Secondary Schizophrenia (35 page)

the latter occurring during a period when seizures
are relatively quiescent. However, this distinction may
to some extent be arbitrary. Phenomenologically, the
Evidence for association

two syndromes are very similar, although the PIP is
The evidence that there is indeed an affinity between
likely to be briefer in duration. The interval between
schizophrenia and epilepsy must come from epidemi-the last seizure and the occurrence of PIP can vary
ological data. The epidemiological evidence for this
considerably and the interval of up to one week is an
association is summarized in
Table 6.4.
The rates must
arbitrary one
[31].
The predisposing factors for the
be interpreted in light of the general prevalences of
two syndromes appear to be similar as well. The neu-psychosis, schizophrenia, and epilepsy in the gen-rophysiological correlates are not necessarily distinc-eral population. Schizophrenia is estimated to have a
tive. Although FN is usually associated with BIP, it is
prevalence of 0.5% to 1% in the general population,
not invariable in this syndrome
[63]
, and can occur
but if we use a broad concept of psychosis, the preva-in PIP
[50].
Patients who experience both PIP and
lence is likely to be much higher
[89].
Epilepsy has a
BIP have been said to have bimodal psychosis
[86],

point prevalence of 0.4% to 1% in the general pop-and in these patients either syndrome may appear
ulation, and the lifetime risk of having at least one
first.

unprovoked seizure is 2% to 5%
[90, 91].
The preva-An examination of possible neurophysiological
lence is low in the first decade of life, increases to a
factors underlying the psychosis suggests that they
plateau in the adult years, and increases further in the
may share common mechanisms. The inhibitory pro-elderly
[91].
Methodological difficulties in the studies
cesses mentioned previously not only bring about the
should also be taken into account in the interpretation
postictal state but are also necessary for maintain-of the prevalence data. For example, the classic study
ing the interictal state. Postictal and brief interic-by Slater and colleagues
[88]
drew its subjects from ter-tal psychoses may therefore not be as different from
tiary centers in two major London hospitals. The over-each other as is generally thought. As the develop-all evidence suggests that SLP is many times higher
ment of seizures may be due to either disinhibition
in epileptic patients than in the general population. A
or hypersynchrony involving enhanced disinhibition
recent study based on the Danish longitudinal registers
[87],
the occurrence of a seizure during psychosis
is particularly noteworthy for its comprehensive cov-may have different pathogenetic mechanisms and may
erage of a population of 2.7 million
[92].
The relative
indicate either disinhibition or increased inhibition.

risk of schizophrenia in patients with epilepsy was 2.48

The effect on psychosis of a seizure could thereby be
(95% CI 2.20–2.8) and of SLP, 2.93 (2.69–3.20). The
either an amelioration of symptoms or their exacer-risk was the same in men and women and increased
bation. This may explain why seizures during postic-with age and with a family history of schizophrenia or
tal psychosis often exacerbate the psychosis, whereas
epilepsy.

those during brief interictal psychosis may improve
the psychiatric status. Different patterns of excitation
and inhibition may also explain why the EEG may
Clinical features

show “forced normalization” in some cases of interic-Slater and colleagues
[88]
reported that they had found
tal psychosis. These speculations can be tested by the
a mean age at onset of about 30 years and that the
longitudinal examination of patients with brief psy-symptoms were largely paranoid-hallucinatory, com-choses, using neurophysiological and neuroimaging
monly associated with catatonia, affective blunting,

methods.

and volitional symptoms. Phenomenologically, the
Chapter 6 – Schizophrenia-like psychosis and epilepsy

Table 6.4
Epidemiological evidence for the association between epilepsy and schizophrenia-like psychosis (SLP)

%

Prevalence of SLP in epilepsy clinic groups:
Gibbs and Gibbs [93]

11,612

2.8

Reliability of psychiatric diagnosis uncertain; majority of subjects young
Currie
et al.
[94]

666

1.8

No criteria described; only patients with temporal lobe epilepsy included
Standage and Fenton [95] 37

8.0

Temporal lobe epilepsy not different from other epilepsies
Mendez
et al.
[4]

1,611

9.25

1.06% of migraine (comparison) subjects had schizophrenia-like psychosis;
comprehensive assessment used DSM-III-R criteria
Schmitz and Wolf [67]

697

4.0

Both generalized and focal epilepsies represented
Onuma
et al.
[96]

1,285

9.1

Point prevalence, 4.0%

Prevalence of SLP in epileptic patients in community-based studies:
Krohn [97]

−

2.0

Additional 9% had incapacitating behavioral disorder
Gudmundsson [98]

987

8.1

Entire epileptic population of Iceland studied
Qin
et al.
[92]

2.27
×
106 RR2.93 (2.69–2.20)
Danish longitudinal register. Risk increased with age and family history of
schizophrenia and epilepsy.

Prevalence of epilepsy in psychotic patients:
Kat [99]

50,000

0.33

Davison and Bagley [100]

1–10

Estimates from

published data

Betts [101]

1,950

2.1

Hospitalized patients

Annual incidence of psychosis in epileptic patients:
Lindsay
et al.
[102]

87

10

Temporal lobe epilepsy subjects followed up for 39 years
Onuma
et al.
[96]

1,285

0.3

up for 39 years

disorder was indistinguishable from schizophrenia,
controlled studies are lacking. Nearly one-half (45%)
although the authors reported a better preservation of
of the patients of Slater and colleagues
[88]
had a
affect, mood swings, mystical experiences, and visual
chronic course. In a 10-year follow-up study in Japan
hallucinations. Investigators in two controlled stud-

[108],
64% of the patients had a chronic course. In the
ies from London
[103, 104]
also noted the largely
absence of an appropriate comparison group, it is dif-paranoid-hallucinatory characteristics of the disorder,
ficult to know if this outcome is different from that in
but they stressed the greater frequency of “organicity.”

schizophrenia with a relatively later age at onset.

One report commented on the rarity of negative symptoms, formal thought disorder, and catatonic symp-Risk factors of chronic SLP with epilepsy
toms
[105],
whereas another reported that visual hallucinations were more prominent than auditory ones
Although a number of studies have examined the risk
[106].
In the study by Mendez and colleagues
[4],
the
factors, the literature remains contentious, without a
epilepsy-with-schizophrenia group did not differ from
clear consensus emerging on many variables. The puta-the nonepileptic schizophrenic comparison subjects
tive risk factors are summarized in
Table 6.5.

on any psychosis item except increased suicidal behavior. In conclusion, except for minor reported differ-

Age at onset, duration, and severity of epilepsy

ences, which may be accounted for by selection biases
Epilepsy beginning at an early age
[109]
and endur-in the comparison group, the chronic psychoses of
ing through puberty
[102, 110, 111]
has been asso-epilepsy are similar to schizophrenia.

ciated with SLP. Other studies have found no rela-Many authors have commented on the relative
tionship to age of onset
[69,
112]
or have associated
lack of negative symptoms and a more benign course
psychosis with later age of onset of epilepsy
[4,
92].

for epileptic schizophrenia [88, 107], but supportive
In the large Danish study
[92],
for every five-year
Organic Syndromes of Schizophrenia – Section 3

Table 6.5
Putative risk factors for chronic schizophrenia-like
ported by a majority of case series
[74, 88,
103, 104,

psychosis of epilepsy

113].
Mendez and colleagues
[4]
reported a higher rate
1. Age: Early age of onset, but evidence conflicting
of partial complex seizures, but not temporal lobe foci,
2. Sex: A female bias reported by one group but not others
in their group with SLP plus epilepsy than in their non-3. Family history: F/H of psychosis or epilepsy
schizophrenic epilepsy comparison subjects. In studies
that compared patients with TLE and those with gen-4. Characteristics of epilepsy:
eralized epilepsy
[114,
115],
the patients with TLE were
a. Many years (usually 10–14) between the onset of
more likely to be psychotic, but the subject groups in
epilepsy and the onset of psychosis

these studies were small. In the large Danish study, the
b. Severe epilepsy: multiple seizure types, history of status
epilepticus, multiple hospital admissions, resistance to
relative risk associated with complex partial epilepsy
drugs

was slightly but nonsignificantly higher than other
c. Partial complex epilepsy, especially of mediobasal
types of epilepsy
[92]
(relative risk, after adjustment for
temporal lobe origin

complex partial epilepsy 3.38, for other partial epilepsy
d. History of secondary generalization

3.18, for generalized epilepsy 2.81). Another argument
e. Left sided focus

[89]
has been that the proportion of TLE in epilepsy-psychosis patients is no different from that in the adult
5. Neuropathology:

epileptic population in general, the latter being esti-a. Presence of neuroembryodysplastic lesions, for example,
mated to be about 60%
[90, 91].
This debate, therefore,
gangliogliomas, hamartomas.

has not resolved but continues to be in favor of a special
b. Bilateral pathology

but not exclusive relationship between SLP and TLE.

6. Neurological examination: sinistrality
Additionally, there are neuroimaging and neuropathological data linking the temporal lobe with psychosis
increase in the age at diagnosis of epilepsy, there was
(discussed in later sections).

an increased relative risk of 1.2 (1.14–1.26, p < 0.0001)
There is also a suggestion that the phenomenology
of SLP. Many years (usually 10–14) are said to inter-of the psychosis associated with TLE is somewhat dif-vene between the onsets of epilepsy and SLP
[88,
104,

ferent from that associated with generalized epilepsy.

107],
but this period is highly variable and patients
The latter are reportedly relatively mild, shorter in
who develop epilepsy after the psychosis are usually
duration, often associated with confusion in the early
excluded from such analyses. Moreover, the peak age
stages, and lacking in Schneiderian first-rank symp-at onset for epilepsy is in any case earlier than that for
toms
[114].

schizophrenia
[90]
, thus making the relevance of this
observation somewhat ambiguous.

It is often noted that the patients who develop psy-

Mediobasal or neocortical temporal lobe epilepsy?

chosis have a severe form of epilepsy involving mul-Kristensen and Sindrup [69] reported that psychotic
tiple seizure types
[95],
a history of status epilepticus
patients had a substantial preponderance of tempo-

[88],
and resistance to drug treatment
[6].
The risk
ral mediobasal spike foci, recorded on sphenoidal
was higher in epileptics with multiple admissions in
electrodes, and an excess of epigastric auras. Her-the Danish study
[92].
The frequency of seizures at
mann and colleagues
[116]
reported a higher fre-the time of development of the psychosis is variable;
quency of schizophrenia and other psychopathology
some authors report an improvement
[102],
whereas
in patients with an aura of fear. Mendez and colleagues
others report a worsening
[4].
Most often, it is not
[4]
reported more psychic and autonomic auras in the
possible in the case of chronic psychosis to relate
psychotic patients. The neuropathological literature
the onset of the psychosis to any change in seizure
(see later section) has supported a predominant abnor-frequency
[88].