Epileptic Disorders
MENUNeurocutaneous melanosis and epilepsy surgery Volume 7, issue 1, March 2005
Auteur(s) : Frederick Andermann1, P. Halász1, A. Holló1, G. Rásonyi1, Dileep Nair2, Armin Mohamed3, Richard Burgess2, Hans Lüders2
1 Neurology and Paediatrics, McGill University,
Epilepsy Service, Montreal Neurological Hospital & Institute,
3801 University Street, Montreal, Quebec H3A 2B4, Canada
Tel: +00 1 514 398 1976 <andermannf@qc.aibn.com>
National Institute of Psychiatry and Neurology, Epilepsy Center,
1021 Budapest, Hüvösvölgyi-út 116, Hungary
<halasz@opni.hu>
2 Department of Neurology, The Cleveland Clinic
Foundation, Cleveland, Ohio
3 Institute of Clinical Neuroscience, Royal Prince
Alfred Hospital, Camperdown, Australia
Neurocutaneous melanosis and epilepsy surgery
To the Editor: Dr Jean Latrelle, an eminent Parisian
neurologist at the time of a meeting in Harriman, N.Y., started his
discussion by saying, “Everything has been written, but not
everything has been read”.
I found the description of a patient with neurocutaneous melanosis
by Oliveira de Andrade et al. (Epileptic Disord,
2004; 6: 145-52) interesting. For the past 30 years we have
been following a similar patient. She was an intelligent,
29 year old, right-handed female. Birth, development, and
family history were normal, and she had no illness that might lead
to cerebral seizures. She had a vest type giant nevus from birth,
which extended from her back to the extremities. Minor seizures
started at the age of 3 years, with a feeling of fear and a
painful grabbing sensation in her stomach followed by trembling of
the body for 10 to 30 sec. She would call or run to her
mother.Minor seizures continued over the years. When she was in her
20s, the attacks started with a sudden uncomfortable feeling in the
head described as “high voltage”. She felt cold or hot, shivered,
and had “goose pimples”. There was a variable degree of impairment
of consciousness, but she was usually not fully unconscious. The
episodes lasted a few seconds, and most were purely subjective.
Only rarely did she fall in her minor attacks. At times they
occurred singly or in brief clusters, but usually they recurred
every 2 to 3 min for up to 12 or 15 hours. These
prolonged bouts occurred once or twice a month. At such times, her
mood changed; she became quiet and irritable. She had a frequent
desire to void and was unable to carry on with her work. When these
episodes occurred at night, she was unable to sleep since the brief
feelings awakened her. She found these prolonged episodes to be
uncomfortable, interfering with her work and social life. She had
five major tonic clonic attacks in her early 20s.
After suitable localization studies she had a right temporal
resection. There was no meningeal melanosis. The amygdala was
replaced by a dark, brownish yellow, rather discrete mass measuring
7-8 mm in diameter. It was adherent to the tip of the pes
hippocampus which was markedly increased in consistency. She had no
further seizures with impairment of awareness but has continued to
have some auras, presumably coming from the same hemisphere. This
patient, with amygdaloid melanosis was described in 1980
(Andermann et al. 1980). She has been fully functional as a
hospital social worker since.This lady illustrates very well, like
the patient presented by Oliviera de Andrade, that seizures in
patients with extensive nevi do not necessarily present with
melanoma, nor is there necessarily malignant deterioration. It is
unlikely that surgical treatment would lead to development of
malignancy and her retardation is also not a reason not to consider
surgical treatment. It is much better to be slow without seizures
than slow and have seizures as well. As we stated in our 1980
report, melanosis of the amygdala occurring as part of
neurocutaneous melanosis may of itself cause symptoms and should be
added to the list of small lesions in the temporal lobe leading to
complex partial seizures.n
References
Andermann F, Mathieson G, Wilkinson RD, Rasmussen TB. Amygdaloid melanosis: A form of neurocutaneous melanosis associated with partial complex seizures. Adv Epileptol 1980.
Oliveira de Andrade D, Dravet C, Raybaud C, Broglin D, Laguitton V, Peretti-Viton P. An unusual case of neurocutaneous melanosis. Epileptic Disord 2004; 6: 145-52.
A critical review of the different conceptual hypotheses framing human focal epilepsy
To the Editor: Nair, Mohamed, Burgess and Lüders
published a theoretical paper (Epileptic Disord 2004; 6:
77-83) analyzing the validity of the “epileptogenic zone” concept
of Penfield and Jasper versus the slightly extended model of
Talairach and Bancaud, involving not only the strict seizure onset
but also the “early seizure spread” zone, and the “large network”
hypothesis of Spencer. They unsheathed their sword to defend the
concept accordingly, in which only the ictal onset zone is relevant
for determining the “epileptogenic zone”, a concept attributed by
them to Penfield and Jasper.
Although the title of the paper (A critical review of the
conceptual hypotheses framing human focal epilepsy) promises a
conceptual approach to the existing models of focal epilepsies,
they address almost entirely the question of how epilepsy surgery
proves or disproves the strict focal or more extended network
models.
The topic they address is of great importance as regards the
grounds for surgical activity. Surgical treatment is, on one hand a
very successful procedure but on the other hand it raises serious
theoretical questions that we are inclined to sweep under the
carpet. We would like to emphasize some of the questions that
surround the conceptual hypotheses of focal epilepsies.
Historical aspects
Penfield and Jasper were operating on epilepsy patients on the basis of the clinical semiology of seizures, interictal EEG (frequently taken from the cortical surface explored during surgery), brain lesions detected by the methods available at that time (pneumoencephalography, angiography, skull x-ray and by visual inspection of the exposed cortex), and seizure symptoms elicited by electrical stimulation of the cortex. They tried to reproduce the “initial phenomenon of the patient’s attacks” by electrical stimulation of the exposed cortex, working under local anaesthesia, but they were rarely able to establish the “initial ictal-onset zone”, and they were not thinking in terms of ictal EEG at all. When, after a local excision guided by a visible cortical abnormality and/or electrical stimulation eliciting seizure phenomena, interictal spiking was still present, they recommended further extension of the excision (Penfield 1954 and 1956, Gloor 1986). So it seems to be rather questionable to attribute the idea of the value of the ictal onset-zone (based on the localization of spontaneous seizure onset) to them.
Can surgical results verify any epilepsy model?
Firstly, it is questionable whether the validity of any concept
of pathophysiology of focal epilepsies can be verified or disproved
by the results of surgical resections.
If we explore the interictal and ictal manifestations of a “focal”
epileptic disorder by different methods, including functional
neuroimaging data, it cannot be restricted to a circumscribed
cortical area, although the resection of a small cortical area
could provide a seizure-free state.
If it were possible to entirely remove the epileptogenic zone
during surgery, the patients would remain seizure-free without
antiepileptic drug treatment. In reality, some of them remain
seizure-free after tapering of the drugs, but some of them never
become seizure-free without drugs. In addition, the phenomena of
“running down” and “building up”: stepwise decreases and the
re-occurrence of seizures long periods after surgery, are also not
understandable within the framework of the “pacemaker resection”
concept.
Our increasing knowledge about the organization of cortical
functions and in particular the higher cognitive functions argues
against the homunculus-like and solitary functional representations
in the majority of the cortical areas, including temporal, frontal
and parietal cortices, apart from the sensori-motor strip. We
should think in terms of networks of multiple, flexible, parallel
functional representations rather than the classical theory
involving cortical representations. If it is true, the epileptic
malfunction should also be considered as a network disorder,
possibly with multiple parallel dysfunctioning of cortical areas,
in contrast to the classical doctrines of cortical functional
localization the sensory-motor areas suggested by Penfield
(Penfield already faced this problem when he stimulated the
temporal cortices as was shown in the classical paper of
Gloor).
Different concepts for focal, versus system-related epilepsies?
The Penfield and Jasper concept fails to explain the
thalamo-cortical pathophysiology of generalized, idiopathic
epilepsies explored by a variety of researchers from Gloor to
Steriade and co-workers in the recent past. In addition, we should
consider that the generalized epilepsy category seems to be more
and more untenable. Instead of “generalized” we should consider a
bilateral partial epilepsy involving the thalamocortical structures
and consequently widespread cortical associative areas. Therefore,
the theories of idiopathic, generalized epilepsies approach more
and more the concepts of focal epilepsies. If we believe that
epilepsies have some common physiopathogenetic features, a theory
explaining the pathophysiology of a focal epileptic disorder should
be applicable to the generalized epilepsies as well.
However, because of the dichotomous way of thinking we can exclude
generalized epilepsies saying that focal and generalized epilepsy
represent quite different categories. However, what of the case of
mesial temporal lobe epilepsy, which is thought to belong to the
focal epilepsies, although there are much data concerning the
involvement of both temporal lobes and the whole limbic system. The
success of different kinds of temporal lobectomies can hardly be
explained by the Penfield-Jasper doctrine since the “pacemaker
area” of temporal lobe seizures is either in the
hippocampus-amygdala complex or in parahippocampal structures,
however the resection necessary for the seizure-free state
involves, in the majority of cases, a more extended area of the
temporal lobe.
Disorder or symptom-(seizure) oriented approach
A further aspect to be discussed is the relationship between
epileptogenesis and seizure genesis related to the epileptogenic
zone.
There is a certain kind of tautological thinking if we say that
the epileptogenic area is an area the excision of which results in
seizure-freedom. Although epilepsy is a disorder producing
seizures, seizures are not synonymous with epilepsy.
The absurdity of identification with or reduction of the epileptic
process to ictogenesis is apparent if we take into consideration
the years without any seizures after identification of the original
epileptogenic lesion and before the first of later, recurring
seizures. We do not know how many structures, and which spatial
network are involved in the development of an established
“pacemaker” area. Furthermore, in certain epilepsies, the seizures
are far less characteristic features of the epileptic disorder, and
the interictal epileptiform activity is the main cause of all the
severe cognitive symptoms that the patients suffer from. It is so
in LKS and ESES, although these are surely partial epilepsies or at
least they have partial features with secondary generalization.
Even the surgical solution in these epilepsies is based upon the
so-called “interictal” and not on ictal symptoms. Seizures cannot
be identified with epilepsy, seizures are just the visible symptoms
of epilepsy. If we want to understand the physiopathogenesis of
epilepsies, or even purely focal epilepsies, we should look beyond
the seizures.
Reply from the Authors: We read with interest the comments of Drs Halász, Holló and Rásonyi and welcome an opportunity to try to answer some of the questions they raised. It is interesting to notice that a good number of the discrepancies pointed out by Halász et al. arise from differences in the definition of the terms.
In our article, we have defined the epileptogenic zone as “the area of cortex from which seizures originate and whose surgical resection leads to seizure-freedom”. Dr Halász and collaborators are correct when they point out that this approach is addressing primarily the question of the approach that should guide the surgeon when he is trying to render a patient seizure-free by resection of cortical tissue. Indeed, the objective of the manuscript was very practical and no attempt was made to discuss the complex pathophysiology of focal seizures. We also agree with Dr Halász that the title of our manuscript “A critical review of the different conceptual hypothesis framing human focal epilepsy”) was probably slightly misleading considering that we did not discuss the pathophysiology of focal epilepsy but only the concept of “epileptogenic zone” as defined above.
On the other hand, it is clear that Penfield and Jasper endorsed the concept of relatively limited epileptogenic zones, even if they usually did not directly measure the “seizure-onset zone”. The main purpose of our article was to compare the concept of epileptic zone (as defined above) and its actual influence on epileptic seizures, between different epilepsy surgery schools. In this regard, it is interesting to note that the concepts of epileptic networks as an integral part of the epileptogenic zone can only arise from the actual recording of ictal EEG discharges. Therefore, it is not surprising at all that Penfield and Jasper actually conceived the epileptogenic zone as a limited cortical area from which seizures arise, without even addressing the possibility that areas of “early” spread could be part of it.
In their letter, Dr Halász et al. argue that the
functional alterations produced by focal epilepsy are not
restricted to a very limited cortical zone, but produce widespread
alterations. There is no doubt that cortical areas are extensively
interconnected and therefore, it is not surprising that any
alteration of function in a given cortical area will affect, to
varying degrees, most of the entire brain. The assumption that
epileptic seizures originate from a limited cortical area certainly
does not contradict this concept! We also feel that it is a gross
misrepresentation to assume that Penfield and Jasper conceived the
cortical areas included in their homunculi as functioning in
isolation. However, the description of the homunculus is certainly
extremely valuable since it guides us as regards the eloquent
cortical areas the resection of which may produce specific
neurological deficits and whose activation by epileptic discharges
produces typical ictal symptoms. This value is independent of the
fact, certainly recognized by any neurophysiologists, that
normally, none of these areas will function in isolation.
Dr Halász et al. argue that it is “absurd to reduce
the epileptic process to epileptogenesis”. Again it is a matter of
definition. We feel that we can only use the term epilepsy if a
certain condition produces epileptic seizures or an equivalent. For
example, let us consider two patients with exactly the same type
and degree of mesial temporal sclerosis, one with seizures and the
other patient just having symptoms related to the pathological
damage (for example, memory difficulties). It would certainly be
difficult to argue that the second patient has epilepsy just
because he has a lesion that frequently produces epileptic
seizures. In other words, there is no such thing as an epileptic
process without seizures. There are only pathological processes
that cause epileptic seizures. The situation of LKS and ESES is
different. In this case, it is assumed that the symptomatology
(cognitive symptoms) is the consequence of interictal discharges.
Interictal discharges are actually “small seizures” that tend to
synchronize only a limited number of neurons and, therefore, do not
produce clinical symptoms or, in the case of LKS or ESES, only
produce cognitive deficits mainly as “ictal” or “postictal
manifestations”.
Dr Halász et al. also argue that the concept of a
“limited” epileptogenic zone is contradicted by the “running down”
or “building up” phenomenon, as well as by the observation that
some patients after surgery only become seizure-free on
medications, or that some patients require more extensive cortical
resections for seizure-freedom. It is difficult to understand why
these observations should contradict the Penfield/Jasper model of
the epileptogenic zone. The “running down” and “building up”
phenomena only imply that the extension of the epileptogenic zone
(area from which seizures may arise) can change over time. The fact
that a patient is only seizure-free if he takes medicine indicates
that surgery did not remove the whole of the epileptogenic zone and
that the remaining epileptogenic zone is less active and can now be
controlled with antiepileptic medication. Finally, those cases in
which selective amygdalo-hippocampectomy was ineffective, point to
the fact that the epileptogenic zone was slightly more extensive
and, therefore, required a more extensive surgery.
Summarizing, we agree with Dr Halász et al. that the
article does not discuss the pathogenesis of focal seizures as
possibly implied in this title. The objective of the article was to
discuss the concept of the epileptogenic zone as clearly defined in
the introduction. None of the arguments presented by Dr Halász
et al. contradicts the fact that Penfield and Jasper
conceived the epileptogenic zone as a very limited cortical area
that did not include cortical regions to which the seizure
discharge would spread after its initiation (Talairach and Bancaud,
and Spencer). n
References
Nair DR, Mohamed A, Burgess R, Lüders H. A critical review of the different conceptual hypotheses framing human focal epilepsy. Epileptic Disord 2004; 6: 77-83.
Penfield W, Jasper H. Epilepsy and the Functional Anatomy of the Human Brain. Little, Brown, 1954.
Penfield W. Epileptogenic Lesions. Acta Neurol Psychiatr Belg 1956; 2: 77-8.