Accepted Manuscript

Efficacy, safety and tolerability of aripiprazole in bipolar disorder:

An updated systematic review and meta-analysis of randomized
controlled trials

Dian-Jeng Li, Ping-Tao Tseng, Brendon Stubbs, Che-Sheng Chu,

Han-Yung Chang, Eduard Vieta, Michele Fornaro, Andre F.
Carvalho, Marco Solmi, Nicola Veronese, Tien-Yu Chen, Yen-
Wen Chen, Pao-Yen Lin, Philip Chik-keung Chow

PII: S0278-5846(17)30238-5
DOI: doi: 10.1016/j.pnpbp.2017.06.023
Reference: PNP 9145
To appear in: Progress in Neuropsychopharmacology & Biological Psychiatry
Received date: 26 March 2017
Revised date: 20 June 2017
Accepted date: 21 June 2017

Please cite this article as: Dian-Jeng Li, Ping-Tao Tseng, Brendon Stubbs, Che-Sheng
Chu, Han-Yung Chang, Eduard Vieta, Michele Fornaro, Andre F. Carvalho, Marco Solmi,
Nicola Veronese, Tien-Yu Chen, Yen-Wen Chen, Pao-Yen Lin, Philip Chik-keung Chow ,
Efficacy, safety and tolerability of aripiprazole in bipolar disorder: An updated systematic
review and meta-analysis of randomized controlled trials. The address for the
corresponding author was captured as affiliation for all authors. Please check if
appropriate. Pnp(2017), doi: 10.1016/j.pnpbp.2017.06.023

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

Efficacy, safety and tolerability of aripiprazole in bipolar disorder: An

updated systematic review and meta-analysis of randomized controlled
trials

Dian-Jeng Li a,b,1 , Ping-Tao Tseng c,d,1, Brendon Stubbs e,f

, Che-Sheng Chug, Han-Yung

PT
Chang h, Eduard Vietai, Michele Fornaroj,k, Andre F. Carvalhol, Marco Solmi m, Nicola

RI
Veronesen, Tien-Yu Cheno, Yen-Wen Chen p, Pao-Yen Lin q,r,*, Philip Chik-keung Chow

SC
s,*

NU
a
: Department of Addiction Science, Kaohsiung Municipal Kai-Syuan Psychiatric
MA

Hospital, Kaohsiung City, Taiwan

ED

b
: Graduate Institute of Medicine, College of Medicine, Kaohsiung Medical

University,Taiwan
T
EP

c
: Department of Psychiatry, Tsyr-Huey Mental Hospital, Kaohsiung Jen-Ai’s Home,

Taiwan
C
AC

d
WinShine Clinics in Specialty of Psychiatry, Kaohsiung City, Taiwan

e
: Physiotherapy Department, South London and Maudsley NHS Foundation Trust,

London, UK

f
: Health Service and Population Research Department, Institute of Psychiatry,

Psychology and Neuroscience (IoPPN), King's College London, De Crespigny Park,

London, UK
1
 ACCEPTED MANUSCRIPT

g
: Department of Psychiatry, Kaohsiung Veterans General Hospital, Kaohsiung City,

Taiwan

h
: Kaohsiung Municipal Kai-Syuan Psychiatric Hospital, Kaohsiung, Taiwan

i
: Bipolar Disorder Unit, Institute of Neuroscience, Hospital Clinic, University of

Barcelona, IDIBAPS, CIBERSAM, Barcelona, Catalonia, Spain.

PT
j
: New York State Psychiatric Institute (NYPSI); Columbia University, NYC, NY, USA.

RI
SC
k
: Outpatient Unit on Treatment Resistant Psychosis, Department of Neuroscience,

University School of Medicine Federico II, Naples, Italy.

NU
l
Department of Clinical Medicine and Translational Psychiatry Research Group,
MA

Faculty of Medicine, Federal University of Ceará, Fortaleza, CE, Brazil

m
: Department of Neuroscience, University of Padova, Padova, Italy; Mental Health
ED

Institute for clinical Research and Education on Medicine, I.R.E.M., Padua, Italy
T

n
: Geriatrics Division, Department of Medicine-DIMED, University of Padova, Padova,
EP

Italy.
C

o
: Department of Psychiatry, Tri-Service General Hospital; School of Medicine,
AC

National Defense Medical Center Taipei, Taiwan

p
: Prospect clinic for otorhinolaryngology & neurology

q
: Department of Psychiatry, Kaohsiung Chang Gung Memorial Hospital and Chang

Gung University College of Medicine, Kaohsiung, Taiwan

r
: Center for Translational Research in Biomedical Sciences, Kaohsiung Chang Gung
2
 ACCEPTED MANUSCRIPT

Memorial Hospital, Kaohsiung, Taiwan

s
: Department of Child and Adolescent Psychiatry, Kaohsiung Municipal Kai-Syuan

Psychiatric Hospital, Kaohsiung City, Taiwan

PT
1
: contributed equally as first authors

RI
*: contributed equally as corresponding authors

SC
NU
Correspondence to:
MA

Pao-Yen Lin, MD, PhD

Department of Psychiatry, Kaohsiung Chang Gung Memorial Hospital

ED

123, Dapi Road, Niaosong District, Kaohsiung City 833, Taiwan

T

Telephone number: 886-7-7317123 ext. 8751

EP

Fax number: 886-7-7326817

C

E-mail address: [email protected]

AC

Philip Chik-keung Chow, MD, PhD

Department of Child and Adolescent Psychiatry, Kaohsiung Municipal Kai-Syuan

Psychiatric Hospital, Kaohsiung City, Taiwan

130, Kaixuan 2nd Road, Lingya District., Kaohsiung City 802, Taiwan

3
 ACCEPTED MANUSCRIPT

Telephone number: 886-7-7513171

Email address: [email protected]

PT
RI
SC
NU
MA
T ED
C EP
AC

4
 ACCEPTED MANUSCRIPT

Abstract

Numerous studies have investigated aripiprazole as a treatment for bipolar disorder

(BD). therefore we conducted this comprehensive meta-analysis to investigate the

efficacy and safety profile of aripiprazole in treating BD. Two authors conducted

systematic searches of PubMed and ScienceDirect from inception until May 14 th,

PT
2017. Randomized controlled trials (RCTs) of people with BD who received

aripiprazole were included. A total of 20 RCTs met the eligibility criteria, including

RI
two which investigated the efficacy of aripiprazole versus haloperidol

SC
(aripiprazole=340; haloperidol=337), three which compared aripiprazole versus

lithium (aripiprazole=208; lithium=212), and 15 with multiple comparisons of

NU
aripiprazole versus a placebo (aripiprazole=1923; placebo=1499). Compared to a
MA

placebo, aripiprazole improved acute mania (Hedges’ g: -0.299, p=0.001) and

psychosis (Hedges’ g: -0.296, p<0.001) in the acute mania state, but did not improve
ED

depressive symptoms (Hedges’ g: -0.127, p=0.054) in the acute depressive state.

Aripiprazole was associated with lower relapse rates in bipolar mania when used in
T

combination versus a placebo in maintenance therapy (odds ratio: 0.522, p<0.029).

EP

Aripiprazole was also associated with higher levels of high density lipoprotein, lower
C

dropout rates, but no difference in extrapyramidal symptoms in the maintenance

AC

phase versus a placebo or in comparison with other medications (haloperidol or

lithium). Our results suggest that aripiprazole is effective and safe in treating bipolar

mania. Further trials are necessary to evaluate the efficacy and tolerability versus

other medications.

Keywords: Aripiprazole; bipolar disorder; efficacy; tolerability; meta-analysis

5
 ACCEPTED MANUSCRIPT

Introduction

Bipolar disorder (BD) affects approximately 2.4% of the general population

(Merikangas et al., 2011) and is easily misdiagnosed or under-diagnosed (Angst,

2006). BD is associated with a greatly impaired quality of life, increased physical

health burden (Vancampfort et al., 2016; Vancampfort et al., 2015) and disability

PT
(Hirschfeld et al., 2003; Yatham et al., 2004). Moreover, BD can impact employment

(Reed et al., 2010) and lead to lower self-esteem (Nilsson et al., 2010) and reduced

RI
social interaction (Judd and Akiskal, 2003). Furthermore, approximately 23% of

SC
patients with BD report a history of suicide attempts (Kattimani et al., 2016)
NU
BD has a complicated, multifactorial etiology including genetic, neuro-endocrine,

and environmental factors (Craddock and Sklar, 2013). Pharmacological therapy is

MA

often the first-line treatment for BD, followed by psychological (Oud et al., 2016) and

psychosocial interventions (Goodwin et al., 2008). In addition to lithium and

ED

anticonvulsants (van der Loos et al., 2011; Yildiz et al., 2015), antipsychotic agents

have also shown promising efficacy for bipolar mania (Yildiz et al., 2011) and
T
EP

depression (Goodwin et al., 2016).

C

Aripiprazole, a novel antipsychotic agent combining partial agonist activity at

AC

dopamine-2 (D2), dopamine-3 (D3), and serotonin-1A (5-HT1A) receptors with

antagonist activity at serotonin-2A (5-HT2A) and D2 receptors (Burris et al., 2002;

Jordan et al., 2002). It has shown beneficial effects in patients with schizophrenia and

possibly in mood disorders including BD, including good tolerability in patients with

schizophrenia (Oya et al., 2015), good efficacy in patients with acute mania

(Fountoulakis et al., 2011; Young et al., 2009), and good efficacy in preventing

relapses (Marcus et al., 2011), although there is limited evidence regarding bipolar

6
 ACCEPTED MANUSCRIPT

depression (Vieta et al., 2010). Aripiprazole is known to have a beneficial metabolic

profile, which is an important consideration given that these patients are at an

increased risk of cardiometabolic disorders (Vancampfort et al., 2016; Vancampfort

et al., 2015), although it may be associated with akathisia.

Two previous meta-analyses (Brown et al., 2013; Meduri et al., 2016)

PT
investigated the efficacy or safety profile of aripiprazole. Brown et al (Brown et al.,

2013) discussed the efficacy and adverse reactions in BD; however, the authors only

RI
addressed the effects of treatment with regards to episodes of mania without

SC
considering the depression or maintenance phase, and their search only included

trials before 2013. Furthermore, previous analysis of adverse reactions have

NU
contained less than three trials, precluding the ability to make definitive conclusions
MA

(Davey et al., 2011). Another more recent meta-analysis (Meduri et al., 2016)

included 16 trials and observational studies of BD, however their analysis focused on
ED

the severity of changes in mania and response rates, and did not include subgroup

analyses for depression, remission rates or relapse rates. Following these two
T

meta-analyses, another two preliminary meta-analyses have also been published

EP

(Dundar et al., 2016; Lindstrom et al., 2017). Of them, the meta-analysis by Dundar
C

et al (2016) focused on pharmacologic treatment for acute agitation but not regular
AC

treatment. Their results revealed that olanzapine was superior to haloperidol, and

that no other treatment, including aripiprazole, was more effective than any other

(Dundar et al., 2016). On the other hand, the meta-analysis by Lindstrom et al (2017),

focused on the treatment effect of second-generation antipsychotics in the

maintenance phase, and found that aripiprazole was beneficial in preventing

relapses (Lindstrom et al., 2017). However, although their results showed statistical

significance, the results for aripiprazole were based on only two datasets, thereby

7
 ACCEPTED MANUSCRIPT

limiting the strength of their meta-analysis (Davey et al., 2011).

Given these gaps in the literature, we conducted this comprehensive systematic

review and meta-analysis to investigate the efficacy and safety profile of aripiprazole

in the treatment of BD, specifically focusing on sub-group analyses of depression,

laboratory data, adverse reactions, and preventive efficacy of mood episodes

PT
including relapse rates.

RI
SC
NU
MA
T ED
C EP
AC

8
 ACCEPTED MANUSCRIPT

Method and Materials

The current meta-analysis was conducted according to the Preferred Reporting

Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Liberati et al.,

2009) (Supplement Table 1) following a pre-specified but unpublished protocol. This

meta-analysis met the requirements of the Institutional Review Board of Tri-Service

PT
General Hospital (TSGHIRB: B-105-12).

RI
SC
Inclusion criteria NU
The inclusion criteria were: (1) published randomized controlled trials (RCTs)

exploring the efficacy of aripiprazole as monotherapy or combination therapy versus

MA

a placebo or other antipsychotic medication; (2) articles including patients with the

diagnosis of BD with either manic/mixed or depressed episodes based on the

ED

Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria or International

Classification of Diseases (ICD) codes; and (3) articles that were clinical trials
T
EP

including adult and adolescent populations, for whom aripiprazole has been

approved to treat BD by the Food and Drug Administration (FDA) in the USA and the
C

European Medicines Agency (EMA) in the EU. No limit was set for the length of study
AC

follow-up. The exclusion criteria were: (1) articles that were not controlled

intervention studies; and (2) animal studies.

Database searches and identification of eligible papers

Two independent authors searched PubMed and ScienceDirect using the

9
 ACCEPTED MANUSCRIPT

keywords (aripiprazole) AND (bipolar disorder OR mania OR depress) from inception

to May 14th, 2017 with the limitations of “humans; clinical trial” for PubMed and

“bipolar disorder/antipsychotic; journal article” for ScienceDirect. In addition, the

reference lists of review articles relevant to this topic were manually searched to

identify potentially eligible papers (Arbaizar et al., 2009; Biederman et al., 2005;

Brown et al., 2013; Fountoulakis et al., 2009; Fountoulakis et al., 2011; McKeage,

PT
2014; Meduri et al., 2016; Miura et al., 2014; Perlis, 2007; Perlis et al., 2006; Scherk

RI
et al., 2007; Smith et al., 2007; Taylor, 2003; Vieta et al., 2010; Yatham et al., 2013;

SC
Yildiz et al., 2015).

After completing the searches, duplicate studies were removed, and two
NU
independent authors screened the titles and abstracts for eligibility. A list of
MA

potentially eligible studies was developed by consensus, and the full texts were

assessed by the two authors. Both authors then applied the eligibility criteria and
ED

developed a final list of studies to be included. A third reviewer was available for

mediation in the event of any inconsistencies.

T
C EP

Methodological quality appraisal

AC

The methodological quality of the included studies was determined by two

authors using the Jadad scale (Jadad et al., 1996). In brief, the Jadad score was

calculated for each study and included three categories of study quality:

randomization, blindness, and withdrawals and dropouts. The Jadad score ranged

from zero (poor quality) to five (high quality).

10
 ACCEPTED MANUSCRIPT

Primary outcomes

The primary outcomes were differences in changes of measures of mania,

depression, relapse rates, or general severity as measured using the Clinical Global

Impressions Scale-Severity (CGI-S) from pre-treatment to post-treatment between

aripiprazole and a placebo or other medications (haloperidol or lithium). The

PT
outcomes of interest had to be recorded using a validated scale such as the Hamilton

Rating Scale for Depression (17-items) (HAM-D-17) (Williams, 1988) or

RI
Montgomery-Åsberg Depression Rating Scale (MADRS) (Montgomery and Asberg,

SC
1979) for depressive symptoms, and Young Mania Rating Scale (YMRS) for mania

symptoms (Young et al., 1978).

NU
MA

Secondary outcomes
ED

The secondary outcomes of interest included response rates, remission rates,

drop-out rates, rates of discontinuation of trials due to side effects, cognitive

T
EP

subscales of the Positive and Negative Syndrome Scale (PANSS), and the incidence of

any adverse reactions. Other measurements of disease severity including the PANSS
C

(Kay et al., 1987) for psychotic features, and CGI-S (Busner and Targum, 2007) for
AC

mania or depression were also recorded.

Response was defined as ≥ 50% reduction from baseline in YMRS total score

(Jeong et al., 2012) or ≥ 50% decrease from baseline in MADRS total score (Thase et

al., 2008). Remission was defined as a MADRS total score ≤ 8 (Thase et al., 2008),

YMRS total score ≤ 12 or CGI-S ≤ 2 (El Mallakh et al., 2010; Findling et al., 2009).

Relapse was defined as discontinuation of the study due to a lack of efficacy (Keck et

11
 ACCEPTED MANUSCRIPT

al., 2007), the need for hospitalization (Carlson et al., 2012; Jeong et al., 2012), YMRS

total score ≥ 18 or MADRS total score ≥ 18 or ≥ 4-point increase from baseline

(El-Mallakh et al., 2012), YMRS total score > 14 and a MADRS total score ≤ 16 for a

relapse of a manic episode, a YMRS total score > 14 and a MADRS total score > 16 for

a relapse of a mixed episode, and a YMRS total score ≤ 14 and a MADRS total score >

16 for a relapse of a depressive episode (Thase et al., 2008).

PT
RI
SC
Data extraction and management

Two independent authors extracted data from the eligible studies into a
NU
database. The variables of interest included mean age (years), age at onset, gender,
MA

duration of illness, treatment duration, dosage of prescribed aripiprazole, educa tion,

marriage status, occupation, ethnicity (Caucasian, Africa American, Hispanic, Asian,

ED

or Native America), Jadad score (Jadad et al., 1996), and all primary and secondary

outcomes as listed above. When data were not available in the articles, we contacted
T

the corresponding authors twice over a month-long period in an attempt to acquire

EP

the variables of interest.

C
AC

Meta-analysis

Under the presumed heterogeneity of the sample populations among all of the

recruited studies, the data were analyzed using random-effects meta-analysis

models rather than fixed effect models (Borenstein et al., 2010) with Comprehensive

Meta-Analysis software, version 3 (Biostat, Englewood, NJ). Continuous outcomes

12
 ACCEPTED MANUSCRIPT

were analyzed by calculating the effect sizes (ESs) as Hedges’ g and 95% confidence

intervals (95% CIs) comparing outcomes between aripiprazole and a placebo or other

psychotropic medications. Odds ratios (ORs) and 95% CIs were calculated for

dichotomous data. When measuring the efficacy with regards to psychotic features,

total PANSS scores were included into the analysis for a more complete evaluation, if

the study provided both total PANSS and PANSS scores of positive symptoms. Data on

PT
adverse events were also pooled across studies and compared versus a placebo

RI
where possible.

SC
A two-tailed p value of less than 0.05 was considered to be statistically

significant. Heterogeneity was assessed using the Q statistic to evaluate the

NU
dispersion of the true effect among the recruited studies (Higgins et al., 2003).
MA

Publication bias was evaluated by visual inspection of funnel plots and Egger’s

regression tests (Egger et al., 1997). In cases of publication bias, the trim and fill test
ED

was conducted to impute potential missing trials and re-calculate the effect size

(Duval and Tweedie, 2000). Finally, meta-regression and subgroup meta-analyses

T

were conducted to investigate the effects of clinical variables as possible sources of

EP

heterogeneity. Meta-regression analyses were performed using the unrestricted

C

maximum likelihood method only when data were available across more than five
AC

studies. The variables of interest for meta-regression were mean age, female gender,

body mass index (BMI), treatment duration, dosage of prescribed medications,

duration of illness, education, marriage status, occupation, ethnicity (Caucasian,

Africa American, Hispanic, Asian, or Native America), drop-out rate, and Jadad score.

Subgroup meta-analysis was performed when at least three sets of data were

available (Davey et al., 2011). Subgroup analyses included monotherapy or

13
 ACCEPTED MANUSCRIPT

combination therapy, acute or maintained course, and different categories of

treatment doses in order to clarify the efficacy in clinical practice.

PT
RI
SC
NU
MA
T ED
C EP
AC

14
 ACCEPTED MANUSCRIPT

Results

Study selection

In the initial search, 373 studies were identified after removing duplicates (59

from PubMed and 341 from ScienceDirect). After screening the titles/abstracts, 43

full-text articles were assessed for eligibility, of which 22 were excluded (Figure 1 and

PT
Supplement Table 2). Therefore, 20 articles were eligible for the current

RI
meta-analysis (Carlson et al., 2012; El-Mallakh et al., 2012; El Mallakh et al., 2010;

SC
Findling et al., 2013; Findling et al., 2009; Findling et al., 2012; Jeong et al., 2012;

Kanba et al., 2014; Keck et al., 2007; Keck et al., 2003; Keck et al., 2009; Muzina et al.,
NU
2008; Quante et al., 2010; Sachs et al., 2006; Thase et al., 2008; Tramontina et al.,

2009; Vieta et al., 2005; Vieta et al., 2008; Woo et al., 2011; Young et al., 2009) (Table
MA

1).
ED

Among the 20 eligible articles, two investigated the efficacy of aripiprazole

monotherapy in comparison with haloperidol (Vieta et al., 2005; Young et al., 2009)
T

(aripiprazole participants = 340; haloperidol participants = 337). One article (Keck et

EP

al., 2009) compared aripiprazole monotherapy and lithium in the acute phase of
C

illness (aripiprazole = 155; lithium = 160), another article (El-Mallakh et al., 2012)
AC

compared aripiprazole monotherapy versus lithium in the maintenance stage

(aripiprazole = 25; lithium = 38), and another article (Jeong et al., 2012) investigated

the combination of aripiprazole with valproic acid versus haloperidol with valproic

acid (aripiprazole = 28; haloperidol = 14). The remaining 15 articles (Carlson et al.,

2012; El Mallakh et al., 2010; Findling et al., 2013; Findling et al., 2009; Findling et al.,

2012; Kanba et al., 2014; Keck et al., 2007; Keck et al., 2003; Muzina et al., 2008;

Quante et al., 2010; Sachs et al., 2006; Thase et al., 2008; Tramontina et al., 2009;

15
 ACCEPTED MANUSCRIPT

Vieta et al., 2008; Woo et al., 2011) compared aripiprazole

monotherapy/combination therapy with a placebo in the acute or maintenance stage

(aripiprazole = 1923; placebo = 1499).

Methodological quality of the included studies

PT
Across the 20 studies, the average Jadad score was 3.35 with a standard

RI
deviation (SD) of 0.49 (Supplementary Table 3).

SC
NU
Meta-analysis investigating the effect of aripiprazole on BD compared to a placebo
MA

or other medication

1. Meta-analysis comparing the treatment efficacy of aripiprazole monotherapy

ED

and a placebo in the acute stage of BD

T

1.1. Primary outcome: comparing treatment efficacy for depression and mania
EP

1.1.1. Treatment efficacy for depression

C
AC

Across four articles (Findling et al., 2009; Sachs et al., 2006; Tramontina et al.,

2009; Young et al., 2009), there was no significant difference in treatment effect

between aripiprazole monotherapy (case number, n = 519 in five comparative arms)

and a placebo (n = 412 in four arms) for the acute stage of bipolar depression

(Hedges’ g: -0.127, 95% CI: -0.257 to 0.002, p = 0.054, RCTs = 4 with five datasets)

(Figure 2A). There was no significant heterogeneity (Q value = 4.41, df = 4, p = 0.353;

I2 = 9.287%, tau = 0.046) or publication bias according to Egger’s test (t = 0.512, df = 3,

16
 ACCEPTED MANUSCRIPT

2-tailed p = 0.644). Following a sensitivity test and removal of the study by Young et

al (Young et al., 2009), aripiprazole was found to be more effective in treating

depression versus a placebo (Hedges’ g: -0.208, 95% CI: -0.355 to -0.061, p = 0.006).

We did not perform further subgroup meta-analysis of trials with adults only as

fewer than three datasets were available (Sachs et al., 2006; Young et al., 2009). We

PT
divided the trials into two categories by daily dosage (< 20 mg/day and > 20 mg/day)

to evaluate whether or not there was a dose-dependent effect. Only two studies

RI
used a dose of < 20 mg/day, thus precluding meta-analysis (Findling et al., 2009;

SC
Tramontina et al., 2009). Across three RCTs, no significant difference between

aripiprazole (n = 403 in three arms) and placebo (n = 387 in three arms) was noted in
NU
the > 20 mg/day group (Findling et al., 2009; Sachs et al., 2006; Young et al., 2009)
MA

(Hedges’ g: -0.097, 95% CI: -0.279 to 0.086, p = 0.299).

Meta-regression analysis showed that age, female sex, dosage of aripiprazole,

ED

ethnicity (Caucasian), and Jadad scores (randomization, blind, total) were all

non-significant (data not shown, available on request).

T
C EP

1.1.2 Treatment efficacy and relapse rates for mania

AC

To determine the efficacy for mania, five articles (El Mallakh et al., 2010;

Findling et al., 2009; Sachs et al., 2006; Tramontina et al., 2009; Young et al., 2009)

were eligible for the analysis. Aripiprazole monotherapy (n = 786 in seven arms) was

significantly superior to placebo (n = 546 in five arms) in treating mania (Hedges’ g:

-0.299, 95% CI: -0.476 to -0.123, p = 0.001) (Figure 2B). Egger`s regression test

showed no significant publication bias (t = 1.519, df = 5, 2-tailed p = 0.189); however,

17
 ACCEPTED MANUSCRIPT

heterogeneity was significant (Q value = 17.51, df = 6, p = 0.008; I2 = 65.73%, tau =

0.188). With regards to sensitivity, the significance of the results did not change after

removing any one of the recruited studies.

The results of subgroup meta-analysis of trials with adults only revealed no

significant difference in the treatment effect on manic symptoms between

PT
aripiprazole and a placebo (Hedges’ g: -0.162, 95% CI: -0.328 to 0.003, p = 0.055) (El

Mallakh et al., 2010; Sachs et al., 2006; Young et al., 2009). However, after removing

RI
data from El Mallakh et al’s study (2010) which included aripiprazole 15 mg only (El

SC
Mallakh et al., 2010), the results showed a significantly better treatment effect on

manic symptoms in the aripiprazole arms than in the placebo arms (Hedges’ g: -0.218,
NU
95% CI: -0.392 to -0.043, p = 0.014). In addition, in the lower dose group (< 20
MA

mg/day) (El Mallakh et al., 2010; Findling et al., 2009; Tramontina et al., 2009),

aripiprazole (n = 247 in three arms) was not effective compared with a placebo (n =
ED

258 in three arms) (Hedges’ g: -0.386, 95% CI: -0.859 to -0.087, p = 0.109). However,

in the higher dose group (> 20 mg/day) (El Mallakh et al., 2010; Findling et al., 2009;
T

Sachs et al., 2006; Young et al., 2009), aripiprazole (n = 539 in four arms) was
EP

superior to a placebo in treating mania (n = 521 in four arms) (Hedges’ g: -0.272, 95%
C

CI: -0.444 to -0.101, p = 0.002).

AC

It was possible to perform subgroup meta-analysis of studies of a shorter

duration (< 4 weeks) (El Mallakh et al., 2010; Young et al., 2009), and the results

revealed no significant difference between aripiprazole (n = 434 in three arms) and a

placebo (n = 287 in two arms) (Hedges’ g: -0.097, 95% CI: -0.231 to 0.037, p = 0.157).

There were insufficient data for studies of a longer treatment duration (> 4 weeks)

since only one article (Tramontina et al., 2009) met the inclusion criteria.

18
 ACCEPTED MANUSCRIPT

For relapse rates of mania, three datasets from two articles (Thase et al., 2008;

Young et al., 2009) revealed no significant difference (OR: 0.829, 95% CI: 0.407 to

1.688, p = 0.605) between aripiprazole (n=538 in 3 arms) and a placebo (n = 529 in

three arms) without significant heterogeneity (Q value = 0.718, df = 2, p = 0.698; I2 <

0.001%, tau < 0.001) but significant publication bias (t value = 135.796, df = 1, p =

0.005). However, after the trim and fill test (with two potentially missing studies to

PT
the left of the mean), no change in the results of meta-analysis was noted (OR: 0.650,

RI
95% CI: 0.358 to 1.180, p = 0.157). In addition, no change was noted in a sensitivity

SC
analysis after removing one study.

In meta-regression analysis, older age was positively correlated with a greater

NU
effect size (coefficient = 0.0137, p = 0.0081) in treatment efficacy for mania. No
MA

relationship was noted for other variables (female sex, status of BD (percentage of

manic mood state, rapid-cycling, or mixed episodes), dosage of aripiprazole, and

ED

ethnicity (Caucasian, Hispanic).

T
EP

1.2 Secondary outcomes: other measurements of efficacy and adverse reactions

C

1.2.1 Efficacy profile: response rates, remission rates, and other measurements of
AC

severity (CGI-S, PANSS)

There were insufficient data to analyze response rates or remission rates for

depression. In three articles which reported remission rates for mania (Findling et al.,

2009; Tramontina et al., 2009; Young et al., 2009), aripiprazole was significantly more

effective than a placebo (OR: 4.893, 95% CI: 1.373 to 17.432, p = 0.014) with

significant heterogeneity (Q value = 25.721, df = 3, p < 0.001; I2 = 88.336%, tau =

19
 ACCEPTED MANUSCRIPT

1.201) with no significant publication bias (t value = 2.344, df = 2, p = 0.144). The

significance of the results became non-significant after removing the dataset of

Findling et al (10 mg aripiprazole/day) (OR: 4.689, 95% CI: 0.858 to 25.618, p = 0.075)

(Findling et al., 2009) or Tramontina et al (OR: 4.786, 95% CI: 0.987 to 23.214, p =

0.052) (Tramontina et al., 2009). Response rates for mania were reported in five

articles (El Mallakh et al., 2010; Findling et al., 2009; Sachs et al., 2006; Tramontina et

PT
al., 2009; Young et al., 2009), and aripiprazole was significantly better than a placebo

RI
(OR: 2.054, 95% CI: 1.366 to 3.089, p = 0.001) with significant heterogeneity (Q value

SC
= 21.062, df = 6, p = 0.002; I2 = 71.513%, tau = 0.450) but no significant publication

bias (t value = 1.625, df = 5, p = 0.165). The results did not change after the removal
NU
of one study.
MA

The CGI-S was used to estimate the general severity of illness for mania,

depressed or mixed episodes across nine articles (El Mallakh et al., 2010; Findling et
ED

al., 2009; Kanba et al., 2014; Keck et al., 2003; Keck et al., 2009; Sachs et al., 2006;

Thase et al., 2008; Tramontina et al., 2009; Young et al., 2009). After treatment,
T

aripiprazole was associated with a greater improvement in CGI-S scores than a

EP

placebo (Hedges’ g: -0.322, 95% CI: -0.418 to -0.225, p < 0.001) without evidence of

heterogeneity (Q value = 16.678, df = 10, p = 0.082; I2 = 40.040%, tau = 0.102) but

C
AC

with significant publication bias (t value = 2.317, df = 9, p = 0.046). However, after the

trim and fill test (with two potentially missing studies to the right of the mean), the

results of the meta-analysis did not change (Hedges’ g: -0.293, 95% CI: -0.391 to

-0.195, p < 0.001). The results did not change after the removal of one study.

In four articles (Kanba et al., 2014; Keck et al., 2009; Sachs et al., 2006; Young et

al., 2009), aripiprazole was more effective than a placebo in improving psychotic

20
 ACCEPTED MANUSCRIPT

features (Hedges’ g: -0.296, 95% CI: -0.411 to -0.181, p < 0.001). No significant

heterogeneity was noted (Q value = 0.101, df = 3, p = 0.992; I2 < 0.001%, tau < 0.001)

although there was evidence of publication bias (t value = 12.390, df = 2, p = 0.006).

However, the trim and fill test (with two potentially missing studies to the right of the

mean) did not change the main result (Hedges’ g: -0.288, 95% CI: -0.393 to -0.184, p

< 0.001). No change in the result was noted after the removal of one study.

PT
RI
SC
Adverse outcomes and dropout rates

1.2.2 Metabolic profile: fasting glucose, BMI, body weight, high density lipoprotein
NU
(HDL), total cholesterol, triglyceride (TG), and increased appetite
MA

Compared to a placebo, aripiprazole was associated with a lower risk of elevated

fasting glucose (Hedges’ g: 0.447, 95% CI: 0.048 to 0.846, p = 0.028, RCTs = 2 with
ED

three datasets), increased appetite (OR: 2.114, 95% CI: 1.058 to 4.226, p = 0.034,

RCTs = 2 with four datasets), and total cholesterol (OR: 1.686, 95% CI: 1.032 to 2.754,
T
EP

p = 0.037, RCTs = 2 with three datasets). No significant differences were noted

between aripiprazole and a placebo in changes in BMI (Hedges’ g: 0.170, 95% CI:
C

-0.017 to 0.357, p = 0.074, RCTs = 2 with three datasets) and body weight (Hedges’ g:
AC

0.051, 95% CI: -0.109 to 0.210, p = 0.535, RCTs = 4 with five datasets). Furthermore,

aripiprazole did not affect levels of high density lipoprotein (HDL) (Hedges’ g: -0.233,

95% CI: -0.488 to 0.022, p = 0.074, RCTs = 2 with three datasets) or TG (Hedges’ g:

0.082, 95% CI: -0.064 to 0.228, p = 0.273, RCTs = 2 with three datasets) compared to

a placebo (Table 3).

21
 ACCEPTED MANUSCRIPT

1.2.3 Adherence profile: dropout rates and rates of discontinuation due to side

effects

There was no significant difference in dropout rate due to any reason between

aripiprazole and a placebo (OR: 0.923, 95% CI: 0.723 to 1.178, p = 0.521, RCTs = 9

with 12 datasets). In contrast, aripiprazole was associated with higher rates of

PT
discontinuation due to side effects compared to a placebo (OR: 1.551, 95% CI: 1.155

to 2.082, p = 0.004, RCTs = 8 with 11 datasets) (Table 3).

RI
SC
1.2.4 Gastro-intestinal profile (constipation, diarrhea, nausea, vomiting, dyspepsia)
NU
Aripiprazole was associated with higher rates of constipation (OR: 1.720, 95% CI:
MA

1.251 to 2.365, p = 0.001, RCTs = 6 with nine datasets), nausea (OR: 1.816, 95% CI:

1.427 to 2.310, p < 0.001, RCTs = 7 with 10 datasets), and vomiting (OR: 2.083, 95%
ED

CI: 1.354 to 3.203, p = 0.001, RCTs = 6 with nine datasets) compared to a placebo.

However, there were no significant differences in diarrhea (OR: 0.956, 95% CI: 0.694
T
EP

to 1.319, p = 0.785, RCTs = 5 with seven datasets) and dyspepsia (OR: 1.444, 95% CI:

0.891 to 2.339, p = 0.136, RCTs = 4 with six datasets) between aripiprazole and a
C

placebo (Table 3).

AC

1.2.5 Other adverse events: agitation, akathisia, anxiety, dizziness, over-sedation,

light headedness, headache, fatigue, insomnia, pain in the extremities, and profile

of extrapyramidal symptoms (EPS and sialorrhea)

22
 ACCEPTED MANUSCRIPT

Higher incidence rates of akathisia (OR: 5.104, 95% CI: 3.733 to 6.980, p < 0.001,

RCTs = 8 with 11 datasets), anxiety (OR: 1.526, 95% CI: 1.058 to 2.203, p = 0.024, RCTs

= 4 with six datasets), EPS (Hedges’ g: 0.538, 95% CI: 0.070 to 1.007, p = 0.024 , RCTs

= 4 with five datasets), hyper-salivation (OR: 4.696, 95% CI: 1.535 to 14.369, p =

0.007, RCTs = 2 with three datasets), fatigue (OR: 2.272, 95% CI: 1.443 to 3.577, p <

0.001, RCTs = 2 with four datasets), insomnia (OR: 1.496, 95% CI: 1.028 to 2.177, p =

PT
0.035, RCTs = 5 with seven datasets), over-sedation (OR: 3.607, 95% CI: 2.267 to

RI
5.739, p < 0.001, RCTs = 5 with seven datasets), and pain in the extremities (OR:

SC
2.281, 95% CI: 1.260 to 4.127, p = 0.006, RCTs = 2 with three datasets) were observed

with aripiprazole than with a placebo. However, there were no significant differences
NU
between aripiprazole and a placebo in dizziness (OR: 1.361, 95% CI: 0.719 to 2.575, p

= 0.344, RCTs = 2 with four datasets), agitation (OR: 1.079, 95% CI: 0.773 to 1.508, p =
MA

0.654, RCTs = 3 with four datasets), headache (OR: 1.029, 95% CI: 0.857 to 1.235, p =

0.762, RCTs = 8 with 11 datasets), and light headedness (OR: 1.108, 95% CI: 0.737 to
ED

1.666 p = 0.621, RCTs = 3 with four datasets) (Table 3).

T
EP

2. Meta-analysis comparing the treatment efficacy of aripiprazole monotherapy

C

and a placebo in the maintenance stage of BD

AC

2.1. Primary outcome: rates of total relapse and general severity (CGI-S)

Data were available from three studies to analyze the treatment efficacy during

the maintenance phase (Findling et al., 2013; Keck et al., 2007; Muzina et al., 2008).

In two of the articles (Findling et al., 2013; Keck et al., 2007), aripiprazole (n= 185 in

three arms) was significantly superior to a placebo (n = 91 in two arms) in reducing

23
 ACCEPTED MANUSCRIPT

CGI-S (Hedges’ g: -0.483, 95% CI: -0.674 to -0.292, p < 0.001) (Figure 2C) without

significant heterogeneity (Q value = 1.436, df = 2, p = 0.488; I2 < 0.001%, tau < 0.001)

or publication bias (t value = 9.461, df = 1, p = 0.067). The results were unchanged

after the removal of one study.

We could not perform further meta-analysis of total relapse rate in the

PT
maintenance stage of aripiprazole monotherapy and placebo monotherapy because

only two studies contained relevant data (Keck et al., 2007; Muzina et al., 2008). In

RI
these studies, the total relapse rates were significantly lower in the aripiprazole

SC
group than in the placebo group in the study by Keck et al (2007) (33% in the

aripiprazole group and 52% in the placebo group) (Keck et al., 2007), and in the study
NU
by Muzina et al the time to relapse was significantly longer in the aripiprazole group
MA

than in the placebo group at week 100 (hazard ratio = 0.18, log-rank p = 0.017)

(Muzina et al., 2008).

T ED

2.2 Secondary outcome: other measurements of efficacy and adverse reactions

EP

2.2.1 Metabolic profile: fasting glucose, body weight, HDL, total cholesterol, and TG
C
AC

No significant differences were noted between treatment with aripiprazole and

a placebo in fasting glucose (Hedges’ g: 0.169, 95% CI: -0.013 to 0.361, p = 0.069,

RCTs = 3 with four datasets), increase in body weight (OR: 2.228, 95% CI: 0.856 to

5.801, p = 0.101, RCTs = 2 with three datasets), change in total cholesterol (Hedges’ g:

0.179, 95% CI: -0.156 to 0.514, p = 0.295, RCTs = 3 with four datasets), and changes in

TG (Hedges’ g: 0.013, 95% CI: -0.606 to 0.631, p = 0.968, RCTs = 3 with four datasets).

However, aripiprazole was associated with an increased level of HDL compared to a

24
 ACCEPTED MANUSCRIPT

placebo (Hedges’ g: -0.710, 95% CI: -1.165 to -0.254, p = 0.002, difference in means =

-2.622, 95% CI: -4.370 to -0.875, RCTs = 3 with four datasets) (Table 3).

2.2.2 Adherence profile: dropout rates, and rates of discontinuation due to side

effects

PT
Aripiprazole was associated with significantly higher rates of discontinuation of

RI
treatment due to side effects compared to a placebo (OR: 7.897, 95% CI: 1.740 to

SC
35.848, p = 0.007, RCTs = 3 with four datasets). On the other hand, aripiprazole was

associated with a significantly lower dropout rate compared to a placebo (OR: 0.417,
NU
95% CI: 0.249 to 0.696, p = 0.001, RCTs = 3 with four datasets) (Table 3).
MA
ED

2.2.3 Gastro-intestinal profile: diarrhea, nausea, and dry mouth

Aripiprazole was associated with significantly higher rates of diarrhea (OR: 7.472,
T
EP

95% CI: 1.322 to 42.234, p = 0.023, RCTs = 2 with three datasets), and dry mouth (OR:

5.801, 95% CI: 1.473 to 22.839, p = 0.012, RCTs = 3 with four datasets) compared to a
C

placebo (Table 3).

AC

2.2.4 Other adverse events: akathisia, back pain, over-sedation, headache, tremor,

EPS, and upper respiratory tract infection

Aripiprazole was associated with a significantly higher rate of akathisia (OR:

7.711, 95% CI: 2.162 to 27.499, p = 0.002, RCTs = 3 with four datasets) compared to a
25
 ACCEPTED MANUSCRIPT

placebo. However, there were no significant differences in the rates of headache (OR:

1.501, 95% CI: 0.853 to 2.641, p = 0.159, RCTs = 2 with three datasets) and upper

respiratory-tract infection (OR: 2.308, 95% CI: 0.925 to 5.763, p = 0.073, RCTs = 2 with

three datasets) between aripiprazole and a placebo (Table 3).

PT
3. Meta-analysis comparing the treatment efficacy of aripiprazole combination

RI
therapy and a placebo in the acute or maintenance stage of BD

SC
3.1 Primary outcome: relapse rates for maniaNU
Only two articles compared aripiprazole in combination therapy in the acute

phase of BD thus precluding meta-analysis (Quante et al., 2010; Vieta et al., 2008).
MA

Three trials (Carlson et al., 2012; Findling et al., 2012; Woo et al., 2011) were
ED

eligible in the maintenance phase with 248 patients in three arms receiving

aripiprazole and 246 in three arms receiving a placebo. Aripiprazole tended to have a
T

lower mania relapse rate (OR: 0.522, 95% CI: 0.291 to 0.937, p = 0.029) compared to
EP

a placebo without significant heterogeneity (Q value = 1.254, df = 2, p = 0.534; I2 <

C

0.001%, tau < 0.001) or publication bias (t value = 0.477, df = 1, p = 0.717). After
AC

removal of the study by Carlson et al (Carlson et al., 2012) (OR: 0.453, 95% CI: 0.107

to 1.926, p = 0.284) or Findling et al (Findling et al., 2012) (OR: 0.567, 95% CI: 0.310

to 1.040, p = 0.067), the results were not significant.

3.2 Secondary outcome: other measurements of adverse reactions

26
 ACCEPTED MANUSCRIPT

In terms of side effect profiles, aripiprazole was associated with a significantly

greater increase in total cholesterol (difference in means: 2.000, 95% CI: 1.993 to

2.007, p < 0.001, RCTs = 3 with three datasets). No significant differences were

observed between aripiprazole and a placebo in changes in fasting glucose

(difference in means: -1.967, 95% CI: -8.447 to 4.514, p = 0.552, RCTs = 3 with three

datasets), changes in TG (difference in means: 3.238, 95% CI: -8.162 to 14.638, p =

PT
0.578, RCTs = 3 with three datasets), incidence of headache (OR: 0.721, 95% CI: 0.262

RI
to 1.983, p = 0.526, RCTs = 3 with three datasets) and dropout rate (OR: 0.748, 95%

SC
CI: 0.381 to 1.468, p = 0.398, RCTs = 3 with three datasets) (Table 3).
NU
4. Meta-analysis comparing the treatment efficacy of aripiprazole monotherapy
MA

or combination therapy with other medications (lithium or haloperidol) in the

acute or maintenance stage of BD

ED

Meta-analysis comparing the treatment efficacy of aripiprazole monotherapy

T

and other medications during the maintenance phase was not possible becaus e only
EP

one article (El-Mallakh et al., 2012) contained sufficient relevant data. Another short
C

trial (Jeong et al., 2012) did not have sufficient data for meta-analysis in the acute or
AC

maintenance phase of BD.

4.1. Primary outcome: treatment efficacy for depressed symptoms, general

severity (CGI-S), response rates, and manic symptoms

Of three articles (Keck et al., 2009; Vieta et al., 2005; Young et al., 2009) with

27
 ACCEPTED MANUSCRIPT

data of the acute stage, aripiprazole monotherapy (n = 497 in three arms) did not

significantly differ from other medications (haloperidol = 337 in two arms and lithium

= 160 in one arm) in the treatment efficacy of depressed symptoms during the acute

phase (Hedges’ g: -0.039, 95% CI: -0.193 to 0.115, p = 0.620). No significant

heterogeneity (Q value = 3.085, df = 2, p = 0.214; I2 = 35.173%, tau = 0.081) or

publication bias (t = 0.2854, df = 1, 2-tailed p = 0.823) were observed. The results

PT
were unchanged after the removal of one study. Furthermore, aripiprazole was

RI
significantly better than the other medications (haloperidol or lithium) in CGI-S

SC
(Hedges’ g: -0.191, 95% CI: -0.315 to -0.067, p = 0.003) without significant

heterogeneity (Q value = 1.648, df = 2, p = 0.439; I2 < 0.001%, tau < 0.001) or

NU
publication bias (t = 0.044, df = 1, 2-tailed p = 0.972). Three articles reported the

efficacy for mania symptoms (Keck et al., 2009; Vieta et al., 2005; Young et al., 2009),
MA

in which aripiprazole showed no significant difference in response rates (OR: 1.271,

95% CI: 0.647 to 2.496, p = 0.486) compared to haloperidol or lithium. Subgroup

ED

analysis by medication (haloperidol vs. aripiprazole and lithium vs. aripiprazole) could

not be performed because of insufficient data (less than three RCTs).

T
C EP

4.2. Secondary outcome: other adverse reactions

AC

There were no significant differences between aripiprazole and other

medications (haloperidol or lithium) in akathisia (OR: 0.789, 95% CI: 0.241 to 2.586, p

= 0.696, RCTs = 3 with three datasets), dropout rate (OR: 0.812, 95% CI: 0.375 to

1.759, p = 0.598, RCTs = 3 with three datasets), rates of discontinuation due to side

effects (OR: 0.710, 95% CI: 0.212 to 2.378, p = 0.578, RCTs = 3 with three datasets),

symptoms of EPS (Hedges’ g: -0.436, 95% CI: -1.032 to 0.159, p = 0.151, RCTs = 3 with

28
 ACCEPTED MANUSCRIPT

three datasets), headache (OR: 1.079, 95% CI: 0.748 to 1.555, p = 0.685, RCTs = 3

with three datasets), and tremor (OR: 0.684, 95% CI: 0.433 to 1.082, p = 0.105, RCTs =

3 with three datasets) (Table 3).

PT
RI
SC
NU
MA
T ED
C EP
AC

29
 ACCEPTED MANUSCRIPT

Discussion

To the best of our knowledge, this is first meta-analysis to comprehensively

investigate the efficacy tolerability, and adverse reactions to aripiprazole from the

acute to maintenance stage within bipolar manic/mixed state to bipolar depression.

Comparison to two recent meta-analyses (Brown et al., 2013; Meduri et al., 2016),

PT
our study provided more information on the analysis of adverse reactions and

preventive efficacy as shown by relapse rates from bipolar mania to bipolar

RI
depression. The more recent search dates and stricter inclusion criteria (only RCTs in

SC
this meta-analysis) mean that the results of this meta-analysis are more up-to-date

compared to the previous reports. The main differences between the current study
NU
and the two previous meta-analyses are listed in Table 2. It should be noted that two
MA

(Findling et al., 2009; Tramontina et al., 2009) of the included articles reported the

efficacy of BD in adolescents. In addition, the dose-dependent responses and

ED

adverse reactions to antipsychotic drugs should be considered with caution; however,

aripiprazole has been approved for the treatment of BD in pediatric patients by the
T

FDA in the USA and the EMA in the EU for its efficacy and safety. Furthermore, the
EP

previous two studies (Brown et al., 2013; Meduri et al., 2016) also included above
C

two studies into their analyses. However, despite the apparent good tolerability, the
AC

results of this meta-analysis should be interpreted with caution for application in

clinical practice in pediatric patients.

To differentiate the does-effect from primary outcome, we set the cutoff point of

a higher and lower dose at 20 mg/day. A previous meta-analysis (Brown et al., 2013)

divided the enrolled trials into three groups at 10 mg/day, 15 mg/day, and 30 mg/day.

However, in order to include more trials in each group, we only divided them into

30
 ACCEPTED MANUSCRIPT

two groups. Furthermore, an animal study (Li et al., 2004) demonstrated that a

higher dose (30 mg/kg) in rodents could act as an antagonist to decrease cortical

dopamine release, whereas a lower dose (0.3 mg/kg) would increase dopamine

release in the medial prefrontal cortex and hippocampus. Setting a relatively higher

cut-off point of 20 mg/day could prevent the effect of dopamine agonism with

aripiprazole, although no human trials have replicated this finding.

PT
We summarized the efficacy of treating depression with aripiprazole. There was

RI
no significant difference in the efficacy between aripiprazole monotherapy and a

SC
placebo for acutely bipolar depression, although there was an insignificant trend

favoring aripiprazole (effect size: -0.127, p = 0.054). However, its efficacy reached
NU
statistical significance when removing the study by Young et al (Young et al., 2009)
MA

during the sensitivity test. This may be due to the relatively large sample (n = 320)

compared to other studies (n = 272 in Sachs et al; 43 in Tramontina et al; 296 in

ED

Findling et al with three comparison arms). Furthermore, a relatively higher dropout

rate (43.1%) and discontinuation rate due to adverse reactions (14.4%) may have
T

excluded potential participants responsive to aripiprazole. A higher treatment dose

EP

of aripiprazole (> 20 mg/day) did not show increased efficacy (effect size: -0.097, p =
C

0.299), although an insufficient number of trials with a lower treatment dose

AC

prevented further analysis. Of three previous meta-analyses (Fountoulakis et al.,

2011; Meduri et al., 2016; Vieta et al., 2010), none showed positive results with the

efficacy of aripiprazole at the endpoint; however, one study mentioned early efficacy

for bipolar depression (Fountoulakis et al., 2011). Aripiprazole monotherapy did not

significantly differ from haloperidol or lithium in the treatment efficacy of depressive

symptoms during the acute phase, as assessed by reduced scores in the MADRS. In a

short trial on combination therapy with aripiprazole in the acute stage, Quante et al

31
 ACCEPTED MANUSCRIPT

(Quante et al., 2010) reported that aripiprazole plus citalopram was beneficial in

treating bipolar depression compared to a placebo plus citalopram. These results

may demonstrate the potential efficacy of aripiprazole in treating bipolar depression,

and further studies focusing on bipolar depression may help to clarify the efficacy of

aripiprazole.

PT
For trials on mania with sufficient data for analysis, aripiprazole monotherapy

was significantly superior to a placebo with a higher dose (> 20 mg/day), but not

RI
with a lower dose (< 20 mg/day) or shorter treatment duration (< 4 weeks). From

SC
response and remission rates as secondary outcomes in the acute stage, aripiprazole

was significantly better than a placebo. For sub-groups without enough trials to allow
NU
for analysis, one study with a longer treatment group (> 4 weeks) (Tramontina et al.,
MA

2009) and another trial (Vieta et al., 2008) reported that aripiprazole combination

therapy in the acute stage showed significant efficacy. Previous systemic reviews and
ED

meta-analyses (Arbaizar et al., 2009; Brown et al., 2013; Fountoulakis et al., 2011;

Meduri et al., 2016; Yildiz et al., 2015) reported significant efficacy in treating mania
T

with aripiprazole, which is consistent with our results. Compared to the study
EP

focusing on mania (Brown et al., 2013), we used grouping to evaluate

C

dose-dependent and time effects. As a result, our results suggest that a higher dose
AC

(> 20 mg/day) and longer treatment period (> 4 weeks) may achieve better outcomes

for bipolar mania, although previous publications (Dhillon, 2012; Schatzberg and

Nemeroff, 2009) suggested that a therapeutic dose of 15 to 30 mg/day may be

appropriate for bipolar mania. In comparisons of aripiprazole monotherapy versus a

placebo in acute mania, our meta-regression revealed that an older age was

associated with a poorer outcome with aripiprazole treatment than with a placebo

with a significant but relatively small effective coefficient ratio (coefficient = 0.0137,

32
 ACCEPTED MANUSCRIPT

p = 0.0081). A previous study investigating predictors for recovery from mania

showed that an older age was associated with a faster recovery from mania (Tohen et

al., 2003), although this may be not meaningful for our results due to the relatively

small effect size.

Few studies (Fountoulakis et al., 2011) have investigated relapse rates under

PT
aripiprazole treatment for BD. We focused on the relapse rates as preventive efficacy

within acute or maintenance states. No significance differences in relapse rates were

RI
observed between aripiprazole monotherapy and a placebo in the acute stage of BD,

SC
however, combination aripiprazole therapy showed a better preventive effect as

measured by relapse rates (OR: 0.522, p = 0.029) versus a placebo in the

NU
maintenance stage of BD. Although there were insufficient trials to allow for
MA

subgroup analysis of aripiprazole monotherapy versus a placebo in maintenance

therapy, both studies (Keck et al., 2009; Muzina et al., 2008) showed promising
ED

efficacy of aripiprazole. Therefore, it may be more appropriate to use relapse rates as

the primary outcome in the maintenance phase than in the acute phase in clinical
T

practice, because the maintenance phase has a longer period of follow-up than the
EP

acute phase. Furthermore, our results differ from a previous meta-analysis (Miura et
C

al., 2014) which demonstrated no significant efficacy in relapse rates in the

AC

maintenance state of BD. In summary, the current study clarifies the efficacy of

relapse rates, and suggests that they can be applied to clinical practice in the future.

We also analyzed the CGI-S as an alternative measurement of efficacy of

aripiprazole treatment. Comparing aripiprazole monotherapy and a placebo in the

acute stage of BD, aripiprazole was associated with significant improvements in the

CGI-S than a placebo across manic/mixed and depressed states, and also with

33
 ACCEPTED MANUSCRIPT

aripiprazole monotherapy and a placebo in the maintenance stage. These results are

comparable to previous studies (Brown et al., 2013; Fountoulakis et al., 2011) which

reported categories of CGI, however further clinical trials are necessary for subgroup

analysis within different categories of mood state. In three trials (Keck et al., 2009;

Vieta et al., 2005; Young et al., 2009), aripiprazole was significantly better than other

medications (haloperidol or lithium) in the CGI-S (effect size: -0.191, p = 0.003) with

PT
regards to treatment efficacy of BD during the acute phase. In addition, the efficacy

RI
of treatment for psychotic features is also important due to the effect on recovery in

SC
BD. Psychotic mania has been associated with longer periods of remission (Rosenthal

et al., 1979), higher rates of relapse, worse social functioning, and occupational
NU
status (Tohen et al., 1990) compared to non-psychotic mania. Moreover, psychotic

features are highly associated with suicide attempts in patients with BD (McGrady et
MA

al., 2017). Aripiprazole monotherapy tended to reduce PANSS scores compared to a

placebo in the acute phase, indicating a beneficial effect in treating psychotic

ED

features. Another meta-analysis (Fountoulakis et al., 2009) also supports the

usefulness of aripiprazole for psychotic symptoms during the acute manic and
T
EP

maintenance phases of BD, which is consistent with our results.

C

Antipsychotic medications are associated with a wide range of potential adverse

AC

effects, including metabolic syndrome. A population-based cohort study (Chien et al.,

2010) found that the use of antipsychotics was hazardous for patients with BD

suffering from diabetes. We comprehensively analyzed associations between

aripiprazole and the metabolic syndrome/diabetes. Compared to a placebo in the

acute phase, aripiprazole was significantly associated with a lower risk of elevated

fasting glucose, increased appetite, and level of total cholesterol. No significant

differences in levels of HDL, TG, BMI or body weight were found. In the maintenance

34
 ACCEPTED MANUSCRIPT

stage of BD, aripiprazole was still not associated with significant differences in fasting

glucose, increased body weight, level of total cholesterol, or level of TG. Of note,

aripiprazole was associated with a significantly higher level of HDL than a placebo

(effect size: -0.710, p = 0.002) with a mean increment of 2.662 mg/dl after

aripiprazole treatment. HDL has been shown to be beneficial as a cardiovascular

protector (Gomaraschi et al., 2016) and to exert anti-diabetogenic functions on beta

PT
cells of the endocrine pancreas (von Eckardstein and Widmann, 2014). In

RI
combination therapy, aripiprazole was associated with lower levels of fasting glucose

SC
and TG, and a smaller increase in body weight than a placebo. However, a significant

increase in total cholesterol of 2 mg/dl was found in the aripiprazole group compared
NU
with a placebo. Considering the normal ranges of total cholesterol (< 200 mg/dl) and

HDL (40-60 mg/dl), an increase in total cholesterol of 2 mg/dl may not be clinically
MA

meaningful. In conclusion, aripiprazole seems to be a safe therapeutic option with

regards to metabolic profile. We then evaluated possibly associations between

ED

gastro-intestinal disturbances and aripiprazole treatment (Taylor, 2003) with regards

to short-term nausea, vomiting, and constipation. Our result showed that

T
EP

aripiprazole should be used cautiously due to gastro-intestinal problems. In the acute

stage, aripiprazole was significantly associated with higher rates of constipation,

C

nausea, and vomiting than a placebo, although no harmful cases of diarrhea or

AC

dyspepsia were observed. When extending the follow-up period to the maintenance

stage, aripiprazole continued to be associated with a higher incidence of diarrhea

and dry mouth than a placebo. Our findings are inconsistent with a previous study

(Brown et al., 2013), which suggested insignificant differences of rate of diarrhea

compared to placebo, and clinicians should carefully explain these side effects before

prescribing aripiprazole to patients with BD.

35
 ACCEPTED MANUSCRIPT

EPS is an undesirable adverse reaction to all antipsychotics, although atypical

antipsychotics may have a lower risk than typical antipsychotics (Pierre, 2005).

Aripiprazole was associated with a significantly higher incidence of EPS and

hyper-salivation in the acute phase than a placebo. However, there were no

significant differences in EPS among the patients receiving aripiprazole monotherapy

and other medications (haloperidol or lithium). Specifically, aripiprazole was better

PT
than haloperidol, but not different than lithium. No significance differences were

RI
noted in tremors between haloperidol or lithium. Hence, aripiprazole may be

SC
beneficial compared to other medications, which is consistent with the findings of a

previous meta-analysis (Brown et al., 2013) focusing on bipolar mania. Moreover,

NU
other undesirable symptoms including anxiety and akathisia should be addressed

due to the high incidence of akathisia in several trials (El Mallakh et al., 2010; Thase
MA

et al., 2008). Aripiprazole was associated with higher rates of akathisia and anxiety

than a placebo in the acute stage, and a higher rate of akathisia in the maintenance
ED

stage. However, compared to other medications (lithium or haloperidol), there was

no significant difference in akathisia (effect size: -0.130, p = 0.696). Therefore,

T
EP

aripiprazole may not be as hazardous as previously thought, and further clinical trials

are warranted to investigate this issue, especially for comparisons with

C

anticonvulsants or other second-generation antipsychotics.

AC

For other adverse reactions including, dizziness, agitation, headache, light

headedness, and upper respiratory tract infection, there were no significant

differences between aripiprazole and a placebo in the acute or maintenance stage.

However, aripiprazole was associated with significantly higher incidence rates of

fatigue, insomnia, over-sedation, and pain in the extremities compared to a placebo.

However, there was no significant difference in over-sedation between aripiprazole

36
 ACCEPTED MANUSCRIPT

and a placebo in the maintenance phase. There was also no significant difference in

the incidence of headache between aripiprazole and haloperidol or lithium. We

assessed adherence using dropout rates and rates of discontinuation due to side

effects. Compared with a placebo in the acute stage, no significance was noted for

aripiprazole in dropout rate, however there was a significantly lower dropout rate

with aripiprazole in the maintenance stage. In combination therapy, there was no

PT
significant difference between aripiprazole and a placebo in dropout rate. The results

RI
of recent meta-analyses (Brown et al., 2013; Meduri et al., 2016; Yildiz et al., 2015)

SC
are consistent with our findings with regards to drop-out rates. However, aripiprazole

was associated with a higher rate of discontinuation due to side effects than a
NU
placebo in acute and maintenance stages, which may have been due to the relatively

higher incidence of intolerable side effects such as akathisia and gastro-intestinal

MA

disturbances. However, there were no significant differences in discontinuation rates

due to side effects between aripiprazole and lithium or haloperidol, which is

ED

consistent with the study by Brown et al (Brown et al., 2013). Our results suggest that

aripiprazole may not be more hazardous than other medications

T
EP

A strength of the current study is that we focused our analysis on aripiprazole

C

treatment across mania to depression in BD. Our results show inconsistencies in the
AC

efficacy or adverse reactions of aripiprazole treatment, which is also a strength.

37
 ACCEPTED MANUSCRIPT

Limitations:

There are several limitations to this study. First, the lack of trials (fewer than

three) for subgroup analysis, including aripiprazole monotherapy versus other

medications in the maintenance stage, aripiprazole combination therapy versus

other medications in the acute and maintenance stages, and aripiprazole

PT
combination therapy versus a placebo in the acute stage, limits the strength of our

results. Second, due to language limitations, we may have missed articles published

RI
in other languages. Third, we cannot rule out the possible confounding effects of the

SC
concomitant use of other medications with combination therapy of aripiprazole and

a placebo/other medications. Fourth, due to the limited number of included studies,

NU
we could not undertake meaningful meta-regression analyses of specific comparison
MA

groups (maintenance stage, etc.) to investigate the influence of variables that may

have influenced the observed results and the heterogeneity of these findings. Finally,
ED

due to the absence of head to head studies considering the advantages or

disadvantages of aripiprazole over other second-generation antipsychotics, we could

T

not perform meta-analyses between them, which may be a concern in clinical

EP

practice.
C
AC

38
 ACCEPTED MANUSCRIPT

Conclusion:

In summary, the current meta-analysis adds to the current knowledge on

treating bipolar mania, especially with regards to secondary outcomes and safety

profiles. Our results showed that aripiprazole was beneficial in preventing bipolar

mania. Unfortunately, aripiprazole has a minimal effect on bipolar depression. In

PT
addition, aripiprazole has outstanding tolerability with regards to the metabolic

syndrome, EPS, and dropout rates. Furthermore, aripiprazole may be beneficial in

RI
clinical practice for akathisia compared to haloperidol or lithium, although it was

SC
more hazardous than a placebo. Caution should be taken with regards to

gastro-intestinal disturbances when prescribing aripiprazole. Future trials should

NU
carefully detail information on the incidence of adverse events and treatment
MA

efficacy compared to other medications, especially second-generation antipsychotics.

T ED
C EP
AC

39
 ACCEPTED MANUSCRIPT

Acknowledgments

There was no research funding for this work.

Conflict of Interest

PT
One of the first author, Ping-Tao, Tseng had ever received speaker fees by the

RI
Otsuka for the seminar of aripiprazole. Dr. Eduard Vieta has received grants and

served as consultant, advisor or speaker for the following entities: AB-Biotics, Almirall,

SC
Allergan, AstraZeneca, Bristol-Myers Squibb, Cephalon, Dainippon Sumitomo Pharma,
NU
Elan, Eli Lilly, Farmindustria, Ferrer, Forest Research Institute, Gedeon Richter,

GlaxoSmithKline, Janssen, Jazz, Johnson and Johnson, Lundbeck, Merck, Novartis,

MA

Organon, Otsuka, Pfizer, Roche, Sanofi-Aventis, Servier, Schering-Plough, Shire, the

Brain and Behavior Foundation, the Seventh European Framework Programme

ED

(ENBREC), the Stanley Medical Research Institute, Sunovion, Takeda, Teva, and

United BioSource Corporation. All other authors declare no biomedical conflicts of

T

interest.
EP

Part of current meta-analysis had been orally presented in PRCP 2016.

C
AC

40
 ACCEPTED MANUSCRIPT

Figure legends

Figure 1: Flowchart of the current meta-analysis

Figure 1 The databases used, the inclusion criteria, the search

strategy and selection process of the current meta-analysis

PT
RI
Figure 2 (A) Forest plot of the meta-analysis of differences in changes in

SC
depressive symptoms in patients receiving aripiprazole and a placebo

in acute bipolar depression; (B) forest plot of the meta-analysis of

NU
differences in changes in manic symptoms in patients receiving

aripiprazole and a placebo in acute bipolar mania; (C) forest plot of

MA

the meta-analysis of differences in changes in the general severity in

patients with bipolar disorder receiving aripiprazole and a placebo in

ED

the maintenance stage

T

Figure 2A There were no significant differences between changes in

EP

depressive symptoms between the patients receiving aripiprazole or

C

a placebo in acute bipolar depression (Hedges’ g=-0.127, 95%

AC

CI=-0.257 to 0.002, p=0.054); Figure 2B There was a significant

improvement in manic symptoms in the patients receiving

aripiprazole than those receiving a placebo in acute bipolar mania

(Hedges’ g=-0.299, 95% CI=-0.476 to -0.123, p=0.001). Figure 2C

There was a significant improvement in the general severity in

patients with bipolar disorder receiving aripiprazole than in those

receiving a placebo in the maintenance stage (Hedges’ g=-0.483, 95%

41
 ACCEPTED MANUSCRIPT

CI=-0.674 to -0.292, p<0.001).

PT
RI
SC
NU
MA
T ED
EP
C
AC

42
 ACCEPTED MANUSCRIPT

References

Angst, J., 2006. Do many patients with depression suffer from bipolar disorder?
Can J Psychiatry 51(1), 3-5.
Arbaizar, B., Dierssen-Sotos, T., Gomez-Acebo, I., Llorca, J., 2009. Aripiprazole in
major depression and mania: meta-analyses of randomized
placebo-controlled trials. Gen Hosp Psychiatry 31(5), 478-483.
Biederman, J., McDonnell, M.A., Wozniak, J., Spencer, T., Aleardi, M., Falzone, R.,

PT
Mick, E., 2005. Aripiprazole in the treatment of pediatric bipolar disorder: a
systematic chart review. CNS Spectr 10(2), 141-148.

RI
Borenstein, M., Hedges, L.V., Higgins, J.P., Rothstein, H.R., 2010. A basic
introduction to fixed-effect and random-effects models for meta-analysis.

SC
Res Synth Methods 1(2), 97-111.
Brown, R., Taylor, M.J., Geddes, J., 2013. Aripiprazole alone or in combination for
acute mania. Cochrane Database Syst Rev(12), CD005000.
NU
Burris, K.D., Molski, T.F., Xu, C., Ryan, E., Tottori, K., Kikuchi, T., Yocca, F.D.,
Molinoff, P.B., 2002. Aripiprazole, a novel antipsychotic, is a high-affinity
MA

partial agonist at human dopamine D2 receptors. J Pharmacol Exp Ther

302(1), 381-389.
Busner, J., Targum, S.D., 2007. The clinical global impressions scale: applying a
research tool in clinical practice. Psychiatry (Edgmont) 4(7), 28-37.
ED

Carlson, B.X., Ketter, T.A., Sun, W., Timko, K., McQuade, R.D., Sanchez, R.,
Vester-Blokland, E., Marcus, R., 2012. Aripiprazole in combination with
T

lamotrigine for the long-term treatment of patients with bipolar I disorder

EP

(manic or mixed): a randomized, multicenter, double-blind study

(CN138-392). Bipolar Disord 14(1), 41-53.
Chien, I.C., Chang, K.C., Lin, C.H., Chou, Y.J., Chou, P., 2010. Prevalence of
C

diabetes in patients with bipolar disorder in Taiwan: a population-based

AC

national health insurance study. Gen Hosp Psychiatry 32(6), 577-582.

Craddock, N., Sklar, P., 2013. Genetics of bipolar disorder. Lancet 381(9878),
1654-1662.
Davey, J., Turner, R.M., Clarke, M.J., Higgins, J.P., 2011. Characteristics of
meta-analyses and their component studies in the Cochrane Database of
Systematic Reviews: a cross-sectional, descriptive analysis. BMC Med Res
Methodol 11, 160.
Dhillon, S., 2012. Aripiprazole: a review of its use in the management of mania in
adults with bipolar I disorder. Drugs 72(1), 133-162.
Dundar, Y., Greenhalgh, J., Richardson, M., Dwan, K., 2016. Pharmacological
43
 ACCEPTED MANUSCRIPT

treatment of acute agitation associated with psychotic and bipolar disorder:

a systematic review and meta-analysis. Hum Psychopharmacol 31(4),
268-285.
Duval, S., Tweedie, R., 2000. Trim and fill: A simple funnel-plot-based method of
testing and adjusting for publication bias in meta-analysis. Biometrics 56(2),
455-463.
Egger, M., Davey Smith, G., Schneider, M., Minder, C., 1997. Bias in meta-analysis
detected by a simple, graphical test. BMJ 315(7109), 629-634.

PT
El-Mallakh, R.S., Marcus, R., Baudelet, C., McQuade, R., Carson, W.H., Owen, R.,
2012. A 40-week double-blind aripiprazole versus lithium follow-up of a
12-week acute phase study (total 52 weeks) in bipolar I disorder. J Affect

RI
Disord 136(3), 258-266.
El Mallakh, R.S., Vieta, E., Rollin, L., Marcus, R., Carson, W.H., McQuade, R., 2010.

SC
A comparison of two fixed doses of aripiprazole with placebo in acutely
relapsed, hospitalized patients with bipolar disorder I (manic or mixed) in
NU
subpopulations (CN138-007). Eur Neuropsychopharmacol 20(11), 776-783.
Findling, R.L., Correll, C.U., Nyilas, M., Forbes, R.A., McQuade, R.D., Jin, N.,
Ivanova, S., Mankoski, R., Carson, W.H., Carlson, G.A., 2013. Aripiprazole for
MA

the treatment of pediatric bipolar I disorder: a 30-week, randomized,

placebo-controlled study. Bipolar Disord 15(2), 138-149.
Findling, R.L., Nyilas, M., Forbes, R.A., McQuade, R.D., Jin, N., Iwamoto, T.,
ED

Ivanova, S., Carson, W.H., Chang, K., 2009. Acute treatment of pediatric
bipolar I disorder, manic or mixed episode, with aripiprazole: a randomized,
T

double-blind, placebo-controlled study. J Clin Psychiatry 70(10), 1441-1451.

Findling, R.L., Youngstrom, E.A., McNamara, N.K., Stansbrey, R.J., Wynbrandt, J.L.,
EP

Adegbite, C., Rowles, B.M., Demeter, C.A., Frazier, T.W., Calabrese, J.R., 2012.
Double-blind, randomized, placebo-controlled long-term maintenance study
C

of aripiprazole in children with bipolar disorder. J Clin Psychiatry 73(1), 57-63.

AC

Fountoulakis, K.N., Gonda, X., Vieta, E., Schmidt, F., 2009. Treatment of psychotic
symptoms in bipolar disorder with aripiprazole monotherapy: a
meta-analysis. Ann Gen Psychiatry 8, 27.
Fountoulakis, K.N., Vieta, E., Schmidt, F., 2011. Aripiprazole monotherapy in the
treatment of bipolar disorder: a meta-analysis. J Affect Disord 133(3),
361-370.
Gomaraschi, M., Calabresi, L., Franceschini, G., 2016. Protective Effects of HDL
Against Ischemia/Reperfusion Injury. Front Pharmacol 7, 2.
Goodwin, G.M., Anderson, I., Arango, C., Bowden, C.L., Henry, C., Mitchell, P.B.,
Nolen, W.A., Vieta, E., Wittchen, H.U., 2008. ECNP consensus meeting.

44
 ACCEPTED MANUSCRIPT

Bipolar depression. Nice, March 2007. Eur Neuropsychopharmacol 18(7),

535-549.
Goodwin, G.M., Haddad, P.M., Ferrier, I.N., Aronson, J.K., Barnes, T., Cipriani, A.,
Coghill, D.R., Fazel, S., Geddes, J.R., Grunze, H., Holmes, E.A., Howes, O.,
Hudson, S., Hunt, N., Jones, I., Macmillan, I.C., McAllister-Williams, H.,
Miklowitz, D.R., Morriss, R., Munafo, M., Paton, C., Saharkian, B.J., Saunders,
K., Sinclair, J., Taylor, D., Vieta, E., Young, A.H., 2016. Evidence-based
guidelines for treating bipolar disorder: Revised third edition

PT
recommendations from the British Association for Psychopharmacology. J
Psychopharmacol 30(6), 495-553.
Higgins, J.P., Thompson, S.G., Deeks, J.J., Altman, D.G., 2003. Measuring

RI
inconsistency in meta-analyses. BMJ 327(7414), 557-560.
Hirschfeld, R.M., Lewis, L., Vornik, L.A., 2003. Perceptions and impact of bipolar

SC
disorder: how far have we really come? Results of the national depressive
and manic-depressive association 2000 survey of individuals with bipolar
NU
disorder. J Clin Psychiatry 64(2), 161-174.
Jadad, A.R., Moore, R.A., Carroll, D., Jenkinson, C., Reynolds, D.J., Gavaghan, D.J.,
McQuay, H.J., 1996. Assessing the quality of reports of randomized clinical
MA

trials: is blinding necessary? Control Clin Trials 17(1), 1-12.

Jeong, H.G., Lee, M.S., Ko, Y.H., Han, C., Jung, I.K., 2012. Combination treatment
with aripiprazole and valproic acid for acute mania: an 8-week, single-blind,
ED

randomized controlled trial. Clin Neuropharmacol 35(3), 97-102.

Jordan, S., Koprivica, V., Chen, R., Tottori, K., Kikuchi, T., Altar, C.A., 2002. The
T

antipsychotic aripiprazole is a potent, partial agonist at the human 5-HT1A

receptor. Eur J Pharmacol 441(3), 137-140.
EP

Judd, L.L., Akiskal, H.S., 2003. The prevalence and disability of bipolar spectrum
disorders in the US population: re-analysis of the ECA database taking into
C

account subthreshold cases. J Affect Disord 73(1-2), 123-131.

AC

Kanba, S., Kawasaki, H., Ishigooka, J., Sakamoto, K., Kinoshita, T., Kuroki, T., 2014.
A placebo-controlled, double-blind study of the efficacy and safety of
aripiprazole for the treatment of acute manic or mixed episodes in Asian
patients with bipolar I disorder (the AMAZE study). World J Biol Psychiatry
15(2), 113-121.
Kattimani, S., Subramanian, K., Sarkar, S., Rajkumar, R.P., Balasubramanian, S.,
2016. History of Lifetime suicide attempt in bipolar I disorder: its correlates
and effect on illness course. Int J Psychiatry Clin Pract, 1-7.
Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome
scale (PANSS) for schizophrenia. Schizophr Bull 13(2), 261-276.

45
 ACCEPTED MANUSCRIPT

Keck, P.E., Jr., Calabrese, J.R., McIntyre, R.S., McQuade, R.D., Carson, W.H.,
Eudicone, J.M., Carlson, B.X., Marcus, R.N., Sanchez, R., Aripiprazole Study, G.,
2007. Aripiprazole monotherapy for maintenance therapy in bipolar I
disorder: a 100-week, double-blind study versus placebo. J Clin Psychiatry
68(10), 1480-1491.
Keck, P.E., Jr., Marcus, R., Tourkodimitris, S., Ali, M., Liebeskind, A., Saha, A.,
Ingenito, G., Aripiprazole Study, G., 2003. A placebo-controlled, double-blind
study of the efficacy and safety of aripiprazole in patients with acute bipolar

PT
mania. Am J Psychiatry 160(9), 1651-1658.
Keck, P.E., Orsulak, P.J., Cutler, A.J., Sanchez, R., Torbeyns, A., Marcus, R.N.,
McQuade, R.D., Carson, W.H., Group, C.N.S., 2009. Aripiprazole monotherapy

RI
in the treatment of acute bipolar I mania: a randomized, double-blind,
placebo- and lithium-controlled study. J Affect Disord 112(1-3), 36-49.

SC
Li, Z., Ichikawa, J., Dai, J., Meltzer, H.Y., 2004. Aripiprazole, a novel antipsychotic
drug, preferentially increases dopamine release in the prefrontal cortex and
NU
hippocampus in rat brain. Eur J Pharmacol 493(1-3), 75-83.
Liberati, A., Altman, D.G., Tetzlaff, J., Mulrow, C., Gotzsche, P.C., Ioannidis, J.P.,
Clarke, M., Devereaux, P.J., Kleijnen, J., Moher, D., 2009. The PRISMA
MA

statement for reporting systematic reviews and meta-analyses of studies that

evaluate health care interventions: explanation and elaboration. PLoS Med
6(7), e1000100.
ED

Lindstrom, L., Lindstrom, E., Nilsson, M., Hoistad, M., 2017. Maintenance
therapy with second generation antipsychotics for bipolar disorder - A
T

systematic review and meta-analysis. J Affect Disord 213, 138-150.

Marcus, R., Khan, A., Rollin, L., Morris, B., Timko, K., Carson, W., Sanchez, R.,
EP

2011. Efficacy of aripiprazole adjunctive to lithium or valproate in the

long-term treatment of patients with bipolar I disorder with an inadequate
C

response to lithium or valproate monotherapy: a multicenter, double-blind,

AC

randomized study. Bipolar Disord 13(2), 133-144.

McGrady, A., Lynch, D., Rapport, D., 2017. Psychosocial Factors and Comorbidity
Associated with Suicide Attempts: Findings in Patients with Bipolar Disorder.
Psychopathology.
McKeage, K., 2014. Aripiprazole: a review of its use in the treatment of manic
episodes in adolescents with bipolar I disorder. CNS Drugs 28(2), 171-183.
Meduri, M., Gregoraci, G., Baglivo, V., Balestrieri, M., Isola, M., Brambilla, P.,
2016. A meta-analysis of efficacy and safety of aripiprazole in adult and
pediatric bipolar disorder in randomized controlled trials and observational
studies. J Affect Disord 191, 187-208.

46
 ACCEPTED MANUSCRIPT

Merikangas, K.R., Jin, R., He, J.P., Kessler, R.C., Lee, S., Sampson, N.A., Viana, M.C.,
Andrade, L.H., Hu, C., Karam, E.G., Ladea, M., Medina-Mora, M.E., Ono, Y.,
Posada-Villa, J., Sagar, R., Wells, J.E., Zarkov, Z., 2011. Prevalence and
correlates of bipolar spectrum disorder in the world mental health survey
initiative. Arch Gen Psychiatry 68(3), 241-251.
Miura, T., Noma, H., Furukawa, T.A., Mitsuyasu, H., Tanaka, S., Stockton, S.,
Salanti, G., Motomura, K., Shimano-Katsuki, S., Leucht, S., Cipriani, A.,
Geddes, J.R., Kanba, S., 2014. Comparative efficacy and tolerability of

PT
pharmacological treatments in the maintenance treatment of bipolar
disorder: a systematic review and network meta-analysis. Lancet Psychiatry
1(5), 351-359.

RI
Montgomery, S.A., Asberg, M., 1979. A new depression scale designed to be
sensitive to change. Br J Psychiatry 134, 382-389.

SC
Muzina, D.J., Momah, C., Eudicone, J.M., Pikalov, A., McQuade, R.D., Marcus,
R.N., Sanchez, R., Carlson, B.X., 2008. Aripiprazole monotherapy in patients
NU
with rapid-cycling bipolar I disorder: an analysis from a long-term,
double-blind, placebo-controlled study. Int J Clin Pract 62(5), 679-687.
Nilsson, K.K., Jorgensen, C.R., Craig, T.K., Straarup, K.N., Licht, R.W., 2010.
MA

Self-esteem in remitted bipolar disorder patients: a meta-analysis. Bipolar

Disord 12(6), 585-592.
Oud, M., Mayo-Wilson, E., Braidwood, R., Schulte, P., Jones, S.H., Morriss, R.,
ED

Kupka, R., Cuijpers, P., Kendall, T., 2016. Psychological interventions for adults
with bipolar disorder: systematic review and meta-analysis. Br J Psychiatry
T

208(3), 213-222.
Oya, K., Kishi, T., Iwata, N., 2015. Efficacy and tolerability of aripiprazole once
EP

monthly for schizophrenia: a systematic review and meta-analysis of

randomized controlled trials. Neuropsychiatr Dis Treat 11, 2299-2307.
C

Perlis, R.H., 2007. Treatment of bipolar disorder: the evolving role of atypical
AC

antipsychotics. Am J Manag Care 13(7 Suppl), S178-188.

Perlis, R.H., Welge, J.A., Vornik, L.A., Hirschfeld, R.M., Keck, P.E., Jr., 2006.
Atypical antipsychotics in the treatment of mania: a meta-analysis of
randomized, placebo-controlled trials. J Clin Psychiatry 67(4), 509-516.
Pierre, J.M., 2005. Extrapyramidal symptoms with atypical antipsychotics :
incidence, prevention and management. Drug Saf 28(3), 191-208.
Quante, A., Zeugmann, S., Luborzewski, A., Schommer, N., Langosch, J., Born, C.,
Anghelescu, I., Wolf, J., 2010. Aripiprazole as adjunct to a mood s tabilizer and
citalopram in bipolar depression: a randomized placebo-controlled pilot
study. Hum Psychopharmacol 25(2), 126-132.

47
 ACCEPTED MANUSCRIPT

Reed, C., Goetz, I., Vieta, E., Bassi, M., Haro, J.M., Board, E.A., 2010. Work
impairment in bipolar disorder patients--results from a two-year
observational study (EMBLEM). Eur Psychiatry 25(6), 338-344.
Rosenthal, N.E., Rosenthal, L.N., Stallone, F., Fleiss, J., Dunner, D.L., Fieve, R.R.,
1979. Psychosis as a predictor of response to lithium maintenance treatment
in bipolar affective disorder. J Affect Disord 1(4), 237-245.
Sachs, G., Sanchez, R., Marcus, R., Stock, E., McQuade, R., Carson, W.,
Abou-Gharbia, N., Impellizzeri, C., Kaplita, S., Rollin, L., Iwamoto, T.,

PT
Aripiprazole Study, G., 2006. Aripiprazole in the treatment of acute manic or
mixed episodes in patients with bipolar I disorder: a 3-week
placebo-controlled study. J Psychopharmacol 20(4), 536-546.

RI
Schatzberg, A.F., Nemeroff, C.B., 2009. The American Psychiatric Publishing
textbook of psychopharmacology, 4th ed. American Psychiatric Pub.,

SC
Washington, D.C.
Scherk, H., Pajonk, F.G., Leucht, S., 2007. Second-generation antipsychotic agents
NU
in the treatment of acute mania: a systematic review and meta-analysis of
randomized controlled trials. Arch Gen Psychiatry 64(4), 442-455.
Smith, L.A., Cornelius, V., Warnock, A., Tacchi, M.J., Taylor, D., 2007.
MA

Pharmacological interventions for acute bipolar mania: a systematic review

of randomized placebo-controlled trials. Bipolar Disord 9(6), 551-560.
Taylor, D.M., 2003. Aripiprazole: a review of its pharmacology and clinical use. Int
ED

J Clin Pract 57(1), 49-54.

Thase, M.E., Jonas, A., Khan, A., Bowden, C.L., Wu, X., McQuade, R.D., Carson,
T

W.H., Marcus, R.N., Owen, R., 2008. Aripiprazole monotherapy in

nonpsychotic bipolar I depression: results of 2 randomized,
EP

placebo-controlled studies. J Clin Psychopharmacol 28(1), 13-20.

Tohen, M., Waternaux, C.M., Tsuang, M.T., 1990. Outcome in Mania. A 4-year
C

prospective follow-up of 75 patients utilizing survival analysis. Arch Gen

AC

Psychiatry 47(12), 1106-1111.

Tohen, M., Zarate, C.A., Jr., Hennen, J., Khalsa, H.M., Strakowski, S.M.,
Gebre-Medhin, P., Salvatore, P., Baldessarini, R.J., 2003. The McLean-Harvard
First-Episode Mania Study: prediction of recovery and first recurrence. Am J
Psychiatry 160(12), 2099-2107.
Tramontina, S., Zeni, C.P., Ketzer, C.R., Pheula, G.F., Narvaez, J., Rohde, L.A., 2009.
Aripiprazole in children and adolescents with bipolar disorder comorbid with
attention-deficit/hyperactivity disorder: a pilot randomized clinical trial. J Clin
Psychiatry 70(5), 756-764.
van der Loos, M.L., Mulder, P., Hartong, E.G., Blom, M.B., Vergouwen, A.C., van

48
 ACCEPTED MANUSCRIPT

Noorden, M.S., Timmermans, M.A., Vieta, E., Nolen, W.A., LamLit Study, G.,
2011. Long-term outcome of bipolar depressed patients receiving
lamotrigine as add-on to lithium with the possibility of the addition of
paroxetine in nonresponders: a randomized, placebo-controlled trial with a
novel design. Bipolar Disord 13(1), 111-117.
Vancampfort, D., Correll, C.U., Galling, B., Probst, M., De Hert, M., Ward, P.B.,
Rosenbaum, S., Gaughran, F., Lally, J., Stubbs, B., 2016. Diabetes mellitus in
people with schizophrenia, bipolar disorder and major depressive disorder: a

PT
systematic review and large scale meta-analysis. World Psychiatry 15(2),
166-174.
Vancampfort, D., Stubbs, B., Mitchell, A.J., De Hert, M., Wampers, M., Ward, P.B.,

RI
Rosenbaum, S., Correll, C.U., 2015. Risk of metabolic syndrome and its
components in people with schizophrenia and related psychotic disorders,

SC
bipolar disorder and major depressive disorder: a systematic review and
meta-analysis. World Psychiatry 14(3), 339-347.
NU
Vieta, E., Bourin, M., Sanchez, R., Marcus, R., Stock, E., McQuade, R., Carson, W.,
Abou-Gharbia, N., Swanink, R., Iwamoto, T., Aripoprazole Study, G., 2005.
Effectiveness of aripiprazole v. haloperidol in acute bipolar mania:
MA

double-blind, randomised, comparative 12-week trial. Br J Psychiatry 187,

235-242.
Vieta, E., Locklear, J., Gunther, O., Ekman, M., Miltenburger, C., Chatterton, M.L.,
ED

Astrom, M., Paulsson, B., 2010. Treatment options for bipolar depression: a
systematic review of randomized, controlled trials. J Clin Psychopharmacol
T

30(5), 579-590.
Vieta, E., T'Joen, C., McQuade, R.D., Carson, W.H., Jr., Marcus, R.N., Sanchez, R.,
EP

Owen, R., Nameche, L., 2008. Efficacy of adjunctive aripiprazole to either

valproate or lithium in bipolar mania patients partially nonresponsive to
C

valproate/lithium monotherapy: a placebo-controlled study. Am J Psychiatry

AC

165(10), 1316-1325.
von Eckardstein, A., Widmann, C., 2014. High-density lipoprotein, beta cells, and
diabetes. Cardiovasc Res 103(3), 384-394.
Williams, J.B., 1988. A structured interview guide for the Hamilton Depression
Rating Scale. Arch Gen Psychiatry 45(8), 742-747.
Woo, Y.S., Bahk, W.M., Chung, M.Y., Kim, D.H., Yoon, B.H., Lee, J.H., Ahn, Y.M.,
Chung, S.K., Kim, J.G., Lee, K.H., Paik, K.C., 2011. Aripiprazole plus divalproex
for recently manic or mixed patients with bipolar I disorder: a 6-month,
randomized, placebo-controlled, double-blind maintenance trial. Hum
Psychopharmacol 26(8), 543-553.

49
 ACCEPTED MANUSCRIPT

Yatham, L.N., Kennedy, S.H., Parikh, S.V., Schaffer, A., Beaulieu, S., Alda, M.,
O'Donovan, C., Macqueen, G., McIntyre, R.S., Sharma, V., Ravindran, A.,
Young, L.T., Milev, R., Bond, D.J., Frey, B.N., Goldstein, B.I., Lafer, B., Birmaher,
B., Ha, K., Nolen, W.A., Berk, M., 2013. Canadian Network for Mood and
Anxiety Treatments (CANMAT) and International Society for Bipolar Disorders
(ISBD) collaborative update of CANMAT guidelines for the management of
patients with bipolar disorder: update 2013. Bipolar Disord 15(1), 1-44.
Yatham, L.N., Lecrubier, Y., Fieve, R.R., Davis, K.H., Harris, S.D., Krishnan, A.A.,

PT
2004. Quality of life in patients with bipolar I depression: data from 920
patients. Bipolar Disord 6(5), 379-385.
Yildiz, A., Nikodem, M., Vieta, E., Correll, C.U., Baldessarini, R.J., 2015. A network

RI
meta-analysis on comparative efficacy and all-cause discontinuation of
antimanic treatments in acute bipolar mania. Psychol Med 45(2), 299-317.

SC
Yildiz, A., Vieta, E., Leucht, S., Baldessarini, R.J., 2011. Efficacy of antimanic
treatments: meta-analysis of randomized, controlled trials.
NU
Neuropsychopharmacology 36(2), 375-389.
Young, A.H., Oren, D.A., Lowy, A., McQuade, R.D., Marcus, R.N., Carson, W.H.,
Spiller, N.H., Torbeyns, A.F., Sanchez, R., 2009. Aripiprazole monotherapy in
MA

acute mania: 12-week randomised placebo- and haloperidol-controlled study.

Br J Psychiatry 194(1), 40-48.
Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale for mania:
ED

reliability, validity and sensitivity. Br J Psychiatry 133, 429-435.

T
C EP
AC

50
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
T ED
EP
C
AC

51
 ACCEPTED MANUSCRIPT

PT
RI
SC
NU
MA
T ED
EP
C
AC

52
 ACCEPTED MANUSCRIPT

Table 1: Summary of characteristics of studies in current

meta-analysis

Gender
Criter Diagnosis/m Dropo Ag Countr
Study Comparison N (%fema Phase
ia ood state ut (%) e y
le)

PT
13 40.
BD-I,
Keck, P.E. DSM- Aripiprazole 0 58.0 55.0 5
manic/mixe Acute USA
(2003) IV Placebo 13 79.0 58.0 40.

RI
d
2 5

SC
17 42. Multip
BD-I,
Vieta, E. DSM- Aripiprazole 5 49.4 56.6 6 le
manic/mixe Acute
(2005) IV Haloperidol 17 71.8 66.9 41. countr
d
NU
2 0 y
13 37.
BD-I,
Sachs, G. DSM- Aripiprazole 7 45.0 50.0 3
MA

manic/mixe Acute USA

(2006) IV Placebo 13 48.0 53.0 40.
d
5 4
39. Multip
ED

Keck, P.E. DSM- Aripiprazole 77 91.0 62.0 0 Maintena le

BD-I
(2007) IV Placebo 81 92.8 72.0 40. nce countr
T

3 y
EP

Thase, 18 39.
M.E. DSM- BD-I, Aripiprazole 6 46.8 62.0 0
Acute USA
(2008)-st IV depressed Placebo 18 35.1 63.0 39.
C

udy A 8 0
AC

Thase, 18 41.
M.E. DSM- BD-I, Aripiprazole 7 41.2 60.0 0
Acute USA
(2008)-st IV depressed Placebo 18 29.8 60.0 40.
udy B 8 0
37.
Muzina,
DSM- BD-I, rapid Aripiprazole 14 79.0 64.3 6 Maintena
D.J. USA
IV cycling Placebo 14 100.0 71.4 38. nce
(2008)
8
Vieta, E. DSM- BD-I, Aripiprazole 25 52.0 42. Multip
n/a Acute
(2008) IV manic/mixe + Li or Val 3 58.0 2 le

53
 ACCEPTED MANUSCRIPT

d Placebo + Li 13 41. countr

or Val 1 7 y
15 39.
5 6
BD-I, Aripiprazole 20.0 49.0
Keck, P.E. DSM- 16 39.
manic/mixe Li 15.0 48.0 Acute USA
(2009) IV 0 6
d Placebo 19.0 48.0
16 39.
5 8

PT
13.
Aripiprazole
7
Findling, BD-I, 10 mg 98 14.3 46.9
DSM- 13.

RI
R.L. manic/mixe Aripiprazole 99 22.2 48.5 Acute USA
IV 3
(2009) d 30 mg 99 23.2 43.4
13.

SC
placebo
3
11.
Tramonti BD-I,
NU
DSM- Aripiprazole 18 5.6 66.7 7
na, S. manic/mixe Acute Brazil
IV Placebo 25 4.0 44.0 12.
(2009) d
2
MA

16 40.
7 5 Multip
Young, BD-I, Aripiprazole 43.1 57.0
DSM- 16 41. le
ED

A.H. manic/mixe haloperidol 42.4 56.0 Acute

IV 5 6 countr
(2009) d Placebo 45.1 54.0
15 40. y
T

3 2
Aripiprazole
EP

+ (Li or Val)
53.
Quante, +
C

DSM- BD-I or II, 12 45.4 6 Germa

A. Citalopram n/a Acute
IV depressed 11 50.0 48. ny
AC

(2010) Placebo + (Li

2
or Val) +
Citalopram
13 39.
Aripiprazole
El 1 3 Multip
BD-I, 15 mg 57.0 50.0
Mallakh, DSM- 13 41. le
manic/mixe Aripiprazole 60.0 54.0 Acute
R.S. IV 6 4 countr
d 30 mg 60.0 51.0
(2010) 13 40. y
Placebo
4 6
Woo, Y.S. DSM- BD-I, Aripiprazole 40 42.5 67.5 38. Maintena Korea

54
 ACCEPTED MANUSCRIPT

(2011) IV manic/mixe + Val 43 41.9 55.8 3 nce

d Placebo + 38.
Val 3
El 37.
BD-I,
Mallakh, DSM- Aripiprazole 25 75.0 60.0 2 Maintena
manic/mixe USA
R.S. IV Li 38 65.8 50.0 41. nce
d
(2012) 2
Aripiprazole
17 39.

PT
Carlson, BD-I, +
DSM- 8 63.5 67.4 6 Maintena
B.X. manic/mixe Lamotrigine USA
IV 17 69.4 62.4 38. nce
(2012) d Placebo +

RI
3 6
Lamotrigine
Aripiprazole 37.

SC
Jeong, BD-I,
DSM- + Val 28 7.1 67.9 2
H.G. manic/mixe Acute Korea
IV Haloperidol 14 7.1 57.1 36.
(2012) d
NU
+ Val 1
Aripiprazole
+ other
MA

psychostimu
Findling,
DSM- lant 30 80.0 37.0 7.1 Maintena
R.L. BD-I or II USA
IV Placebo + 30 100.0 23.0 6.7 nce
ED

(2012)
other
psychostimu
T

lant
13.
EP

Aripiprazole
6
Findling, BD-I, 10 mg 75 54.7
DSM- 13. Maintena
C

R.L. manic/mixe Aripiprazole 71 69.0 n/a USA

IV 1 nce
(2013) d 30 mg 64 81.3
AC

13.
Placebo
3
12 37. Multip
BD-I,
Kanba, S. DSM- Aripiprazole 8 43.8 57.4 5 le
manic/mixe Acute
(2014) IV Placebo 13 50.8 60.0 37. countr
d
0 8 y

Abbreviation: n/a: not available; Bipolar I disorder: BD-I; Bipolar II disorder: BD-II; DSM-IV:
Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; DSM-IV: Diagnostic and Statistical
th
Manual of Mental Disorders, 4 Edition; n/a: not applicable; Li: lithium; Val: valproate

55
 ACCEPTED MANUSCRIPT

Table 2: Summary of comparison between current study and

two recently meta-analysis on aripiprazole

Meduri, M. et al.
Articles Current Brown, R. et al. 2013
2016
No. of included
20 10 22

PT
articles
Open labeled
Not included Not included 5
trails

RI
RCCTs 2 2 2

SC
RPCTs 20 9 15
th th th
End of search January 23 ,2017 July 31 ,2013 June 30 ,2015
manic/mixed,
Mood states manic/mixed, depressed
NU
manic/mixed
depressed
MA of acute phase Included Included Included
MA of maintenance
MA

Included n/a n/a

phase
MA of efficacy (result)
2
NS (Arp vs. Pla, mono-tx in NS (Arp vs. Pla,
ED

changes of
AP) mono-tx)
severity 2
n/a
NS (Arp vs. Hal or Li, NS (Arp vs. Hal or Li,
(depression)
T

mono-tx in AP) mono-tx)

EP

Favor Arp (Arp vs. Pla, Favor Arp (Arp vs. Pla, Favor Arp (Arp vs.
changes of
mono-tx in AP) mono-tx) Pla, mono-tx)
severity
n/a (Arp vs. Hal or Li, NS (Arp vs. Hal or Li, NS (Arp vs. Hal or Li,
C

(mania)
mono-tx) mono-tx) mono-tx)
AC

changes of Favor Arp (Arp vs. Pla,

severity Favor Arp (Arp vs. Pla, mono-tx )
2
n/a
(psychotic mono-tx in AP) NS (Arp vs. Hal or Li,
features) mono-tx)
Favor Arp in mania
Favor Arp in mania (Arp vs. Favor Arp in mania
(Arp vs. pla,
Pla, mono-tx in AP) (Arp vs. Pla, mono-tx )
response rates mono-tx)
NS (Arp vs. Hal or Li, NS in mania (Arp vs.
NS in mania (Arp vs.
mono-tx) Hal or Li, mono-tx)
Hal or Li, mono-tx)
remission rates Favor Arp in mania (Arp vs. NS in mania (Arp vs. n/a

56
 ACCEPTED MANUSCRIPT

Pla, mono-tx in AP) Pla, mono-tx )

Favor Arp (Arp vs. Pla,
relapse rates n/a n/a
com-tx in MP)
Favor Arp in mania
(Arp vs. Pla, mono-tx )
Favor Arp (Arp vs. Pla, NS in depression (Arp
mono-tx in AP) vs. Pla, mono-tx )
1
Favor Arp (Arp vs. Pla, Favor Arp overall (Arp
CGI-S n/a

PT
mono-tx in MP) vs. Pla, mono-tx )
2
Favor Arp (Arp vs. Hal or Li, NS in mania (Arp vs.
mono-tx in AP) Hal or Li, mono-tx )

RI
2
NS in depression (Arp

SC
vs. Hal or Li, mono-tx )
MA of adverse drug reaction
Increased appetite, Sugar
NU
Metabolic profile AC, BMI, BW change, level weight gain n/a
of HDL, level of TG
Adherenc es
MA

dropout rates, Dis SE dropout rates, Dis SE n/a

profile
constipation, diarrhea, nausea, dyspepsia,
Gastro-intestinal
nausea vomiting, dyspepsia, vomiting, constipation, n/a
ED

profile
dry mouth diarrhea
EPS EPS, salivation, tremor EPS, tremor n/a
T

agitation, akathisia, anxiety, hypertension

EP

dizziness over-sedation, akathisia, headache,

light-headedness, UTI, anxiety, insomnia,
Other ADRs n/a
insomnia, headache, fatigue light-headedness, pain
C

dyspnea, pain extremities, extremities,

AC

back pain over-sedation

1
: mania plus depression
2
: less than three trials

Abbreviation: n/a: not available; MA: meta-analysis; ADR: adverse reaction;

randomized placebo-controlled trials; RPCTs: randomized placebo-controlled trials;
RCCTs: randomized case-controlled trials; CGI-S: The Clinical Global Impression –
Severity scale; EPS: extrapyramidal symptoms; Sugar AC: blood sugar before meals;
BMI: body mass index; BW: body weight; HDL; high density lipoprotein; TG: total

57
 ACCEPTED MANUSCRIPT

cholesterol, triglyceride; Dis SE: rates of discontinuation of trials due to side effect;
UTI: upper respiratory tract infection; vs.: versus; Arp: aripiprazole; Hal: haloperidol;
Li: lithium; Pla: placebo; mono-tx: monotherapy; com-tx: combination therapy; AP:
acute phase; MP: maintenance phase; NS: non-significance

PT
RI
SC
NU
MA
T ED
C EP
AC

58
 ACCEPTED MANUSCRIPT

Table 3: Summary of comparison of adverse reactions in

current meta-analysis

Maintenance
Acute phase Acute/Maintenance phase
phase
Adverse Reaction ARP ARP ARP ARP combination

PT
monotherapy monotherapy combination therapy/monotherapy vs
vs PLA vs PLA therapy vs PLA other Rx
Metabolic profile

RI
Worse by
fasting glucose N.S. N.S. n/a

SC
PLA
BMI N.S. n/a n/a n/a
BW N.S. N.S. n/a n/a
NU
HDL N.S. Worse by ARP n/a n/a
Worse by
T-chol N.S. Worse by ARP n/a
PLA
MA

TG N.S. N.S. N.S. n/a

Increased Worse by
n/a n/a n/a
appetite PLA
ED

Adherences profile n/a

dropout rate N.S. Worse by PLA N.S. N.S.
T

Worse by
dis-side-effect Worse by ARP n/a N.S.
EP

ARP
Gastro-intestinal
n/a
profile
C

Worse by
constipation n/a n/a n/a
AC

ARP
diarrhea N.S. Worse by ARP n/a n/a
Worse by
nausea n/a n/a n/a
ARP
Worse by
vomiting n/a n/a n/a
ARP
dyspepsia N.S. n/a n/a n/a
dry mouth n/a Worse by ARP n/a n/a
Other profile n/a
agitation N.S. n/a n/a n/a

59
 ACCEPTED MANUSCRIPT

Worse by
akathisia Worse by ARP n/a N.S.
ARP
Worse by
anxiety n/a n/a n/a
ARP
dizziness N.S. n/a n/a n/a
Worse by
over-sedation n/a n/a n/a
ARP
light-headedness N.S. n/a n/a n/a

PT
headache N.S. N.S. N.S. N.S.
Worse by
fatigue n/a n/a n/a
ARP

RI
Worse by
insomnia n/a n/a n/a

SC
ARP
Worse by
pain extremities n/a n/a n/a
ARP
NU
Worse by
EPS n/a n/a N.S.
ARP
Worse by
MA

sialorrhea n/a n/a n/a

ARP
tremor n/a n/a n/a N.S.
URI n/a N.S. n/a n/a
ED

Abbreviation: n/a: not available; BMI: body mass index; BW: body weight; HDL: high density
T

lipoprotein; T-chol: total cholesterol; TG: triglyceride; ARP: Aripiprazole; PLA: placebo; NS:
EP

Non-significant difference; dis-side-effect Rates of discontinuation due to side effect; EPS:

extrapyramidal symptoms; URI: upper respiratory-tract infection; Rx: medication
C
AC

60