain function
and structure in a number of ways which can result in altered behaviour. For
instance, postnatally or in early childhood, abnormal recurrent synaptic bom-
bardment of distant projection areas by repeated ictal and interictal epileptiform
discharges can adversely aect the development of normal neuronal integration
(Engel and Shewmon, 1991; Morrell et al., 1995). In the mature brain such
abnormal synaptic activity can induce plastic changes through kindling mecha-
nisms (Corcoran and Mosh, 1998; Engel et al., 1986), and the naturally occur-
18
ring homeostatic seizure-suppressing mechanisms which maintain the interictal
state may have adverse consequences on normal neuronal function (Engel et al.,
1986; Engel and Shewmon, 1991). In addition, epileptic seizures are known to
disrupt sleep patterns (Shouse et al., 1997) and endocrine function (Herzog,
1997), which can alter behaviour; and severe epileptic activity results in neuro-
nal cell death and consequent synaptic reorganization (Sloviter and Damiano,
1981). Given the importance of the limbic systemin the development and main-
tenance of cognition, emotion, motivation and other essential behaviours, it is
not surprising that epileptogenic lesions capable of disrupting limbic connectiv-
ity are associated with a great variety of clinically signicant behavioural abnor-
malities.
The limbic system
The term limbic system was coined by Maclean (1952) to designate a series of
structures originally delineated by Broca (1878) and Papez (1937), plus the amyg-
dala and its connections, which were believed to play a crucial role in mediating
the exchange of information between the thinking brain (cortex) and the more
primal animal brain (diencephalon and brain stem). Broca rst described the
limbic lobe as the area of brain making up the rim of the cortex, including the
hippocampus, cingulate cortex and certain frontal lobe structures. Papez more
precisely dened circuits that connected these structures, including the septal
area, mamillary bodies, and parts of the thalamus and hypothalamus. The addi-
tion of the amygdala and its connections by Maclean provided the functional sub-
strate for human emotional behaviours now attributed to the limbic system
(Figure 3.1). The hippocampus, amygdala and their projection elds are central
to the concept of the limbic system, and also are the most epileptogenic regions
of the mammalian brain. The perirhinal and piriform regions of the parahippo-
campal cortex, along with the central nucleus of the amygdala, are the most
rapidly kindled structures in the brain (McIntyre and Plant, 1993), and an area of
deep piriform complex, area tempesta, has the lowest threshold in the brain to sei-
zures induced by local application of antagonists to gamma amino butyric acid
(GABA), the primary inhibitory neurotransmitter in the brain (Piredda and Gale,
1985). The limbic system, therefore, is uniquely susceptible to the development
of epileptiform abnormalities, not only as a result of hippocampal sclerosis, but
due to any irritating disturbance. Furthermore, distant epileptogenic lesions
which preferentially project to mesial temporal structures also have a high pro-
pensity to induce mesial temporal epileptiform activity resulting in both tempo-
ral lobe seizures and their behavioural consequences (Williamson and Engel,
1997).
19 Anatomical substrates of behavioural disturbances
20 J. Engel Jr et al.
Cingulate
Thalamus
Hypothalamus septum
SMA
Frontomedial
Anterior cingulate
Fronto-lateral
Fronto-orbital
NA
Parietal
Post lateral
temporal
Insula
Inferotemporal and
anterior temporal
CC VHC
DHC
AC
Contralateral
hippocampal
formation
Mid-brain tegm.
reticular formation
Second generalization
Ento.
Parahipp.
Fornix
MB
Hipp.
Figure 3.1. Diagrammatic representation of the human limbic system and its afferent and efferent
connections. Note the major connections between the hippocampus, entorhinal cortex
and amygdala (NA), and the components of the Papez circuit, described in the text. SMA,
supplementary motor area; HIPP, hippocampus; ENTO, entorhinal cortex; PARAHIPP,
parahippocampal gyrus; CC, corpus callosum; VHC, ventral hippocampal commissure;
DHC, dorsal hippocampal commissure; AC, anterior commissure. (With permission,
adapted from Wieser, 1988.)
Hippocampus
Normal structure
The hippocampus is perhaps the most studied region of the mammalian brain, due
in part to its highly organized laminar structure, and its propensity for plastic
change, such as long-term potentiation and long-term depression, which may
underlie mechanisms of normal learning and memory (Gloor, 1997;
Schwartzkroin and McIntyre, 1997). The hippocampal formation consists of the
dentate gyrus and the hippocampus proper or cornu ammonis (CA), which is
divided into CA1, CA2 and CA3 (Figure 3.2). The major excitatory hippocampal
input is glutaminergic from the entorhinal cortex via the perforant path, which ter-
minates on the distal two-thirds of granule cell dendrites in the dentate gyrus. The
medial portion of the entorhinal cortex projects predominantly to the middle third
21 Anatomical substrates of behavioural disturbances
Regio superior
(CA1)
Basket cell
Lateral
entorhinal
cortex
Perforant
path
Medial entorhinal
cortex
Exposed
blade
of fascia
dentata
Buried
blade
of fascia
dentata
Mossy
fibres
Regio
inferior
(CA3)
Schaffer
collateral
To
septum
Figure 3.2. Illustration of the intrinsic circuitry of the hippocampal formation. The primary focus of
this illustration is the trisynaptic circuit. The first synapse consists of the mossy fibre to
CA3 pyramidal cell connection derived from the granule cells of the dentate gyrus. The
second synapse is the connection of the CA3 cell output via the Schaffer collaterals to
CA1, and the third, not shown, is the CA1 axonal output to subiculum and entorhinal
cortex. Entorhinal cortex also provides the initial input to the dentate gyrus. (With
permission from OKeefe and Nadel, 1978.)
of the granule cell dendritic layer (middle molecular layer), while the lateral ento-
rhinal cortex projections, which also contain opioids, terminate on the distal third
of granule cell dendrites (outer molecular layer). These latter projections may play
a role in epileptogenicity due to the opioid disinhibitory eect on granule cell excit-
ability (Xie et al., 1992). The neuronal structure of the dentate gyrus provides the
substrates for strong feed-forward and feedback inhibition which resists the prop-
agation of epileptiform activity, leading to its designation as the dentate gate
(Lothman et al., 1991).
The classical excitatory trisynaptic pathway through the hippocampus is glutam-
atergic and utilizes NMDA and non-NMDA receptors, while inhibitory interneu-
rons are GABAergic. The pathway begins with the dentate granule cell axons, called
mossy bres, which project through the dentate hilus to area CA3. Mossy bres
contain two types of terminals: the rst type are the large, complex mossy terminals
which give the bres their name, and the second type are small terminals and lo-
podia extending from the mossy terminals. The primary target of the mossy termi-
nals are the CA3 pyramidal cells, with which they form multiple excitatory synapses,
while the primary target of the small terminals and lopodia are GABAergic inter-
neurons. The ratio of dentate granule axon terminals contacting GABAergic inter-
neurons to those terminals contacting pyramidal and mossy cells (excitatory
neurons in the dentate hilus) has been estimated at approximately 5: 1, which means
that the major postsynaptic targets of granule cells are inhibitory neurons (Acsdy
et al., 1998). Although the large proportion of granule cell output which activates
interneurons has a signicant inhibitory eect, the mossy terminals exciting CA3
pyramidal cells are quite powerful. Because CA3 pyramidal cell intrinsic connec-
tions support recurrent excitation, focused discharge of specic CA3 pyramidal cells
can occur. Excitation is further supported by granule cell activation of mossy cells
in the dentate hilus, which in turn project back onto the proximal one-third of the
granule cell dendrites (inner molecular layer). This excitatory feedback loop is oset
by similar inhibitory feedback loops via a variety of hilar inhibitory neurons. CA3
pyramidal cells have a unique propensity to burst synchronously, thus excitatory
epileptiform input that traverses the dentate gate activates synchronously bursting
pacemaker cells in CA3, giving rise to interictal spikes (Wong and Traub, 1983).
The second segment of the hippocampal trisynaptic pathway consists of the
Schaer collaterals, CA3 pyramidal cell axons which terminate on CA1 pyramidal
neurons. Under epileptogenic conditions, synchronous bursting in CA3 can
produce continuous discharges in CA1, the usual site of ictal onset in the hippo-
campal slice preparation. CA2 is a transition zone that does not appear to par-
ticipate in the transmission of epileptiform discharges in the nonsclerotic
hippocampus. However, pyramidal cells in this region are relatively spared in the
process of neuronal loss that characterizes hippocampal sclerosis (Mathern et al.,
22 J. Engel Jr et al.
1997), and these remaining neurons may play a more important role in mesial tem-
poral lobe epilepsy (Williamson and Spencer, 1994).
The third segment of the hippocampal trisynaptic pathway is its primary output,
from CA1 to the adjacent subiculum. The subiculum in turn projects back to the
entorhinal cortex, completing a closed hippocampal excitatory loop. This pathway
is not absolutely unidirectional, however; there are also signicant connections
from entorhinal cortex to CA1, and some bres even reach CA3. The subiculum
predominantly projects to perirhinal and piriform cortex, but also to the mamil-
lary bodies. The perirhinal cortex has connections with frontal motor cortical
areas, whereas the piriform cortex projects reciprocally to olfactory areas and
the brain stem. These projections are predominantly responsible for mediating the
clinical manifestations of temporal lobe seizures; ictal discharges conned to the
hippocampus may have no overt signs and symptoms (Sperling and OConnor,
1990). Parahippocampal areas may also become an important site of seizure gen-
eration; success in eliminating seizures by amygdalohippocampectomy, performed
to treat medically intractable temporal lobe epilepsy, appears to depend upon the
amount of parahippocampal tissue removed (Siegel et al., 1990).
CA1 axons also project via the mbria and fornix along the traditional Papez
circuit to the septal area, midline thalamus, amygdala, hypothalamus and brain
stem autonomic centres. The amygdala, hypothalamic and brain stem projections
mediate aspects of emotional behaviours, and the hippocampus may play a role in
the adrenohypothalamicpituitary axis, in view of the fact that adrenal steroid
receptors are abundant in the hippocampus (Gould et al., 1991).
The hippocampus also receives direct cholinergic and GABAergic input from the
septal area, which mediates the classical hippocampal theta rhythm, and these pro-
jections may have disinhibitory as well as inhibitory inuences on epileptic activ-
ity. With respect to brain stem aerents, direct noradrenergic input to the
hippocampus from the locus coeruleus can be excitatory (Madison and Nicoll,
1986), while serotonergic input from the raph nucleus and dopaminergic input
from the substantia nigra and ventral tegmental area are predominantly inhibitory
(Andrade and Nicoll, 1987), but biogenic amine eects can be varied, depending
on the postsynaptic target. Indirect aerent inuences on the hippocampus, via
entorhinal cortex, derive from a large area of neocortex, including frontal limbic
regions responsible for goal-directed behaviour, cingulate cortex inuencing
memory, all sensory cortical areas for integration of polymodal sensory informa-
tion, and perirhinal and piriform cortex, as well as from amygdala.
Hippocampal sclerosis
Much is known about the structural and functional abnormalities that exist in the
epileptic hippocampus, due in part to the ready availability of human tissue from
23 Anatomical substrates of behavioural disturbances
epilepsy surgery programmes, and in part to numerous animal models of this con-
dition; however, the cause of these abnormalities, and the reasons why these abnor-
malities should result in spontaneous seizures, are not well understood. The
characteristic nding in hippocampal sclerosis is cell loss, most prominent in the
dentate hilus and CA1, but it is also marked in CA3 and, to a lesser extent, the
dentate gyrus and subiculum (Mathern et al., 1997). Although specic hilar inter-
neurons are lost, namely those containing somatostatin and neuropeptide Y, in
general there is relative preservation of hippocampal inhibitory interneurons, com-
pared to principal cells (Babb, 1992). The greatest proportion of hilar cell loss is
from mossy cell death, which results in loss of excitatory inputs to dentate granule
cell proximal dendrites, and consequent mossy bre sprouting. These collateral
granule cell axons, which caneasily be seenwith Timms stain, reinnervate the inner
molecular layer, but most likely end on inhibitory interneurons, as well as on
granule cell dendrites (Figure 3.3). There is also evidence of inhibitory interneuron
sprouting (Babb, 1992). Electrophysiological studies in patients have conrmed
that inhibition is actually increased, interictally, in the epileptiformsclerotic hippo-
campus (Wilson et al., 1998). This enhanced inhibition may act to suppress seizure
generation; however, it has been postulated that such enhanced inhibition together
with enhanced excitation, brought about by pathological neuronal reorganization,
results in a predisposition to abnormal neuronal hypersynchronization (Figure
3.4). In addition, single input bres sprout to innervate wider targets, and reduc-
tion of the dendritic domain places more excitatory input closer to the soma and
axon hillock. Hypersynchronization underlies interictal epileptiform EEG spikes,
but is also the hallmark of the typically hypersynchronous hippocampal ictal EEG
onset (Velasco et al., 2000). Consequently, enhanced inhibition in the sclerotic hip-
pocampus may contribute to its epileptogenicity (Engel et al., 1997). During the
process of epileptogenesis, once the local anatomical substrate for abnormal hyper-
synchrony has been established, these hypersynchronous discharges may act to
kindle distant structures responsible for the manifestation of ictal behaviour.
Amygdala
The amygdala has been implicated in a number of important primal functions,
including fear, aggression, learning, reproduction and appetitive drive. In contrast
to the hippocampus, it consists of a number of loosely organized subnuclei with
diverse neurons and cytoarchitecture, which have been divided into an olfactory
group, a centromedial group and a basolateral group (Gloor, 1997; Schwartzkroin
and McIntyre, 1997). The olfactory group has predominant reciprocal connections
with the hippocampus, piriform cortex and olfactory bulb, with less strong con-
nections to the septal area, brain stem nuclei, hypothalamus, thalamus and the
24 J. Engel Jr et al.
25 Anatomical substrates of behavioural disturbances
Normal
FD
Hilus
OML
MML
IML
SG
Mossy fibre reinnervation
FD
Hilus
CML
MML
IML
SG
Figure 3.3. Schematic diagram showing the reciprocal innervation between principal neurons of the
hilus (squares), and dentate granule cells of the hippocampal formation (circles). In the
normal situation shown above, granule cell axons (mossy fibres) synapse on principal
neurons of the hilus and CA3, while hilar neurons make synapses on proximal dendrites
of the granule cells. The reinnervation that occurs with neuronal loss, such as occurs with
hippocampal sclerosis, is shown below. When hilar input to granule cells is decreased, the
mossy fibres of surviving granule cells sprout and make contact with the vacated
postsynaptic membrane, creating monosynaptic recurrent excitatory circuits. FD, fascia
dentata; OML, outer molecular layer; MML, middle molecular layer; IML, inner molecular
layer; SG, stratum granulosum. (With permission from Engel, 1990.)
nucleus accumbens of the basal forebrain. The centromedial group has reciprocal
connections with extensive areas of the brain stem, including the periaqueductal
grey, but also with the entorhinal cortex, hippocampus, hypothalamus, thalamus
and striatum. The basolateral group has strong reciprocal connections with soma-
tosensory and motor cortex, with only weak brain stem connections. It is likely,
therefore, that the centromedial nuclei, particularly the central nucleus of the
amygdala, are most important in mediating aective ictal and interictal symptoms
observed with amygdala involvement in the propagation of epileptiform dis-
charges.
26 J. Engel Jr et al.
Figure 3.4. Reciprocal innervation of a hypothetical neuronal system is shown schematically above.
This is the typical synaptic organization of neocortex and hippocampus. Excitatory afferent
input terminates on the dendrites of principal neurons (filled triangles). Axon collaterals
from these principal neurons terminate on inhibitory interneurons (empty circles), which
in turn make hyperpolarizing synapses on the soma of the same, and adjacent (not
shown), principal neurons. With cell loss, a number of synaptic reorganizations are likely
to occur, as shown schematically on the right. Fewer afferent input fibres sprout to
innervate more principal neurons, predisposing to hypersynchronization. Because the
dendrites of principal neurons are shorter, these excitatory influences are closer to the
axon hillock and more likely to induce neuronal firing. Neuronal excitability is further
increased by the establishment of monosynaptic excitatory recurrent circuits, as also
shown in Figure 3.3. Although not definitively demonstrated, it is possible that inhibitory
interneurons sprout terminals to produce more powerful, and/or more extensive,
recurrent inhibitory influences, further enhancing the potential for hypersynchronization.
(With permission from Engel, 1990.)
Interhemispheric limbic connections
Due to the enormous expansion of the human cerebral cortex in comparison with
that of subprimate mammals, the limbic structures, with the exception of the cin-
gulate gyrus, have moved far from the midline. Interhemispheric connections are
necessarily increased in length and changed in trajectory by this major reorganiza-
tion (Wilson, 1995). Three pathways exist for interhemispheric communication
between limbic areas: corpus callosum, anterior commissure and hippocampal
commissure.
Corpus callosum
The corpus callosum carries bres from the cingulate gyrus, but also from the
medial orbital frontal cortex, the posterior two-thirds of the temporal cortex and
the posterior parahippocampal gyrus (Demeter et al., 1990). As a component of the
Papez circuit, cingulate cortex is in a position to relay output activity from limbic
structures to the contralateral hemisphere, as well as back to the ipsilateral ento-
rhinal cortex. The callosal connections of the orbital frontal cortex are signicant
for interhemispheric limbic interaction because of the major projection this area
receives from the amygdala (Amaral et al., 1984).
Anterior commissure
In primates, the anterior commissure carries bres from olfactory and other areas
of the orbital frontal cortex, piriform cortex, the anterior third of the temporal
cortex and posterior parahippocampal gyrus, and a few sparse bres from the cor-
tical and deep subnuclei of the amygdala. For the most part, the widespread corti-
cal projections of the amygdala are ipsilateral only, and the small interhemispheric
projection which exists terminates in contralateral temporal cortex, not in contra-
lateral amygdala (Demeter et al., 1990).
Hippocampal commissure
Phylogenetically, commissural bres have constituted a major group of extrinsic
connections joining the two hippocampi and their subelds, but the connections are
substantially reduced in the human brain. There are two components of the inter-
hemispheric pathway: the ventral hippocampal commissure, which carries bres
from the hippocampal formation, and the dorsal hippocampal commissure, which
carries bres from the entorhinal cortex, presubiculum and parahippocampal
gyrus. In the subhuman primate, the ventral hippocampal commissure is greatly
reduced and limited to the anterior hippocampus, while the dorsal commissure
carries primarily presubicular and parahippocampal gyrus bres to the contralat-
eral entorhinal cortex (Amaral et al., 1984). In the human brain, dissection of
27 Anatomical substrates of behavioural disturbances
autopsy specimens has revealed the presence of a dorsal hippocampal commissure,
but not the ventral (Gloor et al., 1993), and electrophysiology studies (Wilson et al.,
1990, 1991) provide additional functional evidence that hippocampal connections
are absent in humans.
The relative reduction of commissural interchange from lower animal to primate
may provide the basis for the hemispheric specialization unique to the human
brain, particularly relating to dierences in verbal versus spatial memory perfor-
mance, which is lateralized to the left and right temporal lobes, respectively
(Amaral et al., 1984; Demeter et al., 1985). The absence of any signicant inter-
hemispheric amygdala connections suggests that emotion and aect may have sim-
ilarly lateralized or hemispherically specialized components (Demeter et al., 1990),
which has important implications for patients with temporal lobe seizures.
Interictal and ictal behavioural and emotional manifestations of mesial temporal
lobe epilepsy can be very dierent depending on whether the epileptogenic abnor-
mality is on the right or left.
Possible anatomic substrates of some behavioural disturbances associated
with epilepsy
The behavioural disturbances most associated with epilepsy include depression
and other aective disorders and personality traits, schizophrenia, aggressivity and
anxiety.
Affective disturbances
The prevalence of aective disorders is higher among epileptic patients than the
general population, or neurological patients with the same degree of disability
(Mendez et al., 1986). No neurobiological mechanism has successfully explained
this correlation between aective disorders and epilepsy. However, several hypoth-
eses have been proposed.
Reciprocal connections between limbic structures and those brain stem nuclei
with biogenic amine and opioid projections to the forebrain provide ample path-
ways for the mediation of depression and other abnormalities of aect. In particu-
lar, endogenous opioids are known to be elevated during seizures (Vindrola et al.,
1981), and mu-opiate-receptor binding is increased in the human epileptic tempo-
ral lobe (Frost et al., 1988). It is postulated that opioid peptides act as endogenous
euphorogens, which may explain why electroconvulsive treatment is an eective
therapy for depression (Kline et al., 1977). Conceivably, patients with frequent
limbic seizures develop a dependence on high levels of circulating opioid peptides,
and the appearance of aective symptoms reects a form of withdrawal when sei-
zures do not regularly recur (Engel et al., 1991). Indeed, transient depression lasting
28 J. Engel Jr et al.
a few months is commonly observed in patients who become suddenly seizure-free
following successful surgical treatment (Glosser et al., 2000; Krahn et al., 1996).
Depression is more common among patients with complex partial seizures than
with generalized seizures (Devinsky and Vasquez, 1993). This nding is not sur-
prising, because complex partial seizures originate from those limbic structures in
which dysfunction has been strongly correlated with depression. For example, hip-
pocampal damage after stress, induced by elevated levels of cortisol, has been pro-
posed as a possible mechanism for depression (Stokes, 1995). Aective symptoms
have been associated with dysfunction in a circuit that travels from the anterior cin-
gulate and medial orbito-frontal cortices to the ventral striatum (nucleus accum-
bens and olfactory tubercle), the ventral pallidum, the medial dorsal nucleus of the
thalamus, and back to the anterior cingulate and medial orbito-frontal cortices
(Alexander et al., 1990; Ketter et al., 1996). Abnormalities in the components of this
circuit have also been described with complex partial seizures (Bromeld et al.,
1992; Savic et al., 1997). A common nding among patients with a primary aec-
tive disturbance, as well as in epileptic patients with depression, is decreased meta-
bolic activity of the prefrontal cortex (Bromeld et al., 1992; Rogers et al., 1998).
Whether the nding of decreased prefrontal metabolism is a consequence of the
depression in patients with complex partial seizures, or alternatively, complex
partial seizures have secondarily aected the activity of the prefrontal cortex, with
a consequent development of depression, is not clear. However, based on the exis-
tence of strong connections between the mesial temporal structures and the pre-
frontal cortex, this second hypothesis is anatomically and functionally plausible
(Drevets, 1998). Abnormalities of metabolism in the amygdala and anterior cingu-
late have also been repeatedly reported to be associated with depression (Drevets,
1998; Mayberg et al., 1994) and temporal lobe epilepsy (Henry et al., 1993a,b).
Therefore, complex partial seizures could induce functional changes in limbic areas
that overlap with those structures involved in the manifestations of aective
psychopathology.
Schizophrenia
The dopaminergic hypothesis of schizophrenia originated initially from two indi-
rect ndings: except for clozapine, all eective antipsychotics are blockers of D2
dopamine (DA) receptors, and stimulants that increase DA release in the limbic
system frequently induce psychosis (Farde, 1997). This hypothesis has been applied
mostly to the positive symptoms of schizophrenia such as hallucinations and delu-
sions, because typical antipsychotic medications, which act almost exclusively on
D2 receptors, are not very ecient in treating negative symptoms and thought dis-
organization. This last fact, together with the delay in action of antipsychotic med-
ications (although D2 antagonism is immediate), put the dopaminergic theory into
29 Anatomical substrates of behavioural disturbances
question (Goldstein and Deutch, 1992). However, recent studies have shown that
schizophrenic patients experience increased DA release after amphetamine chal-
lenge, when compared with control subjects, and this hyperactivity of DA trans-
mission was found to be associated with activation of psychotic symptomatology
(Laruelle and Abi-Dargham, 1999).
In cats, kindling of the dopaminergic ventral tegmental area of the brain stem,
which projects to the nucleus accumbens of the basal forebrain, does not produce
seizures, but alters behaviour in a way that has been interpreted to resemble aspects
of schizophrenia (Stevens, 1973). The nucleus accumbens and the ventral tegmen-
tal area both receive input from the mesial temporal limbic system, and are thus
subject to kindling-like aerent input from seizures originating from the hippo-
campus or perihippocampal regions. An enduring eect of limbic seizures on
dopaminergic function has been demonstrated experimentally by the fact that
amygdala kindling in cats enhances methamphetamine-induced stereotypy, sug-
gesting upregulation of DA receptors (Sato, 1983), although the location of these
receptors has not been identied.
Furthermore, disturbances of limbic structures have been associated with
schizophrenia (Weinberger et al., 1992). Pathological studies of brains of schizo-
phrenic patients revealed a disorganization of the typical laminar structure of the
hippocampus (Kovelman and Scheibel, 1984). Studies from discordant monozy-
gotic twins (i.e., one twin is aected by schizophrenia whereas the other is healthy),
have shown that the schizophrenic twin had smaller hippocampi, and that the size
of the hippocampi correlated with hypofrontality (assessed by blood-ow studies
during performance of the Wisconsin Card Sorting Test) in the aected twins
(Weinberger et al., 1992). This nding is consistent with the possibility that path-
ological abnormalities in the hippocampus might result in a widespread disruption
of other corticolimbic structures, such as the prefrontal cortex, with a consequent
development of schizophrenic symptoms.
Aggression
Kindling is known to induce postictal hyperreactivity (which has been mistaken for
aggression), as well as other postictal behaviours which can all be modulated by
pretreatment with opioid agonists and antagonists (Caldecott-Hazard and Engel,
1987). Some of these behaviours may be direct opioid eects, while others may rep-
resent withdrawal phenomena. Mesial limbic structures project to the brain stem
periaqueductal grey, a major source of forebrain opioid projections, and this area
has been implicated in an experimental model of aggression in cats termed defen-
sive rage. Electrical stimulation of this area, and of the lateral hypothalamus, pro-
duces a stereotypic Halloween cat response, with piloerection, arched back,
pupillary dilation and attack behaviour, lasting only as long as the duration of the
30 J. Engel Jr et al.
stimulus (Bandler, 1988). This behaviour can be modulated by lesions, or stimula-
tion, of hippocampus and amygdala.
Production of experimental temporal lobe epilepsy in cats, by hippocampal
injection of kainic acid which causes a lesion similar to hippocampal sclerosis, has
a profound eect on this behavioural model (Engel et al., 1991; Grith et al., 1987).
Immediately following kainic acid injection, cats experience continuous limbic sei-
zures lasting many hours, and during this time also exhibit spontaneous defensive
rage. This is followed by a latent period when the damaged hippocampus under-
goes epileptogenic neuronal reorganization and ultimately gives rise to spontane-
ous limbic seizures. During the latent period, which can last from weeks to months,
there are no spontaneous seizures and behaviour is normal. Once spontaneous sei-
zures occur, however, not only is the threshold to electrical stimulation-induced
defensive rage lowered, but cats also exhibit defensive rage in response to environ-
mental stimuli, in the absence of periaqueductal grey or lateral hypothalamic
electrical stimulation. Furthermore, the defensive rage reaction can be elicited only
when the cat is approached contralateral to the epileptic hippocampus (Figure 3.5).
Clearly, hippocampal epileptogenesis has induced distant structural changes, pre-
sumably involving amygdala, hypothalamus and periaqueductal grey, resulting in
enduring aberrant behaviour.
These experiments should not be interpreted to suggest that patients with mesial
temporal lobe epilepsy are susceptible to defensive rage or other aggressive behavi-
ours. Defensive rage is a hardwired species-specic behaviour of the cat, and the
brain stem and diencephalic structures responsible for its manifestation undoubt-
edly come under strong neocortical control in the human. A variety of human cor-
relates, other than rage, might be anticipated, including impulsivity, insecurity,
paranoia and perhaps even depression.
Anxiety
The amygdala is believed to provide an aective tone to perceptions and experi-
ences, and human lesion and neuroimaging studies have demonstrated a role for
the amygdala in verbal and nonverbal recognition of emotions (Adolphs et al.,
1998; Hariri et al., 2000), as well as in fear conditioning (Furmark et al., 1997) and
provoked anxiety (Davidson et al., 1999). In animals, there is also extensive evi-
dence to support a role for the amygdala in fear and fear conditioning (LeDoux,
1993). Other structures that have been found to be repeatedly correlated with
anxiety are the orbito-frontal cortex, anterior insula and anterior cingulate
(Malizia, 1999).
The projections of the central nucleus of the amygdala to the locus coeruleus and
hypothalamus appear to be responsible for the arousal and autonomic responses
associated with fear and anxiety (Goldstein et al., 1996) which, together with the
31 Anatomical substrates of behavioural disturbances
32 J. Engel Jr et al.
Figure 3.5. Demonstration of lateralized hypereactivity. Photographs were taken from a videotape
recording of a cat with a chronic, kainic acid-induced epileptogenic lesion in the left
hippocampus. Top: when the cat was approached on the left (ipsilateral) side, it
responded by rubbing up against the offered hand and purring. Bottom: when an attempt
was made to touch the cat on the right (contralateral) side, it assumed a defensive
posture (ears back, piloerection, and pupillary dilatation), hissed, and struck at the offered
hand with the right paw. (With permission from Engel et al., 1991.)
well-established function of the amygdala in emotion recognition and fear condi-
tioning, imply a key role of the amygdala in the integration of behavioural and
neuroendocrine components of the stress response. Amygdala-kindled rats show
behaviours suggestive of increased fear, such as decreased exploration in the open
eld (Depaulis et al., 1997). Consequently, there are ample anatomical and patho-
physiological substrates that can be invoked to explain both ictal and interictal
anxiety as well as panic disorder in patients with temporal lobe epilepsy.
Acknowledgements
Original research reported by the author was supported in part by Grants NS-
02808, NS-15654, NS-33310, and GM-24839, from the National Institutes of
Health, and Contract DE-AC03-76-SF00012 from the Department of Energy.
REFERENCES
Acsdy, L., Kamondi, A., Sik, A., Freund, T. and Buzski, G. (1998). GABAergic cells are the major
postsynaptic targets of mossy bers in the rat hippocampus. J Neurosci, 18, 3386403.
Adolphs, R., Tranel, D. and Damasio, A.R. (1998). The human amygdala in social judgment [see
comments]. Nature, 393, 4704.
Alexander, G.E., Crutcher, M.D. and DeLong, M.R. (1990). Basal ganglia-thalamocortical cir-
cuits: parallel substrates for motor, oculomotor, prefrontal and limbic functions. Prog
Brain Res, 85, 11946.
Amaral, D.G., Insausti, R. and Cowan, W.M. (1984). The commissural connections of the
monkey hippocampal formation. J Comp Neurol, 224, 30736.
Andrade, R. and Nicoll, R.A. (1987). Pharmacologically distinct actions of serotonin on single
pyramidal neurones of the rat hippocampus recorded in vitro. J Physiol, 394, 99124.
Babb, T.L. (1992). GABA neurons, synapses and inhibition in human hippocampal epilepsy. In
Epilepsy and Inhibition, ed. E.-J. Speckmann and M.J. Gutnick, pp. 37597. Munich: Urban and
Schwarzenberg Press.
Bandler, R. (1988). Brain mechanisms of aggression as revealed by electrical and chemical
stimulation: suggestion of a central role for the midbrain periaqueductal gray region. In
Progress in Psychobiology, ed. A. Epstein & A. Morrison, pp. 67153. New York: Academic
Press.
Broca, P. (1878). Anatomie compare des circonvolutions cerebrales. Le grand lobe limbique et
la scissure limbique dans le serie des mammiferes. Rev Anthropol, Ser 2, 1, 385498.
Bromeld, E.B., Altshuler, L., Leiderman, D.B. et al. (1992). Cerebral metabolism and depression
in patients with complex partial seizures. Arch Neurol, 49, 61723.
Caldecott-Hazard, S. and Engel, J. Jr (1987). Limbic postictal events: Anatomical substrates and
opioid receptor involvement. Prog Neuropsychopharmacol Biol Psychiatry, 11, 389418.
Corcoran, M.E. and Mosh, S.L. (Eds) (1998). Kindling V. New York: Plenum Press.
33 Anatomical substrates of behavioural disturbances
Davidson, R.J., Abercrombie, H., Nitschke, J.B. and Putnam, K. (1999). Regional brain function,
emotion and disorders of emotion. Curr Opin Neurobiol, 9, 22834.
Demeter, S., Rosene, D.L. and Van Hoesen, G.W. (1985). Interhemispheric pathways of the hip-
pocampal formation. Presubiculum, entorhinal and posterior hippocampal cortices in the
rhesus monkey: the structure and function of the hippocampal commissures. J Comp Neurol,
233, 3047.
Demeter, S., Rosene, D.L. and Van Hoesen, G.W. (1990). Fields of origins and pathways of the
interhemispheric commissures in the temporal lobe of macaques. J Comp Neurol, 302, 29.
Depaulis, A., Helfer, V., Deransart, C. and Marescaux, C. (1997). Anxiogenic-like consequences
in animal models of complex partial seizures. Neurosci Biobehav Rev, 21, 76774.
Devinsky, O. and Vazquez, B. (1993). Behavioral changes associated with epilepsy. Neurol Clin,
11, 12749.
Drevets, W.C. (1998). Functional neuroimaging studies of depression: the anatomy of melancho-
lia. Annu Rev Med, 49, 34161.
Engel, J. Jr (1990). Functional explorations of the human epileptic brain and their therapeutic
implications. Electroencephalogr Clin Neurophysiol, 76, 296316.
Engel, J. Jr (1998). Etiology as a risk factor for medically refractory epilepsy: a case for early sur-
gical intervention. Neurology, 51, 12434.
Engel, J. Jr and Shewmon, D.A. (1991). Impact of the kindling phenomenon on clinical epilep-
tology. In Kindling and Synaptic Plasticity: The Legacy of Graham Goddard, ed. F. Morrell, pp.
195210. Cambridge, MA: Birkhuser Boston.
Engel, J. Jr, Caldecott-Hazard, S. and Bandler, R. (1986). Neurobiology of behavior: anatomic
and physiologic implications related to epilepsy. Epilepsia, 27(Suppl. 2), S3S11.
Engel, J. Jr, Bandler, R., Grith, N.C. and Caldecott-Hazard, S. (1991). Neurobiological evidence
for epilepsy-induced interictal disturbances. In Advances in Neurology, ed. D. Smith, D.
Treiman & M. Trimble, pp. 97111. New York: Raven Press. Vol. 55.
Engel, J. Jr, Dichter, M. and Schwartzkroin, P. (1997). Basic mechanisms of human epilepsy. In
Epilepsy: A Comprehensive Textbook, ed. J. Engel Jr and T.A. Pedley, pp. 499512. Philadelphia,
New York: Lippincott-Raven.
Farde, L. (1997). Brain imaging of schizophrenia the dopamine hypothesis. Schizophr Res, 28,
15762.
Frost, J.J., Mayberg, H.S., Fisher, R.S. et al. (1988). Mu-opiate receptors measured by positron
emission tomography are increased in temporal lobe epilepsy. Ann Neurol, 23, 2317.
Furmark, T., Fischer, H., Wik, G., Larsson, M. and Fredrikson, M. (1997). The amygdala and
individual dierences in human fear conditioning. Neuroreport, 8, 395760.
Gloor, P. (1997). The Temporal Lobe and the Limbic System. New York: Oxford University Press.
Gloor, P., Salanova, V., Olivier, A. and Quesney, L.F. (1993). The human dorsal hippocampal
commissure: an anatomically identiable and functional pathway. Brain, 116, 124973.
Glosser, G., Zwil, A.S., Glosser, D.S., OConnor, M.J. and Sperling, M.R. (2000). Psychiatric
aspects of temporal lobe epilepsy before and after anterior temporal lobectomy [In Process
Citation]. J Neurol Neurosurg Psychiatry, 68, 538.
Goldstein, M. and Deutch, A.Y. (1992). Dopaminergic mechanisms in the pathogenesis of
schizophrenia. FASEB J, 6, 241321.
34 J. Engel Jr et al.
Goldstein, L.E., Rasmusson, A.M., Bunney, B.S. and Roth, R.H. (1996). Role of the amygdala in
the coordination of behavioral, neuroendocrine, and prefrontal cortical monoamine
responses to psychological stress in the rat. J Neurosci, 16, 478798.
Gould, E., Woolley, C.S and McEwen, B.S. (1991). The hippocampal formation: morphological
changes induced by thyroid, gonadal, and adrenal hormones. Psychoneuroendocrinology, 16,
6784.
Grith, N., Engel, J. Jr and Bandler, R. (1987). Ictal and enduring interictal disturbances in emo-
tional behaviour in an animal model of temporal lobe epilepsy. Brain Res, 400, 3604.
Hariri, A.R., Bookheimer, S.Y. and Mazziotta, J.C. (2000). Modulating emotional responses:
eects of a neocortical network on the limbic system. Neuroreport, 11, 438.
Henry, T.R., Engel, J. Jr and Mazziotta, J.C. (1993a). Clinical evaluation of interictal uorine-18-
uorodeoxyglucose PET in partial epilepsy. J Nucl Med, 34, 18928.
Henry, T.R., Mazziota, J.C. and Engel, J. Jr (1993b). Interictal metabolic anatomy of mesial tem-
poral lobe epilepsy. Arch Neurol, 50, 5829.
Herzog, A.G. (1997). Disorders of reproduction and fertility. In Epilepsy: A Comprehensive
Textbook, ed. J. Engel Jr & T.A. Pedley, pp. 201319. Philadelphia, New York: Lippincott-Raven.
Ketter, T.A., George, M.S., Kimbrell, T.A., Benson, B.E. and Post, R.M. (1996). Functional brain
imaging, limbic function, and aective disorders. Neuroscientist, 2, 5565.
Kline, N.S., Li, C.H., Lehmann, H.E., Lajtha, A., Laski, E. andCooper, T. (1977). Beta-endorphine-
induced changes in schizophrenic and depressed patients, Arch Gen Psychiatry, 34, 111113.
Kovelman, J.A. and Scheibel, A.B. (1984). A neurohistological correlate of schizophrenia. Biol
Psychiatry, 19, 160121.
Krahn, L.E., Rummans, T.A. and Peterson, G.C. (1996). Psychiatric implications of surgical treat-
ment of epilepsy. Mayo Clin Proc, 71, 12014.
Laruelle, M. and Abi-Dargham, A. (1999). Dopamine as the wind of the psychotic re: new evi-
dence from brain imaging studies. J Psychopharmacol, 13, 35871.
LeDoux, J.E. (1993). Emotional memory systems in the brain. Behav Brain Res, 58, 6979.
Lothman, E.W., Bertram, E.H. III and Stringer, J.L. (1991). Functional anatomy of hippocampal
seizures. Prog Neurobiol, 37, 182.
Maclean, P.D. (1952). Some psychiatric implications of physiological studies on frontotemporal
portion of limbic system (visceral brain). EEG Clin Neurophysiol, 4, 40718.
Madison, D.V. and Nicoll, R.A. (1986). Actions of noradrenaline recorded intracellularly in rat
hippocampal CA1 pyramidal neurones, in vitro. J Physiol (Lond), 372, 22144.
Malizia, A.L. (1999). What do brain imaging studies tell us about anxiety disorders? J
Psychopharmacol, 13, 3728.
Mathern, G.W., Babb, T.L. and Armstrong, D.L. (1997). Hippocampal sclerosis. In Epilepsy: A
Comprehensive Textbook, ed. J. Engel Jr & T.A. Pedley, pp. 13355. Philadelphia, New York:
Lippincott-Raven.
Mayberg, H.S, Lewis, P.J., Regenold, W. and Wagner, H.N.J. (1994). Paralimbic hypoperfusion in
unipolar depression. J Nucl Med, 35, 92934.
McIntyre, D.C. and Plant, J.R. (1993). Long-lasting changes in the origin of spontaneous dis-
charges from amygdala-kindled rats: piriform vs. perirhinal cortex in vitro. Brain Res, 624,
26876.
35 Anatomical substrates of behavioural disturbances
Mendez, M.F., Cummings, J.L. and Benson, F. (1986). Depression in epilepsy: signicance and
phenomenology. Arch Neurol, 43, 76670.
Morrell, F., Whisler, W.W., Smith, M.C. et al. (1995). Landau-Klener syndrome: treatment with
subpial intracortical transection. Brain, 118, 152946.
OKeefe, J. and Nadel, L. (1978). The Hippocampus as a Cognitive Map. Oxford: Clarendon Press.
Papez, J.W. (1937). A proposed mechanism of emotion. Arch Neurol Psychiatry, 38, 72543.
Piredda, S. and Gale, K. (1985). A crucial epileptogenic site in the deep prepiriform cortex.
Nature, 317, 62325.
Rogers, M.A., Bradshaw, J.L., Pantelis, C. and Phillips, J.G. (1998). Frontostriatal decits in uni-
polar major depression. Brain Res Bull, 47, 297310.
Sato, M. (1983). Long-lasting hypersensitivity to methamphetamine following amygdaloid kin-
dling in cats: the relationship between limbic epilepsy and the psychotic state. Biol Psychiatry,
18, 52536.
Savic, I., Altshuler, L., Baxter, L. and Engel, J. Jr (1997). Pattern of interictal hypometabolism in
PET scans with udeoxyglucose F18 reects prior seizure types in patients with mesial tempo-
ral lobe seizures. Arch Neurol, 54, 12936.
Schwartzkroin, P.A. and McIntyre, D.C. (1997). Limbic anatomy and physiology. In Epilepsy: A
Comprehensive Textbook, ed. J. Engel Jr and T.A. Pedley, pp. 32340. Philadelphia, New York:
Lippincott-Raven.
Shouse, M.N., Martins da Silva, A. and Sammaritano, M. (1997). Sleep. In Epilepsy: A
Comprehensive Textbook, ed. J. Engel Jr and T.A. Pedley, pp. 192942. Philadelphia, New York:
Lippincott-Raven.
Siegel, A.M., Wieser, H.G., Wichmann, W. and Yasargil, G.M. (1990). Relationships between MR-
imaged total amount of tissue removed, resection scores of specic mesiobasal limbic subcom-
partments and clinical outcome following selective amygdalohippocampectomy. Epilepsy Res,
6, 5665.
Sloviter, R.S. and Damiano, B.P. (1981). Sustained electrical stimulation of the perforant path
duplicates kainate-induced electrophysiological eect and hippocampal damage in rats.
Neurosci Lett, 24, 27984.
Sperling, M.R. and OConnor, M.J. (1990). Auras and subclinical seizures: characteristics and
prognostic signicance. Ann Neurol, 28, 3208.
Stevens, J.R. (1973). An anatomy of schizophrenia. Arch Gen Psychiatry, 29, 17789.
Stokes, P.E. (1995). The potential role of excessive cortisol induced by HPA hyperfunction in the
pathogenesis of depression. Eur Neuropsychopharmacol, 5, 7782.
Velasco, A.L., Wilson, C.L., Babb, T.L. and Engel, J. Jr (2000). Functional and anatomic correlates
of two frequently observed temporal lobe seizure-onset patterns. Neural Plasticity, 7, 4963.
Vindrola, O., Briones, R., Asai, M. and Fernandez-Guardiola, A. (1981). Brain content of leu 5-
and met 5-enkephalin changes independently during the development of kindling in the rat.
Neurosci Lett, 26, 12530.
Weinberger, D.R. Berman, K.F., Suddath, R. and Torrey, E.F. (1992). Evidence of dysfunction of
a prefrontal-limbic network in schizophrenia: a magnetic resonance imaging and regional
cerebral blood ow study of discordant monozygotic twins. Am J Psychiatry, 149, 8907.
36 J. Engel Jr et al.
Wieser, H.G. (1988). Human limbic seizures: EEG studies, origin, and patterns of spread. In
Anatomy of Epileptogenesis, ed. B.S. Meldrum, J.A. Ferrendelli and H.G. Wieser, pp. 12738.
London: John Libbey.
Williamson, A. and Spencer, D.D. (1994). Electrophysiological characterization of CA2 pyrami-
dal cells from epileptic humans. Hippocampus, 2, 22637.
Williamson, P.D. and Engel, J. Jr (1997). Complex partial seizures. In Epilepsy: A Comprehensive
Textbook, ed. J. Engel Jr and T.A. Pedley, pp. 55766. Philadelphia, New York: Lippincott-
Raven.
Wilson, C.L. (1995). Functional pathways underlying ipsilateral and contralateral spread of tem-
poral lobe seizures. In Epilepsy and the Corpus Callosum, ed. A. Reeves, Vol. II, pp. 15373. New
York: Plenum Press.
Wilson, C.L., Isokawa-Akesson, M., Babb, T.L. and Crandall, P.H. (1990). Functional connec-
tions in the human temporal lobe: I. Analysis of limbic system pathways using neuronal activ-
ity evoked by electrical stimulation. Exp Brain Res, 82, 27992.
Wilson, C.L., Isokawa, M., Babb, T.L., Crandall, P.H., Levesque, M.F. and Engel, J. Jr (1991).
Functional connections in the human temporal lobe: II. Evidence for a loss of functional
linkage between contralateral limbic structures. Exp Brain Res, 85, 17487.
Wilson, C.L., Khan, S.U., Engel, J. Jr, Isokawa, M., Babb, T.L. and Behnke, E.J. (1998). Paired
pulse suppression and facilitation in human epileptogenic hippocampal formation. Epilepsy
Res, 31, 21130.
Wong, R.K.S. and Traub, R.D. (1983). Synchronized burst discharge in disinhibited hippocam-
pal slice. I. Initiation in CA2-CA3 region. J Neurophysiol, 49, 44258.
Xie, C.W., Morrisett, R.A. and Lewis, D.V. (1992). Mu opioid-receptor mediated modulation of
synaptic currents in the dentate granule cells of rat hippocampus. J Neurophysiol, 68, 111320.
37 Anatomical substrates of behavioural disturbances
Part II
Clinical aspects
4
The psychiatry of idiopathic generalized
epilepsy
Dieter Janz
Burgunder Strasse 8, Berlin, Germany
Temporal Lobe Epilepsy (TLE) is associated with an increase in psychiatric mor-
bidity compared to other types of epilepsy. This hypothesis is central to the discus-
sion of the relationship between epilepsy and psychopathology, according to the
statement of Anne Stub-Naylor, a young Danish doctor, in her thesis on Epilepsy
and psychiatric disorder a comorbidity study (1996). And she continues: Despite
the scarce evidence it appears to have become a dogma that such a relationship
exists. In contemporary medicine especially three schools have nourished this
view, represented by Elliot Slater, David Bear and Michael Trimble. Indeed, the
description of schizophrenia-like psychoses in patients with TLE (Slater et al.,
1963) has encouraged the expectation of nding a biological-based explanation
for schizophrenia. For this reason, the discussions on psychiatric disorder and espe-
cially those on psychoses in the case of epilepsy have long been dominated by the
limbic hypothesis.
It seems that the time has come to take a critical look at the above thesis, because
the understanding that other forms of epilepsy also show a clinical prole and a
biological quality is gaining ground. To have a stake in biological psychiatry, the
psychiatric aspects should be taken into consideration in regard to other forms and
syndromes of epilepsy as well.
For the following reasons, the syndromes of idiopathic generalized epilepsy
(IGE) are especially suitable for consideration. Grouped together they are at least
as frequent as the localization-related temporal and extra-temporal epilepsies.
Phenomenologically and electroencephalographically they are well dened; there-
fore they can easily be distinguished from focal epilepsies and symptomatic genera-
lized epilepsies. The coexistence of generalized and focal epilepsies is extremely
rare, although transitions from one to the other are possible due to a long and
severe course of disease, and they might well be relevant for the cause of psychiat-
ric disorder. The comparison of subsyndromes within families shows that the
various syndromes of idiopathic generalized epilepsy make up one genetic entity,
41
which diers clearly from genetic focal epilepsies in view of their phenotype (Janz,
1997) as well as from the molecular-genetic view (Sander et al., 2000).
According to the Commission on Classication and Terminology of the
International League against Epilepsy (1989), within IGE belong apart from
benign neonatal convulsions and benign myoclonic epilepsy in infancy, which
are irrelevant for our topic childhood absence epilepsy (CAE), juvenile absence
epilepsy (JAE), juvenile myoclonic epilepsy (JME) and epilepsy with grand mal sei-
zures (GTCS) on awakening (GMA). For these four syndromes of IGE of child-
hood and adolescence electroencephalographic regular or fast generalized
spike-wave discharges are the typical EEG patterns.
When scanning the literature for the frequency of psychiatric disorders in
various types of epilepsy, diculties of classifying become obvious. The earlier lit-
erature does not recognize the modern syndrome classication. Yet, in some recent
studies, especially those from the psychiatric side, the occurrence of generalized sei-
zures is equated with the supposition of generalized epilepsy (Mndez et al., 1993).
Epilepsy with exclusively grand mal seizures (TCS) can be taken for IGE, even
without EEG, if it concerns an epilepsy with grand mal predominantly on awaken-
ing (GMA). But such details are still frequently neglected, although the
International Classication has existed for 12 years.
Psychosis
Looking at the most serious, yet rarest, psychiatric disorder, the psychoses, and
paying attention to their frequency in generalized epilepsies in comparison to focal
epilepsies, most authors of the relevant literature (Table 4.1), found that psychoses
in focal epilepsies are slightly more frequent than in generalized epilepsies, but the
dierence proved to be signicant in only one study (Onuma, 1983). Two studies
using the International Classication did not nd the frequencies of psychoses in
IGE and in TLE to be dierent (Schmitz and Wolf, 1991, 1995; Sengoku et al.,
1997). With regard to specic epileptic syndromes, CAE was signicantly asso-
ciated with the presence of psychosis. CAE occurred in 7/28 (25%) of patients with
psychosis compared with 55/669 (8%) of patients without psychosis (Schmitz and
Wolf, 1991).
Relation to seizure type
With regard to specic seizure types, absences in combination with GTCS are asso-
ciated with an increased risk of psychosis. With this seizure combination psychosis
occurred in 11.3% (7/62) of patients in comparison with only 1% (1/91) of patients
with myoclonic jerks in combination with GTCS (Schmitz, 1988). This gure is in
good correspondence with ndings of Genton et al. (2000) who reported 5 (2.9%)
42 D. Janz
cases with psychotic disorders among 170 patients with JME. Unfortunately, we do
not know the inuence that other seizure variables, such as the length of the disease,
or frequency and intensity of the seizures, may have on these dierences between
the various IGE-syndromes.
In their retrospective studies of 697 and 611 consecutive patients with unselected
epilepsies (Wolf, 1976; Schmitz, 1988), Schmitz and Wolf (1995) found a puzzling
relation between absences and complex focal seizures regarding an increased risk
of psychosis. Contrary to this, patients with only GTCS had no increased risk of
43 Psychiatry of idiopathic generalized epilepsy
Table 4.1. Controlled studies on the frequency of psychosis in different types of epilepsy
Studies Type of psychosis/population Results
Gastaut (1956) Mixed GE, TLE, non-TLE: not dierent
Small et al. (1962) Schizophrenia TLE, 24%; non-TLE, 12%: not
signicantly dierent
Stevens (1966) History of admission to GE, 29%; TLE, 31%; FOC non-TLE, 8%
psychiatric hospital
Mignone et al. (1970) MMPI: schizophrenia score Psychomotor vs. nonpsychomotor: not
dierent
Bruens (1971) Mixed GE, 3.3%; SIM FOC, 0%; COM FOC,
3.5%; GETLE, 19%
Standage and Fenton (1975) Present State Examination TLE vs. non-TLE: not dierent
Shukla et al. (1979) Schizophrenia GE, 4%; TLE, 17%
Onuma (1983) Paranoid symptoms PGE, 3.5%; TLE, 9.2%; non-TLE, 2.5%:
PGE vs. TLE, signicantly dierent
Sengoku et al. (1983) Mixed, epilepsy institution GE, 4.3%; TLE, 6.0%; non-TLE, 2.0%:
GE vs. TLE not signicantly dierent
Schmitz (1988) Mixed, Department of Neurology IGE, 3.5%; FOC,4.2%; GEFOC, 4%:
not signicantly dierent
Sengoku et al. (1997) Mixed, Department of Psychiatry IGE, 19.4%; TLE, 15.2%: not
signicantly dierent
Bredkjaer et al. (1998) Nonorganic Psychomotor epilepsy 3.57 SIR, petit
Nonaective psychosis mal epilepsy 2.3 SIR
Notes:
GE, generalized epilepsies; PGE, primary GE; IGE, idiopathic GE; TLE, temporal lobe epilepsies; non-TLE,
nontemporal lobe epilepsies; FOC, focal epilepsies; SIM FOC, epilepsies with simple focal seizures; COM
FOC, epilepsies with complex partial seizures; SIR, standardized incidence ratios; MMPI, Minnesota
Multiphasic Personality Inventory.
psychosis. They speculated therefore that both seizure types testify to the ability of
a patients brain to produce epileptic events that are widespread but still limited. If
prolonged but strongly limited seizure discharge is indeed present during epileptic
psychoses of many types, as has been proposed, a brains ability to produce such a
discharge would be a prerequisite for psychosis. The brains of patients seem to have
learned the compromise in between if they cannot produce or give an epileptic all-
or-nothing response.
Ictal psychosis
The classical example of psychotic states due to epileptic events that are widespread
but still limited is the nonconvulsive status epilepticus. The symptoms which are
common to both types of nonconvulsive status, the so-called absence- (or petit
mal-) status and the focal status, are disturbance of consciousness and incomplete
or complete amnesia. This psychotic appearance of absence status is completed by
lack of responsitivity and stuporous behaviour. Absence status is by far more fre-
quent in IGE than its analogue, the complex focal status, in focal epilepsies.
Gibbs and Gibbs (1952) and Lennox (1960) reported a history of petit mal status
in 27 (2.6%) of their 1039 petit mal patients. We have seen it in 15 (5%) of our
approximately 300 patients with pyknoleptic petit mal (Janz, 1998), a percentage
which corresponds to the 5.8% of Loiseau and Cohadon (1970). Dalby (1969) has
reported petit mal status in 21 (6.2%) of his series of 346 patients with 3-Hz spike-
wave and 15 (9.3%) of the 160 patients with a history of petit mal.
Absence status which lasts for hours or days usually occurs only after adoles-
cence. As isolated events, they may even occur in middle or later age (Ja, 1962;
Lee, 1985). According to its clear clinical and electroencephalographic picture, the
almost negative interictal psychopathology, and the almost positive electric activ-
ity, the absence status can be characterized as a specic ictal psychosis of IGE.
Postictal psychosis
Furthermore, postictal and interictal psychoses are to be distinguished. Postictal
psychoses are clearly demarcated by a lucid interval from seizures themselves and
from immediate postictal confusion. The phenomenology is pleomorphic, with
uctuating combinations of delirium, delusions, hallucinations, thought disorders,
or/and aective change. Postictal psychosis is not specic for a type of epilepsy.
Dongier (1959/60) found a preponderance of generalized epilepsies in cases of post-
ictal psychosis. Others have found a higher rate among patients with localization-
related epilepsies (Logsdail and Toone, 1988; Savard et al., 1991). Devinsky et al.
(1995) noted in their case-control study of 20 patients with postictal psychoses no
signicant dierence between the percentage of patients with focal or primary gen-
eralized epilepsy in cases and controls. It seems as if long duration of epilepsy
44 D. Janz
(Kanemoto et al., 1996) and high number of lifetime GTCS (Devinsky et al., 1995)
contribute to the pathogenesis of postictal psychosis.
Interictal psychosis
Interictal psychotic episodes show a paranoid hallucinatory symptomatology.
According to some authors the delusions are less systematized than in true schizo-
phrenia (Janzarik, 1955; Tellenbach, 1965), personality deterioration is lacking
(Slater et al., 1963) and thought disorder unusual (Korzeniowski, 1965).
Unfortunately, there are no controlled data on the frequency of acute and chronic
interictal psychoses comparing IGE and TLE. According to Sengoku et al. (1997)
chronic psychoses can only be found in patients with TLE. Moreover, only in TLE
do psychoses relate to the frequency of seizures. Correspondingly, Schmitz (1992)
found that ve of six chronic psychoses occurred in patients with TLE, but she says
also this relation was not signicant due to small numbers and that psychoses in
TLE tend to be more severe than those in generalized epilepsy.
Unfortunately a comparison of symptomatology in Sengoku et al.s (1997) study
cannot be made for evaluation of interictal psychoses, as the semiological rating is
not dened. The preponderance of paranoid symptoms in TLE psychoses is sur-
prising, as is the frequent occurrence of perplexed behaviour, which is not dened
in the text. Similarly to Wolf (1982), Sengoku et al. (1997) have noted an associa-
tion between visual hallucinations and generalized epilepsies, whilst other authors
have also noticed auditory hallucinations in psychoses in TLE.
Unfortunately there is also no systematic comparison of the psychopathological
contents of interictal psychoses in both forms of epilepsy, but only impressions.
Thus Sengoku et al. (1983) wrote that with regard to the contents of psychosis, the
episodic paranoid states were most commonly found in the TLE group. In the gen-
eralized epilepsy (GE) group, however, the types of psychosis were more varied.
Wolf (1982) noted an association between olfactory hallucinations and TLE and
between visual hallucinations and generalized epilepsies, suggesting that the form
of the psychosis is inuenced by symptoms of the epileptic disorder.
Alternative psychosis with forced normalization of the EEG
Some of the interictal psychoses according to Schmitz (1988) 11%, to Wolf (1985)
15%, to Dongier (1959/60) 25% belong to the group of psychiatric disorders with
so-called forced normalization of the EEG (Landolt, 1955, 1958, 1963). With this
term Landolt described the condition that with spontaneous or, more frequently,
drug-induced disappearance of clinical and subclinical seizure manifestations,
psychiatric symptoms appear. These may be (Table 4.2) prepsychotic dysphoria
characterized by insomnia, anxiety, feelings of oppression and social withdrawal or
may proceed to psychotic symptoms characterized by hallucinatory paranoid states.
45 Psychiatry of idiopathic generalized epilepsy
The phenomenon of an inverse relationship between well-being on one hand
and seizure and specic discharges in the EEG on the other occurs with both gen-
eralized and focal epilepsy. Landolts rst observations were with temporal lobe
epilepsy (1955). He felt himself supported by Gibbs (1951), who from his experi-
ence with Phenurone drew the conclusion that the epileptic and psychiatric com-
ponents of psychomotor epilepsy are physiologically antithetic.
Later, with the advent of the succinimides, the focus turned towards the gener-
alized epilepsies. The fundamental study of Tellenbach (1965) gives a detailed
description of as he called it alternative psychoses of a paranoid type in 12 cases,
9 with IGE, in 4 cases in combination with pyknoleptic (childhood) absences.
According to Wolf and Trimble (1985) alternative psychoses occurred three times
more frequently in IGE than in epilepsies with complex focal seizures and in IGE
have been the predominant type of psychoses. Nearly the same relationship holds
true in the series of Schmitz (1988) (Table 4.3).
In a subgroup patients with IGE with absences beyond childhood it may be
seen in as many as 8% of patients, depending on the medication given (Wolf, 1986).
According to Wolf s opinion (1986), in this group we seem to know some details of
the pathogenesis. He mentioned in this respect drug inuences, as for instance the
sleep deprivation and increase of light sleep with ethosuximide, and social status,
particularly regarding the frequent vocational disintegration. He points to the
observations of Tellenbach (1965) of the frequent familial conict situations, and
on the patients personality being aected by having to cope with disabling drug
eects. Tellenbach has argued that patients with awakening epilepsy are particularly
sensitive to a drug-induced process of slowing down, the feeling of being con-
strained, of being immured because of their personality trait of extraversion.
46 D. Janz
Table 4.2. Psychiatric syndromes in 44 cases of forced
normalization observed in 36 patients
Paranoid hallucinatory psychosis 19
Prepsychotic dysphoria 9
Hysterical episode 5
Hypochondrial episode 3
Depressive episode 2
Manic episode 2
Twilight state 1
Depersonalization 1
Source: Wolf (1984).
Anything which threatens their stability will immediately and recklessly be
expressed. In the number of conditions which in the case of forced normalization
of the EEG (forcierter Normalisierung des EEG) leads to psychiatric disorders,
Tellenbach includes special personality traits of patients with awakening epilepsy.
What did the scholars understand by this in 1965, and what do we know about it
currently?
Personality
The heart of the matter
When, in the beginning of the 1950s, the clinical picture of awakening epilepsy
became clear to me, including that of all epilepsies with seizures mainly after awak-
ening, i.e. todays IGE, I became more and more interested in the conditions which
led to the triggering of seizures. According to Niedermeyer (1996), in IGE the role
of arousal is the heart of the matter. The period after awakening, at whatever time
of awakening, is the most critical biological situation in any variety of IGE, both
for the appearing of generalized spike wave-discharges and for the manifestation of
the generalized tonic-clonic seizures, the absences and the myoclonic jerks.
Therefore the early-chosen termepilepsy on awakening (Janz, 1953, 1962) focuses
the attention to the crucial point which is pathophysiologically common to all
forms of IGE.
It has been noted that all seizures in IGE were frequently precipitated by exter-
nal factors such as emotional and physical stress especially in association with lack
of sleep. These factors seem to have little inuence on seizures in other forms of
epilepsy, as shown in comparison with sleep epilepsy (SE) (Table 4.4).
47 Psychiatry of idiopathic generalized epilepsy
Table 4.3. Psychosis and related syndromes in patients of a neurological outpatient clinic (N611)
Generalized Focal Epilepsies with Unclassied
epilepsies epilepsies generalized and focal epilepsies
Syndrome groups (n237) (n270) features (n46) (n58)
Ictal and peri-ictal 3 9 2 1
Forced normalization 4 1
Intoxicated 1 2
Drug withdrawal 1
Unrelated to epilepsy 3 4 1
Multifactorial and unknown 2 1 1 1
Source: Schmitz (1988).
In JME in particular, we have reported unusual lack of sleep and sudden awak-
ening, and/or excessive alcohol consumption, as the most common precipitating
factors (Janz and Christian, 1957). Other triggering factors were less common.
Pedersen and Petersen (1998) registered, although retrospectively, alcohol con-
sumption in 51%, stress in 70% and sleep deprivation in 84% of patients as the
most frequent precipitating factors.
Lifestyle
Most patients are not in a position to sleep as much as they would like to every day;
they are forced to get up earlier than they would like to, and therefore suer from
a chronic sleep decit. Additional stress forces them to pass their limits, and sei-
zures result.
Everyone who knows patients with epilepsy on awakening can produce numer-
ous examples of seizures occurring after late nights spent in preparing for exam-
inations, attending parties, watching television or preparing to leave early for
holidays. Some of the causes are so extreme that one is tempted to say that it is no
wonder that seizures occur. However, the patients describe the events as unavoida-
ble, and one tends to pardon these excesses at rst. One becomes suspicious when
seizures continue to occur after repetitions of the same or similar events. It is sur-
prising that the patients often do not avoid the situations which provoke seizures,
and it is not clear why the patients are unable to learn from experience. One gets
the impression that the patients are unable to draw conclusions from their own
negative experiences and to change their lifestyles, and understands Niedermeyers
48 D. Janz
Table 4.4. Seizure precipitants in sleep epilepsy (SE) (n127) and in awakening epilepsy
(AE) (n90)
First seizures (%) Following seizures (%)
Precipitant SE AE SE AE
Sleep deprivation 5 25 6 42
Alcohol use 1 10 3 23
Abrupt wakening 12 28
Strong emotion 2
Physical stress 3 9 4 13
Menarche 2* 6*
Note:
* Female patients only.
Source: Janz (1953).
(1996) moaning: Unfortunately, some patients are incorrigible and all exhorta-
tions fall on deaf ears.
Sleeping habits
Of all the precipitating factors, sleep deprivation is certainly the most important, and
it would appear that other factors become eective in the presence of sleep depriva-
tion. The question arises what is the reason for this relation: is it a faulty behaviour
of the patients, is it a particular biological weakness or probably both? Patients with
IGE, particularly those withJME, tendto awake slowly andwithdiculty andremain
sleepy for some time afterwards (Table 4.5) (Janz and Christian, 1957; Janz, 1985,
2000). Nearly every patient can report that he or she is dicult to arouse, and would
love to stay in bed longer, that sleep still sits in their limbs, that they initially act
automatically, like in a dream, or are numb. Many have adjusted to this by sleep-
ing longer and having breakfast in bed. Others who cannot aord this, dunk their
heads in cold water or drink strong coee. All emphasize that they need much sleep
and most claimthat they sleep very deeply. However, inour experience, these patients
do not admit or do not recognize that they often retire too late and that, therefore,
they often do not get enough sleep, i.e. that they easily suer fromlack of sleep.
Interviews by questionnaire which have recently been carried out in two loca-
tions in Berlin indicate that there is a signicant dierence in well-being during
day-time between the patients with JME and those with TLE (Pung, 2000) (Table
4.6). JME patients feel less ecient in the morning than TLE patients. The fact that
perception of well-being in the evening met only partly but not signicantly with
expectations, might depend on a lack of acceptance or on an adaptation to social
conventions. It is useful to complete the interview of the patients together with
their near relatives or friends, because JME patients particularly tend to adapt their
answers to conventional attitudes.
49 Psychiatry of idiopathic generalized epilepsy
Table 4.5. Modes of falling asleep and awakening in patients with awakening epilepsy
(AE) and in patients with sleep epilepsy (SE)
AE SE
n % n %
Falling asleep Quick, unmolested 19 40 36 86
Delayed, dicult 28 60 6 14
Awakening Quick, unmolested 11 18 32 68
Delayed, dicult 50 82 15 32
Source: Janz (1953).
Personality traits in patients with epilepsy on awakening
Phenomenology
Patients with IGE, especially those with JME, seem to lack a certain degree of self-
control. This is evident in the fact that they seem to neglect their physical needs and
their health. They are quite likable. They are not hypochondriacs, like many
patients with temporal lobe epilepsy, and do not behave hysterically, but it is impor-
tant to recognize that these patients fail to avoid situations which might provoke
seizures, and also forget to take their medications or even discontinue treatment.
This lack of self-control goes with a tendency to deny problems and conicts, and
besides a shortlasting depression postictally they tend to take seizures, symptoms
and complaints lightly, some even believing that they can avoid the disease by not
taking their tablets. These patients, particularly if teenagers, often become obsti-
nate, claiming something to the eect that I would not consider going to bed at
regular hours, abstaining from alcohol and living like a monk. I am not a child
anymore. If I cant do what I want, then I might as well hang myself . This is an
expression of an immature and childish attitude which must be met with compre-
hension and rmness rather than authoritarian exhortations.
Another characteristic that one often encounters in patients with IGE is their
impressionability. They comprehend and judge quickly, and they are versatile and
adaptable as well as impressionable, which facilitates contact with the physician but
endangers the continuity of the therapeutic relationship. Characteristics such as
impressionability, openness and awareness to the point of being easily distracted
make these patients appear alert and intelligent, but cause them problems when
tasks requiring resoluteness, patience and perseverance are concerned.
50 D. Janz
Table 4.6. Answers of 20 patients with juvenile myoclonic epilepsy (JME) and 20 patients
with temporal lobe epilepsy (TLE)
Question Time period JME TLE
What time of the day do you feel at top?
1621.00 h 3 1
1016.00 h 16 12
810.00 h 1 6
58.00 h 1
Are you more a morning type or an evening type?
Morning 7 14
Evening 13 6
Source: Pung (2000).
Patients with IGE are somewhat unstable and tend to lose their emotional
balance easily. They may be moody, though these phases do not last long. Lack of
self-condence leads them to underestimate themselves though overcondence
may also be observed.
Since my rst description (1953), I have pointed out repeatedly (1962, 1969,
1974, 1989, 1998) that patients with this epileptic syndrome are, as a rule, very
dierent from what had long been considered the typical personality of genuine
epileptics that is, slow, fussy, viscous, circumstantial and irritable, utterly pedan-
tic and obstinate, prone to hypochondria and practically incapable of changing the
subject of conversation. Some of these traits are now grouped together as the
Geschwind syndrome (Benson, 1991; Blumer, 1999) and associated with TLE.
Patients with IGE are inclined to the opposite more unstable, less reliable, incon-
siderate and neglectful of duties and self-interest, more ready to yield to the slight-
est temptation even against better judgement, insensitive to future consequences.
They behave sometimes, as Tellenbach put it, like an adult child or as Pellock
(1999) described JME patients, as perpetual adolescents.
Investigations
Peter Wolf (1985) wrote
The clinical observation of Janz was followed by a psychological investigation of Leder (1967),
who studied a group of 34 patients with GMA and 55 with GMS. From these, two groups of 10
persons each were selected for statistical evaluation. These groups were matched for sex, age,
seizure frequency, association with minor seizures, intelligence, absence of etiological clues and
social status. They were investigated with two personality tests, the Rorschach and Szondi tests,
and several signicant dierences between the groups were found. Their interpretation describes
patients with GMA as extroverts who have diculty in recognizing the limits of their own person
in relation to the external world. Often they have little ability to suppress, to contradict and to
renounce conicts; tensions and disinclinations are usually momentarily disposed often by
denial. These patients will follow simultaneously the most divergent and irreconcilable aims
without being aware of any diculties.
Wolf remarks that the Leder investigation, which has been conrmed by similar
studies in Japan (Aoki and Kawai, 1982; Kawai and Aoki, 1983; Yamashita et al.,
1984), has been criticized for its methodology, since many psychologists have res-
ervations about Rorschach and Szondi tests. However, his description will remind
unprejudiced readers of many of their own patients with IGE whose lack of disci-
pline is often an obstacle to successful therapy. Their unstable sleep behaviour may
precipitate seizures, and sleep withdrawal is often the cause of the rst seizure.
Therefore, it seems possible that unstable personalities of this kind are common in
GMAbecause sucha personality is a factor whichfavours the manifestationof a pre-
disposition to seizures which may often remain latent in people with more regular
51 Psychiatry of idiopathic generalized epilepsy
lifestyles. It is the opinion of Peter Wolf that comparative psychological investiga-
tions of patients andpredisposedbut unaectedrelatives couldclarify this question.
Personality in patients with pyknolepsy (childhood absence epilepsy)
Phenomenology
In the case of two subsyndromes of IGE, namely in pyknolepsy or CAE and in JME,
the description of personality traits has its own tradition. Before the era of EEG, the
observation that children with pyknolepsy are usually lively to over-lively, mentally
active and markedly intelligent (lebhaft bis berlebhaft, geistig regsam und ausgespro-
chen intelligent; Rosenthal 1935), has often served as an argument that due to this
unepileptic behaviour, pyknolepsy is not an epileptic disease. But even later, when
the EEG had given an explanation for the true nature of the absences, the friendli-
ness and brightness of the children was emphasized, and they were often called
model children (Bridge, 1949), even though their positive traits could not hide the
critical ones. Bridge, for example, noted that children with pyknolepsy are over-
lively, never stop in order to relax and show no sign of fatigue when playing with
other children. At home and at school, they are very keen to full their duties due to
their high sensitivity to criticism and scolding. As a consequence, these children
strain their energies excessively. I have myself pointed out that often, in the course of
their further development, the parents expectations are disappointed (Janz, 1955,
1998). The increasing demands at school during their education reveal that the chil-
drens liveliness and cleverness hide a lack of profoundness and perseverance. The
performance at school decreases. The parents complain that their children do not
perform any better than average pupils even though they used to be among the best,
and they explain this as follows: despite their intelligence they are too light-headed,
they are not attentive and they have too many other things in their heads. What
once seemed like keen interest and brightness, then reveals itself to be a lack of con-
centration and carelessness, the liveliness is nervousness, and the quick perception a
lack of profoundness. This metamorphosis from the intellectual miraculous child to
the average pupil or even a loser reects a frequent misjudgement by the adults.
These observations of the psychological characteristics and the development of chil-
dren with pyknolepsy or childhood absence epilepsy show an important proximity
to the generally described psychological structure of awakening epilepsy.
Investigations
For this reason the experimentalpsychological investigations of absence epilepsy
which Mirsky and his group have pursued since the 1960s are of special interest.
They have found in absence epilepsy decits in sustained auditory and visual atten-
tion that are distinct from those seen in other seizure disorders (Mirsky et al., 1995).
The event-related potential studies indicate that an impaired sensory input, seen
52 D. Janz
primarily in the auditory modality, not only occurs during spike-wave bursts, but
during interictal periods as well. The clinical impression of increased distractibil-
ity in absence patients is psychologically explained by a signicant score on a test
assessing the sustaining function of attention, whereas other components of the
complex process of attention like shift or focus/execute are signicantly more
impaired in patients with complex focal seizures (Mirsky et al., 1991) (Figure 4.1).
Levav (1991) has investigated the question posed by Wolf (1985), as to whether
the manifestations of IGE are associated with personality traits, that healthy rela-
tives do not show. She investigated the attention performance of children aected
with absence epilepsy and that of their rst-degree relatives. She found signicant
impairment in tests of sustained attention in comparison with unaected siblings;
moreover, particularly in female probands. That mothers tended to perform more
poorly than fathers, which is in accordance with the known maternally transmit-
ted inuence on seizure susceptibility (Ottmann et al., 1988) gave rise to the
assumption that measures of attention may serve as markers of vulnerability to the
disorder (Mirsky et al., 1995). This hypothesis needs investigating in relation to
subclinical seizure activity.
53 Psychiatry of idiopathic generalized epilepsy
Normal controls
(n 76)
Absence seizures
(n 76)
Complex partial seizures
(n 7)
1.0
0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Shift Focus/Execute Sustain Encode
S
t
a
n
d
a
r
d
s
c
o
r
e
s
Figure 4.1. Component scores extracted by principal components analysis of attention tests on
normal control subjects (white columns), patients with absence seizures (black columns)
and complex partial seizures (cross-hatched columns). The asterisks appearing below the
columns indicate that the group so designated differed significantly from the other two
groups on that component. The absence group performed significantly worse than the
other two groups on the sustain component, whereas the complex partial group was
impaired, in comparison with the other groups, on the shift and focus/execute
components. (Reproduced with permission from Mirsky et al., 1991.)
Personality traits in patients with juvenile myoclonic epilepsy
Phenomenology
With regard to the behavioural phenomenology of patients with JME, we described
the observations in our rst article (1957) as follows:
Their psychological style as was described for awakening epileptics is very frequently character-
ized by unsteadiness, lack of discipline, hedonism and indierence towards their disease, in con-
trast to typical epileptic conduct. Since patients with impulsive petit mal (nowadays called JME)
behave rather similar we feel that it is possible to establish a typology. Only one patient was men-
tally retarded. Most were intellectually average, and none was particularly gifted. But since they
were all quick to learn and judge, exible and adaptable, school and professional or occupational
training were easy for them. But they promise more than they deliver. Of 19 men whose occupa-
tional training was known to us in some detail, ten descended to the level of unskilled laborers after
they had been employees with business training, craftsmen, skilled workers and students. Although
the psychological and social eects of their disease itself were certainly responsible for this to some
extent, especially since the onset of the condition occurs very often during the period of occupa-
tional training, insucient motivation and endurance remain possible explanations for this nega-
tive development. The conduct of these patients often has an unfavorable inuence on therapy.
Although the patients are ready to swear that they have done everything that they were instructed
to do, they frequently fail to appear at follow-up visits or to take their medications regularly. Many
handle themselves with great assurance and demanding, but they may also be decidedly mistrust-
ful and shy, fearful and inhibited. Their labile feeling of self worth also leads them to be both eager
to help, to invite, to give, on the one hand and to be able to react in an exaggeratedly sensitive way
on the other hand. Their mood changes rapidly and frequently. This makes their contact both
charming and dicult. They are easy to encourage and discourage, they are gullible and unreliable.
Their suggestibility makes contacts easy but makes trust dicult. This personality prole plays
along a scale from a likable nonchalance or timidity, through a psychoasthenic syndrome to the
extremes represented by sensitive or reckless psychopathy behaviour (Janz and Christian, 1957).
Investigations
These impressions, gained only by observations, were completed in 1976 by
psychological investigations in Denmark (Bech et al., 1976, 1977; Lund et al., 1976;
Reintoft et al., 1976; Simonsen et al., 1976). In comparison to patients with idio-
pathic grand mal epilepsy no dierence could be found concerning social back-
ground, intelligence and education. However, patients with JME more often than
controls tended to experience a social decline, diculties in nding contacts and a
feeling of being discriminated against. Signs of character-neurosis, that were only
diagnosed in patients with JME, were associated to the psychological test ndings
of substability and subvalidity. These results have been integrated as signs for
instability and psychasthenie.
Since the introduction of JME to the Anglo-Saxon scientic orbit through
Asconap and Penry (1984) and Delgado-Escueta and Enrile-Bascal (1984), many
54 D. Janz
papers on JME have appeared, but comments on the psychological prole have
remained scarce and controversial. Psychiatric disorders in JME have been reported
in comparison with TLE to be relatively high (Vasquez et al., 1993) or relatively low
but higher than in controls (Perini et al., 1996). According to the latter authors, who
performed psychological tests, JME patients only showed one signicant dierence
from controls, by having high scores on a trait anxiety scale.
At the occasion of the 1989 meeting of the American Epilepsy Association in
Boston, I was asked by Kin Penry to outline before the young epileptologists the
personality prole of patients with JME from my view as a clinician.
At the end of my speech I raised two questions to the audience:
1. Is it possible to verify this behavioural pattern with psychological tests?
2. Is it possible to correlate this behavioural pattern with discrete brain structures?
And I cautiously gave the following answer:
Some of these characteristics such as limited rational self-control, suggestibility, distractibility
and indierence to physical needs suggests involvement of frontal regions of the brain. This con-
clusion is compatible with the fact of accentuated bilateral frontal or fronto-central spike-wave
activity in this type of epilepsy. Also morphological ndings of bilateral cortical microdysgene-
sis which is probably more pronounced in the frontal regions (Meencke, 1985; Meencke and Janz,
1984, 1985) correlate with the described psychological syndrome, which therefore one might des-
ignate as a mild frontal lobe syndrome.
Since then two groups have studied this correlation with specic frontal psycholog-
ical tests.
Swartz et al. (1994) examined memory function in 9 patients with JME, 15
patients with frontal lobe epilepsy (FLE) and 15 controls. The key test was the
delayed match to sample, which they refer to as a measure of a primary or working
memory. The JME group performed intermediate between the controls and the
patients with clear frontal pathology. The authors explained their ndings as
related to impairment in selective attention. In a second study with identical tasks
they performed a FDG positron emission tomography (PET) comparing JME
patients and normal controls (Swartz et al., 1996). Controls showed activation in
the dorso-lateral prefrontal area while in JME patients there was a decrease in this
area. Instead, JME patients showed increased activations in the brain stem, the
medio-temporal region and the lateral orbital cortex. The authors explained this as
an attempt to compensate for the failure of activation in the orbito-frontal region.
Devinsky et al. (1997) compared the scores of a number of frontal tasks taken
from 16 patients with JME with those from 15 patients with TLE, matched for IQ.
The results showed evidence for a pattern of impaired frontal functioning in
patients with JME. This was signicantly dierent from the frontal functioning in
patients with TLE. It is the higher order cognitive functions of planning, reasoning
55 Psychiatry of idiopathic generalized epilepsy
and exibility of thought that are compromised, and to lesser extent, the ability to
focus attention and inhibit habitual but inappropriate responses (Devinsky et al.,
1997).
Taken together, these data do suggest subtle but nonetheless real frontal impair-
ments of psychological functions in people with JME (Trimble, 2000).
This raises some questions:
Are there essential dierences in the personality traits of dierent subsyndromes of
IGE? According to the information we have up to now, the dierences seem to
be less important than compared with focal epilepsies, especially TLE.
If we assume certain psychological characteristics, what is then the common basic
disorder?
Is it a disturbance of the function of attention which is noted in the sustained atten-
tion test and in the visual working memory test ?
What is the relation between the assumed psychological basic disturbance and the
circadian sleepwaking cycle and maybe even the ultradian restactivity changes?
And nally, what is the relation between the assumed disturbance of a cognitive
performance that of continuous attention and a biological performance
that of a rhythmic change of rest and activity in view of the development of
those two performances during adolescence?
I think that such questions help best to prevent the aected from being stigma-
tized. If we see that these patients often behave like adult children (Tellenbach,
1965), perpetual adolescents (Pellock, 1999), or better, in some way like still ado-
lescents, then we can nd ways to understand their weaknesses and to help them
eectively.
REFERENCES
Aoki, K. and Kawai, I. (1982). Awake epilepsy and sleep epilepsy: a psychological study with
Rohrschach test. Rohrschach Res, 24, 10117.
Asconap, J. and Penry, J.K. (1984). Some clinical and EEG aspects of benign juvenile myoclonic
epilepsy. Epilepsia, 25, 10814.
Bech, P., Kjaersgard Pedersen, K.K., Simonsen, N. and Lund, M. (1977). Personality traits in epi-
lepsy. In Epilepsy, the Eighth International Symposium, ed. J.K. Penry, pp. 257263. New York:
Raven Press.
Bech, P., Kjaersgard Pedersen, K., Simonsen, N. and Lund, M. (1976). Personality traits in epi-
lepsy. A multidimensional study of personal traits ad modum Sjbring. Acta Neurol Scand, 54,
34858.
Benson, D.F. (1991). The Geschwind syndrome. In Advances in Neurology, ed. D. Smith, D.
Treiman and M. Trimble, Vol. 55, pp. 42141. New York: Raven Press.
56 D. Janz
Blumer, D. (1999). Evidence supporting the temporal lobe epilepsy personality syndrome.
Neurology, 53 (Suppl. 2), S912.
Bredkjaer, S.R., Mortensen, P.B. and Parnas, J. (1998). Epilepsy and non-organic non-aective
psychosis. National epidemiological study. Br J Psychiatry, 172, 2358.
Bridge, E.M. (1949). Epilepsy and Convulsive Disorders in Children. New York: McGraw-Hill.
Bruens, H.J. (1971). Psychoses in epilepsy. Psychiatr Neurol Neurochir, 17492.
Commission on Classication and Terminology of the International League Against Epilepsy
(1989). Proposal for revised classication of epilepsies and epileptic syndromes. Epilepsia, 30,
38999.
Dalby, M.A. (1969). Epilepsy and 3 per second spike and wave rhythms; a clinical, electroen-
cephalographic, and prognostic analysis of 346 patients. Acta Neurol Scand, 45 (Suppl. 40),
1183.
Delgado-Escueta, A.V. and Enrile-Bascal, F. (1984). Juvenile myoclonic epilepsy of Janz.
Neurology, 34, 28594.
Devinsky, O., Abramson, H., Alper, K. et al. (1995). Postictal psychosis: a case control series of 20
patients and 150 controls. Epilepsy Res, 20, 24753.
Devinsky, O., Gershengorn, J., Brown, E., Perrine, K., Vasquez, B. and Luciano, D. (1997). Frontal
functions in juvenile myoclonic epilepsy. Neurol Neuropsychol Behav Neurol, 10, 2436.
Dongier, S. (1959/60). Statistical study of clinical and electroencephalographic manifestations
of 536 psychotic episodes occurring in 516 epileptics between clinical seizures. Epilepsia, 1,
11742.
Gastaut, H. (1956). Colloque de Marseille, 1519 Octobre 1956. Compte rendu du colloque sur
ltude lectroniques en dehors des crises cliniques. Revue Neurol, 95, 587616.
Genton, P., Glisse, P. and Thomas, P. (2000). Juvenile myoclonic epilepsy today: current deni-
tion and limits. In Juvenile Myoclonic Epilepsy: The Janz Syndrome, ed. B. Schmitz and T.
Sander, pp. 1132. Peterseld, UK and Philadelphia, USA: Wrightson Biomedical Publishing.
Gibbs, F.A. (1951). Ictal and non-ictal psychiatric disorders in temporal lobe epilepsy. J Nerv
Ment Dis, 113, 5228.
Gibbs, F.A. and Gibbs, E.L. (1952). Atlas of Electroencephalography. Cambridge, MA: Addison-
Wesley.
Ja, R. (1962). Ictal behaviour disturbance as the only manifestations disorder: Case report. J
Nerv Ment Dis, 135, 4705.
Janz, D. (1953). Aufwach-epilepsien (als ausdruck einer den nacht- oder schlaf epilepsien
gegenberstehenden verlaufsform epileptischer erkrankungen. Arch Psychiat Nervenkr, 191,
7398.
Janz, D. (1955). Die klinische Stellung der Pyknolepsie. Dtsch med Wchschr, 80, 13924,
1399400.
Janz, D. (1962). The grand mal epilepsies and the sleepingwaking cycle. Epilepsia, 3, 69109.
Janz, D. (1969). Die Epilepsien. Spezielle Pathologie und Therapie. Stuttgart: Thieme (1998), 2. unv
Au.
Janz, D. (1985). Epilepsy with impulsive petit mal (juvenile myoclonic epilepsy). Acta Neurol
Scand, 72, 44959.
57 Psychiatry of idiopathic generalized epilepsy
Janz, D. (1989). Juvenile myoclonic epilepsy. Epilepsy with impulsive petit mal. Cleveland Clin J
Med, 56 (Suppl.), 52333.
Janz, D. (1997). The idiopathic generalized epilepsies of adolescence with childhood and juvenile
age of onset. Epilepsia, 38, 411.
Janz, D. (2000). Epilepsy with grand mal on awakening and sleep-waking cycle. Clin Neuro-
physiol, 111 (Suppl. 2), S10310.
Janz, D. and Christian, W. (1957). Inpulsiv-Petit mal. Dtsch Z Nervenheilk (J Neurol), 176,
34686. English (1994) in Idiopathic Generalized Epilepsies, ed. A. Malafosse, P. Genton, E.
Hirsch, C. Marescaux, D. Broglin and R. Bernasconi, pp. 22951. London: John Libbey.
Janzarik, W. (1955). Der wahn schizophrener prgung in den psychotischen episoden der epilep-
tiker und die schizophrene wahnwahrnehmung. Fortschr Neurol Psychiat, 23, 53346.
Kanemoto, K., Kawasaki, J. and Kawai, I. (1996). Postictal psychosis: a comparison with acute
interictal and chronic psychoses. Epilepsia, 37, 5516.
Kawai, I. and Aoki, K. (1983). Primary generalized epilepsy and temporal lobe epilepsy: a psycho-
logical study using Rohrschach test. Folia Psychiat Neurol Jap, 37, 24551.
Korzeniowski, L. (1965). Les problemes diagnostiques concernant les psychoses paranoiaques
schizophreniformes en pilepsie. Ann Md-psychol, 123, 3542.
Landolt, H. (1955). ber verstimmungen, dmmerzustnde und schizophrene zustandsbilder
bei epilepsie. Ergebnisse klinischer und elektroencephalographischer untersuchungen.
Schweiz Arch Neurol Psychiatr, 76, 31321.
Landolt, H. (1958). Serial electroencephalographic investigations during psychotic episodes in
epileptic patients and during schizophrenic attacks. In Lectures on Epilepsy, ed. A.M. Lorentz
de Haas. Folia Psychiatr Neurol Neurochir, (Suppl. 4), pp. 91133. Amsterdam: Elsevier.
Landolt, H. (1963). Die dmmer- und verstimmungszustnde bei epilepsie und ihre elektroen-
cephalographie. Dtsch Z Nervenheilk, 185, 41130.
Leder, A. (1967). Zur psychopathologie der schlaf- und aufwachepilepsie (eine psychodiagnos-
tische untersuchung). Nervenarzt, 38, 43442.
Lee, S.I. (1985). Nonconvulsive status epilepticus: Ictal confusion in later life. Arch Neurol, 42,
77881.
Lennox, W.G. (1960). Epilepsy and Related Disorders. Boston, MA: Little & Brown.
Levav, M.L. (1991). Attention performance in children aected with absence epilepsy and their
rst-degree relatives. Doctoral dissertation. University of Maryland, College Park.
Logsdail, S.J. and Toone, B.K. (1988). Postictal psychosis: a clinical and phenomenological
description. Br J Psychiatry, 152, 24652.
Loiseau, P. and Cohadon, F. (1970). Le Petit Mal et ses Frontires. Paris: Masson.
Lund, M., Reintoft, H. and Simonsen, N. (1976). Eine kontrollierte soziologische und psycholo-
gische untersuchung von patienten mit juveniler myoklonischer epilepsie. Nervenarzt, 47,
70812.
Meencke, H.-J. (1985). Neuron density in the molecular layer of the frontal cortex in primary
generalized epilepsy. Epilepsia, 26, 4504.
Meencke, H.-J. and Janz, D. (1984). The signicance of microdysgenesis in primary generalized
epilepsies: a study of eight cases. Epilepsia, 25, 821.
58 D. Janz
Meencke, H.-J. and Janz, D. (1985). The signicance of microdysgenesis in primary generalized
epilepsy: an answer to the considerations of Lyon and Gastaut. Epilepsia, 26, 36871.
Mndez, M.F., Doss, R.C., Taylor, J.L. and Arguello, R. (1993). Relationship of seizure variables
to personality disorders in epilepsy. J Neuropsychiatry Clin Neurosci, 5, 2836.
Mignone, R.J., Donelly, E.F. and Sadowsky, D. (1970). Psychological and neurological compari-
son of psychomotor epileptic patients. Epilepsia, 11, 34559.
Mirsky, A.F., Anthony, B.J., Duncan, C.C., Ahearn, M.B. and Kellam, S.G. (1991). Analysis of the
elements of attention: a neuropsychological approach. Neuropsychol Rev, 2, 10945.
Mirsky, A.F., Duncan, C.C. and Levav, M.L. (1995). Neuropsychological and psychophysiologi-
cal aspects of absence epilepsy. In Typical Absences and Related Syndromes, ed. J.S. Duncan and
C.P. Panayiotopoulos, pp. 11219. London: Churchill Communications.
Niedermeyer, E. (1996). Primary (idiopathic) generalized epilepsy and underlying mechanisms.
Clin Electroencephalogr, 27, 121.
Onuma, T. (1983). Limbic lobe epilepsy with paranoid symptoms: Analysis of clinical features
and psychological tests. Folia Psychiat Neurol Jpn, 37, 2537.
Ottmann, R., Annegers, J.F., Hauser, W.A. and Kurland, L.T. (1988). Higher risk of seizures in
ospring of mothers than of fathers with epilepsy. Am J Hum Genet, 43, 25764,
Pedersen, S.B. and Petersen, K.A. (1998). Juvenile myoclonic epilepsy: Clinical and EEG features.
Acta Neurol Scand, 97, 1603.
Pellock, J.M. (1999). JME: The elusive epilepsy. Perspect Pediat Neurol, 1, 16.
Perini, G.I., Tosin, C., Carraro, C. et al. (1996). Interictal mood and personality disorders in tem-
poral lobe epilepsy and juvenile myoclonic epilepsy. J Neurol Neurosurg Psychiatry, 61, 6015.
Pung, T. (2002). Schlaf-wachrhythmus und persnlichkeitbei juveniler myoklonischer epilepsie
und patienten mit temporaler epilepsie. Thesis, Humboldt Universitt Berlin.
Reintoft, H., Simonsen, N. and Lund, M. (1976). A controlled sociological study of juvenile myo-
clonic epilepsy. In Epileptology, ed. D. Janz, pp. 4850. Stuttgart: Thieme.
Rosenthal, C. (1935). Die gehuften kleinen Anflle des Kindesalters (Pyknolepsie). Ergebn inn
Med Kinderheilk, 48, 77124.
Sander, T., Schulz, H., Saar, K. et al. (2000). Genome search for susceptibility loci of common
idiopathic generalised epilepsies. Hum Mol Genet, 9, 146572.
Savard, G., Andermann, F., Olivier, A. and Remillard, G.M. (1991). Postictal psychosis after
partial complex seizures: a multiple case study. Epilepsia, 32, 22531.
Schmitz, B. (1988). Psychosen bei epilepsie. Eine epidemiologische untersuchung. Thesis, Freie
Universitt, Berlin.
Schmitz, B. (1992). Psychosis and epilepsy: the link to the temporal lobe. In The Temporal Lobes
and the Limbic System, ed. M.R. Trimble and T.G. Bolwig, pp. 14967. Peterseld, UK and
Bristol, PA, USA: Wrightson Biomedical Publishing.
Schmitz, B. and Wolf, P. (1991). Psychosis in epilepsy. In Epilepsy and Behaviour. Frontiers of
Clinical Neuroscience, ed. O. Devinsky and W.H. Theodore, Vol. 12, pp. 97128. New York:
Wiley-Liss.
Schmitz, B. and Wolf, P. (1995). Psychosis in epilepsy: frequency and risk factors. J Epilepsy, 8,
295305.
59 Psychiatry of idiopathic generalized epilepsy
Sengoku, A., Yagi, K., Seino, M. and Wada, T. (1983). Risk of occurrence and psychosis in rela-
tion to the types of epilepsies and epileptic seizures. Folia Psychiat Neurol Jpn, 37, 2216.
Sengoku, A., Toichi, M. and Murai, T. (1997). Comparison of psychotic states in patients with
idiopathic generalized epilepsy and temporal lobe epilepsy. Epilepsia, 38 (Suppl. 6), 225.
Shukla, G.D., Srivastava, O.N., Katiyar, B.C. et al. (1979). Psychiatric manifestations in temporal
lobe epilepsy: A controlled study. Br J Psychiatry, 135, 41117.
Simonsen, N., Mollgaard, V. and Lund, M. (1976). A controlled clinical and electroencephalo-
graphical study of myoclonic epilepsy (Impulsiv-Petit-Mal): preliminary report. In
Epileptology, ed. D. Janz, pp. 418. Stuttgart: Thieme.
Slater, E., Beard, A.W. and Glithero, E. (1963). The schizophrenia-like psychoses of epilepsy. Br J
Psychiatry, 109, 95150.
Small, J.G., Milstein, V. and Stevens, J.R. (1962). Are psychomotor epilepsies dierent? Arch
Neurol, 7, 18794.
Standage, K.F. and Fenton, G.W. (1975). Psychiatric symptom proles of patients with epilepsy:
a controlled investigation. Psychol Med, 5, 15260.
Stevens, J.R. (1966). Psychiatric implications of psychomotor epilepsy. Arch Gen Psychiatry, 14,
46171.
Stub-Naylor, A. (1996). Epilepsy and Psychiatric Disorder A Comorbidity Study. Copenhagen:
FADL Publishers.
Swartz, B.E., Halgren, E., Simpkins, F. and Syndulko, K. (1994). Primary memory in patients with
frontal and primary generalized epilepsy. J Epilepsy, 7, 23241.
Swartz, B.E., Simpkins, F., Halgren, E. et al. (1996). Visual working memory in primary general-
ized epilepsy. An FDG-PET study. Neurology, 47, 120312.
Tellenbach, H. (1965). Epilepsie als anfallsleiden und als psychose. ber alternative psychosen
paranoider prgung bei forcierter normalisierung (LANDOLT) des elektroencephalo-
gramms epileptischer. Nervenarzt, 36, 190202.
Trimble, M. (2000). Cognitive and personality proles in patients with juvenile myoclonic epi-
lepsy. In Juvenile Myoclonic Epilepsy: The Janz Syndrome, ed. B. Schmitz and T. Sander, pp.
1019. Peterseld, UK and Philadelphia, USA: Wrightson Biomedical Publishing.
Vazquez, B., Devinsky, O., Luciano, D., Alper, K. and Perrine, K. (1993). Juvenile myoclonic epi-
lepsy: clinical features and factors related to misdiagnosis. J Epilepsy, 6, 2338.
Wolf, P. (1976). Psychosen bei Epilepsie. Ihre Bedingungen und Wechselbeziehungen zu Anfllen.
Habilitationsschrift. Berlin: Freie Universitt.
Wolf, P. (1982). Halluzinationen in Rahmen epileptischer Psychosen. In Halluzinationen bei
Epilepsien und ihre Dierentialdiagnose, ed. K. Karbowski, pp. 5866. Bern: Huber.
Wolf, P. (1984). The clinical syndromes of forced normalization. Folia Psychiatr Neurol Jpn, 38,
18792.
Wolf, P. (1985). Epilepsy with grand mal on awakening. In Epileptic Syndromes in Infancy,
Childhood and Adolescence, 2nd edn (1992), ed. J. Roger, M. Bureau, Ch. Dravet, F.E. Dreifuss,
A. Perret and P. Wolf, pp. 32941. London: John Libbey.
Wolf, P. (1986). Forced normalization. In Aspects of Epilepsy and Psychiatry, ed. M.R. Trimble and
T.G. Bolwig, pp. 10112. Chichester: John Wiley & Sons.
60 D. Janz
Wolf, P. and Trimble, M.R. (1985). Biological antagonism and epileptic psychosis. Br J Psychiatry,
146, 2726.
Yamashita, K., Aoki, K. and Kawai, I. (1984). Primary generalized epilepsies and temporal lobe
epilepsies: a psychological study using PF study. J Jap Clin Psychology, 2, 3843.
61 Psychiatry of idiopathic generalized epilepsy
5
Epilepsy and learning disorders
Cesare Maria Cornaggia
1
and Giuseppe Gobbi
2
1
University of Milano Bicocca, Milan, Italy
2
Maggiore Hospital, Bologna, Italy
Introduction
In the practice of epilepsy, the association between epilepsy and learning disorder
is very important, even if sometimes underestimated. A correct and early diagno-
sis of a learning disorder in a person with epileptic seizures will often determine
future development and prognosis.
Definitions
Cognitive function denes the capacity of the human brain to process all informa-
tion coming from the outside and internal world of the individual, and program
ongoing behaviour (Aldenkamp and Bronswijk, 1999). This capacity involves the
ability to remain in contact with the outside world (through the function of vigi-
lance), to select and focus information (through the function of attention), and to
memorize data (through the function of memory). In this way, cognitive function
gives humans the opportunity of becoming aware of themselves and to solve prob-
lems something that we also call intelligence. The latter is the generic capacity of
using all of the elements of thinking necessary to recognize, plan and solve new
problems in a directed and correct manner.
More than one cortical area of the human brain is involved in cognitive function
and related processes, and impaired cognitive function has been observed in the
presence of a lesion or stable dysfunction in the temporal, frontal or parietal lobes
of the dominant and nondominant hemispheres. An impairment in cognitive func-
tion may be seen as a reduction in the capacity of learning of children or reduction
in the intellectual ability of adults. We normally refer to this as learning disability in
children (which, at least in some countries, is dierent from mental retardation),
and deterioration or dementia in adults or people who have already acquired mental
capacity.
From the terminological point of view, learning disorder is dierent from mental
retardation. Actually, in some countries, such as the UK, learning disability is used
62
for dening both a mental retardation and the learning disorders. But, in the
Diagnostic and Statistical Manual of Mental Disorders (DSMIV; American
Psychiatric Association, 1994) learning disorders (DSMIV 315.00315.09) include
a signicant disturbance in academic achievement or daily living activities that
require reading, mathematical or writing skills. In contrast, mental retardation is
dened as a not necessarily lifelong disorder characterized by the combination of a
low cognitive ability and diminished social and adaptive competence (DSMIV
317319) due to signicantly subaverage intellectual functioning, with an onset
before the age of 18 years.
Actually, in most parts of the world, mental retardation denes a situation
involving an abnormal IQ, and a learning disorder one involving a normal IQ.
On the other hand, there are some doubts about the denition of mental retar-
dation, especially in childhood epilepsy. In childhood epilepsy there is a risk of
diagnosing the presence of mental retardation and failing to identify patients who
are only aected by a specic learning disorder. Thus, it is known that an epileptic
focus may lead to the dysfunction of a given specic performance, that coincides
with the cortical area involved, without causing mental retardation. But, at a devel-
opmental age, when cognitive performance is progressing, a dysfunction in a spe-
cic cognitive area may cause a disequilibrium of the complex system of mental
(cognitive) development, causing an apparent mental retardation, that may com-
pletely resolve if this specic cognitive dysfunction is adequately treated.
The term retardation itself may suggest that it is possible to make up the leeway,
but we know that in many or perhaps the majority of cases of so-called mental
retardation this is not to be expected, especially in the presence of brain lesions or
some special epileptic syndromes. In some countries (including Italy), in fact, the
termmental deciency is used instead of mental retardation for dening situations
assuming a denite decit.
This is the case, also, with the term disability. Disability means a restriction of
the ability to perform an activity and it is generally considered to be stable and irre-
versible, at least in some countries such as Italy. In the case of epilepsy it could be
more appropriate to use the word disorder because this opens up the possibility that
the loss of capacity may be transitory or reversible, so-called state dependency as
discussed by Besag (1989, 1995, Chapter 6).
Learning disorders are linked to various types of other behaviour disturbances
that must be separately classied: for example, the Attention Decit/Hyperactivity
Disorder (ADH; DSMIV 314.00314.01), which may or may not be associated
with behavioural or mood disturbances (DSMIV 312.8). We also know that learn-
ing disorders are associated with other disorders, such as conduct disorders, major
depressive disorders, dysthymic disorders and opposition deant disorder, in 10 to
25% of the cases.
63 Epilepsy and learning disorders
Learning and behavioural disorders are linked both in childhood and in adults.
Although it is true that behavioural or mood disorders may be seen as a sympto-
matic expression of a cognitive impairment during epilepsy, it is also true that
behavioural or mood disorders may lead to reduced learning ability.
Prevalence
There is no doubt that a disproportionate percentage of children with epilepsy have
learning and behavioural problems (Bourgeois, 1998), and children with behav-
ioural disturbances are certainly a minority in respect to the number of those with
learning disorders.
The concept that epilepsy is invariably associated with progressive intellectual
deterioration both in children and in the adult population, which prevailed during
the rst half of the twentieth century (Bourgeois, 1998), cannot be accepted any
more. In fact, a large proportion of children with epilepsy have some schooling
diculties (Ross et al., 1980, Sillanp et al., 1999), but fewer than 1% attend
special epilepsy schools. Nakken and Brodtkorb (1999) have recently reported that
about 20% of the epilepsy population are mentally retarded (Forsgren et al., 1990,
1996), and about 20% of mentally retarded people have epilepsy (Wester, 1982). In
reality, only a small subgroup of children with epilepsy shows stable decreases in
IQ, as recently shown by Besag (1995), who demonstrated that the observation of
a declining IQ at any given moment does not necessarily indicate a loss of skills, but
simply implies that the child is not developing at the expected rate.
According to some authors, this lag between mental and chronological age is
more pronounced in older children, and in children with an earlier seizure onset,
a higher lifetime total number of seizures, or multiple seizure types (Seidenberg et
al., 1986). There is some evidence indicating that children aged less than ten years
at the time of assessment gain less than those with an age at onset of more than ten
years (Neyens et al., 1999).
Causes of learning disorders
There are three specic ways in which epilepsy and learning disorders may be
related. First, epilepsy and learning disorders are the result of brain damage and/or
permanent brain dysfunction. Secondly, epilepsy may cause brain damage or per-
manent brain dysfunction, which in turn leads to a learning disorder. Status epi-
lepticus would be the classical example, as there is no doubt that epilepsy can cause
deterioration in children experiencing prolonged bouts of status epilepticus.
Thirdly, epilepsy may cause learning disorders directly, without brain damage or
permanent brain dysfunction, such as in the case of continuous epileptiform EEG
64 C.M. Cornaggia and G. Gobbi
discharges or subtle seizures, and there is also some evidence that certain sources
of seizure discharge are associated with selective cognitive decits (Stores, 1985)
for example the LandauKlener syndrome.
Classification of learning disorders
Learning disorders occurring during epilepsy can be classied into two categories:
state-dependent (potentially treatable and reversible) and permanent.
The prevalence of state-dependent or permanent learning disorders is not satis-
factorily known. Brain damage or stable brain dysfunction is mainly associated
with permanent learning disorders. Epilepsy itself or the medication used to treat it
may lead to state-dependent learning disorders. The picture is further complicated
by associated mood disorders or psychoses, a low level of self-perception and
expectation and reduced learning opportunities.
Regarding epilepsy itself, various situations can be recognized as leading to a
state-dependent learning disorder.
Direct eects of seizures. There is evidence that single complex partial and secon-
darily generalized seizures are associated with neuronal damage (Rabinowicz et
al., 1996), and that brain extracellular glutamate may build up to neurotoxic
levels in partial seizures (During and Spencer, 1993). Perhaps, not surprisingly, a
reduction in seizure frequency over time is associated with an improvement in
cognitive functioning (Seidenberg et al., 1981), and the presence of discharges
aects discrete scholastic performances (Kasteleijin-Nolst Trenite et al., 1988).
Special epileptic syndromes. Typical examples are West syndrome (WS) and
LennoxGastaut syndrome (Gokyigit and Caliskan, 1995; Oguni et al., 1996).
Ictal changes. Nonconvulsive status epilepticus, which may continue for months
or years is an example.
Peri-ictal changes. This situation refers to patients with so many seizures every day
that they do not have time to recover from one to another. Such patients are eec-
tively in a continuous postictal state, which may be misinterpreted as a perma-
nent learning disorder.
The presence of frequent subtle seizures and the direct eects of interictal EEG epi-
leptiform discharges. Children with epilepsy who show epileptiform discharges
during IQ tests do less well than those who do not (Siebelink et al., 1988). The
Transitory Cognitive Impairment (TCI) due to the direct eects of frequent
interictal EEG epileptiform discharges or frequent subtle seizures, is a typical
example. If these are very frequent, they may represent a nonconvulsive status
epilepticus.
In fact, state-dependent learning disorders may be a consequence of TCI, asso-
ciated with either focal or generalized specic epileptiform EEG discharges. In 1939
65 Epilepsy and learning disorders
Schwab demonstrated that generalized spike-wave discharges without any apparent
clinical concomitants may be associated with a simultaneous decline in cognitive
function. TCI was also reported in 1957 by Kooi and Hovey, who studied a case
involving focal discharges. These subclinical or larval discharges are accompanied
by a TCI which, if not recognizable from the patients spontaneous behaviour, can be
detected by appropriate psychological tests during EEG monitoring (Binnie, 1979).
TCI is most readily demonstrated by dicult tasks and during generalized
regular spike-wave discharges lasting more than 3 seconds, but it can also be found
with briefer and focal discharges. In this last case, the errors are more specic: left-
sided focal discharges are more likely to produce errors in verbal tasks, whereas
right-sided discharges are associated with the impaired handling of non-verbal
material. This means that TCI is not only a consequence of global inattention, and
it has been hypothesized that the errors may be due to a working memory disorder
(Aarts et al., 1984). The eect of a TCI accompanying subclinical EEG discharges
on everyday functioning is still uncertain and needs further examination, but there
is experimental evidence that subclinical discharges may be accompanied by a dis-
ruption of educational skills in children, and the AED suppression of discharges is
concomitant with cognitive improvement.
It has been suggested that, if allowed to continue for a long time, state-dependent
learning disorders may produce permanent learning disorders. Strong evidence for
this comes from some rare models of epilepsy, e.g. LandauKlener syndrome and
continuous spike-wave in slow wave sleep (CSWS).
Certainly, the extent to which subtle seizure manifestations aect learning and
behaviour remains unknown. This suggests that there are major controversial ques-
tions to be asked regarding the importance of subtle manifestations of epilepsy on
learning and behaviour. Further, it is possible that some children diagnosed as
having ADHD or autism may actually be aected by undiagnosed manifestations
of epilepsy.
It is certainly a common experience to observe that many normally underesti-
mated or undiagnosed severe epileptic conditions, such as some frontal lobe epi-
lepsies with frequent specic EEG activity (including rapid activity lasting between
one and one-and-a-half seconds), are subsequently associated with the onset of
severe behavioural disturbances (sometimes with psychotic symptoms), or schizo-
phrenia-like or autism-like syndromes.
The existence of situations such as those reported above underlines the need to
clarify the causative role of epilepsy or the presence of nonconvulsive EEG epilep-
tiform discharges in childhood and adult behavioural problems. We must not be
afraid to use the EEG in children and in adults showing behavioural or learning dis-
turbances. Similarly, we believe that the issue of intervening to ameliorate interic-
tal EEG discharges has to be reconsidered. Of course, that needs care and has to be
66 C.M. Cornaggia and G. Gobbi
evaluated in each single patient, since it is dicult to ensure that the drugs them-
selves do not disrupt cognitive processing.
The role of treatments
The data available in the literature indicate that the administration of antiepileptic
drugs (AEDs) can impair cognitive function and aect the behaviour of children
and adults, but the extent and severity of this is still unknown. Moreover, AEDs may
cause state-dependent learning disorders, but evidence is anecdotal.
Some recent studies have compared the ecacy and presence of the cognitive
and behavioural side eects of various AEDs. For example, it has been found that
the ecacy of vigabatrin, lamotrigine and gabapentin is similar, but that vigaba-
trin is associated with a higher incidence of behavioural problems (Bhaumik et al.,
1997). On the other hand, an overall improvement in attention and school perfor-
mance has been reported in a group of children treated with vigabatrin monother-
apy in respect to a carbamazepine monotherapy control group (Gobbi et al., 1999).
The results of other studies support the hypothesis that the cognitive prole of
lamotrigine is similar to that of carbamazepine (Aldenkamp et al., 1997).
In general, an AED may improve learning by reducing the number of EEG dis-
charges or the frequency of seizures but, at the same time, it may impair learning as a
result of its side eects sleepiness, slowed reactions, attention decit and so on. It is
important not to forget clinical observations and manage each single case separately.
Finally, learning disorders need to be considered in the preoperative evaluation
and postoperative rehabilitation of patients with epilepsy.
In the preoperative phase, it is important to consider the eect that the surgical
ablation of the area of the epileptic focus may have on the processes of learning.
The ablation of an epileptogenic area generally produces positive eects because it
eliminates the negative consequences of the epileptic discharges that go from this
to other parts of the brain. However, if the area of the epileptogenic focus has not
completely lost its own primary function, the same ablation may lead to a func-
tional decit. In this case, even a satisfactory improvement in seizure frequency
may lead to very negative postoperative results.
Postoperative rehabilitation must take into account the developmental level of
the individual, and may take weeks or months of intensive specialist input but, pro-
vided that this is carefully planned and provided, the outcome can be very good.
Conclusions
Tentatively, we may summarize the actual picture as follows. Epilepsy may be asso-
ciated with learning disorders, and also behavioural or mood disorders. Behaviour
67 Epilepsy and learning disorders
and learning are often linked; behavioural changes may result from temporary or
permanent cognitive decits. Learning disorders in epilepsy may be state depen-
dent or permanent; state-dependent learning disorders may produce permanent
learning disorder. The prevalence of both conditions is not known. Strong evidence
now shows that very frequent overnight and day-time epileptiform discharges may
eventually result in permanent learning disorder. Epilepsy itself, AEDs and social
factors may inuence the onset of learning disorders. Although a large percentage
of children with epilepsy have school diculties, only a minority shows a loss of
skills or a decreasing IQ. For children in whom actual loss of skills occurs, it is
essential to look for a specic cause, and to plan an ad hoc management pro-
gramme.
REFERENCES
Aarts, J.H.P., Binnie, D.J., Smit, A.M. et al. (1984). Selective cognitive impairment during focal
and generalized epileptiform EEG activity. Brain, 107, 293380.
Aldenkamp, A.P. (1997). Eect of seizures and epileptiform discharges on cognitive function
Epilepsia, 38 (Suppl. 1), 525.
Aldenkamp, A.P. and Bronswijk, K. (1999). Cognitive side eects as an outcome measure in anti-
epileptic drug treatment: the current debate. In Epilepsy and Mental Retardation, ed. M.
Sillanpaa, L. Gram, S.I. Johannessen and T. Tomson, pp. 135146. Peterseld: Wrightson
Biomedica Publishing.
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders
(Fourth edition) (DSMIV). Washington, DC: APA.
Besag, F.M.C. (1989). Epilepsy, learning and behaviour. Educ Child Psychol, 6.
Besag. F.M.C. (1995). Epilepsy, learning, and behavior in childhood. Epilepsia, 36 (Suppl. 1), 5863.
Bhaumik, S., Branford, D., Duggirala, C. and Ismail, I.A. (1997). A naturalistic study of the use
of vigabatrin, lamotrigine and gabapentin in adults with learning disabilities. Seizure, 6,
12733.
Binnie, C.D. (1979). Direction of transitory cognitive impairment during epileptiform EEG dis-
charges: problems in clinical practice. In Epilepsy and Behaviour, ed. B.M. Kulig, H. Meinardi
and G. Stores. Lisse: Swets & Zeitlinger.
Bourgeois, B.F.R.D. (1998). Antiepileptic drugs, learning, and behavior in childhood epilepsy.
Epilepsia, 39, 91321.
During, M.J. and Spencer, D.D. (1993). Extracellular hippocampal glutamate and spontaneous
seizure in the conscious human brain. Lancet, 341, 160710.
Forsgren, L., Edvinsson, S.-O., Blomquist, H.K., Heijbel, I. and Sidenvall, R. (1990). Epilepsy in
a population of mentally retarded children and adults. Epilepsy Res, 6, 23448.
Forsgren, L., Bucht, G., Eriksson, S. and Bergmark, L. (1996). Incidence and clinical character-
ization of unprovoked seizures in adults: a prospective population-based study. Epilepsia, 37,
2249.
68 C.M. Cornaggia and G. Gobbi
Gobbi, G., Pini, A., Bertani, G. et al. (1999). Prospective study of rst-line vigabatrin monother-
apy in childhood partial epilepsies. Epilepsy Res , 35, 2937.
Gokyigit, A. and Caliskan, A. (1995). Diuse SpikeWave status of nine year duration without
behavioural change or intellectual decline: Epilepsia, 36, 21013.
Kasteleijin-Nolst Trenite, D.G., Bakker, D.J., Binnie, C.D., Buerman, A. and Van Raaij, M. (1988).
Psychological eects of subclinical epileptiform EEG discharges. I Scholastic skills. Epilepsy
Res, 2, 11116.
Kooi, K.A. and Hovey, H.B. (1957). Alterations in mental functions and paroxysmal cerebral
activity. Arch Neurol Psychiatry, 1978, 26471.
Nakken, K.A. and Brodtkorb, E. (1999). Epilepsy services for the patient with mental retardation
in Norway. In Epilepsy and Mental Retardation, ed. M. Sillanp, L. Gram, S.I. Johannessen and
T. Tomson, pp. 193200. Peterseld: Wrightson Biomedical Publishing.
Neyens, L.G.J., Aldenkamp, A.P. and Meinardi, H.M. (1999). Prospective follow-up of intellec-
tual development in children with a recent onset of epilepsy. Epilepsy Res, 34, 8590.
Oguni, H., Hayashi, K. and Osawa, M. (1996). Long-term prognosis of LennoxGastaut syn-
drome. Epilepsia, 37 (Suppl. 3), 447.
Rabinowicz, A.L., Carreale, J., Buotros, R.B., Coulwell, W.T., Henderson, C.W. and DeGiorgio,
C.M. (1996). Neuron-specic enolase is increased after single seizures during inpatient video-
EEG monitoring. Epilepsia, 37, 1225.
Ross, E.M., Peckham, C.S. and West, P.B. (1980). Epilepsy in childhood: ndings from the
National Child Development Study. Br Med J, 1, 20710.
Schwab, R.S. (1939). A method of measuring consciousness on petit-mal epilepsy. J Nerv Ment
Dis, 89, 6901.
Seidenberg, M., OLeary, D.S., Berent, S. and Boll, T. (1981). Changes in seizure frequency and
test-retest scores on Wechler adult intelligence scale. Epilepsia, 22, 7583.
Seidenberg, M., Beck, N. and Geisser, M. (1986). Academic achievement of children with epi-
lepsy. Epilepsia, 27, 7539.
Siebelink, B.M., Bakker, D.J., Binnie, C.D. and Kasteleijn-Nolst Trenite, D.G. (1988).
Psychological eects of subclinical epileptiform EEG discharges in children. II general intelli-
gence tests. Epilepsy Res, 2, 11721.
Sillanp, M., Gram, L., Johannessen, S.I. and Tomson, T. (1999). Epilepsy and mental retarda-
tion. Peterseld: Wrightson Biomedical Publishing.
Stores, G. (1985). Clinical and EEG evaluation of seizures and seizures-like disorders. J Am Acad
Child Psychiatry, 24, 1016.
Wester, K. (1982). Epilepsi hos psykik utviklingshemmede. Tidsskr Nor Laegeforen, 102, 2258.
69 Epilepsy and learning disorders
6
Subtle cognitive and behavioural effects of
epilepsy
Frank M.C. Besag
Bedfordshire and Luton Community NHS Trust, Bedford, UK
Introduction
Both cognitive and behavioural problems are common in people with epilepsy.
Epidemiological studies have indicated that around 50% of children with epilepsy
have some schooling diculties and that many have behavioural problems (Besag
et al., 1999; Pazzaglia and Frank-Pazzaglia, 1976; Ross et al., 1980; Sillanp, 1992).
The National Child Development Study in the UK (Ross et al., 1980) revealed that
only 67% of children with epilepsy were attending a mainstream school at age 11.
Using this broad-brush measure it was evident that a large proportion of the chil-
dren with epilepsy had signicant schooling problems. The work of Pazzaglia and
Pazzaglia in Cesena, Italy, (Pazzaglia and Frank-Pazzaglia, 1976) also conrmed
that a high proportion of children were underachieving at school. The excellent
epidemiological studies of Sillanp (Sillanp, 1992) in Finland showed that
31.4% of this unselected sample of children had mental retardation.
In a recent study carried out in the London Borough of Lambeth Besag et al.
(1999) surveyed 127 children with epilepsy. The parents perceived that the children
had schooling problems or learning disability in 65% and 48% were disturbed on
Rutter Behavioural Scales. Sixty-two per cent of the scores indicated a signicant
impact on quality of life which was moderate or great in 35%. These studies suggest
that schooling and behavioural diculties constitute a major problem in child-
hood epilepsy. There is a lack of data, however, to indicate the causes of the prob-
lems encountered, either in children or adults.
A systematic framework for assessing behaviour
The current author has suggested that, when faced with an individual who has epi-
lepsy and behavioural problems, the best approach is to use a systematic framework
for assessing the possible cause or causes of the behavioural disturbance (Besag et
al., 1989a; Besag, 1995). This framework consists of ve main categories: the epi-
70
lepsy itself, treatment of the epilepsy, reactions to the epilepsy, associated brain
damage/dysfunction and causes equally applicable to people without epilepsy.
The rst category, the epilepsy itself, may be broken down into the peri-ictal phe-
nomena of prodrome, aura, automatism and postictal changes, interictal psycho-
ses, focal discharges and frequent absence seizures. This list is not necessarily
comprehensive.
Examples of the way in which the epilepsy itself may aect behaviour have been
provided elsewhere(Besag et al., 1989a). A subtle cognitive or behavioural eect of
epilepsy may be dened as an eect that is not immediately obviously attributable
to an epileptic seizure. This does not imply that the epileptic activity itself is neces-
sarily subtle. For example, prodromal mood change preceding a tonic-clonic
seizure could be regarded as a subtle eect of the epilepsy, although the seizure itself
is not at all subtle.
On the other hand, the subtle cognitive and behavioural eects of epilepsy may
be the result of subtle seizure activity. In this context, subtle seizure activity is taken
to mean a clinical manifestation of epileptiform activity that is not immediately
obviously identiable as a seizure. Absence seizures and many complex partial sei-
zures would fall into this category as would transitory cognitive impairment. There
is a growing realization of the fact that epilepsy may aect cognition and behavi-
our in a variety of ways (Aldenkamp, 1997; Deonna, 1995; Stores and Hart, 1975;
Stores, 1978; Trimble, 1988). There is often an overlap between cognitive and
behavioural eects of epileptiform activity.
Frequent frontal discharges may lead to a high degree of social disinhibition and
consequent behavioural disturbance.
A teenager had a long history of gross behavioural disturbance. The EEG showed very fre-
quent left frontal discharges, typically occurring every second. He underwent a left frontal
lobectomy, following which the epileptiform discharges were abolished, his behavioural dis-
turbance resolved and he returned to his pleasant premorbid personality.
Frequent left temporal discharges may be associated with aggressive behaviour.
There is evidence from the literature suggesting that young men who have left tem-
poral discharges and subsequently undergo a left temporal lobectomy show
improvements both in seizure control and in behaviour (Falconer, 1973).
Further examples of both cognitive and behavioural disturbance that can result
from subtle manifestations of epilepsy will be discussed in the next section.
State-dependent cognitive impairment
It is important to distinguish permanent cognitive impairment on one hand from
state-dependent cognitive impairment on the other. The concept of permanent
71 Subtle effects of epilepsy
cognitive impairment is readily understood. This may arise from a wide variety of
causes of permanent brain damage or dysfunction that may be prenatal, perinatal
or postnatal. The concept of state-dependent cognitive impairment is less widely
acknowledged (Besag, 1994).
What is state-dependent cognitive impairment? State-dependent cognitive
impairment may be dened as cognitive impairment that depends on current
factors aecting the individual, for example antiepileptic medication or epileptic
activity, that are not necessarily permanent. State-dependent cognitive impairment
is potentially reversible and treatable. Failure to treat state-dependent cognitive
impairment is failure to provide an adequate service to the patient. Some might put
the case even more strongly, suggesting that failure to recognize and treat state-
dependent cognitive impairment is a reection on professional competence.
The rst step in managing state-dependent cognitive impairment is to think of
the diagnosis. Regrettably, professionals often fail to take this rst step. The result
is that the condition is neither recognized nor treated. State-dependent cognitive
impairment may be divided into two broad categories: drug-induced and epilepsy-
induced. As already indicated, the epilepsy itself may cause state-dependent cogni-
tive impairment in a number of dierent ways.
Frequent absence seizures
Frequent absence seizures, by interrupting awareness, can aect both cognitive per-
formance and behaviour. People who are having frequent absence seizures may
present as having withdrawn behaviour, fragmented thought processes which may
be mistaken for a psychosis, attention-decit disorder with motor overactivity or,
if the frequency of the seizures is variable, attention-seeking behaviour. The last of
these behaviours tends to be seen when the person emerges from a bout of very fre-
quent absence seizures. It is almost as if the child is making up for lost time in
being badly behaved when he has the opportunity of doing so, having been unable
to misbehave when the absence seizures were very frequent.
In some cases the absence seizures may become so frequent as to amount to non-
convulsive status epilepticus. The person is eectively cut o from his or her sur-
roundings because he or she does not have the opportunity to function adequately
between the absence seizures. Complex partial seizure status epilepticus may also
cause a continuous state of behavioural and cognitive change until the status epi-
lepticus is either treated or resolves spontaneously.
It is clear that absence seizures can aect both performance and condence in a
major way, particularly if they are occurring frequently. It is not possible to count
absence seizures by simple observation alone. To overcome this diculty, an auto-
matic spike-wave monitor, the Monolog was developed (Besag et al., 1989b). The
Monolog is so-called because it monitors and logs spike-wave episodes automat-
72 F.M.C. Besag
ically. Two recording electrodes and a reference electrode are attached to the scalp.
The leads are brought down to a small solid-state recording device, which is in a
belt-worn pouch.
This device provides three pieces of hard data: the total duration of the spike-
wave episodes over the period of recording, the number of spike-wave episodes
and, by playing the solid state circuitry into a chart recorder, the position of each
spike-wave episode in time. In addition, there is an optional plug-in lapel badge
with a ashing light and bleeper, activated by the presence of the spike-wave epi-
sodes. The teacher or carer is alerted by the ashing light/bleeper and he or she will
realize that the child is having a spike-wave episode; it may be appropriate for the
teacher/carer to repeat what has been said to the child during that episode. The light
and bleeper also serve to alert the teachers and carers to the high frequency of
absence seizures occurring in some children. Because absence seizures are often
quite subtle in their manifestation, the teacher may be surprised to note how often
the badge indicates that spike-wave discharges are occurring. This in itself may
result in an alteration of attitude by the teacher who may have an increased incli-
nation to repeat information if the child has not obviously taken it in on the rst
occasion. Using this monitor it has been possible to show that some children have
not only hundreds but thousands of spike-wave episodes daily. The device also con-
rms whether treatment is eective or not.
A 13-year-old boy with a history of epilepsy was reported as having reached the stage at
which the epilepsy was not a problem for him. He was having relatively infrequent overt sei-
zures. However, he presented as a withdrawn child who would not join group activities. He
preferred to sit in the corner of the room sucking his thumb. In the clinic the examiner
uttered the childs name on four occasions in succession. On the first three occasions there
was no response because he was momentarily unconscious in a subtle absence seizure. On
the fourth occasion, when he was not having an absence seizure, he responded promptly
and was more than willing to attend to whatever task the examiner asked him to do.
Prolonged EEG monitoring revealed that around three thousand spike-wave episodes
occurred daily. He responded very well to treatment.
Some individuals have both more obvious seizures, such as tonic-clonic or tonic
seizures and absence seizures. The antiepileptic medication may improve both the
obvious and the subtle seizures. However, in some cases the obvious seizures may
not be aected by medication while the subtle seizures are greatly reduced. Such
individuals may become bright, alert and in control of their lives as a result of the
medication change, even though the obvious seizures have not been reduced in fre-
quency. When patients are assessed in the outpatient clinic the doctor will typically
ask how many seizures have been recorded or will consult the patients seizure diary.
If there has been no reduction in the obvious seizures the doctor may choose to dis-
continue the medication with which the patient has been treated and try another
73 Subtle effects of epilepsy
antiepileptic drug. However, on this basis the doctor might stop a drug that has
been highly eective in treating subtle absence seizures, even though there has been
no major eect on the obvious seizures. An example is provided in Figures 6.1 and
6.2. In Case A there has been a reduction both in the obvious seizures and in the
spike-wave episodes with the addition of lamotrigine. In Case B, however, the
obvious seizures have not decreased. The spike-wave episodes, on the other hand,
have been reduced by over 2000 per day. The parents of this teenager were delighted
with the transformation. He was able to relate much more readily to the world
around him and appeared much more alert. It is important not to discontinue
medication on the basis that obvious seizures have not responded if the individual
has had a good response in terms of reduction of subtle seizure activity.
Transitory cognitive impairment
A number of papers describing this phenomenon have been published by Binnie
and coworkers (Aarts et al., 1984; Marston et al., 1993). The basic concept is that
an epileptiform discharge that does not appear, on simple observation, to be mani-
festing as a seizure, may nevertheless cause a transitory impairment of cognitive
function. It has been shown that discharges on the left side may impair language
function and on the right side may impair visuo-spatial skills. The quality of the
impairment is not always straightforward. For example, a child who is having epi-
74 F.M.C. Besag
120
100
80
60
40
20
0
4000
3000
2000
1000
0
Baseline 13 46 79 Baseline 2 4 10 8
(a) Overt seizures
(a) in 3 months
(b) Spike-wave events
(b) in 24 hours
N
u
m
b
e
r
o
f
s
e
i
z
u
r
e
s
N
u
m
b
e
r
o
f
e
v
e
n
t
s
Months Weeks
Figure 6.1. The effect of lamotrigine administration on number of overt seizures (a) and spike-wave
events (b) experienced by Subject A.
leptiform discharges during reading may pause or may read even more quickly but
make additional mistakes when discharges are occurring.
Binnie and his co-workers have shown that reduction of epileptiform discharges
can result in improvement of psychosocial functioning (Marston et al., 1993),
although this study was confounded by the fact that obvious seizures were
improved as well as the discharges causing the transitory cognitive impairment.
Observation of video tapes taken during testing of young people who have transi-
tory cognitive impairment leaves no doubt about the fact that they nd their lapses
in performance irritating and frustrating. It is highly likely that this phenomenon
aects self-esteem and self-condence. This implies that, although the phenome-
non itself is transitory, there may be an ongoing negative eect on the attitude and
behaviour of the individual.
Frequent localized discharges
Reference has already been made to frequent frontal discharges and frequent left
temporal discharges aecting behaviour. Localized discharges can also aect
ongoing cognitive performance. This eect is more than just transitory if the dis-
charges are frequent.
A girl had a history of a benign left-sided temporal lobe tumour. She had very frequent epi-
leptiform discharges arising from the left temporal lobe. She underwent left temporal lobec-
tomy at 13 years of age. The epileptiform discharges disappeared after the neurosurgery. In
a single year her speech development improved from around the 4-year level to the 7-year
75 Subtle effects of epilepsy
(a) Overt seizures in 3 months (b) Spike-wave events in 24 hours
Baseline 13 46
Months
79
Baseline
120
100
80
60
40
20
0
3000
2500
2000
1500
1000
500
0
2 4 6 10 32 48
Weeks
N
u
m
b
e
r
o
f
e
v
e
n
t
s
N
u
m
b
e
r
o
f
s
e
i
z
u
r
e
s
Figure 6.2. The effect of lamotrigine administration on number of overt seizures (a) and spike-wave
events (b) experienced by Subject B.
level. The temporal lobectomy released the remaining brain from the effect of very frequent
epileptiform discharges, allowing rapid progress to be made.
Frequent hemispheric discharges
The extreme example of frequent localized discharges is provided by individuals
who have an abnormal hemisphere which is the source of such discharges. A
number of dierent pathological conditions can give rise to this phenomenon,
including hemimegalencephaly, a porencephalic cyst and the unilateral cerebral
atrophy that accompanies Rasmussens encephalitis. These individuals often
improve markedly after hemispherectomy. Not only are the seizures and epilepti-
form discharges abolished but the behaviour also improves greatly(Goodman,
1986). If there is a gross structural abnormality that is causing very frequent epi-
leptiform discharges then early surgery should be considered. If the lesion respon-
sible has been present from a very early age, for example if it is a congenital
porencephalic cyst, then the outcome is likely to be good.
A dilemma arises in the case of the child who develops Rasmussens encephalitis
that is causing progressive hemiatrophy of one hemisphere. The function subserved
by the aected hemisphere is likely to worsen and the longer the operation is
delayed the less opportunity there will be for the healthy hemisphere to take over
function. However, hemispherectomy usually results in loss of nger function on
the aected side and there is a tendency to delay the operation until it is clear that
the nger function has already been lost, implying that the surgery will not impair
function further.
Is it appropriate to wait? Because brain plasticity is greatest when the child is
young, there is a better chance that the remaining hemisphere will take over func-
tion if the surgery is performed early. There is a strong argument, in such cases, for
carrying out the surgery sooner rather than later if the assessment indicates that the
disease in the aected hemisphere is progressive.
Postictal state-dependent cognitive impairment
At rst sight it may appear trivial to discuss postictal cognitive impairment because
this is such an obvious phenomenon. However, it is not always as obvious as one
might imagine.
An apparently dementing teenager sat rocking in his chair, not knowing where he was,
barely able to carry out a rudimentary conversation and unable to perform his skills of daily
living. He had been accepted as a boarding pupil in a special residential centre for children
with epilepsy and other needs. The staff felt that he had been placed inappropriately. They
commented that he could not learn and that he should not have been accepted. At that
76 F.M.C. Besag
stage he was having between three and five seizures a day. Following a medication review
he became seizure-free. He was then fully orientated and began progressing very well in his
educational programmes. Both his parents and the staff were delighted with his progress.
The staff then commented how appropriate the placement was. They had previously
assumed that his cognitive impairment was permanent whereas, in fact, a large component
of the cognitive impairment was state-dependent, treatable and reversible. When he had
been having frequent daytime seizures he had not had the chance to recover from one
seizure before he had another. He was in a constant postictal state. When he emerged from
this constant postictal state he was able to learn again and progressed rapidly.
It is important to be aware of the possibility of reversible, state-dependent cogni-
tive impairment in an individual who is in a constant postictal state from frequent
seizures. Reducing the seizure frequency can improve cognitive function markedly.
Overnight video-telemetry has revealed that some people have very frequent
unsuspected and unobserved nocturnal seizures. These seizures may be brief, silent
and easily missed, even by awake night sta. In a series of 15 patients examined by
the author and his coworkers, large numbers of nocturnal seizures were recorded
by video-telemetry, that had been unobserved by awake night sta. In one case over
200 brief nocturnal tonic seizures were recorded. Some individuals are woken by
the seizures. Frequent nocturnal seizures may aect daytime performance not only
because of the direct after-eects of the seizures themselves but also because they
may cause a very broken nights sleep.
Electrical status epilepticus of slow wave sleep
The LandauKlener syndrome of acquired epileptic aphasia is classically asso-
ciated with electrical status epilepticus of slow wave sleep (ESES) (Beaumanoir,
1992; Landau and Klener, 1957). However, a series of six cases of ESES examined
by the author and colleagues conrmed that language is not necessarily the func-
tion that is impaired. Although some of these children clearly had language impair-
ment in association with the ESES, others did not.
A boy with a right congenital porencephalic cyst and an accompanying left hemiparesis had
ESES. He had good language skills but his visuo-spatial skills became increasingly impaired.
He was unable to find his way from his bedroom to the bathroom at home. Following neuro-
surgery his visuo-spatial skills improved greatly.
The LandauKlener syndrome provides an important model of state-dependent
cognitive impairment. Some of the older textbooks suggested that antiepileptic
treatment was only of value in treating the seizures but would not aect the cogni-
tive impairment. Such statements were probably based on burnt-out cases in which
the damage had become permanent. It is quite clear that early, energetic treatment
77 Subtle effects of epilepsy
may improve the cognitive impairment in some children with LandauKlener syn-
drome. The implication is that not only may cognition be improved by early treat-
ment but that permanent impairment may be avoided.
Treatments include antiepileptic medication such as sodium valproate and
lamotrigine, steroid treatment and neurosurgery. Multiple subpial transection,
pioneered by Morrell (Morrell et al., 1989), has been particularly eective in some
cases. This can abolish the ESES and allow the language function to return. The
results obtained from multiple subpial transection in this condition have made it
clear that past statements that antiepileptic treatment could not help cognitive loss
were wrong. It is worth repeating that early treatment may not only result in a
return of these skills but may also prevent long-term cognitive impairment.
The issues raised by ESES have led to the suggestion that any child who loses skills
should be investigated fully and, if another cause is not found, investigations
should include overnight EEG monitoring. It should be noted, in this context, that
it is said that around 25% of the children with LandauKlener syndrome do not
have a previous history of obvious seizures (Beaumanoir, 1992).
Preventing cognitive and behavioural problems
In the examples just given, it was emphasized that some children with ESES may
present with impaired cognition. They often also present with markedly disturbed
behaviour. Both the cognitive and behavioural disturbance are reversible with early
treatment.
It has become evident that a number of children who present with markedly
autistic features have unsuspected epileptiform discharges either during the day or
at night (ESES or continuous spike-wave in slow wave sleep CSWS). Treating
these children allows them to emerge from their apparently autistic state.
However, follow-up of teenagers who had very frequent absence seizures in
childhood has indicated that some longer-term problems in social interaction or
behaviour may be evident. Some teenagers who have had very frequent epilepti-
form EEG discharges earlier in life may subsequently still have autistic features
when the epileptiform discharges are no longer present. It would appear that these
young people have been unable to interact adequately with the world around them
during a critical developmental phase because of the frequent epileptiform dis-
charges at that time. If they have not had the opportunity to develop two-way social
interaction during this developmental phase they may present with a very
Asperger-like picture during teenage years. This suggests that early treatment of the
epileptiform discharges, so as to allow the child to gain fully from the early devel-
opmental years, may have avoided these social impairments.
In summary, it appears that there is now an increasing body of evidence to
78 F.M.C. Besag
suggest that frequent epileptiform discharges should not be allowed to continue for
very long periods because they might cause not only permanent cognitive impair-
ment but also permanent social/behavioural problems.
Conclusions
Systematic assessment of cognitive and behavioural problems often allows the
cause to be found and rational management to be provided. It is very important to
consider the possibility of state-dependent cognitive impairment, especially in the
child who has lost skills. Unless another obvious cause is found, the investigations
should include overnight EEG monitoring. If frequent epileptiform discharges are
found in association with cognitive or behavioural problems then early, energetic
treatment should be initiated. Such treatment may not only improve the impair-
ments at the time but may also prevent permanent cognitive or social impairments.
REFERENCES
Aarts, J.H., Binnie, C.D., Smit, A.M. and Wilkins, A.J. (1984). Selective cognitive impairment
during focal and generalized epileptiform EEG activity. Brain, 107, 293308.
Aldenkamp, A.P. (1997). Eect of seizures and epileptiform discharges on cognitive function.
Epilepsia, 38 (Suppl. 1), S525.
Beaumanoir, A. (1992). The LandauKlener Syndrome. In Epileptic Syndromes in Infancy,
Childhood and Adolescence, ed. J. Roger, M. Bureau, Ch. Dravet, F.E. Dreifuss, A. Perret and P.
Wolf, pp. 23143. London: John Libbey.
Besag, F.M.C. (1994). Epilepsy, education and the role of mental handicap. In Epilepsy, ed. E.M.
Ross and R.C. Woody, pp. 56183. Baillieres Clin Paediatr Int Practice Res, 2, 56183.
Besag, F.M.C. (1995). Epilepsy, learning, and behavior in childhood. Epilepsia, 36, S5863.
Besag, F.M.C., Loney, G., Waudby, E., Fowler, M. and Brooks, N. (1989a). A multidisciplinary
approach to epilepsy, learning diculties and behavioural problems. Educational Child
Psychol, 6, 1824.
Besag, F.M., Mills, M., Wardale, F., Andrew, C.M. and Craggs, M.D. (1989b). The validation of a
new ambulatory spike and wave monitor. Electroencephalogr Clin Neurophysiol, 73, 15761.
Besag, F., ONeill, C. and Ross, E. (1999). A comparison between children with epilepsy in an
inner-city region and those within a special centre, using measures of educational diculty,
behavioural problems and quality of life. Epilepsia, 40, 243.
Deonna, T. (1995). Cognitive and behavioral disturbances as epileptic manifestations in children:
an overview. Semin Pediatr Neurol, 2, 25460.
Falconer, M.A. (1973). Reversibility by temporal-lobe resection of the behavioral abnormalities
of temporal-lobe epilepsy. N Eng J Med, 289, 4515.
Goodman, R. (1986). Hemispherectomy and its alternatives in the treatment of intractable epi-
lepsy in patients with infantile hemiplegia. Dev Med Child Neurol, 28, 2518.
79 Subtle effects of epilepsy
Landau, W.M. and Klener, F.R. (1957). Syndrome of acquired aphasia with convulsive disorder
in children. Neurology, 7, 52330.
Marston, D., Besag, F., Binnie, C.D. and Fowler, M. (1993). Eects of transitory cognitive impair-
ment on psychosocial functioning of children with epilepsy: a therapeutic trial. Dev Med Child
Neurol, 35, 57481.
Morrell, F., Whisler, W.W. and Bleck, T.P. (1989). Multiple subpial transection: a new approach
to the surgical treatment of focal epilepsy [see comments]. J Neurosurg, 70, 2319.
Pazzaglia, P. and Frank-Pazzaglia, L. (1976). Record in grade school of pupils with epilepsy: an
epidemiological study. Epilepsia, 17, 3616.
Ross, E.M., Peckham, C.S., West, P.B. and Butler, N.R. (1980). Epilepsy in childhood: ndings
from the National Child Development Study. Br Med J, 280, 20710.
Sillanp, M. (1992). Epilepsy in children: prevalence, disability, and handicap. Epilepsia, 33,
4449.
Stores, G. (1978). School-children with epilepsy at risk for learning and behaviour problems. Dev
Med Child Neurol, 20, 5028.
Stores, G. and Hart, J.A. (1975). Proceedings: reading skills of children with generalized and focal
epilepsy attending ordinary school. Electroencephalogr Clin Neurophysiol, 39, 42930.
Trimble, M.R. (1988). Cognitive hazards of seizure disorders. Epilepsia, 29 (Suppl. 1), S1924.
80 F.M.C. Besag
7
Aggression and epilepsy
L. Tebartz van Elst
Institute of Neurology, London, UK and Albert-Ludwigs-University, Freiburg, Germany
Introduction
Human aggression is an important social and clinical problem (Fenwick, 1986; Saver
et al., 1996; Swartz et al., 1998; Trimble, 1996). One diculty of studying aggression
is its phenomenological and probably neurobiological heterogeneity, leading to di-
culties in assessment and classication. An important distinction has emerged
between the terms violence and aggression. Following Treiman, violence has been
dened as forceful iniction of abuse or damage onto another individual or object,
but which is not necessarily the result of intentional aggression. In contrast, aggres-
sion is considered an oensive action directed toward another individual or object
with the intent to harm, threaten or control (Treiman, 1991). But even this distinc-
tion leaves researchers with the problem of assessing intentionality, which in clinical
practice often is impossible. Therefore, it is important to dene phenomenological
criteria of the specic behavioural syndrome of interest before embarking on
research or discussion into the neurobiology and psychology of aggression.
Classification of aggression
In the animal kingdom aggressive behaviour has been classied according to the
context in which it can be observed. Table 7.1 summarizes dierent subtypes of
aggressive behaviour in animals following Moyer (1987). In contrast to animal
behaviour, human behaviour is less preformed and it depends less on external cues.
Thus a simple transfer of this classication to human aggressive behaviour is not
valid (Kalin, 1999). Nevertheless, there is general agreement that at least two dier-
ent phenomenological and neurobiological subtypes of aggressive behaviour can
be dierentiated in humans: predatory and defensive aggression (Goldstein, 1974;
Kalin, 1999; Moyer, 1987; Mungas, 1983; Vitiello and Sto, 1997).
Predatory aggression is characterized phenomenologically as a well-structured
and goal-directed behaviour performed in an emotionally calm and concentrated
state of mind. In contrast, defensive aggression is typically being seen in the context
81
of high emotional arousal associated with vocalizations and signs of fear or anger.
The behaviour itself is less structured and defensive (Valzelli, 1981). Apart from the
planned and goal-directed aggression often conducted by persons with antisocial
personality disorder (predatory aggression), most forms of human aggression are
thought to be a reaction toward a perceived threat, be it adequate or not (Albert et
al., 1993). Obviously, the perception as to whether a stimulus is threatening or not
is decisive in the information processing leading to the aggressive behaviour.
Clinical relevance of aggression
Aggressive behaviour can be observed in the context of dierent medical, neuro-
logical and psychiatric disorders and diseases. It is a common problem in patients
with mental retardation, possibly due to impaired social perception or decits in
expressing personal needs (Barratt et al., 1997; Gunn, 1977; Kligman and Goldberg,
1975; Saver et al., 1996). Aggressive behaviour in the context of organic brain
disease like frontal or hypothalamic brain tumours, neuro-degenerative disease,
delirium or drug abuse is often malstructured, defensive and tends to occur in the
context of states of confusion and diuse emotional arousal. Goal-directed and
well-planned acts of aggression can occur on the background of psychiatric dis-
orders like psychosis with delusional states, attention-decit hyperactivity disorder
(ADHA) or bipolar disorder. It is frequently observed in patients with antisocial
personality disorder (APD) where it is part of the characteristic trait-like behaviour
(Barratt et al., 1997; Miller et al., 1997; Stein et al., 1993).
82 L. Tebartz van Elst
Table 7.1. Subtypes of animal aggression
Aggression type Example of animal behaviour Behavioural characteristic
Predatory aggression Predator kills prey Calm, concentrated, goal directed, well
structured
Intermale (territorial) Fight for supremacy of two lions Arousal, concentrated, goal directed, well
aggression structured
Maternal aggression Swan protects ospring from Arousal, concentrated, goal directed, well
predator structured
Sex-related aggression Mantis kills male after copulation Calm, behavioural stereotypes
Fear-induced aggression FlightFightReaction Arousal, vocalization, less well structured,
Gnu ghts predator diverse behavioural pattern, reactive
behaviour
Source: Moyer (1987).
The only clinical syndrome of aggression that has been included in an interna-
tional classicatory system is intermittent explosive disorder (IED) according to the
guidelines of DSMIV (American Psychiatric Association, 1994). IED is character-
ized by several discrete episodes of failure to resist aggressive impulses that result
in serious assaultive acts or destruction of property. The behaviour is out of pro-
portion to any precipitating psychosocial stressor and is not due to substance abuse,
another mental disorder like personality disorder, any other rst axis psychiatric
disorder or a general medical condition like head trauma or neuro-degenerative
diseases. The phenomenological criteria are basically those of episodic dyscontrol, a
psychopathological entity that has been described by many authors in the past
(Bach-Y-Rita et al., 1971).
Neurobiology of aggression
Various brain structures have been implicated in the mediation of aggressive
behaviour in animals and humans, the most important being the periaqueductal
grey (Behbehani, 1995; Brandao et al., 1994), the hypothalamus (Andy and Jurko,
1972), the amygdala and associated limbic structures (Aggleton, 1993; Dicks et al.,
1969; Halgren, 1992; Kling and Brothers, 1992; Rolls, 1992), and the frontal lobes
(Damasio et al., 1990; Miller et al., 1997; Raine et al., 1998).
While an understanding of the complex interplay of these dierent brain circuits
in regulating dierent aggressive behaviours is still poor, the rst elements of a
functional anatomy of aggression can be outlined (see Table 7.2). In animals, brain
stem structures like the periaqueductal grey are crucial for the activation of evolu-
tionary preformed behavioural programs like attacking or defensive behaviour
83 Aggression and epilepsy
Table 7.2. Functional relevance of different brain structures for aggressive behaviour
Brain structure Assumed function
Frontal lobe Inhibitory function
Suppression of aggressive behavioural drive
Amygdala and limbic circuits Emotional evaluation of multimodal sensory and cognitive
input; emotional drive and arousal
Hypothalamus Control of brain stem behavioural programs; regulation of
internal environment; coordination of behavioural programs
and internal environment in ightght situations
Brain stem structures i.e. Evolutionary performed behavioural ightght programs
periaqueductal grey
(Behbehani, 1995; Brandao et al., 1994). These structures are controlled by higher
neuronal centres in the hypothalamus (Bhatnagar and Dallman, 1998; Van de Poll
and Van Goozen, 1992) which in addition to controlling these behavioural brain-
stem programs adjust the internal endocrinological and immunological environ-
ment to aggressive behaviour in ightght situations (Luo, 1998; Reis, 1969;
Shaikh, 1997; Zanchetti, 1968).
Frontal lobe functions are known to be crucial for the ability to suppress behav-
ioural impulses. Consequently patients with frontal lobe lesions lose their ability to
suppress aggressive impulses and thus might present with severe aggressive and
violent psychopathology (Damasio et al., 1990; Krakowski and Czobor, 1997; Petty
et al., 1996; Stein et al., 1993).
Within the network of critical brain structures for aggressive behaviour, the
amygdala are thought to play a crucial role for the mediation of fear-induced
aggression, a subtype of defensive aggression (Aggleton, 1993; Charney and
Deutch, 1996; Gallagher and Chiba, 1996; LeDoux, 1995). They receive extensive
input from various levels of sensory information processing and project to most of
the other critical brain structures i.e. the brain stem, hypothalamus, thalamus and
frontal lobe (Alheid et al., 1995; Amaral et al., 1992). From a neurophysiological
point of view they are in a predestined position for the aective evaluation of multi-
modal sensory input. Thus pathology within the circuits aecting the amygdala
might lead to mental states where the misinterpretation of sensory input as threat-
ening leads to aggressive outbursts. In agreement with this assumption, electrical
stimulation of the amygdala can lead to experiences like fear, anxiety or anger
(Chapman et al., 1954; Gloor et al., 1982) and lesioning of the amygdala severely
impairs fear conditioning in animals (Davis et al., 1994) and humans (LaBar et al.,
1995). Furthermore, in an open retrospective study of 481 cases of bilateral amyg-
dalotomies performed for the control of conservatively untreatable aggressiveness
moderate to excellent improvement of aggressive behaviour was reported in
7076% (Ramamurthi, 1988).
Aggression and epilepsy
The relationship between epilepsy and aggressive behaviour is a particularly con-
troversial issue (Geschwind, 1975). While in patients with episodic aective aggres-
sion, a history of epilepsy is reported to be more common (Bach-Y-Rita et al., 1971;
Elliot, 1982) most of the community-based studies did not nd an increased prev-
alence of aggressive behaviour in patients with epilepsy (Kligman and Goldberg,
1975; Lishman, 1998).
The prevalence of aggression in epilepsy in general, not regarding the specic
epileptic subsyndrome, varies between 4.8% (Rodin, 1973) and 50% (Gastaut et al.,
84 L. Tebartz van Elst
1955). In a large survey of 666 patients with TLE, Currie et al. (1971) reported
aggression in 7% of the patients. Falconer reviewed 100 patients from Londons
Maudsley Hospital referred for temporal lobectomy and found a prevalence of out-
bursts of aggressive behaviour in 27% of their patients (Falconer, 1973). However,
these studies were hampered by selection bias and the real prevalence of aggressive
behaviour in epilepsy remains controversial (Lishman, 1998).
In epilepsy three dierent types of aggressive behaviours should be distinguished
on the basis of their relationship toward the seizures: ictal, postictal and interictal
aggression. Ictal aggression occurs with extreme rarity (Gunn, 1971; Saver et al.,
1996). In a large survey of several thousand seizures documented on video-
telemetry an incidence of about 1/1000 seizures was found (Delgado-Escueta et al.,
1981). In ictal aggression hostile and verbal or physical aggressive behaviour is often
directed towards nearby objects or persons and may be provoked or not. The
patients are usually amnestic for these aggressive episodes and express remorse for
their behaviour (Devinsky and Bear, 1984; Fenwick, 1989).
Postictal aggression is more common than ictal aggression but it is still believed
to be rare (Treiman, 1991). It often occurs following a cluster of complex partial
seizures or very severe secondary generalized seizures. Ictal pain or dysphoria may
predispose individuals to the development of postictal aggressive behaviour
(Gerard, 1998). Postictal aggression is frequently observed in the context of post-
ictal confusional states or postictal psychosis but it also occurs without any signs of
delusion or hallucination (Kanemoto, 1999; Lancman, 1999; Szabo and Lancman,
1996). If postictal aggression is part of a postictal confusional state the disruptive
behaviour immediately follows the seizure without a lucid interval. The violent
behaviour tends to be resistive, poorly structured and patients usually are very
aroused, angry and fearful (Kanemoto, 1999; Lancman, 1999).
Postictal psychosis follows a cluster of complex partial and secondary general-
ized seizures after a lucid interval, which might vary in duration between hours up
to days (Fenwick, 1986; Kanemoto, 1996; Logsdail and Toone, 1988; Trimble,
1991). Aggressive behaviour in the context of a delusional state may be well-
structured and goal-directed and even though patients often feel angry and
aroused, they may appear calm and concentrated to the observer (Kanemoto, 1999;
Szabo and Lancman, 1996). Kanemoto et al. (1999) recently pointed out that well-
directed and self-destructive behaviour was not a feature of epileptic psychosis in
general but the hallmark of postictal psychosis in particular.
Finally, interictal aggression can be seen in the context of an antisocial personal-
ity disorder which, in turn, might be the consequence of the sometimes dicult
psychosocial background of patients with epilepsy. It might also be part of a psy-
chotic episode (Logsdail and Toone, 1988) but as Kanemoto (1999) pointed out
this is rare compared with postictal psychosis.
85 Aggression and epilepsy
Interictal aggression is frequently seen in patients with epilepsy and mental
handicap, but in these patients aggressive behaviour is often a result of poor social
and communication competence in expressing personal needs, and very rarely
results in severe violence (Gunn, 1977). Apart from that, an interictal syndrome of
episodic aective aggression, independent of observable ictal activity, major
psychiatric disorder or antisocial personality disorder, is well described and has
been referred to as episodic dyscontrol (Bach-Y-Rita et al., 1971; Elliott, 1984;
Leicester, 1982; Maletzky, 1973; Ratner and Shapiro, 1979; Stone et al., 1986).
Episodic dyscontrol is characterized by several discrete episodes of extreme
unprovoked arousal and rage resulting in severe aggressive and violent behaviour.
The behaviour is out of proportion to any precipitating psychosocial stressor and
patients often feel remorse for their deeds. The phenomenological criteria are those
of DSMIV intermittent explosive disorder (IED) described above (Elliott, 1984).
Because of the emotional arousal typically seen in episodic dyscontrol, this behav-
ioural syndrome has to be classied as a subtype of aective, defensive aggression.
The high level of arousal with signs of anxiety or fear is also the reason why amyg-
dala pathology is thought to contribute to this behavioural syndrome (Elliot, 1992;
Fenwick, 1986; Trimble et al., 1996).
Intermittent explosive disorder in epilepsy
In the past, few studies have addressed the relationship between dierent psycho-
biological factors like brain pathology, IQ and demographic background, and
aggression in epilepsy. While Rodin found more evidence of organic brain disease
(Rodin, 1973), and Falconer (1973) reported an increased incidence of mesial tem-
poral lobe sclerosis in aggressive patients with temporal lobe epilepsy (TLE),
Herzberg and Fenwick did not nd any relationship between specic electroen-
cephalography (EEG) or computerized tomography (CT) pathology and aggres-
sion in patients with TLE (Herzberg and Fenwick, 1988). All three studies found a
relationship between low IQ and aggression and two reported an association
between male sex and aggression (Falconer, 1973; Rodin, 1973).
Since none of these studies used modern magentic resonance imaging (MRI)
techniques to assess the relationship between brain pathology and aggressive
behaviour in patients with epilepsy, in two recent projects we studied this relation-
ship using quantitative MRI (Tebartz van Elst et al., 2000a,b; Woermann et al.,
2000). The aims of our studies were rst, to investigate amygdala pathology, and
second, to assess cortical abnormalities in patients suering from TLE and addi-
tional aective aggression, specically IED.
The most common pathology underlying TLE in general is hippocampal scler-
osis (HS) often in the context of mesial temporal sclerosis (MTS) (Gloor, 1991;
86 L. Tebartz van Elst
Margerison and Corsellis, 1966; Wieser, 1983). A radiological in vivo diagnosis of
mesial temporal lobe sclerosis is possible by demonstrating atrophy of the mesial
temporal lobe structures on T1-weighted anatomical MRI images and increased
signal on conventional spin echo T2 MRI sequences (Duncan et al., 1996; Jackson
et al., 1990; Woermann et al., 1998). Since HS seems to be diuse rather than focal
in most of the cases (Kim et al., 1995) an involvement of the amygdala by the patho-
logical process underlying hippocampal sclerosis might be expected, and indeed is
reported in the literature (Hudson et al., 1993; Miller et al., 1994). In vivo identi-
cation of amygdala sclerosis by measuring the amygdala T2 relaxation time has
been reported in patients with TLE (Kalviainen et al., 1997; Van Paesschen et al.,
1996). Furthermore amygdala volumetry has been validated as a reliable method
(Cendes et al., 1993; Kalviainen et al., 1997; Soininen et al., 1994; Watson et al.,
1992).
In our study we hypothesized that, in patients with TLE and intermittent aec-
tive aggression, amygdala sclerosis in the context of hippocampal sclerosis would
be more common as compared to control patients. Furthermore, we wanted to test
if there is an association between aggression on the one hand and hippocampal
sclerosis, low IQ, and poor social adjustment on the other hand in patients with
TLE.
In a second step we analysed cortical grey matter abnormalities in these patients
in order to investigate frontal lobe pathology, since from a phenomenological point
of view episodic dyscontrol or IED can be characterized as a hyperarousaldyscon-
trol syndrome.
Amygdala pathology in patients with TLE and IED
Twenty-ve patients with TLE and IED diagnosed according to the DSMIV crite-
ria described above and 25 control patients with TLE without any psychopathol-
ogy were recruited from a tertiary referral centre (National Hospital for Neurology
and Neurosurgery and the associated Chalfont Centre for Epilepsy). The clinical
syndrome of interest was dened as complex partial seizures with a semiology, EEG
and MRI ndings compatible with TLE. On the basis of the discharge summaries
patients with TLE with and without a history of aggression were identied, con-
tacted and seen by a neuropsychiatrist (LTVE). Patients with extratemporal or gen-
eralized epilepsy were excluded as were those with a history of mental handicap or
psychoses. Patients with TLE with and without a history of IED diagnosed accord-
ing to DSMIV criteria were included in the study.
A neurological and psychiatric history and examination were obtained, as well as
routine EEG investigations and neuropsychological investigations. Full (FIQ), verbal
(VIQ), and performance IQ (PIQ) were measured and patients with a FIQ below 70
were excluded from the study to avoid selection bias. All patients were asked to ll in
87 Aggression and epilepsy
the Beck Depression Inventory (BDI13) and the State Trait Anxiety Inventory
(STAI). Both questionnaires are self-rating instruments for depression and anxiety
respectively (Thomson, 1989a, b). In order to assess aggression, carers were asked to
ll in the Social Dysfunction and Aggression Scale (SDAS21), a well validated and
recommended questionnaire (European Rating Aggression Group, 1992; Mak and de
Konning, 1995). Twenty healthy volunteers, matched for age and sex were scanned
and measured twice in order to assess the reliability of the volumetric measurements.
The MRI images were obtained at the Chalfont Centre for Epilepsy on a 1.5 T GE
Signa scanner (GE Medical Systems, Milwaukee, USA) using a T1weighted
inversion-recovery prepared volume acquisition (IRSPGR: TI/TR/TE/ip
450/15/4.2/20; 121.5 mm thick contiguous coronal slices; matrix 256192,
24 cm18 cm FOV), and a conventional spin echo sequence (TR/TE1/TE2/NEX
2000/30/120/1, 256192 matrix, 2418 cm FOV, 5 mm slices) for computation
of T2 values. Volumetric measurements were performed using a locally developed
interactive software program MRreg (Lemieux et al., 1998; Moran et al., 1999) fol-
lowing the established protocol described by Watson et al. (1992). The rater (LTVE)
was blind to the subject grouping. The amygdala volumes were corrected for total
brain size by division by the intracranial volume. Intrarater reliability gures were
carefully assessed and proved to be satisfactory. Amygdala atrophy was dened as a
volume smaller than 3 standard deviations (SD) below the average amygdala
volume of the control group. Amygdala T2 values were measured using DispImage
image analysis software (Plummer, 1992) by placing the largest possible elliptic
region of interest within the amygdala while avoiding anatomical boundaries.
Amygdala T2 signals were dened as pathological if they were greater than 2 SD
above the mean of the control population. Details of the methodology have been
published elsewhere (Tebartz van Elst et al., 1999, 2000b).
The demographic data of both groups are summarized in Table 7.3. The two
patient groups were matched for age, sex, demographic background, duration of
epilepsy and seizure severity. There was no signicant group dierence regarding
the history of birth complications, febrile convulsions or status epilepticus.
However, the incidence of encephalitic brain disease (Fishers Exact Test: P0.05)
and left-handedness (Chi-square Test: P0.05) was signicantly increased in
aggressive patients. There was less right-sided focal EEG abnormality and more
bilateral EEG abnormality in the aggressive group compared to nonaggressive
patients with TLE. Left- as well as right-sided hippocampal sclerosis was signi-
cantly less common in patients with TLE and IED. Other left temporal pathology,
including three patients with amygdala pathology (amygdala sclerosis, amygdala
glioma, amygdala DNT), two patients with multiple small temporal infarctions and
two patients with diuse left temporal atrophy of unknown origin, was signi-
cantly more common in patients with TLE plus IED (see Table 7.4).
88 L. Tebartz van Elst
89 Aggression and epilepsy
Table 7.3. Demographic and historical data of patient groups with temporal lobe epilepsy
(TLE), with TLE and intermitttent explosive disorder (IED) or without IED (TLE alone)
(n25 in each group)
TLE and IED TLE alone
Age (in years) [range] 30.1 [1849] 33.8 [1956]
Sex: f/m 8/17 10/15
Work: number unemployed 21 15
Living: number living independently 14 10
Income: number on social support 18 16
Social: number living in stable relationship 8 8
Therapy: monotherapy polytherapy 322 322
Mean duration of TLE (in years) [range] 22.4 [545] 24.5 [746]
Mean seizure frequency (estimated frequency 13.4 [0.560] 21 [1.5190]
per month) [range]
Birth complications 9 7
Febrile convulsions 5 9
Status epilepticus 0 2
Left-handed 7 1
History of encephalitis 5 0
Source: Tebartz van Elst et al. (2000a).
Table 7.4. Radiological MRI pathology on visual assessment
Other left
Overall signicance: No Right Left Bilateral temporal
P0.002 pathology hemisphere hemisphere hemispheres pathology
TLE and IED 10 1 4 3 7
TLE alone 6 8 10 1 0
Signicance ** ** **
Notes:
TLE, temporal lobe epilepsy; IED, intermittent explosive disorder.
Other left temporal pathology: amygdala sclerosis1, amygdala glioma1, amygdala DNT
1; multiple small temporal infarctions2; generalized left temporal atrophy2.
Closed test procedure: *P0.5, **P0.01).
Source: Tebartz van Elst et al. (2000b).
There was no evidence of an increased amygdala T2 relaxation time in the
aggressive group. Using the denition of amygdala T2 pathology as a T2 time
greater than 2 SD above the mean of the control group, there was no signicant
group dierence in amygdala pathology.
A group comparison did not reveal a signicant overall dierence in amygdala
volumes (right side: aggressive1893 mm
3
, SD435 mm
3
; nonaggressive1909
mm
3
, SD231 mm
3
; left side: aggressive1840 mm
3
, SD398 mm
3
; nonaggres-
sive1868 mm
3
, SD290 mm
3
). However, in the aggressive patients, a subgroup
of ve patients (20%) showed amygdala atrophy as compared to only one in the
nonaggressive group (Chi-square: P0.04). Two patients had left-sided amygdala
atrophy, two had bilateral atrophy and the only patient who exhibited right-sided
amygdala atrophy was left-handed. An increased incidence of encephalitis (Chi-
square Test: P0.05; Fishers Exact Test: P0.1) was the only clinical feature that
distinguished patients with amygdala atrophy from those with normal amygdala
volumes. Since there was no overlap between the ve patients with severe amygdala
atrophy and those seven patients with other amygdala pathology, in a total number
of 12 out of 25 aggressive patients there was some evidence of amygdala pathology
as compared to only one in the nonaggressive group (see Figure 7.1).
There was a highly signicant group dierence in IQ gures with the verbal IQ
(VIQ), the performance IQ (PIQ) and hence the full IQ (FIQ) all being lower in the
aggressive group (Table 7.5). The verbal IQ diered more than the performance IQ.
As expected there was a highly signicant group dierence in the Social
Dysfunction and Aggression Scores (SDAS 9, SDAS 21) since this was the criterion
for group denition (Table 7.5). There was a signicant group dierence in BDI
and STAI scores with the aggressive group rating much higher in depression
(P0.05) state (P0.05) and trait anxiety (P0.01).
Cortical abnormalities in patients with TLE and IED
In order to detect possible subtle cortical brain pathology, that was not present on
visual assessment of the MRI scans, in a second approach we analysed the same
groups of patients using a voxel-by-voxel-comparison of cortical grey matter. After
automated segmentation of cerebral grey matter from T1-weighted MRI, the objec-
tive technique of statistical parametric mapping (SPM) was applied to study the
patient groups described above and comparison made with 35 healthy control sub-
jects (Woermann, 1999). Both TLE patient groups were compared with each other
and with the control subjects on a voxel-by-voxel basis for increases and decreases
of grey matter. SPM 96 characterizes signicant regional dierences in image
parameters, while allowing for global dierences to be taken into account. Details
of the methodology are published elsewhere (Woermann, 1999; Woermann et al.,
1999). The resulting signicant dierences through the 3D image space were dis-
90 L. Tebartz van Elst
played collapsed into three orthogonal planes (glass brain, see Figure 7.2). Regions
of signicant dierence were overlaid on normalized T1-weighted images to facil-
itate correlation with anatomy (see Figure 7.3).
In patients with TLE with IED compared as a group with healthy control sub-
jects, reductions of grey matter were found over large areas of the left extra-
temporal neocortex, maximal in the left frontal neocortex; one maximum
dierence projection had a Z score of 5.67 at Talairach coordinates x58, y36, z
9 mm(left anterior frontolateral cortex), the other a Zscore of 4.78 ina more pos-
terior left frontal lobe location (Talairach coordinates x66, y0, z28 mm,
Figures 7.2 and 7.3). Patients with TLE who did not have IEDshowed no signicant
decrease of cortical grey matter compared with control subjects. Patients with TLE
with IEDalso had reductionof left frontal grey matter, compared with patients with
TLE without IED, although less marked than when compared with control subjects
(Z score3.49 at Talairach coordinates x66, y2, z26 mm). The SPM-based
91 Aggression and epilepsy
Figure 7.1. Amygdala pathology in patients with TLE with and without IED. AGGaggressive.
voxel-wise correlation of SDAS scores and automatically segmented grey matter in
all patients with TLE showed the left frontal grey matter area as negatively corre-
lated with these scores which expressed social consequences of interictal aective
aggression (Z score 3.65 at Talairach coordinates x66, y2, z26 mm). Age,
scores of depression and anxiety, IQ measures, or scores of verbal uency did not
signicantly correlate with specic decreases in grey matter in all patients with TLE.
Dual brain pathology in patients with affective aggression and epilepsy
In our studies we have demonstrated that amygdala-related brain pathology could
be recognized in about half of the patients with TLE and IED. Contrary to our
hypothesis however, there was no evidence for increased mesial temporal lobe
sclerosis or amygdala sclerosis in the aggressive patient group. Brain pathology in
patients with epilepsy and aggression was more diverse in nature and more diuse
in distribution. Interestingly, we found an increased prevalence of a history of
encephalitis in patients with epilepsy plus aggression. Encephalitis in the past might
have been a pathogenic element for the more diuse and widespread pathology in
the temporal lobes seen in our aggressive patients. Furthermore the increased prev-
alence of left-handedness in our aggressive patients may indicate early brain
pathology like encephalitis aecting the left hemisphere (i.e. lateralization of dom-
inance to the right hemisphere).
92 L. Tebartz van Elst
Table 7.5. Neuropsychological and psychometric parameters
Variable TLE and IED TLE alone Group comparison
Mean FIQ (SD) 80.6 (8.5) 93.0 (13.9) ** (P0.000)
Mean VIQ (SD) 81.0 (8.6) 93.8 (14.1) ** (P0.000)
Mean PIQ (SD) 83.3 (11.8) 94.7 (15.1) ** (P0.005)
Mean BDI (SD) 8.8 (4.8) 4.2 (5.8) ** (P0.007)
Mean S-STAI 44.1 (14.9) 32.1 (10.9) ** (P0.005)
Mean T-STAI 50.1 (10.1) 34.7 (11.6) ** (P0.000)
Mean SDAS-9 (SD) 14.9 (7.5) 0.4 (0.7) ** (P0.000)
Mean SDAS-21 (SD) 30.9 (12) 3.1 (4.5) ** (P0.000)
Notes:
TLE, temporal lobe epilepsy; IED, intermittent explosive disorder; FIQ, full IQ; VIQ, verbal IQ;
PIQ, performance IQ; BDI, Beck Depression Inventory; STAI, State Trait Anxiety Inventory;
SDAS, Social Dysfunction and Aggression Scale; SD, standard deviation.
*P0.5, **P0.01 after Bonferroni correction.
Source: Tebartz van Elst et al. (2000a).
Eleven of twelve patients with amygdala-related brain pathology displayed this
pathology on the left-hand side and the only patient with right-sided amygdala
atrophy alone was left-handed. Thus the dominant hemisphere seems to play a
more important role in the mediation of aective aggression than the nondomi-
nant hemisphere.
The nding of frontal cortical grey matter loss was clearly lateralized too.
Patients with TLE plus IED displayed highly signicant left frontal grey matter loss
that correlated with the SDAS scores. This, together with the left lateralized nding
of amygdala-related brain pathology, could support a theory of dual brain pathol-
ogy in patients with IED (see Figure 7.4): pathology within the amygdala or amyg-
daloid circuits might result in hyperarousal states where patients become angry and
93 Aggression and epilepsy
TLE with IED controls
100
80
60
40
20
Figure 7.2. Glass brain view of decreased grey matter in 24 patients with TLE with episodes of
affective aggression compared with 35 control subjects; displayed after correction for
multiple comparisons (Woermann et al., 2000).
aroused without a sucient external stimulus (hyperarousal syndrome). This dys-
functional arousal resulting in aggressive behavioural impulses normally can be
suppressed by learned behavioural rules. However, associated with additional
frontal lobe pathology, the capacity of the aected patients to suppress behavioural
impulses arising from the emotional brain is limited and this leaves the patients
vulnerable to the development of hyperarousaldyscontrol syndromes i.e. episodic
dyscontrol or intermittent explosive disorder.
This theory is in line with earlier functional imaging and MR spectroscopy studies
showing a reduced prefrontal glucose metabolism in murderers and signicantly
lower neuronal markers in the frontal lobes of repetitively violent patients with learn-
ing disabilities, although without clear lateralizing eects (Raine et al., 1997, 1998).
From the fact that in our patient sample frontal as well as limbic brain pathol-
ogy was clearly lateralized to the left i.e. the dominant hemisphere, one might con-
clude that only dual brain pathology of the dominant hemisphere results in
hyperarousaldyscontrol syndromes while a crossover dual brain pathology with
94 L. Tebartz van Elst
Figure 7.3. Area of decreased grey matter (overlaid on normalized T1-weighted MRI) comparing 24
patients with TLE with episodes of affective aggression with 24 patients with TLE without
such episodes. At the location of maximum difference (Talairach coordinates: x66,
y2, z26 mm) the Z-score was 3.49 (Woermann et al., 2000).
pathology of the right frontal lobe or the right amygdala does not produce pheno-
menological states like this. However, this assumption is very speculative and
further investigations, possibly comparing patient groups with left- and right-sided
dual brain pathology may help clarify this question.
There is a controversy regarding the importance of hemispheric specialization
for aggressive behaviour (Bear, 1983; Nachson, 1991) with the majority of studies
pointing to a more important role of the left hemisphere (Saver et al., 1996). Our
MRI and neuropsychological ndings (i.e. particularly low verbal IQ) support this
assumption, which has been reported earlier (Herzberg and Fenwick, 1988).
It is important to note that in our sample there was a strong link between aggres-
sion and high levels of depression and anxiety, conrming other reports of such an
association in the nonpsychiatrically ill population (Bjork et al., 1997). It seems
plausible that high levels of anxiety result in states of hyperarousal that might be
facilitated by amygdala pathology as suggested by other authors (Cendes et al.,
1994). Regarding the relationship between depression and aggression there are only
few and nonconclusive reports (Braconnier and Jeanneau, 1997). Our ndings
point to a clear association between depression and IED in TLE.
Social and psychological aspects of aggression in epilepsy
Even though our studies concentrated on research into the neurobiological basis of
aggressive behaviour in patients with epilepsy, there is no question about the critical
95 Aggression and epilepsy
(Experience)
Evaluation
= hyperarousal = dyscontrol
Emotional
Hipp.
Sensory
input
Input
(Perception)
Output
(Behaviour)
Thalamus
Polymodal
cortex
Unimodal
cortex
Primary
cortex
Brainstem Amygdala
Frontal
cortex
Hypothalamus
Cognitive
Figure 7.4. Dual brain pathology in episodic dyscontrol a theory of intermittent explosive disorder
as hyperarousaldyscontrol syndrome. Hipp, hippocampus.
role of social and psychological intervening variables for the development of aggres-
sive behaviour in general, be it in the context of epilepsy or not. Social disadvantage,
prejudice, poor housing, poverty, poor communication skills, all are factors that
make hyperarousal states and states of discontentment and anger more likely and
thus might increase the probability of aggressive behaviour. On the other hand a
precise and correct diagnosis always is the rst step in managing dicult behaviour,
and sociological, psychological and neurobiological views of the same problem are
not necessarily contradictory.
For example, we were able to demonstrate a close link between episodic dyscon-
trol, reduced IQ, depression and anxiety. Even though disentangling the complex
interaction between these dierent psychobiological elements is very dicult, it
nevertheless may suggest a possible way of treatment, irrespective of which of these
elements is the most important one from an aetiological point of view.
Treatment of aggression in epilepsy
If aggression is a problem in the clinical management of patients with epilepsy the
most important point is to establish a correct diagnosis (Figure 7.5). A careful
neurological, psychiatric and medical history and examination should be per-
formed to answer the following questions: 1. Is there any medical condition that
contributes to the aggressive behaviour such as endocrinological or immunologi-
cal diseases? Is there any medication that might contribute to the aggressive behav-
iour? 2. What is the correct neurological diagnosis? Are there any other cerebral
problems in addition to the epilepsy? 3. Are there any psychiatric diagnoses which
possibly are independent of the epilepsy, like bipolar disease or antisocial person-
ality disorder? If the epilepsy started early in life it is in fact often impossible to
establish if, for example, a clinical picture that fulls the criteria for an antisocial
personality disorder has to be judged as independent of the organic brain disease
indicated by the epilepsy or alternatively is an organic personality disorder. 4.
Finally, a careful behavioural analysis and possibly video-telemetry should clarify
if the aggressive behaviour is ictal, postictal or interictal and whether it occurs in
the context of altered states of consciousness or psychosis.
Following syndromatic and possibly nosological diagnosis, treatment should be
causal if possible i.e. intervening medical problems like endocrinological disorders
should be treated adequately. Neurological syndromes like the epilepsy itself should
be treated eectively but, as with all drugs that inuence cerebral functioning, the
question as to whether anticonvulsants might contribute to the behavioural
problem should be considered. For example benzodiazepines are well known to
have paradoxical eects in a minority of patients and may cause states of arousal
and aggression (Binder, 1987; Daderman, 1999; Marcus, 1995; Sheth, 1994).
Likewise phenobarbital is well known to cause behavioural problems with aggres-
96 L. Tebartz van Elst
sion in a considerable subgroup of patients with epilepsy often with learning dis-
ability (File, 1990). Besides, in individuals any given drug might have dierent
eects than those described in large groups and thus a careful behavioural analysis
of the sequence of events is the only way to establish any possible side eect, for
example of antiepileptic drugs.
Care should be taken to establish signs of depression or anxiety, since a close link
between these psychopathological states and aective aggression in epilepsy has been
established. Both should be treated medically and with psychotherapy at the same
time (Goldstein, 1997; Lorenzen, 1973). Behavioural therapy in particular in patients
with epilepsy and learning disability has been proven to be very eective (Davis,
1984; Holzapfel, 1998; Rapport, 1983). In the medical treatment of depression in
patients with epilepsy, SSRIs or other new antidepressants like venlafaxine should be
preferred to the old tricyclic antidepressants (TCA) since the latter are more likely to
provoke seizures (Blumer, 1997; Lambert, 1999). In fact, an anticonvulsant eect of
SSRIs is well documented in animal models of epilepsy (Browning, 1997; Lu, 1998;
Pasini, 1996; Wada, 1995) and is also described in humans (Favale, 1995).
Following treatment of all medical, neurological and psychiatric conditions that
may or may not contribute to the aggressive psychopathology a symptomatic
97 Aggression and epilepsy
Treatment of medical, neurological and
psychiatric condition (look for depression and anxiety)
Possibly avoid
tricyclic antidepressants
Possibly avoid
convulsant antipsychotics
Consider paradox
side effects
Consider psychotherapy,
social support
Prophylactic
interventions:
Psychotherapy, social support,
lithium, SSRIs, antipsychotics
Management of
acute aggression:
Benzodiazepines,
antipsychotics
Neuropsychiatric
diagnosis
Causal
treatment
Symptomatic
treatment
Medical
history and exam
Neurological
history and exam
Psychiatric
history and exam
Behavioural analysis and diagnosis of
aggressive syndrome (ictal, postictal, interictal)
Figure 7.5. Therapeutic guidelines for the treatment of aggression in patients with epilepsy.
treatment of the aggression is mandatory. However, this depends on whether the
aggression is an ictal, postictal or interictal phenomenon.
Ictal aggression does not need symptomatic treatment but one might consider
interrupting a nonconvulsive status for example with benzodiazepines. Apart from
that, a patient who displays agitation and aggression during a seizure should not be
restricted since defensive violence is more common in such situations and the
aggressive behaviour is self-limited, as is the seizure.
The same is true for postictal confusional states, and even aggressive behaviour
in the context of postictal psychosis is self-limited. However, if the aggression is
severe and disturbing or self-harming, medical treatment with benodiazepines like
diazepam or clobazan and/or antipsychotics, for example risperidone, olanzapine
or quetiapine should be given. Good seizure control, avoiding severe complex
partial or secondary generalized seizures, is the best prophylactic intervention since
postictal confusional and psychotic states are more common after severe seizures.
In interictal aggression, prophylactic and acute symptomatic treatment of
aggression and agitation should be dierentiated. For the treatment of acute hyper-
arousaldyscontrol syndromes a combination of benzodiazepines like diazepam
and antipsychotics like haloperidol or sulpiride are still the most eective and safest
interventions. In cases of interictal psychosis however, the antipsychotic medica-
tion should eventually be switched to one of the atypical antipsychotic agents with
little proconvulsant potential like risperidone, olanzapine or quetiapine, since these
drugs are better tolerated. A good control of the psychosis is the best way to prevent
aggression if it is part of the psychosis.
In the case of interictal aggressive syndromes like IED there is no well-established
medical prophylactic therapy; however, there are many anecdotal reports of eec-
tive use of substances like lithium, valproate, carbamazepine, antipsychotics, beta-
blockers, clonidine and even psychostimulants (Fava, 1997; Grith, 1985;
Yudofsky, 1990). However, since there are hardly any well-conducted systematic
treatment studies at the moment the medical treatment still is very experimental
and single-case driven. Nevertheless, in the light of the very severe burden that is
put on patients and their relatives and caregivers by the sometimes devastating
behavioural episodes, a systematic trial of these agents seems to justied in special
cases. The use of cognitivebehaviour therapy for anger management should also
be considered, either as a rst-choice treatment, or combined with psychotropic
drugs.
Conclusion
Research into the neurobiological basis of aggression is still hampered by the di-
culty in dening phenomenological and nosological homogeneous study groups.
98 L. Tebartz van Elst
Nevertheless, it is important to develop a more precise understanding of the
complex interplay of social, psychological and neurobiological factors all contrib-
uting to the aggressive and violent behaviour. Aggression in epilepsy is rare.
However, if it occurs it imposes an immense burden on the patient, relatives and
caregivers. Clinically, it is crucial to rst of all establish a correct diagnosis.
Intervening medical, neurological and psychiatric disorders, in particular depres-
sion and anxiety, should be recognized and treated adequately. A correct syndro-
matic diagnosis of the aggressive syndrome and its relation to the seizures should
be made. Causal treatment should aim at any underlying medical, neurological or
psychiatric disorders. In symptomatic treatment of aggressive outbursts, treatment
of acute aggression and prophylactic treatment should be dierentiated. In acute
arousal states, if possible, patients should be left alone if aggression is a symptom
of a seizure since restrictive aggression is the most common form in this situation.
In cases of postictal or interictal aggression, it is important to establish a possible
relationship to postictal or interictal psychosis and treat adequately. For interictal
syndromes of aective aggression like IED (episodic dyscontrol), there are no gen-
erally established treatment protocols. However, tentative treatment with dierent
substances as mentioned above seems justied in single cases. Furthermore,
psychotherapeutic and social interventions may help prevent arousal states and
thus can be very eective.
REFERENCES
Aggleton, J.P. (1993). The contribution of the amygdala to normal and abnormal emotional
states. Trends Neurosci, 16, 32833.
Albert, D.J., Walsh, M.L. and Jonik, R.H. (1993). Aggression in humans: what is its biological
foundation? Neurosci Biobehav Rev 17, 40525.
Alheid, G.F., De Olmos, J.S. and Beltramino, C.A. (1995). Amygdala and extended amygdala. In
The Rat Nervous System, ed. G Paxinos, 495578. San Diego: Academic Press.
Amaral, D.G., Price, J.L., Pitknen, A. and Carmichael, S.T. (1992). Anatomical organisation of
the primate amygdaloid complex. In The Amygdala: Neurobiological Aspects of Emotion,
Memory, and Mental Dysfunction, ed. J.P. Aggleton, pp. 166. New York: Wiley-Liss.
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders
(Fourth Revision) (DSMIV). Washington, DC: APA.
Andy, O.J. and Jurko, M.F. (1972). Hyperresponsive syndrome. In Psychosurgery, ed. E.
Hitchcock, L. Laitinen and K. Vaernet, pp. 11726. Springeld, ILL: Charles C. Thomas.
Bach-Y-Rita, G., Lion, J.R., Climent, C.E. and Ervin, F.R. (1971). Episodic dyscontrol: a study of
130 violent patients. Am J Psychiatry, 127, 14738.
Barratt, E.S., Stanford, M.S., Kent, T.A. and Felthous, A. (1997). Neuropsychological and cogni-
tive psychophysiological substrates of impulsive aggression. Biol Psychiatry, 41, 104561.
99 Aggression and epilepsy
Bear, D. (1983). Hemispheric specialisation and the neurology of emotion. Arch Neurol, 40,
195202.
Behbehani, M.M. (1995). Functional characteristics of the midbrain periaquaeductal gray. Prog
Neurobiol, 46, 573605.
Bhatnagar, S. and Dallman, M. (1998). Neuroanatomical basis for facilitation of hypothalamic-
pituitary-adrenal responses to a novel stressor after chronic stress. Neuroscience, 84, 102539.
Binder, R.L. (1987). Three case reports of behavioral disinhibition with clonazepam. Gen Hosp
Psychiatry, 9, 1513.
Bjork, J.M., Dougherty, D.M. and Moeller, F.G. (1997). A positive correlation between self-
ratings of depression and laboratory-measured aggression. Psychiatry Res, 69, 338.
Blumer, D. (1997). Antidepressant and double antidepressant treatment for the aective disor-
der of epilepsy. J Clin Psychiatry, 58, 311.
Braconnier, A. and Jeanneau, A. (1997). Anxiety, aggression, agitation and depression: psycho-
pathologic aspects. Encephale, 23, 437.
Brandao, M.L., Cardoso, S.H., Melo, L.L., Motta, V. and Coimbra, N.C. (1994). Neural substrate
of defensive behavior in the midbrain tectum. Neurosci Biobehav Rev, 18, 33946.
Browning, R.A. (1997). Enhancement of the anticonvulsant eect of uoxetine following block-
ade of 5-HT1A receptors. Eur J Pharmacol, 336, 16.
Cendes, F., Andermann, F., Gloor, P. et al. (1993). MRI volumetric measurement of amygdala and
hippocampus in temporal lobe epilepsy. Neurology, 43, 71925.
Cendes, F., Andermann, F., Gloor, P. et al. (1994). Relationship between atrophy of the amygdala
and ictal fear in temporal lobe epilepsy. Brain, 117, 73946.
Chapman, W.P., Schroeder, H.R., Geyer, G. et al. (1954). Physiological evidence concerning
importance of amygdala nuclear region in the integration of circulatory functions and
emotion in man. Science, 120, 94954.
Charney, D.S. and Deutch, A. (1996). A functional neuroanatomy of anxiety and fear: implications
for the pathophysiology and treatment of anxiety disorders. Crit Rev Neurobiol, 10, 41946.
Currie, S., Heatheld, W., Henson, R. and Scott, D. (1971). Clinical course and prognosis of tem-
poral lobe epilepsy: a survey of 666 patients. Brain, 94, 17390.
Daderman, A.M. (1999). Flunitrazepam (Rohypnol) abuse in combination with alcohol causes
premeditated, grievous violence in male juvenile oenders. J Am Acad Psychiatry Law, 27,
8399.
Damasio, A.R., Tranel, D. and Damasio, H. (1990). Individuals with sociopathic behavior caused
by frontal damage fail to respond autonomically to social stimuli. Behav Brain Res, 41, 8194.
Davis, G.R. (1984). Cognitive-behavioral treatment of depressed aect among epileptics: prelim-
inary ndings. J Clin Psychology, 40, 9305.
Davis, M., Rainnie, D. and Cassell, M. (1994). Neurotransmission in the rat amygdala related to
fear and anxiety. Trends Neurosci, 17, 20814.
Delgado-Escueta, A.V., Mattson, R.H., King, L. et al. (1981). Special report. The nature of aggres-
sion during epileptic seizures. N Engl J Med, 305, 71116.
Devinsky, O. and Bear, D. (1984). Varieties of aggressive behavior in temporal lobe epilepsy. Am
J Psychiatry, 141, 6516.
100 L. Tebartz van Elst
Dicks, P., Myers, R.E. and Kling, A. (1969). Uncus and amygdala lesions: eects on social behav-
ior in the free ranging rhesus monkey. Science, 165, 6971.
Duncan, J.S., Bartlett, P. and Barker, G.J. (1996). Technique for measuring hippocampal T2 relax-
ation time. Am J Neuroradiol, 17, 180510.
Elliot, F.A. (1982). Neurological ndings in adult minimal brain dysfunction and the dyscontrol
syndrome. J Nerv Ment Dis, 170, 6807.
Elliott, F.A. (1984). The episodic dyscontrol syndrome and aggression. Neurol Clin, 2, 11325.
Elliot, F.A. (1992). Violence the neurologic contribution: an overview. Arch Neurol, 49,
595603.
European Rating Aggression Group (1992). Social dysfunction and aggression scale (SDAS21)
in generalised aggression and in aggressive attacks: a validity and reliability study. Int J Methods
Psychiatric Res, 2, 1529.
Falconer, M.A. (1973). Reversibility by temporal-lobe resection of the behavioral abnormalities
of temporal-lobe epilepsy. N Engl J Med, 289, 4515.
Fava, M. (1997). Psychopharmacologic treatment of pathologic aggression. Psychiatr Clin N Am,
20, 42751.
Favale, E. (1995). Anticonvulsant eect of uoxetine in humans [see comments]. Neurology, 45,
19267.
Fenwick, P.B.C. (1986). Aggression and epilepsy. In Epilepsy and Psychiatry, ed. M.R. Trimble and
T. Bolwig, pp. 3160. Chichester: John Wiley & Sons.
Fenwick, P. (1989). The nature and management of aggression in epilepsy. J Neuropsychiatry Clin
Neurosci, 1, 41825.
File, S.E. (1990). Changes in seizure threshold and aggression during chronic treatment with
three anticonvulsants and on drug withdrawal. Psychopharmacology, 100, 23742.
Gallagher, M. and Chiba, A.A. (1996). The amygdala and emotion. Curr Opin Neurobiol, 6,
2217.
Gastaut, H., Morrin, G. and Lesevre, N. (1955). Etudes du compartment des epileptiques psycho-
moteur dans Linterval de leurs crises. Ann Med Psychol, 113, 129.
Gerard, M.E. (1998). Subacute postictal aggression [see comments]. Neurology, 50, 3848.
Geschwind, N. (1975). The clinical setting of aggression in temporal lobe epilepsy. In Neural Basis
of Violence and Aggression, ed. W.S. Fields and W.H. Sweet, pp. 27381. St. Louis, MI: Warren
H. Green.
Gloor, P. (1991). Mesial temporal sclerosis: historical background and an overview from a
modern perspective. In Epilepsy Surgery, ed. H.O. Luders, pp. 689703. New York: Raven Press.
Gloor, P., Olivier, A., Quesney, L.F., Andermann, F. and Horowitz, S. (1982). The role of the
limbic system in experiential phenomena of temporal lobe epilepsy. Ann Neurol, 12, 12944.
Goldstein, L.H. (1997). Eectiveness of psychological interventions for people with poorly con-
trolled epilepsy. J Neurol Neurosurg Psychiatry, 63, 13742.
Goldstein, M. (1974). Brain research and violent behavior: a summary and evaluation of the
status of biomedical research on brain and aggressive violent behavior. Arch Neurol, 30, 135.
Grith, J.L. (1985). Treatment of episodic behavioral disorders with rapidly absorbed benzodi-
azepines. J Nerv Ment Dis, 173, 31215.
101 Aggression and epilepsy
Gunn, J. (1971). Epilepsy, automatism, and crime. Lancet, 1, 11736.
Gunn, J. (1977). Criminal behaviour and mental disorder. Br J Psychiatry, 130, 31729.
Halgren, E. (1992). Emotional neurophysiology of the amygdala within the context of human
cognition. In The Amygdala: Neurobiological Aspects of Emotion, Memory, and Mental
Dysfunction, ed. J.P. Aggleton, pp. 191228. New York: Wiley-Liss.
Herzberg, J.L. and Fenwick, P.B .(1988). The aetiology of aggression in temporal-lobe epilepsy.
Br J Psychiatry, 153, 505.
Holzapfel, S. (1998). Behavioral psychophysiological intervention in a mentally retarded epilep-
tic patient with brain lesion. Appl Psychophysiol Biofeedback, 23, 189202.
Hudson, L.P., Munoz, D.G., Miller, L., McLachlan, R.S., Girvin, J.P. and Blume, W.T. (1993).
Amygdala sclerosis in temporal lobe epilepsy. Ann Neurol, 33, 62231.
Jackson, G.D., Berkovic, S.F., Tress, B.M. et al. (1990). Hippocampal sclerosis can be reliably
detected by magnetic resonnance imaging. Neurology, 40, 186975.
Kalin, N.H. (1999). Primate models to understand human aggression. J Clin Psychiatry, 60
(Suppl. 15), 2932.
Kalviainen, R., Salmenpera, T., Partanen, K., Vainio, P., Riekkinen, P.S. and Pitkanen, A. (1997).
MRI volumetry and T2 relaxometry of the amygdala in newly diagnosed and chronic temporal
lobe epilepsy. Epilepsy Res, 28, 3950.
Kanemoto, K. (1996). Postictal psychosis: a comparison with acute interictal and chronic psycho-
ses. Epilepsia, 37, 5516.
Kanemoto, K. (1999). Violence and epilepsy: a close relation between violence and postictal
psychosis. Epilepsia, 40, 1079.
Kim, J.H., Tien, R.D., Felsberg, G.J. et al. (1995). Fast spin-echo MR in hippocampal sclerosis:
correlation with pathology and surgery. Am J Neuroradiol, 16, 62736.
Kligman, D. and Goldberg, D.A. (1975). Temporal lobe epilepsy and aggression. J Nerv Ment Dis,
160, 32441.
Kling, A.S. and Brothers, L.A. (1992). The amygdala and social behavior. In The Amygdala:
Neurobiological Aspects of Emotion, Memory, and Mental Dysfunction, ed. J.P. Aggleton, pp.
35377. New York: Wiley-Liss.
Krakowski, M. and Czobor, P. (1997). Violence in psychiatric patients: the role of psychosis,
frontal lobe impairment, and ward turmoil. Compr Psychiatry, 38, 2306.
LaBar, K.S., Ledoux, J.E., Spencer, D.D. and Phelps, E.A. (1995). Impaired fear conditioning fol-
lowing unilateral temporal lobectomy in humans. J Neurosci, 15, 684655.
Lambert, M.V. (1999). Depression in epilepsy: etiology, phenomenology, and treatment.
Epilepsia, 40 (Suppl. 10), S2147.
Lancman, M. (1999). Psychosis and peri-ictal confusional states. Neurology, 53, S338.
LeDoux, J.E. (1995). Emotions: clues from the brain. Annu Rev Psychol, 46, 20935.
Leicester, J. (1982). Temper tantrums, epilepsy and episodic dyscontrol. Br J Psychiatry, 141,
2626.
Lemieux, L., Wieshmann, U.C., Moran, N.F., Fish, D.R. and Shorvon, S.D. (1998). The detection
and signicance of subtle changes in mixed-signal brain lesions by serial MRI scan matching
and spatial normalisation. Med Image Analysis, 2, 22742.
102 L. Tebartz van Elst
Lishman, W.A. (1998). Organic Psychiatry The Psychological Consequences of Cerebral Disorder.
Oxford: Blackwell Science.
Logsdail, S.J. and Toone, B.K. (1988). Post-ictal psychoses. A clinical and phenomenological
description. Br J Psychiatry, 152, 24652.
Lorenzen, D. (1973). Initial results of behavior therapy in Psychiatric Regional Hospital
Weinsberg. [In German]. Nervenarzt, 44, 4237.
Lu, K.T. (1998). Endogenous serotonin inhibits epileptiform activity in rat hippocampal CA1
neurons via 5hydroxytryptamine 1A receptor activation. Neuroscience, 86, 72937.
Luo, B. (1998). Cholecystokinin B receptors in the periaqueductal gray potentiate defensive rage
behavior elicited from the medial hypothalamus of the cat. Brain Res, 796, 2737.
Mak, M. and De Konning, P. (1995). Clinical research in aggressive patients: pitfalls in study design
and measurements of aggression. Progr Neuropsychopharmacol Biol Psychiatry, 19, 9931017.
Maletzky, B.M. (1973). The episodic dyscontrol syndrome. Dis Nerv System, 34, 17885.
Marcus, A. (1995). Benzodiazepine administration induces exogenic psychosis: a case of child
abuse. Child Abuse Neglect, 19, 8336.
Margerison, J.H. and Corsellis, J. (1966). Epilepsy and the temporal lobe: a clinical, electroence-
phalographic and neuropathological study of the brain in epilepsy, with particular reference
to the temporal lobes. Brain, 89, 499530.
Miller, B.L., Darby, A., Benson, D.F., Cummings, J.L. and Miller, M.H. (1997). Aggressive, and
socially disruptive and antisocial behaviour associated with fronto-temporal dementia. Br J
Psychiatry, 170, 1505.
Miller, L.A., McLachlan, R.S., Bouwer, M.S., Hudson, L.P. and Munoz, D.G. (1994). Amygdala
sclerosis: preoperative indicators and outcome after temporal lobectomy. J Neurol, Neurosurg
Psychiatry, 57, 1099105.
Moran, N.F., Lemieux, L., Maudgil, D.D., Kitchen, N.D., Fish, D.R. and Shorvon, S.D. (1999).
Analysis of temporal lobe resections in MR images. Epilepsia, 40, 107784.
Moyer, K.E. (1987). Violence and Aggression A Physiological Perspective. New York: Paragon
House Publishers.
Mungas, D. (1983). An empirical analysis of specic syndromes of violent behavior. J Nerv Ment
Dis, 171, 35461.
Nachson, I. (1991). Neuropsychology of violent behavior: controversial issues and new develop-
ments in the study of hemisphere function. In Neuropsychology of Aggression, ed. J.S. Miller,
pp. 93116. Boston: Kluwer Academic Publishers.
Pasini, A. (1996). The anticonvulsant action of uoxetine in substantia nigra is dependent upon
endogenous serotonin. Brain Res, 724, 848.
Petty, R.G., Bonner, D., Mouratoglou, V. and Silverman, M. (1996). Acute frontal lobe syndrome
and dyscontrol associated with bilateral caudate nucleus infarctions. Br J Psychiatry, 168,
23740.
Plummer, D.L. (1992). DispImage, a display and analysis tool for medical images. Revues
Neuroradiol, 5, 48995.
Raine, A., Buchsbaum, M. and Lacasse, L. (1997). Brain abnormalities in murderers indicated by
positron emission tomography. Biol Psychiatry, 42, 495508.
103 Aggression and epilepsy
Raine, A., Stoddard, J., Bihrle, S. and Buchsbaum, M. (1998). Prefrontal glucose decits in mur-
derers lacking psychosocial deprivation. Neuropsychiatry, Neuropsychol Behav Neurol, 11, 17.
Ramamurthi, B. (1988). Stereotactic operation in behaviour disorders. Amygdalotomy and
hypothalamotomy. Acta Neurochirurgica Suppl, 44, 1527.
Rapport, M.D. (1983). Carbamazepine and behavior therapy for aggressive behavior. Treatment
of a mentally retarded, postencephalitic adolescent with seizure disorder. Behav Modication,
7, 25565.
Ratner, R.A. and Shapiro, D. (1979). The episodic dyscontrol syndrome and criminal responsibil-
ity. Bull Am Acad Psychiatry Law, 7, 42231.
Reis, D.J. (1969). Brain norepinephrine: evidence that neuronal release is essential for sham rage
behavior following brainstem transection in cat. Proc Nat Acad Sci USA, 64, 10812.
Rodin, E.A. (1973). Psychomotor epilepsy andaggressive behavior. Arch GenPsychiatry, 28, 21013.
Rolls, E.T. (1992). Neurophysiology and functions of the primate amygdala. In The Amygdala:
Neurobiological Aspects of Emotion, Memory, and Mental Dysfunction, ed. J.P. Aggleton, pp.
14365. New York: Wiley-Liss.
Saver, J.L., Salloway, S.P., Devinsky, O. and Bear, D.M. (1996). Neuropsychiatry of Aggression. In
Neuropsychiatry, ed. B.S. Fogel, R.B. Schier and S.M. Rao, pp. 52348. Baltimore: Williams &
Wilkins.
Shaikh, M.B. (1997). Serotonin 5-HT1A and 5-HT2/1C receptors in the midbrain periaqueduc-
tal gray dierentially modulate defensive rage behavior elicited from the medial hypothalamus
of the cat. Brain Res, 765, 198207.
Sheth, R.D. (1994). Aggression in children treated with clobazam for epilepsy. Clin Neuro-
pharmacol, 17, 3327.
Soininen, H.S., Partanen, K., Pitkanen, A. et al. (1994). Volumetric MRI analysis of the amygdala
and the hippocampus in subjects with age-associated memory impairment: correlation to
visual and verbal memory. Neurology, 44, 16608.
Stein, D.J., Hollander, E., Cohen, L. et al. (1993). Neuropsychiatric impairment in impulsive per-
sonality disorders. Psychiatry Res, 48, 25766.
Stone, J.L., McDaniel, K.D., Hughes, J.R. and Hermann, B.P. (1986). Episodic dyscontrol disor-
der and paroxysmal EEG abnormalities: successful treatment with carbamazepine. Biol
Psychiatry, 21, 20812.
Swartz, M.S., Swanson, J.W., Hiday, V.A., Borum, R., Wagner, H.R. and Burns, B.J. (1998).
Violence and severe mental illness: the eects of substance abuse and nonadherence to medi-
cation. Am J Psychiatry, 155, 22631.
Szabo, C.A. and Lancman, M.S.S. (1996). Postictal psychosis: a review. Neuropsychiatry,
Neuropsychol Behav Neurol, 9, 25864.
Tebartz Van Elst, L., Woermann, F., Lemieux, L. and Trimble, M.R. (1999). Amygdala enlarge-
ment in dysthymia A volumetric study of patients with temporal lobe epilepsy. Biol
Psychiatry, 46, 161423.
Tebartz Van Elst, L., Woermann, F., Lemieux, L., Thompson, P.J. and Trimble, M.R. (2000a).
Aective aggression in patients with temporal lobe epilepsy a quantitative magnetic reso-
nance study of the amygdala. Brain, 123, 23443.
104 L. Tebartz van Elst
Tebartz Van Elst, L., Woermann, F., Lemieux, L. and Trimble, M.R. (2000b). Increased amygdala
volumes in female and depressed patients with temporal lobe epilepsy. Neurosci Lett, 281,
103106.
Thomson, C. (1989a). Aective disorders. In The Instruments of Psychiatric Research, ed. C.
Thomson, pp. 3446. New York: John Wiley & Sons.
Thomson, C. (1989b). Anxiety. In The Instruments of Psychiatric Research, ed. C. Thomson, pp.
4754. New York: John Wiley & Sons.
Treiman, D.M. (1991). Psychobiology of ictal aggression. In Advances in Neurology, ed. D. Smith,
D. Treiman and M.R. Trimble, Vol. 55, pp. 34156. New York: Raven Press.
Trimble, M.R. (1991). The Psychoses of Epilepsy. New York: Raven Press.
Trimble, M.R. (1996) Biological Psychiatry. Chichester: John Wiley & Sons.
Trimble, M.R., Ring, H.A. and Schmitz, B. (1996). Neuropsychiatric aspects of epilepsy. In
Neuropsychiatry, ed. B.S. Fogel, R.B. Schier and S.M. Rao, pp. 771803. Baltimore: Williams
& Wilkins.
Valzelli, L. (1981). Psychobiology of Aggression and Violence. New York: Raven Press.
Van de Poll, N.E. and Van Goozen, S.H.M. (1992). Hypothalamic involvement in sexuality and
hostility: comparative psychological aspects. Progr Brain Res, 93, 34361.
Van Paesschen, W., Connelly, A., Johnson, C.L. and Duncan, J.S. (1996). The amygdala and
intractable temporal lobe epilepsy: a quantitative magnetic resonance imaging study [pub-
lished erratum: Neurology (1997) 48, 1751]. Neurology, 47, 102131.
Vitiello, B. and Sto, D.M. (1997). Subtypes of aggression and their relevance to child psychia-
try. J Am Acad Child Adolesc Psychiatry, 36, 30715.
Wada, Y. (1995). Prolonged but not acute uoxetine administration produces its inhibitory eect
on hippocampal seizures in rats. Psychopharmacology, 118, 3059.
Watson, C., Andermann, F., Gloor, P. et al. (1992). Anatomic basis of amygdaloid and hippocam-
pal volume measurement by magnetic resonance imaging. Neurology, 42, 174350.
Wieser, H.G. (1983). Electroclinical Features of the Psychomotor Seizures. Stuttgart: Fisher
Butterworths.
Woermann, F.G. (1999). Voxel-by-voxel comparison of automatically segmented cerebral gray
matter a rater-independent comparison of structural MRI in patients with epilepsy.
NeuroImage, 10, 37384.
Woermann, F.G., Barker, G.J., Birnie, K.D., Meencke, H.J. and Duncan, J.S. (1998). Regional
changes in hippocampal T2 relaxation and volume: a quantitative magnetic resonance
imaging study of hippocampal sclerosis. J Neurol Neurosurg Psychiatry, 65, 65664.
Woermann, F.G., Tebartz Van Elst, L., Free, S.L., Thompson, P.J., Trimble, M.R. and Duncan, J.S.
(1999). Reduction of frontal neocortical grey matter associated with aective aggression in
patients with temporal lobe epilepsy an objective voxel-by-voxel analysis of automatically
segmented MRI. Epilepsia, 40, 2056. (Abstract.)
Woermann, F., Tebartz Van Elst, L., Koepp, M.J. et al. (2000). Reduction of frontal neocortical
grey matter associated with aective aggression in patients with temporal lobe epilepsy. An
objective voxel-by-voxel analysis of automatically segmented MRI. J Neurol Neurosurg
Psychiatry, 68, 1629.
105 Aggression and epilepsy
Yudofsky, S.C. (1990). Pharmacologic management of aggression in the elderly. J Clin Psychiatry,
51 (Suppl.), 2228.
Zanchetti, A. (1968). Reex and brain stem inhibition of sham rage behaviour. Progr Brain Res,
22, 195205.
106 L. Tebartz van Elst
8
Epilepsy and suicide: a neuropsychiatric
analysis
Dietrich Blumer
University of Tennessee College of Medicine, Memphis, USA
Introduction
According to eight reports, death by suicide occurs in 5% of patients with epilepsy,
compared with 1.4% in the general population (Matthews and Barabas, 1981).
Based on four reports, the four- to vefold increase in suicides among patients with
epilepsy over the rate in the general population is magnied to approximately
25fold among patients with temporal lobe epilepsy (Barraclough, 1987). In a
Danish study covering 14 years (Henriksen et al., 1970), 164 of 2763 patients with
epilepsy died (an excess mortality rate of 273% compared to the number of deaths
expected in the general population). While epilepsy was the cause of death in 26%,
suicide was the cause of death in 20% (an excess mortality rate of 300%) at an
average age at death of 32 years. For studies of an association between epilepsy and
suicide to be statistically valid, the cohort must be standardized by age and sex,
according to Stenager and Stenager (1992); most studies have not met these stan-
dards. The same authors pointed out that the variety of forms of epilepsy necessi-
tates an examination of each syndrome singly for risk of suicide. Their second point
is very signicant, because the data available indicate that suicide is associated par-
ticularly with temporal lobe epilepsy.
Suicide and attempted suicide
Murphy (1994) points out that it is important to distinguish between suicide and
attempted suicide. Dierences between epileptic patients who attempt suicide and
those who complete the act are not established. In the general population, suicides
are three times as likely to be men as women, while attempted suicides are predom-
inantly women at a 2:1 ratio. Suicides are about equally frequent before and after
age 40, and the victims are most commonly suering from a major psychiatric
illness; attempted suicides are mostly under age 40 and are less likely to be suer-
ing from one of the psychiatric illnesses most commonly associated with suicide.
107
Attempted suicide is about 10 times more frequent than suicide. The lifetime risk
of completed suicide after a suicide attempt is estimated to be slightly above 10%.
Suicide is carried out by eective means, while attempted suicides tend to use what
is at hand (8090% of the attempts are by overdose).
Several reports document the high frequency of suicide attempts by overdose
among epileptic patients. The incidence of self-poisoning in epilepsy has been
reported as sevenfold that of the general population (Mackay, 1979). Sixty-ve
percent of suicide attempts were carried out by ingesting antiepileptic drugs, 60%
of which were barbiturates (Hawton et al., 1980). The psychotoxic role of the bar-
biturates in suicide attempts is emphasized by two further studies. In a population
of 126 children and adolescents admitted to an emergency room for suicide
attempt, nine had epilepsy a 15-fold increase compared with the prevalence of
epilepsy in the same age group. Of those nine epileptic children and adolescents,
eight had been treated with phenobarbital (Brent, 1986). A much higher preva-
lence of suicidal ideation (47% vs. 4%) was noted among patients treated with
phenobarbital compared with those treated with carbamazepine (Brent et al.,
1987).
The generally dicult psychosocial circumstances of patients with chronic epi-
lepsy have been considered the leading factor responsible for their elevated suicide
rate, more important even than the presence of psychiatric illness or the availabil-
ity of drugs (Editorial, 1980). However, in general, psychiatric illness has been iden-
tied as the nearly universal antecedent of suicide (Murphy, 1994). Mendez et al.
(1989) studied the causative factors for suicide attempts by overdose in epilepsy and
concluded that interictal psychopathological factors were of primary importance.
A comparison of suicide attempts among patients with epilepsy and comparably
handicapped controls with other chronic disabilities found that 30% of those with
epilepsy had attempted suicide as compared with 7% of the controls (Mendez et al.,
1986). Psychosocial circumstances cannot be considered a sucient precipitant for
suicide attempts or suicide.
In his review of the topic, Diehl (1986) ranked the risk factors for suicide in epi-
lepsy as follows: (1) psychiatric disorders (psychotic episodes, dysphoric episodes,
twilight states, personality disorders); (2) relatively young males (ages 2549 years);
(3) generalized and temporal lobe seizures (with brain lesions); (4) prolonged
duration of the seizure disorder and inadequate therapy; (5) personal, social or
occupational diculties; and (6) availability of large amounts of antiepileptic
drugs.
While a good number of papers list statistical data on the topic, there is a regret-
table paucity of neuropsychiatric case reports that allow for a better understanding
of the psychopathology and pathogenesis specic to suicide in epilepsy and that
might point at methods to prevent the fatal outcome.
108 D. Blumer
Series of suicides providing neuropsychiatric data
Taylor and Marsh (1977) reported on the occurrence of suicide among 193 patients
who had undergone temporal lobectomy and who were followed for 5 to 24 years.
Of 37 deaths, nine were by suicide (six poisoned themselves). Including an addi-
tional six patients who died in unclear circumstances would have raised the suicide
rate observed to 50-fold of that expected (Taylor, 1987). Five of the nine who de-
nitely had committed suicide had been rendered seizure-free by the surgery. The
mental state of the victims was not described. The paper is of exceptional interest
for the statistics of suicide in epilepsy but is of limited value for our understanding
of the neuropsychiatric problem.
Mendez and Doss (1992) reported on the psychiatric aspects of four patients
who died by suicide out of 1611 patients with epilepsy followed in a neurology
clinic over a period of 8 years: two male patients with chronic psychosis and good
seizure control; one male patient with brief psychotic episodes associated with con-
fusion and increased bitemporal spikes and diuse slowing on EEG in the absence
of seizures; and one female patient with profound ictal and postictal depression
who committed suicide after three witnessed staring spells. The patient with brief
psychotic episodes and one of the patients with chronic psychosis experienced
voices commanding them to commit suicide. All four patients had suered from
complex partial seizures since childhood. All four patients committed suicide by
medication overdose.
Of our entire population treated for epilepsy at the Epi-Care Center in Memphis
over the past 12 years (10739 patients), ve patients have committed suicide. All
had a history of early onset (mean age 9.5 years) of longstanding complex partial
seizures (mean duration 29 years), with very high (often daily) seizure frequency
in all but one. Suicide occurred in all patients after a short interval (mean 13
months) of having obtained full control of seizures for the rst time by temporal
lobectomy (three patients), medication (one patient), or vagus nerve stimulation
(one patient). Three patients had a previous history of suicidal moods or suicide
attempts, but in three of the ve, the suicidal act was precipitate and not anticipated
at the time. Four were male and one was female. None of the deaths was caused by
overdose: one by hanging, one by drowning and three by gunshot.
Role of the interictal and peri-ictal psychopathology in suicide
The premodern psychiatrists who established the basis for our modern classica-
tion of mental disorders noted that specic mental changes were associated with
epilepsy. They had the advantage of observing institutionalized patients with
chronic epilepsy over prolonged periods and were familiar with the characteristic
109 Epilepsy and suicide
intermittent and pleomorphic changes that have eluded the modern cross-sectional
psychological assessment of patients with epilepsy. Modern assessments are usually
carried out with methods that have been validated for use in patients who do not
have epilepsy and that are insensitive to the task. Premodern psychiatrists had
arrived at the concept of the dysphoric disorder as the most common psychiatric
disorder of epilepsy, a distinct disorder that has only recently been rediscovered
(Blumer et al., 1995). One of its key symptoms is associated with the episodic sui-
cidal moods of patients with chronic epilepsy, i.e. mesial temporal lobe epilepsy
(Blumer, 2000; Blumer et al., 1995; Gastaut, 1956).
Kraepelin (1923) precisely described the intermittent dysphoric disorder of
patients with epilepsy. Dysphoric episodes present with depressive moods (very
frequently with utter disgust of life and suicidal bent), irritability, anxiety, head-
aches, insomnia or at times with euphoric moods. These polysymptomatic dys-
phoric episodes occur without external triggers with rapid onset and termination
and recur fairly regularly in a uniform manner in the absence of clouding of con-
sciousness. Dysphoric symptoms can be observed as prodromes of an attack or in
the aftermath of an attack, but they most commonly appear as phenomena inde-
pendent of the seizures, with a frequency varying from every few days to every few
months. A patient just awakens one day dysphoric, or the dysphoria develops insid-
iously through the course of a day. As a rule, the dysphoric state lasts from 1 to 2
days but might dissipate after just a few hours. Based on our own observations, we
have added anergia and phobic fears to Kraepelins six key symptoms of the dys-
phoric disorder and have dened it by the presence of at least three of the eight key
symptoms, each present to a troublesome degree. We have noted an average of ve
key symptoms among our patients with dysphoric disorder (Blumer et al., 1995).
The risk of suicide in patients with epilepsy is primarily associated with the epi-
sodes of intense depressive mood that occur during the interictal phase of patients
with a dysphoric disorder, and suicide in epilepsy tends to occur in a precipitate
manner. As already noted by Kraepelin, the dysphoric symptoms also tend to occur
peri-ictally, during the prodrome or in the aftermath of a seizure. The postictal
phase in particular may be associated with marked depressive mood (Blumer,
1992). A high suicidal risk has been observed in patients who experience ictal
depressive mood that extends into the postictal phase for a period of 1 hour to 3
days. Williams (1956) reported 21 such cases among his 100 patients with ictal
emotional experience, and 5 of the 21 patients made suicide attempts during their
postictal phase.
As noted by Kraepelin, interictal psychoses tend to develop among patients
with interictal dysphoric disorder (Blumer et al., 2000; Kraepelin, 1923). The dys-
phoric disorder persists during the psychotic state, and intense depressive moods
may occur in the course of an interictal psychosis. The presence of the hallucina-
110 D. Blumer
tion of voices commanding patients to kill themselves represents a particular sui-
cidal risk.
The psychopathology of four patients (Mendez and Doss, 1992) who commit-
ted suicide is not reported in detail beyond the psychotic episodes in three of them
and the ictal and postictal depression in the fourth patient; the researchers state
only that two of those with psychosis also experienced depressive episodes. Four of
the ve patients from the Epi-Care series had longstanding dysphoric disorders.
The fth patient who had been found free of dysphoric symptoms earlier was not
examined during the 3-month interval from nally (after 20 years) becoming
seizure-free to his death by suicide; it was learned, however, that he had begun to
experience episodes of rage, symptomatic of a de novo dysphoric disorder.
Pathogenesis of the psychiatric disorders of epilepsy
Suicide in epilepsy appears to be caused by the interictal and sometimes by post-
ictal psychopathology. Since the early reports by Gibbs (1951), Landolt (1953), and
Hill (1953), evidence has been increasing that the interictal psychopathology tends
to emerge or to worsen upon improvement of the epilepsy as measured by seizure
frequency and EEG abnormalities. There is persuasive evidence that the psychiat-
ric disorders of epilepsy may result from the inhibitory activity that develops in
reaction to the excessive excitatory activity of the chronic seizure disorder, as pos-
tulated by Stevens (1975) and Engel (1989; Chapter 3). The precise nature of the
seizure-suppressing mechanisms is insuciently understood, and the phenome-
non is usually referred to as forced normalization (Landolt, 1955, 1958, 1963). The
evidence of a linkage of the psychiatric changes to inhibitory mechanisms can be
summarized as follows:
1. The development of interictal dysphoric and psychotic disorders is delayed fol-
lowing the onset of epilepsy (Gastaut et al., 1955; Slater and Beard, 1963) as
inhibitory mechanisms become increasingly established; this situation accords
with the particular linkage of the psychiatric disorders of epilepsy with its most
prominent chronic form mesial temporal lobe epilepsy.
2. Upon decrease and particularly upon full control of seizures, dysphoric symp-
toms and psychosis tend to be exacerbated or to emerge de novo.
3. Psychiatric changes emerge also at times when acute exacerbation of the seizure
activity engages an enhanced inhibitory response, commonly in the prodromal
and postictal phases, and rarely upon increased seizure activity as a seeming
opposite of forced normalization (Blumer et al., 2000).
4. There is a delayed phasing-out of the psychiatric changes after surgical elimina-
tion of the epileptogenic zone, presumably with only gradual fading of the inhib-
itory mechanism (Blumer et al., 1998).
111 Epilepsy and suicide
It is hypothesized that the dysphoric and psychotic symptoms, as well as the suici-
dality of epilepsy, are related, as are the seizures themselves, to the homeostatic ebb
and ow of excitatory and inhibitory inuences.
All patients of the two series of suicides providing neuropsychiatric details
(Mendez and Doss, 1992; Epi-Care patients noted above) had an onset of epilepsy
in early life, with a mean duration of the seizure disorder of 25 and 29 years, respec-
tively. This interval exceeds the mean interval from onset of epilepsy to the mani-
festation of interictal psychosis, reported as 14 years (Slater and Beard, 1963). Two
of the three patients who committed suicide in a psychotic state were under good
seizure control. The third patient showed the rare nding (Demers-Desrosiers et
al., 1978) of what seems the opposite of forced normalization: psychotic episodes
coinciding with the presence of increased electroencephalographic epileptiform
potentials in the absence of seizures, presumably resulting from the forceful engage-
ment of inhibitory mechanisms in response to enhanced subclinical seizure activ-
ity (Blumer et al., 2000). The increased suicide risk among patients with ictal and
persistent postictal depression (as in the fourth case of Mendez and Doss, 1992) was
previously documented by Williams (1956).
In our Epi-Care series, the striking nding of all ve suicides, which occurred
shortly after a longstanding seizure disorder had become controlled, points to
the risk of worsening (or de novo appearance) of a marked dysphoric disorder
once predominance of inhibitory mechanisms is established. It must be noted
again that ve of the nine patients fromthe British series who committed suicide
after temporal lobectomy had been rendered seizure-free (Taylor and Marsh,
1977).
Preventing suicide in epilepsy
Preventing suicide in epilepsy patients consists of eectively treating both the dys-
phoric disorder and the psychosis of the interictal phase (Blumer, 1997; Blumer et
al., 2000; Blumer and Zielinski, 1988). We now treat both the patients with suici-
dal dysphoric moods and those with interictal psychoses with double antidepress-
ant medication, enhanced if necessary with an atypical neuroleptic drug, e.g. with
the combination of 100150 mg imipramine, 2040 mg paroxetine and 24 mg ris-
peridone daily. The same treatment has been eective for patients with severe post-
ictal depressive mood, although we have not had the occasion to treat a patient with
ictal depression and suicidal intensity of the postictal phase. The dysphoric disor-
der is endogenous, and psychotherapy without pharmacotherapy leaves the patient
with suicidal moods at risk.
The bias against using antidepressants for the psychiatric disorders of epilepsy
on the grounds that they may lower the seizure threshold (McConnell and Duncan,
112 D. Blumer
1998) is erroneous on both empirical and theoretical grounds. As our experience
over some 15 years has shown, the modest amounts of antidepressant medication
required do not increase seizure frequency in patients with chronic epilepsy whose
interictal dysphoric disorder or psychosis indicates the presence of marked inhibi-
tion (Blumer, 1997; Blumer et al., 2000; Blumer and Zielinski, 1988). Gastaut et al.
(1955) have pointed out that patients with temporal lobe epilepsy (in contrast to
those with primary generalized epilepsy) show a higher interictal seizure threshold
than do individuals without epilepsy. The proconvulsant eect of antidepressants
does not provoke seizures but may serve to mitigate the psychotoxic eect of exces-
sive inhibition. Avoiding antidepressants for a patient with suicidal dysphoric
moods may have fatal consequences.
The 10739 patients with epilepsy seen at the Epi-Care Center for the past 12 years
were evaluated and treated by a comprehensive team consisting of a neurologist,
neurosurgeon, psychiatrist and neuropsychologist. The ve suicides among this
population represent a much lower fatality rate if compared with the population
reported by Mendez and Doss (1992). This decreased rate may result from the ready
availability of a psychiatrist for treatment of every patient with dysphoric disorder
or psychosis.
There are limitations to suicide prevention, even if one is alerted to the risk. Of
our patients who committed suicide, two were noncompliant with our treatment
and two, for geographic reasons, were followed elsewhere prior to their suicide. In
retrospect, our fth patient, who had become dysphoric after control of his seizures
by medication, should have been brought in for treatment upon the emergence of
an episode of rage that preceded his suicidal act. In our experience, timely psycho-
pharmacological treatment of the dysphoric disorder or of an interictal psychosis
can usually prevent a suicidal outcome of the epileptic disorder.
Conclusion
Further neuropsychiatric studies of patients with epilepsy who commit suicide
need to be made available before one can draw more denite conclusions about all
the modes of suicide in epilepsy. The current review allows a number of conclu-
sions, similar to those reached earlier by Diehl (1986), but more specic in nature.
Suicide in epilepsy results from the psychiatric disorder of temporal lobe epi-
lepsy; that is, from a severe dysphoric disorder, from interictal psychosis (associated
with preceding and concomitant dysphoric disorder, and at times with command
hallucinations), or from a severe postictal depressive state. These psychiatric disor-
ders develop gradually as seizure-suppressing mechanisms become established or,
at times, upon acute engagement of the inhibition. Suicide in epilepsy has increased
with our improved ability to suppress seizures.
113 Epilepsy and suicide
Patients with early onset of temporal lobe epilepsy and prolonged duration of the
illness (more than 20 years) are at particular risk of suicide once their seizures are
suppressed. Males in the age range of 3050 are more at risk. Treatment with barbit-
urates, availability of drugs, loss of loved ones or of jobs, and other dicult psycho-
social predicaments are not aetiological factors but may contribute to the suicide risk.
Suicide in epilepsy tends to occur precipitately during a t of melancholy (as
van Gogh described the depressive mood of his dysphoric episodes) and is often
not anticipated. However, there are usually warnings that precede a suicide. Upon
the occurrence of episodes of suicidal moods, prompt intervention is required with
psychotropic medication, chiey of the antidepressant type, and with careful
follow-up that includes adjusting the psychotropic medication as needed.
Although transient suicidal moods among epileptic patients were observed fre-
quently by premodern psychiatrists, completed suicide was not often reported
(Delay et al., 1957). Deaths from seizures have been markedly decreased by our
progress in seizure control but may be surpassed by now in numbers by deaths from
suicide. Our ability to suppress seizures must become paired with our ability to
treat the psychiatric consequences of improved seizure control.
Acknowledgement
The study of the patients seen at the Epi-Care Center was made possible by the col-
laboration of Drs Keith Davies, Bruce Hermann, Georgia Montouris and Allen
Wyler.
REFERENCES
Barraclough, B.M. (1987). The suicide rate of epilepsy. Acta Psychiatr Scand, 76, 33945.
Blumer, D. (1992). Postictal depression: signicance for the treatment of the neurobehavioral
disorder of epilepsy. J Epilepsy, 5, 21419.
Blumer, D. (1997). Antidepressant and double antidepressant treatment for the aective disor-
der of epilepsy. J Clin Psychiatry, 58, 311.
Blumer, D. (2000). Dysphoric disorders and paroxysmal aects: recognition and treatment of
epilepsy-related psychiatric disorders. Harvard Rev Psychiatry, 8, 817.
Blumer, D. and Zielinski, J. (1988). Pharmacologic treatment of psychiatric disorders associated
with epilepsy. J Epilepsy, 1, 13550.
Blumer, D., Montouris, G. and Hermann, B. (1995). Psychiatric morbidity in seizure patients on
a neurodiagnostic monitoring unit. J Neuropsychiatry Clin Neurosci, 7, 44556.
Blumer, D., Wakhlu, S., Davies, K. and Hermann, B. (1998). Psychiatric outcome of temporal
lobectomy for epilepsy: incidence and treatment of psychiatric complications. Epilepsia, 39,
47886.
114 D. Blumer
Blumer, D., Wakhlu, S., Montouris, G. and Wyler, A. (2000). Treatment of the interictal psy-
choses. J Clin Psychiatry, 61, 11022.
Brent, D.A. (1986). Overrepresentation of epileptics in a consecutive series of suicide attempts
seen at a childrens hospital, 19781983. J Am Acad Child Psychiatry, 25, 2426.
Brent, D.A., Crumrine, P.K., Varma, R.R., Allan, M. and Allman, C. (1987). Phenobarbital treat-
ment and major depressive disorder in children with epilepsy. Pediatrics, 80, 90917.
Delay, J., Deniker, P. and Barande, R. (1957). Le suicide des pileptiques. Encphale, 46, 40136.
Demers-Desrosiers, L.A., Nestoros, J.N. and Vaillancourt, P. (1978). Acute psychosis precipitated
by withdrawal of anticonvulsant medication. Am J Psychiatry, 135, 9812.
Diehl, L.W. (1986). Epilepsie und Suizid. Psychiat Neurol Med Psychol, 38, 62533.
Editorial (1980). Suicide and epilepsy. Br Med J, 281, 530.
Engel, J. (1989). Seizures and epilepsy. Philadelphia: Davis.
Gastaut, H. (1956). tat actuel des connaissances sur lanatomie pathologique des pilepsies. Acta
Neurol Psychiatr Belg, 56, 520.
Gastaut, H., Morin, G. and Lesvre, N. (1955). tude du comportement des pileptiques psycho-
moteurs dans lintervalle de leurs crises: les troubles de lactivit globale et de la sociabilit. Ann
Med Psychol, 113, 127.
Gibbs, F.A. (1951). Ictal and non-ictal psychiatric disorders in temporal lobe epilepsy. J Nerv
Ment Dis, 113, 5228.
Hawton, K., Fagg, J. and Marsack, P. (1980). Association between epilepsy and attempted suicide.
J Neurol Neurosurg Psychiatry, 43, 16870.
Henriksen, B., Juul-Jensen, P.P. and Lund, M. (1970). The mortality of epileptics. In Life
Assurance Medicine, ed. R.D.C. Brackenridge, pp. 13948. London: Pitman.
Hill, D. (1953). Psychiatric disorders of epilepsy. Med Press, 229, 4735.
Kraepelin, E. (1923). Psychiatrie, 8th edn. Leipzig: Barth.
Landolt, H. (1953). Some clinical electroencephalographical correlations in epileptic psychoses
(twilight states) [abstract]. Electronencephalogr Clin Neurophysiol, 5, 121.
Landolt, H. (1955). ber Verstimmungen, Dmmerzustnde und schizophrene Zustandsbilder
bei Epilepsie: Ergebnisse klinischer und electroenzephalographischer Untersuchungen.
Schweiz Arch Neurol Psychiatr, 76, 31321.
Landolt, H. (1958). Serial electroencephalographic investigations during psychotic episodes in
epileptic patients and during schizophrenic attacks. In Lectures on Epilepsy, ed. A.M. Lorentz
de Haas, pp. 91133. Amsterdam: Elsevier.
Landolt, H. (1963). Die Dmmer- und Verstimmungszustnde bei Epilepsie und ihre
Electroenzephalographie. Dtsch Z Nervenheilkunde, 185, 41130.
Mackay, A. (1979). Self-poisoning a complication of epilepsy. Br J Psychiatry, 134, 27782.
McConnell, H.W. and Duncan, D. (1998). Treatment of psychiatric comorbidity in epilepsy. In
Psychiatric Comorbidity in Epilepsy: Basic Mechanisms, Diagnosis, and Treatment, ed. H.W.
McConnell and P.J. Snyder, pp. 245361. Washington, DC: American Psychiatric Press.
Matthews, W.S. and Barabas, G. (1981). Suicide and epilepsy: a review of the literature.
Psychosomatics, 22, 51524.
Mendez, M.F. and Doss, R.C. (1992). Ictal and psychiatric aspects of suicide in epileptic patients.
Int J Psychiatry Med, 22, 2317.
115 Epilepsy and suicide
Mendez, M.F., Cummings, J.L. and Benson, D.F. (1986). Depression in epilepsy. Signicance and
phenomenology. Arch Neurol, 43, 76670.
Mendez, M.F., Lanska, D.J., Manon-Espaillat, R. and Burnstine, T.H. (1989). Causative factors
for suicide attempts by overdose in epileptics. Arch Neurol, 46, 10658.
Murphy, G.E. (1994). Suicide and attempted suicide. In The Medical Basis of Psychiatry, ed. G.
Winokur and P.J. Clayton, pp. 52944. Philadelphia: W.B. Saunders.
Slater, E. and Beard, A.W. (1963). The schizophrenia-like psychoses of epilepsy. Br J Psychiatry,
109, 95112.
Stenager, E.N. and Stenager, E. (1992). Suicide and patients with neurologic diseases.
Methodologic problems. Arch Neurol, 49, 1296303.
Stevens, J.R. (1975). Interictal clinical manifestations of complex partial seizures. In Advances in
Neurology, Vol. 11: Complex Partial Seizures and their Treatment, ed. J.K. Penry and D.D. Daly,
pp. 85112. New York: Raven.
Taylor, D.C. (1987). Psychiatric and social issues in measuring the input and outcome of epilepsy
surgery. In Surgical Treatment of the Epilepsies, ed. J. Engel Jr, pp. 485503. New York: Raven
Press.
Taylor, D.C. and Marsh, S.M. (1977). Implications of long-term follow-up studies in epilepsy:
with a note on the cause of death. In Epilepsy, the Eighth International Symposium, ed. J.K.
Penry, pp. 2735. New York: Raven Press.
Williams, D. (1956). The structure of emotions reected in epileptic experiences. Brain, 79,
2967.
116 D. Blumer
9
Postictal psychoses, revisited
Kousuke Kanemoto
Aichi Medical University, Aichi, Japan
Historical background
Except for the immediate eects of a seizure on mental function, such as complex
partial status epilepticus and postictal confusion, modern epileptic psychoses can be
categorized into three main types: chronic, acute interictal and postictal psychoses. In
1953, Landolt stressed a seesaw relationship between epileptic seizures and psychoses,
and proposed the concept of forced normalization. In 1963, Slater made a rather com-
prehensive report on chronic psychoses in patients with epilepsy (Slater and Beard,
1963). In contrast, it was as late as 1988 before the concept of postictal psychoses was
revived by Logsdail and Toone. This delay in conceptual formation is all the more
peculiar, when considering the very old root of the concept of postictal psychosis.
In 1860, a French psychiatrist, Farlet, classied epileptic psychoses into three
categories: transient peri-ictal, chronic and true epileptic psychosis (Farlet,
1860/1961). As there was a lack of strict distinctions between preictal, intraictal and
postictal events at that time, it is not easy to compare Farlets classication to those
of the present. While transient peri-ictal psychosis overlaps postictal confusion and
the meaning of chronic psychosis is evident, Farlets true epileptic psychosis has no
clear counterpart in modern classications. Farlet assigned extreme psychomotor
agitation as well as extraordinarily aggressive and self-destructive behaviour to his
unique classication. As I will discuss later, these are salient psychopathological
traits of postictal psychoses. Indeed, John Hughlings Jackson (1875), a pioneer of
modern epileptology, stressed that the true epileptic psychosis described by Farlet
often followed clusters of seizures. However, authors of subsequent psychiatric lit-
erature have confused Farlets true epileptic psychosis with other types of epileptic
psychosis and even postictal confusion. While the literature has misunderstood the
true nature of the epileptic psychosis of Farlet for more than 100 years (Kolb and
Brodie, 1982; Taylor, 1972), that in the domain of epileptology has gradually for-
gotten it. As a result, and with the advent of antiepileptic drugs, the alternative
psychosis of Landolt (1963) became very popular, obscuring postictal psychosis
completely (Mendez et al., 1993; Onuma et al., 1995).
117
This longstanding neglect, however, is being rapidly rectied, resulting in several
review articles (Lancman, 1999; Sachdev, 1998). Episodes of postictal psychosis can
no longer be ignored, as the intensive seizure monitoring under the diminution or
discontinuation of antiepileptic drugs currently practiced has dramatically
increased the chances for those who are involved in epileptic surgery to observe
postictal psychosis directly (Devinsky et al., 1995; Kanner et al., 1996). In this
chapter, we pay special attention to the natural history of postictal psychoses
outside of the seizure monitoring unit, in the belief that such a patient sampling
method would present in-vivo postictal psychoses, instead of the quasi-in-vitro
ones produced in the intensive care unit.
Following a description of two representative cases of postictal psychoses clini-
cal data concerning postictal psychosis are reviewed.
Case reports
CASE 1
A 28-year-old kimono shop manager had a 20-year history of paroxysmal fearful feelings of
being left alone. At age 11, complex partial seizures began to follow these moments of fear.
As his age advanced, his seizures increased in intensity as well as frequency, despite
maximum drug therapy. At age 26, the first manifest postictal mental derangement occurred,
after several bouts of complex partial seizures. One day after this cluster of seizures, he struck
his father, the owner of the kimono shop, who had asked whether he was all right, out of
uncontrollable rage. This peculiar dysphoric state lasted for a week, during which he was con-
tinuously prone to violent behaviour over minor matters. He reported that alien ideas had
invaded him and that opposing thoughts battled each other during this state. Afterwards, he
could recall perfectly the details of his violent behaviour, but could not understand why he
had acted in such a manner. Such postictal episodes and repeated violent behaviours tor-
mented those around him, and recurred every 2 months before admission. An MRI revealed
a marked asymmetry of the hippocampi (the left side was smaller than the right) with a lower
signal intensity from the left hippocampus on a reversed T2 condition. Although he was right-
handed, a 60 mg injection of Amytal into the right, but not the left, carotid artery caused the
patient to become aphasic. Ictal EEG recordings, including those with depth, plus subdural
electrodes, unanimously suggested that the left hippocampus was the origin of both the
simple and complex partial seizures. Subsequently, a left inferior lobectomy with a hippo-
campoamygdalotomy was performed. The resected tissue revealed Ammons horn sclerosis.
On awakening from anaesthesia, the patient showed a peculiar dysphoria, just as he had
during his postictal psychoses, which spontaneously disappeared within 3 days. However,
one month after the lobectomy, he became increasingly euphoric and elevated in mood. His
speech was loud, rapid and difficult to interrupt, while full of jokes, puns and other amusing
irreverence. He even exposed his genitalia in public during his elevated moods and courted
several copatients. Since he sang throughout the night, we needed strong sedatives to put
118 K. Kanemoto
him to sleep. This manic state, lasting for 4 weeks, gradually turned into a depressive one.
Two months after the operation, he felt quite upset and lost interest in all activities. He
became so agitated that he could not stand still for a moment, and walked around restlessly.
He complained of slowed thinking and difficulties in making decisions. After treatment with
tricyclic antidepressants, his mood improved steadily, but only gradually, over half a year. Six
months after the operation, he set out to do his previous work and his relations with col-
leagues improved dramatically. A year after the operation, his mood was stabilized without
the help of antidepressants and he was accepted once again as the manager of the kimono
shop. He has been completely seizure-free for 4 years postoperatively. Moreover, neither the
previous dysphoric episodes nor the mood disorders have recurred.
CASE 2
At age 3, a 38-year-old housewife suffered from an episode of prolonged febrile convulsions,
which lasted for more than 30 minutes and resulted in transient paralysis of the left upper
extremities. At age 12, the first episode of complex partial seizures occurred, in which she
unknowingly handed over an examination paper to a classmate who happened to be sitting
next to her. After that, every time during a seizure, she would unconsciously reiterate the
same phrase; Hes coming to collect my examination paper. What should I do?. At age 15,
paroxysmal feelings of a peculiar familiarity began to precede the complex partial seizures,
during which she felt that the atmosphere of her environment suddenly changed and it
seemed as if she had dissolved into the immediate surroundings. Her seizures remained
uncontrollable despite intensive medication. As the patients age advanced, she joined a
local religious sect as a devoted member and became increasingly eccentric. The first mani-
fest postictal psychosis occurred after bouts of complex partial seizures at age 29. After an
intervening 36-hour lucid interval, she would rapidly become more and more elevated in
mood, with loud rapid speech that was difficult to interrupt.
She would change subjects kaleidoscopically from one to another. She screamed to her
husband repeatedly, I love you, darling, and hugged and kissed him in public in a sensual
manner. Three days after the cluster of seizures, the euphoric state culminated in agitated
exaltation. She said, I am directly feeling all that is happening in every corner of the world
through the palpitating movement of my teeth. The circular movements of my teeth are syn-
chronized with the circular movements of the world. Through nerves in my teeth, I can sense
the future 2000 years from now. Because of extreme psychomotor agitation, a short stay
in the psychiatric ward as well as potent sedatives were required. This state disappeared
completely within 10 days. Throughout the episode, her orientation and memory remained
intact. After several episodes of such postictal psychotic states, she agreed to surgery. An
MRI revealed a marked asymmetry of the hippocampi (the left side was smaller than the
right) with a lower signal intensity from the left hippocampus on a reversed T2 condition.
Although she was right-handed, her dominant language side proved to be right. Ictal EEG
recordings unanimously suggested that the left hippocampus was the origin of both the
simple and complex partial seizures. During the course of intensive seizure monitoring,
postictal psychosis recurred once after a cluster of complex partial seizures. Subsequently,
119 Postictal psychoses, revisited
a left inferior lobectomy with a hippocampoamygdalotomy was performed and the resected
tissue revealed Ammons horn sclerosis. She recovered steadily without complications. A
year after the operation, the patient began to work as a manager of a Japanese restaurant.
She completely lost interest in the religious activity as if exorcised. She has been completely
seizure-free for 7 years postoperatively and no episodes of postictal psychosis have
recurred.
Clinical studies
In our work, we re-examined all of the outpatient cases from 1984 to 1999 at the
Kansai Regional Epilepsy Center who were known to have had epilepsy with psy-
chotic episodes (n177). Epilepsy and seizure classications were based on de-
nitions proposed by the International League against Epilepsy (Commission on
Classication and Terminology of the International League Against Epilepsy, 1981;
1989). In our study, psychosis was dened according to the following ICD10 cri-
teria: the presence of hallucinations, delusions, or a limited number of severe
abnormalities of behaviour, such as gross excitement and overactivity, and cata-
tonic behaviour (World Health Organization, 1992). However, we excluded
psychomotor retardation from the original denition. Ictal psychotic episodes
directly corresponding to ictal epileptiform discharge, such as nonconvulsive status
epilepticus, were also excluded from psychotic episodes. Postictal psychosis was
dened as one that occurred within 7 days after the last generalized tonic-clonic sei-
zures or clusters of complex partial seizures. We excluded those patients who exhib-
ited postictal psychosis only during or immediately after intensive seizure
monitoring. In this study, all episodes of psychosis, except for postictal psychosis,
were included in the category of interictal psychosis. Patients with continuous hal-
lucinations or delusions without remission were regarded as having chronic
psychosis. Patients with interictal psychosis but without chronic psychosis were
included with the acute interictal psychosis cases. We assessed the psychopatholog-
ical features of psychotic episodes according to a modied scale for assessing posi-
tive symptoms (SAPS; Andreasen, 1984), for which the test procedure details have
been described in a previous study (Kanemoto et al., 1996a). Statistical analyses
were made with Chi-square tests, with Yates modication for small numbers.
Incidence
Fifty-one (2%) out of 2905 patients with epilepsy treated at Kansai Regional
Epilepsy Center experienced postictal psychoses that were not articially induced
(Table 9.1). It is dicult to compare our data with previous studies, as they are
either multiple case reports (Lancman et al., 1994; Levin, 1952; Logsdail and Toone,
1988; Savard et al., 1991; Umbricht et al., 1995) or based on observations during
120 K. Kanemoto
the seizure monitoring in preparation for epilepsy surgery (Devinsky et al., 1995;
Kanner et al., 1996). However, our nding seems to have a certain reliability,
because the prevalence of interictal psychosis in patients with epilepsy in the
current study (5%) agreed well with that seen in other specialized epilepsy clinics
(49%) (Edeh and Toone, 1987; Mendez et al., 1993; Onuma et al., 1995). The prev-
alence of postictal psychosis in patients with temporal lobe epilepsy (11%) proved
to be far higher than that in the general epilepsy population. Kanner et al. (1996)
reported that the annual incidence of postictal psychiatric events, including post-
ictal psychosis, at their monitoring unit was 7.8%. Seven out of the 13 patients in
their series had their rst-ever postictal psychiatric event during the monitoring
study, therefore, at most only 4% experienced a truly spontaneous postictal psychi-
atric event. Considering that their study was limited to patients with symptomatic
localization-related epilepsy, this gure is approximate to ours.
Lucid interval, recurrence and duration of postictal psychosis
Postictal psychosis has long been confused with the clouded consciousness
observed following complex partial seizures. Recent studies, including ours, have
proven that postictal psychosis cannot be reduced to a mere extension of postictal
confusions. The primary argument in support of this is twofold: (1) there is pre-
served orientation and memory seen during the episodes, and (2) lucid intervals
occur between the end of the seizures and the start of postictal psychoses
(Kanemoto et al., 1996a; Levin, 1952; Logsdail and Toone, 1988; Savard et al.,
1991). It is very dicult to explain the delayed manifestation of positive symptoms
after a short period of normality as a psychic equivalent of Todds paralysis, since a
mere exhaustion of the nervous system would have steadily recovered without
reversion, just as in postictal confusion. In 18 out of 51 patients, we conrmed the
presence of a lucid interval, which lasted from 1 to 3 days in 15 out of the 18 (83%)
(Table 9.2). This agrees well with previous studies (Kanemoto et al., 1996b; Kanner
121 Postictal psychoses, revisited
Table 9.1. Incidence of epileptic psychosis
Patients with psychotic disorder 177 (6%)
Postictal psychosis* 51 (2%)
Acute interictal psychosis* 75 (3%)
Chronic psychosis* 57 (2%)
Patients without psychotic disorder 2728 (94%)
Total 2905 (100%)
Note:
* Six patients experienced both postictal and interictal psychoses.
et al., 1996; Logsdail and Toone, 1988; Savard et al., 1991). In comparison with
acute interictal psychosis, the duration of postictal psychosis is relatively short. In
more than half of the patients, psychotic episodes disappeared within a week, and
psychotic states lasting for more than one week but less than a month were seen in
another one third (Table 9.3). In a certain proportion of patients, postictal psycho-
sis has been shown to have a tendency to recur (Kanner et al., 1996; Lancman et al.,
1994). In our series, 49% of the patients with postictal psychosis experienced one
or more recurrences.
Patient background
In view of the common features that patients with interictal and postictal psycho-
sis share, such as a comparatively long latent period between epilepsy and psycho-
sis onset (longer than 10 years; Table 9.4) and a close association with temporal lobe
epilepsy, the prevailing view has been that interictal and postictal psychoses are
probably similar (Savard et al., 1991). However, we have found that postictal and
interictal psychoses dier in several fundamental demographic data. First, age at
epilepsy onset was signicantly younger in patients with interictal psychosis than
in those with postictal psychosis. Second, the latent period between psychosis and
epilepsy onset was still longer in postictal than interictal psychosis. Third, the pro-
portion of those patients with reduced intelligence quotient (IQ) was signicantly
122 K. Kanemoto
Table 9.2. Presence of a lucid interval in 18
of 51 patients observed
Less than 24 hours 1
1 day 4
2 days 6
3 days 5
47 days 2
Undetermined 33
Table 9.3. Duration of the postictal psychosis
1224 hours 4
17 days 16
1 week to 1 month 13
Longer than 1 month 3
Undetermined 15
higher in patients with interictal psychosis than in those with postictal psychosis.
These data agree well with the report of Umbricht et al. (1995), which conrmed
lower IQ and younger age at onset of epilepsy in interictal psychosis than in post-
ictal psychosis.
Association with temporal lobe epilepsy
Our data support the suggestion of a majority of previous studies that psychosis
in epilepsy might be preferentially associated with the temporal lobe (Bruens,
1971; Edeh and Toone, 1987; Gibbs, 1951; Perez and Trimble, 1980; Toone et al.,
1980). However, a direct comparison between postictal and interictal psychosis
revealed that the close link between psychotic episodes and temporal lobe epi-
lepsy was signicantly more remarkable in postictal than in interictal psychosis
(Table 9.5). Our data provided additional supportive evidence in this respect.
First, patients with postictal psychosis exhibited a temporal lobe pathology sig-
nicantly more often on an MRI than those with interictal psychosis. Second,
electroencephalographically, generalized spikes and waves were signicantly
more often recorded in interictal psychosis than in postictal psychosis patients
(Table 9.6). Furthermore, as for seizures, complex partial seizures were more
closely associated with postictal psychosis than interictal psychosis (Table 9.7).
These data support the view that the involvement of the limbic structure is almost
always indicated in cases of postictal psychosis and that postictal psychosis, in
contrast to the alternative psychosis of Landolt, almost never occurs in idiopathic
generalized epilepsy.
123 Postictal psychoses, revisited
Table 9.4. Patient background
Postictal psychosis Interictal psychosis
Demographic data
a
(n45) (SD) (n126) (SD)
M/F 29/16 78/48
Age (year) 41.4 (12.5) 32.2 (9.8)
Age at epilepsy onset (year) 16.0 (11.3) 11.2 (6.2)
Duration of epilepsy (year) 25.6 (14.1) 20.5 (10.8)
Age at psychosis onset (year) 34.5 (10.3) 24.8 (7.3)
Latent period 18.1 (10.7) 13.2 (8.5)
Low intelligence
b
4/45 39/126
Notes:
a
There was a highly signicant dierence between the groups (P0.005) in all items except
for sex.
b
Full Scale IQ70.
124 K. Kanemoto
Table 9.5. Epilepsy classification
Postictal Interictal No psychotic
psychosis psychosis episode
(n45
a
) (n126
a
) (n2728)
Idiopathic localization-related epilepsy 42
Symptomatic localization-related epilepsy
Temporal lobe epilepsy
b
39 74 422
Extra-temporal lobe epilepsy 10 35 857
Idiopathic generalized epilepsy 5 308
Symptomatic/cryptogenic generalized epilepsy 7 369
Others 2 11 730
Notes:
a
Six patients with both postictal and interictal psychoses were excluded.
b
Chi-square5.14, statistically signicant dierence (p0.05).
Table 9.6. Laterality and localization of the laboratory findings
Postictal psychosis Interictal psychosis
(n45) (n126)
EEG ndings
Temporal foci 33 78
Extra-temporal foci 8 15
Sidedness (L/R) 11/18 43/41
Diuse SWC 1 21
b
MRI localization
Temporal
a
16 25
b
Extra-temporal
a
4 14
Sidedness (L/R) 12/9 34/20
Notes:
a
Patients with both temporal and extra-temporal pathology were excluded.
b
Statistically signicant dierence (P0.05).
SWC, spike and wave complex.
In a previous study based on MRI results (Kanemoto et al., 1996b), we suggested
that medial temporal lesions with additional neocortical involvement were especially
closely linked with postictal psychosis. We have also stressed the close association of
postictal psychosis with psychic auras, such as dj vu and ictal fear (Kanemoto et al.,
1996). In the present study as well, we conrmed a statistically signicant predomi-
nance of dj vu (Table 9.8). Considering the results of a cortical stimulation study
conducted by Gloor et al. (1982), which demonstrated involvement of the lateral as
well as medial temporal structure in the genesis of dj vu, this predominance of dj
vu aura agrees with the nding obtained from our MRI study. Although Savard et al.
(1991) suggested that the high incidence of ictal fear in patients with postictal psychic
aura was proof of the similarity between postictal and interictal psychoses, the
current study contradicts this, demonstrating that such psychic auras were a salient
feature of patients with postictal but not interictal psychosis. Furthermore, several
authors have pointed out that bilateral interictal epileptiform discharges predisposed
patients to develop postictal psychosis (Savard et al., 1991; Umbricht et al., 1995).
125 Postictal psychoses, revisited
Table 9.7. Seizures in patients with postictal or interictal psychoses
Postictal psychosis (n45) Interictal psychosis (n126)
SPS 25 79
CPS 37 84
a
GTC 36 97
Minor GS 0 9
Notes:
a
Statistically signicant dierence (P0.05).
SPS, simple partial seizures; CPS, complex partial seizures; GTC, generalized tonic-clonic
seizures; minor GS, minor generalized seizures including absence, generalized myoclonic
seizure and generalized tonic seizure.
Table 9.8. Simple partial seizures in localization-related epilepsy
Postictal psychosis (n43) Interictal psychosis (n103)
Autonomic 16 40
Dj vu 10 10
a
Motor 4 15
Elementary visual 0 11
Note:
a
Statistically signicant dierence (P0.05).
Psychopathological features
The most striking dierences between interictal and postictal psychoses lay in the
domain of psychopathological phenomenology (Table 9.9). In a series presented by
Logsdail and Toone (1988), only one of 14 patients had primary delusions or thought
disorders (7%), whereas as many as nine exhibited a markedly abnormal mood
(64%). Our previous study, comparing the psychopathological features of postictal
psychoses with those of interictal psychoses, supported their data. The rst-rank
symptoms of Schneider, such as delusions of perception and voice commenting,
occurred signicantly less often in postictal psychoses than in acute interictal psy-
choses, whereas sexual indiscretions, religious delusions, and grandiosity, often in
the setting of an elevated mood, were observed in postictal psychosis ve times more
often than acute interictal psychosis. Illusions of familiarity, mental diplopia, and
feelings of impending death, which Jackson and Stewart (1899) described as hall-
marks of the dreamy state, occurred almost exclusively in postictal psychosis. The
frequent occurrence of grandiose delusions and religious delusions, in a setting of
markedly elevated moods and feelings of mystic fusion of the body with the universe,
characterizes the psychopathology of postictal psychosis. In the present extended
series of patients, we were able to amplify these ndings. In Table 9.7, psychopatho-
logical traits that were proved to show a highly signicant dierence are listed.
Violence and postictal psychosis
Violent behaviour elicited in the course of postictal psychosis deserves a special
comment. The argument against the view that epilepsy is closely related to a libera-
126 K. Kanemoto
Table 9.9. Psychopathological findings (Modified SAPS)
a
Postictal psychosis (n45) Interictal psychosis (n126)
Delusion of perception 0 37
Delusions of reference 3 113
Persecutory delusions 5 111
Verbal hallucinations 3 82
Visual hallucination 9 2
Grandiose delusions 12 1
Religious delusions 10 3
Pressure of speech 22 1
Illusion of familiarity 13 1
Mental diplopia 8 1
Notes:
a
Only features that demonstrated signicant dierence (P0.005) are listed.
SAPS, Scale for the Assessment of Positive Symptoms (Andreason, 1984).
tion of aggressive impulses has marked modern epileptology, with the result that epi-
leptologists have almost succeeded in dismissing this old view. However, in the course
of our investigation of postictal psychosis, the sporadic episodes of abrupt violent
behaviour that we observed impressed us greatly. In a previous study (Kanemoto et
al., 1999), we compared violent attacks during episodes of postictal psychosis, acute
interictal psychosis, and postictal confusion immediately following complex partial
seizures in patients with temporal lobe epilepsy (TLE), and conrmed that severe
violent confrontational behaviour towards surrounding people with impending
danger occurred only rarely during the postictal confusions as previous studies have
also pointed out (Ashford et al., 1980; Rodin, 1973; Treiman, 1991). In contrast,
patients proved to be quite prone to violent behaviour during episodes of postictal
psychosis. Recently, Gerard et al. (1998) reported six cases who were identied as
having subacute postictal aggression, which supports our postulation that stresses a
close link between postictal psychosis and violent behaviour.
Postoperative de novo manic depressive illness
Hill et al. (1957) were one of the rst to recognize that depression could occur after
a temporal lobectomy. In a series evaluated by Taylor (1972), ve patients commit-
ted suicide. In another follow-up study, Taylor and Marsh (1977) reported that the
mortality rate during the rst 2 years postoperatively was twice as high as that in
any subsequent 2-year period. Further, in a Danish series investigated by Jensen and
Larsen (1979), all suicide attempts occurred within the rst postoperative month.
A literature search failed to nd any descriptions of postoperative hypomanic or
manic states, except for our own recent report (Kanemoto et al., 1998). However,
we were able to conrm the presence of a substantial number of cases with postop-
erative transient manic or hypomanic states, and a close relationship between post-
operative mood disorder and preoperative history of postictal psychoses.
Considering the intrinsic interrelatedness of postictal psychosis with dramatic
aective changes (Kanemoto et al., 1996a; Logsdail and Toone, 1988; Savard et al.,
1991), this preponderance of postoperative mood changes among patients with
preoperative postictal psychosis was all the more instructive. As Trimble (1991) has
warned, in view of the high incidence of suicide during the rst few years after a
temporal lobectomy alone, the need for continuing psychiatric observations of
patients who receive an operation, especially when they have a history of postictal
psychosis prior to surgery, is apparent.
Treatment
Treatment for postictal psychosis should be directed at two dierent stages. First,
once an episode of postictal psychosis appears, a direct shortening or alleviation of
127 Postictal psychoses, revisited
postictal psychosis should be attempted. While Kanner et al. (1996) recommend
dopamine blockers, Lancman et al. (1994) advise the use of benzodiazepines and
sedation with chloral hydrate.
However, this dierence in opinion is more apparent than real. In a typical case,
postictal psychosis begins with an initial hypomanic state, which develops rapidly
into a psychotic state with marked psychomotor agitation within 1248 hours. If
we succeed in making patients sleep amply during the initial hypomanic state,
appearance of frank psychosis could be nipped in the bud. In this way, a certain
proportion of the postictal psychosis could be prevented at the stage of the hypo-
manic state, especially in the seizure monitoring unit, where trained psychiatrists
could recognize the initial signs without delay. Indeed, postictal psychosis is self-
limited and lasts at most only for several days in a majority of cases. However, some
patients tend to commit violent behaviours or even suicide during the episodes.
Therefore, we strongly recommend prompt arrest of this state. Second, in contrast
to the alternative psychosis of Landolt, eorts to reduce complex partial and gen-
eralized tonic-clonic seizures could prevent recurrence of postictal psychosis in a
substantial number of patients. Because the direction of the treatment is opposite
at times as a function of alternative and postictal psychosis, the dierential diagno-
sis between these states is all the more important.
Conclusion
In conclusion, a concept of postictal psychosis is certainly contributing to the elu-
cidation of several longstanding neurobehavioural problems in patients with epi-
lepsy that have caused much controversy. Considering the conspicuous aective
nature of postictal psychosis as well as the preponderance of postoperative mood
disorders in patients with a preoperative history of postictal psychosis, further
investigations of this state could reveal underlying neurochemical changes which
could serve as a model for bipolar mood disorders. However, there remains much
controversy on the pathophysiological mechanism underlying postictal psychosis
(Wieser et al., 1985; Wolf, 1991). Although So et al. (1990) conrmed increased
spiking activity in the medial temporal region with depth-EEG recording during
an episode of postictal psychosis, they were sceptical about the causal relationship
between this nding and occurrence of frank psychosis, since increased spiking is
commonly observed as a postictal phenomenon in patients without psychosis. In a
similar study using depth EEG, Mathern et al. (1995) failed to conrm increased
postictal spike activity. Possibly, postictal psychosis should be further subdivided.
While some cases, in which peculiar feelings such as mental diplopia or dj vu are
prominent with the underlying tone of dysphoria, might be attributed to subclin-
ical limbic status epilepticus such that Wieser et al. (1985) conrmed with a depth-
128 K. Kanemoto
EEG study, other cases in the setting of markedly elevated moods would not be
directly related to seizure activity but caused by some alterations in pathways of
neurotransmitters. In this regard, the serotoninergic mechanism should also be
considered in view of the marked aective nature of the postictal psychosis along
with dopaminergic hypersensitivity (Kanner et al., 1996; Logsdail and Toone, 1988;
Savard et al., 1991; So et al., 1990) as well as a GABA-mediated mechanism (Ring
et al., 1994; Szabo et al., 1996). Certainly, postictal psychosis deserves further atten-
tion.
REFERENCES
Andreasen, N.C. (1984). Scale for the Assessment of Positive Symptoms (SAPS). Iowa: The
University of Iowa.
Ashford, J.W., Schulz, S.C. and Walsh, G.O. (1980). Violent automatism in a partial complex
seizure. Arch Neurol, 37, 1202.
Bruens, J.H. (1971). Psychoses in epilepsy. Psychiatr Neurol Neurochir, 74, 17492.
Commission on Classication and Terminology of the International League Against Epilepsy
(1981). Proposal for revised clinical and electroencephalographic classication of epileptic sei-
zures. Epilepsia, 11, 10213.
Commission on Classication and Terminology of the International League Against Epilepsy
(1989). Proposal for revised classication of epilepsies and epileptic syndromes. Epilepsia, 30,
38999.
Devinsky, O., Abramson, H., Alper, K. et al. (1995). Postictal psychosis: a case control series of 20
patients and 150 controls. Epilepsy Res, 20, 24753.
Edeh, J. and Toone, B. (1987). Relationship between interictal psychopathology and the type of
epilepsy. Br J Psychiatry, 151, 95101.
Farlet, J. (1860/1961). De ltat des pileptiques. Arch Gn Md, 16, 66199; 17, 46191; 18,
42343.
Gerard, M.E., Spitz, M.C., Towbin, J.A. and Shantz, D. (1998). Subacute postictal aggression.
Neurology, 50, 3848.
Gibbs, F.A. (1951). Ictal and non-ictal psychiatric disorders in temporal lobe epilepsy. J Nerv
Ment Dis, 113, 5228.
Gloor, P., Olivier, A. and Quensney, L.F. (1982). The role of the limbic system in experiential phe-
nomena of temporal lobe epilepsy. Ann Neurol, 12, 12944.
Hill, D., Pond, D.W., Mitchell, W. and Falconer, M.A. (1957). Personality changes following tem-
poral lobectomy for epilepsy. J Ment Sci, 103, 1827.
Jackson, J.H. (1875). On temporary mental disorders after epileptic paroxysms. West Riding
Lunatic Asylum Med Rep, 5, 10529.
Jackson, J.H. and Stewart, P. (1899). Epileptic attack with a warning of a crude sensation of smell
and intellectual aura (dreamy state) in a patient who has symptoms pointing to gross organic
disease of right temporo-sphenoidal lobe. Brain, 22, 53449.
129 Postictal psychoses, revisited
Jensen, I. and Larsen, J.K. (1979). Mental aspects of temporal lobe epilepsy follow-up of 74
patients after resection of a temporal lobe. J Neurol Neurosurg Psychiatry, 42, 25665.
Kanemoto, K., Kawasaki, J. and Kawai, I. (1996a). Postictal psychoses: a comparison with acute
interictal and chronic psychoses. Epilepsia, 37, 5516.
Kanemoto, K., Takeuchi, J., Kawasaki, J. and Kawai, I. (1996b). Characteristics of temporal lobe
epilepsy with mesial temporal sclerosis, with special reference to psychotic episodes.
Neurology, 47, 1199203.
Kanemoto, K., Kawasaki, J. and Mori, E. (1998). Postictal psychosis as a risk factor for mood dis-
orders after temporal lobe surgery. J Neurol Neurosurg Psychiatry, 65, 5879.
Kanemoto, K., Kawasaki, J. and Mori, E. (1999). Violence and epilepsy: a close relation between
violence and postictal psychosis. Epilepsia, 40, 1079.
Kanner, A.M., Stagno, S., Kotagal, P. and Morris, H.H. (1996). Postictal psychiatric events during
prolonged video-electroencephalographic monitoring studies. Arch Neurol, 53, 25863.
Kolb, L.C. and Brodie, H.K.H. (1982). Modern Clinical Psychiatry. Philadelphia: W.B. Saunders.
Lancman, M.E. (1999). Psychosis and peri-ictal confusional states. Neurology, 53 (Suppl. 2),
S338
Lancman, M.E., Craven, W.J., Asconap, J.J. and Penry, J.K. (1994). Clinical management of
recurrent postictal psychosis. J Epilepsy, 7, 4751.
Landolt, H. (1953). Einige klinisch-elektroencephalographische Korrelationen bei epileptischen
Dmmerzustnden. Nervenarzt, 24, 479.
Landolt, H. (1963). Die dammer- und verstimmungszustande bei epilepsie und ihre
Elektroencephalographie. Dtsch Z Nervenheilkunde, 185, 41130.
Levin, S. (1952). Epileptic clouded states. A review of 52 cases. J Nerv Ment Dis, 116, 21525.
Logsdail, S.J. and Toone, B.K. (1988). Postictal psychoses. A clinical and phenomenological
description. Br J Psychiatry, 152, 24652.
Mathern, G.W., Pretorius, J.K., Babb, T.L. and Quinn, B. (1995). Unilateral hippocampal mossy
ber sprouting and bilateral asymmetric neuron loss with episodic postictal psychosis. J
Neurosurg, 82, 22833.
Mendez, M.F., Grau, R., Doss, R.C. and Taylor, J.L. (1993). Schizophrenia in epilepsy: seizure and
psychosis variables. Neurology, 43, 10737.
Onuma, T., Adachi, N., Ishida, S., Katou, M. and Uesugi, S. (1995). Prevalence and annual inci-
dence of psychoses in patients with epilepsy. Epilepsia, 36 (Suppl. 3), S218
Perez, M.M. and Trimble, M.R. (1980). Epileptic psychosis-diagnostic comparison with process
schizophrenia. Br J Psychiatry, 141, 25661.
Ring, H.A., Trimble, M.R., Costa, D.C., Moriarty, J., Verhoe, N.P. and Ell, P.J. (1994). Striatal
dopamine receptor binding in epileptic psychosis. Biol Psychiatry, 35, 37580.
Rodin, E.A. (1973). Psychomotor epilepsy and aggressive behavior. Arch Gen Psychiatry, 28,
21013.
Sachdev, P. (1998). Schizophrenia-like psychosis and epilepsy: the status of the association. Am J
Psychiatry, 155, 32536.
Savard, G., Andermann, F., Olivier, A. and Remillard, G.M. (1991). Postictal psychosis after
partial complex seizures: a multiple case study. Epilepsia, 32, 22531.
130 K. Kanemoto
So, N.K., Savard, G., Andermann, A., Olivier, A. and Quensney, L.F. (1990). Acute postictal
psychosis: a stereo EEG study. Epilepsia, 31, 18893.
Slater, E. and Beard, A.W. (1963). The schizophrenia-like psychoses of epilepsy. Br J Psychiatry,
109, 95150.
Szabo, C.A., Lancman, M.L. and Stagno, S. (1996). Postictal psychosis: a review. Neuropsychiatry,
Neuropsychol Behav Neurol, 4, 25864.
Taylor, D.C. (1972). Mental state and temporal lobe epilepsy. A correlative account of 100 patients
treated surgically. Epilepsia, 13, 72765.
Taylor, D.C and Marsh, S.M. (1977). Implication of long-term-follow-up in epilepsy: with a note
on the cause of death. In Epilepsy: The 8th International Symposium, ed. J.K. Penry, pp. 2734.
New York: Raven Press.
Toone, B.K., Garralda, M.E. and Ron, M.A. (1980). The psychoses of epilepsy and the functional
psychoses. Br J Psychiatry, 137, 2459.
Treiman, D.M. (1991). Psychobiology of ictal aggression. In Advances in Neurology, Vol. 55, ed.
D. Smith, D. Treiman and M.R. Trimble, pp. 34156. New York: Raven Press.
Trimble, M.R. (1991). The Psychoses of Epilepsy. New York: Raven Press.
Umbricht, D., Degreef, G., Barr, W.B., Lieberman, J.A., Pollack, S. and Schaul, N. (1995). Postictal
and chronic psychoses in patients with temporal lobe epilepsy. Am J Psychiatry, 152, 22431.
Wieser, H.G., Hailemariam, S. and Regard, M. (1985). Unilateral limbic epileptic status activity:
stereo-EEG, behavioral, and cognitive data. Epilepsia, 26, 1929.
Wolf, P. (1991). Acute behavioral symptomatology at disappearance of epileptiform EEG abnor-
mality: paradoxical or forced normalization. In Neurobehavioral Problems in Epilepsy, ed. D.
Smith, D. Treiman and M.R. Trimble, pp. 12742. New York: Raven Press.
World Health Organization (1992). The International Classication of Mental and Behavioural
Disorders: Clinical and Diagnostic Guidelines (Tenth Revision) (ICD10). Geneva: WHO.
131 Postictal psychoses, revisited
Part III
Cognitive aspects
10
Dementia and epilepsy
Stephen W. Brown
Plymouth Postgraduate Medical School, Plymouth, UK
Definitions and context
As Lishman (1998) has pointed out, the term dementia has two potential mean-
ings in medical practice: rst, it may refer to a group of specic diseases, and
second, it is used to describe a clinical syndrome that can have many causes.
Although diseases in the former group are characterized by irreversible decline in
function, the latter includes conditions in which decline can be arrested, or in some
cases reversed.
Both ICD10 (World Health Organization, 1992) and DSMIV (American
Psychiatric Association, 1994) oer detailed diagnostic criteria for the syndrome,
with DSMIV dening additional principles for diagnosing dierent varieties of
dementia. Both of these, in attempting to ensure uniformity of populations that
might be used for research purposes, adopt a cookbook style of approach.
In this chapter, the term is used in the second of the meanings described by
Lishman. Dementia is regarded as an acquired global impairment of intellect,
memory and personality, which is independent of any impairment of conscious-
ness. The symptoms of dementia are typically of long duration, usually progressive
and often irreversible, but none of these latter features are essential to the concept.
How the concept of epilepsy came to be linked historically with that of demen-
tia, how these concepts became uncoupled, and how new connections came to be
made between them, are components of a story that contains much of the history
of psychiatry and of neurology.
Historical aspects
Aretaeus, in the second century AD (quoted by Temkin, 1971) described people in
whom epilepsy had become chronic as being languid, spiritless, stupid, inhuman,
unsociable, . . . not disposed to hold intercourse and slow to learn, from torpidity
of the understanding and of the senses. There was therefore a view that a continu-
ing propensity to seizures led to deterioration of those faculties that we would
probably refer to as intellect and personality. According to Berrios (1995) the term
135
dementia acquired a medical connotation in the second half of the eighteenth
century. He quotes the French encyclopaedia of 1765, which lists as causes:
1. damage to the brain caused by excessive usage, congenital causes or old age, 2. failure of the
spirit, 3. small volume of the brain, 4. violent blows to the head causing brain damage, 5. incur-
able diseases such as epilepsy, or exposure to venoms . . . Dementia is dicult to cure as it is
related to damage of brain bres and nervous uids; it becomes incurable in cases of congenital
defect or old age . . . [otherwise] treatment must follow the cause. (Diderot and dAlambert 1765)
A later concept, which was to carry much inuence, was the theory of degeneracy,
especially as that propounded by Morel (1857). He suggested that although mental
disorders might have an external or environmental cause in the rst instance, the
persons biological state is then modied, so that the disorder becomes hereditary.
Following this, each generation displays an increasing degree of pathology until the
line becomes extinguished in idiocy. Although degeneration was exemplied by the
deterioration across generations, it could also take place within the persons life-
time. Morel expressed a particular interest in epilepsy, and was responsible for sug-
gesting that there could exist a masked form (epilepsie larve), in which the main
features were not seizures but insanity. Thus a link was formalized between epilepsy
and deterioration. By the beginning of the twentieth century many European
medical writers took this as received fact. Thus:
The mental state of epileptics, as is well known, frequently presents deterioration, . . . mischie-
vous restlessness and irritability in childhood may develop to vicious and even criminal tenden-
cies in adult life. Every grade of intellectual defect may be met with, down to actual imbecility.
(Gowers, 1885)
And
The most numerous class of epileptics show, after the lapse of years, a slowly progressive dimming
of the active perceptions of the mind, a loss of memory, a blunting of the aections, a permanent
mental obtuseness which increases and grows, until, if the patient lives long enough, there is a
more or less absolute annihilation of all the faculties. (Berkley, 1901)
Most authorities seemed to believe that decline was consequent upon seizures. While
Gowers acknowledged that in a minority of cases deterioration is the expression of
a cerebral imperfection of which the epilepsy is another manifestation, he also stated
that in the majority of cases the failure must be regarded as a consequence of the
disease. Henry Maudsley also considered that seizures had a direct eect:
The mind is slowly weakened by the storms of fury through which it passes, and they sink nally
into the apathy of dementia a state of mere oblivion, in which they cease to hope or care more.
(Maudsley, 1874)
However, during the twentieth century the view that epilepsy alone might cause
intellectual deterioration has by and large been discredited, although replaced by
136 S.W. Brown
observations regarding the eects of concomitant brain disease, recurrent head
injury due to seizures, and undesirable eects of antiepileptic treatment. It is easy
to dismiss nineteenth century writings as the consequence of prejudice-based,
evidence-free speculation, but the question really has to be considered as to
whether the patient population was dierent to that which we see today. Perhaps
aspects of the natural history of seizure disorders have changed with time. A
common cause for admission to an asylum in the late nineteenth century was
general paralysis of the insane. Neurosyphilis was very common then, but is now
rare. Seizures and dementia can both occur in this condition, which we now know
to be a late complication of syphilis, but the aetiology was not fully recognized until
the twentieth century. Indeed, speculation was rife about the possibility of psycho-
social causation (in some ways analogous to the debate in the 1960s and 1970s con-
cerning the cause of schizophrenia). Objectifying the cognitive consequences of
seizure disorders was a task that had to wait until the advent of psychometrics.
Origins of psychometry
Visitors and residents in nineteenth century London could attend an
Anthropometric Laboratory in South Kensington, and pay to have exact measure-
ments made of their height, weight, breathing power, strength of pull and squeeze,
colour sense and much else. This was the brainchild of Sir Francis Galton
(18221911), a gentleman-scholar, explorer, meteorologist and cousin of Charles
Darwin. Galton was interested in explaining the dierences between individuals.
He held views about the distribution of human abilities, and became convinced
that heredity was of prime importance in dening them. He was mentor to Karl
Pearson (18571936), whose name is well known to scientists for his signicant
contribution to the study of statistics, and who occupied the post of Galton
Professor of Eugenics at University College London after Galtons death. Both
Galtons and Pearsons views about the normal distribution of human abilities were
inuential in the subsequent design of intelligence quotient (IQ) tests. When Binet,
Wechsler and others were to develop standardized tests purporting to measure
intelligence, these tests were adjusted in order to provide a bell-shaped curve of dis-
tribution of results, and tests tended to be constructed to demonstrate previous
underlying assumptions rather than to discover new ones.
When the rst IQ tests were applied to people with epilepsy, there was sometimes
evidence of intellectual deterioration on retesting (Dawson and Conn, 1929)
although this was found to be part of a wider pattern of uctuation in testretest
performance in epilepsy that could occur in either direction (Fetterman and
Barnes, 1934; Patterson and Fonner, 1928; Sullivan and Gahagan, 1935). One study
found that improvement in seizure control was related to improvement in test per-
formance (Kugelmass et al., 1938) raising the possibility that observed uctuations
137 Dementia and epilepsy
in IQ could be related to seizure frequency. Some studies did suggest a slight decline
in mean IQ with increasing duration of seizure disorder, but no very clear picture
emerged (Brown and Reynolds, 1981). Such observations, although supercially
interesting, were of limited value in developing a sophisticated model of the rela-
tionship between seizure disorders and cognitive function. This was to require
developments from an initially parallel, but later converging discipline, that of
neuropsychology.
Origins of neuropsychology
The origins of neuropsychology lie in direct clinical observation. Two nineteenth
century landmarks in the understanding of structural and functional relationships
in the brain were the observations of Broca and Wernicke. In 1861 Paul Broca iden-
tied the third frontal convolution of the left hemisphere as an area that if damaged,
would result in a specic impairment of expressive language. In 1874 Karl Wernicke
described specic impairment of receptive language associated with damage to an
area in the left hemisphere extending from the rst temporal convolution into the
parietal lobe. The recognition of clinically denable psychological syndromes
related to discrete brain pathology continued to develop along with a greater
understanding of cerebral localization. Modern neuropsychological testing
involves specic observation of memory, language, verbal and nonverbal uency
together with visuo-spatial and motor abilities. These assessments complement
tests of general cognitive functioning. Neuropsychological assessment is of course
used as part of the work-up for epilepsy surgery, but has been an essential tool in
studying the interaction between cognitive impairment and seizure-related vari-
ables (Goldstein 1997; Klviainen et al., 1992; Piccirilli et al., 1994; Rugland, 1990).
An issue in interpretation
In the process of scoring IQ tests in children, the raw test scores are subjected to
adjustment, which takes into account the age of the subject. This then gives a result
that is a comparison against the general population of the same age. As children
grow older, the mean raw scores for various items will rise as a result of learning
and normal development, but the adjustment will ensure that the mean population
IQ remains the same, and that IQ scores of the population of the same age t the
expected bell-shaped curve. This could lead to a situation where an individual child
improves test performance over a period of time and therefore improves raw test
scores, but does not improve as much as the rest of the population overall, and
therefore the IQ is seen to fall. This is of course an example of slow learning com-
pared to the rest of the population, but it does not represent deterioration. The
interpretation of serial IQ scores in children therefore requires attention to the raw
score changes. Clinical and educational psychologists will routinely take this into
138 S.W. Brown
account as part of their everyday practice, but the matter is perhaps worth drawing
to the attention of other clinicians who may not be so familiar with the details of
test construction, scoring and interpretation.
Possible relationships
As Gowers implied, dementia and epilepsy may both be consequences of the same
underlying disorder, rather than one being a consequence of the other. Some spe-
cic clinical epilepsy syndromes are associated with acquired disorders of intellect
and memory, although it is not clear whether seizure activity is responsible for the
cognitive changes or whether, like the previous group, there is a shared aetiology. It
is however known that individual seizures may result in a cognitive penalty, and
that interictal epileptiform EEG discharges can sometimes disrupt cognitive func-
tioning. Some antiepileptic drugs can also play a part. These potential relationships
are considered in turn.
Neurological disease in which both dementia and seizures occur
It is well recognized that symptomatic seizures occur in the context of dementia
syndromes in older people (Forsgren et al., 1996). Breteler et al. (1995) found that
people aged 5075 with a diagnosis of epilepsy had an overall relative risk of 1.5 of
subsequently developing a dementia, which they described as a moderately
increased risk over the expected rate. People with Downs syndrome are at partic-
ular risk of developing early Alzheimers disease, and this is frequently associated
with seizures (Collacott, 1993; Lott and Lai, 1982). In people with mental retarda-
tion (learning disabilities) who do not have Downs syndrome, the incidence and
prevalence of dementia is higher than expected, and is often associated with poorly
controlled epilepsy (Cooper, 1997). There are also isolated case reports showing an
apparent association between seizures, dementia and a pathological lesion. It may
well be that there are many such very rare or even one-o disorders (Yerby et al.,
1986). One example of a rare condition in which the relationship between cerebral
lesions, seizures and intellectual deterioration is clear is Sneddon syndrome. This
is of presumed autoimmune origin and is characterized by cerebral infarction,
livedo reticularis, hypertension, epilepsy and progressive dementia (ORiordan et
at., 1995; Stephens, 1992).
One specic pathology that is associated with epilepsy and cognitive deteriora-
tion is the hypothalamic hamartoma (Kuzniecky et al., 1997). This can give rise to
gelastic epilepsy in which brief, frequent and mechanical laughing seizures begin in
infancy against a background of otherwise normal development. Other seizure
types, together with cognitive deterioration, start to appear between the ages of
4 and 10 years (Berkovic et al., 1988).
139 Dementia and epilepsy
Cognitive deterioration and worsening epilepsy also co-exist in Rasmussen syn-
drome (Rasmussen et al., 1958), a devastating progressive focal encephalitis that
has hitherto eluded aetiological explanation. Hart and her colleagues have recently
described some comorbidity with other cerebral pathologies (Hart et al., 1998).
Another hopeful lead has been the discovery by Rogers et al. (1994) of circulating
antibodies to the glutamate GLUR3 receptor in a patient with this condition who
showed some response to removal of the same antibodies by recurrent plasma
exchange.
There are also a number of specic genetic syndromes in which seizures and
dementia occur. Many show a Mendelian pattern, such as the autosomal dominant
hereditary prion disorders (Prusiner, 1994) or the progressive myoclonus of
Unverricht and Lundberg (Lehesjoki et al., 1992), which has a recessive mode of
inheritance. The number of these syndromes continues to expand, as does our
understanding of the pathological processes involved. Much more information will
become available as the map of the human genome continues to be elaborated.
Some specific epilepsy syndromes
Loss of previously acquired skills or abilities is a known adverse association with
certain epilepsy syndromes, most notably West syndrome, LennoxGastaut syn-
drome (LGS), the syndrome of continuous spike-wave discharges in slow wave
sleep (CSWS) and the LandauKlener syndrome (LKS). At rst sight, these con-
ditions all share the feature of frequent epileptiform EEG discharges, and it would
be tempting to suggest that such activity, by disrupting learning, interferes with
cognitive development. Although this may be partly true, it has been shown that
diuse electrical status can occur without apparent cognitive consequence
(Gokyigit and Caliskan, 1995). Besides, some authorities attribute at least part of
the deterioration in LGS to treatment eects, both pharmacological and psycho-
social (Genton and Dravet, 1997). LGS is also characterized in many people by fre-
quently disabling drop attacks (Oguni et al., 1996) which may have cumulative
adverse eects on cognitive function. Goldsmith et al. (2000) found that individu-
als with LGS with a normal cognitive outcome tended to have onset of LGS at a later
age than those who deteriorated.
EEG paroxysmal activity occurring during critical periods in development may
account for some of the specic cognitive and behavioural changes seen in LKS and
CSWS. Synaptic contacts that should have degenerated by apoptosis as part of
normal development may instead be strengthened. If this occurs in the temporop-
arietal cortex, this may lead to the acquired aphasia of LKS. If this is predominantly
frontal, higher cognitive and executive functions will be aected rst, leading to the
disintegrative psychotic-like presentation of CSWS (Roulet Perez et al., 1993;
Smith, 1997). This is consistent with the observation that the earlier the onset, the
140 S.W. Brown
worse the symptoms. It does not, however, explain the cause of the paroxysmal dis-
charges.
Cognitive effects of seizures
Seizures can cause disruption in cognitive functioning lasting for days after other-
wise apparent recovery (Dodrill, 1986), and a similar phenomenon lasting for
several hours has been described following other seizure types (Aldenkamp et al.,
1992). Related to this is the observation that seizures occurring during a nights
sleep may aect learning ability the next day (Aldenkamp, 1995).
Single complex partial or secondarily generalized seizures may be associated with
neuronal damage (Rabinowicz et al., 1996) and in humans, brain extracellular glu-
tamate builds up to potentially neurotoxic levels in partial seizures (During and
Spencer, 1993). Some other seizure-related variables that have been described as
adversely aecting cognitive function over time include: the presence of tonic-
clonic or complex partial seizures (Dodrill, 1986; Seidenberg et al., 1981), early
onset of seizures (OLeary et al., 1983, Strauss et al., 1995), the total number of sei-
zures experienced (Dodrill, 1986; Rodin et al., 1986), and repeated nonconvulsive
status epilepticus (Aldenkamp, 1997).
It has recently been reported that repetitive head injury in young adults is asso-
ciated with neocortical neurobrillary tangles (Geddes et al., 1999). Whether this
may be of clinical signicance in people with epilepsy who suer repeated head
trauma as a consequence of their seizures remains to be seen.
Critical periods in development may render cerebral functioning more suscep-
tible to interference (Holmes, 1997; Johnston, 1996). Not surprisingly, improve-
ment in seizure frequency is related to improvement in cognitive functioning
(Seidenberg et al., 1981). Jambaque et al. (2000) found that children with infantile
spasms due to tuberous sclerosis whose seizures became well controlled with vigab-
atrin showed signicant improvement in cognition and behaviour, in contrast to
the usual poor prognosis of this condition if seizures are not well controlled.
Cognitive effects of EEG discharges
Subclinical epileptiform EEG discharges can aect cognitive processes both gener-
ally and specically (Aarts et al., 1984). However, this may not be inevitable, and at
least one case has been reported in which diuse electrical status epilepticus
occurred without demonstrable learning problems (Gokyigit and Caliskan, 1995).
Nevertheless, other studies have suggested that EEG epileptiform discharges occur-
ring during IQ testing can aect test results (Siebelink et al., 1988) and such dis-
charges may also have a discreet eect on school performance (Kasteleijn-Nolst
Trenite et al., 1988). This phenomenon has been called transitory cognitive impair-
ment (TCI) and has been extensively studied. TCI can be demonstrated in about
141 Dementia and epilepsy
half the cases in which testing is accompanied by discharges (Binnie, 1993).
Although the presence of generalized bursts of 3-Hz spike-wave lasting three
seconds or more is most likely to cause TCI, it can be demonstrated in some cases
with discharges of shorter duration. Focal spikes can also produce TCI, those on
the left being mainly associated with verbal errors and those on the right with non-
verbal task impairment (Binnie and Marston, 1992).
Bailet and Turk (2000) found that children with epilepsy and abnormal inter-
ictal EEGs scored lower than children with epilepsy with normal interictal EEGs
on reading and spelling measures. They concluded that overall long-term risks of
learning problems exist among children with epilepsy compared with controls. This
is true even for those children with normal IQs and whose seizures are well con-
trolled. This observation lends some support to the view previously expressed by
other authors that in some cases TCI can result in deterioration in psychological and
social functioning. It has been suggested that specic antiepileptic drug therapy
aimed at reducing the interictal EEG discharges, and not just clinical seizures, may
have a benecial eect (Besag, 1995; Marston et al., 1993). The issue of state-depen-
dent cognitive deterioration is discussed in more detail by Besag, Chapter 6.
Treatment effects
The eects of antiepileptic treatment on cognitive function remain ill-understood
despite much research activity, and the subject remains to some extent controver-
sial. A well-reasoned account of how this topic should be tackled has recently been
given by Aldenkamp and van Bronswijk (1999). In a previous review by Vermeulen
and Aldenkamp (1995) the authors concluded that the assembled evidence so far
is hardly a basis for denitive statements. The following is a synopsis of published
reports of varying scientic rigour, which nevertheless probably represents the
current consensus. It serves to complement the more extensive review of
Aldenkamp, Chapter 17.
Phenobarbitone can signicantly impair learning ability (Calandre et al., 1990),
and impairs attention and memory at even low therapeutic doses (Riva and Devoti,
1996).
A dementia-like picture sometimes seen with phenytoin has been long recog-
nized (Rosen, 1968); this may be related to the serum level of the drug (Matthews
and Harley, 1975) or to phenytoin-induced folate deciency (Neubauer, 1970). In
the latter case, folate supplementation can result in improved cognitive perfor-
mance (Froscher et al., 1995). Phenytoin has also been implicated in thiamine de-
ciency, resulting in discrete performance decits in visuo-spatial analysis,
visuo-motor speed and verbal abstracting ability. These improve with thiamine
supplementation (Botez et al., 1993). Phenytoin has also been reported as causing
a specic memory impairment (Butlin et al., 1984; Gillham et al., 1990), most
142 S.W. Brown
marked with visual memory (Pulliainen and Jokelainen, 1994), and it can also have
adverse eects on motor and mental speed (Aldenkamp et al., 1994). Phenytoin
also diminishes the practice eect expected to be seen on retesting in some cases
(Sabers et al., 1995).
Some authors have reported carbamazepine to aect adversely memory
(Forsythe et al., 1991) and psychomotor speed (Gillham et al., 1990), while others
report both carbamazepine and valproate as having negligible eects (Prevey et al.,
1996; Stores et al., 1992). Recently, reversible pseudodementia with cortical
pseudoatrophy induced by valproate has been described (Guerrini et al., 1998;
Papazian et al., 1995; Straussberg et al., 1998).
There are relatively few published studies in this area that relate to the newer
antiepileptic drugs. The evidence so far, such as it is, suggests that with standard-
ized testing procedures, vigabatrin, lamotrigine and tiagabine have no deleterious
eect on cognitive functioning, and in some cases may enhance performance, pre-
sumably by preventing adverse consequences of seizures (Banks and Beran, 1991;
Dodrill et al., 1997; Klviainen et al., 1996; Meador and Baker, 1997; Monaco, 1996;
Provinciali et al., 1996; Sveinbjornsdottir et al., 1994).
Psychosocial effects
It has been observed in a number of studies that children with epilepsy as a group
underachieve academically compared to that expected from objective measures of
their cognitive ability (Rutter et al., 1970). There are many possible reasons for this,
whichwill include loss of time at school due toillness andhospital appointments, and
denial of opportunity to take part in normal activities. It has also been shown (Long
and Moore, 1979) that signicant people in the lives of children with epilepsy (such
as their parents) have lower expectations of the childrens achievements than they do
of children without epilepsy. This may also have an eect on cognitive outcome.
Some observations in temporal lobe epilepsy
The author has for some years reported observations on some people with tem-
poral lobe epilepsy who show evidence of acquired cognitive problems in the
absence of the factors listed above (Brown, 1989, 1992, 1999; Brown and
Abeyasinghe, 1984; Brown and Vaughan, 1988, 1990). It seems that such problems
may occur more often in males than females, where there is a long history of poorly
controlled epilepsy of temporal lobe origin. The neuropsychological prole in
females is that of frontal and left temporal impairment; males may show impair-
ment of more cortical areas, especially the parietal lobes. In all cases the epilepsy
syndrome is temporal lobe epilepsy, and the EEG epileptic foci are temporal, not
frontal. A full account is given in Brown (1999).
143 Dementia and epilepsy
A syndrome is described seen in people who develop temporal lobe epilepsy
before puberty, and who may come to the attention of psychologists or other
behavioural professionals after puberty because of behaviour problems. The di-
cult behaviour does not typically persist into adult life. Intellectual decline seems
to date from the late teenage years and is followed in most (but not all cases) by rel-
ative stability into adult life. It is not related to repeated minor head injuries caused
by seizures, although some people (mainly male) may have some dysarthria and
ataxia.
We identied two broad syndromes. A frontal subcortical picture was seen,
mainly in males with a prepubertal age of onset of epilepsy, where there is some dis-
turbance around or just after the time of puberty in which a stepwise decline of
intellectual functioning occurs. In addition, a cortical frontotemporal picture was
also seen, more commonly in females. These sometimes had a later age of onset of
epilepsy, and were more likely to acquire a psychiatric diagnosis. Other people with
localization-related epilepsy and discrete neuropsychological decits are also seen
with dierent histories and who show no intellectual deterioration.
Previous reports of metabolic changes in temporal lobe epilepsy have suggested
that frontal lobe functioning can also be aected. Ingvar (1984) comments:
. . . in temporal lobe epilepsy, the pathways from the deep temporal structures to the frontal lobes
do not function normally. The temporal gate to the frontal lobes may be closed. This may
explain symptoms of psychopathology in this common type of focal epilepsy.
More recently, Mackenzie and Miller (1994) examined senile plaques in temporal
lobectomy specimens and a control group. They found the age-related incidence of
senile plaques was signicantly greater in the temporal lobe epilepsy group, and
concluded that this suggested some aspects of TLE have a positive inuence on the
formation of senile plaques. The exact signicance of this nding is unclear.
Temporal Gate hypothesis
We have proposed a possible mechanism to account for the above observations,
which in deference to Ingvar (1984) we have called the Temporal Gate hypothesis
(Brown, 1999). This may be summarized as follows:
1. Certain neuropsychological functions of the frontal lobes come to maturity
during and after puberty, the process being inuenced by the neuroendocrine
axis.
2. This maturation of functioning is also dependent on frontal lobe input from an
intact temporo-limbic system.
3. If seizures of temporal lobe originoccur before andduring puberty, the temporo-
limbic input to the frontal lobes during this critical period of development may
be disrupted. Such disruption might lead to a period of behavioural or adjust-
144 S.W. Brown
ment problems during and after puberty. Also, because frontal lobe function
does not develop in the normal way, this might lead to the observed neuro-
psychological eects. These eects may reach a plateau, while cases may exhibit
further decline.
4. In our studies, the neuropsychological disruption so caused is sucient to
impair some previously acquired skills, such as reading.
Some indirect, or collateral, support for the hypothesis comes from a study recently
reported by Metz-Lutz et al. (1999) who investigated EEG and neuropsychological
outcomes in children with an initial diagnosis of benign partial epilepsy. Their nd-
ings suggested to them that maturing cognitive functions subserved by a cortical
area distant from the epileptic focus are nevertheless susceptible to interference by
epileptic activity, and this may aect cognitive outcomes.
In other neurological conditions associated with deterioration, the phenomenon
of oxidative stress (the production of oxygen radicals beyond a threshold for proper
antioxidant neutralization) has been implicated. These include Alzheimers disease
(Sims, 1996), Parkinsons disease (Jenner and Olanow, 1994), amyotrophic lateral
sclerosis (Gorman et al., 1994), Picks disease (Castellani et al., 1995) and schizo-
phrenia (Ramchand et al., 1996). Various intracellular messenger systems involv-
ing glutamate are implicated in oxidative radical production. These systems are
involved in neuronal growth, dierentiation and apoptosis (Michaelis, 1998).
Glutamate is also known to play an important role in epilepsy. The author has
observed substantial improvement in cognitive functioning in two patients (one
male subcortical, one female temporal) after using lamotrigine, a glutamate
release inhibitor (Meldrum, 1994). There are various other reports, generally of an
anecdotal nature, of the positive eects of lamotrigine on cognitive functioning
and scholastic ability (Buoni et al., 1998; Meador and Baker, 1997). One could spec-
ulate that oxidative stress plays a part in the maintenance of the acquired frontal
syndrome decits described above.
Conclusions
True intellectual deterioration is now known to be a relatively uncommon associ-
ation of epilepsy. Where it occurs it is usually associated with specic causes. These
include concomitant degenerative brain disease, some specic epilepsy syndromes
of uncertain aetiology, the direct eects of seizures (especially tonic-clonic sei-
zures), transitory cognitive impairment due to interictal EEG epileptiform dis-
charges, and the consequences of some antiepileptic treatments. In some cases it
is possible that a seizure disorder with temporal lobe focus, if present during a crit-
ical developmental period, may adversely inuence maturation and maintenance
of normal frontal lobe function, giving rise to a clinical picture of cognitive
145 Dementia and epilepsy
impairment or decline. The potential role of oxidative stress in this process
deserves further study.
Jokeit and Ebner (1999), as explored in Chapter 11, have described slow,
ongoing deterioration in some people with temporal lobe epilepsy, and state (with
an echo of Henry Maudsleys words from 125 years previously, about the mind
being weakened by the storms of fury through which it passes) It is assumed that
epilepsy-related noxious events and agents exhaust the compensatory capacity of
brain functions. Thus has our understanding come full circle, with still much left
to provide a good seam for investigators to mine.
REFERENCES
Aarts, J.H., Binnie, C.D., Smit, A.M. and Wilkins, A.J. (1984). Selective cognitive impairment
during focal and generalized epileptiform EEG activity. Brain, 107, 293308.
Aldenkamp, A.P. (1995). The impact of epilepsy on cognitive development and learning behav-
iour. In Epilepsy in Children and Adolescents, ed. A.P. Aldenkamp, W.O. Renier, F.E. Dreifuss
and T.P.B.M. Suurmeijer, pp. 22538. Boca Raton, FL: CRC Press.
Aldenkamp, A.P. (1997). Eect of seizures and epileptiform discharges on cognitive function.
Epilepsia, 38 (Suppl. 1), S525.
Aldenkamp, A.P. and van Bronswijk (1999). Cognitive side-eects as outcome measures in anti-
epileptic drug treatment: the current debate. In Epilepsy and Mental Retardation, ed. M.
Sillanpaa et al., pp. 135146. Peterseld: Wrightson Biomedical Publishing.
Aldenkamp, A.P., Gutter, T. and Beun, A.M. (1992). The eect of seizure activity and paroxysmal
electroencephalographic discharge on cognition. Acta Neurol Scand, 86, 11122.
Aldenkamp, A.P., Alpherts, W.C., Diepman, L., vant Slot, B., Overweg, J. and Vermeulin, J.
(1994). Cognitive side-eects of phenytoin compared with carbamazepine in patients with
localization-related epilepsy. Epilepsy Res, 19, 3743.
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders
(Fourth Edition) (DSMIV). Washington, DC: APA.
Bailet, L.L. and Turk, W.R. (2000). The impact of childhood epilepsy on neurocognitive and
behavioral performance: a prospective longitudinal study. Epilepsia, 41, 42631.
Banks, G.K. and Beran, R.G. (1991). Neuropsychological assessment in lamotrigine treated epi-
leptic patients. Clin and Exp Neurol, 28, 2307.
Berkley, H.J. (1901). A Treatise on Mental Disease. London: Henry Kimpton.
Berkovic, S.F., Andermann, F., Melanson, D., Ethier, R.E., Feindel, W. and Gloor, P. (1988).
Hypothalamic hamartomas and ictal laughter: evolution of a characteristic epileptic syndrome
and diagnostic value of magnetic resonance imaging. Ann Neurol, 23, 42939.
Berrios, G.E. (1995). Dementia. In A History of Clinical Psychiatry: The Origin and History of
Psychiatric Disorders, ed. G.E. Berrios and R. Porter, pp. 3451. London: The Athlone Press.
Besag, F.M. (1995). The therapeutic dilemma: treating subtle seizures or indulging in electroen-
cephalogram cosmetics? Semin Pediatr Neurol, 2, 2618.
146 S.W. Brown
Binnie, C.D. (1993). Signicance and management of transitory cognitive impairment due to
subclinical EEG discharges in children. Brain Develop, 15, 2330.
Binnie, C.D. and Marston, D. (1992). Cognitive correlates of interictal discharges. Epilepsia, 33
(Suppl. 6), S1117.
Botez, M.I., Botez, T., Ross-Chouinard, A. and Lalonde, R. (1993). Thiamine and folate treat-
ment of chronic epileptic patients: a controlled study with the Wechsler IQ scale. Epilepsy Res,
16, 15763.
Breteler, M.M., de Groot, R.R., van Romunde, L.K. and Hofman, A. (1995). Risk of dementia in
patients with Parkinsons disease, epilepsy, and severe head trauma: a register-based follow-up
study. Am J Epidemiol, 142, 13005.
Brown, S.W. (1989). Cognitive impairment in epilepsy. Educ Child Psychol, 6, 2532.
Brown, S.W. (1992). Organic brain syndromes in chronic epilepsy what is the truth about epi-
leptic dementia? Seizure, 1 (Suppl. A), S26/1.
Brown, S.W. (1999). Epilepsy dementia: intellectual deterioration as a consequence of epileptic
seizures. In Epilepsy and Mental Retardation, ed. M. Sillanpaa et al., pp. 11534. Peterseld:
Wrightson Biomedical Publishing.
Brown, S.W. and Abeyasinghe, R. (1984). Is there an epileptic dementia? Acta Neurol Scand Suppl,
70, 233.
Brown, S.W. and Reynolds, E.H. (1981). Cognitive impairment in epileptic patients. In Epilepsy
and Psychiatry, ed. E.H. Reynolds and M.R. Trimble. London: Churchill Livingstone.
Brown, S.W. and Vaughan, M. (1988). Dementia in epileptic patients. In Epilepsy, Behaviour and
Cognitive Function, ed. M.R. Trimble and E.H. Reynolds, pp. 17788. Chichester: John Wiley
& Sons.
Brown, S.W. and Vaughan, M. (1990). The nature of cognitive changes in people with chronic
epilepsy. Acta Neurol Scand Suppl, 82, 13.
Buoni, S., Grosso, S. and Fois, A. (1998). Lamotrigine treatment in childhood drug resistant epi-
lepsy. J Child Neurol, 13, 1637.
Butlin, A.T., Danta, G. and Cook, M.L. (1984). Anticonvulsant eects on the memory perfor-
mance of epileptics. Clin Exp Neurol, 20, 2735.
Calandre, E.P., Dominguez-Granados, R., Gomez-Rubio, M. and Molina-Font, J.A. (1990).
Cognitive eects of long-term treatment with phenobarbital and valproic acid in school chil-
dren. Acta Neurol Scand, 81, 5046.
Castellani, R., Smith, M.A., Richey, P.L., Kalaria, R., Gambetti, P. and Perry, G. (1995). Evidence
for oxidative stress in Pick disease and corticobasal degeneration. Brain Res, 696, 26871.
Collacott, R.A. (1993). Epilepsy, dementia and adaptive behaviour in Downs syndrome. J
Intellect Disabil Res, 37, 15360.
Cooper, S.A. (1997). High prevalence of dementia among people with learning disabilities not
attributable to Downs syndrome. Psychol Med, 27, 60916.
Dawson, S. and Conn, J.C.N. (1929). The intelligence of epileptic children. Arch Dis Child, 4,
14251.
Diderot & dAlambert (ed.) (1765). Encyclopdie ou Dictionnaire Raisonn des Sciences, des Arts
et des Mtieres, par une Societ de Gens de Lettres, Vol. 4, Paris: David Briasson; Le Breton:
Durand (quoted in Berrios, 1995).
147 Dementia and epilepsy
Dodrill, C.B. (1986). Correlates of generalized tonic-clonic seizures with intellectual, neuro-
psychological, emotional and social function in patients with epilepsy. Epilepsia, 27, 399411.
Dodrill, C.B., Arnett, J.L., Sommerville, K.W. and Shu, V. (1997). Cognitive and quality of life
eects of diering dosages of tiagabine in epilepsy. Neurology, 48, 102531.
During, M.J. and Spencer, D.D. (1993). Extracellular hippocampal glutamate and spontaneous
seizure in the conscious human brain. Lancet, 341, 160710.
Fetterman, J. and Barnes, R.R. (1934). Serial studies of the intelligence of patients with epilepsy.
Neuropsychologia, 7, 287300.
Forsgren, L., Bucht, G., Eriksson, S. and Bergmark, L. (1996). Incidence and clinical characteriza-
tionof unprovokedseizures inadults: a prospective population-basedstudy. Epilepsia, 37, 2249.
Forsythe, I., Butler, R., Berg, I. and McGuire, R. (1991). Cognitive impairment in new cases of
epilepsy randomly assigned to carbamazepine, phenytoin and sodium valproate. Dev Med
Child Neurol, 33, 52434.
Froscher, W., Maier, V., Laage, M. et al. (1995). Folate deciency, anticonvulsant drugs, and
psychiatric morbidity. Clin Neuropharmacol, 18, 16582.
Geddes, J.F., Vowles, G.H., Nicoll, J.A. and Revesz, T. (1999). Neuronal cytoskeletal changes are
an early consequence of repetitive head injury. Acta Neuropathol, 98, 1718.
Genton, P. and Dravet, C. (1997). LennoxGastaut syndrome and other childhood epileptic
encephalopathies. In Epilepsy: A Comprehensive Textbook, ed. J. Engel Jr and T.A. Pedley, pp.
235566. Philadelphia, PA: Lippincott-Raven.
Gillham, R.A., Williams, N., Wiedmann, K.D., Butler, E., Larkin, J.G. and Brodie, M.J. (1990).
Cognitive function in adult epileptic patients established on anticonvulsant monotherapy.
Epilepsy Res, 7, 21925.
Gokyigit, A. and Caliskan, A. (1995). Diuse spike-wave status of 9-year duration without behav-
ioral change or intellectual decline. Epilepsia, 36, 21013.
Goldsmith, I.L., Zupanc, M.L. and Buchhalter, J.R. (2000). Long-term seizure outcome in 74
patients with LennoxGastaut syndrome: eects of incorporating MRI head imaging in den-
ing the cryptogenic subgroup. Epilepsia, 41, 3959.
Goldstein, L.H. (1997). Neuropsychological assessment. In The Clinical Psychologists Handbook
of Epilepsy, ed. C. Cull and L.H. Goldstein, pp. 1834. London: Routledge.
Gorman, A.M., McGowan, A., ONeill, C. and Cotter, T. (1994). Oxidative stress and apoptosis
in neurodegeneration. J Neurol Sci, 139 (Suppl.), 4552.
Gowers, W.R. (1885). Epilepsy and Other Chronic Convulsive Diseases. New York: William Wood
and Company.
Guerrini, R., Belmonte, A., Canapicchi, R., Casalini, C. and Perucca, E. (1998). Reversible
pseudoatrophy of the brain and mental deterioration associated with valproate treatment.
Epilepsia, 39, 2732.
Hart, Y.M., Andermann, F., Robitaille, Y., Laxer, K.D., Rasmussen, T. and Davis, R. (1998).
Double pathology in Rasmussens syndrome: a window on the etiology? Neurology, 50, 7315.
Holmes, G.L. (1997). Epilepsy in the developing brain: lessons from the laboratory and clinic.
Epilepsia. 38, 1230.
Ingvar, D.H. (1984). Epilepsy related to cerebral blood ow and metabolism. Acta Psychiatr Scand
Suppl, 313, 216.
148 S.W. Brown
Jambaque, I., Chiron, C., Dumas, C., Mumford, J. and Dulac, O. (2000). Mental and behavioural
outcome of infantile epilepsy treated by vigabatrin in tuberous sclerosis patients. Epilepsy Res,
38, 15160.
Jenner, P. and Olanow, C.W. (1994). Oxidative stress and the pathogenesis of Parkinsons disease.
Neurology, 47 (Suppl. 3), S16170.
Johnston, M.V. (1996). Developmental aspects of epileptogenesis. Epilepsia, 37 (Suppl. 1),
S29.
Jokeit, H. and Ebner, A. (1999). Long term eects of refractory temporal lobe epilepsy on cogni-
tive abilities: a cross sectional study. J Neurol Neurosurg Psychiatry, 67, 4450.
Klviainen, R., Aikia, M., Helkala, EL., Mervaala, E. and Riekkinen, P.J. (1992). Memory and
attention in newly diagnosed epileptic seizure disorder. Seizure, 1, 25562.
Klviainen, R., Aikia, M., Mervaala, E., Saukkonen, A.M., Pitknen, A. and Riekkinen, P.J. Sr
(1996). Long-term cognitive and EEG eects of tiagabine in drug-resistant partial epilepsy.
Epilepsy Res, 25, 2917.
Kasteleijn-Nolst Trenite, D.G., Bakker, D.J., Binnie, C.D., Buerman, A. and Van Raaij, M. (1988).
Psychological eects of subclinical epileptiform EEG discharges. I. Scholastic skills. Epilepsy
Res, 2, 11116.
Kugelmass, I.N., Poull, L.E. and Rudnick, J. (1938). Mental growth of epileptic children. Am J Dis
Child, 55, 295303.
Kuzniecky, R., Guthrie, B., Mountz, J. et al. (1997). Intrinsic epileptogenesis of hypothalamic
hamartomas in gelastic epilepsy. Ann Neurol, 42, 607.
Lehesjoki, A.E., Koskiniemi, M., Pandolfo, M. et al. (1992). Linkage studies in progressive myo-
clonus epilepsy: UnverrichtLundborg and Laforas diseases. Neurology, 42, 154550.
Lishman, W.A. (1998). Organic Psychiatry: The Psychological Consequences of Cerebral Disorder,
Third Edition. Oxford: Blackwell Science.
Long, C.G. and Moore, J.R. (1979). Parental expectations for their epileptic children. J Child
Psychol Psychiatry, 24: 299312.
Lott, I.T. and Lai, F. (1982). Dementia in Downs syndrome: observations from a neurology clinic.
Appl Res Ment Retardation, 3, 2339.
Mackenzie, I.R. and Miller, L.A. (1994). Senile plaques in temporal lobe epilepsy. Acta
Neuropathol, 87, 50410.
Marston, D., Besag, F., Binnie, C.D. and Fowler, M. (1993). Eects of transitory cognitive impair-
ment on psychosocial functioning of children with epilepsy: a therapeutic trial. Dev Med Child
Neurol, 35, 57481.
Matthews, C.G. and Harley, J.P. (1975). Cognitive and motor-sensory performances in toxic and
nontoxic epileptic subjects. Neurology, 25, 1848.
Maudsley, H. (1874). Responsibility in Mental Disease. New York: Appleton.
Meador, K.J. and Baker, G.A. (1997). Behavioral and cognitive eects of lamotrigine. J Child
Neurol, 12 (Suppl. 1), S447.
Meldrum, B.S. (1994). The role of glutamate in epilepsy and other CNS disorders. Neurology, 44
(Suppl. 23).
Metz-Lutz, M.N., Kleitz, C., de Saint, M.A., Massa, R., Hirsch, E. and Marescaux, C. (1999).
Cognitive development in benign focal epilepsies of childhood. Dev Neurosci, 21, 18290.
149 Dementia and epilepsy
Michaelis, E.K. (1998). Molecular biology of glutamate receptors in the central nervous system
and their role in excitotoxicity, oxidative stress and aging. Progr Neurobiol, 54, 369415.
Monaco, F. (1996). Cognitive eects of vigabatrin: a review. Neurology, 47 (Suppl. 1), S611.
Morel, B.A. (1857). Trait des Degnrscences Physiques, Intellectuelles et Morales de lEspce
Humaine et des Causes qui Produisent ces Variets Maladive. Paris: Baillire.
Neubauer, C. (1970). Mental deterioration in epilepsy due to folate deciency. Br Med J, 2,
75961.
Oguni, H., Hayashi, K. and Osawa, M. (1996). Long-term prognosis of LennoxGastaut syn-
drome. Epilepsia, 37 (Suppl. 3), 447.
OLeary, D.S., Lovell, M.R., Sackellares, J.C. et al. (1983). Eects of age of onset of partial and
generalized seizures on neuropsychological performance in children. J Nerv Ment Dis, 171,
6249.
ORiordan, J.I., Javed, M., Murphy, R. and Hutchinson, M. (1995). Sneddons syndrome: clinical
course and outcome. Irish Med J, 88, 667.
Papazian, O., Canizales, E., Alfonso, I., Archila, R., Duchowny, M. and Aicardi, J. (1995).
Reversible dementia and apparent brain atrophy during valproate therapy. Ann Neurol, 38,
68791.
Patterson, H.A. and Fonner, D. (1928). Some observations on the intelligence quotient in epilep-
tics. Psychiatr Quart, 31, 5428.
Piccirilli, M., DAlessandro, P., Sciarma, T. et al. (1994). Attention problems in epilepsy: possible
signicance of the epileptogenic focus. Epilepsia, 35, 10916.
Prevey, M.L., Delaney, R.C., Cramer, J.A., Cattanach, L., Collins, J.F. and Mattson, R.H. (1996).
Eect of valproate on cognitive functioning. Comparison with carbamazepine. The
Department of Veterans Aairs Epilepsy Cooperative Study 264 Group. Arch Neurol, 53,
100816.
Provinciali, L., Bartolini, M., Mari, F., Del Pesce, M. and Ceravolo, M.G. (1996). Inuence of
vigabatrin on cognitive performances and behaviour in patients with drug-resistant epilepsy.
Acta Neurol Scand, 94, 1218.
Prusiner, S.B. (1994). Biology and genetics of prion diseases. Annu Rev Microbiol, 48, 65586.
Pulliainen, V. and Jokelainen, M. (1994). Eects of phenytoin and carbamazepine on cognitive
functions in newly diagnosed epileptic patients. Acta Neurol Scand, 89, 816.
Rabinowicz, A.L., Correale, J., Boutros, R.B., Couldwell, W.T., Henderson, C.W. and DeGiorgio,
C.M. (1996). Neuron-specic enolase is increased after single seizures during inpatient
video/EEG monitoring. Epilepsia, 37, 1225.
Ramchand, C.N., Davies, J.I., Tresman, R.L., Griths, I.C. and Peet, M. (1996). Reduced suscep-
tibility to oxidative damage of erythrocyte membranes from medicated schizophrenic patients.
Prostaglandins Leukot Essent Fatty Acids, 55, 2731.
Rasmussen, T., Olszewski, J. and Lloyd-Smith, D. (1958). Focal seizures due to chronic localized
encephalitis. Neurology, 8, 43545.
Riva, D. and Devoti, M. (1996). Discontinuation of phenobarbital in children: eects on neuro-
cognitive behavior. Pediatr Neurol, 14, 3640.
Rodin, E.A., Schmaltz, S. and Twitty, G. (1986). Intellectual functions of patients with childhood-
onset epilepsy. Dev Med Child Neurol, 28, 2533.
150 S.W. Brown
Rogers, S.W., Andrews, P.I., Gahring, L.C. et al. (1994) Autoantibodies to glutamate receptor
GluR3 in Rasmussens encephalitis. Science, 265, 64851.
Rosen, J.A. (1968). Dilantin dementia. Trans AmNeurol Assoc, 93, 273.
Roulet Perez, E., Davido, V., Despland, P.A. and Deonna, T. (1993). Mental and behavioural
deterioration of children with epilepsy and CSWS: acquired epileptic frontal syndrome. Dev
Med Child Neurol, 35, 66174.
Rugland, A.L. (1990). Neuropsychological assessment of cognitive functioning in children with
epilepsy. Epilepsia, 31 (Suppl. 4), S414.
Rutter, M., Graham, P. and Yule, W. (1970). A neuropsychiatric study in childhood. Clinics in
Developmental Medicine Nos. 35/36. London: Spastics International and Heineman Medical.
Sabers, A., Moller, A., Dam, M. et al. (1995). Cognitive function and anticonvulsant therapy:
eect of monotherapy in epilepsy. Acta Neurol Scand, 92, 1927.
Seidenberg, M., OLeary, D.S., Berent, S. and Boll, T. (1981). Changes in seizure frequency and
test-retest scores on the Wechsler Adult Intelligence Scale. Epilepsia, 22, 7583.
Siebelink, B.M., Bakker, D.J., Binnie, C.D. and Kasteleijn-Nolst Trenite, D.G. (1988).
Psychological eects of subclinical epileptiform EEG discharges in children. II. General intel-
ligence tests. Epilepsy Res, 2, 11721.
Sims, N.R. (1996). Energy metabolism, oxidative stress and neuronal degeneration in
Alzheimers disease. Neurodegeneration, 5, 43540.
Smith, M.C. (1997). LandauKlener syndrome and continuous spikes and waves during slow
sleep. In Epilepsy: A Comprehensive Textbook, ed. J. Engel Jr and T.A. Pedley, pp. 236777.
Philadelphia, PA: Lippincott-Raven.
Stephens, C.J. (1992). Sneddons syndrome. Clin Exp Rheumatol, 10, 48992.
Stores, G., Williams, P.L., Styles, E. and Zaiwalla, Z. (1992). Psychological eects of sodium val-
proate and carbamazepine in epilepsy. Arch Dis Child, 67, 13307.
Strauss, E., Loring, D., Chelune, G. et al. (1995) Predicting cognitive impairment in epilepsy:
ndings from the Bozeman epilepsy consortium. J Clin Exp Neuropsychol, 17, 90917.
Straussberg, R., Kivity, S., Weitz, R., Harel, L. and Gadoth, N. (1998). Reversible cortical atrophy
and cognitive decline induced by valproic acid. Eur J Paediatr Neurol, 2, 2138.
Sullivan, E.B. and Gahagan, L. (1935). On intelligence of epileptic children. Genet Soc Gen Psychol
Monogr, 17, 30975.
Sveinbjornsdottir, S., Sander, J.W., Patsalos, P.N., Upton, D., Thompson, P.J. and Duncan, J.S.
(1994). Neuropsychological eects of tiagabine, a potential new antiepileptic drug. Seizure, 3,
2935.
Temkin, O. (1971). The Falling Sickness, 2nd Edition revised. Baltimore: Johns Hopkins Press.
Vermeulen, J. and Aldenkamp, A.P. (1995). Cognitive side-eects of chronic antiepileptic drug
treatment: a review of 25 years of research. Epilepsy Res, 22, 6595.
World Health Organization (1992). The International Classication of Mental and Behavioural
Disorders: Diagnostic Criteria for Research (Tenth Revision) (ICD10). Geneva: WHO.
Yerby, M.S., Shaw, C.M. and Watson, J.M. (1986). Progressive dementia and epilepsy in a young
adult: unusual intraneuronal inclusions. Neurology, 36, 6871.
151 Dementia and epilepsy
11
The risk of cognitive decline in patients with
refractory temporal lobe epilepsy
Hennric Jokeit
1
and Alois Ebner
2
1
Swiss Epilepsy Centre, Zurich, Switzerland
2
Epilepsy Centre Bethel, Bielefeld, Germany
Introduction
Patients with refractory temporal lobe epilepsy (TLE) are at higher risk for mental
and cognitive impairments than healthy controls (Hermann et al., 1987, 1997;
Trimble, 1988). Typically patients with right-sided TLE are frequently impaired in
visuo-spatial retention tasks; patients with left-sided TLE may exhibit decits of
verbal memory. Because of these frequent and prominent memory decits it is
sometimes neglected that many TLE patients perform below healthy control sub-
jects on a variety of neuropsychological tests including intelligence measures
(Hermann et al., 1997). The probable reason is that the temporal epileptogenic
zone is not only malfunctioning but also adversely inuences remote cerebral
structures resulting in additional cognitive decits (Engel et al., 1991; Lders and
Awad, 1991). One of our recent studies conrmed that assumption (Jokeit et al.,
1997).
We investigated 96 TLE patients by FDG-PET and neuropsychological assess-
ment who had a corresponding unilateral temporal hypometabolism, left hemi-
sphere speech dominance, full scale IQ of 70 and no extra-temporal lesion on
MRIs. The regional glucose metabolism was determined in each patient in homol-
ogous regions including prefrontal cortex. A multivariate analysis of variance
revealed that the observed prefrontal metabolic disturbances, that are remote from
the temporal epileptogenic zone, were associated with impaired intellectual abil-
ities. Patients who demonstrated prefrontal metabolic disturbances performed
worse on verbal as well as performance IQ measures than patients without prefron-
tal metabolic disturbances. Although patients who demonstrated prefrontal meta-
bolic disturbances had an earlier epilepsy onset, the revealed association with
cognitive impairment was unrelated to the age at onset. Nevertheless, age at epi-
lepsy onset is a well-documented risk factor for cognitive impairment (Bourgeois
et al., 1983; Glosser et al., 1997). It is assumed that an early epilepsy onset aects
152
considerably the maturation of brain functions and structures as well as the acqui-
sition of complex knowledge and abilities.
One of the most frequent questions asked by epilepsy patients and their relatives
is whether seizures are destructive and contribute to progressive decline of intellec-
tual abilities. Generally dementia is a very rare syndrome in TLE patients and does
not represent the typical course of refractory TLE. However, a growing number of
clinical and experimental studies suggest a slow but ongoing progression of symp-
toms with increasing duration of refractory TLE or total number of lifetime sei-
zures. A long duration of intractable epilepsy is related to a considerable number
of focal or generalized seizures, pathological interictal electric brain activity,
chronic and transient metabolic disturbances (Arnold et al., 1996; Savic et al., 1997;
Theodore et al., 1989) and chronic antiepileptic medication usually with high
serum levels (Hermanns et al., 1996). It is suggested that these noxious factors may
induce secondary neurophysiological and structural long-term changes (Beach et
al., 1995; Ben-Ari and Represa, 1990; Bengzon et al., 1997; Jokeit et al., 1997; Marsh
et al., 1997; Multani et al., 1994; Tasch et al., 1999; Theodore and Gaillard 1999;
Theodore et al., 1999).
A few neuropsychological studies have been aimed at elucidating this question,
whether the cognitive abilities of patients deteriorate with an increasing duration
of intractable epilepsy. But neither short-term longitudinal nor cross-sectional
studies demonstrated a convincing relationship between psychometric intelligence
and the duration of epilepsy in samples of adult patients (Bourgeois et al., 1983;
Brown, 1996; Brown and Vaughan, 1988; Dodrill and Wilensky, 1992; Rodin et al.,
1986; Seidenberg et al., 1981; Selwa et al., 1994; Strauss et al., 1995; Trimble, 1988).
On one hand, the absence of an evident duration eect suggests that probably no
dramatic cognitive changes occur within periods of some years in adult TLE
patients. On the other hand, methodological restrictions, for example, a limited
time range of longitudinal studies which rarely exceeds a decade, an undetected
cohort bias in cross-sectional studies, or confounded variables might cover possible
duration eects. Additionally, in studies with small sample sizes, the possibility of
a Type II error is frequently neglected. The results of densiometric and volumetric
cross-sectional studies in TLE patients demonstrate that dierences in neuro-
psychological measures became signicant only if patients diered in decades of
the duration of epilepsy (Barr et al., 1997; Breier et al., 1997; Jokeit et al., 1999;
Mathern et al., 1995a, b; Multani et al., 1994; Salmenper et al., 1998). Moreover,
these studies indicate that neuronal injury within and beyond the temporal lobes
continues to occur with ongoing seizure activity in TLE patients. However, it is well
known that the brain possesses a large degree of redundancy and plasticity, and has
compensatory mechanisms that may prevent or postpone a cognitive decline due
to small but ongoing brain damage (Calne et al., 1986; Lewin, 1980). Therefore, the
153 Risk of cognitive decline in refractory TLE
individual brain reserve or spare capacity may also considerably inuence the
course of cognitive changes in patients with refractory TLE.
Cross-sectional studies
There are dierent approaches to infer cognitive changes along the time axis. From
a scientic as well as methodological point of view, prospective longitudinal studies
are generally superior to cross-sectional studies. Clinical observations and results
of longitudinal studies suggest that there is, if any, no rapid cognitive decline in
patients with refractory TLE. Consequently, longitudinal studies in TLE patients
are confronted with the problem of probably small eect sizes. Hence, to provide
statistical evidence, large samples of patients have to be observed over long periods
of time, probably exceeding decades. In contrast, cross-sectional studies allow the
recruitment of large sample sizes. However, the existence of a duration eect only
can be inferred from interindividual dierences in the duration of illness. Hence,
the interpretation of duration eects strictly presupposes that the patients were
recruited from the same population. Otherwise a cohort bias might considerably
aect the results. Although conclusions drawn from cross-sectional studies are
limited in some respects they may reveal trends that might be covered in longitu-
dinal studies restricted by time and sample size.
Also, dependent variables, usually intelligence measures, can be treated dieren-
tially as we demonstrate by the following studies that are based on independent
samples. In the rst study we considered dierences between an estimated measure
of former or premorbid intelligence and the current performance in an intelligence
test as a function of duration of epilepsy. This seems to be the only way to infer indi-
vidually a cognitive decline using psychometric instruments during a single neuro-
psychological investigation. In the second study we directly related the duration of
epilepsy with the current IQ test results. However, this approach is only appropri-
ate for sample studies.
Neuropsychological investigations of a patient frequently include so-called intel-
ligence trace tests to estimate the former intelligence in order to compare it with
current test results. Several studies have shown that most patients with cerebral
lesions or early dementia are unimpaired in intelligence trace tests, whereas the
same patients do worse in standard IQ tests. The dierence between the intelligence
trace test and standard IQ tests is considered to be a measure of cognitive decline
(Figure 11.1). If intellectual abilities deteriorate with increasing duration of epi-
lepsy the dierence between both measures should become larger. In our study
(Jokeit et al., 2000) we used the passive vocabulary-intelligence test (MWTB) as
an intelligence trace test (Lehrl, 1995). In this test, patients have to identify a real
word among four pseudo-words in rows of increasing diculty.
154 H. Jokeit and A. Ebner
Patients with a long duration of epilepsy more frequently demonstrated a con-
siderable dierence between both IQs compared with patients with a shorter dura-
tion of TLE.
In a larger study (Jokeit and Ebner, 1999), we examined the eects of duration
of epilepsy on the performance in a standard IQ-test as an indicator of global cog-
nitive abilities and integrity of higher brain functions. Furthermore, the inuence
of the variable education on psychometric intelligence and its interaction with
the duration of epilepsy was investigated. Epidemiological studies identied that
155 Risk of cognitive decline in refractory TLE
30
20
10
0
10
20
30
40
50
I
Q
D
i
f
f
e
r
e
n
c
e
Duration of epilepsy (years)
0 10 20 30 40 50
Figure 11.1. Difference values of the estimated WAIS-R full scale IQ minus MWT-B-IQ (vocabulary
intelligence), as a function of the duration of epilepsy. Data of 37 patients with refractory
temporal lobe epilepsy are shown. The linear regression function confirms a negative
correlation between the difference value and the duration of epilepsy. (With permission
from Jokeit et al. 2000.)
education as an indicator of brain reserve modies the clinical expression of
dementia and Alzheimers disease (Evans et al., 1997; Satz, 1993; Schmand et al.,
1997; Stern et al., 1994; Zhang et al., 1990).
We studied 209 consecutive patients of our epilepsy surgery programme with
TLE who fullled the following criteria: seizures of unilateral temporal origin as
demonstrated by continuous interictal and ictal video/EEG-monitoring, unilateral
lesions within the temporal lobes as demonstrated by MRI, and Full Scale
Intelligence Quotient (FSIQ) greater than 55 to exclude severely mentally impaired
patients. Detailed information on clinical and demographic variables are given in
the original study (Jokeit and Ebner, 1999).
Patients underwent a neuropsychological evaluation designed for patients with
TLE. To have a comprehensive psychometric measure of global cognitive abilities,
which was well normed to age-matched healthy controls, we used the Full Scale
Intelligence Quotient (FSIQ) from the German version of the Wechsler Adult
Intelligence Scale Revised (WAISR) (Tewes, 1991; Wechsler, 1981). The FSIQ
was estimated from the subtests Information, Comprehension, Similarities, Digit
Symbol, Picture Completion and Block Design.
First a data set of 16 independent variables for each patient was submitted to a
linear multiple regression analysis. The analyses on FSIQ revealed signicant con-
tributions by the variables education and duration of epilepsy. The equation
explained 34.6% of total variance. No further variable contributed signicantly.
To reveal possible factor interactions and to exclude linear eects of covariates in
a one-way ANOVA model, the continuous variable duration of epilepsy was
recoded into three values: 15 years0, 1530 years1, and 30 years of refrac-
tory epilepsy2. The side of seizure origin was submitted as the second factor (left,
right). The presence or absence of lesions beyond mesio-temporal structures in
MRI scans was submitted as the third factor (0, 1). Education, age at epilepsy onset,
ranked frequency of interictal epileptiform discharges, ranked frequency of habit-
ual seizures, presence of secondarily generalized seizures in a patients history,
ranked frequency of secondarily generalized seizures within the last year, serum
level for carbamazepine, phenytoin and phenobarbital, and AED mono- or poly-
therapy were controlled as covariates. Only the covariate education was signi-
cantly related to FSIQ. After removing linear eects of covariates the factor
duration of epilepsy was signicant. The factors side of seizure onset and presence
or absence of temporal lesions beyond mesio-temporal structures did not reach sig-
nicance. No interactions reached signicance. Post-hoc contrasts adjusted by
covariates revealed that patients with a duration of epilepsy for more than 30 years
performed worse than patients with less than 15 years of epilepsy and patients with
1530 years of epilepsy. Patients with less than 15 years of epilepsy did not dier
from patients with 1530 years of epilepsy in FSIQ values.
156 H. Jokeit and A. Ebner
Since it is known that variable education explains a considerable amount of var-
iance of FSIQ values, we computed an ANOVA with factors duration of epilepsy (0,
1, 2) and education (low, high) to reveal possible interactions and to specify the
eect of variable education. The educational level was dichotomized into low
(patients who did not attend or did not nish at secondary school, Realschule in
Germany, n109) and high (patients who nished at least at secondary school,
n100). The side of seizure origin, presence or absence of lesions beyond mesio-
temporal structures, age at epilepsy onset, ranked frequency of interictal epilepti-
form discharges, ranked frequency of habitual seizures, presence of generalized
seizures in a patients history, ranked frequency of secondarily generalized seizures
within the last year, serum level for carbamazepine, phenytoin and phenobarbital,
and AED mono- or polytherapy were controlled as covariates. The ANOVA
revealed eects of the factors education and duration of epilepsy on the FSIQ. The
interaction between both factors did not reach signicance. No covariate was sig-
nicantly related to FSIQ. Figure 11.2 shows mean FSIQ values for both educa-
tional groups as a function of duration of TLE. Contrasts adjusted for covariates
revealed that the mean FSIQ values of groups with less than 15 years and 1530
years of TLE did not dier in patients with high educational attainment. However,
patients with more than 30 years of TLE performed worse than patients with less
than 15 years or 1530 years of epilepsy. Patients with low educational attainment
and less than 15 years of TLE performed better than patients with 1530 years of
TLE and patients with more than 30 years of TLE. But there was no signicant
dierence between patients with 1530 years and more than 30 years of TLE in the
low education group. Although not conrmed by an ANOVA interaction, these
contrasts suggest a duration-dependent dierence in mean FSIQ values of patients
with low and high educational attainment. The absence of a signicant ANOVA
interaction probably results from a similar decremental trend of the duration eect
in low- and high-educated patients.
Conclusions
Both studies demonstrate that patients with a long history of intractable TLE were
at higher risk of cognitive impairment than patients with a shorter duration of TLE.
Interestingly, in patients with higher educational attainment, the mean FSIQ was
stable for a longer duration of TLE than in less-educated patients.
We revealed that the educational level and the duration of epilepsy were the best
predictors for psychometric intelligence. Then we provided evidence that only a
long duration of TLE (more than 30 years) was related to impaired psychometric
intelligence in the total sample. The linear inuence of the variables age at epilepsy
onset, educational level of patients, patients serum level of rst-line antiepileptic
157 Risk of cognitive decline in refractory TLE
drugs, polypharmacy, frequency of habitual and secondarily generalised seizures
and frequency of interictal epileptiform discharges on psychometric intelligence
were statistically controlled. Age at testing controlled for duration of TLE was not
signicantly related to FSIQ. The factors side of TLE and presence or absence of
lesions beyond the mesio-temporal structures did not show main eects or inter-
actions. Therefore, the eect of duration of epilepsy cannot be attributed to those
covariates and factors. The patients educational level captured the most amount of
FSIQ variance. This could explain why in contrast to other studies (Strauss et al.,
158 H. Jokeit and A. Ebner
120
110
100
90
80
70
15 1530 30
lowhigh lowhigh lowhigh
Duration of epilepsy (years)
M
e
a
n
F
S
I
Q
Figure 11.2. Mean FSIQ values with 95% confidence intervals of patients with low (n109) and high
(n100) educational attainment for groups with a duration of epilepsy 15 years, 1530
years and 30 years. The factors education and duration of epilepsy were significant
(P0.01). Asterisks (* P0.05; ** P0.01; one tailed) indicate significant contrasts
between adjacent duration groups adjusted for covariates. (With permission from Jokeit
and Ebner, 1999.)
1995) the remaining covariates (e.g. age at epilepsy onset) did not reach signi-
cance.
It is reasonable to assume that human brains develop a functional reserve or have
a spare capacity to cope with a stepwise neuronal loss by eciency, redundancy,
plasticity and reorganization (Lewin, 1980; Meier-Ruge et al., 1991; Stern et al.,
1996). Studies of dierent degenerative brain disorders (e.g. Parkinsons disease,
vascular and Alzheimers dementia) suggest that a functional decline becomes
apparent only if a certain amount of brain parenchyma is insulted (Boone et al.,
1992; Hornykiewicz, 1988; Pasquier and Leys, 1997; Small et al., 1995; Tomlinson et
al., 1970). A long duration of intractable TLE is related to a considerable number of
focal or secondarily generalized seizures, pathological interictal electric brain activ-
ity, chronic and transient metabolic disturbances due to morphological lesions
(Arnold et al., 1996), seizures (Savic et al., 1997) and antiepileptic medication
(Theodore et al., 1989), often the chronic antiepileptic medication with usually high
serum levels (Hermanns et al., 1996). It is suggested that each of these factors may
separately adversely aect cognitive functioning (Dreifuss, 1992; Lesser et al., 1986).
The presence of reactive microglia (Beach et al., 1995), reduced dendritic spine
density, dendritic swellings (Multani et al., 1994) and senile plaques (Mackenzie
and Miller, 1994) in adult temporal lobe specimens suggests that neuronal injury
continues to occur with ongoing seizure activity in TLE patients. Multani et al.
(1994) demonstrated a correlation between decreased dendritic spine density
remote from the epileptogenic zone and duration of seizure history. In patients
with mesial TLE densiometric techniques have revealed secondary declines in hip-
pocampal neuron densities associated with long histories of habitual seizures
(Mathern et al., 1995c, 1996). Recent studies have suggested a secondary decline of
hippocampal volume and temporal lobe metabolism in refractory TLE patients
(Barr et al., 1997; Breier et al., 1997; Jokeit et al., 1999; Salmenper et al., 1998).
Hermann et al. (1997) reported that patients with hippocampal sclerosis had more
generalized cognitive impairment, a signicantly longer history of intractable TLE
and a lower educational level than TLE patients without signicant hippocampal
sclerosis. We assume that a cumulation of small neuro-degenerative eects of
noxious neurochemical agents, abnormal brain electric events, and metabolic dis-
turbances over decades of epilepsy, accompanied by ageing, may increase the prob-
ability that the functional brain reserve or spare capacity is exhausted at a younger
age than expected (mean age of patients with TLE duration 30 years was 44
years), and deterioration of cognitive functions may begin (Meier-Ruge et al., 1991;
Mori et al., 1997; Stern et al., 1996).
The extent of functional reserve and therefore the vulnerability of brain func-
tions may vary considerably between people. It was suggested that higher educa-
tional attainment is related to a higher reserve against cognitive impairment due to
159 Risk of cognitive decline in refractory TLE
stepwise ongoing brain injury (Satz, 1993; Timiras, 1995). Our results of analyses
in patients with lower and higher educational attainment are in accordance with
ndings of epidemiological studies on dementia and Alzheimerss disease (Evans et
al., 1997; Satz, 1993; Schmand et al., 1997; Stern et al., 1994; Zhang et al., 1990). In
patients with higher educational attainment the mean FSIQ was stable for a longer
duration of TLE than in less-educated patients. Thus, higher educational attain-
ment as an indicator of higher cognitive reserve might delay the onset of cognitive
decline in patients with intractable TLE.
The fact that we found signicant eects in the whole sample analysis only in
patients with a history of intractable epilepsy lasting longer than three decades may
question conclusions drawn from negative ndings of studies on adverse eects of
duration of refractory TLE (Mackenzie et al., 1996; Selwa et al., 1994). Based on our
results, only prospective long-term studies which exceed three decades might reveal
the causes of a presumable decline in cognitive functioning of patients with intract-
able TLE. Such studies may solve the important question as to whether certain epi-
leptic syndromes are progressive disorders (Girvin, 1992; Gloor, 1991; Lesser et al.,
1986; Mathern et al., 1996; Sadzot 1997; Sutula et al., 1989). However, today well-
controlled cross-sectional studies comparing dierent well-dened epileptic syn-
dromes and including retest measures to compare dierent treatment strategies
may help to isolate adverse factors and to identify patients with increased risk of
cognitive decline and symptom progression.
REFERENCES
Arnold, S., Schlaug, G., Niemann, H. et al. (1996). Topography of interictal glucose hypo-
metabolism in unilateral mesiotemporal epilepsy. Neurology, 46, 142230.
Barr, W.B., Ashtari, M. and Schaul, N. (1997). Bilateral reductions in hippocampal volume in
adults with epilepsy and a history of febrile seizures. J Neurol Neurosurg Psychiatry, 63, 4617.
Beach, T.G., Woodhurst, W.B., MacDonald, D.B. and Jones, W.M. (1995). Reactive microglia in
hippocampal sclerosis associated with human temporal lobe epilepsy. Neurosci Lett, 191,
2730.
Ben-Ari, Y. and Represa, A. (1990). Brief seizure episodes induce long-term potentiation and
mossy bre sprouting in the hippocamus. Trends Neurosci, 13, 31218.
Bengzon, J., Kokaia, Z., Elmer, E., Nanobashvili, A., Kokaia, M. and Lindvall, O. (1997).
Apoptosis and proliferation of dentate gyrus neurons after single and intermittent limbic sei-
zures. Proc Natl Acad Sci USA, 94, 104327.
Boone, K.B., Miller, B.L., Lesser, I.M. et al. (1992). Neuropsychological correlates of white-matter
lesions in healthy elderly subjects. Arch Neurol, 49, 54954.
Bourgeois, B.F.D., Prensky, A.L., Palkes, H.S., Talent, B.K. and Busch, S.G. (1983). Intelligence in
epilepsy: a prospective study in children. Ann Neurol, 14, 43844.
160 H. Jokeit and A. Ebner
Breier, J.L., Mullani, N.A., Thomas, A.B. et al. (1997). Eects of duration of epilepsy on the
uncoupling of metabolism and blood ow in complex partial seizures. Neurology, 48, 104753.
Brown, S.W. (1996). Epilepsy dementia: Intellectual deterioration as a consequence of epileptic
seizures. Epilepsia, 37 (Suppl. 4), S1223.
Brown, S.W. and Vaughan, M. (1988). Dementia in epileptic patients. In Epilepsy, Behaviour and
Cognitive Function, ed. M.R. Trimble and E.H. Reynolds, pp. 17788. Chichester: John Wiley
& Sons.
Calne, D.B., Eisen, A., McGeer, E. and Spencer, P. (1986). Alzheimers disease, Parkinsons disease,
and motoneurone disease: abiotrophic interaction between ageing and environment? Lancet,
2 (8515), 106770.
Dodrill, C.B. and Wilensky, A.J. (1992). Neuropsychological abilities before and after 5 years of
stable antiepileptic drug therapy. Epilepsia, 33, 32734.
Dreifuss, E.F. (1992). Cognitive function victim of disease or hostage to treatment? Epilepsia,
33 (Suppl. 1), S712.
Engel, J. Jr, Brandler, R., Grith, N.C. and Caldecott-Hazard, S. (1991). Neurobiological evi-
dence for epilepsy-induced interictal disturbances. Adv Neurol, 55, 97111.
Evans, D.A., Hebert, L.E., Becket, L.A. et al. (1997). Education and other measures of socioeco-
nomic status and risk of incident Alzheimer disease in a dened population of older persons.
Arch Neurol, 54, 1399405.
Girvin, J.P. (1992). Is epilepsy a progressive disorder? J Epilepsy, 5, 94104.
Gloor, P. (1991). Mesial temporal sclerosis: historical background and an overview from a
modern perspective. In Epilepsy Surgery, ed. H.O. Lders, pp. 689703. New York: Raven Press.
Glosser, G., Cole, L.C., French, J.A., Saykin, A.J. and Sperling, M.R. (1997). Predictors of intel-
lectual performance in adults with intractable temporal lobe epilepsy. J Int Neuropsychol Soc,
3, 2529.
Hermann, B.P., Wyler, A.R. and Richey, E.T. (1987). Epilepsy frontal lobe, and personality. Biol
Psychiatry, 22, 10557.
Hermann, B.P., Seidenberg, M., Schoenfeld, J. and Davies, K. (1997). Neuropsychological char-
acteristics of the syndrome of mesial temporal lobe epilepsy. Arch Neurol, 54, 36976.
Hermanns, G., Noachtar, S., Tuxhorn, I., Holthausen, H., Ebner, A. and Wolf, P. (1996).
Systematic testing of medical intractability for carbamazepine, phenytoin, and phenobarbital
or primidone in monotherapy for patients considered for epilepsy surgery. Epilepsia, 37,
6759.
Hornykiewicz, O. (1988). Neurochemical pathology and the etiology of Parkinson disease. Mt
Sinai J Med, 55, 1120.
Jokeit, H. and Ebner, A. (1999). Long term eects of refractory temporal lobe epilepsy on cogni-
tive abilities: a cross sectional study. J Neurol Neurosurg Psychiatry, 67, 4450.
Jokeit, H., Seitz, R.J., Markowitsch, H.J., Neumann, N., Witte, O.W. and Ebner, A. (1997).
Prefrontal asymmetric interictal glucose hypometabolism and cognitive impairment in
patients with temporal lobe epilepsy. Brain, 120, 228394.
Jokeit, H., Ebner, A., Arnold, S. et al. (1999). Bilateral depressions of hippocampal volume,
glucose metabolism, and Wada hemispheric memory performance are related to the duration
of mesial temporal lobe epilepsy. J Neurol, 246, 92633.
161 Risk of cognitive decline in refractory TLE
Jokeit, H., Luerding, R. and Ebner, A. (2000). Cognitive impairment in temporal-lobe epilepsy.
Lancet, 355, 101819.
Lehrl, S. (1995). Mehrfachwahl-Wortschatz-Intelligenztest MWT-B [multiple-choice vocabulary
intelligence test]. Balingen: Perimed-spitta.
Lesser, R.P., Lders, H., Wyllie, E., Dinner, D.S. and Morris, III H.H. (1986). Mental deteriora-
tion in epilepsy. Epilepsia, 27 (Suppl. 2), S10523.
Lewin, R. (1980). Is your brain really necessary? Science, 210, 12324.
Lders, H.O. and Awad, I. (1991). Conceptual considerations. In Epilepsy Surgery, ed. H.O.
Lders, pp. 5162. New York: Raven Press.
Mackenzie, I.R.A. and Miller, L.A. (1994). Senile plaques in temporal lobe epilepsy. Acta
Neuropathol, 87, 50410.
Mackenzie, I.R.A., McLachlan, R.S., Kubu, C.S. and Miller, L.A. (1996). Prospective neuro-
psychological assessment of nondemented patients with biopsy proven senile plaques.
Neurology, 46, 4259.
Marsh, L., Morrell, M.J., Shear, P.K. et al. (1997). Cortical and hippocampal volume decits in
temporal lobe epilepsy. Epilepsia, 38, 57687.
Mathern, G.W., Pretorius, J.K. and Babb, T.L. (1995a). Inuence of the type of initial precipitat-
ing injury and at what age it occurs on course and outcome in patients with temporal lobe sei-
zures. J Neurosurg, 82, 2207.
Mathern, G.W., Babb, T.L., Pretorius, J.K., Melendez, M. and Lvesque, M.F. (1995b). The patho-
physiologic relationships between lesion pathology, intracranial ictal EEG onsets, and hippo-
campal neuron losses in temporal lobe epilepsy. Epilepsy Res, 21, 13347.
Mathern, G.W., Babb, T.L., Vickrey, B.G., Melendez, M. and Pretorius, J.K. (1995c). The clini-
cal-pathogenic mechanism of hippocampal neuron loss and surgical outcomes in temporal
lobe epilepsy. Brain, 118, 10518.
Mathern, G.W., Babb, T.L., Leite, J.P., Pretorius, J.K., Yeoman, K.M. and Kuhlman, P.A. (1996).
The pathogenic and progressive features of chronic human hippocampal epilepsy. Epilepsy Res,
26, 15161.
Meier-Ruge, W., Hunziker, O. and Iwango, P. (1991). Senile dementia: a threshold phenome-
non of normal aging? A contribution to the functional reserve hypothesis of the brain. Ann NY
Acad Sci USA, 621, 10418.
Mori, E., Hirono, N., Yamashita, H. et al. (1997). Premorbid brain size as a determinant of
reserve capacity against intellectual decline in Alzheimers disease. Am J Psychiatry, 154,
1824.
Multani, P., Myers, R.H., Blume, H.W., Schomer, D.L. and Sotrel, A. (1994). Neocortical dendritic
pathology in human partial epilepsy: a quantitative Golgi study. Epilepsia, 35, 72836.
Pasquier, F. and Leys, D. (1997). Why are stroke patients prone to develop dementia? J Neurol,
244, 13542.
Rodin, E.A., Schmaltz, S. and Twitty, G. (1986). Intellectual functions of patients with child-
hood-onset epilepsy. Dev Med Child Neurol, 28, 2533.
Sadzot, B. (1997). Epilepsy: a progressive disease? Br Med J, 314, 3912.
Salmenper, T., Klviinen, R., Partanen, K. and Pitknen, A. (1998). Hippocampal damage
caused by seizures in temporal lobe epilepsy. Lancet, 351, 35.
162 H. Jokeit and A. Ebner
Satz, P. (1993). Brain reserve capacity on symptom onset after brain injury: a formulation and
review of evidence for threshold theory. Neuropsychology, 7, 27395.
Savic, I., Altshuler, L., Baxter, L. and Engel, J. Jr (1997). Pattern of interictal hypometabolism in
PET scans with udeoxyglucose F 18 reects prior seizure types in patients with mesial tem-
poral lobe seizures. Arch Neurol, 54, 12936.
Schmand, B., Smit, J.H., Geerlings, M.I. and Lindeboom, J. (1997). The eects of intelligence and
education on the development of dementia. A test of the brain reserve hypothesis. Psychol Med,
27, 133744.
Seidenberg, M., OLeary, D.S., Berent, S. and Boll, T. (1981). Changes in seizure frequency and
test-retest scores on the Wechsler Adult Intelligence Scale. Epilepsia, 22, 7583.
Selwa, L.M., Berent, S., Giordani, B., Henry, T.R., Buchtel, H.A. and Ross, D.A. (1994). Serial cog-
nitive testing in temporal lobe epilepsy: longitudinal changes with medical and surgical ther-
apies. Epilepsia, 35, 7439.
Small, G.W., Mazziota, J.C., Collins, M.T. et al. (1995). Apolipoprotein E type 4 allele and cere-
bral glucose metabolism in relatives at risk for familial Alzheimer disease. J AmMed Assoc, 273,
9427.
Stern, R.A., Silva, S.G., Chaisson, N. and Evans, D.L. (1996). Inuence of cognitive reserve on
neuropsychological functioning in asymptomatic human immunodeciency virus-1 infec-
tion. Arch Neurol, 53, 14853.
Stern, Y., Gurland, B., Tatemichi, T.K., Tang, M.X., Wilder, D. and Mayeux, R. (1994). Inuence of
educationandoccupationonthe incidence of Alzheimers disease. J AmMedAssoc, 271, 100410.
Strauss, E., Loring, D., Chelune, G. et al. (1995). Predicting cognitive impairment in epilepsy:
ndings from the Bozeman Epilepsy Consortium. J Clin Exp Neuropsychol, 17, 90917.
Sutula, T., Cascino, G., Cavazos, J., Parada, I. and Ramirez, L. (1989). Mossy ber synaptic reor-
ganization in the epileptic human temporal lobe. Ann Neurol, 26, 32130.
Tasch, E., Cendes, F., Li, L.M., Dubeau, F., Andermann, F. and Arnold, D.L. (1999).
Neuroimaging evidence of progressive neuronal loss and dysfunction in temporal lobe epi-
lepsy. Ann Neurol, 45, 56876.
Tewes, U. (1991). Manual des Hamburg-Wechsler Intelligenztest fr Erwachsene Revision 1991.
Bern: Verlag Hans Huber.
Theodore, W.H. and Gaillard, W.D. (1999). Association between hippocampal volume and epi-
lepsy duration. Ann Neurol, 46, 800.
Theodore, W.H., Bromeld, E. and Onorati, L. (1989). The eect of carbamazepine on cerebral
glucose metabolism. Ann Neurol, 25, 51620.
Theodore, W.H., Bahita, S., Hatta, J. et al. (1999). Hippocampal atrophy, epilepsy duration, and
febrile seizures in patients with partial seizures. Neurology, 52, 1326.
Timiras, P.S. (1995). Education, homeostasis, and longevity. Exp Gerontol, 30, 18998.
Tomlinson, B.E., Blessed, G. and Roth, M. (1970). Observations on the brains of demented old
people. J Neurol Sci, 11, 20542.
Trimble, M.R. (1988). Cognitive hazards of seizure disorders. Epilepsia, 29 (Suppl. 1), S1924.
Wechsler, D. (1981). WAISR Manual. New York: The Psychological Corporation.
Zhang, M., Katzman, R., Salmon, D. et al. (1990). The prevalence of dementia and Alzheimers
disease in Shanghai, China: impact of age, gender, and education. Ann Neurol, 27, 42837.
163 Risk of cognitive decline in refractory TLE
12
Behavioural and neuropsychological aspects
of frontal lobe epilepsy
Christoph Helmstaedter
University Clinic of Epileptology Bonn, Bonn, Germany
Introduction
To date, neurobiological interest in behaviour and epilepsy has been concerned pri-
marily with temporal lobe epilepsies (TLE), and mesial temporal lobe epilepsies
(mTLE) in particular. This concentration on TLE is mostly due to the fact that this
type of epilepsy represents the majority of the focal epilepsies in the Bonn series
of patients with pharmacoresistant epilepsies this is about 80% and that mTLE
often forms an entity within the focal epilepsies regarding pathology (hippocam-
pal sclerosis), a frequent history of febrile convulsions, an early onset of epilepsy,
and memory problems as the prominent neuropsychological impairment. In TLE,
the aected cerebral structures and epileptogenic region are mostly circumscribed,
and structural pathology can be well quantied by quantitative MRI (T2 relax-
ometry and volumetry) or postoperative histopathological examinations of the
resected specimen. Frequency, homogeneity and quantiable pathology provide
ideal prerequisites for the study of the functional and behavioural correlates of
TLE. Great progress has been made during recent years, at least with respect to the
neuropsychological and cognitive aspects of TLE. Recent developments in the eld,
however, show that it is well recognized that temporal lobe functioning involves
more than memory and that its role in emotion and psychiatric symptoms is being
rediscovered.
The conditions we meet with frontal lobe epilepsy (FLE) are quite dierent.
Frontal lobe epilepsies, in spite of the size of the frontal lobes, are less frequent than
the temporal lobe epilepsies. In our own series, patients with FLE represent about
15% of the patients with pharmacoresistant epilepsies. Site and type of the under-
lying pathology are very heterogeneous. Furthermore, ictal and interictal clinico-
electric manifestations of FLE are infrequently localized, because multiple
connections to most other brain areas enable fast and widely distributed propaga-
tion of epileptic activity. The functional correlates of frontal pathology in epilepsy
are thus less well understood.
164
Frontal lobe and behaviour disorders
Correspondent to its cytoarchitectonic structure, the frontal lobe is traditionally
divided into two parts, which are important in two major areas. The posterior part
controls motor movement and is subdivided into a premotor and a motor area,
which control movement preparation and actual execution of movement respec-
tively. The anterior part of the frontal lobe, the prefrontal cortex, is especially
important in higher mental function, as in anticipation and planning, initiative,
judgement, and in aect control, will power, and the determination of personality
(Bechara et al., 1999; Raine et al., 2000). The prefrontal cortex can be further sub-
divided into the dorso-lateral cortex and the orbito-frontal cortex. This subdivision
of the prefrontal cortex is still simple and it should be noted that the orbito-frontal
cortex itself is a heterogeneous area connected with a wide range of other prefron-
tal, limbic, premotor, sensory areas and subcortical nuclei (Cavada et al., 2000). For
the purpose of this article it is important to know that there is evidence that damage
of the dorso-lateral part of the prefrontal cortex is more associated with impair-
ment of executive functions and functions of working memory, whereas damage of
the orbito-frontal cortex leads to impairment of the choice of behaviour, the estab-
lishment of emotional valences, and the evaluation and balancing of the past and
future consequences of a given behaviour (Bechara et al., 2000; Rolls, 2000; Sarazin
et al., 1998). Studies in common marmosets suggest a dissociation between the
lateral and the orbital-medial division of the prefrontal cortex according to which
the rst selects and controls actions on the basis of higher-order rules while the
latter controls dierent behaviour on the basis of lower-order rules (see Roberts
and Wallis, 2000).
The signicance of the orbito-frontal cortex for social and interpersonal behav-
iour recently came into focus when Anderson et al. (1999) reported two patients,
one with an accident at 15 months, the other with a frontal tumour resection at 16
months, both showing severe impairment of social and moral behaviour.
Traditionally, behavioural dyscontrol has been attributed to temporo-limbic
structures. Evidence for the involvement of the amygdala in aggression comes from
human and animal stimulation studies, from activating and inhibiting eects of anti-
epileptic drugs on aggression, and recently also from direct correlations of amygdala
volumes with aggression in patients with mesial epilepsies (Azouvi et al., 1999; Bearn
and Gibson, 1998; Trimble and Van Elst, 1999; Van Elst et al., 2000). However, aggres-
sion associated with the amygdala seems more defensive than oensive in nature
(Kalynchuk et al., 1999). Disinhibition phenomena, or a loss of impulse control as it
is observed with frontal lesions, may be an additional prerequisite for showing
impulsive aggressive behaviour. The orbito-frontal cortex as the border zone between
the frontal lobe and the limbic system is thus critical for the linkage of the frontal and
165 Behavioural and neuropsychological aspects of FLE
limbic aspects of behavioural dyscontrol disorders. Another important area in this
respect is the anterior-cingulate gyrus, which is also strongly connected to the amyg-
dala, and whose damage has also been associated with deviant social behaviour and
aect states (for overview see Devinsky et al., 1995).
The nding of antisocial and aggressive behaviour with frontal lobe damage is
not in itself new. A prominent and often cited example is the historic case of
Phineas Gage, who after an accident with a severe frontal brain injury changed from
a well-mannered man into an irresponsible and convention-neglecting person
(Damasio et al., 1994; Harlow, 1868). New in the study of Anderson et al. (1999) is
the nding that whether patients not only display severe behavioural disorders but
also fail to see the moral of the behaviour depends on the age at the lesion onset
(Dolan, 1999). Consequently, the orbito-frontal cortex seems not only important
for behaviour control but also for the acquisition of social and interpersonal rules.
It is important to note that irresponsible, aggressive and sociopathic behaviours
can occur independent of intellectual abilities, which are often well preserved in
frontal lesions. Other areas in which the orbital and medial prefrontal cortex are
believed to play a central role are addictive behaviour, attention-decit hyperactiv-
ity disorder, negative emotion and major depression (Drevets and Raichle, 1996;
London et al., 2000; Northo et al., 2000; Rubia et al., 2000). Davidson et al. (2000)
propose a key role of the prefrontal cortex in the regulation of emotion in violent
subjects and those predisposed to violence.
A theoretical basis of the role of the prefrontal cortex in the interplay of cogni-
tion and emotion has been provided with Damasios somatic marker hypothesis
(Damasio, 1996). This proposes that responses to external stimuli do not rely on
either conditioning processes or cognition alone, but on somatic marker signals
or autonomic response sets, which determine the conscious/unconscious connec-
tion between stimulus conditions, feelings and behaviour.
Epilepsy and behaviour disorders
Single case reports of behavioural and personality disorders in patients with severe
brain lesions often appear dramatic. However, with respect to focal epilepsies, these
reports nevertheless raise the question of whether there might be parallels in the
behaviour when epilepsy aects the same brain regions. With the exception of rare
cases of ictal aggression, postictal confusional states or psychosis (Marsh and
Krauss, 2000), behaviour and personality disorders observed in patients with FLE
appear less severe. Furthermore, as with TLE, one can hardly expect to nd the
prototypical frontal epileptic personality or Wesensnderung. Personality is by
denition more trait than state dependent and, particularly in epilepsy, it is quite
dicult to determine whether a given behaviour has trait characteristics or not.
166 C. Helmstaedter
Conclusions about persistence and continuity require follow-up observations with
longer time intervals. In epilepsy several factors can be discerned, which can lead
to more or less reversible changes in patients cognitive abilities and mood states
(see Table 12.1). Furthermore, although we can now look back on a long history of
successful epilepsy surgery, it is still not clear to what degree the fact of having sei-
zures is a prerequisite of behaviour and mood disorders in epilepsy.
The patient with epilepsy must always be seen in his state relative to seizures, e.g.
whether he is ictal, postictal or interictal. According to recent ndings with regard
to seizure prediction by nonlinear measures of complexity loss as recorded by intra-
cranial EEG, signicant seizure-precipitating drops in complexity, i.e. synchroniza-
tion, can be recorded long before the seizure starts (Elger and Lehnertz, 1998).
Accordingly, one must assume also preictal states, which would t well with
patients reports of increased dysphoric mood and cognitive problems before the
seizure starts. Finally, since many patients can now become seizure-free on a long-
term basis by epilepsy surgery, one can suggest an additional state of well-
controlled epilepsy after successful epilepsy surgery.
167 Behavioural and neuropsychological aspects of FLE
Table 12.1. Factors affecting cognitive and mood states in epilepsy
States of epilepsy
preictal
ictal
postictal
interictal
(seizure free after successful surgery)
Seizures
frequency
generalization
nonconvulsive status epilepticus
Epileptic dysfunction
local versus distant eects
Lesion
e.g. alien tissues vs. migration and developmental disorders (confounded with dierent ages
at lesion/epilepsy onset)
extent, location, lateralization
Antiepileptic drugs
positive vs. negative psychotropic eects
individual incompatibility
drug-induced encephalopathy
intoxication
Epileptic activity can aect distant brain areas and cause cognitive and behav-
iour problems not related to the primary lesion or epileptogenic zone (Shulman,
1984). Notwithstanding seizures and epileptic activity, one must also dierentiate
the underlying pathologies, which can be more or less systemic, have dierent ages
of onset in life, and thus have dierent eects on brain maturation and the devel-
opment of cognition and personality. We must nally consider inuences of often
longstanding antiepileptic medication in these patients. Antiepileptic drugs may
have positive or negative psychotropic side eects, and can show incompatibilities
in the individual patient (Schmitz, 1999). Interactive eects of pathology, epilepsy
and treatment must be considered. Apart from idiosyncratic actions, drugs can
have dierent eects in lesion and nonlesion patients, and they may act dierently
dependent on seizure control.
Taking this into consideration, it will be shown in the next sections that there is
nevertheless evidence of specic behavioural abnormalities in patients with FLE,
which can be interpreted within a theoretical framework of frontal lobe dysfunc-
tion. This will be outlined with the example of interictal behaviour as assessed by
neuropsychological examination and self-report measures concerning quality of
life, everyday activities, personality and psychiatric symptoms. In addition, seizure
semiology and impairment during frontal lobe seizures and frontal nonconvulsive
status epilepticus will be considered, to convey an idea of what the behavioural con-
sequences of impaired frontal lobe functions in FLE might be.
The neuropsychology of frontal lobe epilepsy
The development of neuropsychology in FLE is probably best reected by Brenda
Milners (1995) description of her evaluation of Penelds patient K.M., the frontal
counterpart of the temporal patient H.M. This patient had a penetrating head
injury in 1928, developed seizures, and underwent surgery of the anterior parts of
both frontal lobes. Surgery successfully controlled the seizures and led to improved
behaviour as well as improved IQ. However, when re-evaluated with the newly
developed Wisconsin Card Sorting test in 1962 he showed severe impairment in
exible categorical thinking and concept formation whilst his IQ still was average.
This case exemplies how much outcome interpretation depends on both the test
sensitivity and test selection.
Since then there have been surprisingly few attempts to apprehend the cognitive
characteristics of patients with FLE with group studies (Shulman, 1984).
Unfortunately, most of the data from Milners era stem from patients after surgery
and thus tell us more about frontal lobe lesions than about FLE. Furthermore, most
earlier studies focused on single functions, more or less following a rather monis-
tic view of a frontal central executive (Baddeley and Hitch, 1974). Major impair-
168 C. Helmstaedter
ments indicated by these studies are problems in concept formation, response inhi-
bition (Milner, 1964), estimations (Smith and Milner, 1984), conditional associa-
tive learning (Petrides, 1985; Petrides and Milner, 1982), and prot from
information provided in advance in choice reaction tasks (Alivisatos and Milner,
1989). Focusing on memory, Delaney et al. (1980) found no dierences in meas-
ures of memory when nonoperated patients with unilateral frontal lobe foci were
compared to healthy controls. The rst of our own studies found that decits in
attention are the most signicant problem in patients with FLE (Kemper et al.,
1992).
Later systematic group studies in nonresected patients with FLE followed the
theoretical suggestion of dierent frontal subfunctions (Stuss and Benson, 1986)
and met the requirements of the manifold frontal lobe pathology by the use of a
broader range of tests. These addressed dierent aspects of attention, motor coor-
dination, psychomotor speed, uency, response inhibition, conceptual formation
and shift, as well as planning, guessing or estimating.
Between 1996 and 1999 Upton and Thompson published a series of ve articles
reporting dierent ndings on neuropsychology in their sample of 74 subjects with
FLE. Using a test battery with dierent measures of executive functions and motor
skills, they come to the conclusion that patients with epilepsy show a decit pattern
which is similar to that found in frontal lobe dysfunction in general (Upton and
Thompson, 1996a, b). As compared with patients with TLE, frontal patients show
poorer motor coordination, guessing, estimation and response inhibition.
Similarly, we found in 23 patients with FLE that cognitive problems could be
diagnosed with a broad range of 10 frontal tasks with about double as many test
parameters. The great number of test parameters, however, turned out to be highly
redundant and could be statistically reduced to four relatively independent func-
tional areas, namely psychomotor speed/attention, motor coordination, working
memory, and response inhibition. These four factors explained 70% of the total
variance. When compared with patients with TLE, those with FLE were character-
ized by prominent impairment in motor skills and response inhibition
(Helmstaedter et al., 1996). Problems in speed/attention and working memory
were frequent but they appeared rather nonspecic since they were also observed
in the temporal lobe group. This, however, does not necessarily contradict the
assumption that these are frontal functions. An imaging study of Jokeit et al. (1997)
showed in this respect that in patients with TLE, prefrontal metabolic asymmetries
are evident which are associated with frontal lobe measures and intelligence.
In another of our own studies we addressed the cognitive consequences of frontal
lobe surgery. We evaluated 33 patients pre- and postoperatively and were able to
conrm the impairment pattern of impaired motor skills and response inhibition.
We also showed that frontal surgery does not cause considerable additional damage
169 Behavioural and neuropsychological aspects of FLE
if eloquent cortex (SMA; motor and language area) is spared. However, when
surgery included resection of the SMA the most prominent neuropsychological
symptom besides neurological decits directly after surgery was a SMA deciency
syndrome (impairment of initiation) with aphasia (speech arrest and transcortical
aphasia) (Helmstaedter et al., 1998). Additional psychomotor slowing was
observed in lobectomies as compared to lesionectomies.
Looking closer at clinical variables which might explain the impairment pattern
in nonresected patients, no consistent picture emerges. According to Upton and
Thompson (1997a), seizure frequency and the duration of epilepsy have an eect
on performance, but this appears to be a nonspecic eect rather than a consistent
nding over dierent tests. With the exception of motor skills, which were spared
in early right-sided FLE, no systematic eect of the assumed inuence of the age at
the onset of epilepsy on cognitive development could be concluded from their data
(Upton and Thompson, 1997b).
The impact of having epileptic seizures on cognition can be demonstrated by our
postoperative ndings, indicating that in seizure-free patients adjacent functions
recovered after surgery. Comparable release eects have been also reported after
temporal lobe surgery (Hermann et al., 1988). However, one should not go so far
as to conclude that all decits are due to epileptic dysfunction and are thus rever-
sible, as has been suggested by Boone et al. (1988) in a single case report.
In conclusion, with regards to the neuropsychological ndings in FLE, it appears
that indeed dierent frontal subfunctions can be dierentiated. Nevertheless, the
measures which characterize FLE have in common the demand of adequate
response selection and initiation and response inhibition. This holds for tests which
explicitly assess interferences and response inhibition, but also for testing of motor
skills or working memory. Ending up again with a unique central executive func-
tion, one may hypothesize that the particular problem in FLE is the impairment of
response selection/initiation/inhibition with varying emphasis depending on
dierent functional areas. Which area is aected then depends on the type and the
localization of the underlying lesion, including the possibility that symptoms are
overshadowed by spreading epileptic dysfunction.
Besides this, it is important to mention that the development of appropriate test
instruments for the assessment of frontal lobe dysfunction is not yet complete and
still represents a challenge for neuropsychologists. Most psychometric tests which
allow quantication of test behaviour provide patients with a clear structure for
behaviour, i.e. with test instructions, rules, time constraints etc. This enables the
patient to behave in an ordered way and real problems with behaviour organiza-
tion arising from frontal pathology are easily overlooked. If provided with greater
freedom and demands on spontaneous interactions with complex situations, the
same patient would otherwise reveal decits. A possible solution to this dilemma
170 C. Helmstaedter
could be to design tasks which evoke spontaneous behaviour and decisions which
have to be made by the patient, as has been done by Bechara et al. (1998) with the
gambling task, by Goldberg and Podell (2000) with their Cognitive Bias Task, or by
Upton and Thompson (1999) with their Twenty Questions Task.
Ictal behaviour in frontal lobe seizures: positive and negative phenomena
Like others before us, we recently analysed seizure phenomena in patients with FLE
by video-EEG monitoring. The main purpose was to get hints from seizures for
dierential diagnosis. On the other hand, seizures can be studied in terms of tran-
sient dysfunctions which are more or less localized and seizure semiology; pre-
served functions, as well as impaired functions, can tell us something about the
cerebral functional organization of cognition and consciousness.
We studied positive phenomena in terms of seizure semiology, and negative
phenomena in terms of impairment when patients were neuropsychologically
tested during their seizures (Helmstaedter et al., unpublished data; Lux et al., 2000;
Scherrmann and Elger, 1999).
Ictal phenomena in frontal seizures are mostly positive phenomena (see Table
12.2). On the one hand, this means a nearly 1: 1 relationship between discharges and
motor excitation when direct access to motor neurons is possible in primary motor
area seizures, for example. On the other hand, this means release and disinhibition of
171 Behavioural and neuropsychological aspects of FLE
Table 12.2. Ictal frontal seizure semiology (n15)
Localization Positive symptoms
Motor area Nearly 1:1 manifestation of seizure activity in
myoclonic and tonic or clonic motor activity
SMA (speech and motor area) Tonic posturing
Premotor Contraversive head and eye movements
Prefrontal (including gyrus cinguli) Explosive and complex motor automatisms
(including vocalizations); bizarre and hysterical
behaviour; mood change
Negative symptoms
Mesial propagation and secondary Loss of consciousness
generalization
Impaired executive control
Pathological excitation and disinhibition
complex behaviours and behaviour chains when precentral areas are involved.
Examples are posturing andcontraversive movements inSMAandpremotor seizures,
and explosive, bizarre, and emotional unstable behaviours in prefrontal seizures
including its mesial parts. Negative phenomena like loss of consciousness are com-
monly observed in seizures with mesial propagation and secondary generalization.
For frontal seizures one can thus conclude that the prominent feature is impairment
of executive control in terms of a pathological hyper-excitation or disinhibition.
Neuropsychological examination of the cognitive impairment during seizures
can provide additional insight into the ictal event. We performed ictal testing in 116
patients, most of them being candidates for epilepsy surgery. These patients under-
went ictal examinations which included examination of orientation reexes
(verbal, nonverbal, tactile), expressive/receptive language (commands, naming
repetition), nonverbal reception/expression (commands and imitation) and nally
awareness and memory (interview after the seizure). Testing was performed by the
video-EEG monitoring sta and started as soon as possible after seizure onset.
Functions were tested hierarchically according to their complexity and testing was
continued until the seizure ended. About half of the patients had implanted strip
or depth electrodes for invasive EEG recordings.
Table 12.3 shows the impairment pattern which results when the distribution of
ictal EEG activity at the time of testing is considered. In comparison to lateralized
and bilateral temporal lobe seizures frontal lobe seizures are characterized by
prominent impairment of orientation reexes and expressive speech, which are
typical frontal functions. Receptive speech is often preserved. Patients can try, for
example, to follow body commands even when they appear involved in excessive
motor activity. In contrast to left and bitemporal seizures, consciousness (aware-
ness of any kind) and memory for the test situation during the seizures is mostly
preserved.
172 C. Helmstaedter
Table 12.3. Negative ictal symptoms in focal epilepsy (n116)
Location of seizure activity
Percentage impaired when Frontal Right temporal Left temporal Bitemporal
tested ictally n29 n21 n38 n28
Orientation reex 62 10 18 57
Receptive speech (commands) 48 15 59 93
Expressive speech 77 11 47 76
Memory 31 0 46 100
Consciousness 33 12 39 100
A very interesting behaviour and neuropsychological pattern of impairment can
be observed in patients with frontal nonconvulsive status epilepticus. It is impor-
tant to note that in contrast to a generalized tonic-clonic status, which is the repe-
tition of the same seizure, the nature of frontal nonconvulsive status and frontal
seizures is completely dierent. In contrast to frontal lobe seizures, seizure semiol-
ogy of nonconvulsive status is dominated by negative seizure phenomena. Without
EEG recording the epileptic nature of this state is easily overlooked, and patients
appear rather strange, since they are slowed, dysphoric, morose and dicult. When
neuropsychologically examined during the seizure we found in ve cases consis-
tently generally reduced activity, uctuating orientation, reexive and no self-ini-
tiated behaviour, perseverations, intrusions, apractic signs, problems to shift
between tasks, impaired working memory on a higher cognitive level and emo-
tional instability (see Table 12.4). In 1997 we described a single patient with a non-
convulsive status epilepticus who showed a generalized EEG pattern but focal
cognitive decits when neuropsychologically tested during this state. Today, with
better diagnostic tools, we would probably be able to reinterpret this case as frontal
nonconvulsive status (Bauer et al., 1997).
Contrasting frontal seizures to frontal nonconvulsive status, one could interpret
the latter in terms of an impaired executive control by pathological hyper-
inhibition. Impressive recovery to normal behaviour can be observed in these
173 Behavioural and neuropsychological aspects of FLE
Table 12.4. Ictal symptoms in frontal nonconvulsive status epilepticus (n5)
Performance Impairment
Motor functions (including speech) Generally reduced activity
Rarely, automatisms (fumbling etc. . . .)
Orientation Fluctuating
Executive functions (including language) No self-initiated directed actions
Increased perseverations
Intrusions
Apractic signs in object use and imitation
Reasoning Problems with concept formation and shift
(colour/form . . .)
Working memory Only impaired when complex mental
information processing is required
Emotion Emotional instability (dysphoric)
Impaired executive control
Pathological inhibition
patients when the status is successfully cessated by injection of diazepam. This is
one form of state-dependent cognitive impairment, as discussed in more detail in
Chapter 6 by Besag.
Postictally, patients, since they often do not lose consciousness during frontal
lobe seizures, are quickly reoriented. Figure 12.1 shows the course of verbal
memory and decision times in pre- and postictal memory testing after frontal lobe
seizures as compared with left/ right temporal seizures and repeated testing in
healthy controls. After lateralized temporal lobe seizures, material-specic memory
impairment can be observed for at least one hour after complete reorientation.
What is shown for left temporal patients in verbal memory in Figure 12.1 has its
counterpart for right temporal patients in gural memory. As for frontal lobe sei-
zures, it is remarkable that there is no postictal deterioration in memory nor sig-
nicant slowing down of reaction times. However, when seizures secondarily
generalize, lasting memory impairment can be observed following frontal seizures
(Helmstaedter et al., 1994).
We can conclude so far that from frontal lobe seizures a dysexecutive syndrome
results, with mostly preserved awareness and consciousness, only reexive but no
self-initiated behaviour, and a seizure semiology, which is dominated by a state of
hyperexcitation and disinhibition or hyperinhibition. This would conrm the
impression from neuropsychological ndings that the major problem in FLE is
174 C. Helmstaedter
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Figure 12.1. Preictal baseline measures and postictal course of verbal memory and decision times in
patients with frontal and left or right temporal lobe epilepsy. The bars indicate
performance of healthy subjects when tested repeatedly in the same intervals.
appropriate response selection/initiation and inhibition of behaviour. A further
dierentiation according to lesions or foci within particular sites of the frontal
lobes can be suggested but has not yet been proven. From a neuropsychological
point of view it is still dicult to decide whether one central executive function or
dierent executive functions should be assumed. As already mentioned a compro-
mise is favoured at the moment, which suggests that the frontal subfunctions are
constituted by similar processes of response selection/initiation and inhibition in
dierent domains and modalities of behaviour.
Behavioural correlates of frontal lobe epilepsy
If we propose problems with behaviour selection/initiation and inhibition as a
functional complex which is mainly aected in frontal lobe epilepsy, the obvious
question is whether or not this dysfunction has a correlate in personality and
behaviour.
With respect to this question we applied several self-rating scales to a group of
95 patients with either frontal (n18) or mesial temporal lobe epilepsy (n77).
Epilepsy groups were matched regarding sex, the age at the onset of epilepsy (mean
11 years) and the duration of epilepsy (mean 24 years). The tests in use were the
BPSE activity subscale assessing frequencies of activities (Helmstaedter and Elger,
1994); depression and anxiety were assessed by the Beck Depression Inventory
(BDI; Beck et al., 1981) and the Zung Self Rating Anxiety Scale/SAS (Zung, 1971);
personality was assessed by the Neo Five Factor Inventory, a German version of the
NEO personality inventory (Costa and McCrea, 1992). Quality of life in epilepsy
was assessed by a German modied QOLIE-10 (English version: Cramer et al.,
1996); and nally we evaluated education and employment in order to add some
objective data.
Group comparisons considering localization and lateralization of epilepsy
revealed only slight dierences (Table 12.5). Patients with mTLE as a trend showed
poorer mood and signicantly increased anxiety scores, they described themselves
as more active at home, less active with respect to outdoor cultural activities, and
less open for experiences than patients with FLE. It is important to note that, when
compared to normative data of healthy control subjects, the result regarding
outdoor cultural activities must be interpreted in a way that patients with FLE are
more active than the controls and patients with mTLE. Furthermore, when com-
pared to data of a healthy control group, the neuroticism score of patients with
mTLE and the conscientiousness score of patients with FLE appeared elevated.
As regards quality of life (QOL), patients were categorized as having poor QOL
when they showed scores below the 25th percentile. As is shown in Figure 12.2
patients with TLE generally tended to report poorer QOL than patients with FLE.
175 Behavioural and neuropsychological aspects of FLE
Impaired mood, memory problems and social limitations correspond well to the
features of TLE found with the other instruments in this evaluation.
Our current approach to behavioural problems and personality in patients with
focal epilepsies is less led by classication systems, which may be useful in idio-
pathic psychiatric disorders. As already mentioned in the introduction, there is a
long history of personality research in epilepsy and up to now no consistent fea-
tures have been discerned. So far this has been explained by the multifactorial
determination of psychiatric problems in patients with symptomatic epilepsies. As
176 C. Helmstaedter
Table 12.5. Group comparisons considering localization and lateralization of epilepsy
Scale Group Mean SD Signicance
Mood (BDI SAS)
Depression FLE 7.6 7.3 n.s.
mTLE 11.1 9.0
Anxiety FLE 29.7 7.9
a
mTLE 35.9 7.4
Activities (BPSE: activity subscale)
Home activities FLE 25.3 4.9
a
mTLE 27.8 5.7
Social activities FLE 20.2 5.3 n.s.
mTLE 18.7 6.1
Cultural activities FLE 16.3 6.5
a
mTLE 12.8 5.4
Personality (NEO FFI)
Neuroticism FLE 21.4 5.4 n.s.
mTLE 24.7 7.4
Extraversion FLE 26.2 4.3 n.s.
mTLE 26.0 6.2
Open to experiences FLE 28.6 6.6
a
mTLE 25.1 5.3
Agreeableness FLE 31.6 4.6 n.s.
mTLE 30.2 4.2
Conscientiousness FLE 34.9 5.6 n.s.
mTLE 33.2 5.3
Notes:
FLE, frontal lobe epilepsy; mTLE, mesial temporal lobe epilepsy.
For decriptions of scales used, see text.
a
Signicantly dierent; n.s., not signicantly dierent.
far as psychometric approaches are concerned previous studies of temporal lobe
epilepsy mostly used the MMPI (Rose et al., 1996) or more specically the
BearFedio Inventory (Bear et al., 1982; Devinsky and Najjar, 1999). It is our daily
experience that commonly used psychiatric scales or psychological personality
inventories largely fail to reect objectively what seems to the examiner clearly to
be an epilepsy-related change in personality or a behaviour disorder.
At the moment we are evaluating our own clinical personality inventory, which
was empirically designed according to a collection of behavioural problems per-
ceived by the clinical psychological sta in the University Hospital of Epileptology
in Bonn, Germany (Helmstaedter et al., 2000a). For preliminary analysis the ques-
tionnaire was consecutively applied to 59 patients with TLE, 17 patients with FLE,
9 patients with parieto-occipital epilepsy and 44 healthy controls. It consists of 82
questions concerning 15 dierent behavioural domains. The answer style is a six-
stepped frequency of occurrence rating with 1occurs not at all and 6occurs
very frequently. Second order factor analysis resulted in six factors, which were
interpreted as follows: 1. organic personality change with patients reporting com-
munication problems, emotional lability, being indecisive, susceptible to interfer-
ence, perseverative and hypoactive; 2. depressed mood including depressive
mood, reduced vitality, anxiety, and insensitivity; 3. addiction and obsession
including addiction to legal and illegal substances, compulsion, and obsession;
177 Behavioural and neuropsychological aspects of FLE
Figure 12.2. Quality of Life (QOL) in frontal lobe epilepsy (FLE) as compared with mesial temporal
lobe epilepsy (mTLE). Values 25th percentile were considered as reflecting perception
of impaired QOL. Asterisks indicate significant group differences in Chi-square testing.
AED, antiepileptic drug.
4. extraversion comprised sociability, curiosity and self-determined behaviour;
5. aggression comprised aggression, sensation-seeking, nonadaptive behaviour
and violence; 6. hyperactivity and adaptivity. When taking clinical data as well as
sex as independent variables some interesting and comprehensible results could be
obtained (Figure 12.3).
The data rst of all indicate that problems in the respective areas are evident in
2030 % of patients. Organic personality changes are preferentially seen in left epi-
lepsies of either origin, in women more than in men. Addiction and obsession are
more frequent in right epilepsies and in frontal epilepsies in particular. Depressed
mood is preferentially seen in patients with hippocampal sclerosis, a nding which
is in line with one of our recent publications (Quiske et al., 2000). All patients and
patients with parietal epilepsies in particular show reduced extraversion. Aggressive
behaviour seems more frequent in left epilepsies, and patients with FLE show
increased hyperactivity and adaptivity, which may parallel the nding of increased
outdoor/cultural activities and openness for experiences. It is important to note
that these results are preliminary and that larger control groups and validation
studies are still required. However, the data indicate that the often-cited depressive
mood is not the only behavioural problem in patients with focal epilepsy, and that
apart from this there are specic behavioural aspects which appear related to local-
ized and lateralized lesions or epileptic dysfunctions.
Although no dierences between patients with FLE and TLE were obtained, it is
worth reporting the results with regard to the organic personality change scale in
more detail. As shown in Figure 12.4 for selected items, about 20% of the patients
report that they oend others; between 20 and 35% of the patients report problems
178 C. Helmstaedter
*P < 0.05, **P < 0.01 (significantly different from control subjects)
Scale
1. Organic personality change
2. Addiction/obsession
3. Depressed mood
4. Extraversion
5. Aggression
6. Adaptivity/ hyperactivity
32%**
20%*
28%
32%**
31%**
29%*
Impairment
< 1 SD
LEFT*
RIGHT*
/
LEFT/RIGHT*
LEFT*
/
Lateralization
37 right/48 left
TLE/FLE*
FLE*
/
PLE*
/
FLE*
Localization
59 T/17 F/ 9 P
W > M
/
/
/
/
/
Sex
59/ 70
/
/
AHS (right)*
/
/
/
Pathology
28 with AHS
Figure 12.3. Results obtained with the clinical personality questionnaire. TLE, temporal lobe epilepsy;
FLE, frontal lobe epilepsy; PLE, parieto-occipital epilepsy.
with reception, misunderstandings, that they were perceived as perseverative or
circumstantial; and 50% report that their behaviour irritates others. This is similar
to the epileptic personality, and taken together with the depressed mood, one
might well think of the dysphoric and paroxysmal mood disorder as it has been
proposed from a more psychiatric point of view (Blumer, 2000).
Academic achievement and employment in frontal lobe epilepsy
From patients with frontal lesions, it is well known that they may show unimpaired
cognitive functions but nevertheless fail on everyday demands of job and career
because of behavioural problems, unsteadiness, concentration problems, increased
susceptibility to interference and problems with timing and planning. Subjective
data may not detect behavioural problems because patients with frontal lobe
lesions have been reported to underestimate their impairments. With school
achievement and employment, however, we have indirect markers, which allow us
to infer how far patients are adapted to everyday life. As indicated in Table 12.6 it
is not the group with FLE but the one with mTLE which is less educated, and the
job situation is comparable in both groups.
179 Behavioural and neuropsychological aspects of FLE
Organic personality change
Offends others
Receptive problems
Perseverative
Misunderstandings
Circumstantial
Irritates others
0% 10% 20% 30%
18%
23%
26%
33%
34%
51%
% deviant ratings > 75% percentile as compared
with healthy controls
Figure 12.4. Items extracted out of the organic personality change scale of the clinical personality
inventory. Bars represent the per cent of patients with focal epilepsies reporting increased
problems in communication and interpersonal contact.
Trait or state
As shown above, patients with FLE have behavioural disorders which appear very
mild when compared with those reported in patients with frontal mass lesions. With
respect to mood disorders, they appear less aected than patients with TLE, and they
alsoshowbetter academic achievement. The nding that hyperactivity, addictionand
obsession might be a behavioural feature of FLE is of great interest, and can be dis-
cussed as reecting frontal dysfunction in general and to be in line with the behavi-
our observed in neuropsychological examination and during seizures. The question
which remains open, is how consistent the behaviour in focal epilepsies is over time.
We can not yet give a conclusive answer to this question on the basis of long-term
follow-up observations. The impact of epilepsy and seizures on behaviour,
however, can be estimated by comparisons of patients who after surgery still have
seizures with those who became completely seizure-free. We therefore analysed
data from surgically and nonsurgically treated patients who participated in a long-
term follow-up study, which was originally designed to show the cognitive devel-
opment of these patients over time (Helmstaedter et al., 2000b). However, at the
time of the long-term follow-up visit we also assessed depression by use of the BDI
and quality of life by use of a German modied QOLIE10. For the present purpose
we extracted from the total database only the data of the patients with temporo-
mesial epilepsy and hippocampal sclerosis as compared with those with FLE. Fifty-
seven patients had mTLE with hippocampal sclerosis (27 had surgery, 20 were
treated conservatively) and 30 patients had FLE (16 had surgery and 14 were treated
conservatively). Taking depression and quality of life measures as the dependent
variables in a multivariate analysis with consideration of surgery, localization and
lateralization of epilepsy as independent variables, and age and the follow-up inter-
val (mean 56 months; 210 years) as covariates, seizure outcome turned out to be
the only signicant predictor. Only 14% of the seizure-free patients in contrast to
51% of those who still had seizures showed elevated depression scores greater than
180 C. Helmstaedter
Table 12.6. Academic achievement, employment and epilepsy
Academic achievement level %
No regular Low
a
Medium High
a
Percentage
school (Hauptschule) (Realschule) (Gymnasium) employed
FLE (n18) 17 22 22 39 68
mTLE (n83) 10 54 21 15 59
Note:
a
Chi-square: signicant dierence.
the cut-o score of 12 points. It should be noted that 14% is much less than the
normally reported 30% of patients with focal epilepsy and depressive mood, and
that 51% clearly exceed this number. Comparably, 45% of the seizure-free patients
reported good quality of life with QOLIE10, as compared to only 11% of the
patients who continued to have seizures. Although these data are not follow-up
data and although depression and quality of life represent only two facets of the
whole range of behaviour, these data show quite impressively what a dierence the
presence or absence of seizures can make. The nding parallels recent ndings in
children who after successful epilepsy surgery show marked improvement in
behaviour disorders (Lendt et al., 2000). Long-term follow-up studies on person-
ality and behaviour disorders are thus badly needed to complete our understand-
ing of the interaction between brain damage, epilepsy and behaviour.
Conclusion
We can conclude that in FLE, frontal dysfunctions can be suggested which char-
acteristically become evident in cognition, seizures and behaviour. The main
common feature of the behavioural problems met in FLE concerns behaviour
control in terms of response selection/initiation and inhibition. The domains in
which these problems become apparent may vary with clinical features. Following
our own ndings, hyperactivity, conscientiousness, obsession and addiction can be
seen as reecting frontal lobe dysfunction in frontal lobe epilepsy. Depression,
anxiety, neuroticism, cognitive (memory) impairment and social limitations in
contrast seem more a feature of mTLE. However, methodological diculties
regarding the adequacy of the clinical measures in use as well as confounding eects
of lesions, epileptic dysfunction, antiepileptic drugs and psychosocial status do not
yet allow further distinctions such as can be made, for example, in neurobiological
models about the frontal lobes and behaviour. Full-blown personality disorders are
very rare in FLE and symptoms appear rather mild when compared with patients
with mass lesions. As regards the state/trait discussion in epilepsy, the eects of
seizure control indicate that a great part of the observed behavioural problems is
indeed state-dependent. However, follow-up evaluations are badly needed to
partial out the contribution of lesions and epileptic dysfunctions to behaviour dis-
orders and to demonstrate how far these are reversible or not.
REFERENCES
Alivisatos, B. and Milner, B. (1989). Eects of frontal or temporal lobectomy on the use of
advance in information in a choice reaction time task. Neuropsychologia, 27, 495503.
181 Behavioural and neuropsychological aspects of FLE
Anderson, S.W., Bechara, A., Damasio, H. et al. (1999). Impairment of social and moral behav-
ior related to early damage in human prefrontal cortex. Nat Neurosci, 2, 10327.
Azouvi, P., Jokic, C., Attal, N., Denys, P., Markabi, S. and Bussel, B. (1999). Carbamazepine in
agitation and aggressive behavior following severe closed head injury: results of an open trial.
Brain Injury, 13, 797804.
Baddeley, A.D. and Hitch, G. (1974). Working memory. In Recent Advances in Motivation and
Learning, Vol. 8, ed. G.A. Bower, pp. 4790. New York: Academic Press.
Bauer, J., Helmstaedter, C. and Elger, C.E. (1997). Nonconvulsive status epilepticus with gener-
alized fast activity. Seizure, 6, 6770.
Bear, D., Levin, K., Blumer, D., Chetham, D. and Ryder, J. (1982). Interictal behaviour in hospi-
talized temporal lobe epileptics: relationship to idiopathic psychiatric syndromes. J Neurol
Neurosurg Psychiatry, 45, 4818.
Bearn, R.G. and Gibson, R.J. (1998). Aggressive behavior in intellectually challenged patients
with epilepsy treated with lamotrigine. Epilepsia, 39, 2802.
Bechara, A., Damasio, H., Damasio, A.R. and Lee, G.P. (1999). Dierent contribution of the
human amygdala and ventromedial prefrontal cortex to decision making. J Neurosci, 19,
547381.
Bechara, A., Damasio, H. and Damasio, A.R. (2000). Emotion, decision making and the orbito-
frontal cortex. Cereb Cortex, 10, 295307.
Bechara, A., Damasio, H., Tranel, D. and Anderson, S.W. (1998). Dissociation of working
memory from decision making within the human prefrontal cortex. J Neurosci, 18, 42837.
Beck, A.T., Rush, A.J., Shaw, B.F. and Emery, G. (1981). Kognitive Therapie der Depression.
Mnchen: Urban und Schwarzenberg.
Blumer, D. (2000). Dysphoric disorders and paroxysmal aects: recognition and treatment of
epilepsy related disorders. Harv Rev Psychiatry, 8, 817.
Boone, K.B., Miller, B.L., Rosenberg, L., Durazo, A., McIntyre, H. and Weil, M. (1988).
Neuropsychological and behavioural abnormalities in an adolescent with frontal lobe seizures.
Neurology, 38, 5836.
Cavada, C., Compay, T., Tejedor, J., Cruz-Rizzolo, R.J. and Reinoso-Surez, F. (2000). The ana-
tomical connections of the macaque monkey orbitofrontal cortex. A review. Cereb Cortex, 10,
22042.
Costa, P.T. and McCrea, R.R. (1992). Revised NEO Personality Inventory (NEO PI-R) and the NEO
Five Factor Inventory. Professional Manual. Odessa, FL: Psychological Assessment Resources.
Cramer, J.A., Perrine, K., Devinsky, O. and Meador, K. (1996). A brief questionnaire to screen
quality of life in epilepsy the QOLIE10. Epilepsia, 37, 57782.
Damasio, A.R. (1996). The somatic marker hypothesis and the possible functions of the prefron-
tal cortex. Philos Trans R Soc Lond B Biol Sci, 351, 141320.
Damasio, H., Grabowski, T., Frank, R., Galaburda, A.M. and Damasio, A.R. (1994). The return
of Phineas Gage: clues about the brain from the skull of a famous patient. Science, 20/264,
11025.
Davidson, R.J., Putnam, K.M. and Larson, C.L. (2000). Dysfunction in the neural circuitry of
emotion regulation a possible prelude to violence. Science, 289, 5914.
Delaney, R.C., Rosen, A.J., Mattson, R.H. and Novelly, R.A. (1980). Memory function in focal
182 C. Helmstaedter
epilepsy: a comparison of non-surgical, unilateral temporal lobe and frontal lobe samples.
Cortex, 16, 10317.
Devinsky, O. and Najjar, S. (1999). Evidence against the existence of a temporal lobe syndrome.
Neurology, 53, S1325.
Devinsky, O., Morrell, M.J. and Vogt, B.A. (1995). Contributions of anterior cingulate cortex to
behavior. Brain, 118, 279306.
Dolan, R. (1999). On the neurology of morals. Nature Neurosci, 2, 9279.
Drevets, W.C. and Raichle, E. (1996). Positron emission tomographic imaging studies in human
emotional disorders. In The Cognitive Neurosciences, ed. M.S. Gazzaniga, pp. 115362.
Massachusetts: MIT Press.
Elger, C.E. and Lehnertz, K. (1998). Seizure prediction by non-linear time series analysis of brain
electrical activity. Eur J Neurosci, 10, 7869.
Goldberg, E. and Podell, K. (2000). Adaptive decision making, ecological validity, and the frontal
lobes. J Clin Exp Neuropsychol, 22, 5668.
Harlow, J.M. (1868). Recovery from passage of an iron bar through the head. Mass Med Soc Publ,
2, 32746.
Helmstaedter, C. and Elger, C.E. (1994). Cognitive-behavioral aspects of quality of life in presur-
gical patients with epilepsy. J Epilepsy, 7, 22031.
Helmstaedter, C., Elger, C.E. and Lendt, M. (1994). Postictal courses of cognitive decits in focal
epilepsies. Epilepsia, 35, 10738.
Helmstaedter, C., Kemper, B. and Elger, C.E. (1996). Neuropsychological aspects of frontal lobe
epilepsy. Neuropsychologia, 34, 399406.
Helmstaedter, C., Gleiner, U., Zentner, J. et al. (1998). Neuropsychological consequences of epi-
lepsy surgery in frontal lobe epilepsy. Neuropsychologia, 36, 33341.
Helmstaedter, C., Gleiner, U. and Elger, C.E. (2000a). Clinical Personality Scales (CPS) in focal
epilepsy: preliminary results. Epilepsia, 42 (Suppl.), 236.
Helmstaedter, C., Kurthen, M., Lux, S. et al. (2000b). Long-term clinical, neuropsychological, and
psychosocial follow-up in surgically and nonsurgically treated patients with drug-resistant
temporal lobe epilepsy. Nervenarzt, 71, 62942.
Hermann, B.P., Wyler, A.R. and Richey, E.T. (1988). Wisconsin card sorting test performance in
patients with complex partial seizures of temporal lobe origin. J Clin Exp Neuropsychol, 10,
46776.
Jokeit, H., Seitz, R.J., Markowitsch, H.J., Neumann, N., Witte, O.W. and Ebner, A. (1997).
Prefrontal asymmetric interictal glucose hypometabolism and cognitive impairment in
patients with temporal lobe epilepsy. Brain, 120, 228394.
Kalynchuk, L.E., Pinel, J.P. and Treit, D. (1999). Characterization of the defensive nature of -
kindling-induced emotionality. Behav Neurosci, 113, 76675.
Kemper, B., Helmstaedter, C. and Elger, C.E. (1992). Kognitive Prole von prchirurgischen
Patienten mit Frontal- und Temporallappenepilepsie. In Epilepsie 91, ed. D. Schener, pp.
34550. Reinbeck: Einhorn Presse Verlag.
Lendt, M., Helmstaedter, C., Kuczaty, S., Schramm, J. and Elger, C.E. (2000). Behavioural disor-
ders in children with epilepsy: early improvement after surgery. J Neurol Neurosurg Psychiatry,
69, 73944.
183 Behavioural and neuropsychological aspects of FLE
London, E.D., Ernst, M., Grant, S., Bonson, K. and Weinstein, A. (2000). Orbitofrontal cortex
and human drug abuse: functional imaging. Cereb Cortex, 10, 33442.
Lux, S., Helmstaedter, C., Kurthen, M., Hartje, W. and Elger, C.E. (2000). Localizing value of ictal
neuropsychological decits in focal epilepsies. Epilepsia, 41 (Suppl. 7), 153.
Marsh, L. and Krauss, G.L. (2000). Aggression and violence in patients with epilepsy. Epilepsy
Behav, 1, 1608.
Milner, B. (1964). Some eects of frontal lobectomy in man. In The Frontal Granular Cortex, ed.
J.M. Warren and K. Akert, pp. 31334. New York: McGraw-Hill.
Milner, B. (1995). Aspects of human frontal lobe function. In Advances in Neurology, Vol. 66, ed.
H.H. Jasper, S. Riggio and P.S. Goldman-Rakic, pp. 6784. New York: Raven Press.
Northo, G., Richter, A., Gessner, M. et al. (2000). Functional dissociation between medial and
lateral prefrontal cortical spatiotemporal activation in negative and positive emotions: a com-
bined fMRI/MEG study. Cereb Cortex, 10, 93107.
Petrides, M. (1985). Decits on conditional associative learning tasks after frontal- and temporal
lobe lesions in man. Neuropsychologia, 5, 60114.
Petrides, M. and Milner, B. (1982). Decits on subject-ordered tasks after frontal and temporal
lobe lesions in man. Neuropsychologia, 3, 24962.
Quiske, A., Helmstaedter, C., Lux, S. and Elger, C.E. (2000). Depression in patients with temporal
lobe epilepsy is related to mesial temporal sclerosis. Epilepsy Res, 39, 1215.
Raine, A., Lencz, T., Bihrle, S., LaCasse, L. and Coletti, P. (2000). Reduced prefrontal gray matter
volume and reduced autonomic activity in antisocial personality disorder. Arch Gen Psychiatry,
67, 11927.
Roberts, A.C. and Wallis, J.D. (2000). Inhibitory control and aective processing in the prefron-
tal cortex: neuropsychological studies in the common marmoset. Cereb Cortex, 10, 25262.
Rolls, E.T. (2000). The orbitofrontal cortex and reward. Cereb Cortex, 10, 28494.
Rose, K.J., Derry, P.A. and McLachlan, R.S. (1996). Neuroticism in temporal lobe epilepsy: assess-
ment and implications for pre- and postoperative psychosocial adjustment. Epilepsia, 37,
48491.
Rubia, K., Overmeyer, S., Taylor, E. et al. (2000). Functional frontalization with age: mapping
neurodevelopmental trajectories with fMRI. Neurosci Biobehav Rev, 24, 1319.
Sarazin, M., Pillon, B., Giannakopoulos, P., Rancurel, G., Samson, V. and Dubois, B. (1998).
Clinicometabolic dissociation of cognitive functions and social behavior in frontal lobe
lesions. Neurology, 51, 1428.
Scherrmann, J.M. and Elger, C.E. (1999). Generation of seizure phenomena in the frontal and
temporal lobe. Epilepsia, 40 (Suppl. 7), 102.
Schmitz, B. (1999). Psychiatric syndromes related to antiepileptic drugs. Epilepsia, 40 (Suppl. 10),
6570.
Shulman, M.B. (1984). The frontal lobes, epilepsy, and behavior. Epilepsy Behav, 1, 38495.
Smith, M.L. and Milner, B. (1984). Dierential eects of frontal-lobe lesions on cognitive esti-
mation and spatial memory. Neuropsychologia, 22, 697705.
Stuss, T.D. and Benson, T. (1986). The Frontal Lobes. New York: Raven Press.
Trimble, M.R. and Van Elst, L.T. (1999). On some clinical implications of the ventral striatum
and the extended amygdala. Investigations of aggression. Ann NY Acad Sci USA, 29, 63844.
184 C. Helmstaedter
Upton, D. and Thompson, P.J. (1996a). General neuropsychological characteristics of frontal
lobe epilepsy. Epilepsy Res, 23, 16977.
Upton, D. and Thompson, P.J. (1996b). Epilepsy and the frontal lobes: neuropsychological char-
acteristics. J Epilepsy, 9, 21522.
Upton, D. and Thompson, P.J. (1997a). Neuropsychological test performance in frontal-lobe epi-
lepsy: the inuence of aetiology, seizure type, seizure frequency and duration of disorder.
Seizure, 6, 4437.
Upton, D. and Thompson, P.J. (1997b). Age at the onset and neuropsychological function in
frontal lobe epilepsy. Epilepsia, 38, 110313.
Upton, D. and Thompson, P.J. (1999). Twenty question tests and frontal lobe dysfunction. Arch
Clin Neuropsychol, 14, 20316.
Van Elst, L.T., Woermann, F.G., Lemieux, L., Thompson, P.J. and Trimble, M.R. (2000). Aective
aggression in patients with temporal lobe epilepsy: a quantitative MRI study of the amygdala.
Brain, 123, 23443.
Zung, W.W.K. (1971). A rating instrument for anxiety disorders. Psychosomatics, 12, 3719.
185 Behavioural and neuropsychological aspects of FLE
Part IV
Nonepileptic attacks
13
Epilepsy, dissociation and nonepileptic
seizures
Richard J. Brown
Institute of Neurology, London, UK
Introduction
Within the context of psychiatry and neurology, the term dissociation is extremely
dicult to dene satisfactorily. Since its introduction in the late nineteenth century,
the term has been applied to a wide range of neurological, psychiatric and psycho-
logical phenomena. As a result, there is considerable confusion over what actually
constitutes dissociation and the concept is frequently misapplied. This is particularly
true within the eld of epilepsy. Many epileptic phenomena have been labelled as dis-
sociative, including the sensory, aective and cognitive features of partial seizures,
behavioural automatisms, postictal amnesia and fugue (Thomas and Trimble, 1997).
Similarly, certain psychiatric phenomena that mimic epileptic events, so-called non-
epileptic seizures, have been identied as primarily dissociative in nature. Indeed,
many authorities have argued that the dissociative nature of nonepileptic seizures
could provide the basis for their conceptual and practical dierentiation from
genuine epileptic events (Kuyk et al., 1999). If, however, dissociation is experienced
both by individuals with epilepsy and by those with pseudo-epileptic seizures, how
can its occurrence aid in the dierential diagnosis of these conditions?
In this chapter, I will explore what we mean by dissociation and how it relates to
the phenomena of epilepsy and nonepileptic seizures. In this way, I aim to demon-
strate that (i) many epileptic phenomena previously labelled as dissociative should
not be regarded as dissociative at all; (ii) dissociation is a fundamental aspect of
nonepileptic seizures; (iii) those epileptic phenomena that are genuinely dissocia-
tive involve a dierent type of dissociation to that involved in nonepileptic seizures;
and (iv) the concept of dissociation needs to be used far more precisely if it is to
help distinguish between epileptic and nonepileptic seizures.
What is dissociation?
Cardea (1994) has described a useful taxonomy that captures the dierent ways in
which the concept of dissociation has been used. According to this scheme, there
189
are three major facets to the dissociation construct: (i) dissociation as noncon-
scious or nonintegrated mental modules or systems; (ii) dissociation as an altera-
tion in consciousness; and (iii) dissociation as a defence mechanism.
Dissociation as nonintegrated mental modules or systems
This is probably the most widely used denition of dissociation and is reected in
the so-called dissociative disorders categories of current psychiatric taxonomies.
According to DSMIV, the essential feature of the dissociative disorders is a dis-
ruption in the usually integrated functions of consciousness, memory, identity, or
perception of the environment (p. 477; American Psychiatric Association, 1994).
Dissociation in this sense reects the original meaning of the term dsagrgation,
oered by Janet as a label for the putative process whereby previously integrated
memories can become inaccessible to consciousness and control behaviour in a way
that the individual is unaware of (Janet, 1889, 1924). According to Janet, such a
process is the basic psychopathological mechanism underlying so-called hysteri-
cal symptoms, phenomena that have now been reclassied within the dissociative
and somatoform disorders categories in DSMIV and ICD10. The DSMIV dis-
sociative disorders category encompasses dissociative amnesia, dissociative fugue,
dissociative identity disorder (formerly multiple personality disorder), deperson-
alization disorder and dissociative disorder not otherwise specied (see Table 13.1).
A slightly broader denition is oered in ICD10, which identies the loss of
control over bodily movements as an additional dissociative phenomenon (World
Health Organization, 1992). As such, the ICD10 dissociative (conversion) disor-
ders category encompasses dissociative amnesia, dissociative fugue, dissociative
motor disorders, dissociative convulsions, dissociative anaesthesia and sensory
190 R.J. Brown
Table 13.1. Classification of dissociative disorders in ICD10 and DSMIV
ICD10 dissociative (conversion) disorders DSMIV dissociative disorders
Dissociative amnesia Dissociative amnesia
Dissociative fugue Dissociative fugue
Dissociative motor disorders Dissociative identity disorder
Dissociative convulsions Depersonalization disorder
Dissociative anaesthesia and sensory loss Dissociative disorder not otherwise specied
Dissociative stupor
Trance and possession disorders
Mixed dissociative (conversion) disorders
Other dissociative (conversion) disorders
Dissociative (conversion) disorder, unspecied
loss, dissociative stupor, mixed dissociative (conversion) disorders, other dissocia-
tive (conversion) disorders and dissociative (conversion) disorders, unspecied
(see Table 13.1). The somatoform and dissociative disorders are related in that these
phenomena are characterized by symptoms that, on the face of it, resemble those
that occur in certain physical conditions, but which are presumed to be psycholog-
ical in origin. Dissociative and somatoform phenomena dier in that the symptoms
of the former resemble those of neurological illness, while the symptoms of the
latter are more akin to those encountered in internal medicine. As Kihlstrom
(1994) has cogently argued, all of the phenomena identied as dissociative disor-
ders within DSMIV and ICD10 are linked by the fact that each has a temporary
disruption in consciousness or volition as its primary dening feature.
The dierences between DSMIV and ICD10 in their classication of the dis-
sociative and somatoform disorders are readily apparent. First, unlike ICD10,
DSMIV places nonepileptic attacks in the somatoform rather than the dissocia-
tive disorders category, along with other so-called conversion phenomena, such as
unexplained motor and sensory symptoms, that are identied as dissociative in
ICD10. This dierence is more practical than conceptual, with DSMIV placing
greater emphasis on the importance of excluding physical illness in the dierential
diagnosis of these phenomena (American Psychiatric Association, 1994). Second,
unlike DSMIV, ICD10 does not identify depersonalization as a dissociative phe-
nomenon, due to the lack of any signicant loss of control over sensation, memory
or movement in this condition, and its limited eect on personal identity. Third,
DSMIV identies a distinct category for multiple personality disorder, relabelled
dissociative identity disorder in the latest edition of this scheme. In contrast,
ICD10 places multiple personality disorder in the other dissociative (conversion)
disorders category, reecting controversy over whether this syndrome is iatrogenic
or culturally bound to North America. Inconsistencies aside, both DSMIV and
ICD10 explicitly state that physical conditions such as epilepsy should be excluded
in the dierential diagnosis of the dissociative and somatoform disorders.
In addition to the dissociative and somatoform disorders, Cardea (1994) iden-
ties other pathological phenomena characterized by a lack of integration between
mental modules or systems that are caused by neurological rather than psychiatric
events. Blindsight, a rare condition in which the suerer displays above-chance
visual discrimination despite reporting a lack of visual experience, provides one
example of how normally integrated functions can become dissociated through
neurological damage. Many of the unusual behaviours often displayed by patients
following commissurotomy also fall within this category, as do those exhibited by
individuals suering from hemi-neglect. In each of these cases, the dissociation is
between the individuals ongoing behaviour and their introspective verbal report.
Neurological phenomena of this sort are important because they demonstrate the
191 Seizures, epilepsy and dissociation
distributed, semimodular (Fodor, 1983; Kosslyn, 1994) organization of brain,
mind, consciousness and behaviour.
Neurological dissociations such as blindsight are supercially analogous to those
observed in psychiatric instances of dissociation, such as the preservation of
implicit perception
1
in the context of dissociative blindness (Kihlstrom, 1994).
Neurological and psychiatric dissociation dier, however, in that the former is often
permanent, reecting irreversible damage to the underlying neurological sub-
systems in question (Kihlstrom, 1994). Psychiatric instances of dissociation, in
contrast, are thought to be the product of an alteration in the parameters govern-
ing otherwise intact psychological functions; they are, therefore, reversible by de-
nition. Relatedly, neurological and psychiatric dissociations dier in that, unlike
the former, the latter involves symptoms (e.g. glove anaesthesia) that need not, and
typically do not, relate to the actual organization of the nervous system and its
many distributed components. On these grounds, it is apparent the dissociation
in these cases is an entirely dierent phenomenon and the two must not be con-
fused.
The idea that normally integrated psychological processes can become tempor-
arily dissociated and exist in isolation of one another has also been cited as the basis
for other, less pathological, phenomena (Cardea, 1994; Hilgard, 1977; Woody and
Bowers, 1994). Many apparently hypnotic phenomena fall within this category,
including profound amnesia, the loss of perceptual experience and complex beha-
viours characterized by a sense of involuntariness, all of which can be temporarily
produced by appropriate suggestions in certain individuals. The extent to which
similar processes are involved in these phenomena and those displayed by individ-
uals with dissociative psychopathology has been a matter of debate since the time
of Janet. Conceptually, there are good grounds to assume a common mechanism
in hypnotic and dissociative phenomena (Oakley, 1999) and recent functional
imaging evidence provides some support for a link between the two (Halligan et al.,
2000).
According to Cardea (1994), this particular denition of dissociation has also
been inappropriately applied to a number of other normal psychological phenom-
ena. Following Hilgard (1977), the execution of complex behaviours with only
minimal conscious awareness, such as the action of driving a car whilst holding a
conversation, has often been identied as a dissociative phenomenon. As Cardea
has pointed out, however, the dissociation label should not simply be applied to any
behaviour or psychological process that, for whatever reason, occurs without full
192 R.J. Brown
1
Implicit perception is evidenced when an external stimulus produces psychological eects despite not being
perceived consciously (Kihlstrom, 1994). The phenomenon of blindsight provides one such example.
Implicit perception is akin to the concept of implicit memory, in which behaviour is inuenced by learnt
information that the individual cannot consciously recall.
awareness. Such a practice ignores the fact that, in many such cases, the individual
can bring the apparently dissociated process into awareness by an act of selective
attention. Other such cases involve dissociation between systems or processes that
one would not normally expect to operate in an integrated fashion. According to
Cardea, mental modules or systems should only be regarded as truly dissociated
from one another if their dissociation is (i) in contrast to a normal state of integra-
tion; and (ii) cannot be overcome by an act of will.
Dissociation as an alteration in consciousness
A second usage of the dissociation concept refers to an altered state of conscious-
ness characterized specically by a disengagement from the self or the environment
(Cardea, 1994). As Cardea has pointed out, this sense of the dissociation concept
should not be applied to everyday phenomena, such as daydreaming and other
states of distraction, where engagement with the environment is less than complete.
Rather, it should be reserved specically for states that are regarded by the experi-
encing individual as qualitatively dierent to their normal state of awareness.
Although a number of dierent phenomena fall within the bounds of this deni-
tion (e.g. trance and possession states), probably the most commonly reported
are depersonalization and derealization. In depersonalization, the individual
experiences a profound feeling of detachment from their thoughts, perceptions,
actions and emotions, often characterized by a sense of numbness or disembodi-
ment. In derealization, the individual experiences an intact sense of self coupled
with a feeling of detachment from the external environment, which often feels
unreal or at a distance. Such feelings are extremely common, frequently occurring
in the context of psychiatric illnesses such as depression and anxiety; they also
occur as a circumscribed problem in their own right, such as in depersonalization
disorder. Although DSMIV identies depersonalization disorder as a dissociative
phenomenon, this condition clearly relates to a dierent sense of dissociation than
that which applies to the other members of this category; this dierence further jus-
ties the separation of depersonalization disorder from the dissociative disorders
category in ICD10. Depersonalization and derealization are also found in certain
drug states (e.g. those produced by marijuana, LSD and ketamine), neurological
conditions such as temporal lobe epilepsy and can occur spontaneously in the
context of stress or fatigue.
Dissociation as a defence mechanism
Finally, dissociation has been described as a defence mechanism that protects the
individual from potentially overwhelming pain or anxiety. In many respects, this
account of dissociation is indistinguishable from the Freudian concept of repres-
sion (Erdelyi, 1985). This sense of the dissociation concept is typically used to
193 Seizures, epilepsy and dissociation
describe the psychological function served by the creation of a dissociated state of
consciousness, or the dissociation of mental modules or systems from one another
(Cardea, 1994). By this view, exposure to a traumatic situation may trigger the
compartmentalization of memories, which preserves psychological integrity by
preventing the distressing material from entering consciousness after the event.
Alternatively, such traumatic exposure may spontaneously elicit a depersonalized
state that prevents extreme emotion from inhibiting an appropriate behavioural
response. As such, this denition of dissociation may relate to either of the deni-
tions described previously. In both cases, dissociation of this sort could either be
an acute response to an isolated traumatic event or a trait-like characteristic
acquired as a result of repeated exposure to trauma.
Epilepsy and dissociation
Although the dierential diagnosis of ICD10 and DSMIV dissociative disorders
explicitly requires the exclusion of symptoms with an identiable neurological
basis, many of the phenomena associated with epilepsy, particularly temporal lobe
epilepsy, have been regarded as dissociative in nature (Devinsky et al., 1989; Good,
1993). Indeed, ICD10 includes a specic category for dissociative disorders due to
a general medical condition, which encompasses many of the symptoms exhibited
by individuals with epilepsy. The absence of such a category from DSMIV,
however, reects doubt concerning the value of attaching the dissociative label to
these phenomena. In my view, such doubt is well justied in many cases.
The notion that many epileptic phenomena can be regarded as dissociative is
based, to a considerable extent, on the frequent occurrence of amnesia in epilepsy
(Good, 1993). Individuals with complex partial or generalized seizures typically
display profound amnesia for events occurring during the ictus. Moreover, certain
people experience a postictal fugue state characterized by apparently purposeful
behaviour for which they are subsequently amnesic, much like dissociative fugue.
Despite their prima facie resemblance to dissociative phenomena, however, these
events should not be regarded as episodes of dissociation (Good, 1993).
Dissociative amnesia is characterized by an inability to retrieve information that
has been learnt and is present in memory despite its inaccessibility (American
Psychiatric Association, 1994). Amnesia for ictal events, in contrast, reects a dis-
ruption in normal information processing, resulting from uncontrolled neural
activity, that prevents the encoding of new material during the ictus. The ictal
amnesia is not a product of a retrieval failure, therefore, but simply the absence of
memories to retrieve. It is for this reason that this form of amnesia is irreversible,
unlike most cases of dissociative amnesia (American Psychiatric Association,
1994).
194 R.J. Brown
Postictal fugue should not be regarded as a dissociative episode for similar
reasons. Unlike dissociative fugue, postictal fugue is characterized by a disruption
in consciousness associated with signicant confusion and an abnormal EEG
(Thomas and Trimble, 1997). The apparently purposeful behaviour displayed in
postictal fugue is not dissociated from ongoing cognitive activity as it is in dissoci-
ative fugue; rather, it occurs in the relative absence of such activity. The inability to
reverse the amnesia associated with postictal fugue serves as an illustration of this
fact. The behavioural automatisms often observed in the context of complex partial
seizures and regarded as a dissociative phenomenon by some (Good, 1993) are
amenable to a similar interpretation. Like the behaviours exhibited during postictal
fugue, ictal automatisms only occur in the context of a disruption in consciousness
and disturbed behavioural control; genuinely dissociated behaviours are note-
worthy because they occur despite an otherwise intact ability to control action
2
. In
both cases, it is likely that these behaviours result from the uncontrolled activation
of circumscribed motor programs by epileptic discharges in neural sites associated
with behavioural control. The fact that such automatisms are particularly charac-
teristic of seizures originating in the frontal lobes lends support to this view.
Many of the phenomena associated with partial seizures originating in the tem-
poral lobes, such as hallucinations, sensory and cognitive auras, dj vu and dj
veu should also be distinguished from true episodes of dissociation. Although hal-
lucinations and auras have a phenomenology that departs from external reality,
these phenomena involve the paradoxical integration of information within con-
scious awareness. As such, they may be more appropriately regarded as phenom-
ena of association rather than dissociation (Kihlstrom, 1994). The phenomenology
of epileptic hallucinations and auras probably originates in the activation of repre-
sentational structures in the temporal lobes, either directly by seizure activity in
representational networks, or indirectly through seizure-related stimulation of
limbic structures such as the amygdala and anterior cingulate (Bancaud et al.,
1994). Experiences of epileptic dj vu and dj veu are also more associative than
dissociative and may involve a similar neurophysiological process. For example,
stimulation of the amygdala by seizure-related discharges could imbue current per-
ceptions and cognitions with an unwarranted emotional colouring that may be
experienced as a sense of having encountered the situation before (Bancaud et al.,
1994; Sierra and Berrios, 1998).
Certain phenomena associated with temporal lobe epilepsy can, however, be
regarded as genuine examples of dissociation according to the scheme described by
Cardea (1994). Depersonalization and derealization commonly occur in the
195 Seizures, epilepsy and dissociation
2
Dissociated behaviours should be distinguished from normal automatic behaviours because the latter are
generally in accord with system goals and can therefore be considered voluntary (Brown, 1999; see also
Cardea, 1994).
context of temporal lobe epilepsy and involve an alteration in consciousness char-
acterized by dissociation from the self and/or the environment. According to Sierra
and Berrios (1998), depersonalization and derealization are the products of a ves-
tigial defence mechanism evolved to provide the optimum processing conditions
for adaptive behaviour in the face of threat. By this view, extreme anxiety triggers
an inhibitory response from the left prefrontal cortex that dampens output from
the sympathetic nervous system, through inhibition of the amygdala and anterior
cingulate. In turn, the right prefrontal cortex is activated by ascending arousal
systems controlled by uninhibited amygdala circuits, generating further inhibition
of the cingulate. As a result, the individual experiences a sense of vigilant alertness
devoid of any emotional or cognitive content, a state that is ideally adapted for the
control of action in the face of extreme and potentially debilitating danger. If this
response is triggered in the absence of threat, however, the resulting sense of deper-
sonalization and derealization can itself be highly unpleasant and incapacitating.
Given the validity of this account, depersonalization and derealization in the
context of temporal lobe epilepsy may be the result of seizure activity in the amyg-
dala that prevents the emotional tagging of perceptual and cognitive information
prior to its entry into conscious awareness
3
. It may be that this process is the
product of transient disconnection between the amygdala and sensory areas result-
ing directly from seizure activity. Alternatively, it may reect an indirect defensive
response in the face of anxiety elicited by seizure-based stimulation of the amyg-
dala. Intuitively, one suspects that the former is the more plausible possibility,
although the latter cannot be ruled out a priori. Following this account of deper-
sonalization and derealization, these phenomena can be regarded as dissociative in
sense (ii) of the term; whether they should, in the context of epilepsy, be regarded
as the result of a dissociative defence mechanism remains an empirical issue.
The fact that few epileptic phenomena can be regarded as dissociative in any
strict sense reects widespread confusion over what actually constitutes dissocia-
tion. Although widely endorsed, the idea that any breakdown in memory, con-
sciousness, identity, perception or behavioural control is dissociative overextends
the term and diminishes its descriptive validity (Cardea, 1994). Amnesia cannot
be considered genuinely dissociative unless it involves an inability to retrieve intact
information that should, under normal circumstances, be available for recall
(Kihlstrom, 1994). Amnesia resulting from a failure to encode information, includ-
ing that which occurs in the context of epilepsy, does not fall within this category.
Loss of behavioural control can only be considered dissociative if it is within the
context of an otherwise intact ability to control action. Seizure-related motor phe-
196 R.J. Brown
3
Such a process could also be responsible for the paradoxical sense of unfamiliarity that characterizes jamais
vu, a phenomenon commonly observed in epilepsy.
nomena, including complex automatisms, are not dissociative because they occur
only in the context of reduced behavioural control in general. Current psychiatric
taxonomies do not make these distinctions clearly enough, relying instead on a
purely descriptive approach that precludes precise classication based on the
mechanisms underlying dierent phenomena.
Nonepileptic seizures and dissociation
Epileptologists frequently encounter patients who present with paroxysmal events
that, despite resembling epileptic episodes, are actually nonepileptic. Indeed, as
many as 50% of patients referred to specialist epilepsy centres may turn out not to
have epilepsy (Francis and Baker, 1999). While some nonepileptic seizures may be
attributable to physical causes other than epilepsy (see Gates and Erdahl, 1993), a
demonstrable organic basis is absent in many such cases. Of these, some are attrib-
utable to an identiable psychiatric illness, such as psychosis, that can produce
seizure-like symptoms. In other cases, however, nonepileptic seizures
4
occur as an
isolated psychiatric problem in their own right. Identifying such cases represents a
considerable challenge to neurologists working within this domain. At present,
there are few, if any, reliable criteria for an inclusive diagnosis of nonepileptic attack
disorder. As a result, current diagnostic practice is essentially based on the exclu-
sion of epilepsy and other physical disorders (see Brown and Trimble, 2000). The
concept of dissociation is particularly important in relation to nonepileptic attacks
as it sheds light on both the mechanisms and, potentially, the dierential diagnosis
of these phenomena.
Although nonepileptic seizures have a presentation as diverse as that associated
with epilepsy itself, it is possible to impose some order on their general semiology.
In a sample of 110 patients with well-documented nonepileptic attacks, Meierkord
et al. (1991) report that approximately one-third of all cases involved a collapse
with limpness, while two-thirds involved prominent motor activity such as limb
thrashing. In this study, decreased responsivity to verbal stimulation was evident in
three-quarters of all cases, while purposeful or semi-purposeful motor behaviour
197 Seizures, epilepsy and dissociation
4
Much has been written about the relative merits of the various terms used to describe these phenomena.
Many have argued that terms such as pseudoseizures are pejorative because they imply that the events in
question are not subjectively compelling (Betts, 1990). In contrast, the neutral alternative term nonepilep-
tic seizures has been criticized for being imprecise, as it fails to distinguish between the many dierent types
of paroxysmal events that are nonepileptic (Kuyk et al., 1999). My own view is that the term somatoform
seizures would be more appropriate than either of the above, as it is nonpejorative, has descriptive preci-
sion and emphasizes the importance of excluding physical illness in the dierential diagnosis of these
events. For the sake of descriptive continuity, however, the terms nonepileptic seizures and nonepileptic
attacks will be adopted here; the term nonepileptic attack disorder will be used to describe the condition
characterized by these events. In the present context, it should be assumed that the term refers specically
to those events that cannot be attributed to either an organic cause or an identiable psychiatric illness.
was observed in 44%. Just over two-thirds of all cases presented with stereotyped
seizures. These observations are broadly consistent with those reported by Betts
and Boden (1992). In this study, three types of emotional nonepileptic attacks
were identied in addition to those deliberately simulated for primary or secondary
gain and those attributable to a recognizable psychiatric attack disorder (e.g. panic
disorder). Approximately 21% of these individuals experienced so-called swoon
attacks, in which the individual characteristically sinks to the oor and lies inert
and unresponsive for the duration of the attack. Roughly 33% displayed so-called
tantrum attacks, involving a sudden drop to the oor followed by screaming and
thrashing of the limbs. Finally, 46% displayed so-called abreactive attacks, involv-
ing gasping, limb-thrashing, pelvic thrusting and stiening of the body with back
arching.
In all cases, nonepileptic attacks involve a temporary loss of behavioural, sensory
or cognitive control that occurs in the context of intact neuropsychological func-
tioning, as evidenced by a normal EEG during the nonepileptic ictus. The absence
of paroxysmal brain discharges serves as the principal feature that distinguishes
nonepileptic from genuine epileptic events. By itself, however, the EEG cannot
provide a completely reliable basis for the identication of epileptic and nonepilep-
tic seizures (Brown and Trimble, 2000), underlining the potential value of dissoci-
ation as a criterion for an inclusive diagnosis of nonepileptic attack disorder.
Several converging lines of evidence indicate that these events involve a dissoci-
ative psychological mechanism (see Kuyk et al., 1997). In the rst instance, nonepi-
leptic seizures are commonly found in the context of other forms of dissociative
psychopathology. Bowman (1993) and Bowman and Markand (1996) found that
the vast majority of individuals with nonepileptic attacks meet criteria for DSMIV
dissociative disorders such as dissociative amnesia, identity disturbance and deper-
sonalization. Post-traumatic stress disorder, commonly assumed to involve a dis-
sociative mechanism, was also particularly common in this group of patients
(Bowman, 1993; Bowman and Markand, 1996). Other studies have found that
nonepileptic seizures frequently occur alongside other unexplained physical symp-
toms (Krishnamoorthy et al., 2001; Meierkord et al., 1991), suggesting that they
may be one aspect of a broader tendency to express psychological distress somati-
cally, so-called somatization (Lipowski, 1968). A number of authorities have sug-
gested that dissociation is an important aspect of this phenomenon also (Nemiah,
1991). Eating disorder symptoms, which have been linked to a dissociative process
(Pettinati et al., 1985), also appear to be particularly common in patients with non-
epileptic seizures (Krishnamoorthy et al., 2001).
The frequent co-occurrence of dissociative psychopathology in patients with
nonepileptic attacks appears to indicate a general propensity for dissociative exper-
198 R.J. Brown
iences in these individuals. Consistent with this notion is a recent study by Kuyk et
al. (1999), showing that individuals with nonepileptic attacks display elevated levels
of hypnotic susceptibility. In a related vein, nonepileptic attacks can, in many such
individuals, be provoked using suggestion, placebo or hypnosis (Dericioglu et al.,
1999). High hypnotic susceptibility is commonly found in patients with dissocia-
tive psychopathology (Frischholz et al., 1992; Pettinati et al., 1985; Spiegel et al.,
1988), and a dissociative interpretation of hypnosis has been oered by a number
of authorities (Hilgard, 1977; Woody and Bowers, 1994). Bowman (1993) also
found that individuals with nonepileptic seizures yield elevated scores on the
Dissociative Experiences Scale (DES; Bernstein and Putnam, 1986), a self-report
measure assessing everyday occurrences of dissociation, compared with nonclini-
cal controls. However, in a more recent study, Alper et al. (1997) found that DES
scores are also elevated in patients with complex partial seizures (see also Devinsky
et al., 1989); indeed, there was no signicant dierence in overall DES scores
between these patients and a group with nonepileptic seizures. Nevertheless, both
epileptic and nonepileptic groups scored higher on the DES than typically observed
in nonclinical populations. In my view, this particular nding demonstrates the
danger of conating the various denitions of dissociation within a single measure
such as the DES. As the DES treats dissociation as a unitary concept, it cannot
dierentiate between conditions that are characterized by dierent forms of disso-
ciative phenomena, such as epilepsy and nonepileptic attack disorder.
It is widely thought that traumatic experiences precipitate the development of
dissociative symptoms, which serve a defensive function that protects the individ-
ual from extreme anxiety and psychological disintegration. Indeed, both DSMIV
and ICD10 make an explicit link between traumatic events and the onset of dis-
sociative symptoms. Moreover, a number of studies have found disproportionately
high rates of physical, sexual and emotional abuse in patients with dissociative dis-
orders (Chu and Dill, 1990; Irwin, 1994; Pribor et al., 1993). As such, evidence indi-
cating an increased prevalence of traumatic experiences in individuals with
nonepileptic seizures could be viewed as additional support for a dissociative inter-
pretation of this phenomenon. To this end, Bowman (1993) found that 70% and
77% of her sample of 27 nonepileptic seizure patients had experienced physical or
sexual abuse respectively. Similarly, Betts and Boden (1992) obtained positive
sexual abuse histories from 54% of 96 patients with nonepileptic seizures.
Evidence implicating high dissociative comorbidity, hypnotic susceptibility
and exposure to trauma in individuals with nonepileptic seizures provides only
indirect evidence for a dissociative interpretation of this phenomenon. Although
such evidence suggests that a tendency to dissociate may be a common feature of
these individuals, it does not constitute conclusive proof that nonepileptic attacks
199 Seizures, epilepsy and dissociation
are themselves dissociative. A recent study by Kuyk et al. (1999) places such an
interpretation on a rmer footing. Like genuine epileptic seizures, nonepileptic
attacks are often associated with a dense amnesia for events occurring during the
ictus. As I have hopefully demonstrated, genuine epileptic amnesia should not be
considered a dissociative phenomenon because it arises from a seizure-related
disruption in memory encoding rather than an inability to retrieve intact
memory traces. However, Kuyk et al. (1999) have shown that the amnesia asso-
ciated with nonepileptic attacks may actually be the product of such a retrieval
decit. Kuyk et al. (1999) compared a group of individuals with amnesia for
events occurring during well-documented nonepileptic attacks with a group dis-
playing amnesia following complex partial and generalized epileptic seizures. All
subjects were hypnotized and given suggestions designed to facilitate the recov-
ery of ictal events; the experimenter remained blind to group status at all times.
Using a free-recall paradigm, 17 out of 20 patients with nonepileptic seizures
recovered signicant information concerning the designated attack; this informa-
tion was veried by video recordings or third-party reports. In contrast, not one
of the 17 patients with epilepsy retrieved information concerning their attack
during hypnosis. Such a nding appears to demonstrate that, unlike that found
in epilepsy, nonepileptic amnesia results from a process that prevents the individ-
ual from accessing memories successfully encoded during the attack. This appar-
ent separation of intact memorial information from conscious awareness
following a nonepileptic attack, coupled with the phenomenological character of
these events, clearly identies these phenomena as dissociative in sense (i) of the
term.
Dissociative mechanisms of nonepileptic seizures
If one assumes that nonepileptic seizures are a dissociative phenomenon, what
mechanisms might be involved in the generation and maintenance of these events?
Current theorizing in this domain is still very much dominated by nineteenth
century ideas concerning the mechanisms underlying so-called hysterical
5
phe-
nomena. Indeed, the term dissociation originates in the work of Pierre Janet who
proposed one of the earliest systematic accounts of the psychological mechanisms
underlying hysteria. According to Janet (1889, 1924), a fundamental weakness in
the hysterical individuals mental character makes them susceptible to a breakdown
in the normally integrated functions of consciousness when faced by environmen-
tal stress or trauma. As a result, organized sets of knowledge pertaining to the
200 R.J. Brown
5
In recent years, the term hysteria has fallen from favour due to its unwarranted and pejorative connota-
tions. The multiple unexplained symptoms captured by this concept, of which nonepileptic seizures are
one, are now subsumed within the somatoform and dissociative disorder categories of DSMIV and
ICD10.
trauma may become dissociated from the main body of consciousness, and may
serve to take control of behaviour and experience if activated by environmental
events. The automatic activation of these dissociated memories results in a hyster-
ical reaction (or somnambulism) that, in some instances, takes the form of a non-
epileptic attack. According to this view, two aspects of the hysterical individuals
psychophysiological make-up are responsible for the processes of dissociation and
somnambulism. First, the hysterical individual possesses an abnormally high
degree of suggestibility that allows ideas from the external environment to develop
within them in the absence of their own eort or awareness. Second, the hysterical
individual suers from an attentional dysfunction or retraction of the eld of con-
sciousness (Janet, 1924, p. 314) which prevents them from entertaining alternative
states of mind, thereby accentuating their responsivity to external suggestion.
Although a century old, the theoretical analysis of hysteria oered by Janet con-
tinues to inuence theory and research concerning dissociation and the dissociative
disorders. This inuence is particularly evident in Hilgards (1977) neodissociation
theory and its conceptual derivatives (Woody and Bowers, 1994). According to
Hilgard, behaviour is controlled by the operation of a large set of functionally auton-
omous low-level cognitive control systems specialized for the execution of particular
behavioural acts. Each control system comprises an organized set of well-learned
behavioural and cognitive routines or schema
6
developed following extensive expe-
rience with the environment; these control systems are hierarchically organized
beneath an executive ego, a higher-level cognitive structure associated with volition
and consciousness. The executive ego is responsible for selecting the appropriate cog-
nitive control systems for any given task; however, in order to conserve the limited
resources of the executive, once control systems are selected they are able to function
with a considerable degree of autonomy from the ego. As such, they have become dis-
sociated from the executive, with their continued operation occurring largely outside
of awareness; in this way, well-learned behaviours can be performed eortlessly and
concurrently. According to Hilgard, such processing dissociations are fundamental
to human cognition, with the conscious representation of information being largely
unnecessary for the execution of extremely complex behaviours.
Hilgards theory provides an elegant account of the psychological mechanisms
involved in dissociation. Although it was originally developed as a general account
of the mechanisms involved in behavioural control, neodissociation theory has
been most inuential as an account of hypnotic behaviour. According to neodisso-
ciation theory, the hypnotic induction brings about a functional inhibition of the
executive ego, causing it to fractionate into two separate elements (Kirsch and
201 Seizures, epilepsy and dissociation
6
In contemporary cognitive theory, schemas are organized knowledge structures that represent the sequence
of actions and/or processing operations involved in a given behaviour e.g. the movements involved in con-
trolling a car.
Lynn, 1995). Although one part of the fractionated executive continues to function
as normal during hypnosis, the second part is concealed from awareness by the for-
mation of an amnesic barrier. This part of the ego can exert behavioural control in
the usual fashion but such control is prevented from representing itself in con-
sciousness by the amnesia (see Figure 13.1). Hypnotic behaviours result from the
selection of cognitive control systems via the part of the executive ego concealed by
the amnesia; as the hypnotized individual is aware of only the resulting behaviour
and not the process by which it was selected (due to the amnesia), they experience
the execution of hypnotic suggestions as occurring involuntarily.
The neodissociative model of hypnotic behaviours also provides the basis for an
account of more pathological phenomena such as dissociative amnesia, fugue and
multiple personality disorder (Kihlstrom, 1994; Spiegel and Cardea, 1991).
According to such a view, the formation of amnesic barriers within the executive
ego is a common defensive response in the face of trauma. These barriers serve an
adaptive function in that they protect the individual from experiencing potentially
overwhelming negative aect associated with the traumatic event. However, patho-
logical dissociative amnesia can arise if the barrier within the executive ego endures
to the point where the memory loss itself becomes distressing or debilitating. In the
case of dissociative fugue, the amnesic barrier conceals large tracts of autobio-
graphical memory as well as the traumatic events themselves. Without access to this
autobiographical information, the fugue suerer not only reports amnesia but also
202 R.J. Brown
Figure 13.1. The neodissociation account of dissociative behaviours. (Adapted from Kirsch and Lynn,
1995.)
a profound loss of personal identity (Kihlstrom, 1994). Neodissociation theory also
postulates that dierent parts of the executive ego, separated from each other by
amnesic barriers, can be activated at dierent times (e.g. during traumatic events).
If a nonprimary aspect of the ego is activated often enough or for a sucient length
of time, it may become associated with enough autobiographical information to
constitute a separate identity in its own right. Such a process could allow for the
development of multiple identities within a single individual, a phenomenon that
characterizes dissociative identity disorder. Neodissociation theory can also
account for dissociative phenomena characterized by a loss of behavioural control,
such as nonepileptic attacks. By this view, dissociative behaviours are controlled by
that part of the executive ego concealed beneath the amnesic barrier (Kirsch and
Lynn, 1995). As such, neodissociation theory would identify nonepileptic attacks
as essentially voluntary behaviours; however, they are perceived as involuntary
because the individual is amnesic for the executive control involved in their execu-
tion (Kihlstrom, 1994).
Hilgards neodissociation theory represents an important step forward in our
understanding of the dissociation concept. Perhaps most importantly, it recasts
Janets original ideas within a modern cognitive psychological framework that
bridges the gap between pathological (e.g. dissociative amnesia) and nonpatholog-
ical (e.g. hypnotic amnesia) instances of dissociation. While Janet regarded disso-
ciation as an abnormal process provoked by stress in the constitutionally weak
mind of the hysterical individual, neodissociation theory identies dissociation as
a normal psychological process that is utilized by the organism as a defence mech-
anism in the face of trauma. Pathological dissociation arises when this normally
adaptive process becomes generalized to situations where such a response is inap-
propriate. Similarly, hypnotic suggestions capitalize on the dissociative nature of
brain and mind to produce unusual and entertaining phenomena.
Despite its popularity within psychiatry and psychology, there are a number of
problems with neodissociation theory as an account of the mechanisms underly-
ing the dissociative disorders. In the rst instance, neodissociation theory assumes
that pathological dissociation results from a defensive response to trauma that has
become overgeneralized. However, not all individuals who display dissociative
symptomatology have been exposed to trauma. Second, neodissociation theory
assumes that pathological dissociation arises from the generation of an amnesic
barrier. However, amnesia is not apparent in all cases of dissociative symptomatol-
ogy. Many individuals with nonepileptic seizures, for example, are not amnesic for
the loss of behavioural control that they experience during the ictus. Third, neo-
dissociation theory assumes that similar mechanisms underlie both pathological
and nonpathological dissociations, such as those observed during hypnosis.
203 Seizures, epilepsy and dissociation
However, as Woody and Bowers (1994) have pointed out, the amnesic-barrier
concept that forms the basis of neodissociation theory does not provide a good
explanation of hypnotic behaviours and experiences. If such a notion were correct,
it would imply that all hypnotic phenomena involve some degree of spontaneous
amnesia. This is, in fact, extremely uncommon. One would also expect hypnotic
(i.e. suggested) amnesia to be far more common than is actually the case. Moreover,
many individuals are unable to experience hypnotic amnesia despite being respon-
sive to other hypnotic suggestions.
More recently, Woody and Bowers (1994) have oered an alternative account of
the dissociative mechanisms underlying hypnotic behaviours and experiences.
Intrinsic to neodissociation theory is the idea that dissociative behaviours are con-
trolled by part of an executive system that is concealed beneath an amnesic barrier.
In this sense, neodissociation theory is based on the idea of dissociated experience;
that is, the individual experiences a loss of behavioural control that is, in fact, illu-
sory
7
(Woody and Bowers, 1994). Woody and Bowers (1994) have proposed an
alternative account based on the concept of dissociated control. By this view, the dis-
sociation is between the executive ego and the lower-level systems controlling beha-
viour, rather than within the executive itself. According to Woody and Bowers
(1994), this dissociation between higher- and lower-level cognitive control results
from an inhibition of the executive system due to the induction of hypnosis. In this
model, dissociated behaviours are generated by the automatic activation of lower-
level behavioural routines by environmental cues (see Figure 13.2). As such, disso-
ciated behaviours are genuinely involuntary, because they are not controlled by the
executive ego.
Although originally developed to account for the dissociative phenomena
observed in hypnosis, dissociated control theory could also be applied to more
pathological instances of dissociation, such as nonepileptic attacks. By this view,
nonepileptic attacks would be the product of a learned behavioural routine trig-
gered automatically (i.e. without input from the executive control system) by cues
from the environment. The apparently involuntary and stereotyped nature of non-
epileptic attacks (see Meierkord et al., 1991) certainly appears to indicate that this
phenomenon is controlled by the activation of a low-level behavioural routine or
schema. If this were the case, the semiology of any given nonepileptic attack would
correspond to the nature of the behavioural representation underlying it. In addi-
tion to behavioural features (e.g. those representing the movements involved in the
seizure), this schema could also involve a cognitive control component that serves
to inhibit the operation of the executive system and the information that is passed
204 R.J. Brown
7
This is based on the notion that any behaviour controlled by the executive ego is volitional, regardless of
whether the executive is concealed beneath an amnesic barrier. See Brown (1999) for a fuller discussion.
to it by lower-level systems. This could account not only for the behavioural aspect
of nonepileptic seizures but also for the loss of behavioural control and disruption
in subjective awareness that characterizes many such events.
Such an account of the mechanisms involved in nonepileptic seizures raises two
important questions. First, what are the origins of the behavioural routines under-
lying these events? The fact that individuals with nonepileptic events have often
been exposed to epilepsy (either in themselves or others) suggests that some sort of
behavioural modelling is involved in this process. Alternatively, it may be that non-
epileptic attacks are a learned behavioural analogue of a biological reaction such as
the sham death reex (Kretschmer, 1926), perhaps triggered previously as an acute
response to threat.
Second, what are the environmental cues that trigger the nonepileptic event? If
one assumes that nonepileptic seizures serve something of a defensive function, it
is likely that a common precipitant of nonepileptic attacks will be feelings of
anxiety, or thoughts, images or memories associated with such feelings. In princi-
ple, however, triggers could be anything that is associated in memory with the
behavioural representation underlying the attack; these are likely to vary from indi-
vidual to individual depending largely on the circumstances surrounding the initial
occurrence of the attacks.
A dissociated control account of nonepileptic seizures has all the advantages of
neodissociation theory and fewer problems. Like neodissociation theory, it can
account for the basic features of nonepileptic seizures, but does so without recourse
205 Seizures, epilepsy and dissociation
Figure 13.2. The dissociated control theory account of dissociative behaviours. (Adapted from Kirsch
and Lynn, 1995.)
to the much-criticized amnesic barrier notion (Brown, 1999; Woody and Bowers,
1994). Moreover, unlike neodissociation theory, a dissociated control account of
nonepileptic seizures would identify them as genuinely involuntary behaviours.
Such an idea is more intuitively appealing than the neodissociation theory view that
dissociative behaviours are volitional but erroneously perceived as involuntary.
Indeed, the idea that an individual can execute a behaviour deliberately but not be
aware of it (even though they may be paying it full attention) seems something of
a contradiction in terms.
One of the problems with dissociated control theory in its current form is that
the direct activation of low-level cognitive systems by environmental cues is actu-
ally a fundamental aspect of routine behavioural control (Brown, 1999; Kirsch and
Lynn, 1997). This is well illustrated by everyday action-slips (Reason, 1979), such
as dialling a familiar but out-of-date telephone number despite being aware that it
is no longer valid. As such, one need not assume that executive inhibition is neces-
sary for the occurrence of dissociative behaviours. This is entirely consistent with
the present account of nonepileptic attacks, which posits that executive inhibition
occurs after the activation of a low-level behavioural schema (i.e. as part of the
schema itself), not before it.
Summary
Dissociation is a complex and multifaceted concept that is frequently misapplied
within the eld of epilepsy. In this chapter, I have explored the various components
of the dissociation concept and how they relate to the phenomena of epilepsy and
nonepileptic seizures. I have demonstrated why many epileptic phenomena often
thought to be instances of dissociation, such as amnesia, postictal fugue, behav-
ioural automatisms, auras and hallucinations, should not be regarded as dissocia-
tive at all. I have also argued, however, that depersonalization and derealization
occurring in the context of epilepsy can be regarded as genuinely dissociative, in
the sense that they involve an altered state of consciousness characterized by disen-
gagement from the self or environment. I have also presented evidence indicating
that nonepileptic attacks should be considered a dissociative phenomenon, in this
case involving a temporary disruption in behavioural control and subjective aware-
ness despite intact neuropsychological functioning. Although certain aspects of
epilepsy are dissociative, therefore, they do not involve the same type of dissocia-
tion as that underlying nonepileptic attacks.
If the dissociation concept is to prove useful in this area, much greater precision,
both conceptual and methodological, is required. Researchers and clinicians
should be explicit about which denition of dissociation they are referring to and
eorts should be made to construct measures of dissociation that are, unlike the
206 R.J. Brown
DES, phenomenon-pure. Assessing the reversibility of nonepileptic amnesia,
which may prove invaluable as an aid to dierential diagnosis in this area, provides
one illustration of the potential utility of such an endeavour.
REFERENCES
Alper, K., Devinsky, O., Perrine, K. et al. (1997). Dissociation in epilepsy and conversion nonepi-
leptic seizures. Epilepsia, 38, 9917.
American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders
(Fourth Edition) (DSMIV). Washington, DC: APA.
Bancaud, J., Brunet-Bourgin, F., Chavel, P. and Halgren, E. (1994). Anatomical origin of dj vu
and vivid memories in human temporal lobe epilepsy. Brain, 127, 7190.
Bernstein, E. and Putnam, F.W. (1986). Development, reliability and validity of a dissociation
scale. J Nerv Ment Dis, 174, 72735.
Betts, T. (1990). Pseudoseizures: seizures that are not epilepsy. Lancet, 336, 1634.
Betts, T. and Boden, S. (1992). Diagnosis, management and prognosis of 128 patients with non-
epileptic attack disorder. Part I. Seizure, 1, 1926.
Bowman, E.S. (1993). Etiology and clinical course of pseudoseizures: relationship to trauma,
depression and dissociation. Psychosomatics, 4, 33342.
Bowman, E.S. and Markand, O.N. (1996). Psychodynamics and psychiatric diagnoses of pseudo-
seizure patients. Am J Psychiatry, 153, 5763.
Brown, R.J. (1999). An Integrative Cognitive Theory of Suggestion and Hypnosis. PhD thesis,
University College London.
Brown, R.J. and Trimble, M.R. (2000). Dissociative psychopathology, non-epileptic seizures and
neurology. J Neurol Neurosurg Psychiatry, 69, 28591.
Cardea, E. (1994). The domain of dissociation. In Dissociation. Clinical and Theoretical
Perspectives, ed. S.J. Lynn and J.W. Rhue, pp. 1531. New York: Guilford Press.
Chu, J.A. and Dill, D.L. (1990). Dissociative symptoms in relation to childhood physical and
sexual abuse. Am J Psychiatry, 147, 88792.
Dericioglu, N., Saygi, S. and Ciger, A. (1999). The value of provocation methods in patients sus-
pected of having non-epileptic seizures. Seizure, 8, 1526.
Devinsky, O., Putnam, F., Grafman, J., Bromeld, E. and Theodore, W.H. (1989). Dissociative
states and epilepsy. Neurology, 39, 83540.
Erdelyi, M.H. (1985). Psychoanalysis: Freuds Cognitive Psychology. New York: W.H. Freeman.
Fodor, J.A. (1983). The Modularity of Mind. Cambridge, MA: MIT Press.
Francis, P. and Baker, G.A. (1999). Non-epileptic attack disorder (NEAD): a comprehensive
review. Seizure, 8, 5361.
Frischholz, E.J., Lipman, L.S., Braun, B.G. and Sachs, R. (1992). Psychopathology, hypnotizabil-
ity, and dissociation. Am J Psychiatry, 149, 15215.
Gates, J.R. and Erdahl, P. (1993). Classication of non-epileptic events. In Non-Epileptic Seizures,
ed. J.R. Gates and A.J. Rowan, pp. 2130. Oxford: ButterworthHeinemann.
207 Seizures, epilepsy and dissociation
Good, M.I. (1993). The concept of an organic dissociative syndrome: what is the evidence? Harv
Rev Psychiatry, 1, 14557.
Halligan, P.W., Athwal, B.S., Oakley, D.A. and Frackowiak, R.S.J. (2000). Imaging hypnotic par-
alysis: implications for conversion hysteria. Lancet, 355, 9867.
Hilgard, E.R. (1977). Divided Consciousness: Multiple Controls in Human Thought and Action.
New York: John Wiley & Sons.
Irwin, H.J. (1994). Proneness to dissociation and traumatic childhood events. J Nerv Ment Dis,
8, 45660.
Janet, P. (1889). Lautomatisme Psychologique. Paris: Felix Alcan.
Janet, P. (1924). The Major Symptoms of Hysteria (2nd edition). New York: Macmillan.
Kihlstrom, J.F. (1994). One hundred years of hysteria. In Dissociation: Clinical and Theoretical
Perspectives, ed. S.J. Lynn and J.W. Rhue, pp. 36594. New York: Guilford Press.
Kirsch, I. and Lynn, S.J. (1995). The altered state of hypnosis. Am Psychologist, 50, 84658.
Kirsch, I. and Lynn, S.J. (1997). Hypnotic involuntariness and the automaticity of everyday life.
Am J Clin Hypnosis, 40, 32948.
Kosslyn, S.M. (1994). Image and Brain. Cambridge, MA: MIT Press.
Kretschmer, E. (1926). Hysteria. New York: Nervous and Mental Disease Publishing Co.
Krishnamoorthy, E.S., Brown, R.J. and Trimble, M.R. (2001). Personality and psychopathology
in non-epileptic attack disorder (NEAD): a prospective study, Epilepsy Behav, 2, 41822.
Kuyk, J., van Dyck, R. and Spinhoven, P. (1997). The case for a dissociative interpretation of
pseudoepileptic seizures. J Nerv Ment Dis, 184, 46874.
Kuyk, J., Spinhoven, P. and van Dyck, R. (1999). Hypnotic recall: a positive criterion in the
dierential diagnosis between epileptic and pseudoepileptic seizures. Epilepsia, 40, 48591.
Lipowski, Z.J. (1968). Review of consultation psychiatry and psychosomatic medicine. III.
Theoretical issues. Psychosom Med, 30, 395422.
Meierkord, H., Will, B., Fish, D. and Shorvon, S. (1991). The clinical features and prognosis of
pseudoseizures diagnosed using video-EEG telemetry. Neurology, 41, 16436.
Nemiah, J.C. (1991). Dissociation, conversion and somatization. In American Psychiatric Press
Annual Review of Psychiatry, ed. A. Tasman and S. Goldnger, Vol. 10, pp. 24860.
Washington, DC: APA Press.
Oakley, D.A. (1999). Hypnosis and conversion hysteria: A unifying model. Cognitive
Neuropsychiatry, 4, 24365.
Pettinati, H.M., Horne, R.L. and Staats, J.M. (1985). Hypnotizability in patients with anorexia
nervosa and bulimia. Arch Gen Psychiatry, 42, 101416.
Pribor, E.E., Yutzi, S.H., Dean, T.J. and Wetzel, R.D. (1993). Briquets syndrome, dissociation, and
abuse. Am J Psychiatry, 150, 150711.
Reason, J.T. (1979). Actions not as planned. In Aspects of Consciousness, Vol. 1: Psychological Issues,
ed. G. Underwood and R. Stevens, pp. 6789. London: Academic Press.
Sierra, M. and Berrios, G.E. (1998). Depersonalization: neurobiological perspectives. Biol
Psychiatry, 44, 898908.
Spiegel, D. and Cardea, E. (1991). Disintegrated experience: the dissociative disorders revisited.
J Abnorm Psychol, 100, 36678.
208 R.J. Brown
Spiegel, D., Hunt, T. and Dondershine, H.E. (1988). Dissociation and hypnotizability in post-
traumatic stress disorder. Am J Psychiatry, 145, 3015.
Thomas, L. and Trimble, M.R. (1997). Dissociative disorders. In Epilepsy: A Comprehensive
Textbook, ed. J. Engel and T.A. Pedley, pp. 277584. Philadelphia: LippincottRaven
Publishers.
Woody, E.Z. and Bowers, K.S. (1994). A frontal assault on dissociated control. In Dissociation:
Clinical and Theoretical Perspectives, ed. S.J. Lynn and J.W. Rhue, pp. 5279. New York:
Guilford Press.
World Health Organization (1992). The International Classication of Mental and Behavioural
Disorders: Clinical Descriptions and Diagnostic Guidelines (Tenth Revision) (ICD10). Geneva:
WHO.
209 Seizures, epilepsy and dissociation
14
Psychobiology of psychogenic
pseudoseizures
J. Chris Sackellares and Dalma Kalogjera-Sackellares
Malcolm Randall VA Medical Center and University of Florida, Florida, USA
Introduction
Psychogenic pseudoseizures are paroxysmal events which mimic epileptic seizures.
While patients suering from these symptoms are referred to neurologists because
they are mistakenly believed to have epilepsy, neurologists consider the underlying
disorder to be of psychological aetiology. Our observations on the nature of
pseudoseizures are based on a carefully studied sample of 100 patients with pseudo-
seizures evaluated over a period of 5 years at the University of Michigan Medical
Center. The patients were evaluated by intensive neurological, clinical psychologi-
cal and/or neuropsychological investigations. Psychodynamic psychotherapy was
performed in a portion of this group. The results of these eorts have provided us
with insights into the psychopathology of the disorder (Kalogjera-Sackellares,
1995; Kalogjera-Sackellares and Sackellares, 1997a, b, 1999). In addition, we dis-
covered evidence to suggest that neurological disturbances may play a role in the
pathophysiology of the disorder (Kalogjera-Sackellares and Sackellares, 1999). A
recurrent theme among our patients was trauma. This included physical as well as
emotional trauma (Kalogjera-Sackellares, 1995; Kalogjera-Sackellares and
Sackellares, 1999). The prevalence of trauma, as well as the post-traumatic charac-
ter of many of the symptoms, do not t with the current diagnostic framework of
the somatoform disorders (Guggenheim and Smith, 1995) under which pseudosei-
zures are typically considered. The concept of somatoform disorders obscures the
operative impact of both the emotional as well as physical trauma which are cor-
nerstones of our approach to this disorder. Our own observations regarding the
psychopathology as well as the possibility of neurological disturbance led us to an
extensive review of the literature on pseudoseizures and hysteria.
Among the richest sources of information on the subject are the case descrip-
tions by the nineteenth century clinical investigators, Paul Briquet (a psychiatrist)
and Jean-Martin Charcot (a neurologist). Briquet wrote extensively on the subject
of hysterical seizures (a historic term for pseudoseizures) in his classic book on hys-
210
teria (Briquet, 1859). Charcots case descriptions as well as his comments on the
subjects have been captured for posterity in the form of published lectures
(Charcot, 1879; Goetz, 1987; Harris, 1991). These clinicians contributed much to
the foundations of both neurology and psychiatry. As we came to learn from the
review of their work, both believed that hysteria resulted from a neurological dis-
order, but failed to nd neuroanatomical abnormalities to support this view. As a
result, neurological investigations into this disorder all but ceased. The emphasis
shifted to investigations into the psychopathology of the disorder. These psychiat-
ric investigations occurred under the historic impact of the pioneering insights of
Janet, Freud and others. Our analysis of the literature was aided considerably by the
translation of Charcot by G. Siegerson (Charcot, 1879), as well as the more recent
translations (Goetz, 1987; Harris, 1991). Analysis of Briquet, on the other hand,
was largely based on direct translations (by DKS) of his text (Briquet, 1859) because
no English translation of his work was available.
In this chapter, we will review the historical events that led to the commonly held
view that pseudoseizures are solely grounded in psychopathology, in the absence of
detectable neurological disorders. In reviewing the works of Charcot and Briquet,
it is apparent that the importance of trauma in the development of hysteria was rec-
ognized by both Briquet (1859) and Charcot (1879; Harris, 1991). However, focus-
ing on a series of well-described cases of Charcot, we nd that it was emotional
trauma, not physical trauma that he emphasized. Yet, we will provide ample evi-
dence that, in Charcots cases, physical trauma (particularly head injury) may have
played an equally important role. In addition to our review of historical cases, we
will discuss the results of recent studies supporting the concept that neurobiologi-
cal disturbances interact with traumatic environmental stresses, leading to the
development of psychogenic pseudoseizures.
Historically, psychogenic pseudoseizures have been considered to be a classic
manifestation of hysteria. Until the emergence of the modern view of hysteria in
the nineteenth century, hysteria was generally believed to result from disturbances
in the reproductive system, specically the uterus. Nineteenth century investiga-
tors, particularly Pierre Briquet, and Jean-Martin Charcot, rejected the aetiological
role of the uterus. In addition to making this historic contribution, they further-
more hypothesized that hysteria arose from disturbances in the central nervous
system (Briquet, 1859; Goetz, 1987; Harris, 1991). Thus, both of these clinical
investigators believed hysteria to be a neurological disorder (although Briquets
contribution to the conceptualization of hysteria as a disorder of the central
nervous system is rarely acknowledged and has only recently been discussed at
some length in the context of the forgotten avantgarde of neuroscience (Mai, 1982;
Mai and Merskey, 1981). First Briquet (Mai, 1982; Mai and Merskey, 1981), and
later Charcot, attempted to identify lesions in the nervous system to account for the
211 Psychobiology of psychogenic pseudoseizures
symptoms of hysteria. They were unable to nd relevant anatomical lesions in their
patients. However, their quest was limited by the histological methods of the times.
In response to the absence of identiable lesions, both men postulated physiologi-
cal or dynamical neurological disturbances to account for the functional distur-
bances and physical ndings they observed in their patients.
As noted earlier, Charcot emphasized the importance of trauma in patients with
hysteria. However, he focused upon the emotional trauma, or shock rather than
physical injury to the nervous system. Review of his cases fully substantiates the
importance of shock and emotional trauma in these unfortunate individuals
(Harris, 1991). However, his case reports also describe physical trauma, and in par-
ticular, head injury as well as debilitating illnesses in the vast majority of his cases.
While Charcot described these physical insults, he did not postulate a causal rela-
tionship between head injury and hysteria, presumably because of the absence of
neurological lesions in postmortem tissue.
Given the obvious presence of emotional trauma and psychological symptoms
among hysterical patients and the failure to nd neurological lesions, it was natural
that subsequent clinical investigators, led by Janet and Freud, would focus upon
psychodynamics. The success of psychological investigations and failure of biolog-
ical investigations led neurologists to abandon the subject of hysteria. The primary
interest of neurologists became limited to the dierentiation of hysteria from true
neurological disorders.
Apart from recognizing the importance of emotional trauma in his patients,
Charcot was unable to provide further insight into the aetiology of hysteria.
However, he did provide extraordinarily useful descriptions of ndings on neuro-
logical examination as well as vivid description of hysterical seizures. His extraor-
dinarily detailed studies of the phenomenology of hysterical patients can be found
in the lectures chronicled by his students (Charcot, 1879; Harris, 1991). Based on
these observations, Charcot described four phases of the archetypal hysterical
seizure: (1) a prodromal phase, (2) an epileptoid phase, (3) a phase characterized
by bizarre postures and (4) an emotional phase. Less severe or elaborate seizures
could occur. These modied attacks consisted of only parts of the full-blown arche-
typal seizures. The study of these phases was aided by drawings that captured the
important details of hysterical seizures (see Figures 14.1 and 14.2). A large sample
of these drawings was published in a book by Richer (1885). Richers drawings cap-
tured the motor manifestations of pseudoseizures (for example, see Figure 14.1).
Even more impressive was his success in rendering facial expressions which capture
the intense emotional and experiential aspects of certain phases of the hysterical
seizure. An example is illustrated in Figure 14.2.
The question arises as to whether todays patient with psychogenic pseudosei-
zures suers from the same malady as the hysteria cases of Charcots time. Given the
212 J.C. Sackellares and D. Kalogjera-Sackellares
high incidence of nonepileptic seizures (hystero-epilepsy) in hysterical patients of
the nineteenth century, it seems reasonable to assume that Charcot was describing
the same clinical entity as that underlying todays psychogenic pseudoseizure. Citing
Briquet, Charcot stated that hysterical seizures occurred in all but 25% of patients
with hysteria (Harris, 1991). Thus, nonepileptic seizures have been considered a
typical manifestation of hysteria from the beginning of modern neurology. In the
nineteenth century, hysteria was grouped along with other disorders which had no
known anatomical lesion. Examples of such disorders included epilepsy and chorea,
neither of which were associated with neuropathological lesions demonstrable with
the histological techniques of the time. Disorders considered to be neurological, yet
without a demonstrable lesion, were classied as neuroses. Paradoxically, while the
term neurosis has persisted, its modern meaning, in contrast to that of nineteenth
century clinical investigators, indicates psychological causation.
The modern vestige of this early attempt at classication is a tendency among
many clinicians to consider nonepileptic seizures as functional rather than
213 Psychobiology of psychogenic pseudoseizures
Figure 14.1. Reproduction of a drawing by Richer depicting a patient experiencing the epileptoid phase
of a hysterical seizure (from Richer, 1885).
neurological or organic in aetiology. The implication is that nonepileptic pseudo-
seizures do not have a biological substrate. However, careful reading of Charcots lec-
tures suggests that he used the term functional to mean a disorder of the nervous
systemthat is not associatedwithdemonstrable lesions of the nervous system. Today,
with knowledge gained from more powerful investigative tools, we have extended
the concept of organic to include a number of toxic, metabolic, degenerative and
genetic disorders that are not associated with physical lesions. In clinical practice
today, the term functional, applied to a symptom or nding on examination, is
taken to mean that the cause is psychological as opposed to biological in nature. It
214 J.C. Sackellares and D. Kalogjera-Sackellares
Figure 14.2. Reproduction of the period of passionate attitude (the emotional phase), the fourth
stage of a classical hysterical seizure.
is generally accepted that many disorders once considered to be functional are due
to biological causes (e.g. Gilles de la Tourette syndrome, schizophrenia, bipolar
aective disorder). In an analogous manner, it is possible that a biological substrate
for pseudoseizures will be identied, using todays sophisticated investigative tools.
Charcot meticulously applied the prevailing clinicalanatomical method, a
method which he helped perfect, to the investigation of patients with hysteria. The
clinicalanatomical method involved the correlation of clinical signs and symp-
toms to anatomical lesions in the nervous system. His application of this method
to the study of the progressive muscular atrophies, amyotrophic lateral sclerosis,
and multiple sclerosis resulted in major contributions to the understanding of
those disorders. Such cases were classied as organic because they were associated
with localized lesions of the nervous system. Such lesions were detectable by visual
inspection of autopsy material. More detailed localization and description of these
organic lesions was made possible by skilful use of the microscope, a tool that had
been recently introduced into medical science.
In contrast to his success with the organic neurological disorders, Charcot was
unable to unravel the mystery of hysteria through the application of the clini-
calanatomical method. Hysterical patients presented a plethora of observable
physical ndings on neurological examination. In addition to seizures, these nd-
ings included blindness, diplopia, limb weakness, sensory loss, tremors and invol-
untary movements. However, these ndings occurred in patterns that diered from
those found in patients with observable anatomical lesions of the nervous system.
In some cases, the pattern of visual disturbance, weakness, or sensory loss con-
icted with the rules of functional neuroanatomy known at the time. More impor-
tantly, the profound clinical abnormalities were not associated with detectable
lesions in the nervous system. Charcot was convinced that the clinical ndings of
hysterical patients were due to appropriately located anatomical abnormalities.
However, in the absence of observable structural lesions, he referred to these lesions
as dynamical lesions (Harris, 1991).
Although the clinical ndings in hysterical patients diered from those of the
organic neurological disorders, Charcot was able to demonstrate classical clinical
ndings that have become the hallmark of hysteria. These ndings include con-
stricted visual elds, monocular diplopia, monoplegia, strange contractures and
hemianaesthesia. The motor weakness observed in hysterics was not associated
with the reex abnormalities which were seen in patients with demonstrable lesions
of the nervous system. Even today, such ndings are taken as evidence for the pres-
ence of hysteria, although modern psychiatrists prefer to apply more modern
terms, such as conversion disorder.
Charcot preferred not to oer theories to explain neurological disorders, generally
preferring to conne his discussions to empirical ndings. However, his observations
215 Psychobiology of psychogenic pseudoseizures
of the eects of hypnosis and suggestion in hysterics served as a stimulus for the sub-
sequent development of the psychological theories pioneered by Janet and Freud
(Goetz, 1987). The clinical histories he provided during his lectures frequently
described neurological as well as psychiatric illness in family members. His com-
ments suggest that he considered heredity an important factor in the development of
hysteria. Although he did not speculate about this issue, his discussions indicate that
various neurological symptoms appearing in the same family represented dierent
clinical manifestations of an underlying hereditary predisposition (Harris, 1991).
Charcots lectures included descriptions of traumatic events in his female
patients with hysterical seizures (Charcot, 1879); however, we noted that it was in
his male hysterics that he emphasized the history of antecedent trauma and previ-
ous illness (Harris, 1991). In his lectures on seven cases of male hysteria (in all of
whom pseudoseizures gured prominently), he emphasized the history of antece-
dent trauma and its relationship to the onset of symptoms. During his lectures, his
comments focused upon psychological trauma. However, our careful review of his
lectures on seven male patients with hysteria reveals numerous references to signi-
cant closed head injuries. Indeed, he describes closed head injuries in four of the
seven cases described. A fth patient lost consciousness as a result of an accident
that caused severe blood loss. Three were either physically abused or attacked.
Antecedent debilitating illnesses occurred in ve cases. Emotional trauma is
described in all seven cases. Paternal alcohol abuse was described in four cases, but
there is only one patient with a history of past alcohol abuse. Charcots emphasis
on the role of emotional trauma and not physical trauma is understandable, given
the intolerable psychological and social stresses patients experienced. What is less
clear is that he did not consider the impact of the physical trauma which was equally
apparent in his patients.
Consistent with the historic heritage discussed in this chapter, a number of con-
temporary authors have rediscovered serious traumatic experiences in patients
with psychogenic pseudoseizures (Alper et. al., 1993; Arnold and Privitera, 1996;
Bowman, 1993; Cartmill and Betts, 1992; Gross, 1982; Harden, 1997; Kalogjera-
Sackellares, 1995). This recurrent theme of trauma gured so prominently in the
histories of pseudoseizure patients that it played a key role in the classication of
pseudoseizure syndromes proposed by Kalogjera-Sackellares (1995). The continu-
ity of historical and contemporary reference to trauma makes the study of trauma
a natural starting point for investigations into the nature of pseudoseizures
(Kalogjera-Sackellares and Sackellares, 1999). It must be considered relevant to the
understanding of its causes and to the development of approaches to clinical man-
agement (Kalogjera-Sackellares, 1995).
Although psychic trauma is a dominant factor in most cases of psychogenic
pseudoseizures, physical trauma, often involving closed head injuries also is a
216 J.C. Sackellares and D. Kalogjera-Sackellares
prominent part of the history in many patients. Several recent authors have com-
mented on head injuries in their patients with psychogenic pseudoseizures. A high
incidence of head trauma occurring during physical abuse or during accidents was
reported by Bowman (1993). Similarly, Lancman et al. (1993) found many patients
with head trauma associated with loss of consciousness. In addition, Westbrook et
al. (1998) found that 32% of their patients with nonepileptic seizures had experi-
enced an antecedent head trauma which was classied as minor in 91% of the cases.
Upon reviewing the histories of 100 patients with psychogenic pseudoseizures
evaluated at the University of Michigan (Kalogjera-Sackellares and Sackellares,
1999), we found a documented history of signicant closed head injury in 52% (see
Table 14.1). In most instances, the trauma would have been categorized as mild
head injury. These patients sustained head injury as a result of motor vehicle acci-
dents, accidents on the job, falls or physical assault. Often, the physical assault was
part of a chronic pattern of physical abuse. In many instances, these injuries
occurred during childhood. However, there are some cases in which there was no
history of signicant head trauma until adulthood. As in many of Charcots cases,
in those cases with head trauma during adulthood, the injury preceded seizure
onset by days to months and the patients usually ascribed a causal role to the head
injury.
The high incidence of head trauma in patients with psychogenic pseudoseizures
raises the question as to whether the head trauma may have played some role in the
pathogenesis of the disorder. Is it possible that mild brain injury can render a
patient more vulnerable to psychogenic pseudoseizures? In our sample, other
potential causes of brain injury were reported. A history of drug or alcohol abuse
was found in 13% of patients (Table 14.1). Approximately 29% (Table 14.1) were
217 Psychobiology of psychogenic pseudoseizures
Table 14.1. Demographics of 100 patients with psychogenic pseudoseizures referred to
the neuropsychology programme at the University of Michigan
Males 20%
Females 80%
Mean age 33.7%
Pure pseudoseizures 71%
Mixed epilepsy and pseudoseizures 29%
Abnormal neurological examination 4%
Structural brain abnormality on neuroimaging 8%
History of closed head injury 52%
History of substance abuse 13%
Source: Adapted with permission.
found to have a prior or concurrent history of epilepsy. In another sample, we
found evidence for active epilepsy in approximately 25% of patients with psycho-
genic pseudoseizures (Kalogjera-Sackellares and Sackellares, 1997a, b). In their case
series, Wilkus et al. (1984) reported a high incidence of events that may have con-
tributed to neurological impairment. Such events occurred in fully 80% of their
sample. These events included head injury with loss of consciousness or other
sequelae in 7 of their 25 cases; an infectious disorder with sequelae in 2 cases; prior
brain surgery had occurred in 2 cases. Other events such as birth trauma, sun
stroke, severe heat exhaustion, partial drowning or gas exposure also occurred in
their patients.
Further evidence of impaired brain function in patients with psychogenic
pseudoseizures comes from studies of cognitive and intellectual function in these
patients. Neuropsychological test batteries are employed in clinical practice to detect
evidence of organic brain dysfunction. Therefore, it is natural to expect that patients
with a purely functional disorder would perform normally on these tests.
Alternatively, one might expect the pattern of test performance to dier from
patients with organic brain disorders. However, several investigators have reported
impaired neuropsychological performances in patients with pseudoseizures. In fact,
several studies, reported by dierent investigators, have produced very similar
results (Kalogjera-Sackellares and Sackellares, 1999). Wilkus et al. (1984; Wilkus and
Dodrill, 1989) reported results from administration of an intensive clinical neuro-
psychological test battery in 25 patients with psychogenic pseudoseizures evaluated
at the University of Washington. These investigations revealed that their sample of
patients performed in the impaired range on 51.2% of the tests. Sackellares et al.
(1985) analysed a sample of patients evaluated at the University of Virginia and
University of Michigan. The authors found that the cognitive performance of that
group was less than would be expected for normal individuals of similar intelli-
gence. Binder et al. (1998) compared a group of patients with nonepileptic seizures
with a group of patients with epilepsy and with normal control subjects on a battery
of neuropsychological tests. They found that the two seizure groups performed sig-
nicantly more poorly than normal controls. However, there were no signicant
dierences between the two seizure groups on any items of the extensive test battery.
They also found that performance on neuropsychological tests was more strongly
correlated with measures on the MMPI/MMPI-2 in the nonepileptic seizure group.
Based on these observations, the investigators concluded that neuropsychological
tests do not discriminate between patients with epilepsy and those with nonepilep-
tic seizures. These investigators concluded that impairment of neuropsychological
performance was strongly related to emotional and psychosocial factors.
In a more recent study, we examined intellectual and neuropsychological test
performance in 53 patients with psychogenic pseudoseizures (Kalogjera-
218 J.C. Sackellares and D. Kalogjera-Sackellares
Sackellares and Sackellares, 1999). The group comprised 44 patients with pseudo-
seizures who had no evidence of concomitant epilepsy, and no denitive historical
data that would support the diagnosis of antecedent epilepsy. This subgroup was
designated as the pure pseudoseizure subgroup. In addition, there were nine
patients with documented pseudoseizures and concomitant epilepsy, or a well-
documented history of epileptic seizures. This subgroup was designated as the
mixed pseudoseizures and epilepsy subgroup. In both subgroups, there was a pre-
ponderance of women (77% in the pure subgroup and 78% in the mixed sub-
group). Intellectual functioning had been measured, using the WAISR in each of
these patients. The HalsteadReitan Neuropsychological Test Battery (Boll, 1981)
was administered to all but one. These tests were performed as a part of standard
clinical evaluation which included medical and neurological history, physical and
neurological examinations, routine EEG, neuroimaging, neuropsychological eval-
uations, and in most cases, long-term EEG-video monitoring. Interestingly, there
were no statistically signicant dierences in the two subgroups with respect to
WAISR scores. The group as a whole (pure and mixed subgroups combined)
revealed an interesting pattern of scores (see Figure 14.3). The mean verbal IQ, per-
formance IQ, and full scale IQ all fell in the average range. Scores were highly var-
iable and ranged from the mentally decient to the very superior. However, the
overall distribution was skewed toward the low end of the average range. Scores on
the HalsteadReitan Neuropsychological Test Battery were not signicantly dier-
ent for the pure and mixed subgroups. Of particular importance is the nding that
more than 50% of the group as a whole performed in the impaired range on more
than half of individual subtests of the battery (based on published cuto scores of
Jarvis and Barth (1984). The percentage of patients scoring in the impaired range
on each of the HalsteadReitan tests for which cuto scores are available is shown
in Table 14.2. The Halstead Impairment Index (HII) is a summary score for the
HalsteadReitan Neuropsychological Test Battery. The mean HII for the entire
sample and for each subgroup was in the impaired range. However, individual HII
scores varied substantially (Figure 14.4). Scores ranged from 0.1 (normal) to 1.0
(impaired performance on all the constituent tests used to calculate the HII). The
HII fell within the moderately to severely impaired range (0.7 and higher, based on
the criteria of Jarvis and Barth (1984)) in nearly half (48.8%) of the pure pseudo-
seizure group. This nding was particularly important given that this group had
not been diagnosed with any dened neurological disorder and their symptoms
had been assumed to result from purely psychological causes.
This study (Kalogjera-Sackellares and Sackellares, 1999) involved a large sample
of patients with well-documented psychogenic pseudoseizures. The intellectual
and neuropsychological ndings of this study are consistent with those of previ-
ous studies. These studies together underscore the prevalence of measurable
219 Psychobiology of psychogenic pseudoseizures
neuropsychological and cognitive decits in patients with psychogenic pseudo-
seizures. Although some investigators have attributed these decits to emotional
factors (Binder et al., 1998), an equally plausible explanation is that many patients
with pseudoseizures, with or without concomitant epilepsy, have undiagnosed
neurological abnormalities. In the vast majority of cases, structural abnormalities
are not reported on MRI and CT scans. In the vast majority of cases, neurological
examinations do not reveal focal or lateralized abnormalities that t patterns that
are seen with lesions of the nervous system. In fact, focal or lateralized neurologi-
cal ndings do not t characteristic patterns associated with anatomical lesions of
the nervous system. However, this does not exclude the possibility of biological
disturbances aecting brain function. Not all such disturbances cause gross
anatomical changes detectable by routine CT or MRI scans.
A surprising observation in our patient sample provides additional evidence that
biological factors may play an important role in the pathogenesis of psychogenic
220 J.C. Sackellares and D. Kalogjera-Sackellares
IQ range
N
u
m
b
e
r
o
f
s
u
b
j
e
c
t
s
16
14
12
10
8
6
4
2
0
6069 7079 8089 9099 100109 110119 120129 130139
Verbal IQ
Performance IQ
Full scale IQ
Figure 14.3. Distribution of scores on the Wechsler Adult Intelligence Scale Revised (WAISR) in a
sample of 53 patients with psychogenic pseudoseizures. Scores range from the mentally
deficient to the very superior range. However, the overall distribution was skewed toward
the low end of the average range. (Adapted with permission.)
pseudoseizures. We recently discovered the very interesting fact that approximately
30% of our patients with pure pseudoseizures are left-handed (Kalogjera-
Sackellares and Sackellares, 2001). This is markedly increased in comparison with
the 10% prevalence of left-handers reported in the general population (Hardyck
and Petrinovich, 1977). This observation can be explained in one of three ways: (1)
it is a chance observation, (2) left-handers are prone to developing psychogenic
pseudoseizures, or (3) our sample contains pathological left handers. The probabil-
ity of nding this high incidence of left-handers by chance alone is extremely low
(P0.000381) (see Kalogjera-Sackellares and Sackellares, 2001). That normal left-
handers are predisposed to developing some sort of functional disorder character-
ized by seizure-like activity seems quite unlikely. Given the neuropsychological and
cognitive diculties observed in patients with pseudoseizures, it is more likely that
the over-representation of left-handers in our sample is due to the presence of some
pathological left-handers (see Kalogjera-Sackellares and Sackellares, 2001).
Pathological left-handers are individuals who were destined to be right-handers,
but developed left-hand dominance due to an insult to the left frontal lobe pre-
natally, during infancy or early in childhood.
In summary, we traditionally consider psychogenic pseudoseizures to be a mani-
festation of a purely psychological disorder. This view stems largely from the
absence of gross neuroanatomical lesions that can explain the seizures or associated
neurological symptoms and signs. Further, these patients manifest a plethora of
221 Psychobiology of psychogenic pseudoseizures
Table 14.2. Percentage of subjects scoring in the impaired range on HalsteadReitan
Neuropsychological Test Battery
Groups Pure Mixed pseudoseizures
combined pseudoseizures and epilepsy
Test
Halstead Index 63% 61% 75%
Categories 67% 73% 33%
TPT-Total 46% 43% 63%
TPT-Memory 18% 14% 38%
TPT-Location 66% 62% 88%
Speech Sounds Perception Test 60% 59% 63%
Seashore Rhythm Test 62% 63% 57%
Trails A Time 24% 21% 38%
Trails B Time 42% 38% 63%
Finger tapping dominant hand 98% 98% 100%
Source: Reprinted with permission.
emotional and personality disturbances (Bowman, 1993; Cartmill and Betts, 1992;
Kalogjera-Sackellares, 1995; Kalogjera-Sackellares and Sackellares, 1997a, b; Roy,
1979; Vanderzant et al, 1986) which direct the clinician toward psychiatric issues.
The importance of psychological factors is further underscored by the prevalence
of traumatic emotional experiences in these patients. Yet, a persistent question is
why these traumatic experiences trigger pseudoseizures and other pseudoneuro-
logical symptoms in some individuals, but not others. At least one possibility is that
some individuals are biologically susceptible to the development of psychogenic
pseudoseizures in response to traumatic emotional experiences.
222 J.C. Sackellares and D. Kalogjera-Sackellares
Impairment index
N
u
m
b
e
r
o
f
s
u
b
j
e
c
t
s
Figure 14.4 Distribution of the Halstead Impairment Index (HII), the summary score for the
HalsteadReitan Neuropsychological Test Battery, administered to 52 of the same patients
shown in Figure 14.3. A score of 0.1 is normal. Scores of 0.7 and above are in the
moderately to severely impaired range. A score of 1.0 indicates performance in the
pathological range on all constituent tests used to compute the HII. Note that individual
scores ranged from normal to 1.0. However, 46.9% of patients in the entire group (pure
pseudoseizures and mixed pseudoseizures and epilepsy subgroups combined) scored in
the moderately to severely impaired range; and 48.8% of the pure group (n44) scored
in the moderately to severely impaired range. (Adapted with permission.)
We suggest that, at least in some individuals, the biological susceptibility may
result from neurological impairment. As our previous discussion indicates, a
number of investigators have reported a high incidence of closed head injury and
other neurological insults in their patients. Several investigators have reported
impaired performance on tests of neuropsychological and cognitive function.
Finally, we have found an over representation of left-handers in our pseudo-
seizure patient sample (Kalogjera-Sackellares and Sackellares, 2001). This over-
representation of left-handers suggests one of two possibilities. First, it is possible
that left-handers subjected to certain neurological insults or severe emotional
trauma may be predisposed to psychogenic pseudoseizures. A second, and more
plausible explanation is the presence of pathological left-handers in the sample. If
the additional left-handers are pathological left-handers, it suggests that some
patients may have experienced neurological insults to the left hemisphere prior to
the establishment of hand dominance (Kalogjera-Sackellares and Sackellares,
2001).
Taken together, all of these observations suggest that biological factors play an
important aetiological role in the development of psychogenic pseudoseizures.
This possibility in no way negates the role of emotionally traumatic experiences or
other environmental factors in the development of psychiatric disorders character-
ized by pseudoseizures. Nonetheless, it may help to explain why all individuals
experiencing traumatic life events do not develop pseudoseizures.
We feel that it is extremely important to consider the strong possibility that
psychobiological factors may play an important role in the pathophysiology of epi-
lepsy. Because pseudoseizures have long been assumed to stem from purely func-
tional causes, there have been essentially no investigations into the neurobiology
of this disorder.
This disorder causes as much disability as medically intractable epilepsy. Yet,
little progress has been made in the exploration of new therapeutic interventions.
Psychotherapy can be benecial to many of these patients (Kalogjera-Sackellares,
1995). However, not all respond, and many do not have access to psychotherapy.
To our knowledge, there have been no organized scientic investigations into the
use of psychopharmacology in the treatment of this disorder.
REFERENCES
Alper, K., Devinsky, O., Perrine, K., Vazquez, B. and Luciano, D. (1993). Nonepileptic seizures
and childhood sexual and physical abuse. Neurology, 43, 19503.
Arnold, L.M. and Privitera, M.D. (1996). Psychopathology and trauma in epileptic and psycho-
genic seizure patients. Psychosomatics, 37, 43843.
223 Psychobiology of psychogenic pseudoseizures
Binder, L.M., Kindermann, S.S., Heaton, R.K. and Salinsky, M.C. (1998). Neuropsychologic
impairment in patients with nonepileptic seizures. Arch Clin Neuropsychology, 13, 51322.
Boll, T.J. (1981). The HalsteadReitan Neuropsychology Test Battery. In Handbook of Clinical
Neuropsychology, ed. S.B. Filskov and T.J. Boll, pp. 577607. New York: John Wiley & Sons.
Bowman, S. (1993). Etiology and clinical course of pseudoseizures relationship to trauma,
depression and dissociation. Psychosomatics, 34, 33342.
Briquet, P. (1859). Trait Clinique et Thrapeutique de Lhystrie. Paris: J.B. Bailliere.
Cartmill, A. and Betts, T. (1992). Seizure behavior in a patient with post-traumatic stress disor-
der following rape. Notes on the etiology of pseudoseizures. Seizure, 1, 336.
Charcot, J.M. (1879). Lectures on the Diseases of the Nervous System Delivered at La Salpetrire.
Translated by G. Siegerson. Philadelphia: Henry O. Lea.
Goetz, C.G. (1987). Charcot the Clinician: The Tuesday Lessons. New York: Raven Press.
Gross, M. (1982). Incest and hysterical seizures. Med Hypnoanalysis, 3, 14652.
Guggenheim, F.G. and Smith, G.R. (1995). Somatoform disorders. In Comprehensive Textbook of
Psychiatry, Sixth Edition, Vol. 1, ed. H.I. Kaplan and B.J. Sadock, pp. 125170. Baltimore:
Williams & Wilkins.
Harden, C.L. (1997). Pseudoseizures and dissociative disorders: a common mechanism involv-
ing traumatic experiences. Seizure, 6, 1515.
Hardyck, C. and Petrinovich, L.F. (1977). Left-handedness. Psychol Bull, 84, 385404.
Harris, R. (ed.) (1991) Charcot, J.M.: Clinical Lectures on Diseases of the Nervous System. London:
Tavistock/Routledge.
Jarvis, P.E. and Barth, J.T. (1984). The HalsteadReitan Test Battery: an Interpretive Guide. Odessa,
FL: Psychological Assessment Resources Inc.
Kalogjera-Sackellares, D. (1995). Psychological disturbances in patients with pseudoseizures. In
Psychological Disturbances in Epilepsy, ed. J.C. Sackellares and S. Berent, pp. 191217. Boston:
ButterworthHeinemann.
Kalogjera-Sackellares, D. and Sackellares, J.C. (1997a). Personality proles of patients with
pseudoseizures. Seizure, 6, 17.
Kalogjera-Sackellares, D. and Sackellares, J.C. (1997b). Analysis of MMPI patterns in patients
with psychogenic pseudoseizures. Seizure, 6, 41927.
Kalogjera-Sackellares, D. and Sackellares, J.C. (1999). Intellectual and neuropsychological fea-
tures in patients with psychogenic pseudoseizures. Psychiatry Res, 86, 7384.
Kalogjera-Sackellares, D. and Sackellares, J.C. (2001). Impaired motor function in patients with
psychogenic pseudoseizures. Epilepsia, 42, 16006.
Lancman, M.E., Brotherton, T.A., Asconape, J.J. and Penry, J.K. (1993). Psychogenic seizures in
adults: a longitudinal analysis. Seizure, 2, 2816.
Mai, F.M. (1982). The forgotten avant-garde. Trends Neurosci, 5, 678.
Mai, F.M. and Merskey, H. (1980). Briquets treatise on hysteria. Arch Gen Psychiatry, 37, 14015.
Mai, F.M. and Merskey, H. (1981). Briquets concept of hysteria: an historical perspective. Can J
Psychiatry, 26, 5763.
Richer, P. (1885). Etudes Cliniques sur la Grande Hystrie ou lHystropilpsie. Paris: Delahaye et
LeCrosnier.
Roy, A. (1979). Hysterical seizures. Arch Neurol, 36, 447.
224 J.C. Sackellares and D. Kalogjera-Sackellares
Sackellares, J.C., Giordani, B., Berent, S. et al. (1985). Patients with pseudoseizures: intellectual
and cognitive performance. Neurology, 35, 11619.
Vanderzant, C.W., Giordani, B., Berent, S. Dreifuss, F.E. and Sackellares, J.C. (1986). Personality
of patients with pseudoseizures. Neurology, 36, 6648.
Westbrook, L.E., Devinsky, O. and Geocadin, R. (1998). Nonepileptic seizures after head injury.
Epilepsia, 39, 97882.
Wilkus, R.J., Dodrill, C.B. and Thompson, P.M. (1984). Intensive EEG monitoring and psycho-
logical studies of patients with pseudoepileptic seizures. Epilepsia, 25, 1007.
Wilkus, R.J. and Dodrill, C.B. (1989). Factors aecting the outcome of MMPI and neuropsycho-
logical assessments of psychogenic and epileptic seizure patients. Epilepsia, 30, 33947.
225 Psychobiology of psychogenic pseudoseizures
15
Epilepsy and panic disorder
Howard A. Ring and Nuri Gene-Cos
St Bartholomews and the Royal London School of Medicine, London, UK
Introduction
Epilepsy is a neurological condition associated with well-dened abnormalities of
brain electroencephalographic (EEG) activity. It is generally treated with antiepi-
leptic agents and occasionally with resective neurosurgery. Panic disorder (PD) is a
psychiatric condition mostly treated with antidepressant medication and cogni-
tivebehaviour therapy (CBT). Why then should the conditions in any way be con-
sidered together in the same chapter? This paper will initially point out clinical
similarities and dierences and will then discuss the question of whether there are
any pathophysiological similarities between the two conditions.
Epilepsy
Epilepsy is a complex condition characterized by recurrent episodes of paroxysmal
disturbance of normal brain functioning. For a secure diagnosis of epilepsy to be
made there needs to be evidence that these paroxysms of disturbance involve both
disruptions of ongoing behaviour and abnormalities of brain electroencephalo-
graphic (EEG) activity recognized as epileptiform.
The clinical manifestations of epilepsy depend on the site of origin, or focus of
the seizure, the pattern, manner and extent of propagation of epileptiform electri-
cal activity through the brain and the aetiology of the epilepsy. From the point of
view of the person with epilepsy, the subjective experience of repeated seizures is
generally relatively similar, or stereotyped, although the actual nature of the experi-
ence is highly variable between individuals. Hence some may experience hardly any
awareness of a seizure, even if they collapse to the ground unconscious and proceed
to suer a sustained tonic-clonic convulsion. Other people with epilepsy may expe-
rience emotional experiences such as fear, unreality or dj vu which, although
perhaps not associated with any behaviours that an external observer would recog-
nize as abnormal, nevertheless may be very disruptive and unpleasant.
The current International League against Epilepsy (ILAE) classication of sei-
zures is based on EEG ndings and seizure phenomenology. The two largest groups
226
of seizure types are the partial seizures, which have a focal onset, and the primary
generalized seizures, which do not have any identiable focal origin. When consid-
ering the relationship between epilepsy and PD, whether with respect to dieren-
tial diagnosis or putative commonalities in pathophysiology, it is the partial
seizures which are of most relevance.
Panic disorder
In both the ICD10 and DSMIV diagnostic classications PD is considered as an
anxiety disorder. Although these classicatory systems do not represent the last
word in mechanistic understanding of behavioural disorders, it is clear from the
inclusion of PD within the anxiety disorder/neurotic disorder grouping that the
general view is that PD has a psychological rather than a biological aetiology.
However, whilst it is clear that the core subjective experience of PD is one of
extreme fear, this does not in itself prove that the disorder is simply an extreme end
of a continuum that starts with mild anxiety.
Panic attacks may (in common with epileptic seizures) be described as paroxys-
mal events. They are discrete periods of intense fear or emotional discomfort,
accompanied by a range of somatic symptoms including palpitations, trembling, a
feeling of shortness of breath (which may be associated with hyperventilation),
sweating, feelings of choking and psychological symptoms including depersonal-
ization, fear of losing control and fear of dying. The attacks occur spontaneously,
without warning, and although they may occur in situations in which they have
previously occurred, when the patient is concerned that an event may happen, they
may also occur unexpectedly. Individual panic attacks are self-limiting although
estimates of duration vary. Retrospective estimates by suerers suggest an average
duration of between 10 and 20 minutes. However, a prospective study reported
considerably longer attacks, with a mean of between 15 and 50 minutes (Taylor et
al., 1986). Whilst some people experience attacks accompanied by most if not all of
the associated symptoms described above, in others there may be very few experi-
enced apart from paroxysmal fear and anxiety. The frequency of attacks varies, both
between individuals, and over time within individuals, from several attacks in a day
to only occasional attacks over a whole year.
The clinical diagnosis of PD is characterized by panic attacks, avoidance of sit-
uations in which previous panic attacks have occurred and ongoing worry regard-
ing the possibility of future attacks. However, these recurrent attacks of extreme
fear and a feeling of impending death or disaster are not restricted to any partic-
ular environmental setting or set of circumstances. In addition, it is important to
note that although patients with PD worry about having further panic attacks,
this worry is of a lower magnitude than the emotions experienced during an
attack.
227 Epilepsy and panic disorder
There is good evidence, based on clinical accounts, that PD is not a homogene-
ous disorder. In some people pure PD exists with panic attacks in the absence of any
other psychopathology. However, sizeable proportions of those with PD are comor-
bid for agoraphobia or depression or both. It has been reported that women with
PD are more likely to report depression, anxiety or agoraphobic avoidance than men
with this diagnosis (Chambless and Mason, 1986). Increased anxiety and depression
in aected women was also noted by Oie et al. (1990). Whilst women may show
greater agoraphobic avoidance, men may increase their alcohol intake to cope with
their symptoms. Although these dierences in the associated symptoms and beha-
viours between the manifestations of PD in men and women exist, the core features
of panic attacks appear to be relatively similar across the sexes. Hence, although in
therapeutic terms it is clearly important to consider the whole syndrome with which
the patients present, in mechanistic terms it may be that both sexes share a common
aetiology of panic attacks, but that cultural and behavioural responses to these
attacks then determine the sex dierences in associated symptoms.
In a large American epidemiological study, the National Comorbidity Survey, it
was found that the prevalence of PD was greatest in the age range 1525 (Eaton et
al., 1994). However, separating data from males and females revealed that whilst
peak age of onset was within this range for males, for females it was older. This
study also noted that people with less than 12 years of education were up to ten
times more likely to suer from PD than those with more than 16 years of educa-
tion (i.e. including a college education).
Perhaps surprisingly, in recent years it has become clear that panic attacks are
associated with an increased risk of attempted and completed suicide (Markowitz
et al., 1989). The rate of suicide attempts is reported as 20% in those with PD and
12% in those with panic attacks but without the avoidance of panic-inducing situ-
ations and anxiety regarding future attacks that contribute to the diagnosis of PD.
In this context it is also noted that there is an increased risk of suicide and attempted
suicide in people with epilepsy, as discussed by Blumer (Chapter 8).
Potential pathophysiological commonalities
Biochemical pathophysiology
Although there are various models of neurochemical disturbance underlying
seizure genesis and also various models proposed to underlie PD, in both condi-
tions aetiological disruptions of the GABA system have been proposed.
It is well established that GABA exerts an inhibitory control on neuronal excit-
ability by its rapid action on Cl