Epilepsia, 53(6):1088–1094, 2012
doi: 10.1111/j.1528-1167.2012.03438.x
FULL-LENGTH ORIGINAL RESEARCH
Interictal psychotic episodes in epilepsy: Duration and
associated clinical factors
*yNaoto Adachi, yzNozomi Akanuma, yxMasumi Ito, yMitsutoshi Okazaki, y{Masaaki Kato,
and y{Teiichi Onuma
*Adachi Mental Clinic, Sapporo, Japan; yNational Center Hospital for Mental, Nervous, and Muscular Disorders, N.C.N.P., Tokyo,
Japan; zSouth London and Maudley NHS Foundation Trust, London, United Kingdom; xDepartment of Neuropsychiatry, Tenshi
Hospital, Sapporo, Japan; and {Musashino Kokubunji Clinic, Kokubunji, Japan
SUMMARY
Purpose: There have been few reports showing the distri-
bution of the duration of interictal psychosis (IIP) episodes
and their association with clinical characteristics. To clar-
ify the nature of IIP, we studied the duration of IIP
episodes and their related factors.
Methods: One hundred fifty-five patients with epilepsy
exhibited 320 IIP episodes during our follow-up period
(mean 16.9 years). The duration of all the episodes and
the longest episode in each patient during the follow-up
periods were studied. Characteristics of the patients (e.g.,
epilepsy type, age of onset, and family history of psychosis)
and episode-specific factors (e.g., age of the episode, sei-
zure frequency, administrations of antiepileptic drugs
[AEDs] and antipsychotic drugs [APDs]) were analyzed in
association with the duration of the episodes.
Key Findings: Mean duration of the 320 IIP episodes was
82.7 weeks and that of the longest IIP episodes was
Psychotic episodes in patients with epilepsy that occur
without direct relation to seizures have been diagnosed as
interictal psychosis (IIP) (Mellers et al., 1998; Adachi et al.,
2000, 2002). In patients with IIP, their psychotic episodes
negatively affect their quality of life as much as do their sei-
zures per se. Whereas IIP often causes serious conse-
quences, its nature remains uncertain. In general, as seen in
functional psychoses, the duration of psychotic episodes is
associated with multiple issues, for example, individual and
environmental characteristics and treatment strategies (Hafner
& an der Heiden, 2003). Despite the long history of clinical
studies of epilepsy psychosis, there has been little evidence
showing associations between the duration of IIPs and
clinical characteristics.
Accepted January 30, 2012; Early View publication March 16, 2012.
Address correspondence to Dr. Naoto Adachi, Adachi Mental Clinic,
Kitano 7-5-12, Kiyota, Sapporo 004-0867, Japan. E-mail: adacchan@
tky2.3web.ne.jp
Wiley Periodicals, Inc.
ª 2012 International League Against Epilepsy
1088
150.1 weeks. During the follow-up period, 17 patients
(11.0%) showed all episodes remitting within a month and
54 (34.8%) showed all episodes lasting for 6 months or
longer. The IIP episodes that occurred at a younger age
were often prolonged. Patients with a family history of
psychosis or with early onset of psychosis tended to have
more prolonged IIP episodes. Among the episodes
treated with APDs, early administration of APDs was
significantly associated with shorter IIP duration.
Significance: The distribution of the duration of IIP epi-
sodes indicated the broad spectrum and heterogeneity of
the IIP phenomena. The individual vulnerability to psy-
chosis may be associated with prolonged episodes.
Administration of APDs soon after onset of the episodes
appeared to be effective in controlling them. These find-
ings support empirical treatment principles for IIP to
administer APDs at an early stage of its development.
KEY WORDS: Epilepsy, Psychosis, Interictal psychosis,
Duration of psychotic episodes, Treatment.
IIP has often been categorized into two subclassifications,
that is, acute/transient/brief (called ‘‘acute’’ hereafter and
chronic, which may facilitate further clinical and patho-
physiologic investigations (Sachdev, 1998; Adachi, 2006).
No standard criteria, however, have been established. Con-
ventional criteria defined episodes lasting days or weeks as
acute-episodic IIPs and those lasting for months or years as
chronic IIPs (Bruens, 1974; Sachdev, 1998). Some Japanese
studies regarded psychotic episodes for 3 months (Kanemoto
et al., 1996) or longer (Onuma et al., 1991; Matsuura et al.,
1993) as chronic psychosis. More recent studies used the
criteria set for diagnosis of schizophrenia in the Diagnostic
and Statistical Manual of Mental Disorders (DSM) IV
(American Psychiatric Association, 1994) for chronic IIPs
in patients with epilepsy (Mellers et al., 1998). These crite-
ria were set arbitrarily without any examination of validity
and clinical relevance of the cutoff timing. Furthermore, the
current diagnostic schemes for IIP do not address cases with
multiple episodes with varying durations.
In the current study with a large number of IIP episodes,
we reported the distribution of the duration of IIP and its

relation to various clinical characteristics in view of disclos-
ing the nature of IIP.
Methods
Definitions
IIP episodes were defined as the presence of hallucina-
tions, delusions, or a limited number of severe behavioral
abnormalities in accordance with the International Classifi-
cation of Diseases (ICD)-10 Classification of Mental and
Behavioral Disorders (World Health Organization, 1992).
The definition of interictal psychosis is as follows in accor-
dance with Slater and Roth (1969) description and subse-
quent studies: the first psychosis occurs after the
development of epilepsy; their psychotic episodes occur
without a decisive antecedent seizure or cluster of anteced-
ent seizures; and the episodes occur under clear conscious-
ness. Although some episodic manifestations of IIP can be
as short as a few hours or even minutes, the present study
focused on patients who had episodes lasting at least 12 h
for the reliability of evaluation.
Subject selection
IIP episodes have been registered consecutively into our
study database in a series of multicenter studies involving
neuropsychiatric wards for epilepsy or epilepsy outpatient
clinics of National Centre Hospital for Mental, Nervous and
Muscular Disorders, Musashino Kokubunji Clinic, Tenshi
Hospital, or Adachi Mental Clinic since January 1980. Our
epilepsy patients with psychosis attended their clinics every
2–4 weeks in accordance with the national healthcare
guidelines. The decision-making process for the treatment
for IIP episodes was described in detail in our previous arti-
cles (Onuma, 1987; Adachi, 2005). In brief, our treatment
strategies are as follows: (1) establishing a diagnosis of IIP
episodes by excluding ictal/perictal/postictal psychotic phe-
nomena, and evaluating the level of their disturbances; (2)
assessing the patient’s capacity to give the clinician’s con-
sent to treatment and/or to participate in the decision-
making process, and seeking views from family members
when necessary; (3) deciding treatment strategy, that is,
psychotherapy or watchful wait in outpatient clinics, or
pharmacologic treatment (with or without other treatment
options) in outpatient clinics, or in neuropsychiatric wards;
(4) optimizing antiepileptic drug (AED) regimens where
possible by reducing polypharmacy and adjusting the dose
to aim for therapeutic serum levels; and (5) when adminis-
tration of neuroleptics is decided, butyrophenones and ben-
zamides can be used as first-line antipsychotic drugs for IIP
episodes. However, for serious psychomotor excitement,
any neuroleptics can be used to control them rapidly. When
nonpsychotic symptoms are evident, other psychotropics,
for example, mood stabilizers, benzodiazepines, or antide-
pressants, may also be used. More detailed clinical settings
of these institutions were described elsewhere (Adachi
1089
Duration of Interictal Psychosis
et al., 2010). All IIP episodes were assessed and treated by
consultant psychiatrists qualified in clinical psychiatry and
epileptology. Patients with progressive brain diseases,
dementing process, or substance misuse were excluded. The
study was approved by the relevant ethics committees of the
institutions that participated.
Duration of IIP episode
Psychotic symptoms that lasted for 12 h or longer were
assessed. Episodes of duration of 12–36 h were counted as
1 day. Episodes lasting longer than a week were measured
by weeks. In most cases, the timing of the end of IIP epi-
sodes was specified. In the remaining cases, when the dis-
tinct timing of the remission was unclear, the day of
consultation on which the last description of psychotic
symptoms was recorded was regarded as the timing of its
remission. Episodes during which patients showed excite-
ment, disinhibition, or displeasure without distinct psy-
chotic symptoms were not included. In some cases where
the termination of IIP episodes could not be confirmed when
the patients were transferred to another institution or their
episodes were still ongoing at the study endpoint, the period
between the onset of IIP and the last assessment was opera-
tively regarded as the duration of the episode. For analysis,
we used two sets of data: the duration of the all IIP episodes
and each individual’s longest episode during the follow-up
period.
Investigation items
As for patient’s characteristics, the following items were
recorded: (1) age at onset of epilepsy (the age of the first
afebrile seizures); (2) years of follow-up, from the first visit
to the institutes after the development of interictal psychosis
until December 31, 2006; (3) type of epilepsy, based on ictal
semiology, electroencephalography (EEG), and neuroimag-
ing in accordance with the International Classification of
Epilepsies (Commission on Classification and Terminology
of the International League Against Epilepsy, 1989); (4)
cause of epilepsy: presence of known cause for epilepsy as
symptomatic and absence as cryptogenic; (5) lateralization
of epileptiform discharges (spikes and sharp wave variants),
divided into four categories: left, right, bilateral, and none.
Our criteria for lateralization are detailed elsewhere
(Adachi et al., 1998); (6) the presence of mesial temporal
sclerosis (MTS) in brain magnetic resonance imaging
(MRI) scans: MTS was assessed qualitatively and MRI was
performed in accordance with our routine protocol (Adachi
et al., 2005); (7) family history of psychosis or any psy-
chotic disorders in their first-degree relatives (Kitamura
et al., 1984); and (8) intellectual function: mental retarda-
tion (full scale Wechsler Adult Intelligence Scale-Revised
IQ [FIQ] 70 or below); borderline intellectual functioning
(FIQ 71–84); or normal (FIQ 85 or above) in accordance
with the DSM IV (American Psychiatric Association,
1994).
Epilepsia, 53(6):1088–1094, 2012
doi: 10.1111/j.1528-1167.2012.03438.x
1090
N. Adachi et al.
As for episode-specific factors of each IIP episode, the
following were recorded.
1 Age at IIP episode
2 Frequency of habitual seizures around the episode, classi-
fied into six categories, that is, daily, weekly, monthly,
yearly, less than yearly, seizure free for more than 3 years
(Adachi et al., 1998)
3 The number of AEDs taken;
4 AED regimen (the number of AEDs or dosage) within
1 month before onset of the IIP episode, categorized into
increased, decreased, changed in multiple ways (some
increased and others decreased), or unchanged
5 Treatment with antipsychotic drugs (APDs), any APDs
taken before or after the beginning of the IIP episodes
were noted. Prescribing of APDs (timing, kinds, dosages)
was left to the patient’s consultant who made a decision
based on clinical demand in accordance with the treat-
ment strategy for IIP described above (Onuma, 1987; Ad-
achi, 2005)
6 APD time-lag, the duration between onset of the IIP epi-
sode and start of APD administration was noted.
Analysis
Differences in duration of the IIP episodes by each condi-
tion were analyzed with analysis of variance (ANOVA).
The relation between duration of the IIP episode and liner/
rank-ordered clinical factors was analyzed with Pearson’s or
Spearman’s rank correlation coefficient. Significant level
was set <0.05. Bonferroni correction was used to avoid risk
of multiple comparisons: that is, 0.05/8 (approximately
0.0063) patient’s characteristics and 0.05/6 (approximately
0.0083) for episode-specific factors of each IIP episode.
SPSS version 14.0 (SPSS Inc., Chicago, IL, U.S.A.) was
used for all statistical analyses.
Results
Episodes and patients profiles
In 155 patients with epilepsy (89 men and 66 women),
320 IIP episodes were observed in our institutions. The total
follow-up period of the 155 patients was mean 15.9 years
(standard deviation [SD] 10.6, median 13, range 1–55). The
duration between the initial IIP episode and the endpoint
was mean 12.2 years (SD 8.7, median 10, range 1–40). The
patients had mean 2.1 IIP episodes (SD 1.4, median 2, range
1–8); 68 patients had single IIP episodes and 87 patients had
multiple IIP episodes (12 patients developed IIP episodes
five times or more).
Age of onset of the IIP episode was mean 30.9 years (SD
10.5, median 28, range 14–66) and interval between onset
of epilepsy and that of the IIP episode was mean 18.3 years
(SD 9.0, median 17, range 0–55). Seizure frequencies at
onset of the IIP episode were as follows: seizure free in 68
patients, less than yearly in 26, yearly in 77, monthly in 68,
weekly in 63, and daily in 18. The number of AEDs taken
Epilepsia, 53(6):1088–1094, 2012
doi: 10.1111/j.1528-1167.2012.03438.x
was mean 2.2 (SD 1.3, median 2, range 0–7). In 47 IIP epi-
sodes (14.7%), AED regimen was changed within 1 month
before episode onset (increased in 20, decreased in 20, and
multiply changed in 7), whereas 273 episodes occurred on
the same AED regimen. Of the 320 IIP episodes, 48 epi-
sodes were treated without any APD throughout the episode
(non-APD group). The remaining 272 episodes were treated
with some APDs during the course of episodes (APD
group): 118 episodes occurred when the patients were on
prophylactic treatment with APDs and continued the treat-
ment throughout the episode (prophylactic APD group) and
154 episodes were treated with some APDs after their onset
(APD add-on group). Of the APD group, APD time-lag
between onset at the IIP episode and the time APDs were
started was mean 9.7 weeks (SD 25.9, median 0.6, range
0–221).
Of the 155 patients, age at onset of epilepsy was mean
12.4 years (SD 8.6, range 0–49) and age of onset of psycho-
sis was mean 28.5 years (SD 10.5, range 14–66). One hun-
dred thirty patients had partial epilepsies (PE): temporal
lobe epilepsy in 69, frontal lobe epilepsy in 31, parietal lobe
epilepsy in 7, occipital lobe epilepies in 6, multilobe/unde-
termined lobe epilepsy in 14, and benign epilepsy of child-
hood with centrotemporal foci in 3. Twenty-five patients
had generalized epilepsies (GE) idiopathic generalized epi-
lepsy (IGE) in 17, symptomatic generalized epilepsy (SGE)
in 3, and special epilepsy syndromes (e.g., progressive
myoclonus epilepsy) in 5. In addition to IIP episodes, seven
patients had separate postictal psychosis during their course
of illness (bimodal psychosis). The lateralization of scalp
EEG abnormalities was left in 60 patients, right in 60, and
bilateral in 35. MTS was found on the left in 19 patients, on
the right in 14, bilaterally in 4, and unremarkable in 118.
Intellectual functioning was normal in 61 patients, border-
line in 34, and in the range of mental retardation in 60.
The duration of the IIP episodes observed
The duration of all the 320 IIP episodes from the 155
patients was mean 82.7 weeks (SD 194.3, median 17, mode
2, range 0.2–1,874). The distribution of the duration of all
the IIP episodes was shown in Fig. 1: IIP episodes subsided
within 1 month in 82 episodes (25.6%), over 1 month
within 3 months (1–3 months) in 63 episodes (19.7%),
3–6 months in 43 (13.4%), 6–12 months in 39 (12.2%),
12–24 months in 36 (11.3%), 24–60 months in 25 (7.8%),
and 60 months or longer in 32 (10.0%). Thirty patients still
had IIP episodes at the last follow-up consultation: 12
patients continued experiencing IIP episodes at the study
endpoint, 8 were transferred to another institution, 5 discon-
tinued attending the clinic without notices, 2 had a sudden
unexpected death, and 3 committed suicide.
The duration of the longest IIP episodes of each patient
during the follow-up period was mean 150.1 weeks (SD
260.9, median 52, mode 4, range 0.6–1,874). The longest
IIP episodes were for 1 month or less in 17 patients
 1091
Duration of Interictal Psychosis

Table 1. Relations between episode-specific factors

and the duration (weeks) of each IIP episode
Duration (weeks)
Temporal factors (n) Mean, SD, median, range Statistics
Age at onset of IIP r = )0.186,
episodes (320) p = 0.001
Interval from onset r = )0.107,
of epilepsy (320) p = 0.056
Seizure frequency r = 0.026,
p = 0.647
Free (68) 65, 136.3, 13.5, 1–810
Less than yearly (26) 81.6, 159.7, 23.5, 2–754
Yearly (77) 77.1, 221.3, 17, 0.2–1,794
Monthly (68) 71.3, 114.6, 17.5, 0.2–546
Weekly (63) 129.1, 291.2, 15, 0.3–1,874
Daily (18) 56.0, 85.2, 21.5, 0.4–286
Antiepileptic drugs r = 0.099,
Figure 1. (AEDs): number (320) p = 0.079
IIP episodes subsided within 1 month (4 weeks) in 82 episodes Changes of AED regimens F = 1.51,
(25.6%), over 1 month within 3 months (5–13 weeks) in 63 p = 0.213
episodes (19.7%), 3–6 months (14–26 weeks) in 43 (13.4%), Increased (20) 39.5, 79.5, 10, 0.2–323
6–12 months (27–52 weeks) in 39 (12.2%), 12–24 months Decreased (20) 22.3, 28.1, 9.5, 0.4–95
Multiply changed (7) 168.4, 124.7, 4, 2–897
(53–104 weeks) in 36 (11.3%), 24–60 months (105–260 weeks)
Unchanged (273) 88.1, 201.7, 18, 0.2–1,874
in 25 (7.8%), and 60 months or longer (260 weeks+) in 32 Antipsychotic F = 4.38,
(10.0%). drugs (APDs) p = 0.037
Epilepsia ILAE Non-APD (48) 28.9, 38.0, 10.5, 0.2–173
APD (272) 92.2, 208.8, 17, 0.2–1,874
(11.0%), over 1 month to 3 months (1–3 months) in Timing of APD F = 5.32,
18 (11.6%), 3–6 months in 19 (12.3%), 6–12 months in 24 administration p = 0.022
APD prophylactic (118) 59.1, 195.0, 10, 0.2–1,874
(15.5%), 12–24 months in 25 (16.1%), 24–60 months in 21 APD add-on (154) 117.6, 216.0, 34, 1–1,794
(13.5%), and 60 months or longer in 31 (20.0%). APD time lag (272) r = 0.43,
p = 0.000
A longitudinal aspect of the duration of IIP episodes
Provided that the period of acute-episodic psychosis is
defined as a duration of 1 month or shorter, 17 (11.0%) of of the episode. The number of AEDs taken correlated
the 155 patients showed all episodes acute, 102 (65.8%) slightly with the IIP duration, although it did not reach a sta-
showed all episodes chronic, and 36 (23.2%) showed both tistically significant level. Changes of the AED regimen
acute and chronic episodes. When the maximum duration were not associated with the IIP duration. The IIP duration
for acute psychoses is set at 3 months, 35 (22.6%) of the differed significantly between the groups regarding APD
patients showed all episodes acute, 75 (48.4%) showed all treatment: the APD group had a longer duration than did the
episodes chronic, and 45 (29.0%) showed both acute and non-APD group. Among the 272 episodes with APD treat-
chronic episodes. When the maximum duration is set at ments, the prophylactic APD group had shorter duration
6 months or less, 54 (34.8%) of the patients showed all epi- than the APD add-on group. As APD time-lag prolonged,
sodes acute, 54 (34.8%) showed all episodes chronic, and 47 the IIP duration became longer significantly.
(30.3%) showed both acute and chronic episodes.
Relation between patients’ characteristics and the
Relation between episode-specific factors and the duration of the longest IIP episode
duration of each IIP episode Relations between individual characteristics and the lon-
Relations between episode-specific factors and the IIP gest duration were shown in Table 2. The duration of the
duration were shown in Table 1. Two age-related factors longest IIP episode showed a slight inverse correlation with
showed weak associations with the IIP duration. As age of age of onset of psychosis, whereas age of onset of epilepsy
onset of IIP episode advanced, the duration of the episode and the duration between onset of epilepsy and that of psy-
became shorter. When the interval between onset of epi- chosis were not correlated with each other. The IIP patients
lepsy and that of the IIP episode was longer, the duration of with a family history of psychosis showed a slightly longer
the episode tended to be slightly shorter, although it did not duration of the longest IIP episodes than did those without,
reach a significant level. Seizure frequency at onset of the although there was no significant difference. There was no
IIP episode did not correlate significantly with the duration difference in the longest duration between IIP patients with

Epilepsia, 53(6):1088–1094, 2012

doi: 10.1111/j.1528-1167.2012.03438.x
1092
N. Adachi et al.

Table 2. Relation between patients’ characteristics and the duration (weeks) of the longest IIP episode
Duration of the longest IIP
Clinical characteristics (n) Mean, SD, median, Range Statistics
Age of onset of epilepsy r = )0.04, p = 0.613
Age of onset of psychosis r = )0.15, p = 0.058
Duration between epilepsy-psychosis r = )0.03, p = 0.673
Family history of epilepsy F = 0.30, p = 0.586
Presence (12) 189.6, 506.0, 53, 4–1,794
Absence (143) 146.7, 232.0, 52, 0.6–1,874
Family history of psychosis F = 3.16, p = 0.077
Presence (14) 267.5, 464.2, 122, 1–1,794
Absence (141) 138.4, 231.0, 48, 0.6–1,874
Postictal psychotic episode F = 2.10, p = 0.150
Presence (7) 11.0, 8.7, 13, 1–23
Absence (148) 156.6, 265.3, 59, 0.6–1,874
Epilepsy type F = 3.14, p = 0.078
Partial epilepsies (130) 165.6, 279.5, 58, 1–1,874
Generalized epilepsies (25) 65.5, 82.0, 30, 0.6–334
Subtypes of partial epilepsies F = 1.98, p = 0.087
Temporal lobe epilepsy (69) 193.6, 348.2, 61, 1–1,874
Frontal lobe epilepsy (31) 83.1, 102.5, 38, 1–364
Parietal lobe epilepsy (7) 322.7, 207.5, 268, 36–650
Occipital lobe epilepsy (6) 242.7, 166.7, 245, 32–432
Multilobular/undetermined lobe epilepsy (14) 62.2, 88.5, 30, 6–320
BECCT (3) 368.0, 392.5, 234, 60–810
Subtypes of generalized epilepsies F = 0.46, p = 0.635
Idiopathic generalized epilepsies (17) 67.9, 67.5, 56, 0.6–238
Symptomatic generalized epilepsies (3) 24.3, 29.3, 11, 4–58
Progressive myoclonus epilepsies (5) 82.0, 141.8, 30, 3–334
Cause of epilepsy F = 0.02, p = 0.901
Symptomatic (57) 153.5, 289.6, 39, 2–1,794
Cryptogenic (98) 148.0, 244.3, 60.5, 0.6–1,874
EEG; lateralization of abnormalities F = 1.09, p = 0.339
Left (60) 172.4, 296.3, 54, 1–1,874
Right (60) 159.2, 270.3, 60.5, 1–1,794
Bilateral (35) 93.3, 154.7, 34, 0.6–810
MRI; mesial temporal sclerosis F = 0.04, p = 0.848
Presence (37) 156.6, 311.6, 52, 2–1,794
Absence (118) 147.2, 243.8, 53, 0.6–1,874
Intellectual functioning r = )0.02, p = 0.835
Normal (61) 136.7, 254.3, 60, 0.6–1,794
Borderline (34) 108.3, 163.6, 3–650
Mental retardation (60) 185.6, 306.5, 1–1,874

a family history of epilepsy and those without. There was no
difference in the longest duration between IIP patients with
a history of postictal psychosis and those without. Patients
with partial epilepsy tended to have a slightly longer duration
than patients with GE, although the difference was not sta-
tistically significant. There was no significant difference in
the lateralization of EEG abnormalities, MTS on MRI, or
intellectual functioning.
Discussion
The current study demonstrated that the duration and
related factors of a large number of IIP episodes showed a
broad spectrum, suggesting heterogeneous characteristics of
IIPs. Although the duration of IIP episodes tended to be
prolonged, the timing of APD treatment was strongly asso-
ciated with their duration. Because evidence on these issues
is scarce, the findings may provide invaluable information
from clinical and pathophysiologic perspectives.
The duration of each IIP episode
Mean duration of the episode was approximately
80 weeks, and one half of the IIP episodes lasted for
4 months or more, whereas their duration was distributed
widely from a day to more than 30 years (Fig. 1). This
broad distribution of the episode duration, one of the core
elements in the diagnostic process, can lead to difficulty in
understanding IIP by looking into the duration solely. This

Epilepsia, 53(6):1088–1094, 2012

doi: 10.1111/j.1528-1167.2012.03438.x

contrasts strikingly with the duration of postictal psychosis
(PIP), almost all of whose episodes end within a month
(Adachi et al., 2007).
This study considered only psychotic symptoms in accor-
dance with the conventional definition of IIP. Psychiatric
negative symptoms, for example, psychomotor retardations,
blunted affects, and emotional withdrawal in epilepsy
patients with psychosis have long been ignored. In his landmark
study, Slater et al. (1963) observed that a significant number
(40%) of IIP patients showed negative symptoms. These
findings were, however, not noticed for a long time until we
revisited this issue and reported the presence of psychiatric
negative symptoms in some patients with IIP (Adachi et al.,
2000; Adachi, 2006). Negative symptoms generally outlast
positive symptoms in patients with functional psychosis
(Lenzenweger & Dworkin, 1996). Furthermore, thought
disorders in patients with epilepsy with psychosis have not
been systematically studied. Among various thought disor-
der symptoms, delusions were clarified as disordered con-
tent (Cutting & Murphy, 1988); therefore, we were able to
assess them in this study. In contrast, the other symptoms of
thought disorder, that is, disordered forms, including disor-
der of the mechanisms of thinking and disorder of language
and speech (Cutting & Murphy, 1988), were not well evalu-
ated. If the duration of negative symptoms and/or thought
disorders was also measured, their duration of illness should
be prolonged.
A longitudinal aspect of IIP episodes
During the follow-up period (mean 10 years), approxi-
mately 90% of the patients had IIP lasting for more than
1 month. If the conventional duration-related classification
(Bruens, 1974; Sachdev, 1998) was used, the majority of IIP
episodes were classified into ‘‘chronic’’ psychosis. Two
thirds of the IIP patients had psychoses lasting more than
6 month, which fulfilled the DSM IV criteria for schizo-
phrenia regardless of the presence of negative symptoms.
Previous studies have also shown long-lasting psychosis in
many patients with epilepsy (Slater et al., 1963; Bruens,
1971; Onuma et al., 1991). Therefore, IIP tends to take a
long-lasting course, which is similar to other functional psy-
choses (White et al., 2009). On the other hand, the preva-
lence (10%) of brief psychotic disorder among the patients
with IIP appeared to be a little higher in comparison with
the reported figures of brief psychotic disorders, accounting
for 1–6% of all psychotic disorders (Susser & Wanderling,
1994; Perala et al., 2007). These findings may indicate the
heterogeneous nature of IIP phenomena.
The relation between episode-specific factors
and the duration of IIP
The use of APDs was significantly associated with the
duration of IIP. Particularly, those patients who were treated
with APDs during their IIP episodes (the APD add-on
group) showed the longest duration. Although the use of
1093
Duration of Interictal Psychosis
APDs was a consequence of psychopathologic as well as
psychosocial aspects of each episode, the use of APDs defi-
nitely reflects acuity and severity of the episodes (Adachi
et al., 2007). Among the episodes treated with APDs, earlier
APD administration shortened the IIP duration signifi-
cantly. This is similar to findings in functional psychosis
where prognosis is poorer if the first-episode psychosis
becomes prolonged without appropriate treatment (White
et al., 2009). These findings support empirical treatment
principles for IIP to administer APDs at an early stage of its
development (Onuma, 1987).
There were some aging effects on the duration of the
episodes: the younger the patient was at the onset of the IIP
episode, the longer was the duration. This is in line with
findings in studies of functional psychosis (Borga et al.,
1991; Haro et al., 1994), in that those who developed psy-
choses at a younger age showed more chronic and poorer
outcome. Likewise, IIP episodes tended to be shorter when
occurring after a long period since onset of epilepsy. This
may be another age-related finding, mirroring that the
duration became shorter as age advanced.
The relation between patients’ characteristics
and the longest duration of IIP
Although none of the clinical characteristics were signifi-
cantly associated with the longest duration of the IIP,
patients with a family history of psychosis or with onset of
psychosis at a younger age tended to have longer duration of
psychosis. These factors likely reflect individual vulnerabil-
ities to psychosis. People with a family history of psychosis
tend to have an increased risk of psychosis and to exhibit
their first psychotic symptoms earlier than do those without
(Nicholson & Neufeld, 1992; Albus & Maier, 1995).
Patients with early onset of psychosis are also known to
show poor outcome (Sato et al., 2004; Malla & Payne,
2005). In contrast, slightly longer IIP duration in patients
with partial epilepsy may be associated with brain damage
or epilepsy-related adversities (e.g., intractable seizures and
multiple AED administrations) (Adachi et al., 2010).
Limitation of the study
The current study had several limitations. First, whereas
most information used for analysis in this study was
retrieved from the database in that data were entered in a
prospective manner, some additional data were collected
retrospectively. Diagnosis and descriptions of psychopa-
thologies were recorded by qualified neuropsychiatrists,
who were proven reliable in our previous studies (Adachi
et al., 2000; Adachi, 2006); however, data collected retro-
spectively might have inconsistencies. Even if this was the
case, minor assessment instabilities cannot account for the
findings because data entry had been completed before
study-specific hypotheses were formulated or data analyses
were carried out, and the findings were obtained from our
large cohort of patients. Second, the duration of IIP episodes
Epilepsia, 53(6):1088–1094, 2012
doi: 10.1111/j.1528-1167.2012.03438.x
1094
N. Adachi et al.
could change with environment and treatment, particularly
with APDs. To our knowledge, evidence for effective IIP
treatment has been scarce and there has been no standard-
ized treatment guideline for IIP. Because regimen and doses
of APDs were determined on the basis of clinical needs, thus
not controlled, effects of APDs on the IIP duration could not
be analyzed. To confirm our findings, controlled studies
with the APD administration should be undertaken. Third,
because some IIP episodes did not remit at the last follow-
up consultation, mean duration can be longer than reported.
The proportion of the patients whose duration was opera-
tively determined, however, was fairly small; therefore, the
duration of IIP episodes that did not remit would not have
affected the findings significantly.
The current study showed the distribution of IIP duration
and its related factors. Our findings, in particular effects of
the timing of APD administration, would contribute to our
understanding of the nature of the IIP phenomena. Further
studies with prospective approaches are required to confirm
our results and develop more evidence-based strategies in
the treatment and management of IIPs.
Disclosures
None of the authors have any conflicts of interests. We confirm that we
have read the Journal’s position on issues involved in ethical publication
and affirm that this report is consistent with those guidelines.
References
Adachi N. (2005) Treatment for psychiatric symptoms of epilepsy patients.
Jpn J Psychiatry Treat 20(Suppl.):370–373. (Japanese).
Adachi N. (2006) Epilepsy and psychosis. Issues on clinical research in
epilepsy psychosis. Seishin Shinkeigaku Zasshi 108:260–265.
(Japanese).
Adachi N, Alarcon G, Binnie CD, Elwes RDC, Polkey CE, Reynolds EH.
(1998) Predictive value of interictal epileptiform discharges during
non-REM sleep on scalp EEG recordings for lateralization of epilepto-
genesis. Epilepsia 39:628–632.
Adachi N, Nishiwaki S, Onuma T, Murauchi S, Akanuma N, Ishida S,
Takei N. (2000) Interictal and postictal psychoses in frontal lobe
epilepsy: a retrospective comparison with psychoses in temporal lobe
epilepsy. Seizure 9:328–335.
Adachi N, Matsuura M, Hara T, Oana Y, Okubo Y, Kato M, Onuma T.
(2002) Psychoses and epilepsy: are interictal and postictal psychoses
distinct clinical entities? Epilepsia 43:1574–1582.
Adachi N, Kanemoto K, Muramatsu R, Kato M, Akanuma N, Ito M, Kawa-
saki J, Onuma T. (2005) Intellectual prognosis after status epilepticus in
adult epilepsy patients. Analysis with Wechsler Adult Intelligence
Scale-Revised. Epilepsia 46:1502–1509.
Adachi N, Ito M, Kanemoto K, Akanuma N, Okazaki M, Ishida S, Sekimot-
o M, Kato M, Kawasaki J, Tadokoro Y, Oshima T, Onuma T. (2007)
Duration of postictal psychotic episodes. Epilepsia 48:1531–1537.
Adachi N, Akanuma N, Ito M, Kato M, Hara T, Oana Y, Matsuura M, Ok-
ubo Y, Onuma T. (2010) Epileptic, organic, and genetic vulnerabilities
for timing of the development of interictal psychosis. Br J Psychiatry
196:212–216.
Albus M, Maier W. (1995) Lack of gender differences in age at onset of
familial schizophrenia. Schizophr Res 18:51–57.
Epilepsia, 53(6):1088–1094, 2012
doi: 10.1111/j.1528-1167.2012.03438.x
American Psychiatric Association. (1994) Diagnostic and statistical man-
ual of mental disorders. 4th ed. American Psychiatric Association,
Washington DC.
Borga P, Widerlov B, Cullberg J, Stefansson CG. (1991) Patterns of care
among people with long-term functional psychosis in three different
areas of Stockholm County. Acta Psychiatr Scand 83:223–233.
Bruens JH. (1971) Psychoses in epilepsy. Psychiatr Neurol Neurochir
74:174–192.
Bruens JH. (1974) Psychoses in epilepsy. In Vinken PJ, Bruyn G.W (Eds)
Handbook of clinical neurology. vol. 15. North Holland, Amsterdam,
pp. 593–610.
Commission on Classification and Terminology of the International League
Against Epilepsy. (1989) Proposal for revised classification of epilep-
sies and epileptic syndromes. Epilepsia 30:389–399.
Cutting J, Murphy D. (1988) Schizophrenia thought disorder. Br J Psychiatry
152:310–319.
Hafner H, an der Heiden W. (2003) Course and outcome of schizophrenia.
In Hirsch SR, Weinberger D (Eds) Schizophrenia. 2nd ed. Blackwell
Science Ltd, Oxford, pp. 101–141.
Haro JM, Eaton WW, Bilker WB, Mortensen PB. (1994) Predictability of
rehospitalization for schizophrenia. Eur Arch Psychiatry Clin Neurosci
244:241–246.
Kanemoto K, Kawasaki J, Kawai I. (1996) Postictal psychosis: a compari-
son with acute interictal and chronic psychoses. Epilepsia 37:551–556.
Kitamura T, Shima S, Sekino E, Kato M. (1984) Reliability study on Family
History-Research Diagnostic Criteria (FH-RDC) by using case vign-
ettes. Jpn J Soc Psychiatry 7:308–312. (Japanese).
Lenzenweger MF, Dworkin RH. (1996) The dimensions of schizophrenia
phenomenology. Not one or two, at least three, perhaps four. Br J Psy-
chiatry 168:432–440.
Malla A, Payne J. (2005) First-episode psychosis: psychopathology, quality
of life, and functional outcome. Schizophr Bull 31:650–671.
Matsuura M, Senzaki A, Terasaki O, Ohbayashi S, Matsushima E, Okubo
Y, Toru M. (1993) Classification of the clinical course of delusional
and/or hallucinatory states in epilepsy. Jpn J Psychiatry Neurol
47:363–365.
Mellers JDC, Adachi N, Takei N, Cluckie A, Toone BK, Lishman WA.
(1998) A SPET study of verbal fluency in schizophrenia and epilepsy.
Br J Psychiatry 173:69–74.
Nicholson IR, Neufeld RW. (1992) A dynamic vulnerability perspective on
stress and schizophrenia. Am J Orthopsychiatry 62:117–130.
Onuma T. (1987) Therapeutic approach to epileptic patients with psychiat-
ric symptoms. Jpn J Psychiatry Treat 2:553–560. (Japanese).
Onuma T, Adachi N, Hisano T, Uesugi S. (1991) 10-year follow-up study
of epilepsy with psychosis. Jpn J Psychiatry Neurol 45:360–361.
Perala J, Suvisaari J, Saarni SI, Kuoppasalmi K, Isometsa E, Pirkola S,
Partonen T, Tuulio-Henriksson A, Hintikka J, Kiesepp T, Hrknen T,
Koskinen S, Lçnnqvist J. (2007) Lifetime prevalence of psychotic and
bipolar I disorders in a general population. Arch Gen Psychiatry 64:19–28.
Sachdev P. (1998) Schizophrenia-like psychosis and epilepsy: the status
association. Am J Psychiatry 155:325–336.
Sato T, Bottlender R, Schroter A, Moller HJ. (2004) Psychopathology of
early-onset versus late-onset schizophrenia revisited: an observation
neuroleptic-nave patients before and after first-admission treatments.
Schizophr Res 67:175–183.
Slater E, Roth M. (1969) Mayer-Gross, Slater and Roth clinical psychiatry.
3rd ed. Bailliere Tindal, London.
Slater E, Beard AW, Glithero E. (1963) The schizophrenia-like psychoses
of epilepsy. Br J Psychiatry 109:95–150.
Susser E, Wanderling J. (1994) Epidemiology of nonaffective acute remit-
ting psychosis vs schizophrenia. Sex and sociocultural setting. Arch
Gen Psychiatry 51:294–301.
White C, Stirling J, Hopkins R, Morris J, Montague L, Tantam D, Lewis S.
(2009) Predictors of 10-year outcome of first-episode psychosis.
Psychol Med 39:1447–1456.
World Health Organization. (1992) The ICD-10 Classification of mental
and behavioural disorders: clinical descriptions and diagnostic guide-
lines. World Health Organization, Geneva.