Secondary Schizophrenia (32 page)

and hallucinations, are the main features. Because the
term schizophrenia refers to an idiopathic syndrome,
The focus on seizures

the schizophreniform illness associated with epilepsy
is referred to as schizophrenia-like psychosis (SLP) in
The exclusive focus on seizures in the epileptic patient
order to keep the issue of causality open.

may be in error as the epileptic brain is not normal
between seizures. Even when neuroimaging or interictal EEG abnormalities are lacking, some abnormality
The problem of heterogeneity

at the cellular or molecular level is likely to be present
Both epilepsy and schizophrenia are heterogeneous
[9],
and the psychosis may be related to that underly-disorders, and their categorization has varied consid-ing abnormality. Furthermore, epilepsy is not a static
erably in the last century. Epilepsy, by definition, is
process, and the brain of the epileptic patient is under-

80

characterized by recurrent seizures that are symptoms
going structural and neurochemical change before and
Chapter 6 – Schizophrenia-like psychosis and epilepsy

after the development of seizures, to which ictal events
Table 6.1
Schizophrenia-like psychoses (SLP) and
may actually contribute
[10, 11].

epilepsy: a classification

1. Ictal psychosis:

Is there an affinity between epilepsy

a. Partial complex or psychomotor status
and SLP?

b. Simple partial status

c. Petit mal status

Evidence has been presented in the literature on both
“affinity” and “antagonism” between the two disorders,
2. Postictal psychosis:

which may seem mutually inconsistent on the sur-a. Single episode
face, but must be reconciled in any analysis. The two
b. Recurrent

themes that have dominated psychiatric thought on
3. Brief interictal psychosis

the association between SLP and epilepsy are: 1) they
a. Alternating psychosis

occur together more often than by chance
[10];
and
b. Nonperiodic

2) they are antagonistic to each other
[12].
Because
the evidence suggests that both associations are pos-4. Bimodal psychosis
sible, at least in a modified form, the paradox must
5. Chronic interictal psychosis

be addressed, if not resolved. Is it possible that affin-6. Antiepileptic drug-induced psychosis
ity is explained by one mechanism and antagonism by
7. Neuroleptic-induced epilepsy in a schizophrenic patient
another, and the two can coexist?

prompts one to deduce that the psychotic disturbance
The promise of new insights

is etiologically related to the epilepsy “through a phys-Even if we are able to fully understand the associa-iological mechanism” (p. 316,
[14]).
Most often, sepa-tion between the disorders, new insights do not nec-rate diagnoses of epilepsy and the particular psychotic
essarily follow. The majority of epilepsies are idio-syndrome are appropriate. In such circumstances, note
pathic in nature, so that their etiology cannot greatly
should be made of the relationship of the psychosis
inform the understanding of schizophrenia. In rela-with the onset of epilepsy, seizure frequency, recent
tion to the genetic mechanisms, more than 200 sin-seizure episodes, current anticonvulsant medication,
gle gene disorders are known in which epilepsy is an
EEG abnormalities, and the underlying neurological
important phenotypic feature
[13],
and these include
lesion, if known.

neurodegenerative disorders, mental retardation syn-Because clinical seizures are the outstanding
dromes, neuronal migration disorders, and mitochon-feature of epilepsy, psychotic syndromes have tradi-drial encephalopathies. To which kind of epilepsy the
tionally been classified according to their temporal
psychosis relates is therefore important for any mean-relationship to these events, as ictal, postictal (or peri-ingful understanding. It is no longer a novel finding
ictal), and interictal, with the last type being either
simply to demonstrate that psychosis may be related to
brief or chronic. For this review, I will retain this clas-cerebral dysfunction, and the exercise would be sterile
sification without implying that these categories are
if a deeper understanding of the pathogenetic mecha-distinct in their pathophysiology or clinical manifes-nisms were not forthcoming.

tations (see
Table 6.1).

Categorization

Ictal psychosis

How does one best classify SLP associated with
A nonconvulsive status epilepticus can result in symp-epilepsy? A consensus on the classification is lack-toms resembling psychosis. The psychosis is necessar-ing, and neither DSM-IV nor International Classifi-ily brief, usually minutes to hours. When prolonged
cation of Diseases (ICD-10) has addressed this issue
into days, it is likely to be ictal behavior that extends
specifically. Application of DSM-IV criteria results in
postictally. The most common association is with paran ambiguous situation in which one can make the
tial complex (or psychomotor) status, and patients may
diagnosis of a primary psychotic syndrome or sec-present a wide range of perceptual, behavioral, cog-ondary syndrome (due to a general medical condi-nitive, and affective symptoms, often in association
81

tion) depending on whether the nature of the evidence
with automatisms involving oral activity, picking at
Organic Syndromes of Schizophrenia – Section 3

clothes, and paucity of speech or mutism
[15,
16].

esophageal, or foramen ovale) may be necessary. Res-These episodes of automatisms may be recurrent, with
olution of the disturbance with intravenous diazepam
behavior not being normal in the intervening periods,
is not diagnostically foolproof as many nonepileptic
although the patient may respond to simple instruc-behaviors may so resolve
[17].
Some assistance in diag-tions. Hallucinations may be prominent, and paranoid
nosis may come from the examination of serum pro-delusions or overvalued ideas may be present. Con-lactin levels, which rise after epileptic seizures, peaking
sciousness is altered during the episode but may be dif-at about 20 minutes and returning to baseline around 1

ficult to test, and patients are amnesic for the episode.

hour
[21].
With partial complex seizures, the rise is less
The appropriate DSM-IV diagnosis would therefore
than that after generalized convulsions, but rises above
be delirium. There is generally a history of partial or
500 mU/L should be considered as suggestive
[22].
A
generalized seizures, with reports of an aura in most
rise may be insignificant or absent after a simple partial
cases.

seizure.

Simple partial status may produce affective, auto-Symptoms may reflect one of two mechanisms
nomic, and psychic symptoms that may include hal-

[23]:
1) A positive effect of the seizure discharge, that
lucinations and thought disorder in clear conscious-is, the epileptic discharge activates a behavioral mech-ness. Insight is usually maintained, and the manifes-anism represented in the area subjected to the distation is not that of a true psychosis, but the symptoms
charge. This may result in a myriad of symptoms in
may be misinterpreted or embellished by the patient
the behavioral, cognitive, affective, perceptual, or auto-and behavioral disturbance may result
[17].

nomic domains; 2) A negative effect, that is, either: a)
Petit mal status (absence or spike-wave status)
the individual is unable to engage in a certain behav-results in altered consciousness and such motor symp-ior owing to the temporary paralysis of the anatomi-toms as eyelid fluttering and myoclonic jerks, and
cal substrate of that behavior; or b) some behaviors are
it may superficially resemble psychosis with disorga-released by the inactivation of structures that normally
nized behavior, but delusions and hallucinations are
suppress them. Behavioral disturbance due to a nega-lacking
[16].
Patients almost always have a history
tive effect may occur in other situations, for example,
of absence seizures or rarely generalized tonic-clonic
when the whole cerebral cortex is subjected to a rela-seizures, and the onset is usually before the age of 20.

tively mild form of seizure activity represented by gen-If it has a later onset, there is frequently an underly-eralized spike and wave discharges. This negative effect
ing metabolic disturbance
[18]
. The onset and offset
may continue postictally, or it may initiate then. Expe-are abrupt, and the episode may last from minutes to
riential phenomena in ictal psychosis are likely to be
several hours or even days. The alteration of conscious-due to positive effects, whereas automatisms may be
ness is variable, ranging from slowing of thinking and
caused by positive or negative effects.

behavior to marked disorientation to stupor.

The question of whether chronic psychoses in clear
Patients with ictal psychosis usually have a history
consciousness can be a direct consequence of contin-of epilepsy. By definition, ictal psychosis is concur-uous seizure activity restricted to deep brain struc-rently associated with epileptic discharges in the brain,
tures has generated much controversy. Most epilep-and, except in some patients with simple partial sta-tologists consider this to be extremely unlikely,
[17]

tus
[19],
scalp EEG abnormalities are detectable. The
but this long-held idea remains current
[3].
Kendrick
majority of discharges in psychomotor status have a
and Gibbs
[24]
first used implanted electrodes to study
focus in the limbic and isocortical components of the
the electrophysiological disturbance in schizophre-temporal lobe, but the focus is extratemporal in about
nia and in the psychoses of psychomotor epilepsy;
30% of patients
[20],
usually in the frontal or cingu-spike discharges in medial temporal and frontal struc-late cortex. Because the scalp EEG may be normal in
tures were demonstrated in both patient groups. They
simple partial seizures, the behavioral disturbance may
reported that surgery on medial temporal structures
be mistaken to be interictal, and a high index of sus-was nearly always beneficial for schizophrenia. Sem-picion is necessary. In such cases, if the patient is on
Jacobsen
[25]
and Heath
[26]
noted similar abnor-anticonvulsant medication, the dose may have to be
mal discharges that did not spread beyond the amyg-reduced, especially if EEG telemetry is planned. Acti-dala, hippocampus, and septal regions, again in both
vating techniques such as light sleep or sleep depriva-schizophrenic and “epileptic psychosis” patients. Scalp
82

tion may be useful, and special recordings (sphenoidal,
EEGs were normal or only mildly abnormal. These
Chapter 6 – Schizophrenia-like psychosis and epilepsy

findings have not been further examined, perhaps due
of some pathogenetic mechanisms of psychosis. They
to ethical constraints against using depth recordings
usually follow seizure clusters or a recent exacerba-for research purposes. In patients undergoing depth
tion in seizure frequency
[31]
that may be related to
recording for another purpose such as presurgical
withdrawal of anticonvulsants, often as a part of the
evaluation, the chance occurrence of psychotic phe-video-EEG monitoring of patients
[33, 35].
Postictal
nomena may present an opportunity to address this
psychoses are common in epilepsy-monitoring facil-question. Invasive recordings in animals are generally
ities; 6.4% of the patients in one study developed this
not enough to provide the pathogenetic insights to
syndrome
[35],
and nearly 10% did so in another study
complex behavioral disorders
[27].
A noninvasive neu-

[37].
If the psychosis develops gradually and in parallel
rophysiological technique that may be able to detect
with increasing seizure frequency, it may be referred to
deep limbic discharges is magnetoencephalography,
as periictal rather than postictal, but there is no reason
arguably an important tool for future attempts to
to believe that this distinction is meaningful clinically
address this issue
[28].

or for its pathophysiology. A confounding factor in the
Even if continuous ictal activity in depth record-evaluation of PIP in video-EEG monitoring facilities
ings could be demonstrated, appropriate longitudinal
is that these patients have often had a withdrawal of
and controlled studies are necessary to establish its
their anticonvulsants, and it has been suggested that
significance. First, epileptiform activity in deep struc-this itself may result in psychopathology
[38],
although
tures may be common in nonpsychotic patients with
psychosis is not usual in the absence of seizures.