Brain Advance Access published August 8, 2016

doi:10.1093/brain/aww196

BRAIN 2016: Page 1 of 11

| 1

Psychotic disorders induced by antiepileptic

drugs in people with epilepsy
Ziyi Chen,1,2 Ana Lusicic,3 Terence J. OBrien,1 Dennis Velakoulis,3 Sophia J. Adams3 and
Patrick Kwan1

1 Departments of Medicine and Neurology, The Melbourne Brain Centre, The University of Melbourne, The Royal Melbourne
Hospital, Victoria, Australia
2 Department of Neurology, The First Afliated Hospital, Sun Yat-sen University, Guangzhou, China
3 Melbourne Neuropsychiatry Centre, The University of Melbourne, The Royal Melbourne Hospital, Victoria, Australia

Received April 6, 2016. Revised May 30, 2016. Accepted June 20, 2016.
The Author (2016). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved.
For Permissions, please email: [email protected]

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

Antiepileptic drug treatment can induce psychosis in some patients. However, there are no agreed denitions or diagnostic criteria
for antiepileptic drug-induced psychotic disorder in the classication systems of either epileptology or psychiatry. In this study we
investigated the clinical spectrum of antiepileptic drug-induced psychotic disorder in patients with epilepsy. The medical records of
all patients with epilepsy who were diagnosed by a neuropsychiatrist as having a psychotic disorder at the Royal Melbourne
Hospital from January 1993 to June 2015 were reviewed. Data were extracted regarding epilepsy and its treatment, psychotic
symptoms prole and outcome. The diagnosis of epilepsy was established in accordance to the classication system of the
International League Against Epilepsy while that of psychotic disorder was made according to the Diagnostic and Statistical
Manual of Mental Disorders, 5th Edition and the proposal on neuropsychiatric disorders in epilepsy. Patients with antiepileptic
drug-induced psychotic disorder were compared to those with psychotic disorders unrelated to antiepileptic drugs assessed over the
same period (non-antiepileptic drug induced psychotic disorder group). Univariate comparisons were performed and variables with
a value of P 5 0.1 were selected for the multivariate logistic regression analysis. The records of 2630 in-patients and outpatients
with epilepsy were screened, from which 98 (3.7%) with psychotic disorders were identied. Among these, 14 (14.3%) were
diagnosed to have antiepileptic drug-induced psychotic disorder. Excluding one patient who developed psychosis after valproate
withdrawal, 76.9% in the non-antiepileptic drug induced psychotic disorder group were female and the percentage of temporal
lobe involvement was higher in the non-antiepileptic drug induced psychotic disorder group (69.2% versus 38.1%, P 5 0.05).
Current use of levetiracetam was higher in antiepileptic drug-induced psychotic disorder group (84.6% versus 20.2%, P 5 0.01)
while use of carbamazepine was higher in the comparator group (15.4% versus 44.0%, P 5 0.05). Multivariate logistic regression
conrmed four factors associated with antiepileptic drug-induced psychotic disorder: female gender, temporal lobe involvement and
use of levetiracetam, and a negative association with carbamazepine. Disorganized behaviours and thinking were more common in
the antiepileptic drug-induced psychotic disorder group (100% versus 72.6% and 76.9% versus 38.1%, respectively; P 5 0.05).
The percentage of continuous treatment with antipsychotic drugs was lower in the antiepileptic drug-induced psychotic disorder
group (15.4% versus 66.7%, P 5 0.01). No patients experienced a chronic course in antiepileptic drug-induced psychotic disorder
group whereas 40.5% did in non-antiepileptic drug induced psychotic disorder (P 5 0.05). Our ndings indicated that one in seven
patients with epilepsy who developed psychosis had antiepileptic drug-induced psychotic disorder. In these patients, female gender,
temporal lobe involvement and current use of levetiracetam were signicantly associated with antiepileptic drug induced psychotic
disorder compared to other types of psychosis, while carbamazepine had a negative association. Disorganized behaviours and
thinking were predominant in antiepileptic drug-induced psychotic disorder. Patients with antiepileptic drug-induced psychotic
disorder differed from non-antiepileptic drug-induced psychotic disorders in having better outcome.

| BRAIN 2016: Page 2 of 11

Z. Chen et al.

Correspondence to: Patrick Kwan

300 Grattan Street Parkville, Melbourne, Victoria, Australia 3050
E-mail: [email protected]

Keywords: epilepsy; psychosis; antiepileptic drug; AED-induced psychotic disorder

Abbreviations: AED = antiepileptic drug; AIPD = antiepileptic drug induced psychotic disorder; DSM-5 = Diagnostic and
Statistical Manual of Mental Disorders, 5th Edition

Introduction

Materials and methods

Patient sources
Eligible patients were identied from the Department of
Neurology at the Royal Melbourne Hospital between January
1993 and June 2015. Patients were mainly identied from those
admitted electively for a comprehensive epilepsy evaluation,
which included prolonged (5 days or more) video EEG monitoring, clinical assessment by epileptologists, psychiatric evaluation by neuropsychiatrists, and review of neuroimaging by
neuroradiologists. A minority (7.1%) of patients were identied
from the epilepsy outpatient clinics. All patients had undergone
formal psychiatric interview by a specialist neuropsychiatrist.
The inclusion criteria were: (i) onset of the psychotic disorders at 16 years or older; (ii) diagnosis of epilepsy; and
(iii) admission to hospital or attendance at epilepsy outpatient

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

Epilepsy is one of the most common neurological disorders

and ranks as the second leading neurological cause of
reduced disability-adjusted life-years (Murray et al.,
2012). Patients with epilepsy have increased vulnerability
to psychiatric co-morbidity including psychotic disorders
(Clarke et al., 2012; Rai et al., 2012), imposing additional
disease burden. For instance, in a Danish population-based
cohort study, the incidence of schizophrenia and schizophrenia-like psychosis in epilepsy patients was nearly 2.5
times and 3 times higher than in the general population,
respectively (Qin et al., 2005).
Among the various types of psychotic disorders in epilepsy,
antiepileptic drug (AED)-induced psychotic disorder (AIPD)
represents an iatrogenic, adverse drug reaction. Prevalence of
AIPD has been reported to range from 1.0% to 8.4% in
clinical trials of AEDs (Piedad et al., 2012). However, detailed analysis of the clinical prole of the psychotic episodes
was lacking in these studies, which tended to rely on screening questionnaires to ascertain psychiatric symptoms with
few patients undergoing structured interview by psychiatrists
(Clancy et al., 2014). Few studies have reported long-term
outcome of the psychotic episodes, as most randomized trials
reported the psychiatric events within the 1216 weeks of
observation (de la Loge et al., 2010).
Besides methodological limitations, advances in understanding AIDP have been further hampered by the lack of
agreed diagnostic criteria in the existing classication systems
(Lin et al., 2012). Although substance/medication-induced
psychotic disorder is dened in the Fifth Edition of
Diagnostic and Statistical Manual of Mental Disorders
(DSM-5), its applicability to AIPD may be questioned because
the pharmacodynamic mechanisms of AEDs may be different
from other substances or medications (American Psychiatric
Association, 2013). The International League Against
Epilepsy (ILAE) has published a classication scheme for
AED-induced psychiatric disorders, but it is not specic for
psychosis and covers other psychiatric manifestations, such as
affective disorders, that show a different clinical course
(Krishnamoorthy et al., 2007).
As a result of these limitations in knowledge, the management of AIPD in clinical practice is extremely challenging and not evidence-based. By denition, the denitive
diagnosis of AIPD can only be made retrospectively. In
theory, the most valid way to determine whether a given
AED is responsible in causing a particular adverse event

would be to withdraw the culprit drug and observe the

remission of symptoms, followed by rechallenging with
the medication and observing symptom relapse (Edwards
and Aronson, 2000). This approach, however, is rarely
practical in the clinical epilepsy setting, particularly for psychiatric adverse effects. The diagnosis is further compounded by the predisposition towards AIPD in people
with history of psychiatric illnesses (Trimble et al., 2000;
Weintraub et al., 2007). In some cases the episode of AIPD
can resemble recurrence of previous primary psychotic disorder. Therefore, when a patient with epilepsy develops
psychotic symptoms, it is challenging to determine at presentation whether the psychosis is AED induced or not.
Misdiagnosing AIPD as primary psychotic disorder may
lead to inappropriate management, including continuation
of the culprit AED and additional treatment with antipsychotic drugs. Often, the psychotic symptoms of AIPD
persist in a uctuating manner as long as the AED is continued (Amerincan Psychiatric Association, 2013). The patient may endure both the adverse effects of the AED and
potential exacerbation of epilepsy by antipsychotic drug
therapy (Lin et al., 2012). Therefore, identication of reliable factors at presentation that help to differentiate AIDP
from other forms of psychosis in epilepsy is needed.
In this study we aimed to identify these factors by investigating the clinical spectrum of AIPD in patients with epilepsy who presented with psychotic symptoms, including
the clinical features of the epilepsy, AED treatment, the
psychotic symptoms and outcome.

Antiepileptic drug-induced psychosis

BRAIN 2016: Page 3 of 11

| 3

Psychotiic disorder*

Specic to Epilepsy

bstance/medicaaon induced
Sub
psychoc dissorder*

Yes

o
No

Cormorb
bid schizophren
nia
/schizophreniform disord
der#

Induced byy AED

Psycho
osis of epilepsyy

Interictal psycho
osis

Yes

Posctal psychosiis

of epilepsy#

AEED-induced psyychoc
disorder

No

Other
sub
bstance/medicaaoninduced
psycho
i
oc
disorder

Figure 1 Diagnosis scheme of psychotic disorders related to epilepsy modified from DSM-5 and the proposal by ILAE
Commission on Psychobiology of Epilepsy. *As per DSM-5 (American Psychiatric Association, 2013); #as per ILAE proposal
(Krishnamoorthy et al., 2007); zdefined in this study.

Study procedure and diagnostic

approach
The study was approved by the Clinical Research and Ethics
Committee of the Royal Melbourne Hospital (HREC No:
2002.232). Information regarding individual demographic
data, the clinical manifestations of epilepsy and psychosis, prior
psychiatric history, AED usage, and outcomes of epilepsy and
psychosis was retrieved from the medical records using a
standardized case report form. The data were reviewed by an
epileptologist (Z.C.) and a neuropsychiatry fellow (A.L.) who
jointly conrmed the diagnosis of epilepsy and psychotic disorder.
Figure 1 illustrates the diagnostic scheme for the various
types of psychotic disorders in relation to epilepsy in the patients. First, the diagnosis of psychotic disorders was established. As per the ILAE proposal (Krishnamoorthy et al.,
2007), psychotic disorders specic to epilepsy were then classied as psychosis of epilepsy, including interictal psychosis of
epilepsy, postictal psychosis, and AIPD. Psychotic disorders
unrelated to the underlying epilepsy were classied as comorbid schizophrenia/schizophreniform disorder. As AIPD was not
specically dened in the ILAE proposal, its denition was
adopted from the criteria for substance/medication-induced
psychotic disorders in DSM-5, such that if the offending
drug was an AED, the episode was classied as AIPD.
Otherwise it was classied as other substance/medicationinduced psychotic disorder.

Clinical assessments and definitions

According to the classication system of the ILAE, seizures
were classied as generalized or focal. Focal seizures were further classied depending on whether there was impairment of
consciousness or evolution to bilateral convulsion (Berg et al.,
2010). Epilepsy syndromes were broadly classied as genetic,
structural/metabolic, and epilepsy of unknown cause. Specic
structural abnormalities of interest were hippocampal sclerosis,

brain tumour and malformations of cortical development.

Temporal lobe involvement was dened as the epileptogenic
lesion locating in temporal lobe with or without the involvement of other lobes. Drug resistance was dened as the failure
of two appropriately chosen and tolerated AED schedules to
maintain seizure freedom (Kwan et al., 2010).
According to the classication system DSM-5, the diagnosis
of psychotic disorders requires the presence of delusions or
hallucinations, plus possible disorganized thinking and grossly
disorganized or abnormal motor behaviour. Either delusions
or hallucinations must be present and the duration of the
psychotic episode must last at least 1 day. Delusions may be
persecutory, referential, somatic, religious, or grandiose delusions. The characteristics of hallucinations, such as auditory,
visual or tactile hallucinations were recorded in the medical
notes. Disorganized thinking was characterized by derailment
or loose associations, tangentiality, incoherence or word
salad. Grossly disorganized behaviours reported in our
cohort consisted of aggressive behaviour and unusual social
behaviours, such as socially or sexually inappropriate behaviours, e.g. talking to oneself in public, obscene language, or
exposing oneself to others.
The relationship of the psychotic disorder to the patients
underlying epilepsy were established in accordance with the
proposal by ILAE (Krishnamoorthy et al., 2007). Interictal
psychosis of epilepsy was dened as psychotic episodes in accordance with the criteria of psychosis in DSM-5 and independent from seizures (Table 1). Post-ictal psychosis was
dened as psychotic episodes after a lucid interval (up to
48 h) following a cluster of seizures. Comorbid schizophrenia/schizophreniform disorder was diagnosed under DSM-5.
In substance/medication-induced psychotic disorder, either delusions or hallucinations must be present and the psychotic
symptoms develop during or soon after the exposure to a substance or medication or the withdrawal of that substance. The
severity must impair the patients social or occupational function. When the offending agents were AEDs, the disorder was
classied as AIPD.
For the analysis of outcome, the clinical courses were categorized as following: (i) a single episode was dened as the
duration of psychosis of longer than 1 day; (ii) a relapse was
dened as recurrent episodes within 1 year after remittance of

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

clinic for psychotic symptoms. Patients were excluded if they

had (i) psychogenic non-epileptic seizures; (ii) psychotic symptoms as part of the ictal semiology; or (iii) an organic illness
with known psychiatric manifestations, e.g. Wilsons disease.

| BRAIN 2016: Page 4 of 11

Z. Chen et al.

Table 1 Diagnosis criteria of psychotic disorders in epilepsy used in this study

Diagnosis criteria

AIPD

Delusions, hallucinations, disorganized thinking and grossly disorganized or abnormal motor behaviour.
At least delusion or hallucination must be present.
The duration of the psychotic episodes lasted at least 1 day.
The severity reached the level that impaired patients social or occupational function.
The psychotic symptoms developed during or soon after the exposure to an AED or the withdrawal.

Interictal psychosis of epilepsy

Delusions, hallucinations, disorganized thinking and grossly disorganized or abnormal motor behaviour.
At least delusion or hallucination must be present.
The duration of the psychotic episodes lasted at least 1 day.
The severity reached the level that impaired patients social or occupational function.
Psychotic episodes are independent with seizures.

Post-ictal psychosis

Delusions, hallucinations, disorganized thinking and grossly disorganized or abnormal motor behaviour.
At least delusion or hallucination must be present.
The duration of the psychotic episodes lasted at least 1 day.
The severity reached the level that impaired patients social or occupational function.
The psychotic episodes occur after a lucid interval following clusters of seizures.

Comorbid schizophrenia/schizophreniform disorder

Delusions, hallucinations, disorganized thinking and grossly disorganized or abnormal motor behaviour.
At least delusion or hallucination must be present.
The duration of the psychotic episodes lasted at least 1 month.
The severity reached the level that impaired patients social or occupational function.
No distinguishing features separate it from those seen in general population.

Other substance/medicationinduced psychotic disorder

Delusions, hallucinations, disorganized thinking and grossly disorganized or abnormal motor behaviour.
At least delusion or hallucination must be present.
The duration of the psychotic episodes lasted at least 1 day.
The severity reached the level that impaired patients social or occupational function.
The psychotic symptoms developed during or soon after the exposure to substance/medication or the withdrawal except AED.

longer than 2 months; and (iii) a chronic course was dened as

duration of the psychotic state of over 1 year without remittance for more than 2 months (Matsuura et al., 2000).

Statistical analysis
Data were presented as n (%) for categorical/qualitative variables
or mean  standard deviation (SD) or median (interquartile
range, IQR) for continuous/quantitative variables. Cases were
patients with AIPD. Controls were the patients with epilepsy
and psychotic disorders unrelated to AEDs assessed over the
same period (non-AIPD group). Clinical variables of epilepsy
and psychotic disorder were compared between the AIPD
group and the non-AIPD group. Univariate comparisons were
performed with t-test, 2 test or Fishers exact test, as appropriate. Univariate logistic regression analyses were performed to
calculate the odds ratios of the variables. Variables with
P 5 0.1 were selected for multivariate logistic regression analysis.
P 5 0.05 was considered to be statistically signicant. Statistical
analyses were carried out using the statistical software package
SPSS 20.0.

Results
Patient characteristics
A total of 98 patients (53 male) with epilepsy who had
experienced psychotic disorders were identied, 22 of

whom had been reported previously (Adams et al., 2008).

Six patients developed psychosis during hospitalization for
video EEG monitoring while the remaining 92 experienced
the psychotic episode at times separate from the monitoring
admission. The median age of onset of epilepsy was 18.5
years (IQR 931) and the median age of onset of psychosis
was 34.5 years (IQR 2745). Seventy-nine (80.6%) patients
had focal onset seizures and 19 (19.4%) had generalized
onset seizures (Table 2). The epilepsy was classied as genetic in 16 (16.3%), structural/metabolic in 59 (60.3%), and
of unknown cause in 23 (23.5%).
The psychosis was classied as AIPD in 14 (14.3%) patients and unrelated to AED therapy in the other 84
(85.7%). The latter included interictal psychosis of epilepsy
in 33 (33.7%) patients, post-ictal psychosis in 25 (25.5%),
comorbid schizophrenia/schizophreniform disorder in 19
(19.4%), and psychotic disorder induced by substances or
medications other than AEDs in seven (7.1%).

Patients with antiepileptic druginduced psychotic disorder

Table 3 shows the clinical features of the 14 patients with
AIPD. The majority (10 of 14) of patients had temporal
lobe involvement in their seizures with a variety of pathologies. In these patients the most common hallucinations
were auditory and visual. Two patients reported tactile

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

Groups

Antiepileptic drug-induced psychosis

BRAIN 2016: Page 5 of 11

Table 2 Diagnosis classification of epilepsy and psychotic disorders

Diagnosis
Epilepsy
Genetic epilepsy
Structural/metabolic epilepsy
Epilepsy with unknown cause
Psychotic disorders
AEDs-induced psychotic disorders
Interictal psychosis of epilepsy
Post-ictal psychosis
Comorbid schizophrenia/schizophreniform disorder
Other substance/medication-induced psychotic disorder
Total

n (%)
16 (16.3)
59 (60.3)
23 (23.5)
14 (14.3)
33 (33.7)
25 (25.5)
19 (19.4)
7 (7.1)
98 (100)

criteria. Furthermore, although antipsychotic treatment

with olanzapine was prescribed, the psychotic symptoms
resolved without continuous use of olanzapine. Hence,
this case may be diagnosed as AIPD. As psychosis that
develops after AED withdrawal is conceptually distinct
from that induced by drug initiation or dose escalation,
Patient 13 was excluded from the AIPD group in the following statistical comparisons.

Comparison between antiepileptic

drug-induced and non-antiepileptic
drug-induced psychotic disorder
Epilepsy and treatment related factors
The epilepsy and AED-treatment related factors are listed
in Table 4. There were more females in the AIPD group
(76.9%) compared with the non-AIPD group (41.7%;
P 5 0.05). There was no difference in age of onset of
psychosis between the two groups, nor was there difference
in seizure types.
In the analysis of aetiology classication, AIPD was more
often associated with structural/metabolic epilepsy (84.6%
versus 56.0% in non-AIPD group, P 5 0.05). There was a
trend of higher proportion of patients having brain tumour
in the AIPD group but the difference did not reach statistical signicance. In the AIPD group, two patients had craniopharyngioma and one had meningioma. In the nonAIPD group, there were two patients with astrocytoma
and one with metastasis of ovarian cancer. Tumour type
was unknown in the other two patients. Comparisons of
other common causes, including hippocampal sclerosis,
malformations of cortical development, brain traumatic
injury and cerebral vascular disease, showed no signicant
differences.
There was no difference in lateralization of seizure focus
between AIPD and non-AIPD patients. However, the percentage of temporal lobe involvement was higher in the
AIPD group [69.2% versus 38.1% in non-AIPD group,
odds ratio (OR) 4.063, P 5 0.05].
Two (15.4%) patients in the AIPD group had a history
of psychiatric disorder prior to epilepsy onset while 24
(28.6%) in the non-AIPD group did. Interestingly, both
cases in the AIPD group were classied as substance/medication-induced psychosis (antihistamine and overdose of
weight-loss medication, respectively) while only two cases
in the non-AIPD group experienced prior substance/medication induced psychosis (marijuana and steroid, respectively). In the latter group, another two patients had
depression, one had antisocial personality disorder and
the remaining 19 had comorbid schizophrenia/schizophreniform disorder. The differences in prior psychiatric history
in general or that specically related to substance/medication between the two groups were not statistically signicant. Besides epilepsy and psychotic disorders, four
(23.1%) patients in the AIPD group had other medical
comorbidities,
including
pan-hypopituitarism,
liver

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

hallucinations. The most prevalent type of delusion was

persecutory. Other delusions reported by the patients
included referential, religious, grandiose and somatic. All
patients presented with disorganized behaviours, such as
aggressive and unusual social behaviours. Patient 1 had a
previous history of psychosis induced by an overdose of
weight-loss medication. A similar history was reported by
Patient 4 after taking an antihistamine.
Levetiracetam was the most common AED taken by patients with AIPD, either as monotherapy or in combination
with other AEDs, accounting for 8 (57.1%) of the 14 cases.
Three (21.4%) were taking lamotrigine. Two (14.3%) patients experienced two episodes of psychosis associated
with levetiracetam and topiramate separately. One (7.1%)
case was induced by the withdrawal of valproate. The maximal dose of levetiracetam used in the AIDP patients varied
from 500 mg/day to 3000 mg/day (median 2000 mg/day).
Notably, relatively low doses (500 to 1000 mg/day) were
used in the four patients who developed AIPD on
monotherapy levetiracetam. Among the three cases with
lamotrigine induced AIPD, the maximal doses ranged
from 100 to 600 mg/day. For topiramate, one patient
took 400 mg/day and the other developed psychosis while
taking 150 mg/day.
In 11 patients the psychotic symptoms resolved after
withdrawal of the culprit AEDs. Two patients and one patient in one of her episodes recovered after reducing the
dosage of the culprit AEDs. The duration of the AIPD episodes was less than 7 days in nearly half of the patients.
Although eight patients needed treatment with antipsychotic drugs to control the psychiatric symptoms, most of
them did not require antipsychotic medication for longer
than 1 month.
Patient 13 of AIPD associated with valproic acid withdrawal was established in accordance to diagnosis criteria
by ILAE. It is possible this was a primary psychotic disorder, which relapsed because of the abrupt withdrawal of
valproic acid therapy. Review of the record showed no
previous history of psychosis in this patient. The time interval of drug withdrawal to onset of psychosis and that of represcription to symptom resolution fullled the diagnosis

| 5

F
F

M
F

M
F

2
3

7
8

10
11

12

13

14

16

43

5
23

12

55
88

28

10

7
9

22

Age at
epilepsy
onset

29

27

58

45
23

37

55
89

28

36

36

34
38

38

Age at
psychosis
onset

Tactile

Religious

Aggressive

Aggressive

Aggressive

Auditory,
Referential
tactile
Auditory, visual Referential
somatic
Auditory
Persecutory

Aggressive

Aggressive, un- +
usual social

Persecutory

Auditory

Induced by
antihistamine

Induced by
overdose
of diet
pills

12/m
3/w

1/m

2/m

12/w
2/d
1 (3 m after
stroke)
3/y

1/m
1/612 m

24/d
12/d
1/w

No
L TPO

RT

LT

LF
No
L TFP
L TF

No
R F

R TF
R TFP
LT

Haemochromatosis
Infarction
Craniopharyngioma

Astrocytoma
Unknown

LEV

TPM,2004
LEV,2005
LMT,1996
LMT,2004
LMT

54 d

10 d
3m
33 d
68 d
21 d

1m

LEV

Adjustment
of
culprit
AEDs

600 mg to 400 mg

Olanzapine

1
1

PRM, CLZ

34 m

10 m
1m
8y
1m
6m

(continued)

Droperidol,
haloperidol

Risperidone

CLZ
CLZ
PHT

32 m

Olanzapine

5
1

1
1

2
2

Number of
present
AEDs

Continuous Follow-up Outcome

APDs
duration
after
AIPD

5
1

1
1

3
3

Number of
previous
AEDs

Duration APDs
index
psychosis

3000 3000 mg to 2000 mg CBZ

600

400 400 mg to 100 mg

2000 Withdrawn
600 Withdrawn

1000 Withdrawn

Interval DMax
to
(mg)
psychosis
onset

Melanoma

Motor vehicle accident

Cavernous malformation

Craniopharyngioma

HS

FCD

Presumed genetic
Porencephaly

Combined
AEDs

34/w

RT

HS

Seizures
frequency
before
psychosis

Localization
of the
lesion(s)

Cause of
epilepsy

Culprit
AEDs

Focal seizure with impairment of consciousness or

awareness
Myoclonus seizure, generalized tonic-clonic seizure
Focal seizure with impairment of consciousness or
awareness, focal seizures evolving to a bilateral convulsive seizure
Focal seizure with impairment of consciousness or
awareness, focal seizures evolving to a bilateral convulsive seizure
Focal seizure with impairment of consciousness or
awareness, focal seizures evolving to a bilateral convulsive seizure
Focal seizure with impairment of consciousness or
awareness, non-convulsion status epilepticus
generalized tonic-clonic seizure
Focal seizure with impairment of consciousness or
awareness
Focal seizure with impairment of consciousness or
awareness, focal seizures evolving to a bilateral convulsive seizure
Myoclonus seizure, generalized tonic-clonic seizure
Focal seizure without impairment of consciousness or
awareness, focal seizures evolving to a bilateral convulsive seizure
Focal seizure with impairment of consciousness or
awareness
Atonic, focal seizures evolving to a bilateral convulsive
seizure
Focal seizure with impairment of consciousness or
awareness, focal seizures evolving to a bilateral convulsive seizure

Seizure types

Disorganized Previous
thinking
psychotic
history

Disorganized
behaviour

Patient Hallucination Delusion

ID

Psychotic symptoms and antipsychotic treatment

Gender

Patient
ID

| BRAIN 2016: Page 6 of 11

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

Sociodemographic and clinical features

Table 3 Sociodemographic and clinical features of the 14 cases with AIPD

6
Z. Chen et al.

CBZ = carbamazepine; d = day; F = frontal; L = left; m = month; P = parietal; O = occipital; T = temporal; y = year; LMT = lamotrigine; LEV = levetiracetam; TPM = topiramate; VPA = valpropic acid.
Patient 5 experienced the relapse of psychosis 2 years after the episode of AIPD. The relapse was caused by non-compliance of AEDs and her divorce, but no epileptic seizures had been observed before the relapse.
Duration of index psychosis: 1 = 17 d; 2 = 830 d; 3 = 16 m; 4 = 46 m.
Outcome: 1 = a single episode; 2 = a relapse; 3 = a chronic course.

1
1
Cease of VPA 28 d
LEV
30 d
Visual
Visual
13
14

Persecutory
Persecutory,
referential

Unusual social
Unusual social

Auditory
Auditory
11
12

+
+

1000 Represcribed
2000 Withdrawn

Olanzapine

3
1
CBZ, LMT
CBZ, CLB

11 y
6m

1
1

2
3
Withdrawn
Withdrawn
500
100
23 d
21 d
LEV
LMT

+
+

Withdrawn

Aggressive,
unusual social
Aggressive
Unusual social
Persecutory
Auditory
10

Persecutory,
grandiose
Religious

Persecutory

Aggressive
Aggressive
Aggressive

Visual

7
8
9

+
+
+

Withdrawn
Withdrawn
Withdrawn

1000
1000
1250
150
2000
2d
9d
5d
6m
10 + d
LEV
LEV
LEV,2008
TPM,2010
LEV

+
Risperidone
3

5y
7m

1
1
2
Olanzapine

2
1
1

14 m
3m
6y
4y
6y

Risperidone
2

VPA, CLB,
PHT

VPA
VPA
VPA, LMT,
TPM, CLZ

LEV
2000 Withdrawn
13 d
LEV

Aggressive
Auditory
6

Interval DMax
to
(mg)
psychosis
onset
Disorganized
behaviour

Disorganized Previous
thinking
psychotic
history

Culprit
AEDs

Adjustment
of
culprit
AEDs

Duration APDs
index
psychosis

2m

Table 4 Manifestations and treatment of epilepsy in the

cohort of patients with AIPD compared with the cohort
of patients with non-AIPD

Patient Hallucination Delusion

ID

Psychotic symptoms and antipsychotic treatment

| 7

Variables, n(%)

AIPD
(n = 13)a

Non-AIPD
(n = 84)

OR

P-value

Gender: F
Age of onset of epilepsy
Seizure types
Generalized seizures
Focal seizure with impairment
of consciousness or awareness
Focal seizures evolving to a
bilateral convulsive seizure
Aetiology
Genetic
Structural/metabolic
Hippocampal sclerosis
Brain tumour
Malformations of
cortical development
Unknown cause
Lateralization
Left involved
Right involved
Bilateral
Localization
Temporal lobe involved
Febrile convulsion
History of prior psychiatric
disorders
History of prior medication/
substance-induced psychiatric
disorders
Family history of epilepsy
Drug resistance
Brain surgery
Number of previous AEDs
Number of present AEDs

10(76.9%)
16(928)

35(41.7%)
19.5(1031.5)

4.667
1.003

0.033b
0.790

3(21.6%)
9(69.2%)

18(21.4%)
48(57.1%)

0.909
1.688

0.569c
0.549c

6(46.2%)

51(60.7%)

0.555

0.321

1(7.7%)
11(84.6%)
2(15.4%)
3(23.1%)
2(15.4%)

15(17.9%)
47(56.0%)
17(20.2%)
5(6.0%)
12(14.3%)

0.383
4.330
0.717
4.740
1.091

0.323c
0.044b,c
0.510c
0.072c
0.596c

1(7.7%)

22(26.2%)

0.235

0.181c

6(46.2%)
4(30.8%)
3(23.1%)

33(39.3%)
29(34.5%)
22(26.1%)

0.755
0.843
0.441

0.638
0.529c
0.187c

9(69.2%)
2(15.4%)
2(15.4%)

32(38.1%)
4(4.8%)
24(28.6%)d

3.656
3.636
0.455

0.035b,c
0.183c
0.504c

2(15.4%)

2(2.4%)

7.455

0.086c

0(0%)
10(76.9%)
5(38.5%)
3(25)
2(13)

9(10.7%)
55(65.5%)
25(29.8%)
2(13)
2(13)

0.569
1.475
1.336
1.415

0.258a
0.317c
0.369
0.068
0.148

a
Excluding the patient who developed psychosis after withdrawal of valproate (Case 13
in Table 3).
b
Statistically significant.
c
Fishers Exact Test.
d
Among the 24 cases, two had marijuana- or steroid-induced psychotic disorder, two
had depression, one had antisocial personality disorder and the remaining 19 had
comorbid schizophrenia/schizophreniform disorder.

cirrhosis, hypertension and type 2 diabetes mellitus, while

in the non-AIPD group, 31 (36.9%) had medical comorbidities including interstitial nephritis, non-Hodgkins
lymphoma, type 1 diabetes mellitus, asthma and myocardial infarction (OR = 0.722, P = 0.429).
A similar proportion of patients were drug-resistant in
the two groups. However, patients in the AIPD group
had been treated with a greater number of prior AEDs
before the index episode of psychosis, compared to the
non-AIPD group, although the difference was not statistical
[3(IQR: 25) versus 2(IQR: 13), P = 0.068].
Table 5 lists the AEDs taken by the patients during the
psychotic episodes. In the AIPD group, levetiracetam was
the most commonly used AED, followed by lamotrigine
and valproate. In the non-AIPD group, carbamazepine, valproate and phenytoin were most commonly used. Use of
levetiracetam was higher in AIPD group (84.6% versus

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

Table 3 Continued

BRAIN 2016: Page 7 of 11

Combined
AEDs

Continuous Follow-up Outcome

APDs
duration
after
AIPD

Antiepileptic drug-induced psychosis

| BRAIN 2016: Page 8 of 11

Z. Chen et al.

Table 5 AEDs currently used during by the patients

with epilepsy the episode of psychosis
Drug
Valproic acid
Carbamazepine
Phenytoin
Primidone
Levetiracetam
Lamotrigine
Topiramate
Clonazepam

AIPD
(n = 13)a
4(30.8%)
2(15.4%)
2(28.6%)
1(7.7%)
11(84.6%)
4(30.8%)
3(23.1%)
3(23.1%)

Non-AIPD
(n = 84)
29(34.5%)
37(44.0%)
24(28.6%)
2(2.4%)
17(20.2%)
17(20.2%)
9(10.7%)
7(8.3%)

OR
0.843
0.231
0.455
3.417
21.676
1.752
2.500
3.300

P-value
b

0.529
0.044b,c
0.262b
0.354b
0.001b,c
0.297
0.201b
0.130b

a
Excluding the patient who developed psychosis after withdrawal of valproate (Patient
13 in Table 3).
b
Fishers Exact Test.
c
Statistically significant.

Variables

OR

95% CI

P-value

Female gender
Structural/metabolic epilepsy
Brain tumour
Temporal lobe involvement
History of prior substance/
medication-induced
psychotic disorder
Number of previous AEDs
Current use of levetiracetam
Current use of carbamazepine

26.440
2.504
1.118
27.201
5.314

1.45779.731
0.26423.743
0.06619.069
1.65547.105
0.09014.974

0.027
0.424
0.938
0.021
0.423

0.6641.797
3.730121.431
0.0020.454

0.727
0.004
0.011

Excluding the patient who developed psychosis after withdrawal of valproate (Patient
13 in Table 3).
CI = confidence interval.

20.2%, P 5 0.01) while use of carbamazepine was higher

in the non-AIPD group (15.4% versus 44.0%, P 5 0.05).
Eight factors with P 5 0.1 from Tables 4 and 5 were
selected for the multivariate analysis. Multivariate logistic
regression conrmed four factors associated with AIPD:
female gender, temporal lobe involvement and current use
of levetiracetam, and a negative association with carbamazepine (Table 6).

Psychiatric manifestations and outcome

Table 7 summarizes the clinical manifestation of psychotic
disorders and the outcome. There were no signicant between-group differences observed with regard to age of
onset of psychoses, or the prevalence of hallucinations or
delusions. Disorganized behaviours and thinking were more
common in the AIPD group compared to non-AIPD group
(100% versus 72.6% and 76.9% versus 64.3%, respectively; P 5 0.05 for both). There were no signicant differences between the groups in co-morbid depressive mood,
anxiety or cognitive function. The duration of psychotic

Age of onset of psychosis,

years
Interval of epilepsy to
psychosis, years
Follow-up duration after
psychotic episode, years
Hallucination
Auditory hallucination
Visual hallucination
Tactile hallucination
Delusion
Persecutory delusion
Referential delusion
Somatic delusion
Religious delusion
Grandiose delusion
Grossly disorganized or
catatonic behaviour
Aggressive
Unusual social
Disorganized thinking
Depressive mode
Anxiety
Cognitive impairment
Duration of the index
psychosis
1:17 d
2:830 d
3:16 m
4: 46 m
Family history of
psychotic disorders
APD
More than one APD
Continuous treatment
with APDs
Outcome
1: A single episode
2: Recurrent episodes
3: Chronic course

AIPD
(n = 13)a

Non-AIPD
(n = 84)

OR

P-value

37(3445)

34(2346.5)

1.018

0.290

16(127)

13.5(1.523)

1.020

0.365

1.2(0.54.3)

2.7(1.47.3)

0.858

0.127

11(84.6%)
8(61.5%)
3(23.1%)
2(15.4%)
11(84.6%)
7(53.8%)
3(23.1%)
1(7.7%)
2(15.4%)
2(15.4%)
13(100%)

67(79.8%)
58(69.0%)
15(17.9%)
2(2.4%)
69(82.1%)
52(61.9%)
21(25.0%)
4(4.8%)
6(7.2%)
4(4.8%)
61(72.6%)

1.396
0.717
1.380
7.455
1.196
0.718
0.900
1.646
2.333
3.636

0.510b
0.402
0.449b
0.086b
0.593b
0.396
0.594b
0.525b
0.295b
0.183b
0.034b,c

11(84.6%)
4(30.8%)
10(76.9%)
2(15.4%)
2(15.4%)
5(38.5%)

54(64.3%)
7(8.3%)
32(38.1%)
22(26.2%)
17(20.2%)
37(44.0%)

3.056
4.889
5.417
0.512
0.717
0.794

0.209a
0.038c
0.014b,c
0.509b
0.510b
0.473

6(46.2%)
5(38.5%)
2(15.4%)
0
1(7.7%)

19(22.6%)
13(15.5%)
23(27.4%)
29(34.5%)
7(8.4%)

0.905

0.013c

0.705b

7(53.8%)
1(7.7%)
2(15.4%)

68(81.0%)
22(26.2%)
56(66.7%)

0.275
0.235
0.091

0.030c
0.131b
0.001b,c

7(53.8%)
6(42.9%)
0

33(39.3%)
17(20.2%)
34(40.5%)

0.011c

a
Excluding the patient who developed psychosis after withdrawal of valproate
(Patient13 in Table 3).
b
Fishers Exact Test.
c
Statistically significant.
APD = antipsychotic drug.

episodes were less than 1 week in 42.9% of patients in

the AIPD group compared with 22.6% in the non-AIPD
group (P 5 0.05).
Fewer patients with AIPD were treated with antipsychotic
drugs compared with patients with other psychotic disorders (53.8% versus 81.0%, P 5 0.05). Only one patient
with AIPD was treated with more than one antipsychotic
drug. The proportion of patients taking continuous antipsychotic treatment was lower in the AIPD group than in
the non-AIPD group (15.4% versus 66.7%, P 5 0.01).
More patients in the AIPD group experienced a single episode (53.8% versus 39.3%) and fewer experienced

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

Table 6 Multivariate logistic regression of risk factors

for AIPDa

1.093
64.672
0.030

Table 7 Manifestations, treatment and outcome of

psychotic disorders in the cohort of patients with AIDP
compared with those with non-AIDP

Antiepileptic drug-induced psychosis

recurrent episodes (46.2% versus 20.2%, P 5 0.05). No

patient experienced a chronic course of psychosis in the
AIPD group while 40.5% did in the non-AIPD group.

Discussion

Antiepileptic drug-induced psychotic

disorder is associated with female
gender and temporal lobe
involvement
Our ndings showed that female gender and temporal lobe
involvement were signicant risk factors for AEDs-induced
psychotic disorders. More than two-thirds of the patients
with AIPD were female in this study. In previous studies on
the psychiatric side effects of the new AEDs, a similar trend
was reported (Trimble et al., 2000; Mula et al., 2003;
Weintraub et al., 2007).
Results from both univariate and multivariate analyses
demonstrated that temporal lobe involvement was strongly
associated with AIPD. Previous studies showed that the
patients with temporal lobe epilepsy may be susceptible
to develop psychosis (van der Feltz-Cornelis et al., 2008;
Mula and Monaco, 2009). The susceptibility of temporal
lobe epilepsy to AIPD could be linked to the neuroanatomical anomalies, such as hippocampal sclerosis and
the underlying abnormal connections to temporal and
extratemporal cortices (Lin et al., 2012). The epileptic aetiology within temporal lobe involvement varied and brain
tumour was one of the risk factors for AIPD in the univariate comparison. The lesions of all three cases with tumours were located at both temporal and other lobes,
which indicated complicated pathological mechanisms of
intra/extratemporal connection for the relation of AEDinduced psychotic disorder and temporal involvement.
The whole cohort with psychotic disorders shared some
mutual epilepsy characteristics, such as focal seizures, cognitive impairment and drug resistance. It has been reported
that a higher ratio of patients suffered from focal seizures
with impairment of awareness than other epileptic seizure
types in the population with psychiatric symptoms (Trimble

| 9

et al., 2000; van der Feltz-Cornelis et al., 2008). In our

observation, the percentages of this seizure type in both
groups were higher that 50%, but no signicant difference
was demonstrated between them. Hence, focal seizures with
impairment of awareness might be a predictor of psychiatric comorbidity but not specically that of AED-induced
psychosis.
Similarly, cognitive impairment has been reported to be
related to psychosis in patients with epilepsy (Noguchi
et al., 2012). In our study nearly half the patients had intellectual disability but no statistically signicant difference
was found between the AIPD group and the comparator
group. This suggests that intellectual dysfunction might be
associated with psychosis in epilepsy in general but not
with AED-induced psychosis.
Both groups had high percentage of drug resistance, an
observation noted in previous reports. A population-based
study in male adolescents reported that treatmentrefractory epilepsy increased the risk of psychotic disorders
(Fruchter et al., 2014). We further analysed the AEDs used
before the episodes of psychosis. The development of
psychosis during the treatment with the culprit AED was
associated with a higher number of previous AEDs used.
This nding was perhaps not surprising given that the more
AEDs were trialled to control seizures in patients with
drug-resistant epilepsy, the higher the possibility of developing psychiatric adverse effects. However, this factor
was not signicant as an independent variable contributing
to the development of AIPD in the multivariate logistic regression analysis.
Patients with a history of febrile convulsions (Mula et al.,
2003, 2007) and status epilepticus (Mula et al., 2004)
might be more vulnerable to develop psychiatric adverse
effects, as suggested in post-marketing studies of levetiracetam. In our study, neither the history of febrile convulsions
nor status epilepticus was a predictor of AIPD.

Association with specific

antiepileptic drugs
In the multivariate logistic regression, levetiracetam was
more commonly used among patients with AIDP compared
to those with other types of psychosis. Levetiracetam targets the synaptic vesicle glycoprotein SV2A and presynaptic
calcium channels (Shorvon and van Rijckevorsel, 2002;
Lynch et al., 2004). Previous studies concerning levetiracetam-induced psychotic adverse effects were contradictory
(Mula et al., 2003; Noguchi et al., 2012). Many clinical
trials of levetiracetam reported behaviour adverse effects,
such as irritability, aggressive behaviour in both children
and adults. However, patients in these studies were often
not assessed by psychiatrists (Glauser et al., 2006; de la
Loge et al., 2010). By analysing a cohort of patients who
had developed psychosis, our ndings suggest that when a
patient with epilepsy presents with psychotic symptoms,

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

In this study we report the detailed clinical proles of AIPD

in comparison with other psychotic disorders in patients
with epilepsy. In a systematic review and meta-analysis,
the pooled prevalence rate for psychosis in epilepsy patients
was 5.6% (Clancy et al., 2014). Although prevalence of
AIPD has been reported to vary from 1.0% to 8.4% in
drug trials (Piedad et al., 2012), the percentage of AIPD
in the population of psychosis in epilepsy has not been
reported before. In this study, among epilepsy patients
with psychotic disorders, one in seven could be attributed
to AEDs. This highlighted the importance of considering
the possibility of AIPD in patient with epilepsy who develops psychotic symptoms.

BRAIN 2016: Page 9 of 11

10

| BRAIN 2016: Page 10 of 11

current usage of levetiracetam should raise the strong suspicion of AIPD.

Interestingly, logistic regression analysis demonstrated
that the current use of carbamazepine was negatively associated with AIPD, compared with other types of psychosis,
an observation reported by others (Piedad et al., 2012). No
literature has reported carbamazepine-induced psychosis,
although possible psychotropic effects of carbamazepine,
such as anxiety (Berg et al., 1993) or depression
(Pulliainen and Jokelainen, 1995), have long been recognized. Indeed, carbamazepine has been shown to reduce
aggressive behavioural symptoms (Jones et al., 2011).
Hence, in case of AIDP, carbamazepine might be a safe
substitution for the offending agent.

In the proposal by ILAE, clinical features of psychotic disorders in epilepsy may include auditory hallucination and/
or paranoid delusions (Krishnamoorthy et al., 2007). In
DSM-5, the diagnostic criteria of substance/medicationinduced psychotic disorder consist of delusions and hallucinations. Disorganized speech and grossly disorganized
behaviour are the diagnostic criteria of brief psychotic disorder but not specically of substance/medication-induced
psychotic disorder.
We reviewed all the four categories of symptoms in both
groups and found no differences about the two core symptoms of psychotic disorders, i.e. hallucinations or delusions.
However, the AIPD group showed higher occurrence of
grossly disorganized behaviours and disorganized thinking
compared with the non-AIPD group. This is consistent with
previous reports of high incidence of aggression, agitation
or irritability with certain AEDs, such as 2.724.4% with
topiramate (Elterman et al., 1999; Mula and Trimble,
2003; Weintraub et al., 2007), 2.312.5% with levetiracetam (Mula et al., 2003; de la Loge et al., 2010), and 1.3
6.1% with lamotrigine (Weintraub et al., 2007; Labiner
et al., 2009). In contrast, the prevalence of psychosis was
relatively low, as 1.56.3% with topiramate (Mula and
Trimble, 2003), 1.01.3% with levetiracetam (Mula
et al., 2003; Weintraub et al., 2007) and 0.4% with lamotrigine (Weintraub et al., 2007). Therefore, in case of
psychosis with prominent abnormal behaviours in patients
with epilepsy, AIPD should be taken into consideration.
Disorganized thinking was seldom reported in previous studies. In a European multicentre parallel-group double-blind
trial of zonisamide as add-on treatment, the presence of
disorganized thinking was statistically signicant compared
with placebo (Schmidt et al., 1993). Hence the presence of
disorganized thinking or speech should raise the suspicion
of AIPD.
Follow-up observation showed that AIPD had a generally
better outcome than that of the other epileptic psychoses.

Theoretically, AIPD should improve after the cessation of

the culprit medication; hence, if treated properly, the cases
with AIPD might have shorter duration of the psychotic
episode than other types of psychosis. Furthermore none
of the AIPD cases experienced chronic psychiatric course.
For the cases with recurrent course, the main cause was
resumption of the same or similarly acting culprit drugs.
Therefore, the timely cessation of the offending drug and
avoidance of prescription again would offer protection
against further AIPD.
The limitations of this study included a relatively small
sample size and its retrospective design without randomization or blinding. As such, it is possible that certain AEDs
might have been preferentially chosen or not chosen in
patients with prior psychiatric history. Patients were assessed by different epileptologists and psychiatrists. To minimize the heterogeneity in evaluation for the present study,
the clinical information was jointly reviewed by a single
epileptologist (Z.C.) and a neuropsychiatrist (A.L.) using
a standardized approach, based on the combination of
medical notes and clinic letters, to arrive at the nal
diagnosis.
Future research may seek to identify genetic predictors of
psychosis in epilepsy (Helmstaedter et al., 2013). In this
study, the psychotic symptoms induced were not associated
with high dose or fast titration of the offending agents,
suggesting that there was individual susceptibility in terms
of AIPD. Therefore investigation of genetic markers of
AIPD should be the considered for the future studies.

Conclusion
AIPD was common among epilepsy patients who develop
psychotic symptoms. In our study one in seven patients
with epilepsy who presented with psychosis had AIPD. In
these patients, female gender, temporal lobe involvement
and current use of levetiracetam were signicantly associated with AIPD compared to other types of psychosis,
while carbamazepine had a negative association.
Disorganized behaviours and abnormal disorganized thinking were predominant symptoms of AIPD. AIPD had an
overall better outcome than that of other psychotic disorders in people with epilepsy.

Funding
Z.C. was supported by the Australian and New Zealand
Association of Neurologists (ANZAN) Bayer Asia Pacic
Region Neurology Educational Grant.

References
Adams SJ, OBrien TJ, Lloyd J, Kilpatrick CJ, Salzberg MR,
Velakoulis D. Neuropsychiatric morbidity in focal epilepsy. Br J
Psychiatry 2008; 192: 4649.

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

Antiepileptic drug-induced psychotic

disorder characterized by disorganized behaviour and thought

Z. Chen et al.

Antiepileptic drug-induced psychosis

| 11

Lynch BA, Lambeng N, Nocka K, Kensel-Hammes P, Bajjalieh SM,

Matagne A, et al. The synaptic vesicle protein SV2A is the binding
site for the antiepileptic drug levetiracetam. Proc Natl Acad Sci USA
2004; 101: 98616.
Matsuura M, Adachi N, Oana Y, Okubo Y, Hara T, Onuma T.
Proposal for a new ve-axis classication scheme for psychoses of
epilepsy. Epilepsy Behav 2000; 1: 34352.
Mula M, Monaco F. Antiepileptic drugs and psychopathology of epilepsy: an update. Epileptic Disord 2009; 11: 19.
Mula M, Trimble MR. The importance of being seizure free: topiramate and psychopathology in epilepsy. Epilepsy Behav 2003; 4:
4304.
Mula M, Trimble MR, Sander JW. Psychiatric adverse events in patients with epilepsy and learning disabilities taking levetiracetam.
Seizure 2004; 13: 557.
Mula M, Trimble MR, Sander JW. Are psychiatric adverse events of
antiepileptic drugs a unique entity? A study on topiramate and levetiracetam. Epilepsia 2007; 48: 23226.
Mula M, Trimble MR, Yuen A, Liu RS, Sander JW. Psychiatric adverse events during levetiracetam therapy. Neurology 2003; 61:
7046.
Murray CJ, Vos T, Lozano R, Naghavi M, Flaxman AD, Michaud C,
et al. Disability-adjusted life years (DALYs) for 291 diseases and
injuries in 21 regions, 1990-2010: a systematic analysis for the
Global Burden of Disease Study 2010. Lancet 2012; 380: 2197223.
Noguchi T, Fukatsu N, Kato H, Oshima T, Kanemoto K. Impact of
antiepileptic drugs on genesis of psychosis. Epilepsy Behav 2012; 23:
4625.
Piedad J, Rickards H, Besag FM, Cavanna AE. Benecial and adverse
psychotropic effects of antiepileptic drugs in patients with epilepsy: a
summary of prevalence, underlying mechanisms and data limitations. CNS Drugs 2012; 26: 31935.
Pulliainen V, Jokelainen M. Comparing the cognitive effects of phenytoin and carbamazepine in long-term monotherapy: a two-year
follow-up. Epilepsia 1995; 36: 1195202.
Qin P, Xu H, Laursen TM, Vestergaard M, Mortensen PB. Risk for
schizophrenia and schizophrenia-like psychosis among patients with
epilepsy: population based cohort study. BMJ 2005; 331: 23.
Rai D, Kerr MP, McManus S, Jordanova V, Lewis G, Brugha TS.
Epilepsy and psychiatric comorbidity: a nationally representative
population-based study. Epilepsia 2012; 53: 1095103.
Schmidt D, Jacob R, Loiseau P, Deisenhammer E, Klinger D, Despland
A, et al. Zonisamide for add-on treatment of refractory partial epilepsy: a European double-blind trial. Epilepsy Res 1993; 15: 6773.
Shorvon SD, van Rijckevorsel K. A new antiepileptic drug. J Neurol
Neurosurg Psychiatry 2002; 72: 4269.
Trimble MR, Rusch N, Betts T, Crawford PM. Psychiatric symptoms
after therapy with new antiepileptic drugs: psychopathological and
seizure related variables. Seizure 2000; 9: 24954.
van der Feltz-Cornelis CM, Aldenkamp AP, Ader HJ, Boenink A,
Linszen D, Van Dyck R. Psychosis in epilepsy patients and other
chronic medically ill patients and the role of cerebral pathology in
the onset of psychosis: a clinical epidemiological study. Seizure
2008; 17: 44656.
Weintraub D, Buchsbaum R, Resor SR Jr, Hirsch LJ. Psychiatric and
behavioral side effects of the newer antiepileptic drugs in adults with
epilepsy. Epilepsy Behav 2007; 10: 10510.

Downloaded from http://brain.oxfordjournals.org/ by guest on August 8, 2016

American Psychiatric Association. Diagnostic and statistical manual of

mental disorders. Washington, DC: APA; 2013.
Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross JH, van Emde Boas
W, et al. Revised terminology and concepts for organization of seizures
and epilepsies: report of the ILAE Commission on Classication and
Terminology, 2005-2009. Epilepsia 2010; 51: 67685.
Berg I, Butler A, Ellis M, Foster J. Psychiatric aspects of epilepsy in
childhood treated with carbamazepine, phenytoin or sodium valproate: a random trial. Dev Med Child Neurol 1993; 35: 14957.
Clancy MJ, Clarke MC, Connor DJ, Cannon M, Cotter DR. The
prevalence of psychosis in epilepsy; a systematic review and metaanalysis. BMC Psychiatry 2014; 14: 75.
Clarke MC, Tanskanen A, Huttunen MO, Clancy M, Cotter DR,
Cannon M. Evidence for shared susceptibility to epilepsy and psychosis: a population-based family study. Biol Psychiatry 2012; 71:
8369.
de la Loge C, Hunter SJ, Schiemann J, Yang H. Assessment of behavioral and emotional functioning using standardized instruments in
children and adolescents with partial-onset seizures treated with adjunctive levetiracetam in a randomized, placebo-controlled trial.
Epilepsy Behav 2010; 18: 2918.
Edwards IR, Aronson JK. Adverse drug reactions: denitions, diagnosis, and management. Lancet 2000; 356: 12559.
Elterman RD, Glauser TA, Wyllie E, Reife R, Wu SC, Pledger G. A
double-blind, randomized trial of topiramate as adjunctive therapy
for partial-onset seizures in children. Topiramate YP Study Group.
Neurology 1999; 52: 133844.
Fruchter E, Kapara O, Reichenberg A, Yoffe R, Fono-Yativ O, Kreiss
Y, et al. Longitudinal association between epilepsy and schizophrenia: a population-based study. Epilepsy Behav 2014; 31: 2914.
Glauser TA, Ayala R, Elterman RD, Mitchell WG, Van Orman CB,
Gauer LJ, et al. Double-blind placebo-controlled trial of adjunctive
levetiracetam in pediatric partial seizures. Neurology 2006; 66:
165460.
Helmstaedter C, Mihov Y, Toliat MR, Thiele H, Nuernberg P, Schoch
S, et al. Genetic variation in dopaminergic activity is associated with
the risk for psychiatric side effects of levetiracetam. Epilepsia 2013;
54: 3644.
Jones RM, Arlidge J, Gillham R, Reagu S, van den Bree M, Taylor PJ.
Efcacy of mood stabilisers in the treatment of impulsive or repetitive aggression: systematic review and meta-analysis. Br J Psychiatry
2011; 198: 938.
Krishnamoorthy ES, Trimble MR, Blumer D. The classication of
neuropsychiatric disorders in epilepsy: a proposal by the ILAE
Commission on Psychobiology of Epilepsy. Epilepsy Behav 2007;
10: 34953.
Kwan P, Arzimanoglou A, Berg AT, Brodie MJ, Allen Hauser W,
Mathern G, et al. Denition of drug resistant epilepsy: consensus
proposal by the ad hoc Task Force of the ILAE Commission on
Therapeutic Strategies. Epilepsia 2010; 51: 106977.
Labiner DM, Ettinger AB, Fakhoury TA, Chung SS, Shneker B, Tatum
Iv WO, et al. Effects of lamotrigine compared with levetiracetam on
anger, hostility, and total mood in patients with partial epilepsy.
Epilepsia 2009; 50: 43442.
Lin JJ, Mula M, Hermann BP. Uncovering the neurobehavioural
comorbidities of epilepsy over the lifespan. Lancet 2012; 380:
118092.

BRAIN 2016: Page 11 of 11