Drugs (2014) 74:17471755

DOI 10.1007/s40265-014-0293-6

REVIEW ARTICLE

Antiepileptic Drugs for the Treatment of Agitation

and Aggression in Dementia: Do They Have a Place in Therapy?
Damien Gallagher Nathan Herrmann

Published online: 20 September 2014

Springer International Publishing Switzerland 2014

Abstract Antiepileptic drugs (AEDs) are a class of

medications that have received considerable attention as
possible treatments for agitation and aggression in patients
with dementia. This attention has been driven in equal
measure by promising findings from limited trial and
observational data and the desire to find treatments with
improved tolerability. Their use, to date, has been largely
confined to circumstances where first-line treatments have
proven inadequate or are poorly tolerated. In recent years
there has been some growth in the evidence base, and we
can now make more informed recommendations regarding
a number of older AEDs. Carbamazepine continues to have
the best evidence to support its use, although the evidence
base remains relatively small and concerns regarding tol-
erability limit its use. There is now more consistent evi-
dence that valproate preparations should not be used for
agitation and aggression in dementia. Despite a lack of
high-quality data, some results have been reported for
several newer medications, including levetiracetam, ox-
carbazepine, gabapentin, topiramate and lamotrigine, and a
number of these warrant further investigation. Recent
findings and implications for clinical practice are
discussed.
D. Gallagher (&)
N. Herrmann
Sunnybrook Health Sciences Centre, 2075 Bayview Avenue,
Toronto, ON M4N 3M5, Canada
e-mail: gallagherdamien@hotmail.com
Key Points
Antiepileptic drugs (AEDs) have been the focus of
considerable attention as potential alternatives to
existing treatments for agitation and aggression in
patients with dementia.
Carbamazepine continues to have the best evidence
to support its use, although the evidence base
remains relatively small and concerns regarding
tolerability limit its use.
There is now more consistent evidence that valproate
preparations should not be used for agitation and
aggression in dementia.
There are limited data for several newer AEDs that
warrant further investigation.
1 Introduction
Agitation and aggression occur in approximately 2030 %
of people with Alzheimers disease (AD) living in the
community and approximately 4060 % in care facilities
[1, 2]. Common symptoms of agitation include restless-
ness, pacing or other motor activities that may be associ-
ated with anxiety or abnormal vocalizations. Restlessness
and agitation can escalate to include verbal insults and
shouting, as well as physical aggression such as hitting or
biting others. In fact, neuropsychiatric symptoms (NPS) are
one of the most common reasons for presentation to clin-
ical care, with over 90 % of patients developing at least
one NPS in the course of dementia [3]. Agitation and
aggression are particularly distressing for both the patient
1748
and caregiver, with an increased likelihood of institutional
care and increased costs [1].
Non-pharmacological interventions remain the main-
stay of treatment but unfortunately are inadequate in a
sizeable proportion where agitation or aggression persists
or worsens. Antipsychotic medications have been the best
studied and most commonly used medications for the
treatment of agitation and aggression in dementia, but
concerns regarding tolerability and modest efficacy have
limited their use [1]. Cholinesterase inhibitors were
reported to have efficacy in secondary analyses of early
randomized controlled trials (RCTs) conducted in AD
patients with low levels of NPS at baseline. A subsequent
large, well-designed RCT of donepezil in patients with
significant agitation at baseline found no benefit compared
with placebo [4]. Overall, there have been few studies
designed specifically to address this question, and clinical
guidelines currently state that there is insufficient evi-
dence to recommend for or against the use of cholines-
terase inhibitors for this indication [5]. Memantine has
similarly been reported to have significant benefits com-
pared with placebo [6], but recent studies designed spe-
cifically to address this question have failed to find
significant benefits [7, 8]. There have been relatively few
studies of antidepressants for the treatment of agitation in
dementia, with some promising findings, although defini-
tive evidence to recommend their use for this indication is
lacking [9]. Citalopram remains the best-studied selective
serotonin reuptake inhibitor (SSRI) for this indication,
with a recent large placebo-controlled RCT demonstrating
a clinically meaningful reduction in agitation over
9 weeks [10]. However, generalizability of this study was
limited by the use of a dose (30 mg) in excess of rec-
ommendations by the US Food and Drug Administration
(FDA), as FDA recommendations were changed following
the start of the study (to a daily dose limit of 20 mg for
those 60 years of age and over).
Antiepileptic drugs (AEDs) have therefore been the
focus of considerable attention as potential alternatives to
existing treatments. The majority of studies to date have
focused on the older AEDs such as carbamazepine and
valproate, and these are now the best-studied AEDs for this
indication. In the past, there were few controlled studies on
which to base recommendations regarding efficacy and
tolerability, but recent years have seen some additions to
the evidence base, and these are summarized below. There
are also several newer AEDs that have demonstrated some
promise in mostly observational or uncontrolled studies.
This review focuses specifically on non-benzodiazepine
AEDs as a recent review did not support the routine use of
benzodiazepines for the treatment of NPS [11]. We discuss
the current evidence base and the implications that this
may have for clinical care.
D. Gallagher, N. Herrmann
2 Methods
The PubMed database was searched for publications in
English up to April 2014 using the search terms Alzhei-
mers disease, Alzheimer disease or dementia
combined with the medication names (carbamazepine,
sodium valproate, valproic acid, levetiracetam, ox-
carbazepine, gabapentin, topiramate and lamotrigine).
Emphasis was placed on relevant reviews and RCTs
relating to the use of AEDs for treatment of agitation or
aggression in dementia. Uncontrolled trials and case series
were included if there were insufficient or inadequate
RCTs. Reference sections from review papers were hand
searched for additional relevant publications.
3 Etiology of Agitation and Aggression in Dementia
The etiology of agitation in dementia is ordinarily multi-
factorial, and symptoms are best considered as the result of
an interaction between disease-related neurobiological
factors and environmental factors that can precipitate or
perpetuate symptom expression. For example, agitation in
AD has been associated with greater neurofibrillary tangle
density in the orbitofrontal cortex and anterior cingulate
[12]. Similarly hypometabolism in the right frontal/tem-
poral and bilateral cingulate cortices has been associated
with aggressive, hostile and irritable symptoms in AD [13].
From a genetic perspective, the presence of the long allele
of an insertion/deletion polymorphism in the promoter
region of the serotonin transporter has been associated with
both psychosis and aggression in AD [14]. Environmental
factors such as overcrowding, lack of privacy, noise and
poor communication between caregivers and patients may
increase the likelihood of agitation [1]. These studies
suggest that agitation and other behaviors in dementia
occur in the context of both neuropathological and envi-
ronmental factors, which may interact to modulate symp-
tom expression. Symptoms fluctuate but typically increase
in frequency and severity with disease progression [3, 15].
In addition, NPS are a poor prognostic indicator and have
been associated with more rapid disease progression [16].
4 Measurement of Agitation and Aggression
in Dementia
There is some variability in behavior that might be broadly
described as agitated, and a number of instruments have
been used to assess agitation and aggression in dementia.
One of the most commonly used instruments to measure
agitation in dementia is the Cohen-Mansfield Agitation
Inventory (CMAI). A principal components analysis of the
Antiepileptic Drugs for Agitation and Aggression in Dementia
CMAI revealed four factors (aggressive behavior, physi-
cally non-aggressive behavior, verbally agitated behavior
and a fourth factor comprising hiding and hoarding
behaviors) [17]. Similarly, a principal components analysis
of another commonly used measure, the Neuropsychiatric
Inventory (NPI), revealed four components, two of which
describe agitated behaviors: behavioral dyscontrol
(euphoria, disinhibition, aberrant motor behavior, and
sleep/appetite disturbances) and agitation (agitation/
aggression and irritability/lability) [18]. These components
testify to the complexity of agitated or aggressive behavior,
which may have a different presentation and etiological
origin depending on neuropathology and environmental
influences [19]. This variability has implications for clini-
cal care and the therapeutic approach, as outlined below.
5 Overview of Therapeutic Approach and Treatment
Principles
The therapeutic approach to agitation and aggression in
dementia is necessarily broad and requires consideration of
both environmental and neurobiological factors, as outlined
above. Non-pharmacological interventions are considered
first line, and pharmacological treatments are only con-
sidered where non-pharmacological strategies have proven
inadequate.
In the first instance, agitation or aggression may be
considered an expression of unmet need. The assessment
should identify and address unmet physical health needs
such as infection, pain, delirium, dehydration or sensory
impairments. Recent medication changes that may exac-
erbate symptoms should be considered together with any
environmental triggers. Environmental factors that increase
the likelihood of agitation include overcrowding, lack of
privacy, noise and poor communication between caregivers
and patients. Patients with dementia typically function best
in an environment that is safe, calm, and predictable with
consistent and responsive care. The pattern and frequency
of behaviors together with antecedents and reinforcing
factors may be documented to identify key triggers or
exacerbating factors. This facilitates the formulation of a
care plan tailored to the needs of the patient and allows
ongoing monitoring of symptoms to determine the success
of changes in care [20].
In instances where pharmacological treatment is indi-
cated, the treatment chosen should target the specific
symptoms with which the patient is presenting. For
example, if the predominant problem is depression, then an
antidepressant might be chosen. Similarly, if there are
features of psychosis, then an antipsychotic medication
may be the most appropriate choice. Antipsychotics are
still ordinarily considered a first-line agent for treatment of
1749
significant agitation or aggression associated with extreme
distress or risk, while agents with mixed or inclusive evi-
dence such as memantine, antidepressants or cholinesterase
inhibitors are sometimes used on a trial basis in the context
of milder agitation. AEDs have been typically used to
target symptoms of agitation or aggression where alternate
non-pharmacological and first-line pharmacological
approaches for agitation or aggression have proved inef-
fective or are poorly tolerated. It is usual to start at a low
dose and titrate upwards according to tolerability and
response.
6 The Evidence for Older Antiepileptic Drugs (AEDs)
6.1 Carbamazepine
Early reports of carbamazepine to treat emotionally dis-
turbed patients date from the 1970s, with reports of its use
to treat agitation in dementia dating from the 1980s. Car-
bamazepines efficacy for NPS may be related to its
inhibitory effects on limbic system kindling, increased
locus coeruleus firing and enhanced tryptophan levels [21].
It is known to block voltage-dependent sodium channels
and calcium L-type channels. It reinforces repolarization
by potassium outflow, and has GABAergic, adenosinergic
and glutamate antagonistic properties. The first placebo-
controlled study of carbamazepine in 19 elderly dementia
patients was reported in 1982, with negative findings [22].
Tariot and colleagues then followed this up with two
positive placebo-controlled studies [23, 24]. The first was a
placebo-controlled, crossover study in 25 nursing home
residents meeting criteria for probable or possible AD or
vascular dementia. This preliminary study found a statis-
tically significant improvement in agitation and global
ratings of behavior, suggesting that carbamazepine in low
doses (mean dose 300 mg/day) could reduce agitation in
some patients with dementia [23]. A further randomized,
controlled, multisite study was conducted in 51 nursing
home patients with dementia (probable or possible Alz-
heimers dementia, vascular or mixed dementia) [24].
Carbamazepine was started at 100 mg/day and increased
by 50 mg every 24 days according to tolerability. The
modal carbamazepine dose at 6 weeks was 300 mg/day,
and there was a statistically significant advantage in favor
of carbamazepine, with a Brief Psychiatric Rating Scale
(BPRS) score reduction of 7.7 points for the carbamazepine
group and 0.9 for the placebo group. This resulted in a
clinically meaningful difference for the majority of
patients, with a clinical global improvement in 77 % of the
patients taking carbamazepine versus 21 % of those taking
placebo. There was no significant change in cognition
assessed by the Mini-Mental Status Examination or in
1750
function as assessed by the Physical Self Maintenance
Scale. However, side effects were significantly more
common in patients on carbamazepine than in those given
placebo (59 vs. 29 %), although only two were considered
clinically significant, with one subject developing tics and
another ataxia. Other individuals developed drowsiness,
disorientation or ataxia. In 1996, Cooney et al. [25] pub-
lished findings from a small (n = 6) randomized, placebo-
controlled, crossover study that indicated that carbamaze-
pine significantly reduced aggression among patients with
severe dementia in residential settings. One patient devel-
oped sedation at 300 mg, which resolved following
reduction to 200 mg, and there were no adverse hemato-
logical reactions. Finally, a 6-week, randomized, double-
blind, placebo-controlled, parallel-group trial investigated
carbamazepine in 21 agitated patients with dementia (16
completers) who had been treated unsuccessfully with an-
tipsychotics [26]. There was greater improvement for the
carbamazepine group on the Clinical Global Impression of
Change and the BPRS hostility item. A subsequent meta-
analysis of the two parallel, placebo-controlled RCTs
indicated significant benefit both on the BPRS [mean dif-
ference -5.5 points, 95 % confidence interval (CI) -8.5 to
-2.5] and on the Clinical Global Impression Scale [odds
ratio (OR) 10.2, 95 % CI 3.133.1] in comparison with
placebo [1].
In summary, a number of small, placebo-controlled
studies in generally severely impaired patients with
dementia have found evidence of modest benefit for agi-
tation and aggression (Table 1). However, given the rela-
tively small body of evidence from clinical trials, the risk
of drugdrug interactions, and concerns regarding tolera-
bility [24, 27], carbamazepine is not routinely recom-
mended, although its use may be considered in certain
instances where other agents have proven ineffective or are
poorly tolerated.
6.2 Valproate Preparations
The use of valproate preparations for agitation and
aggression in AD was initially based on case reports, case
series and small, open-label studies. Valproate blocks
voltage-dependent sodium channels, calcium channels and
enhances gaba-mediated neurotransmission [28]. At an
intracellular level, it acts on inositol phosphate metabolism,
activates antioxidant cell survival proteins and has an an-
tikindling effect [29]. One randomized, placebo-controlled
study of divalproex sodium in 56 nursing home patients
with dementia, where valproate was titrated to a mean daily
dose of 826 mg, indicated some small benefit for agitation
when several covariates were taken into account, support-
ing the conduct of further placebo-controlled studies [30].
There was also some reduction in agitation in a study that
D. Gallagher, N. Herrmann
included patients with dementia and manic-type symptoms
secondary to dementia where patients were titrated to a
median dose of 1,000 mg per day [31]. However, sub-
sequent RCTs, with mean daily dosing ranging from
480 mg/day to a possible maximum daily dose of 1,500 mg
per day, have failed to provide any support for its use.
A Cochrane systematic review in 2004 included three
randomized, placebo-controlled trials of valproate therapy
for agitation in the context of dementia and concluded that
it could not be recommended, on the basis of unproven
efficacy and poor tolerability [32]. The authors noted the
relative lack of effect for valproate at lower doses, with
unacceptably high frequency of adverse effects such as
somnolence and thrombocytopenia at higher doses. The
review was updated in 2008 to include two additional
studies conducted in 2005 [33] and 2007 [34]. The new
meta-analysis of pooled results showed no improvement in
agitation among valproate users compared with controls,
and showed an increase in adverse events (falls, infection,
gastrointestinal disorders) among valproate-treated
patients. The authors concluded that the review corrobo-
rated earlier findings that valproate preparations were
ineffective for the treatment of agitation in dementia and
that valproate therapy is associated with unacceptable
adverse effects [35]. More recently, the AD cooperative
study published results of a large RCT of divalproex
sodium for prevention of the emergence of agitation or
psychosis in mild to moderate AD [36]. This 24-month,
randomized, placebo-controlled study of 313 patients
showed no difference in the emergence of NPS and
divalproex sodium was poorly tolerated, with significant
toxicity. It is also of concern that, in a sub-study, patients
treated with divalproex were found to experience acceler-
ated brain volume loss and greater cognitive impairment
compared with placebo-treated patients [37]. Thrombocy-
topenia and hyperammonemia are other potential adverse
effects of valproate preparations and have previously been
described in the literature [38, 39]. Based on these new
studies and previous studies, there is now more consistent
evidence that valproate should not be used for agitation or
aggression in AD (Table 1). Recommendations to avoid
the use of valproate have already been incorporated into
evidence-based clinical practice guidelines [5].
7 The Evidence for Newer AEDs
7.1 Gabapentin
Gabapentin is a newer AED that is structurally similar to
GABA and potentiates GABAergic activity. GABA-related
phenotypes such as anxiety, aggression and sleep distur-
bances may be produced or inhibited by agents that bind to
Antiepileptic Drugs for Agitation and Aggression in Dementia 1751

Table 1 Summary of parallel placebo-controlled trials of AEDs for agitation or aggression in dementia
AED Description Findings Interpretation

Carbamazepine Nursing home residents with
dementia (n = 51) (Tariot et al.
[24])
Outpatients with dementia and
agitation treated unsuccessfully
with antipsychotics (n = 21)
(Olin et al. [26])
Valproate Nursing home residents with
preparations dementia and agitation (n = 56)
(Porsteinsson et al. [30])
Nursing home residents with
dementia (n = 172) (Tariot
et al. [31])
Nursing home residents with AD
and agitation (n = 153) (Tariot
et al. [33])
Trial to prevent emergence of
agitation or psychosis in mild to
moderate AD (n = 313) (Tariot
et al. [36])
Oxcarbazepine Institutionalized patients with
dementia (n = 103) (Sommer
et al. [52])
AD Alzheimers disease, AED antiepileptic drug, NPS neuropsychiatric symptoms
Clinical global improvement in 77 % taking Limited trial data, but remains AED with
carbamazepine vs. 21 % on placebo best evidence for agitation and
(p = 0.001). Brief Psychiatric Rating aggression in dementia. Concerns
Scale score decreased 7.7 points for the regarding long-term safety and
treatment group and 0.9 for placebo tolerability limit use to instances where
Greater improvement in treatment group on first-line approaches have proven
Global Impression of Change (p = 0.055) ineffective
and the Brief Psychiatric Rating Scale
hostility item (p = 0.009)
Reduced agitation on the Clinical Global Poor tolerability and limited efficacy
Impression Scale in 68 % on valproate vs. means that valproate should not be used
52 % on placebo (p = 0.06 in adjusted for agitation or aggression in AD
analysis). Side effects in 68 % of
valproate group vs. 33 % on placebo
(p = 0.03)
Agitation decreased in treatment group, but
more dropouts due to adverse events (22
vs. 4 %, p = 0.001). Somnolence was
common adverse reaction and study was
terminated prematurely
No difference in NPS at a mean dose of
800 mg/day over 6 weeks. Diarrhea and
low platelets were more common in
divalproex treated patients
No difference in emergence of NPS. Higher
rates of somnolence, gait disturbance,
tremor, diarrhea, and weakness in
valproate group
No significant differences in NPS, and Only controlled study, but concerns
adverse events occurred more frequently regarding efficacy and tolerability mean
in the treatment group (75 vs. 47 %). that cannot be recommended for this
indication

the GABA receptor. GABA levels in plasma have been
positively correlated with depression and apathy in patients
with AD [40]. Gabapentin undergoes renal excretion,
which is an important consideration and potentially prob-
lematic in older adults with diminished renal function.
However, because it is not hepatically metabolized and
does not bind to serum proteins, drugdrug interactions are
less likely. A pooled analysis of adverse events from
clinical trials involving adult patients with post-herpetic
neuralgia found that the most common adverse effects were
dizziness, somnolence and peripheral edema [41]. A recent
review of gabapentin for the treatment of behavioral
symptoms in dementia found 11 case reports, three case
series and one retrospective chart review, with no con-
trolled studies published to date [42]. In the majority of
cases, gabapentin was reported to be a well tolerated and
effective for the treatment of agitation and aggression in
dementia. One case series conducted in 12 patients with
moderate to severe dementia concluded that gabapentin
may have a role in treating a subgroup of dementia patients
with severe behavioral disorders who have not responded
to antipsychotics [43]. Another case series has recently
been added to the literature and reported successful and
well-tolerated treatment of aggression in seven patients
with vascular or mixed Alzheimers/vascular dementia
[44]. The absence of controlled trial data, however, limits
support for the use of gabapentin for the treatment of
behavioral symptoms in dementia. Controlled studies are
required before more conclusions can be made regarding
the utility of gabapentin in this context.
7.2 Lamotrigine
Lamotrigine has received very little attention as a possible
treatment for NPS in dementia. One retrospective chart
review and a case report indicate some potential benefit for
agitation and aggression in patients with dementia [45, 46].
It shares many cellular mechanisms of action with car-
bamazepine and valproate. It blocks voltage-dependent
sodium and calcium channels and reduces glutamate
1752
release. Its use in older adults appears to be largely con-
fined to the treatment of bipolar depression or seizure
disorders [47]. There are currently no data of sufficient
quality to support its use for the treatment of NPS in
dementia.
7.3 Levetiracetam
Levetiracetam is an AED that is effective in generalized
and partial epilepsy syndromes. Its mechanism of action is
poorly understood, but it is known to bind synaptic vesicle
protein SV2A in the brain. A recent review uncovered two
studies that assessed the utility of levetiracetam for
behavioral symptoms in patients with dementia [48]. One
open-label study explored the utility of levetiracetam in 19
inpatients with AD and manic-like behavior and demon-
strated an improvement [49]. Another open-label, uncon-
trolled study assessed levetiracetam as a treatment for
agitation in 20 outpatients with AD. In this study, leveti-
racetam was commenced at 500 mg twice daily and titrated
upwards to a total daily dose of 1,500 or 2,000 mg
according to symptoms and tolerability. The authors
reported frequent side effects and possibly impaired cog-
nition without significant therapeutic efficacy. Six subjects
terminated the study early, three because of increased
agitation or aggression [50]. Intriguingly, levetiracetam at
lower doses, 125 mg twice daily, has been reported to have
possible cognitive benefits in patients with amnestic mild
cognitive impairment, and these effects remain the subject
of ongoing investigation [51]. However, there is inadequate
evidence to support the use of levetiracetam for NPS in
dementia, with concerns regarding tolerability.
7.4 Oxcarbazepine
Oxcarbazepine is the keto-derivative of carbamazepine and
is thought to act via sodium channel inhibition in the same
way as carbamazepine. Its efficacy as a treatment for agi-
tation and aggression in patients with Alzheimers
dementia, vascular dementia or both has been evaluated in
one randomized, placebo-controlled trial [52]. In this study,
103 institutionalized patients at 35 sites were randomized
to oxcarbazepine or placebo. The agitation and aggression
subscores of the NPI were the primary outcomes, and
secondary outcomes included caregiver burden and the
Brief Agitation Rating Scale (BARS). No statistically
significant differences were found between the two groups,
although a trend was observed in favor of the oxcarbaze-
pine group on the BARS score. Adverse events occurred
significantly more frequently in the treatment group (75 vs.
47 %). Of the 52 patients receiving oxcarbazepine, there
were reports of sedation (n = 13), fainting (n = 8) and
ataxia (n = 5). In other studies, oxcarbazepine has been
D. Gallagher, N. Herrmann
associated with more frequent hyponatremia than carbam-
azepine, especially in older patients [27]. This is the only
controlled study of oxcarbazepine in this context, and so
definitive conclusions cannot be drawn, although early
indications are that oxcarbazepine cannot be recom-
mended, because of concerns regarding efficacy and
tolerability.
7.5 Topiramate
The AED topiramate has sometimes been found to be
useful in the treatment of mania in patients with bipolar
disorder. It modulates sodium conductance, inhibits L-type
calcium channels, potentiates GABAergic inhibition,
decreases AMPA/kainate receptor-mediated currents and is
a weak inhibitor of carbonic anhydrase [53]. It has a good
tolerability profile, but the risk of cognitive impairment has
to be taken into account when used in older patients with
dementia [54]. One retrospective chart review of 15
patients with dementia and aggression who were treated
with topiramate, alone or in combination with antipsy-
chotic medication, concluded that it appeared to be effec-
tive and that controlled studies were required [55]. In 2010,
results from a randomized, controlled study were pub-
lished. In this study, 48 patients were randomized to
receive either topiramate or risperidone. Both groups
showed significant improvement on all outcomes, with no
significant between-group differences on the NPI, CMAI or
Clinical Global Impression Scale. There were no signifi-
cant changes in the cognitive status of patients (assessed by
the Mini-Mental Status Examination) taking topiramate or
risperidone. In the topiramate group, 21 of 25 patients
completed the study, with two withdrawals secondary to
fatigue and another two secondary to loss of appetite. The
authors concluded that treatment with low-dose topiramate
(2550 mg per day) demonstrated comparable efficacy to
risperidone [56]. The small sample size and absence of a
comparison placebo group are limitations, and so it is not
known whether topiramate is superior to placebo. There are
also concerns regarding weight loss, which has been
reported in other populations and which may be of par-
ticular concern in older patients with dementia [57].
However, further study of topiramate is required before
more definitive conclusions can be drawn regarding its
potential as a therapeutic agent.
8 Discussion
Overall, there are still very few RCTs of AEDs for NPS in
dementia. The two best-studied agents remain carbamaze-
pine and valproate, with a number of controlled studies for
these agents and largely uncontrolled data available for
Antiepileptic Drugs for Agitation and Aggression in Dementia
newer agents, with a few exceptions. The use of the latter
has not been supported by trial data, and recent studies
have provided more consistent evidence that this medica-
tion is relatively ineffective at lower doses and so poorly
tolerated at higher doses that it cannot be recommended for
treatment of behavioral symptoms in dementia. These
findings are now reflected in clinical guidelines, which
recommend against the use of valproate for this indication
[5]. Limited trial data provide some support for the use of
carbamazepine. In practice, its use is generally reserved for
circumstances where alternate agents have proved inef-
fective or are poorly tolerated. Concerns remain regarding
its potential for drugdrug interactions, as it is an enzyme
inducer and it also has possible adverse effects such as
drowsiness, disorientation or ataxia. There is a dearth of
data regarding its safety for long-term use.
There are several newer AEDs that may have thera-
peutic potential but as yet lack definitive evidence to make
clear recommendations regarding their use for this indica-
tion. There has been one negative controlled study for
oxcarbazepine, although findings need to be replicated
before definitive conclusions can be drawn. Gabapentin
and lamotrigine have been the subject of largely observa-
tional or uncontrolled studies and are worthy of further
investigation. Topiramate demonstrated comparable effi-
cacy to risperidone in one study, but its superiority to
placebo has not been demonstrated. Further study of the
above medications would require careful patient selection
and monitoring for potential adverse effects according to
the tolerability profile of each agent and individual patient
vulnerability.
Clinical trial data do not clearly indicate which patients
might benefit most from AEDs. In general, AEDs are only
considered when non-pharmacological approaches and
first-line pharmacological approaches have proven inef-
fective, and in all instances, pharmacological approaches
should form part of a complete package of care. The best
evidence for treatment of agitation and aggression in
dementia exists for the antipsychotic medications risperi-
done, olanzapine and aripiprazole [5]. There is limited
evidence to support the use of antidepressant medication,
although a recent study does provide some additional
support for citalopram, albeit the dose used in this study
was beyond the currently licensed maximum dose [9, 10].
There has been mixed and inconclusive evidence for cho-
linesterase inhibitors and memantine [34, 58]. Antipsy-
chotics are still ordinarily considered first-line agents
where agitation or aggression is associated with extreme
distress or risk, while medications with less evidence are
sometimes used on a trial basis in the context of milder
agitation. In general, a trial of an AED is only considered
where alternate approaches have proven ineffective or are
poorly tolerated. Some of the most commonly reported
1753
adverse effects with AEDs are sedation, dizziness or ataxia.
However, each agent has its own particular concerns, and
when clinicians decide that a trial of an AED is indicated,
they should monitor closely for adverse effects according
to the safety profile of each agent. Adverse effects may also
vary according to patient considerations such as individual
vulnerability, medical comorbidities and concomitant
medications. For example, carbamazepine has been asso-
ciated with drowsiness, disorientation, ataxia bone marrow
toxicity and hyponatremia. It is also known to induce its
own metabolism, which means that its clinical effects may
diminish over time, and it also induces the metabolism of
other medications [24, 54]. Commonly reported adverse
effects for valproate include somnolence, diarrhea, gait
disturbance and thrombocytopenia. Laboratory monitoring
is usual for monitoring of possible adverse effects such as
reduced cell blood counts, elevation of liver enzymes or
hyponatremia with carbamazepine. However, levels of
AEDs have not been clearly correlated with efficacy in this
context and so are not indicated unless there are concerns
about toxicity or adherence. It is essential to monitor target
symptoms and tolerability of any AED with a view to
withdrawal if adverse effects outweigh benefits.
9 Conclusion
Overall, carbamazepine and valproate remain the AEDs
that have been most studied in placebo-controlled trials.
There is some evidence to support the use of carbamaze-
pine, although the evidence base remains relatively small
and concerns regarding tolerability limit its use. There is
now more consistent evidence that valproate preparations
should not be used for agitation and aggression in
dementia. There is very limited trial data to make clear
recommendations for or against many of the newer AEDs,
and further randomized, controlled studies of medications
that demonstrate promise are required. Future studies
should be based upon sound neurobiological models of
disease and incorporate biomarkers that might allow more
individualized and targeted pharmacological treatment.
Acknowledgments No industry funding was received to assist with
the preparation of this review. Dr. Gallagher has no conflicts of
interest to declare. Dr. Herrmann has received research Grants from
Lundbeck, Sanofi-Aventis and Roche, consultant fees from Abbvie
and speakers honoraria from Pfizer.
References
1. Ballard C, Corbett A, Chitramohan R, Aarsland D. Management
of agitation and aggression associated with Alzheimers disease:
controversies and possible solutions. Curr Opin Psychiatry.
2009;22(6):53240. doi:10.1097/YCO.0b013e32833111f9.
1754
2. Lyketsos CG, Lopez O, Jones B, Fitzpatrick AL, Breitner J,
DeKosky S. Prevalence of neuropsychiatric symptoms in
dementia and mild cognitive impairment: results from the car-
diovascular health study. JAMA. 2002;288(12):147583.
3. Steinberg M, Shao H, Zandi P, Lyketsos CG, Welsh-Bohmer KA,
Norton MC, et al. Point and 5-year period prevalence of neuro-
psychiatric symptoms in dementia: the Cache County Study. Int J
Geriatr Psychiatry. 2008;23(2):1707. doi:10.1002/gps.1858.
4. Howard RJ, Juszczak E, Ballard CG, Bentham P, Brown RG,
Bullock R, et al. Donepezil for the treatment of agitation in
Alzheimers disease. N Engl J Med. 2007;357(14):138292.
doi:10.1056/NEJMoa066583.
5. Herrmann N, Lanctot KL, Hogan DB. Pharmacological recom-
mendations for the symptomatic treatment of dementia: the
Canadian consensus conference on the diagnosis and treatment of
dementia 2012. Alzheimers Res Ther. 2013;5(Suppl 1):S5.
doi:10.1186/alzrt201.
6. Gauthier S, Loft H, Cummings J. Improvement in behavioural
symptoms in patients with moderate to severe Alzheimers dis-
ease by memantine: a pooled data analysis. Int J Geriatr Psy-
chiatry. 2008;23(5):53745. doi:10.1002/gps.1949.
7. Herrmann N, Gauthier S, Boneva N, Lemming OM. Investiga-
tors. A randomized, double-blind, placebo-controlled trial of
memantine in a behaviorally enriched sample of patients with
moderate-to-severe Alzheimers disease. Int Psychogeriatr.
2013;25(6):91927. doi:10.1017/S1041610213000239.
8. Fox C, Crugel M, Maidment I, Auestad BH, Coulton S, Treloar
A, et al. Efficacy of memantine for agitation in Alzheimers
dementia: a randomised double-blind placebo controlled trial.
PLoS One. 2012;7(5):e35185. doi:10.1371/journal.pone.003
5185.
9. Seitz DP, Adunuri N, Gill SS, Gruneir A, Herrmann N, Rochon P.
Antidepressants for agitation and psychosis in dementia. Coch-
rane Database Syst Rev. 2011;(2):CD008191. doi:10.1002/
14651858.CD008191.pub2.
10. Porsteinsson AP, Drye LT, Pollock BG, Devanand DP, Frangakis
C, Ismail Z, et al. Effect of citalopram on agitation in Alzheimer
disease: the CitAD randomized clinical trial. JAMA.
2014;311(7):68291. doi:10.1001/jama.2014.93.
11. Tampi RR, Tampi DJ. Efficacy and tolerability of benzodiaze-
pines for the treatment of behavioral and psychological symptoms
of dementia: a systematic review of randomized controlled trials.
Am J Alzheimers Dis Other Demen. 2014. doi:10.1177/
1533317514524813.
12. Tekin S, Mega MS, Masterman DM, Chow T, Garakian J, Vinters
HV, et al. Orbitofrontal and anterior cingulate cortex neurofi-
brillary tangle burden is associated with agitation in Alzheimer
disease. Ann Neurol. 2001;49(3):35561.
13. Sultzer D, Melrose R, Leskin L, Narvaez T, Ando T, Walston A,
et al. FDG-PET cortical metabolic activity associated with dis-
tinct agitation behaviors in Alzheimers disease. Poster pre- 31. Tariot PN, Schneider LS, Mintzer JE, Cutler AJ, Cunningham
sented at the Alzheimers Association International Conference
on Alzheimers Disease (ICAD); Paris, France; 2011.
14. Sweet RA, Pollock BG, Sukonick DL, Mulsant BH, Rosen J,
Klunk WE, et al. The 5-HTTPR polymorphism confers liability to
a combined phenotype of psychotic and aggressive behavior in
Alzheimer disease. Int Psychogeriatr. 2001;13(4):4019.
15. Zahodne LB, Ornstein K, Cosentino S, Devanand DP, Stern Y.
Longitudinal relationships between Alzheimer disease progres-
sion and psychosis, depressed mood, and agitation/aggression.
Am J Geriatr Psychiatry. 2013. doi:10.1016/j.jagp.2013.03.014.
16. Rabins PV, Schwartz S, Black BS, Corcoran C, Fauth E, Mielke
M, et al. Predictors of progression to severe Alzheimers disease
in an incidence sample. Alzheimers Dement. 2013;9(2):2047.
doi:10.1016/j.jalz.2012.01.003.
D. Gallagher, N. Herrmann
17. Rabinowitz J, Davidson M, De Deyn PP, Katz I, Brodaty H,
Cohen-Mansfield J. Factor analysis of the Cohen-Mansfield
Agitation Inventory in three large samples of nursing home
patients with dementia and behavioral disturbance. Am J Geriatr
Psychiatry. 2005;13(11):9918. doi:10.1176/appi.ajgp.13.11.991.
18. Hollingworth P, Hamshere ML, Moskvina V, Dowzell K, Moore
PJ, Foy C, et al. Four components describe behavioral symptoms
in 1,120 individuals with late-onset Alzheimers disease. J Am
Geriatr Soc. 2006;54(9):134854. doi:10.1111/j.1532-5415.2006.
00854.x.
19. Geda YE, Schneider LS, Gitlin LN, Miller DS, Smith GS, Bell J,
et al. Neuropsychiatric symptoms in Alzheimers disease: past
progress and anticipation of the future. Alzheimers Dement.
2013;9(5):6028. doi:10.1016/j.jalz.2012.12.001.
20. Cohen-Mansfield J. Assessment of agitation. Int Psychogeriatr.
1996;8(2):23345.
21. Post RM, Susan R, Weiss B. Sensitization, kindling, and car-
bamazepine: an update on their implications for the course of
affective illness. Pharmacopsychiatry. 1992;25(1):413. doi:10.
1055/s-2007-1014386.
22. Chambers CA, Bain J, Rossbottom R, Ballinger BR, McLaren S.
Carbamazepine in senile dementia and overactivity: a placebo
controlled double blind trial IRCS. Med Sci. 1982;10:5056.
23. Tariot PN, Erb R, Leibovici A, Podgorski CA, Cox C, Asnis J,
et al. Carbamazepine treatment of agitation in nursing home
patients with dementia: a preliminary study. J Am Geriatr Soc.
1994;42(11):11606.
24. Tariot PN, Erb R, Podgorski CA, Cox C, Patel S, Jakimovich L,
et al. Efficacy and tolerability of carbamazepine for agitation and
aggression in dementia. Am J Psychiatry. 1998;155(1):5461.
25. Cooney C, Mortimer A, Smith A, Newton K, Wrigley M. Car-
bamazepine use in aggressive behaviour associated with senile
dementia. Int J Geriatr Psychiatry. 1996;11:9015.
26. Olin JT, Fox LS, Pawluczyk S, Taggart NA, Schneider LS. A
pilot randomized trial of carbamazepine for behavioral symptoms
in treatment-resistant outpatients with Alzheimer disease. Am J
Geriatr Psychiatry. 2001;9(4):4005.
27. Dong X, Leppik IE, White J, Rarick J. Hyponatremia from ox-
carbazepine and carbamazepine. Neurology. 2005;65(12):
19768. doi:10.1212/01.wnl.0000188819.45330.90.
28. Rosenberg G. The mechanisms of action of valproate in neuro-
psychiatric disorders: can we see the forest for the trees? Cellular
and molecular life sciences: CMLS. 2007;64(16):2090103.
doi:10.1007/s00018-007-7079-x.
29. Weiss SR, Post RM. Kindling: separate vs. shared mechanisms in
affective disorders and epilepsy. Neuropsychobiology. 1998;38(3):
16780.
30. Porsteinsson AP, Tariot PN, Erb R, Cox C, Smith E, Jakimovich
L, et al. Placebo-controlled study of divalproex sodium for agi-
tation in dementia. Am J Geriatr Psychiatry. 2001;9(1):5866.
MR, Thomas MS, et al. Safety and tolerability of divalproex
sodium in the treatment of signs and symptoms of mania in
elderly patients with dementia: results of a double-blind, placebo-
controlled trial. Curr Ther Res. 2001;62(1):5167.
32. Lonergan ET, Cameron M, Luxenberg J. Valproic acid for agi-
tation in dementia. Cochrane Database Syst Rev. 2004;(2):
CD003945. doi:10.1002/14651858.CD003945.pub2.
33. Tariot PN, Raman R, Jakimovich L, Schneider L, Porsteinsson A,
Thomas R, et al. Divalproex sodium in nursing home residents
with possible or probable Alzheimer disease complicated by
agitation: a randomized, controlled trial. Am J Geriatr Psychiatry.
2005;13(11):9429. doi:10.1176/appi.ajgp.13.11.942.
34. Herrmann N, Lanctot KL, Rothenburg LS, Eryavec G. A placebo-
controlled trial of valproate for agitation and aggression in
Antiepileptic Drugs for Agitation and Aggression in Dementia
Alzheimers disease. Dement Geriatr Cogn Disord. 2007;23(2):
1169. doi:10.1159/000097757.
35. Lonergan E, Luxenberg J. Valproate preparations for agitation in
dementia. Cochrane Database Syst Rev. 2009;(3):CD003945.
doi:10.1002/14651858.CD003945.pub3.
36. Tariot PN, Schneider LS, Cummings J, Thomas RG, Raman R,
Jakimovich LJ, et al. Chronic divalproex sodium to attenuate
agitation and clinical progression of Alzheimer disease. Arch Gen
Psychiatry. 2011;68(8):85361. doi:10.1001/archgenpsychiatry.
2011.72.
37. Fleisher AS, Truran D, Mai JT, Langbaum JB, Aisen PS, Cum-
mings JL, et al. Chronic divalproex sodium use and brain atrophy
in Alzheimer disease. Neurology. 2011;77(13):126371. doi:10.
1212/WNL.0b013e318230a16c.
38. Trannel TJ, Ahmed I, Goebert D. Occurrence of thrombocyto-
penia in psychiatric patients taking valproate. Am J Psychiatry.
2001;158(1):12830.
39. McCall M, Bourgeois JA. Valproic acid-induced hyperammone-
mia: a case report. J Clin Psychopharmacol. 2004;24(5):5216.
40. Lanctot KL, Herrmann N, Rothenburg L, Eryavec G. Behavioral
correlates of GABAergic disruption in Alzheimers disease. Int
Psychogeriatr. 2007;19(1):1518. doi:10.1017/S104161020600
3899.
41. Parsons B, Tive L, Huang S. Gabapentin: a pooled analysis of
adverse events from three clinical trials in patients with pos-
therpetic neuralgia. Am J Geriatr Pharmacother. 2004;2(3):
15762. doi:10.1016/j.amjopharm.2004.09.004.
42. Kim Y, Wilkins KM, Tampi RR. Use of gabapentin in the
treatment of behavioural and psychological symptoms of
dementia: a review of the evidence. Drugs Aging. 2008;25(3):
18796.
43. Herrmann N, Lanctot K, Myszak M. Effectiveness of gabapentin
for the treatment of behavioral disorders in dementia. J Clin
Psychopharmacol. 2000;20(1):903.
44. Cooney C, Murphy S, Tessema H, Freyne A. Use of low-dose
gabapentin for aggressive behavior in vascular and mixed vas-
cular/Alzheimer dementia. J Neuropsychiatry Clin Neurosci.
2013;25(2):1205. doi:10.1176/appi.neuropsych.12050115.
45. Devarajan S, Dursun SM. Aggression in dementia with lamotri-
gine treatment. Am J Psychiatry. 2000;157(7):1178.
46. Ng B, Camacho A, Bardwell W, Sewell DD. Lamotrigine for
agitation in older patients with dementia. Int Psychogeriatr.
2009;21(1):2078. doi:10.1017/S1041610208007898.
47. Sajatovic M, Ramsay E, Nanry K, Thompson T. Lamotrigine
therapy in elderly patients with epilepsy, bipolar disorder or
1755
dementia. Int J Geriatr Psychiatry. 2007;22(10):94550. doi:10.
1002/gps.1784.
48. Dolder CR, Nealy KL. The efficacy and safety of newer anti-
convulsants in patients with dementia. Drugs Aging.
2012;29(8):62737. doi:10.2165/11632480-000000000-00000.
49. Kyomen HH, Whitfield TH, Baldessarini RJ. Levetiracetam for
manic behavior in hospitalized geriatric patients with dementia of
the Alzheimers type. J Clin Psychopharmacol. 2007;27(4):
40810. doi:10.1097/01.jcp.0000264996.08901.3c.
50. Weiner MF, Womack KB, Martin-Cook K, Svetlik DA, Hynan
LS. Levetiracetam for agitated Alzheimers disease patients. Int
Psychogeriatr. 2005;17(2):3278.
51. Bakker A, Krauss GL, Albert MS, Speck CL, Jones LR, Stark
CE, et al. Reduction of hippocampal hyperactivity improves
cognition in amnestic mild cognitive impairment. Neuron.
2012;74(3):46774. doi:10.1016/j.neuron.2012.03.023.
52. Sommer OH, Aga O, Cvancarova M, Olsen IC, Selbaek G,
Engedal K. Effect of oxcarbazepine in the treatment of agitation
and aggression in severe dementia. Dement Geriatr Cogn Disord.
2009;27(2):15563. doi:10.1159/000199236.
53. Meldrum BS. Update on the mechanism of action of antiepileptic
drugs. Epilepsia. 1996;37(Suppl 6):S411.
54. Sommer BR, Fenn HH, Ketter TA. Safety and efficacy of anti-
convulsants in elderly patients with psychiatric disorders: ox-
carbazepine, topiramate and gabapentin. Expert opinion on drug
safety. 2007;6(2):13345. doi:10.1517/14740338.6.2.133.
55. Fhager B, Meiri IM, Sjogren M, Edman A. Treatment of
aggressive behavior in dementia with the anticonvulsant topira-
mate: a retrospective pilot study. Int Psychogeriatr. 2003;15(3):
3079.
56. Mowla A, Pani A. Comparison of topiramate and risperidone for
the treatment of behavioral disturbances of patients with Alz-
heimer disease: a double-blind, randomized clinical trial. J Clin
Psychopharmacol. 2010;30(1):403. doi:10.1097/JCP.0b013e
3181ca0c59.
57. Gupta S, Masand PS, Frank BL, Lockwood KL, Keller PL. To-
piramate in bipolar and schizoaffective disorders: weight loss and
efficacy. Prim Care Companion J Clin Psychiatry. 2000;2(3):
96100.
58. Rodda J, Morgan S, Walker Z. Are cholinesterase inhibitors
effective in the management of the behavioral and psychological
symptoms of dementia in Alzheimers disease? A systematic
review of randomized, placebo-controlled trials of donepezil,
rivastigmine and galantamine. Int Psychogeriatr. 2009;21(5):
81324. doi:10.1017/S1041610209990354.