Secondary Schizophrenia (36 page)

mality in the medial temporal structures, although
more widespread damage has also been reported
[68,

Is greater risk of psychosis particular to

69].
The majority of the evidence, therefore, points to
temporal lobe epilepsy (TLE)?

a mediobasal rather than neocortical temporal lobe
Suggestions that psychosis in epilepsy may be exclu-abnormality underpinning psychosis when the focus

sively or preferentially associated with TLE are sup-is in the temporal lobes.

Chapter 6 – Schizophrenia-like psychosis and epilepsy

Laterality of epileptic focus?

Because the suggestion by Flor-Henry
[117]
of a pre-Patients who have SLP with epilepsy generally have
ponderance of left-sided pathology in patients with
been reported not to have a greater than normal famil-SLP, many studies have examined this issue. In the
ial aggregation of schizophreniform disorders
[88,

EEG studies, the majority opinion favors an excess
104, 107].
In the Danish longitudinal registers study
of left temporal foci in the patients with TLE and
[92],
family history of psychosis was associated with
schizophreniform psychosis
[103,
118]
, although there
a relative risk of 3.12 (2.83–3.43) of schizophrenia-have been some negative laterality studies
[69,
115,

like psychosis associated with epilepsy. Interestingly,
119]
. There are many problems with the available data.

a family history of epilepsy also increased the risk
First, the rigor with which laterality was established
of schizophrenia or SLP, even after adjusting for the
differs across studies, and the use of surface EEG

effects of personal history of epilepsy and other con-recordings to establish laterality is open to question.

founding factors. This familiar aggregation of epilepsy
Second, the presence of an epileptic focus on one side
and SLP suggests shared genetic and/or environmental
does not mean that pathology is restricted to that side.

factors.

Third, left-sided preponderance of temporal lobe foci
may not be restricted to psychotic individuals, as the

evidence supports a left-sided bias for TLE in general
Because patients with primary schizophrenic illness
[94].
Fourth, there is emerging evidence that epilepsy
often have abnormalities in their premorbid personal-patients with schizophrenia have generalized seizures
ities, their assessment has been considered a measure
even when they have a temporal focus
[4,
66].
Fifth,
of vulnerability. Slater and colleagues
[88]
argued that
the instruments and diagnostic criteria used for psy-the premorbid personalities of subjects with psychosis
chosis are language dependent, thus introducing a left-related to epilepsy were normal, suggesting that they
side bias
[89].

were different from primary schizophrenics. How-The neuroimaging studies that examined lateral-ever, assessments of premorbid personalities are noto-ity were inconclusive. The CT
[104, 107]
and MRI
riously unreliable, and other studies have not com-

[120]
studies failed to demonstrate lateralized lesions,
mented on this aspect.

although the patients with hallucinations had higher
T1 values in the left temporal lobe. Two small func-Postlobectomy psychosis
tional imaging studies
[121,
122]
provided preliminary evidence of greater left medial temporal lobe
Schizophrenia-like psychosis may develop de novo
dysfunction in SLP with epilepsy. A proton magnetic
many months or years after temporal lobectomy
resonance spectroscopy study
[123]
showed metabolic
for the treatment of intractable epilepsy. Rates from
abnormalities in the left temporal lobe of patients
3% to 28% have been reported, as summarized in
with SLP and epilepsy. The neuropathological stud-

Table 6.6.
The psychosis is usually a paranoid-ies
[109, 110]
have not supported lateralization of
hallucinatory state with depressive features
[102, 111,

pathology.

119],
the neuropathology is diverse, the patients addi-The laterality issue therefore remains undecided,
tionally have had generalized seizures, and there is an
but the importance of a left-sided focus is not striking.

overrepresentation of right lobectomy
[124, 125, 126].

It is possible that the structural abnormality in epilep-Roberts and colleagues
[109]
argued that the postop-tic psychosis is not lateralized, and is possibly bilateral,
erative onset of psychosis may be an artifact of an ear-but that the functional abnormality is predominantly
lier age at operation, but the clustering of the onset
left-sided. However, right-sided abnormality seems to
of psychosis particularly in the 6 months after the
be sufficient, and generalization of the epileptic distur-lobectomy argues against this
[127].
Shaw and col-bance is commonly present.

leagues
[127]
recently summarized the literature on
postlobectomy psychosis, noting reports of 50 cases
of de novo psychosis following temporal lobectomy,
and added 11 of their own. They reiterate the excess

of congenital lesions such as dysembryoblastic neu-A female sex bias was reported by one group
[111]
but
roepithelial tumors in the excised lobes rather than the

is not generally supported
[88,
104, 105, 107].

typical mesial temporal sclerosis. They also emphasize
Organic Syndromes of Schizophrenia – Section 3

Table 6.6
Schizophrenia-like psychosis in patients with drug-resistant temporal lobe epilepsy before and after temporal lobectomy
Schizophrenia-like psychosis

Duration of

Before

After

lobectomy

lobectomy

after lobectomy

Authors

N

Years

%

%

%

Bailey
et al.
[128]

63

2–6

12

19

19

30

7

11

Taylor [129]

100

>5

16

16

19

19

3

3

Jensen and Larsen [119]

74

2–5

11

15

20

27

9

12

Polkey [124]

40

2–5

0

0

3

8

3

8

Stevens [125]

14

20–30

0

0

4

28

4

28

Roberts
et al.
[109]

249

−
a

16

6

25

10

9

4

Shaw
et al.
[127]

320

8

−

−

−

−

11

3.4

a Not Stated

bilateral temporal lobe abnormalities, reporting bilat-which included two patients with chronic and two
eral EEG abnormalities preoperatively and a small
with postictal SLP, showed higher than normal levels
amygdala on the unoperated side, but right-sided
of dopa decarboxylase activity in SLP and schizophre-lobectomy did not emerge as a significant risk factor
nia, and it was suggested to be due to suppressed tonic
in their series. There are some reports of improvement
release of dopamine in striatum because of low corti-of schizophrenic symptoms with temporal lobectomy;
costriatal glutamatergic input.

it is interesting that these cases were associated with
left-sided surgery
[109,
125].

Neuropathological studies

Neuropathological studies of SLP with epilepsy have
Neuroimaging studies

been limited. A large series from London, drawing on
A few neuroimaging studies must be highlighted in
subjects with histories of temporal lobectomies, has
relation to chronic SLP of epilepsy. An MRI study
been reported
[109, 111].
Taylor
[111]
commented
[120]
already referred to did not show any difference
that epilepsy patients with SLP were less likely to
in T1 relaxation times between epilepsy-psychosis
have mesial temporal sclerosis and more likely to have
patients and schizophrenic comparison subjects. There
alien tissue lesions (small tumors, hamartomas, and
is extensive literature on MRI brain morphometry of
focal dysplasias). In the report by Roberts and col-schizophrenia and TLE, and some of the morpholog-leagues
[109],
40% of patients with SLP and epilepsy
ical abnormalities described (large ventricles, small
had mesial temporal sclerosis, 20% had alien tissue
hippocampus) are common to the disorders
[130].
A
gangliomas, and 20% had no lesions (49%, 4%, and
PET study using 15O-H2O demonstrated lower oxy-15% of the total epileptic group, respectively). Histo-gen extraction ratios in the frontal, temporal, and basal
ries of birth injury, head injury, and febrile convulsions
ganglia regions of psychotic patients with epilepsy
were not overrepresented in the group with SLP, but
than in nonpsychotic epileptic patients
[121],
and a
the frequency of alien tissue tumors and early onset
small study using SPECT showed lower left medial
of seizures suggested a developmental lesion in the
temporal blood flow in psychotic than nonpsychotic
medial temporal structures that had been physiolog-epileptic patients
[122]
. The left temporal lobe abnor-ically active from an early age. Stevens
[134],
on the
mality was supported by another study
[131],
but a
other hand, reported widespread pathology in six cases
more recent SPECT study of interictal psychosis in
of epilepsy and psychosis; the pathology included the
patients with TLE failed to find a significant difference
hippocampus, hypothalamus, thalamus, pallidum, and
from controls, with a trend for increased blood flow in
cerebellum. In a study by Bruton and colleagues
[110],

the posterior cingulated region
[132].
The PET study of
epileptic patients with SLP had larger ventricles, more

patients with psychosis by Reith and colleagues
[133],

periventricular gliosis, more focal damage, and more
Chapter 6 – Schizophrenia-like psychosis and epilepsy

periventricular white matter softenings than nonpsy-

Table 6.7
The cellular and molecular consequences of seizures
chotic epileptic comparison subjects, but similar rates
with relevance to the pathogenesis of chronic psychosis
of mesial sclerosis, suggesting greater nonspecific neu-1. Neurogenesis from precursor or stem cells, especially in the
ropathology.

granular layer of hippocampus

2. Neuronal cell death by necrosis or apoptosis, especially in
Possible pathophysiological mechanisms for
various hippocampal sectors

3. Expansion of glutamatergic presynaptic mossy fibers
psychosis in epilepsy

4. Sprouting and altered dendritic morphology, resulting in
Discussion has centered broadly on two mechanisms:
“miswiring”

i) the psychosis is due to the repeated electrical dis-5. Increase in perforated postsynaptic densities on granule cell
charges, either directly or through the development
dendrites

of neurophysiological or neurochemical abnormali-6. Glial cell activation and proliferation
ties; or ii) the epilepsy and psychosis share a com-7. Change in ion channels, for example, increased expression
mon neuropathology that may be localized (emphasis
of the
α
1H-subunit of the voltage-sensitive Ca
++
channel
on temporal lobe but also frontal lobe and the cere-8. Changes in synaptic function.

bellum) or widespread in the brain. Both mechanisms
may be operative, the latter being primary and the for-the antagonism
[136],
but there are limitations to the
mer modifying the presentation, determining exacer-hypothesis: the relationship of the psychosis to seizures
bations and remissions, or being the proximate cause.