Journal of Affective Disorders 269 (2020) 154–184

Contents lists available at ScienceDirect

Journal of Affective Disorders

journal homepage: www.elsevier.com/locate/jad

Review article

Comparative efficacy and tolerability of pharmacological treatments for the T

treatment of acute bipolar depression: A systematic review and network
meta-analysis
Anees Bahjia,b, , Dylan Ermacorab, Callum Stephensonc, Emily R. Hawkena,d, Gustavo Vazqueza
⁎

a
Department of Psychiatry, Queen's University, Kingston, Ontario, Canada
b
Department of Public Health Sciences, Queen's University, Kingston, Ontario, Canada
c
School of Kinesiology and Health Studies, Queen's University, Kingston, Ontario, Canada
d
Providence Care Hospital, Kingston, Ontario, Canada

ARTICLE INFO ABSTRACT

Keywords: Objective: We investigated the comparative efficacy and tolerability of pharmacological treatment strategies for
Bipolar disorder the treatment of acute bipolar depression.
Pharmacotherapies Data sources: A systematic review and network meta-analysis was conducted by searching eight registries for
Meta-analysis published and unpublished, double-blind, randomized controlled trials of pharmacotherapies for the acute
Depression
treatment of bipolar depression.
Review
Data extraction and synthesis: PRISMA guidelines were used for abstracting data, while the Cochrane Risk of Bias
Comparative effectiveness
Tool was used to assess data quality. Data extraction was done independently by two reviewers, with dis-
crepancies resolved by consensus. Data were pooled using a random-effects model.
Main outcomes and measures: Primary outcomes were efficacy (response and remission rate) and acceptability
(completion of treatment and dropouts due to adverse events). Summary odds ratios (ORs) were estimated using
pairwise and network meta-analysis with random effects.
Results: Identified citations (4,404) included 50 trials comprising 11,448 participants. Escitalopram, phenelzine,
moclobemide, carbamazepine, sertraline, lithium, paroxetine, aripiprazole, gabapentin and ziprasidone appear
to be ineffective as compared to placebo in treatment of bipolar depression. Divalproex, olanzapine/fluoxetine,
olanzapine, quetiapine, cariprazine, and lamotrigine, appear to be effective as compared to placebo in treatment
of bipolar depression according to the network meta-analysis. Aripiprazole showed higher discontinuation rates
versus placebo due to the appearance of any adverse event. Quetiapine was better than placebo at reducing
treatment-emergent affective switches. For Bipolar I Disorder, cariprazine, fluoxetine, imipramine, lamotrigine,
lurasidone, olanzapine-fluoxetine, and olanzapine were significantly better than placebo at response, while
fluoxetine, imipramine, cariprazine, lurasidone, olanzapine-fluoxetine, and olanzapine were significantly better
than placebo at remission.
Conclusions and relevance: These results could serve evidence-based practice and inform patients, physicians,
guideline developers, and policymakers on the relative benefits of the different antidepressants, antipsychotics,
and mood-stabilizing agents for the treatment of bipolar depression.
Registration: PROSPERO (CRD42019122172).

1. Introduction irrespective of ethnic and national origins as well as socioeconomic

Bipolar disorder (BD) is a complex disorder characterized by re-
current episodes of depression and mania (bipolar I disorder) or hy-
pomania (bipolar II disorder) (American Psychiatric Association 2013).
Bipolar disorder affects more than 1% of the world's population,
status (Grande et al., 2016). The most recent global estimates of the
lifetime prevalence rates of bipolar I disorder (BD-I), bipolar II disorder
(BD-II), subthreshold bipolar disorder, and bipolar spectrum (BDS)
were 0.6%, 0.4%, 1.4%, and 2.4%, respectively (Merikangas et al.,
2011). Due to its early onset, severity and chronicity, bipolar disorder is

⁎
Corresponding author at: PGY5, Department of Psychiatry, Queen's University, MSc Candidate, Department of Public Health Sciences, Queen's University,
Abramsky Hall, Room 328, 21 Arch Street, Kingston, K7L 3N6 ON, Canada
E-mail address: 0ab104@queensu.ca (A. Bahji).

https://doi.org/10.1016/j.jad.2020.03.030
Received 20 January 2020; Received in revised form 12 March 2020; Accepted 12 March 2020
Available online 20 March 2020
0165-0327/ © 2020 Elsevier B.V. All rights reserved.
A. Bahji, et al.
one of the main causes of disability among young people, leading to
cognitive and functional impairment and raised mortality, particularly
death by suicide (Grande et al., 2016). A high prevalence of psychiatric
and medical comorbidities is also common among affected individuals
(Grande et al., 2016). Additionally, population growth and aging are
leading to an increase in the burden of bipolar disorder over time.
Accordingly, it is important that resources be directed towards im-
proving the coverage of evidence-based intervention strategies for bi-
polar disorder (Ferrari et al., 2016).
Overall, BD poses a large economic burden on the affected in-
dividual. In the United States alone, the total costs of BD-I were esti-
mated at over $200 billion in 2015, corresponding to an average of
roughly $80,000 per person (Cloutier et al., 2018). To reduce economic
strain, accurate diagnosis (Grande et al., 2016), effective pharmacologic
and psychologic treatments (Miura et al., 2014, Jauhar et al., 2016,
Jauhar et al., 2016), identification of biomarkers (Phillips and
Kupfer, 2013), optimizing resources, and improving overall care are
some of the prevailing needs in the field. Currently, pharmacologic
therapies are the mainstay of BD treatment and represent the standard
of care. Therefore, having reliable estimates of comparative efficacy,
tolerability, and acceptability could be considered both clinically and
economically advantageous.
While several recent reviews and meta-analyses in the literature
have attempted to answer this question (Sidor and Macqueen, 2011,
McGirr et al., 2016, Fornaro et al., 2018), there is still controversy
about the comparative efficacy and tolerability of acute treatment
agents. Most existing reviews and meta-analyses of treatments for bi-
polar depression have focused on the use of antidepressants and con-
tinue to reach the same conclusions, which are largely self-evident (e.g.,
that the four currently FDA-approved pharmacotherapies for bipolar
depression are all superior to placebo).
However, few prior reviews have considered both antidepressants
and ‘non-antidepressant’ pharmacotherapies for bipolar depression
(e.g., second generation antipsychotics, anti-epileptic medications) in
side-by-side fashion. The need to support clinical decision making and
cost-effectiveness analyses in medicine, despite a dearth of head-to-
head treatment comparisons, has encouraged the development of
methods enabling indirect comparisons of treatment alternatives, in-
cluding network meta-analysis (NMA) (Miura et al., 2014, Yildiz et al.,
2014, Cipriani et al., 2018). Valid application of NMA requires close
similarity of compared trials, including their design, patient char-
acteristics, and methods of diagnosis and symptomatic assessment. In
the absence of direct comparisons between all available pharmacolo-
gical treatments, a NMA could synthesize the available direct and in-
direct evidence (Salanti, 2012). With thoughtful and critical applica-
tion, NMA can add useful information concerning the comparative
benefits, risks, and costs of specific treatments in psychiatry
(Yildiz et al., 2014). This method has been previously applied to guide
clinical practices in medicine and psychiatry (Miura et al., 2014,
Cipriani et al., 2018, Cipriani et al., 2009, Cipriani et al., 2016,
Cortese et al., 2018, Leucht et al., 2013, Yildiz et al., 2015,
Ostacher et al., 2018).
Here, we performed a systematic review and network meta-analysis
of the efficacy and tolerability of pharmacologic treatments for bipolar
depression to provide a synthesis of the available evidence and to in-
form decisions about its acute treatment.
2. Methods
2.1. Protocol and registration
Before beginning the review, the study protocol was registered with
the PROSPERO database of systematic reviews (CRD42019122172)
(National Institute for Health Research 2019). We did our systematic
review in accordance with the PRISMA Extension Statement for Re-
porting of Systematic Reviews Incorporating Network Meta-analyses
155
Journal of Affective Disorders 269 (2020) 154–184
(Hutton et al., 2015). Ethics approval was not required for this study as
it was a meta-analysis of published trials.
2.2. Search strategy
This systematic review and network meta-analysis closely follows
the approach taken by several, previous high-quality network meta-
analyses of affective disorders (Miura et al., 2014, Cipriani et al., 2018,
Cipriani et al., 2009, Yildiz et al., 2015, Ostacher et al., 2018,
Furukawa et al., 2016). Several databases were searched, including
Cochrane Central Register of Controlled Trials, CINAHL and Pre-CI-
NAHL, Embase, LILACS database, MEDLINE, PsycINFO, and PubMed
from the date of their inception to April 2019 with no language or age
restrictions; an updated search was performed in December 2019, with
no interval studies identified. The electronic database searches were
supplemented with manual searches for published, unpublished, and
ongoing randomized controlled trials (RCT) in international trial reg-
isters, websites of drug approval agencies, and key scientific journals in
the field (Furukawa et al., 2016). For example, we searched Clinical-
Trials.gov using the search term “bipolar depression” to obtain sup-
plemental, unpublished information about both premarketing and post-
marketing studies. We also searched the grey literature for unpublished
studies using the World Health Organization's International Clinical
Trials Registry Platform (ICTRP). Additionally, we reviewed the re-
ference lists of all eligible articles—as well as reviews—for additional
studies. Our full search strategy is described in Appendix 1. Three in-
vestigators (AB, CS, and DE) independently selected the studies, re-
viewed the main reports and supplementary materials, extracted the
relevant information from the included trials, and assessed the risk of
bias. Any discrepancies were resolved by consensus and arbitration by a
panel of investigators within the review team (AB, CS, DE, EH, and GV).
2.3. Eligibility criteria
This is an intentionally selective review of monotherapies for bi-
polar depression; as such, trials involving augmentation or adjunct
therapies were excluded, with the exception of the combination treat-
ment, olanzapine/fluoxetine. However, to best inform clinical practice,
we purposefully included multiple classes of psychotropic medications.
We included all RCTs comparing any antidepressant, antipsychotic, or
mood-stabilizing agent versus placebo or active comparator as oral
monotherapy for the acute treatment of patients with a primary diag-
nosis of bipolar disorder—currently in the depressed phase—according
to standard operationalized diagnostic criteria (Feighner criteria,
Research Diagnostic Criteria, DSM-III, DSM-III-R, DSM-IV, DSM-5, and
ICD-10). We considered only double-blind trials because we included
placebo in the network meta-analysis, and because this study design
increases methodological rigor by minimizing performance and ascer-
tainment biases (Hróbjartsson et al., 2013). Studies were excluded
when participants were primarily in a non-depressed mood episo-
de—including manic, hypomanic, and mixed states—as well as those
with prophylaxis or relapse prevention designs for maintenance treat-
ment.
2.4. Assessed outcomes
Consistent with previous network meta-analyses, our primary out-
comes were efficacy and acceptability (Furukawa et al., 2016). Efficacy
was defined as response rate, measured by the total number of patients
who had a reduction of ≥ 50% of the total score on a standardized
observer-rating scale for depression, such as the Montgomery-Åsberg
Depression Rating Scale (MADRS) (Montgomery and Asberg, 1979) or
the Hamilton Depression Rating Scale (HDRS) (Hamilton, 1960). In the
absence of information or supplemental data from the authors, response
rate was calculated according to a validated imputation method
(Furukawa et al., 2005). Conversely, acceptability was defined by the
A. Bahji, et al.
treatment completion (defined as the proportion who remained in the
study until the primary endpoint), as well as all-cause treatment dis-
continuation (defined as the proportion of patients who withdrew due
to adverse events).
Secondary outcomes were remission (defined as total endpoint
scores below 10 on the MADRS or below 7 on the HDRS), tolerability
(defined as the proportion who dropped out due to any cause), affective
switch (defined as the proportion experiencing a treatment-emergent
manic or hypomanic episode), and reduction in depression severity
(defined as the change in depression rating score at study endpoint
relative to baseline).
2.5. Data collection process
Data extraction was done independently by one author and con-
firmed by the remaining three review authors. Where necessary, cor-
responding authors were contacted to confirm data. A standardized tool
was used to extract information about authors, study objectives, sample
characteristics, inclusion/exclusion criteria, study design, experimental
processes, treatment protocols, outcome variables, and analytic strategy
in Cochrane's Covidence, a web-based systematic review manager
(Veritas Health Innovation 2019).
2.6. Risk of bias assessment
To ascertain the validity of eligible randomized trials, a pair of re-
viewers (AB and CS) working independently determined the adequacy
of randomization and concealment of allocation, blinding, extent of loss
to follow-up (i.e. proportion of patients in whom the investigators were
not able to ascertain outcomes), selective reporting, and other sources
of potential bias using the Cochrane Risk of Bias Tool for randomized
controlled trials (Higgins et al., 2011). Each study domain was assigned
a rating of low, high, or unclear risk; consensus ratings between the two
reviewers are presented. These individual domains were pooled to de-
termine an overall study-level risk of bias.
2.7. Summary measures
Quantitative analyses were performed on an intention-to-treat basis
and were confined to data derived from the period immediately after
treatment termination (i.e., at the completion of the trial's primary
endpoint). For each pairwise comparison of dichotomous outcomes
(response, remission, completion of treatment, all-cause dropouts,
withdrawal due to adverse events, and affective switch), we calculated
odds ratios (OR) with their 95% confidence intervals (CI) immediately
after treatment termination (i.e., completion of primary endpoint of the
trial). For response, remission, and completion of treatment, an OR of
greater than one was considered a positive effect; for all-cause drop-
outs, withdrawal due to adverse events, and affective switch, an OR of
less than one was considered a positive effect. For each pairwise com-
parison of continuous outcomes (e.g., reduction in severity of depres-
sion), we calculated standardized mean differences (SMD) with their
95% CI immediately after treatment termination. For all continuous
outcomes, a negative SMD was considered a positive effect. We as-
sumed a 2-sided P < .05 to indicate statistical significance: 95% CIs not
containing 1 were considered statistically significant for ORs, and those
not containing 0 were considered statistically significant for SMDs.
2.8. Planned methods of analysis
All statistical analyses were conducted using R studio
(Viechtbauer, 2010), while flow diagrams and risk of bias graphs were
generated using Cochrane's Review Manager 5.3 (The Cochrane
Collaboration 2014). The NMA was based on a frequentist random ef-
fects model, which preserves randomised treatment comparisons within
trials. To conduct network meta-analyses within a frequentist
156
Journal of Affective Disorders 269 (2020) 154–184
framework, we used the package netmeta version 0.9-7 from the open-
source software environment R Studio, version 3.5.1 (Rücker et al.,
2019). The R function pairwise transformed the data set to the contrast-
based format, which was needed for conducting the NMA
(Appendix 2).
A network was created using jointly randomizable pharmacologic
treatments. We assumed that any patient who met all inclusion criteria
was likely, in principle, to be randomized to any of the interventions in
the synthesis comparator set. We addressed the assumption of transi-
tivity in the network meta-analysis by first assessing whether the in-
cluded interventions were similar across studies using a different design
(Rouse et al., 2017). We considered random-effects models rather than
fixed-effect models because we assumed that the included studies dif-
fered with respect to clinical and other factors. Pairwise ORs were
calculated for the relevant comparisons. In addition, indirect evidence
was estimated using the entire network of evidence.
Heterogeneity was visualized with forest plots, and quantified using
the τ2 and I2 statistics (Higgins and Thompson, 2002). A value of
τ2 = 0.04 was considered as low heterogeneity; 0.09 as moderate;
and 0.16 as high heterogeneity (Higgins and Thompson, 2002). Like-
wise, a value of I2 = 0% to 50% was considered as low heterogeneity;
50% to 75% as moderate heterogeneity; and 75% to 90% as high het-
erogeneity (Cochrane Collaboration 2014). We assumed a common
estimate for the between-study heterogeneity variance across all in-
cluded comparisons.
2.9. Assessment of inconsistency
To determine the amount of heterogeneity explained by incon-
sistency (between-study heterogeneity), we used local and global net-
work methods (Higgins et al., 2003, Higgins et al., 2012). First locally,
using the netsplit command (i.e., splitting direct and indirect evidence);
and second globally, using the decomp.design command (i.e., using the
design-by-treatment interaction model). Consistency was mainly as-
sessed by the comparison of the conventional NMA model, for which
consistency is assumed, with a model that does not assume consistency
(a series of pairwise meta-analyses analysed jointly). When the trade-off
between model fit and complexity favoured the model with assumed
consistency, this model was preferred. We also calculated the difference
between direct and indirect evidence in all closed loops in the network;
inconsistent loops were identified with a significant disagreement be-
tween direct and indirect evidence (e.g., 95% credible interval that
excluded 0 for SMD or 1 for OR).
2.10. Risk of bias across studies
We assessed the possibility of publication bias by evaluating a
funnel plot of the trial effect sizes for asymmetry (Appendix 3), which
can result from the non-publication of small trials with negative results
(Sedgwick and Marston, 2015). The symmetry of funnel plots was as-
sessed visually, with Egger's test (Egger et al., 1997), with adjusted rank
correlation and regression asymmetry tests (Begg and Mazumdar, 1994,
Duval and Tweedie, 2000). For asymmetric funnel plots, we applied the
trim and fill method, acknowledging that other factors, such as differ-
ences in trial quality or true study heterogeneity, could produce
asymmetry in funnel plots (Begg and Mazumdar, 1994, Duval and
Tweedie, 2000).
2.11. Additional analyses
To delineate the impact of bipolar disorder subtype on measures of
treatment effectiveness, we conducted subgroup network meta-analyses
which restricted the analyses to results from trials with participants
who had exclusively Bipolar I Disorder or Bipolar II Disorder.
A. Bahji, et al.
Fig. 1. PRISMA systematic review flow diagram
3. Results
3.1. Study selection
A total of 50 trials were identified for inclusion in the review
(Fig. 1). The systematic search provided a total of 4,403 citations. After
adjusting for duplicates, 3,129 remained. Of these, 2,947 studies were
discarded because after reviewing the abstracts, it appeared that these
papers clearly did not meet the eligibility criteria. The full texts of the
remaining 182 citations were examined in more detail. Of these, 55
(McGirr et al., 2016, Anand et al., 2012, Bauer et al., 1999,
Bocchetta et al., 1993, Brennan et al., 2013, Brown et al., 2014,
Calabrese et al., 2010, Calabrese et al., 2014, Chengappa et al., 2000,
Diazgranados et al., 2010, Eden Evins et al., 2006, Frangou et al., 2006,
Frye et al., 2007, Frye et al., 2015, Geddes et al., 2016,
Geretsegger et al., 2008, Goldberg et al., 2004, Juruena et al., 2009,
Keck et al., 2006, Ketter et al., 2015, Lee et al., 2014, Loebel et al.,
2014, McElroy et al., 2015, McIntyre et al., 2002, McIntyre et al., 2019,
Mehrpooya et al., 2018, Murphy et al., 2014, Nemeroff et al., 2001,
Nery et al., 2008, Nierenberg et al., 2006, Nolen et al., 2007, Park et al.,
2017, Parker et al., 2006, Post et al., 2006, Quante et al., 2010,
Sachs et al., 1994, Sachs et al., 1994, Sachs et al., 2007, Sachs et al.,
157
Journal of Affective Disorders 269 (2020) 154–184
2011, Saricicek et al., 2011, Savitz et al., 2018, Schaffer et al., 2006,
Shelton and Stahl, 2004, Silverstone, 2001, Smeraldi et al., 1999,
Stoll et al., 1999, Suppes et al., 2016, Toniolo et al., 2017, Uhl et al.,
2014, van der Loos et al., 2009, Yatham et al., 2016, Zarate et al., 2012,
Zeinoddini et al., 2015, Zhang et al., 2007, Vieta et al., 2002) were
excluded as they were trials with adjunctive medication (i.e., in addi-
tion to a primary mood-stabilizing agent); 29 were review articles
(Miura et al., 2014, Sidor and Macqueen, 2011, McGirr et al., 2016,
Cortese et al., 2018, Yildiz et al., 2015, Ostacher et al., 2018,
Amann et al., 2011, Bartoli et al., 2017, Chiesa et al., 2012, Cruz et al.,
2010, Fountoulakis, 2012, Fountoulakis et al., 2017, Geddes et al.,
2009, Gijsman et al., 2004, Goldsmith et al., 2003, Parsaik et al., 2015,
Prabhavalkar et al., 2015, Reinares et al., 2013, Sarris et al., 2012,
Selle et al., 2014, Smith et al., 2010, Suttajit et al., 2014, Taylor et al.,
2014, Tränkner et al., 2013, Vázquez et al., 2013, Vázquez et al., 2011,
Yatham et al., 2018, Zhang et al., 2013); 17 were secondary analyses of
previously-published trials (Calabrese et al., 2008, Altshuler et al.,
2006, Amsterdam et al., 2004, Amsterdam and Shults, 2008,
Brown et al., 2009, Hirschfeld et al., 2006, Ionescu et al., 2015,
Leverich et al., 2006, Lorenzo-Luaces et al., 2016, Perlis et al., 2010,
Pilhatsch et al., 2010, Post et al., 2001, Rajagopalan et al., 2016,
Tohen et al., 2014, van der Loos et al., 2010, van der Loos et al., 2011,
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

Table 1
Summary of Characteristics of Randomized Controlled Trials for Bipolar Depression Treatment
Study Age Male (%) Treatment comparisons of interest Weeks

Agosti and Stewart 2007 (Agosti and 36.2 43.3 Imipramine (Hutton et al., 2015) vs Phenelzine (Hróbjartsson et al., 2013) vs Placebo 6
Stewart, 2007) (National Institute for Health Research 2019)
Altshuler 2017 (Altshuler et al., 2017) 38.7 44.4 Lithium (Calabrese et al., 2014) vs Sertraline (Bocchetta et al., 1993) 16
Amsterdam 2005 (Amsterdam and Shults, 2005) 41.0 77.8 Fluoxetine (Jauhar et al., 2016) vs Olanzapine/Fluoxetine (Jauhar et al., 2016) vs Olanzapine 8
(Phillips and Kupfer, 2013) vs Placebo (Phillips and Kupfer, 2013)
Amsterdam 2015 (Amsterdam et al., 2015) 42.0 48.1 Venlafaxine (McIntyre et al., 2002) vs Lithium (McElroy et al., 2015) 4
Baumhackl 1989 (Baumhackl et al., 1989) 54.2 24.0 Moclobemide (Cipriani et al., 2016) vs Imipramine (Salanti, 2012) 5
Baskaran 2013 (Baskaran et al., 2013) 44.0 14.3 Ziprasidone (Jauhar et al., 2016) vs Placebo (Miura et al., 2014) 4
Brown 2006 (Brown et al., 2006) 37.0 40.0 Olanzapine/Fluoxetine (Stokes et al., 1971) vs Lamotrigine (Stokes et al., 1971) 7
Calabrese 1999 (Calabrese et al., 1999) 37.0 42.4 Lamotrigine (Perlis et al., 2010) vs Placebo (McIntyre et al., 2019) 7
Calabrese 2005 (Calabrese et al., 2005) 37.5 40.1 Quetiapine (361) vs Placebo (Tohen et al., 2003) 8
Cohn 1989 (Cohn et al., 1989) 39.7 33.7 Fluoxetine (Higgins et al., 2011) vs Imipramine (Higgins et al., 2011) vs Placebo (Veritas Health 6
Innovation 2019)
Davis 2005 (Davis et al., 2005) 41.0 88.0 Divalproex (Yildiz et al., 2014) vs Placebo (Fornaro et al., 2018) 8
De Wilde and Doogan 1982 (De Wilde and 45.8 30.4 Fluvoxamine (Cloutier et al., 2018) vs Clomipramine (Ferrari et al., 2016) 4
Doogan, 1982)
DelBello 2009 (DelBello et al., 2009) 15.5 31.3 Quetiapine (Cipriani et al., 2016 27) vs Placebo (Salanti, 2012) 8
DelBello 2017 (DelBello et al., 2017) 14.3 51.0 Lurasidone (Swartz et al., 2012) vs Placebo (Li et al., 2016) 6
Detke 2015 (Detke et al., 2015) 14.8 51.0 Olanzapine/Fluoxetine (Li et al., 2016) vs Placebo (Smeraldi et al., 1999) 8
Durgam 2016 (Durgam et al., 2016) 41.9 41.8 Cariprazine (436) vs Placebo (Zarate et al., 2005) 8
Earley 2019 (Earley et al., 2019) 42.8 40.8 Cariprazine (322) vs Placebo (Watson et al., 2012) 6
Findling 2014 (Findling et al., 2014) 13.9 50.1 Quetiapine (Zeinoddini et al., 2015) vs placebo (Fountoulakis, 2012) 8
Frye 2000 (Frye et al., 2000) 39.2 44.0 Lamotrigine (Hróbjartsson et al., 2013) vs Placebo (Hróbjartsson et al., 2013) 6
Ghaemi 2007 (Ghaemi et al., 2007) 38.3 52.5 Divalproex (Sidor and Macqueen, 2011) vs Placebo (Jauhar et al., 2016) 6
Grossman 1999 (Grossman et al., 1999) 41.0 38.0 Idazoxan (Jauhar et al., 2016) vs Bupropion (Phillips and Kupfer, 2013) 6
Himmelhoch 1991 (Himmelhoch et al., 1991) 38.7 39.3 Imipramine (Goldberg et al., 2004) vs Tranylcypromine (Furukawa et al., 2005) 6
Li 2016 (Li et al., 2016) 33.1 48.0 Quetiapine (Baron et al., 1975) vs Placebo (Donnelly et al., 1978) 8
Loebel 2014 (Loebel et al., 2014) 41.8 52.4 Lurasidone (Patkar et al., 2015) vs Placebo (Patkar et al., 2012) 6
Lombardo 2012 (Lombardo et al., 2012) 39.8 42.1 Ziprasidone (362) vs Placebo (Thase et al., 2006) 6
Lombardo 2012b (Lombardo et al., 2012) 40.2 42.5 Ziprasidone (Himmelhoch et al., 1991) vs Placebo (Muzina et al., 2011) 6
McElroy 2010 (McElroy et al., 2010) 38.8 38.8 Quetiapine (461) vs Paroxetine (Calabrese et al., 2008) vs Placebo (Amsterdam and Shults, 2008) 8
Murasaki 2018 (Murasaki et al., 2018) 39.0 45.5 Quetiapine (Patkar et al., 2015) vs Placebo (Baskaran et al., 2013) 8
Muzina 2011 (Muzina et al., 2011) 42.6 42.6 Divalproex (Montgomery and Asberg, 1979) vs Placebo (Furukawa et al., 2005) 6
Park 2017 (Park et al., 2017) 46.5 68.4 Riluzole (Jauhar et al., 2016) vs Placebo (McGirr et al., 2016) 8
Parker 2006 (Parker et al., 2006) 29.0 50.0 Escitalopram (Miura et al., 2014) vs Placebo (Ferrari et al., 2016) 36
Patkar 2015 (Patkar et al., 2015) 32.6 32.6 Ziprasidone (Calabrese et al., 2014) vs Placebo (Calabrese et al., 2014) 13
Riesenberg 2012 (Riesenberg et al., 2012) 39.0 52.5 Quetiapine Extended-Release (Nemeroff et al., 2001) vs Quetiapine (Nery et al., 2008) 1
Sachs 1994 (Sachs et al., 1994) 37.0 57.9 Bupropion (Phillips and Kupfer, 2013) vs Desipramine (Sidor and Macqueen, 2011) 8
Sachs 2001 (Cipriani et al., 2013) 37.8 37.8 Divalproex (Hutton et al., 2015) vs placebo (National Institute for Health Research 2019) 8
SCAA2010 2005 (Calabrese et al., 2008) 40.7 52.2 Lamotrigine (Gijsman et al., 2004) vs Placebo (Gijsman et al., 2004) 10
SCA40910 2008 (Calabrese et al., 2008) 37.5 45.0 Lamotrigine (Suttajit et al., 2014) vs Placebo (Selle et al., 2014) 8
SCA100223 2006 (Calabrese et al., 2008) 36.5 37.1 Lamotrigine (Tohen et al., 2014) vs Placebo (Pilhatsch et al., 2010) 8
SCA30924 2009 (Calabrese et al., 2008) 39.4 46.0 Lamotrigine (van der Loos et al., 2011) vs Placebo (Ionescu et al., 2015) 8
Suppes 2010 (Suppes et al., 2010) 39.5 35.1 Quetiapine (Baron et al., 1975) vs Placebo (Ballenger and Post, 1980) 8
Swartz 2012 (Swartz et al., 2012) 36.1 37.8 Quetiapine (McGirr et al., 2016) vs Placebo (Cipriani et al., 2018) 12
Thase 1992 (Thase et al., 1992) 35.9 31.3 Tranylcypromine (Fornaro et al., 2018) vs Imipramine (Ferrari et al., 2016) 6
Thase 2006 (Thase et al., 2006) 37.7 44.9 Quetiapine (341) vs Placebo (DelBello et al., 2009) 8
Thase 2008 (Thase et al., 2008) 39.8 38.4 Aripiprazole (Earley et al., 2019) vs Placebo (Loebel et al., 2014) 8
Thase 2008b (Thase et al., 2008) 38.4 38.4 Aripiprazole (DelBello et al., 2017) vs Placebo (Loebel et al., 2014) 8
Tohen 2003 (Tohen et al., 2003) 37.0 37.0 Olanzapine (370) vs Olanzapine/Fluoxetine (Stoll et al., 1999) vs Placebo (377) 8
Tohen 2012 (Tohen et al., 2012) 41.6 41.6 Olanzapine (343) vs Placebo (McElroy et al., 2010) 6
Wang 2014 (Wang et al., 2014) 41.2 41.2 Olanzapine (Rouse et al., 2017) vs Placebo (Rouse et al., 2017) 8
Young 2010 (Young et al., 2010) 40.4 40.4 Quetiapine (533) vs Lithium (Amsterdam, 1998) vs Placebo (van der Loos et al., 2011) 8
Zhang 2007 (Zhang et al., 2007) 34.1 45.6 Carbamazepine (Calabrese et al., 2014) vs Placebo (Hróbjartsson et al., 2013) 12

Wang et al., 2016, Xu et al., 2015); 13 were single-arm, open-label trials
(Amsterdam, 1998, Amsterdam and Garcia-España, 2000,
Ballenger and Post, 1980, Baron et al., 1975, Donnelly et al., 1978,
Dunn et al., 2008, Goodwin et al., 1969, Kemp et al., 2014,
McElroy et al., 2007, Milev et al., 2006, Post et al., 1986,
Schoeyen et al., 2015, Zarate et al., 2005); six involved maintenance
treatment (Vieta et al., 2002, Amsterdam and Shults, 2005,
Amsterdam and Shults, 2010, Amsterdam et al., 2015, Berk et al., 2008,
Daryani et al., 2014); six involved the wrong outcomes (Lally et al.,
2014, Licht et al., 2008, Peters et al., 2018, Stevens et al., 2013,
Watson et al., 2012, Young et al., 2000); two involved mixed states
(Berk et al., 2015, Patkar et al., 2012); two were not in English (粟幼嵩
陈 俊, Szádóczky and Füredi, 2002); two only involved individuals with
unipolar depression (Gillin et al., 1995, Lépine et al., 2000); and finally,
one did not involve an intervention (Kantrowitz et al., 2015). In total,
50 double-blind, parallel group, controlled RCTs (comprising 11,471
patients) were identified.
3.2. Study characteristics
Table 1 summarizes the characteristics of included studies. Studies
were conducted and published between 1989 and 2019. The majority of
trials were conducted in the USA, however, a few trials occurred in
Australia, Austria, Canada, China, and Japan. Across studies, a total of
27 distinct psychotropic agents were represented. Six unique second-
generation antipsychotics were identified: quetiapine (Calabrese et al.,
2005, DelBello et al., 2009, Findling et al., 2014, Li et al., 2016,
McElroy et al., 2010, Suppes et al., 2010, Swartz et al., 2012,
Thase et al., 2006, Young et al., 2010, Riesenberg et al., 2012); zipra-
sidone (Baskaran et al., 2013, Lombardo et al., 2012, Patkar et al.,

158
A. Bahji, et al.
2015); olanzapine (Amsterdam and Shults, 2005, Tohen et al., 2003,
Tohen et al., 2012, Wang et al., 2014); aripiprazole (Thase et al., 2008);
cariprazine (Durgam et al., 2016, Earley et al., 2019); and lurasidone
(DelBello et al., 2017, Loebel et al., 2014). 13 antidepressants were
identified: imipramine (Agosti and Stewart, 2007, Baumhackl et al.,
1989, Cohn et al., 1989, Himmelhoch et al., 1991, Thase et al., 1992),
tranylcypromine (Himmelhoch et al., 1991, Thase et al., 1992,
Himmelhoch et al., 1982), fluoxetine (Amsterdam and Shults, 2005,
Cohn et al., 1989), paroxetine (McElroy et al., 2010), sertraline
(Altshuler et al., 2017), moclobemide (Baumhackl et al., 1989), esci-
talopram (Parker et al., 2006), venlafaxine (Amsterdam et al., 2015),
phenelzine (Agosti and Stewart, 2007), clomipramine (De Wilde and
Doogan, 1982), fluvoxamine (De Wilde and Doogan, 1982), idazoxan
(Grossman et al., 1999), and bupropion (Grossman et al., 1999). Five
mood-stabilizing agents were located: divalproex (Davis et al., 2005,
Ghaemi et al., 2007, Muzina et al., 2011, Sachs et al., 2001), lamo-
trigine (Calabrese et al., 2008, Brown et al., 2006, Calabrese et al.,
1999, Frye et al., 2000), carbamazepine (Zhang et al., 2007), gaba-
pentin (Frye et al., 2000), and lithium (Amsterdam et al., 2015,
Young et al., 2010, Altshuler et al., 2017, Stokes et al., 1971). Finally,
four studies used the combination medication, olanzapine-fluoxetine
(Amsterdam and Shults, 2005, Tohen et al., 2003, Brown et al., 2006,
Detke et al., 2015).
The mean study sample size was 104 participants (SD = 114). In
total, 7,528 participants with bipolar depression were randomly as-
signed to an active drug while 3,920 were randomly assigned to pla-
cebo. The mean age was 38 years (SD = 8) for both men and women;
6,548 (57.2%) of the sample population were women. The median
duration of acute treatment was 8 weeks (interquartile range = 6–8).
3.3. Network quality
Eight (McElroy et al., 2010, Young et al., 2010, Amsterdam and
Shults, 2005, Tohen et al., 2003, Agosti and Stewart, 2007, Cohn et al.,
1989, Altshuler et al., 2017, Frye et al., 2000) trials involved three or
more groups. After exclusion of closed-loop networks from four studies
(Park et al., 2017, Riesenberg et al., 2012, De Wilde and Doogan, 1982,
Grossman et al., 1999), 46 studies—involving 22 interventions and 63
comparisons—were eligible for network meta-analysis. Fig. 2 details
the network of eligible comparisons for efficacy and acceptability. With
the exceptions of sertraline, moclobemide, and venlafaxine, all agents
had at least one placebo-controlled trial. The estimated value of be-
tween-study variance for the network meta-analysis was minimal across
comparisons. Tests of heterogeneity (within designs) and consistency
(between designs) were not statistically significant for response rates
(p = 0.9697) or tolerability (p = 0.7878).
3.4. Efficacy
Listed as decreasing in efficacy, tranylcypromine, venlafaxine,
fluoxetine, divalproex, imipramine, olanzapine-fluoxetine, lurasidone,
quetiapine, cariprazine, lamotrigine and olanzapine were more effec-
tive than placebo in improving outcomes, with ORs ranging between
1.49 (95% credible interval [CrI] 1.19–1.86) for olanzapine and 16.87
(4.52–62.91) for tranylcypromine (Fig. 3).
3.5. Acceptability
Completion of treatment was reported by all studies, and mea-
surement was supported by the inclusion of CONSORT (Schulz et al.,
2010) diagrams outlining study flow from randomization to completion
of treatment. Across trials, only aripiprazole was less acceptable than
placebo, with an OR for discontinuation due to adverse events of 2.25
(1.18–4.30). For all other agents, there was no significant difference in
dropouts due to adverse events relative to placebo (Fig. 3).
159
Journal of Affective Disorders 269 (2020) 154–184
3.6. Remission
The relative efficacy of pharmacotherapies compared with placebo
was also conducted for remission (Fig. 3). In order of decreasing effi-
cacy, tranylcypromine, fluoxetine, venlafaxine, olanzapine-fluoxetine,
quetiapine, lurasidone, olanzapine and lamotrigine were more effective
than placebo, with ORs ranging from 1.54 (1.07–2.21) for lamotrigine
and 18.38 (3.32–101.65) for tranylcypromine.
3.7. Treatment-emergent mania & hypomania
In terms of the relative tolerability of pharmacotherapies compared
with placebo, the only agent that was significantly less likely to trigger
a treatment-emergent affective switch was quetiapine, with an OR of
0.55 (0.36–0.84). None of the other agents were significantly different
from placebo (Fig. 3).
3.8. Depression severity
In terms of changes in the depression rating scale scores at the
completion of treatment, fluoxetine, divalproex, lurasidone, car-
iprazine, and quetiapine were more effective at lowering the absolute
severity of depression symptoms compared to placebo, with SMDs
ranging from -1.41 (-2.55, -0.27) for fluoxetine to -0.48 (-0.82, -0.14)
for quetiapine (Fig. 3).
3.9. Publication bias
For most outcomes, all statistical tests of funnel plot asymmetry
were non-significant, indicating low risk of publication bias
(Appendix 3).
3.10. Risk of bias within studies
The methodological features evaluated for each trial, as well as a
summary results of the judged risk of bias across studies, are presented
in Appendices 4–5. Five (10%) trials were rated as having a high risk of
bias, 8 (16%) trials as moderate, and 37 (74%) as low.
3.11. Subgroup network meta-analyses
We undertook subgroup network meta-analyses to delineate differ-
ences in response and remission rates by bipolar disorder subtype (i.e.,
BAD-I vs. BAD-II); these are graphically depicted in Appendix 6. There
were 21 trials involving participants with BAD-I only. In terms of re-
sponse, cariprazine, fluoxetine, imipramine, lamotrigine, lurasidone,
olanzapine-fluoxetine, and olanzapine were significantly better than
placebo. However, aripiprazole, divalproex, moclobemide, quetiapine,
and ziprasidone were not significantly better for depression in response.
In terms of remission, fluoxetine, imipramine, cariprazine, lurasidone,
olanzapine-fluoxetine, and olanzapine were significantly better than
placebo; aripiprazole, divalproex, moclobemide, quetiapine, ziprasi-
done, imipramine, and lamotrigine were not significantly better for
depression in terms of remission relative to placebo.
As there were only 6 trials that explored only participants with BAD-
II, a subgroup network meta-analysis could not be conducted because of
a lack of shared comparators across the six trials (Parker et al., 2006,
Amsterdam et al., 2015, Swartz et al., 2012, Agosti and Stewart, 2007,
Altshuler et al., 2017) as only three of the trials were placebo-con-
trolled. Similarly, a subgroup analysis could not be done for partici-
pants with either BAD-I or BAD-II due to the lack of shared comparators
across these trials—this led to the creation of sub-networks, which
subsequently would violate the assumptions of network meta-analysis.
A. Bahji, et al.
Fig. 2. Network graph of the included studies to enable visualization of the geometry of the treatment network. OFC = olanzapine/fluoxetine combination.
4. Discussion
4.1. Summary of evidence
This network meta-analysis is based on 50 trials (46 in the analytic
sample), which included 11,471 patients randomly assigned to 27 in-
dividual antidepressant, antipsychotic, and mood-stabilising drugs or
placebo. The project extends previous work that has addressed the ef-
ficacy and tolerability of lurasidone versus other atypical antipsychotic
monotherapy agents in patients with bipolar depression
(Ostacher et al., 2018). The larger evidence base (about 11,000 versus
6,000) obtained through an exhaustive search for published and un-
published information, allowed us to investigate additional important
outcomes, including remission and the occurrence of treatment-emer-
gent mania. To our knowledge, the present review is the most com-
prehensive network meta-analysis of treatments for bipolar depression
to date.
For bipolar depression, the present network meta-analysis can be
considered clinically useful specifically because of the vast number of
available treatment options. Coupled with multiple independent pla-
cebo-controlled or head-to-head trials, it is impossible to determine
which treatment is the most efficacious (Dias et al., 2010). Network
meta-analysis provides consistent estimates of the relative intervention
effects compared with all others using both direct and indirect evidence
without double counting the patients. The inclusion of active com-
parators (i.e., other drug therapies) is important as it encourages op-
timal treatment selection by clinicians for patients with bipolar de-
pression. Even in the absence of head-to-head randomized controlled
trials, this network meta-analysis provides clinicians with the bottom-
line regarding the best available evidence on the comparative efficacy
160
Journal of Affective Disorders 269 (2020) 154–184
of different pharmacologic agents for the acute treatment of bipolar
depression.
Unlike previous network meta-analyses, our analysis was not lim-
ited to atypical antipsychotics used as mono-therapy for the treatment
of bipolar depression. Our review strengthens the evidence base as it
identified antidepressants—mainly tranylcypromine, venlafaxine,
fluoxetine, and imipramine—that are also efficacious and well-toler-
ated for the treatment of bipolar depression, without a significant in-
crease in the rate of treatment-emergent mania. These findings chal-
lenge conventional wisdom regarding the risk of treatment-emergent
mania and particular classes of antidepressant medication
(Fornaro et al., 2018, PEJr et al., 2005, Henry et al., 2001).
Regarding the primary efficacy outcome, escitalopram, phenelzine,
moclobemide, carbamazepine, sertraline, lithium, paroxetine, ar-
ipiprazole, gabapentin and ziprasidone appear to be ineffective as
compared to placebo in treatment of bipolar depression according to
the network meta-analysis. As some of these results may be surprising
to clinicians, we must first emphasize that these results are for agents
delivered in monotherapy—as such, these results may not be applicable
when considering these medications as adjuncts or in augmentation to
others.
Conversely, divalproex, olanzapine/fluoxetine, olanzapine, quetia-
pine, cariprazine, and lamotrigine, appear to be effective as compared
to placebo in treatment of bipolar depression according to the network
meta-analysis. These results are somewhat in agreement with the recent
Canadian Network for Mood and Anxiety Disorder Treatments
(CANMAT) and International Society of Bipolar Disorders (ISBD)
guidelines, which recommend quetiapine, lurasidone, lithium, lamo-
trigine, lurasidone as first-line treatments, and divalproex, cariprazine,
and olanzapine-fluoxetine as second-line treatments (Yatham et al.,
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

2018). Of the antidepressant agents included in this NMA, tranylcy- Regarding the complex issue of treatment-emergent affective
promine, venlafaxine, fluoxetine, and imipramine were more effica- switch, the present study demonstrated that quetiapine reduced the risk
cious than placebo for the treatment of acute bipolar depression, and of switch relative to placebo, while no other agent was clearly superior
the summary effect sizes were moderate to large. to placebo at reducing affective switch. However, it is important to

Fig. 3. Contrast plots for odds ratios of depression response, remission, completion of treatment, and dropout due to adverse events versus placebo.
OFC = olanzapine-fluoxetine. combination.

161
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

Fig. 3. (continued)

162
A. Bahji, et al.
Fig. 3. (continued)
mention the long-term risk for cycle acceleration is an entirely separate
issue. A previous study by Prien and colleagues explored long-term
affective switches in individuals with unipolar and bipolar depression
treated with lithium plus imipramine, lithium alone, or imipramine
alone (Prien et al., 1984). Therein, the authors found that the combi-
nation treatment provided no advantage over imipramine alone, with
the lithium carbonate-treated group having fewer manic episodes than
the other groups. This is an important consideration when choosing to
use an antidepressant for acute bipolar depression because the present
meta-analysis’ results alone leaves unanswered questions about the
safety and wisdom of continuing an intervention beyond the acute
phase.
4.2. Strengths
One of the strengths of this study is that relative and absolute effect
sizes are both presented for medication efficacy. For example, on first
glance, fluoxetine appears to have a larger effect than quetiapine for
bipolar depression, partly because we have presented a relative mea-
sure; however, the accompanying absolute measure provides a useful
comparison. This is particularly relevant for tranylcypromine, as having
the largest relative effect may appear somewhat problematic on first
glance, yet, the standardized mean difference in depression severity
demonstrated a more modest effect size. Furthermore, this NMA pro-
vide more than one measure of acceptability. As most previous reviews
have used all-cause discontinuation as a proxy for acceptability, how-
ever, we also included dropout due to adverse events and treatment-
emergent affective switches as these were measure of real-world tol-
erability for patients and clinicians, respectively.
4.3. Limitations
Provided that similarity and consistency assumptions hold, network
meta-analysis can yield useful information about the relative efficacy
and tolerability of different treatments that have not been directly
163
Journal of Affective Disorders 269 (2020) 154–184
compared to one another (Yildiz et al., 2014). However, there are a
number of methodologic problems and idiosyncrasies in existing studies
of bipolar depression that historically have made it very difficult to
undertake comparisons across agents, as has been noted in previous
reviews and meta-analyses.
First, agents used in small, single-site, proof of concept studies are
often combined with larger, multi-site randomized trials, effectively
reducing the basis for gauging the rigor and robustness of findings re-
ported from one study to another. For example, studies that report large
effects that have gone replicated (e.g., tranylcypromine or venlafaxine)
are accorded the same stature as very large-multi-site, replicated trials
(e.g., the four FDA-approved treatments). Mostly these involve un-
balanced accounting for outcome moderators, such as bipolar I vs II
subtype (Altshuler et al., 2006), subthreshold mixed features
(Frye et al., 2009), rapid cycling, variable episode number, medication
dosing, and the use of concomitant pharmacotherapies. For example,
regarding the latter of these factors, the Himmelhoch et al. study of
tranylcypromine notoriously lacked systematic data on concomitant
mood stabilizer therapy (Himmelhoch et al., 1991). Similarly, in the
Nemeroff et al. adjunctive paroxetine trial (not included in the present
review), the outcome was moderated (post hoc) by therapeutic vs.
nontherapeutic lithium level (Nemeroff et al., 2001). Furthermore,
fluoxetine was more efficacious when combined with olanzapine
(Tohen et al., 2003) but not with lithium, as in the Cohn et al. study
(Cohn et al., 1989).
Second, there are many well-recognized pharmacotherapies for bi-
polar depression for which at least one published placebo-controlled
trial exists that are neither included nor mentioned in this paper, in-
cluding ketamine (Zarate et al., 2012, Ionescu et al., 2015,
Kantrowitz et al., 2015), modafinil/armodafinil (Calabrese et al., 2010,
Calabrese et al., 2014, Frye et al., 2007, Ketter et al., 2015), prami-
pexole (Goldberg et al., 2004, Lattanzi et al., 2002), isradipine
(Ostacher et al., 2014), supraphysiologic levothyroxine (Uhl et al.,
2014), specific combinations of lamotrigine (Geddes et al., 2016,
van der Loos et al., 2009), and inositol (Chengappa et al., 2000,
A. Bahji, et al.
Eden Evins et al., 2006, Nierenberg et al., 2006), among others. The
decision to pursue a review of only monotherapies—rather than com-
bination with adjunctive pharmacologic agents—affects the overall
network meta-analysis results. However, these trials were intentionally
excluded from this analysis because combining participants who are in
receipt of adjunctive medications with those receiving a single medi-
cation would violate the consistency assumption of network meta-
analysis (Higgins et al., 2012).
A third important methodological problem worth noting for those
trying to interpret the literature is that some studies lump together two
or more antidepressants within a class without separating out the ef-
fects of one versus another, such as the STEP-BD study from 2007,
which did not differentiate outcomes with paroxetine versus bupropion
(Sachs et al., 2007).
A fourth methodological problem that raises more questions than
answers about treatment efficacy is the notorious spate of failed versus
negative RCTs in bipolar depression, notably, the two failed ar-
ipiprazole trials (Thase et al., 2008) and the five lamotrigine industry
trials that all suffered more from high placebo response rates than from
failure to demonstrate intrinsic value of a particular active drug arm
(Calabrese et al., 2008). This methodological dilemma warrants parti-
cular emphasis in order to give more context to the present meta-ana-
lysis, otherwise the absence of evidence could become confused with
evidence of absence (Steinert, 2009).
Consequently, if the evidence base informing these analyses is found
to be heterogeneous, this may modify observed treatment effects. Meta-
regression, which can potentially adjust for such effect modifiers, could
not be run due to the absence of a sufficiently large number of trials per
comparison that is required to render meta-regression meaningful at the
aggregate level (Berkey et al., 1998, López-López et al., 2017). Simi-
larly, our ability to run meaningful subgroup analyses was limited due
to the lack of a sufficient number of shared comparator agents across
trials—this led to the creation of undesirable, closed loop subnetworks
that prevented the delineation of treatment outcomes for participants
with only BAD-I or only BAD-I/BAD-II only. To that end, the subgroup
analyses we could conduct showed that quetiapine was not a statisti-
cally significant agent for the treatment of bipolar depression for those
with BAD-I; however, this finding was due to the fact that only one trial
was restrictive to only BAD-I, and the remaining 11 quetiapine trials
included both BAD-I and BAD-II patients.
Although it is not known whether trial duration is a treatment effect
modifier, it is possible that these differences may have biased results
and subsequent interpretation of the findings. Finally, while our study
expanded upon the tolerability outcomes of other meta-analyses, we did
not include metabolic parameters—such as hyperglycemia, diabetes
mellitus, and dyslipidemia—or extrapyramidal symptoms.
4.4. Research implications
While this review identified a wide variety of pharmacotherapies, it
purposefully excluded trials of adjunctive medication strategies, which
have therapeutic potential for individuals with bipolar de-
pression—particularly those with treatment resistance. This could be a
topic of future review studies. Additionally, future studies could explore
functional (rather than symptom-based) depression outcomes that
correlate with meaningful improvements in depression, or by using
biological measures of depression, such as BDNF levels. The effective-
ness, safety, and tolerability of larger and longer trials would also be
helpful in understanding the therapeutic potential of medications be-
yond the acute treatment window. Alternatively, real-world measures
of bipolar depression treatment may be provided through phase IV ef-
fectiveness studies using administration datasets. Finally, some under-
studied subpopulations (i.e., older adults, individuals with concurrent
psychiatric conditions and substance use disorders) should be included
164
Journal of Affective Disorders 269 (2020) 154–184
to increase generalizability of findings.
4.5. Clinical implications
On the basis of this review, several pharmacotherapies appear to
show promise for the treatment of bipolar depression in terms of re-
sponse and remission of depression symptoms, completion of treatment,
the occurrence of adverse events, the likelihood of treatment-emergent
mania, and for absolute reductions in depression severity. In conjunc-
tion with evidence-based psychotherapies and neurostimulation treat-
ments, these pharmacotherapies may one day be incorporated into in-
dividualized treatment algorithms. In parallel, research exploring the
molecular mechanisms of these diverse agents may help us refine our
understanding on the pathophysiology of bipolar depression and its
treatment. For example, the differential activation and involvement of
serotonergic, dopaminergic, and noradrenergic receptors may be re-
levant to understanding which patients are better suited for particular
agents, while the alteration of downstream second-messenger path-
ways, neuronal-membrane stabilization (through antiepileptic agents),
and other novel mechanisms may work synergistically for some pa-
tients—but not others.
5. Conclusions
Results from this NMA suggest that cariprazine, lurasidone, que-
tiapine, and olanzapine are more efficacious than aripiprazole and zi-
prasidone monotherapies, and that, tranylcypromine, fluoxetine, ven-
lafaxine, and imipramine, are more effective than paroxetine and
phenelzine for the management of bipolar depression. The mood sta-
bilizers lamotrigine, and divalproex were similarly efficacious, and both
performed better than lithium and placebo. Overall, this NMA suggests
that several pharmacological agents are efficacious treatment options in
bipolar depression.
Data sharing
The full dataset will be freely available online in Mendeley Data, a
secure online repository for research data, which allows archiving of
any file type and assigns a permanent and unique digital object iden-
tifier (DOI) so that the files can be easily referenced.
Role of the funding source
This study received no external funding. All authors had full access
to all the data in the study and had final responsibility for the decision
to submit for publication following author consent.
Author Statement
All authors contributed wholly and equally to the development of
the manuscript, from its conceptualization to writing of the final draft.
With regard to specific roles, Dr. Vazquez and Dr. Hawken undertook
supervisory roles, while Mr. Stephenson and Mr. Ermacora supported
data extraction and data analysis. Dr. Bahji supported all phases of the
manuscript's development.
Declaration of Competing Interest
The authors declare that there are no relevant conflicts of interest.
Acknowledgements
Nil.
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jad.2020.03.030.

Appendix 1. Search Strategy

EMBASE: inception to April 29, 2019

Search step Hits

1. Exp bipolar depression 5654

2. Exp drug therapy 2582411
3. Exp antidepressant agent 425647
4. Exp neuroleptic agent/ or exp anticonvulsive agent/ or exp lithium/ or exp mood stabilizer/ or exp carbamazepine/ or exp tranquilizer/ or exp valproic acid 715479
5. 2 or 3 or 4 333186
6. 1 and 5 3797

MEDLINE: inception April 29, 2019

Search step Hits

1. Exp bipolar depression 37,841

2. Exp antidepressant agents 136,837
3. Exp antipsychotic agents 116,433
4. Exp anticonvulsants 132,520
5. Exp lithium 214,64
6. Exp Valproic acid/ or exp antimanic agents 31,021
7. Exp carbamazepine 10,700
8. Exp “Hypnotics and Sedatives”/ or exp anti-anxiety agents/ or exp tranquilizing agents 278,615
9. 2 or 3 or 4 or 5 or 6 or 7 or 8 456,765
10. Exp depression 1011
11. 1 and 10 3058
12. 9 and 11 1011
13. Limit 12 to (English language and humans and randomized controlled trial) 45

PsycINFO: inception to April 29, 2019

Search step Hits

1. Exp bipolar disorder 25,423

2. Exp major depression 120,470
3. 1 and 2 5,781
4. Ex drug therapy/ or exp neuroleptic drugs/ or exp antidepressant drugs 158,097
5. Exp lithium 6,256
6. Exp anticonvulsive drugs 11,284
7. Exp valproic acid 1,691
8. Exp carbamazepine 1,414
9. Exp sedatives/ or exp tranquilizing drugs/ or exp minor tranquilizers 1,052
10. 4 or 5 or 6 or 7 or 8 or 9 315,321
11. 3 and 10 1,652
12. Limit 11 to (human and English language and “0300 clinical trial”) 135

Cochrane Library: inception April 29, 2019

Search step Hits

1. MeSH descriptor: [Bipolar Disorder] explode all trees 23,18

2. MeSH descriptor: [Psychotropic Drugs] explode all trees 12,770
3. MeSH descriptor: [Antidepressive Agents] explode all trees 5,360
4. MeSH descriptor: [Antipsychotic Agents] explode all trees 4,289
5. MeSH descriptor: [Lithium] explode all trees 647
6. MeSH descriptor: [Anticonvulsants] explode all trees 2,268
7. MeSH descriptor: [Valproic Acid] explode all trees 867
8. MeSH descriptor: [Antimanic Agents] explode all trees 386
9. MeSH descriptor: [Hypnotics and Sedatives] explode all trees 3,356
10. 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 18,725
11. MeSH descriptor: [Depressive Disorder, Major] explode all trees 4,044
12. 1 and 11 198
13. 10 and 12 72
14. Limit 13 to (Trials) 72

165
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

PubMed: inception to April 29, 2019

Search step Hits

1. (“bipolar depression”) AND (((((((((antidepressants) OR mood stabilizer) OR neuroleptic) OR lithium) OR valpro*) OR anticonvulsant) OR antiepileptic) OR carbama- 1,276
zepine) OR sedative)
2. Limit to Randomized Controlled Trial 165

CINAHL and Pre-CINAHL: inception to April 29, 2019

Search step Hits

1. (MH "Bipolar Disorder") OR “"bipolar depression" 10,390

2. (MH "Antidepressive Agents") OR (MH "Antipsychotic Agents") OR "neuroleptic" OR (MH "Lithium") OR "lithium" OR (MH "Lithium Carbonate") OR (MH "Lithium 60,919
Compounds") OR (MH "Anticonvulsants") OR (MH "Valproic Acid") OR (MH "Topiramate") OR (MH "Tiagabine Hydrochloride") OR (MH "Tiagabine") OR (MH "Pri-
midone") OR (MH "Pregabalin") OR (MH "Phenytoin") OR (MH "Phenobarbital") OR (MH "Lamotrigine") OR (MH "Gabapentin") OR "anticonvulsant" OR "bipolar
depression" OR (MH "Antianxiety Agents, Benzodiazepine") OR (MH "Temazepam") OR (MH "Quazepam") OR "benzodiazepines" OR (MH "Drug Therapy") OR "pha-
rmacotherapy"
3. (MH "Randomized Controlled Trials") 81,466
4. 1 and 2 and 3 101

LILACS: inception to April 29, 2019

Search step Hits

1. "bipolar depression" [Words] and randomized controlled trial 5

WHO's International Clinical Trials Registry Platform (ICTRP): inception to April 30, 2019

Search step Hits

1. URL: https://www.who.int/ictrp/en/
2. List by Health Topics: http://apps.who.int/trialsearch/ListBy.aspx?TypeListing=0
3. Search: “bipolar” and “depression”, restrict to “Results available” 0

US Food and Drug Administration Clinicaltrials.gov Database: inception to April 30, 2019

Search step Hits

1. URL: https://clinicaltrials.gov/
2. Advanced search: “Bipolar Disorder Depression” AND “Interventional Studies (Clinical Trials)” AND “Studies with Results” 83

166
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

Appendix 2. Meta-analysis code (attached)

167
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

168
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

169
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

170
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

171
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

Appendix 3. Funnel plots and tests of symmetry for publication bias

172
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

173
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

174
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

Appendix 4. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included
studies

Appendix 5. Risk of bias summary: review authors' judgements about each risk of bias item for each included study

175
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

176
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

Appendix 6. Subgroup analyses for Bipolar I Disorder

177
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

178
A. Bahji, et al. Journal of Affective Disorders 269 (2020) 154–184

179
A. Bahji, et al.
References
American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental
Disorders, Fifth Edition. American Psychiatric Association Publishing.
Grande, I, Berk, M, Birmaher, B, Vieta, E, 2016 Apr 9. Bipolar disorder. The Lancet 387
(10027), 1561–1572.
Merikangas, KR, Jin, R, He, J-P, Kessler, RC, Lee, S, Sampson, NA, et al., 2011 Mar Mar.
Prevalence and correlates of bipolar spectrum disorder in the world mental health
survey initiative. Arch. Gen. Psychiatry 68 (3), 241–251.
Ferrari, AJ, Stockings, E, Khoo, J-P, Erskine, HE, Degenhardt, L, Vos, T, et al., 2016. The
prevalence and burden of bipolar disorder: findings from the Global Burden of
Disease Study 2013. Bipolar Disord. 18 (5), 440–450.
Cloutier, M, Greene, M, Guerin, A, Touya, M, Wu, E, 2018. The economic burden of
bipolar I disorder in the United States in 2015. J. Affect Disord. 15 (226), 45–51.
Miura, T, Noma, H, Furukawa, TA, Mitsuyasu, H, Tanaka, S, Stockton, S, et al., 2014 Oct
Oct. Comparative efficacy and tolerability of pharmacological treatments in the
maintenance treatment of bipolar disorder: a systematic review and network meta-
analysis. Lancet Psychiatry 1 (5), 351–359.
Jauhar, S, McKenna, PJ, Laws, KR, 2016 Apr Apr. NICE guidance on psychological
treatments for bipolar disorder: searching for the evidence. Lancet Psychiatry 3 (4),
386–388.
Jauhar, S, McKenna, PJ, Laws, KR, 2016 Apr Apr. Psychosocial treatments in bipolar
disorder. Lancet Psychiatry 3 (4), 321.
Phillips, ML, Kupfer, DJ, 2013 May 11. Bipolar disorder diagnosis: challenges and future
directions. Lancet 381 (9878), 1663–1671.
Sidor, MM, Macqueen, GM., 2011 Feb Feb. Antidepressants for the acute treatment of
bipolar depression: a systematic review and meta-analysis. J Clin Psychiatry 72 (2),
156–167.
McGirr, A, Vöhringer, PA, Ghaemi, SN, Lam, RW, Yatham, LN, 2016. Safety and efficacy
of adjunctive second-generation antidepressant therapy with a mood stabiliser or an
atypical antipsychotic in acute bipolar depression: a systematic review and meta-
analysis of randomised placebo-controlled trials. Lancet Psychiatry 3 (12),
1138–1146.
Fornaro, M, Anastasia, A, Novello, S, Fusco, A, Solmi, M, Monaco, F, et al., 2018.
Incidence, prevalence and clinical correlates of antidepressant-emergent mania in
bipolar depression: a systematic review and meta-analysis. Bipolar Disord. 20 (3),
180
Journal of Affective Disorders 269 (2020) 154–184
195–227.
Yildiz, A, Vieta, E, Correll, CU, Nikodem, M, Baldessarini, RJ, 2014 Dec. Critical issues on
the use of network meta-analysis in psychiatry. Harv. Rev. Psychiatry 22 (December
(6)), 367–372.
Cipriani, A, Furukawa, TA, Salanti, G, Chaimani, A, Atkinson, LZ, Ogawa, Y, et al., 2018.
Comparative efficacy and acceptability of 21 antidepressant drugs for the acute
treatment of adults with major depressive disorder: a systematic review and network
meta-analysis. Lancet 07;391 (10128), 1357–1366.
Salanti, G, 2012 Jun Jun. Indirect and mixed-treatment comparison, network, or mul-
tiple-treatments meta-analysis: many names, many benefits, many concerns for the
next generation evidence synthesis tool. Res. Synth. Methods 3 (2), 80–97.
Cipriani, A, Furukawa, TA, Salanti, G, Geddes, JR, Higgins, JP, Churchill, R, et al., 2009
Feb 28. Comparative efficacy and acceptability of 12 new-generation antidepressants:
a multiple-treatments meta-analysis. Lancet 373 (9665), 746–758.
Cipriani, A, Zhou, X, Del Giovane, C, Hetrick, SE, Qin, B, Whittington, C, et al., 2016 Aug
27. Comparative efficacy and tolerability of antidepressants for major depressive
disorder in children and adolescents: a network meta-analysis. Lancet 388 (10047),
881–890.
Cortese, S, Adamo, N, Del Giovane, C, Mohr-Jensen, C, Hayes, AJ, Carucci, S, et al., 2018
Sep Sep. Comparative efficacy and tolerability of medications for attention-deficit
hyperactivity disorder in children, adolescents, and adults: a systematic review and
network meta-analysis. Lancet Psychiatry 5 (9), 727–738.
Leucht, S, Cipriani, A, Spineli, L, Mavridis, D, Orey, D, Richter, F, et al., 2013.
Comparative efficacy and tolerability of 15 antipsychotic drugs in schizophrenia: a
multiple-treatments meta-analysis. Lancet 382 (September 14(9896)), 951–962.
Yildiz, A, Nikodem, M, Vieta, E, Correll, CU, Baldessarini, RJ, 2015 Jan Jan. A network
meta-analysis on comparative efficacy and all-cause discontinuation of antimanic
treatments in acute bipolar mania. Psychol. Med. 45 (2), 299–317.
Ostacher, M, Ng-Mak, D, Patel, P, Ntais, D, Schlueter, M, Loebel, A, 2018 Dec Dec.
Lurasidone compared to other atypical antipsychotic monotherapies for bipolar de-
pression: a systematic review and network meta-analysis. World J. Biol. Psychiatry 19
(8), 586–601.
National Institute for Health Research. PROSPERO international prospective register of
systematic reviews [Internet]. 2019. Available from:http://www.crd.york.ac.uk/
PROSPERO/display_record.asp?ID=CRD42014006602.
Hutton, B, Salanti, G, Caldwell, DM, Chaimani, A, Schmid, CH, Cameron, C, et al., 2015
Jun 2. The PRISMA Extension Statement for Reporting of Systematic Reviews
A. Bahji, et al.
Incorporating Network Meta-analyses of Health Care Interventions: checklist and
Explanations. Ann. Intern. Med. 162 (11), 777.
Furukawa, TA, Salanti, G, Atkinson, LZ, Leucht, S, Ruhe, HG, Turner, EH, et al., 2016.
Comparative efficacy and acceptability of first-generation and second-generation
antidepressants in the acute treatment of major depression: protocol for a network
meta-analysis. BMJ Open 08;6 (7), e010919.
Hróbjartsson, A, Thomsen, ASS, Emanuelsson, F, Tendal, B, Hilden, J, Boutron, I, et al.,
2013. Observer bias in randomized clinical trials with measurement scale outcomes: a
systematic review of trials with both blinded and nonblinded assessors. CMAJ 185
(March 5(4)), E201–E211.
Montgomery, SA, Asberg, M, 1979 Apr Apr. A new depression scale designed to be sen-
sitive to change. Br. J. Psychiatry 134, 382–389.
Hamilton, M., 1960 Feb. A rating scale for depression. J. Neurol. Neurosurg. Psychiatry
23 (February(1)), 56–62.
Furukawa, TA, Cipriani, A, Barbui, C, Brambilla, P, Watanabe, N, 2005 Jan. Imputing
response rates from means and standard deviations in meta-analyses. Int. Clin.
Psychopharmacol. 20 (January(1)), 49–52.
Veritas Health Innovation, 2019. In: Covidence systematic review software. Melbourne,
Australia.
Higgins, JPT, Altman, DG, Gøtzsche, PC, Jüni, P, Moher, D, Oxman, AD, et al., 2011 Oct
18. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials.
BMJ 343, d5928.
Viechtbauer, W., 2010. Conducting meta-analyses in r with the metafor package. J Stat.
Soft. 36 (August 5(1)), 1–48.
The Cochrane Collaboration, 2014. Review Manager (RevMan). The Nordic Cochrane
Centre, Copenhagen.
Rücker G, Krahn U, König J, Efthimiou O, Schwarzer G. Netmeta: network meta-analysis
using frequentist methods [internet]. 2019[cited 2019 Oct 8]. Available
from:https://CRAN.R-project.org/package=netmeta.
Rouse, B, Chaimani, A, Li, T, 2017 Feb. Network meta-analysis: an introduction for
clinicians. Intern. Emerg. Med. 12 (February(1)), 103–111.
Higgins, JPT, Thompson, SG., 2002. Quantifying heterogeneity in a meta-analysis. Stat.
Med. 21 (June 15(11)), 1539–1558.
Cochrane Collaboration. Cochrane Handbook: General Methods for Cochrane Reviews.
[Internet]. Heterogeneity. 2014[cited 2019 Jul 17]. Available from:https://
handbook-5-1.cochrane.org/chapter_9/9_5_heterogeneity.htm.
Higgins, JPT, Thompson, SG, Deeks, JJ, Altman, DG, 2003. Measuring inconsistency in
meta-analyses. BMJ 327 (September 6(7414)), 557–560.
Higgins, JPT, Jackson, D, Barrett, JK, Lu, G, Ades, AE, White, IR, 2012 Jun. Consistency
and inconsistency in network meta-analysis: concepts and models for multi-arm
studies. Res. Synth. Methods 3 (June(2)), 98–110.
Sedgwick, P, Marston, L., 2015 Sep 16. How to read a funnel plot in a meta-analysis. BMJ
351, h4718.
Egger, M, Davey Smith, G, Schneider, M, Minder, C, 1997. Bias in meta-analysis detected
by a simple, graphical test. BMJ 315 (September 13(7109)), 629–634.
Begg, CB, Mazumdar, M., 1994 Dec. Operating characteristics of a rank correlation test
for publication bias. Biometrics 50 (December(4)), 1088–1101.
Duval, S, Tweedie, R, 2000 Jun. Trim and fill: A simple funnel-plot-based method of
testing and adjusting for publication bias in meta-analysis. Biometrics 56 (June(2)),
455–463.
Anand, A, Gunn, AD, Barkay, G, Karne, HS, Nurnberger, JI, Mathew, SJ, et al., 2012. Early
antidepressant effect of memantine during augmentation of lamotrigine inadequate
response in bipolar depression: a double-blind, randomized, placebo-controlled trial.
Bipolar Disord. 14 (1), 64–70.
Bauer, M, Zaninelli, R, Müller-Oerlinghausen, B, Meister, W, 1999 Apr. Paroxetine and
amitriptyline augmentation of lithium in the treatment of major depression: a double-
blind study. J. Clin. Psychopharmacol. 19 (April(2)), 164–171.
Bocchetta, A, Bernardi, F, Burrai, C, Pedditzi, M, Del Zompo, M, 1993 Dec. A double-blind
study of L-sulpiride versus amitriptyline in lithium-maintained bipolar depressives.
Acta Psychiatr Scand. 88 (December(6)), 434–439.
Brennan, BP, Jensen, JE, Hudson, JI, Coit, CE, Beaulieu, A, Pope, HGJ, et al., 2013 Oct. A
placebo-controlled trial of acetyl-L-carnitine and alpha-lipoic acid in the treatment of
bipolar depression. J. Clin. Psychopharmacol. 33 (October(5)), 627–635.
Brown, ES, Park, J, Marx, CE, Hynan, LS, Gardner, C, Davila, D, et al., 2014 Nov Nov. A
randomized, double-blind, placebo-controlled trial of pregnenolone for bipolar de-
pression. Neuropsychopharmacology 39 (12), 2867–2873.
Calabrese, J, Ketter, T, Youakim, J, Tiller, J, Yang, R, Frye, M, 2010. Adjunctive armo-
dafinil for major depressive episodes associated with bipolar I disorder: a rando-
mized, multicenter, double-blind, placebo-controlled, proof-of-concept study. J. Clin.
Psychiatry 71 (10), 1363‐1370.
Calabrese, JR, Frye, MA, Yang, R, Ketter, TA, 2014 Oct. Armodafinil Treatment Trial
Study Network. Efficacy and safety of adjunctive armodafinil in adults with major
depressive episodes associated with bipolar I disorder: a randomized, double-blind,
placebo-controlled, multicenter trial. J. Clin. Psychiatry 75 (October(10)),
1054–1061.
Chengappa, KN, Levine, J, Gershon, S, Mallinger, AG, Hardan, A, Vagnucci, A, et al., 2000
Mar. Inositol as an add-on treatment for bipolar depression. Bipolar Disord. 2 (March
(1)), 47–55.
Diazgranados, N, Ibrahim, L, Brutsche, NE, Newberg, A, Kronstein, P, Khalife, S, et al.,
2010 Aug. A randomized add-on trial of an N-methyl-D-aspartate antagonist in
treatment-resistant bipolar depression. Arch. Gen. Psychiatry 67 (August(8)),
793–802.
Eden Evins, A, Demopulos, C, Yovel, I, Culhane, M, Ogutha, J, Grandin, LD, et al., 2006
Apr. Inositol augmentation of Lithium or valproate for bipolar depression. Bipolar
Disord. 8 (April(2)), 168–174.
Frangou, S, Lewis, M, McCrone, P, 2006 Jan Jan. Efficacy of ethyl-eicosapentaenoic acid
181
Journal of Affective Disorders 269 (2020) 154–184
in bipolar depression: randomised double-blind placebo-controlled study. Br. J.
Psychiatry 188, 46–50.
Frye, MA, Grunze, H, Suppes, T, McElroy, SL, Keck, PEJ, Walden, J, et al., 2007 Aug. A
placebo-controlled evaluation of adjunctive modafinil in the treatment of bipolar
depression. Am. J. Psychiatry 164 (August(8)), 1242–1249.
Frye, MA, Amchin, J, Bauer, M, Adler, C, Yang, R, Ketter, TA, 2015 Dec. Randomized,
placebo-controlled, adjunctive study of armodafinil for bipolar I depression: im-
plications of novel drug design and heterogeneity of concurrent bipolar maintenance
treatments. Int. J. Bipolar Disord. 3 (December(1)), 34.
Geddes, JR, Gardiner, A, Rendell, J, Voysey, M, Tunbridge, E, Hinds, C, et al., 2016.
Comparative evaluation of quetiapine plus lamotrigine combination versus quetia-
pine monotherapy (and folic acid versus placebo) in bipolar depression (CEQUEL): a
2 × 2 factorial randomised trial. The Lancet Psychiatry 3 (January 1(1)), 31–39.
Geretsegger, C, Bitterlich, W, Stelzig, R, Stuppaeck, C, Bondy, B, Aichhorn, W, 2008 Feb.
Paroxetine with pindolol augmentation: a double-blind, randomized, placebo-con-
trolled study in depressed in-patients. Eur. Neuropsychopharmacol. 18 (February(2)),
141–146.
Goldberg, J, Burdick, K, Endick, C, 2004. Preliminary randomized, double-blind, placebo-
controlled trial of pramipexole added to mood stabilizers for treatment-resistant bi-
polar depression. Am. J. Psychiatry 161 (3), 564‐566.
Juruena, MF, Ottoni, GL, Machado-Vieira, R, Carneiro, RM, Weingarthner, N, Marquardt,
AR, et al., 2009 Feb. Bipolar I and II disorder residual symptoms: oxcarbazepine and
carbamazepine as add-on treatment to lithium in a double-blind, randomized trial.
Prog. Neuro-Psychopharmacol. Biolog. Psychiatry 33 (February(1)), 94–99.
Keck, PEJ, Mintz, J, McElroy, SL, Freeman, MP, Suppes, T, Frye, MA, et al., 2006. Double-
blind, randomized, placebo-controlled trials of ethyl-eicosapentanoate in the treat-
ment of bipolar depression and rapid cycling bipolar disorder. Biol. Psychiatry 60
(November 1(9)), 1020–1022.
Ketter, TA, Yang, R, Frye, MA, 2015. Adjunctive armodafinil for major depressive epi-
sodes associated with bipolar I disorder. Bastien C Calabrese, Calabrese, Calabrese,
Calabrese, First, Fountoulakis, Frye, Geddes, Ketter, Loebel, Loebel, Lombardo,
Posner, Post, Rush, Rush, Sachs, Sidor, Thase, Thase, Tohen, Young, editor. Journal of
Affective Disorders 181, 87–91 (American Psychiatric Association. (2000). Diagnostic
and Statistical Manual of Mental Disorders. Text Revision: DSM-IV-TR (4th ed.).
Washington, DC: American Psychiatric Association.):.
Lee, S-Y, Chen, S-L, Chang, Y-H, Chen, PS, Huang, S-Y, Tzeng, N-S, et al., 2014 Jun Jun.
The effects of add-on low-dose memantine on cytokine levels in bipolar II depression:
a 12-week double-blind, randomized controlled trial. J. Clin. Psychopharmacol. 34
(3), 337–343.
Loebel, A, Cucchiaro, J, Silva, R, Kroger, H, Sarma, K, Xu, J, et al., 2014 Feb Feb.
Lurasidone as adjunctive therapy with lithium or valproate for the treatment of bi-
polar I depression: a randomized, double-blind, placebo-controlled study. Am. J.
Psychiatry 171 (2), 169–177.
McElroy, SL, Martens, BE, Mori, N, Blom, TJ, Casuto, LS, Hawkins, JM, et al., 2015 Jan
Jan. Adjunctive lisdexamfetamine in bipolar depression: a preliminary randomized,
placebo-controlled trial. Int. Clin. Psychopharmacol. 30 (1), 6–13.
McIntyre, RS, Mancini, DA, McCann, S, Srinivasan, J, Sagman, D, Kennedy, SH, 2002 Jun
Jun. Topiramate versus bupropion SR when added to mood stabilizer therapy for the
depressive phase of bipolar disorder: a preliminary single-blind study. Bipolar Disord.
4 (3), 207–213.
McIntyre, RS, Subramaniapillai, M, Lee, Y, Pan, Z, Carmona, NE, Shekotikhina, M, et al.,
2019 May 8. Efficacy of Adjunctive Infliximab vs Placebo in the Treatment of Adults
With Bipolar I/II Depression: a randomized clinical trial. JAMA Psychiatry [Internet]
[cited 2019 Jun 18]; Available from: https://jamanetwork.com/journals/
jamapsychiatry/fullarticle/2732872.
Mehrpooya, M, Yasrebifar, F, Haghighi, M, Mohammadi, Y, Jahangard, L, 2018 Oct Oct.
Evaluating the effect of coenzyme q10 augmentation on treatment of bipolar de-
pression: a double-blind controlled clinical trial. J. Clin. Psychopharmacol. 38 (5),
460–466.
Murphy, BL, Babb, SM, Ravichandran, C, Cohen, BM, 2014. Oral SAMe in persistent
treatment-refractory bipolar depression: a double-blind, randomized clinical trial.
Berg B Champe, Hardy, Papakostas, Reynolds, Siegel, editor.. J. Clin.
Psychopharmacol. 34 (3), 413–415.
Nemeroff, CB, Evans, DL, Gyulai, L, Sachs, GS, Bowden, CL, Gergel, IP, et al., 2001 Jun
Jun. Double-blind, placebo-controlled comparison of imipramine and paroxetine in
the treatment of bipolar depression. Am. J. Psychiatry 158 (6), 906–912.
Nery, FG, Monkul, ES, Hatch, JP, Fonseca, M, Zunta-Soares, GB, Frey, BN, et al., 2008 Mar
Mar. Celecoxib as an adjunct in the treatment of depressive or mixed episodes of
bipolar disorder: a double-blind, randomized, placebo-controlled study. Hum.
Psychopharmacol. 23 (2), 87–94.
Nierenberg, A, Ostacher, M, Calabrese, J, Ketter, T, Marangell, L, Miklowitz, D, et al.,
2006. Treatment-resistant bipolar depression: a STEP-BD equipoise randomized ef-
fectiveness trial of antidepressant augmentation with lamotrigine, inositol, or ris-
peridone. Am. J. Psychiatry 163 (2), 210‐216.
Nolen, WA, Kupka, RW, Hellemann, G, Frye, MA, Altshuler, LL, Leverich, GS, et al., 2007
May May. Tranylcypromine vs. lamotrigine in the treatment of refractory bipolar
depression: a failed but clinically useful study. Acta Psychiatr Scand. 115 (5),
360–365.
Park, LT, Lener, MS, Hopkins, M, Iadorola, N, Machado-Vieira, R, Ballard, E, et al., 2017
Jun Jun. A double-blind, placebo-controlled, pilot study of riluzole monotherapy for
acute bipolar depression. J. Clin. Psychopharmacol. 37 (3), 355–358.
Parker, G, Tully, L, Olley, A, Hadzi-Pavlovic, D, 2006 Jun Jun. SSRIs as mood stabilizers
for Bipolar II disorder? a proof of concept study. J. Affect Disord. 92 (2–3), 205–214.
Post, RM, Altshuler, LL, Leverich, GS, Frye, MA, Nolen, WA, Kupka, RW, et al., 2006 Aug
Aug. Mood switch in bipolar depression: comparison of adjunctive venlafaxine, bu-
propion and sertraline. Br. J. Psychiatry 189, 124–131.
A. Bahji, et al.
Quante, A, Zeugmann, S, Luborzewski, A, Schommer, N, Langosch, J, Born, C, et al., 2010
Mar Mar. Aripiprazole as adjunct to a mood stabilizer and citalopram in bipolar
depression: a randomized placebo-controlled pilot study. Hum. Psychopharmacol. 25
(2), 126–132.
Sachs, GS, Lafer, B, Stoll, AL, Banov, M, Thibault, AB, Tohen, M, et al., 1994 Sep Sep. A
double-blind trial of bupropion versus desipramine for bipolar depression. J. Clin.
Psychiatry 55 (9), 391–393.
Sachs, GS, Nierenberg, AA, Calabrese, JR, Marangell, LB, Wisniewski, SR, Gyulai, L, et al.,
2007 Apr 26. Effectiveness of adjunctive antidepressant treatment for bipolar de-
pression. N. Engl. J. Med. 356 (17), 1711–1722.
Sachs, G, Ice, K, Chappell, P, Schwartz, J, Gurtovaya, O, Vanderburg, D, et al., 2011.
Efficacy and safety of adjunctive oral ziprasidone for acute treatment of depression in
patients with bipolar I disorder: a randomized, double-blind, placebo-controlled trial.
J. Clin. Psychiatry 72 (10), 1413‐1422.
Saricicek, A, Maloney, K, Muralidharan, A, Ruf, B, Blumberg, HP, Sanacora, G, et al., 2011
Jun Jun. Levetiracetam in the management of bipolar depression: a randomized,
double-blind, placebo-controlled trial. J. Clin. Psychiatry 72 (6), 744–750.
Savitz, JB, Teague, TK, Misaki, M, Macaluso, M, Wurfel, BE, Meyer, M, et al., 2018 Jan 24.
Treatment of bipolar depression with minocycline and/or aspirin: an adaptive, 2x2
double-blind, randomized, placebo-controlled, phase IIA clinical trial. Transl.
Psychiatry 8 (1), 27.
Schaffer, A, Zuker, P, Levitt, A, 2006 Nov Nov. Randomized, double-blind pilot trial
comparing lamotrigine versus citalopram for the treatment of bipolar depression. J.
Affect Disord. 96 (1–2), 95–99.
Shelton, RC, Stahl, SM., 2004 Dec Dec. Risperidone and paroxetine given singly and in
combination for bipolar depression. J. Clin. Psychiatry 65 (12), 1715–1719.
Silverstone, T, 2001 Aug Aug. Moclobemide vs. imipramine in bipolar depression: a
multicentre double-blind clinical trial. Acta Psychiatr Scand. 104 (2), 104–109.
Smeraldi, E, Benedetti, F, Barbini, B, Campori, E, Colombo, C, 1999 Apr Apr. Sustained
antidepressant effect of sleep deprivation combined with pindolol in bipolar de-
pression. A placebo-controlled trial. Neuropsychopharmacology. 20 (4), 380–385.
Stoll, AL, Severus, WE, Freeman, MP, Rueter, S, Zboyan, HA, Diamond, E, et al., 1999
May 1. Omega 3 fatty acids in bipolar disorder: a preliminary double-blind, placebo-
controlled trial. Arch. Gen. Psychiatry 56 (5), 407–412.
Suppes, T, Kroger, H, Pikalov, A, Loebel, A, 2016 Jul Jul. Lurasidone adjunctive with
lithium or valproate for bipolar depression: a placebo-controlled trial utilizing pro-
spective and retrospective enrolment cohorts. J. Psychiatr Res. 78, 86–93.
Toniolo, RA, Fernandes, F, de, BF, Silva, M, Dias R da, S, Lafer, B, 2017 Dec 15. Cognitive
effects of creatine monohydrate adjunctive therapy in patients with bipolar depres-
sion: results from a randomized, double-blind, placebo-controlled trial. J. Affect
Disord. 224, 69–75.
Uhl, I, Bez, JA, Stamm, T, Pilhatsch, M, Assion, H-J, Norra, C, et al., 2014 Jul Jul.
Influence of levothyroxine in augmentation therapy for bipolar depression on central
serotonergic function. Pharmacopsychiatry 47 (4–5), 180–183.
van der Loos, M, Mulder, P, Hartong, E, Blom, M, Vergouwen, A, de Keyzer, H, et al.,
2009. Efficacy and safety of lamotrigine as add-on treatment to lithium in bipolar
depression: a multicenter, double-blind, placebo-controlled trial. J. Clin. Psychiatry
70 (2), 223‐231.
Yatham, LN, Vieta, E, Goodwin, GM, Bourin, M, de Bodinat, C, Laredo, J, et al., 2016 Jan
Jan. Agomelatine or placebo as adjunctive therapy to a mood stabiliser in bipolar I
depression: randomised double-blind placebo-controlled trial. Br. J. Psychiatry 208
(1), 78–86.
Zarate, CAJ, Brutsche, NE, Ibrahim, L, Franco-Chaves, J, Diazgranados, N, Cravchik, A,
et al., 2012 Jun 1. Replication of ketamine's antidepressant efficacy in bipolar de-
pression: a randomized controlled add-on trial. Biol. Psychiatry 71 (11), 939–946.
Zeinoddini, A, Sorayani, M, Hassanzadeh, E, Arbabi, M, Farokhnia, M, Salimi, S, et al.,
2015 Mar Mar. Pioglitazone adjunctive therapy for depressive episode of bipolar
disorder: a randomized, double-blind, placebo-controlled trial. Depression Anxiety 32
(3), 167–173.
Zhang, Z-J, Kang, W-H, Tan, Q-R, Li, Q, Gao, C-G, Zhang, F-G, et al., 2007 Jun Jun.
Adjunctive herbal medicine with carbamazepine for bipolar disorders: a double-
blind, randomized, placebo-controlled study. J. Psychiatr Res. 41 (3–4), 360–369.
Vieta, E, Martinez-Arán, A, Goikolea, JM, Torrent, C, Colom, F, Benabarre, A, et al., 2002
Jun Jun. A randomized trial comparing paroxetine and venlafaxine in the treatment
of bipolar depressed patients taking mood stabilizers. J. Clin. Psychiatry 63 (6),
508–512.
Amann, B, Born, C, Crespo, JM, Pomarol-Clotet, E, McKenna, P, 2011 Oct Oct.
Lamotrigine: when and where does it act in affective disorders? A systematic review.
J. Psychopharmacol. (Oxford) 25 (10), 1289–1294.
Bartoli, F, Dell'Osso, B, Crocamo, C, Fiorillo, A, Ketter, TA, Suppes, T, et al., 2017. Benefits
and harms of low and high second-generation antipsychotics doses for bipolar de-
pression: a meta-analysis. J. Psychiatry Res. 88, 38–46.
Chiesa, A, Chierzi, F, De Ronchi, D, Serretti, A, 2012 Mar Mar. Quetiapine for bipolar
depression: a systematic review and meta-analysis. Int. Clin. Psychopharmacol. 27
(2), 76–90.
Cruz, M, Pincus, HA, Welsh, DE, Greenwald, D, Lasky, E, Kilbourne, AM, 2010 Feb Feb.
The relationship between religious involvement and clinical status of patients with
bipolar disorder. Bipolar Disord. 12 (1), 68–76.
Fountoulakis, KN., 2012 Mar Mar. Refractoriness in bipolar disorder: definitions and
evidence-based treatment. CNS Neurosci. Ther. 18 (3), 227–237.
Fountoulakis, KN, Yatham, L, Grunze, H, Vieta, E, Young, A, Blier, P, et al., 2017 Feb 1.
The International College of Neuro-Psychopharmacology (CINP) Treatment
Guidelines for Bipolar Disorder in Adults (CINP-BD-2017), Part 2: review, Grading of
the Evidence, and a Precise Algorithm. Int. J. Neuropsychopharmacol. 20 (2),
121–179.
Geddes, JR, Calabrese, JR, Goodwin, GM, 2009 Jan Jan. Lamotrigine for treatment of
182
Journal of Affective Disorders 269 (2020) 154–184
bipolar depression: independent meta-analysis and meta-regression of individual
patient data from five randomised trials. Br. J. Psychiatry 194 (1), 4–9.
Gijsman, HJ, Geddes, JR, Rendell, JM, Nolen, WA, Goodwin, GM, 2004 Sep Sep.
Antidepressants for bipolar depression: a systematic review of randomized, con-
trolled trials. Am. J. Psychiatry 161 (9), 1537–1547.
Goldsmith, DR, Wagstaff, AJ, Ibbotson, T, Perry, CM, 2003. Lamotrigine: a review of its
use in bipolar disorder. Drugs 63 (19), 2029–2050.
Parsaik, AK, Singh, B, Khosh-Chashm, D, Mascarenhas, SS, 2015 Nov Nov. Efficacy of
Ketamine in Bipolar Depression: Systematic Review and Meta-analysis. J. Psychiatry
Pract. 21 (6), 427–435.
Prabhavalkar, KS, Poovanpallil, NB, Bhatt, LK, 2015. Management of bipolar depression
with lamotrigine: an antiepileptic mood stabilizer. Front. Pharmacol [Internet].[cited
2019 Jun 18];6. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC4615936/.
Reinares, M, Rosa, AR, Franco, C, Goikolea, JM, Fountoulakis, K, Siamouli, M, et al., 2013
Mar Mar. A systematic review on the role of anticonvulsants in the treatment of acute
bipolar depression. Int. J. Neuropsychopharmacol. 16 (2), 485–496.
Sarris, J, Mischoulon, D, Schweitzer, I, 2012 Jan Jan. Omega-3 for bipolar disorder: meta-
analyses of use in mania and bipolar depression. J. Clin. Psychiatry 73 (1), 81–86.
Selle, V, Schalkwijk, S, Vázquez, GH, Baldessarini, RJ, 2014 Mar Mar. Treatments for
acute bipolar depression: meta-analyses of placebo-controlled, monotherapy trials of
anticonvulsants, lithium and antipsychotics. Pharmacopsychiatry 47 (2), 43–52.
Smith, LA, Cornelius, VR, Azorin, JM, Perugi, G, Vieta, E, Young, AH, et al., 2010 Apr Apr.
Valproate for the treatment of acute bipolar depression: systematic review and meta-
analysis. J. Affect Disord. 122 (1–2), 1–9.
Suttajit, S, Srisurapanont, M, Maneeton, N, Maneeton, B, 2014. Quetiapine for acute bi-
polar depression: a systematic review and meta-analysis. Drug Des. Devel. Ther. 8,
827–838.
Taylor, D, Sparshatt, A, Varma, S, Olofinjana, O, 2014 Mar 19. Antidepressant efficacy of
agomelatine: meta-analysis of published and unpublished studies. BMJ 348, g1888.
Tränkner, A, Sander, C, Schönknecht, P, 2013. A critical review of the recent literature
and selected therapy guidelines since 2006 on the use of lamotrigine in bipolar dis-
order. Neuropsychiatry Dis. Treat. 9, 101–111.
Vázquez, GH, Tondo, L, Undurraga, J, Baldessarini, RJ, 2013 Aug Aug. Overview of an-
tidepressant treatment of bipolar depression. Int. J. Neuropsychopharmacol. 16 (7),
1673–1685.
Vázquez, G, Tondo, L, Baldessarini, RJ, 2011 Jan Jan. Comparison of antidepressant re-
sponses in patients with bipolar vs. unipolar depression: a meta-analytic review.
Pharmacopsychiatry 44 (1), 21–26.
Yatham, LN, Kennedy, SH, Parikh, SV, Schaffer, A, Bond, DJ, Frey, BN, et al., 2018 Mar
14. Canadian Network for Mood and Anxiety Treatments (CANMAT) and
International Society for Bipolar Disorders (ISBD) 2018 guidelines for the manage-
ment of patients with bipolar disorder. Bipolar Disord.
Zhang, Y, Yang, H, Yang, S, Liang, W, Dai, P, Wang, C, et al., 2013 Nov 5. Antidepressants
for bipolar disorder: A meta-analysis of randomized, double-blind, controlled trials.
Neural Regen. Res. 8 (31), 2962–2974.
Calabrese, JR, Huffman, RF, White, RL, Edwards, S, Thompson, TR, Ascher, JA, et al.,
2008 Mar Mar. Lamotrigine in the acute treatment of bipolar depression: results of
five double-blind, placebo-controlled clinical trials. Bipolar Disord. 10 (2), 323–333.
Altshuler, LL, Suppes, T, Black, DO, Nolen, WA, Leverich, G, Keck, PE, et al., 2006 Feb 1.
Lower switch rate in depressed patients with bipolar ii than bipolar i disorder treated
adjunctively with second-generation antidepressants. AJP 163 (2), 313–315.
Amsterdam, JD, Shults, J, Brunswick, DJ, Hundert, M, 2004 Feb Feb. Short-term fluox-
etine monotherapy for bipolar type II or bipolar NOS major depression - low manic
switch rate. Bipolar Disord. 6 (1), 75–81.
Amsterdam, JD, Shults, J., 2008 Apr Apr. Comparison of short-term venlafaxine versus
lithium monotherapy for bipolar II major depressive episode: a randomized open-
label study. J. Clin. Psychopharmacol. 28 (2), 171–181.
Brown, E, Dunner, DL, McElroy, SL, Keck, PE, Adams, DH, Degenhardt, E, et al., 2009 Jul
Jul. Olanzapine/fluoxetine combination vs. lamotrigine in the 6-month treatment of
bipolar I depression. Int. J. Neuropsychopharmacol. 12 (6), 773–782.
Hirschfeld, R, Weisler, R, Raines, S, Macfadden, W, 2006. Quetiapine in the treatment of
anxiety in patients with bipolar I or II depression: a secondary analysis from a ran-
domized, double-blind, placebo-controlled study. J. Clin. Psychiatry 67 (3), 355‐362.
Ionescu, DF, Luckenbaugh, DA, Niciu, MJ, Richards, EM, Zarate, CAJ, 2015 Jun Jun. A
single infusion of ketamine improves depression scores in patients with anxious bi-
polar depression. Bipolar Disord. 17 (4), 438–443.
Leverich, GS, Altshuler, LL, Frye, MA, Suppes, T, McElroy, SL, Keck, PE, et al., 2006 Feb
Feb. Risk of switch in mood polarity to hypomania or mania in patients with bipolar
depression during acute and continuation trials of venlafaxine, sertraline, and bu-
propion as adjuncts to mood stabilizers. Am. J. Psychiatry 163 (2), 232–239.
Lorenzo-Luaces, L, Amsterdam, JD, Soeller, I, DeRubeis, RJ, 2016. Rapid versus non-rapid
cycling bipolar II depression: Response to venlafaxine and lithium and hypomanic
risk. Akiskal A, editor.. Acta Psychiatrica Scandinavica 133 (Clinical
Psychopharmacology [3340]), 459–469.
Perlis, RH, Adams, DH, Fijal, B, Sutton, VK, Farmen, M, Breier, A, et al., 2010 May May.
Genetic association study of treatment response with olanzapine/fluoxetine combi-
nation or lamotrigine in bipolar I depression. J. Clin. Psychiatry 71 (5), 599–605.
Pilhatsch, M, Wolf, R, Winter, C, Lewitzka, U, Bauer, M, 2010. Comparison of paroxetine
and amitriptyline as adjunct to lithium maintenance therapy in bipolar depression: A
reanalysis of a randomized, double-blind study. Anderson A Bauer, Bauer, Bocchetta,
Christiansen, Cipriani, Grunze, Grunze, Judd, Katona, Keck, Ketter, Kuhs, Moller,
Moller, Muller-Oerlinghausen, Muzina, Guy, Nemeroff, Nolen, Okamoto, Post,
Stuppaeck, Tondo, van der Loos, Wegener, Yatham, Young, editor. J. Affect. Disord.
126 (Clinical Psychopharmacology [3340]), 453–457.
Post, RM, Altshuler, LL, Frye, MA, Suppes, T, Rush, AJ, Keck, PE, et al., 2001 Oct Oct.
A. Bahji, et al.
Rate of switch in bipolar patients prospectively treated with second-generation an-
tidepressants as augmentation to mood stabilizers. Bipolar Disord. 3 (5), 259–265.
Rajagopalan, K, Bacci, ED, Ng-Mak, D, Wyrwich, K, Pikalov, A, Loebel, A, 2016 May 23.
Effects on health-related quality of life in patients treated with lurasidone for bipolar
depression: results from two placebo controlled bipolar depression trials. BMC
Psychiatry 16, 157.
Tohen, M, Kanba, S, McIntyre, RS, Fujikoshi, S, Katagiri, H, 2014. Efficacy of olanzapine
monotherapy in the treatment of bipolar depression with mixed features. Bauer B
Bowden, Goldberg, Judd, Katagiri, Montgomery, Nusslock, Tohen, Tohen, Vieta,
Vieta, Young, editor.. J. Affect. Disord. 164 (Clinical Psychopharmacology [3340]),
57–62.
van der Loos, MLM, Mulder, P, Hartong, EGTM, Blom, MBJ, Vergouwen, AC, van
Noorden, MS, et al., 2010 Sep Sep. Efficacy and safety of two treatment algorithms in
bipolar depression consisting of a combination of lithium, lamotrigine or placebo and
paroxetine. Acta Psychiatry Scand. 122 (3), 246–254.
van der Loos, MLM, Mulder, P, Hartong, EGTM, Blom, MBJ, Vergouwen, AC, van
Noorden, MS, et al., 2011 Feb Feb. 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.
Wang, G, Cheng, Y, Wang, JN, Wu, SH, Xue, HB, 2016. Efficacy and safety of olanzapine
for treatment of patients with bipolar depression: Chinese subpopulation analysis of a
double-blind, randomized, placebo-controlled study. Altshuler B Goldberg,
Hamilton, Hlastala, Judd, Katagiri, Katagiri, Keller, Montgomery, Niufan, Post,
Sheehan, Spearing, Tohen, Tohen, Tohen, Tohen, Tohen, Tohen, Xue, Young, editor..
Neuropsychiatric Disease and Treatment 12 (Altshuler, L. L., Gitlin, M. J., Mintz, J.,
Leight, K. L., Frye, M. A. (2002). Subsyndromal depression is associated with func-
tional impairment in patients with bipolar disorder. J Clin Psychiatry, 63, 9, 807-811.
http://dx.doi.org/10.4088/JCP.v63n0910).
Xu, AJ, Niciu, MJ, Lundin, NB, Luckenbaugh, DA, Ionescu, DF, Richards, EM, et al., 2015.
Lithium and valproate levels do not correlate with Ketamine's antidepressant efficacy
in treatment-resistant bipolar depression. Neural Plast. 2015, 858251.
Amsterdam, J., 1998. Efficacy and safety of venlafaxine in the treatment of bipolar II
major depressive episode. J. Clin. Psychopharmacol. 18 (5), 414‐417.
Amsterdam, J, Garcia-España, F., 2000. Venlafaxine monotherapy in women with bipolar
II and unipolar major depression. J. Affect. Disord. 59 (3), 225‐229.
Ballenger, JC, Post, RM., 1980 Jul Jul. Carbamazepine in manic-depressive illness: a new
treatment. Am. J. Psychiatry 137 (7), 782–790.
Baron, M, Gershon, ES, Rudy, V, Jonas, WZ, Buchsbaum, M, 1975 Sep 1. Lithium
Carbonate response in depression: prediction by unipolar/bipolar illness, average-
evoked response, catechol-o-methyl transferase, and family history. Arch. Gen.
Psychiatry 32 (9), 1107–1111.
Donnelly, EF, Goodwin, FK, Waldman, IN, Murphy, DL, 1978 May May. Prediction of
antidepressant responses to lithium. Am. J. Psychiatry 135 (5), 552–556.
Dunn, RT, Stan, VA, Chriki, LS, Filkowski, MM, Ghaemi, SN, 2008 Sep Sep. A prospective,
open-label study of Aripiprazole mono- and adjunctive treatment in acute bipolar
depression. J. Affect. Disord. 110 (1–2), 70–74.
Goodwin, FK, Murphy, DL, Bunney, WE, 1969 Oct Oct. Lithium-carbonate treatment in
depression and mania. A longitudinal double-blind study. Arch. Gen. Psychiatry. 21
(4), 486–496.
Kemp, DE, Schinagle, M, Gao, K, Conroy, C, Ganocy, SJ, Ismail-Beigi, F, et al., 2014 Jun
Jun. PPAR-γ agonism as a modulator of mood: proof-of-concept for pioglitazone in
bipolar depression. CNS Drugs 28 (6), 571–581.
McElroy, SL, Suppes, T, Frye, MA, Altshuler, LL, Stanford, K, Martens, B, et al., 2007 Aug
Aug. Open-label aripiprazole in the treatment of acute bipolar depression: a pro-
spective pilot trial. J. Affect. Disord. 101 (1–3), 275–281.
Milev, R, Abraham, G, Zaheer, J, 2006 Jul Jul. Add-on quetiapine for bipolar depression:
a 12-month open-label trial. Can. J. Psychiatry 51 (8), 523–530.
Post, RM, Uhde, TW, Roy-Byrne, PP, Joffe, RT, 1986 Jan Jan. Antidepressant effects of
carbamazepine. Am. J. Psychiatry 143 (1), 29–34.
Schoeyen, HK, Kessler, U, Andreassen, OA, Auestad, BH, Bergsholm, P, Malt, UF, et al.,
2015 Jan Jan. Treatment-resistant bipolar depression: a randomized controlled trial
of electroconvulsive therapy versus algorithm-based pharmacological treatment. J.
Psychiatry 172 (1), 41–51.
Zarate, CA Jr, JA, Quiroz, Singh, JB, Denicoff, KD, De Jesus, G, Luckenbaugh, DA, et al.,
2005. An Open-Label Trial of the Glutamate-Modulating Agent Riluzole in
Combination with Lithium for the Treatment of Bipolar Depression. Benoit B
Berman, De Sarro, First, Frizzo, Hebert, Lacomblez, Maj, Montgomery, Paul, Tonen,
Wagner, Wang, Young, Zarate, Zarate, Zarate, editor.. Biol. Psychiatry 57 (Affective
Disorders [3211]), 430–432.
Amsterdam, J, Shults, J., 2005. Fluoxetine monotherapy of bipolar type II and bipolar
NOS major depression: a double-blind, placebo-substitution, continuation study. Int.
Clin. Psychopharmacol. 20 (5), 257‐264.
Amsterdam, JD, Shults, J., 2010 Jul Jul. Efficacy and safety of long-term fluoxetine versus
lithium monotherapy of bipolar II disorder: a randomized, double-blind, placebo-
substitution study. Am. J. Psychiatry 167 (7), 792–800.
Amsterdam, JD, Lorenzo-Luaces, L, Soeller, I, Li, SQ, Mao, JJ, DeRubeis, RJ, 2015 Oct 1.
Safety and effectiveness of continuation antidepressant versus mood stabilizer
monotherapy for relapse-prevention of bipolar II depression: a randomized, double-
blind, parallel-group, prospective study. J. Affect Disord. 185, 31–37.
Berk, M, Copolov, DL, Dean, O, Lu, K, Jeavons, S, Schapkaitz, I, et al., 2008 Sep Sep. N-
acetyl cysteine for depressive symptoms in bipolar disorder–a double-blind rando-
mized placebo-controlled trial. Biol. Psychiatry 64 (6), 468–475.
Daryani K, Narendra Kumar B, OP R. A Comparative Study of Efficacy of Lamotrigine and
Levetiracetam in the Continuous Maintenance Phase of Patients with Bipolar
Depressive Disorder. 2014;4.
183
Journal of Affective Disorders 269 (2020) 154–184
Lally, N, Nugent, AC, Luckenbaugh, DA, Ameli, R, Roiser, JP, Zarate, CA, 2014 Oct 14.
Anti-anhedonic effect of ketamine and its neural correlates in treatment-resistant
bipolar depression. Transl. Psychiatry 4, e469.
Licht, RW, Gijsman, H, Nolen, WA, Angst, J, 2008 Nov Nov. Are antidepressants safe in
the treatment of bipolar depression? a critical evaluation of their potential risk to
induce switch into mania or cycle acceleration. Acta Psychiatr. Scand. 118 (5),
337–346.
Peters, EM, Bowen, R, Balbuena, L, 2018 Oct Oct. Melancholic symptoms in bipolar ii
depression and responsiveness to lamotrigine in an exploratory pilot study. J. Clin.
Psychopharmacol. 38 (5), 509–512.
Stevens, J, Bies, RR, Shekhar, A, Anand, A, 2013 Mar Mar. Bayesian model of Hamilton
Depression Rating Score (HDRS) with memantine augmentation in bipolar depres-
sion. Br. J. Clin. Pharmacol. 75 (3), 791–798.
Watson, S, Gallagher, P, Porter, RJ, Smith, MS, Herron, LJ, Bulmer, S, et al., 2012 Dec 1. A
randomized trial to examine the effect of mifepristone on neuropsychological per-
formance and mood in patients with bipolar depression. Biol. Psychiatry 72 (11),
943–949.
Young, LT, Joffe, RT, Robb, JC, MacQueen, GM, Marriott, M, Patelis-Siotis, I, 2000 Jan
Jan. Double-blind comparison of addition of a second mood stabilizer versus an an-
tidepressant to an initial mood stabilizer for treatment of patients with bipolar de-
pression. Am. J. Psychiatry 157 (1), 124–126.
Berk, M, Tiller, JWG, Zhao, J, Yatham, LN, Malhi, GS, Weiller, E, 2015. Effects of ase-
napine in bipolar I patients meeting proxy criteria for moderate-to-severe mixed
major depressive episodes: a post hoc analysis. J. Clin. Psychiatry 76 (6), 728–734.
Patkar, A, Gilmer, W, Pae, C, Vöhringer, P, Ziffra, M, Pirok, E, et al., 2012. A 6 week
randomized double-blind placebo-controlled trial of ziprasidone for the acute de-
pressive mixed state. Plos one 7 (4), e34757.
粟幼嵩陈 俊, Su You-song CJ. 碳酸锂联合阿立哌唑治疗双相障碍抑郁发作的疗效和安全性.
上海交通大学学报 (医学版). 31(11):1536.
Szádóczky, E, Füredi, J., 2002. Efficacy and acceptability of tianeptine and sertraline in
the acute treatment phase of depression. Encephale 28 (4), 343‐349.
Gillin, JC, Lauriello, J, Kelsoe, JR, Rapaport, M, Golshan, S, Kenny, WM, et al., 1995. No
antidepressant effect of biperiden compared with placebo in depression: a double-
blind 6-week clinical trial. Psychiatry Res. 58 (2), 99–105.
Lépine, JP, Goger, J, Blashko, C, Probst, C, Moles, MF, Kosolowski, J, et al., 2000 Sep Sep.
A double-blind study of the efficacy and safety of sertraline and clomipramine in
outpatients with severe major depression. Int. Clin. Psychopharmacol. 15 (5),
263–271.
Kantrowitz, JT, Halberstam, B, Gangwisch, J, 2015. Single-dose ketamine followed by
daily D-cycloserine in treatment-resistant bipolar depression. Beck C Crane,
Diazgranados, Epstein, Goble, Hamilton, Heresco-Levy, Hwang, Kantrowitz,
Montgomery, Murrough, Nathanson, Torun, Zarate, editor.. J. Clin. Psychiatry 76 (6),
737–738.
Calabrese, JR, Keck, PEJ, Macfadden, W, Minkwitz, M, Ketter, TA, Weisler, RH, et al.,
2005 Jul Jul. A randomized, double-blind, placebo-controlled trial of quetiapine in
the treatment of bipolar I or II depression. Am. J. Psychiatry 162 (7), 1351–1360.
DelBello, MP, Chang, K, Welge, JA, Adler, CM, Rana, M, Howe, M, et al., 2009 Aug Aug. A
double-blind, placebo-controlled pilot study of quetiapine for depressed adolescents
with bipolar disorder. Bipolar Disord. 11 (5), 483–493.
Findling, RL, Pathak, S, Earley, WR, Liu, S, DelBello, MP, 2014 Aug Aug. Efficacy and
safety of extended-release quetiapine fumarate in youth with bipolar depression: an 8
week, double-blind, placebo-controlled trial. J. Child Adolesc. Psychopharmacol. 24
(6), 325–335.
Li, H, Gu, N, Zhang, H, Wang, G, Tan, Q, Yang, F, et al., 2016 Apr Apr. Efficacy and safety
of quetiapine extended release monotherapy in bipolar depression: a multi-center,
randomized, double-blind, placebo-controlled trial. Psychopharmacology (Berl) 233
(7), 1289–1297.
McElroy, SL, Weisler, RH, Chang, W, Olausson, B, Paulsson, B, Brecher, M, et al., 2010
Feb Feb. A double-blind, placebo-controlled study of quetiapine and paroxetine as
monotherapy in adults with bipolar depression (EMBOLDEN II). J. Clin. Psychiatry 71
(2), 163–174.
Suppes, T, Datto, C, Minkwitz, M, Nordenhem, A, Walker, C, Darko, D, 2010 Feb Feb.
Effectiveness of the extended release formulation of quetiapine as monotherapy for
the treatment of acute bipolar depression. J. Affect. Disord. 121 (1–2), 106–115.
Swartz, CM, Krohmer, R, Michael, N, 2012 Jun Jun. ECT stimulus dose dependence on
current separately from charge. Psychiatry Res. 198 (1), 164–165.
Thase, ME, Macfadden, W, Weisler, RH, Chang, W, Paulsson, B, Khan, A, et al., 2006 Dec
Dec. Efficacy of quetiapine monotherapy in bipolar I and II depression: a double-
blind, placebo-controlled study (the BOLDER II study). J. Clin. Psychopharmacol. 26
(6), 600–609.
Young, AH, McElroy, SL, Bauer, M, Philips, N, Chang, W, Olausson, B, et al., 2010 Feb
Feb. A double-blind, placebo-controlled study of quetiapine and lithium mono-
therapy in adults in the acute phase of bipolar depression (EMBOLDEN I). J. Clin.
Psychiatry 71 (2), 150–162.
Riesenberg, RA, Baldytcheva, I, Datto, C, 2012 Nov Nov. Self-reported sedation profile of
quetiapine extended-release and quetiapine immediate-release during 6-day initial
dose escalation in bipolar depression: a multicenter, randomized, double-blind, phase
IV study. Clin. Ther. 34 (11), 2202–2211.
Baskaran, A, Summers, D, Willing, S, Jokic, R, Milev, R, 2013. Sleep architecture in zi-
prasidone-treated bipolar depression: a pilot study. Ther. Adv. Psychopharmacol.
Lombardo, I, Sachs, G, Kolluri, S, Kremer, C, Yang, R, 2012 Aug Aug. Two 6-week, ran-
domized, double-blind, placebo-controlled studies of ziprasidone in outpatients with
bipolar I depression: did baseline characteristics impact trial outcome? J. Clin.
Psychopharmacol. 32 (4), 470–478.
Patkar, AA, Pae, C-U, Vohringer, PA, Mauer, S, Narasimhan, M, Dalley, S, et al., 2015. A
13-week, randomized double-blind, placebo-controlled, cross-over trial of
A. Bahji, et al.
ziprasidone in bipolar spectrum disorder. Berman B Berman, Dunner, Fava, Ghaemi,
Ghaemi, Ghaemi, Ghaemi, Ghaemi, Goodwin, Goodwin, Koukopoulos, Louis,
Maclure, Marcus, Papakostas, Papakostas, Parker, Parker, Patkar, Patkar, Perlis,
Rush, Shelton, Woods, editor.. J. Clin. Psychopharmacol. 35 (Clinical
Psychopharmacology [3340]), 319–323.
Amsterdam, JD, Shults, J., 2005 Jul Jul. Comparison of fluoxetine, olanzapine, and
combined fluoxetine plus olanzapine initial therapy of bipolar type I and type II
major depression–lack of manic induction. J. Affect Disord. 87 (1), 121–130.
Tohen, M, Vieta, E, Calabrese, J, Ketter, TA, Sachs, G, Bowden, C, et al., 2003 Nov Nov.
Efficacy of olanzapine and olanzapine-fluoxetine combination in the treatment of
bipolar I depression. Arch. Gen. Psychiatry 60 (11), 1079–1088.
Tohen, M, McDonnell, DP, Case, M, Kanba, S, Ha, K, Fang, YR, et al., 2012 Nov Nov.
Randomised, double-blind, placebo-controlled study of olanzapine in patients with
bipolar I depression. Br. J. Psychiatry 201 (5), 376–382.
Wang, M, Tong, J, Huang, D, Zhu, G, Liang, G, Du, H, 2014 Jul Jul. Efficacy of olanzapine
monotherapy for treatment of bipolar I depression: a randomized, double-blind,
placebo controlled study. Psychopharmacology (Berl) 231 (14), 2811–2818.
Thase, M, Jonas, A, Khan, A, Bowden, C, Wu, X, McQuade, R, et al., 2008. Aripiprazole
monotherapy in nonpsychotic bipolar I depression: results of 2 randomized, placebo-
controlled studies. J. Clin. Psychopharmacol. 28 (1), 13‐20.
Durgam, S, Earley, W, Lipschitz, A, Guo, H, Laszlovszky, I, Nemeth, G, et al., 2016 Mar
Mar. An 8-week randomized, double-blind, placebo-controlled evaluation of the
safety and efficacy of cariprazine in patients with bipolar i depression. J. Psychiatry
173 (3), 271–281.
Earley, W, Burgess, MV, Rekeda, L, Dickinson, R, Szatmári, B, Németh, G, et al., 2019 Jun
1. cariprazine treatment of bipolar depression: a randomized double-blind placebo-
controlled phase 3 study. Am. J. Psychiatry 176 (6), 439–448.
DelBello, MP, Goldman, R, Phillips, D, Deng, L, Cucchiaro, J, Loebel, A, 2017 Dec Dec.
Efficacy and Safety of Lurasidone in Children and Adolescents With Bipolar I
Depression: A Double-Blind, Placebo-Controlled Study. J Am Acad Child Adolesc
Psychiatry. 56 (12), 1015–1025.
Loebel, A, Cucchiaro, J, Silva, R, Kroger, H, Hsu, J, Sarma, K, et al., 2014 Feb Feb.
Lurasidone monotherapy in the treatment of bipolar I depression: a randomized,
double-blind, placebo-controlled study. Am. J. Psychiatry 171 (2), 160–168.
Agosti, V, Stewart, JW., 2007. Efficacy and safety of antidepressant monotherapy in the
treatment of bipolar-II depression. Int. Clin. Psychopharmacol. 22 (5), 309–311.
Baumhackl, U, Bizière, K, Fischbach, R, Ch, Geretsegger, et al., 1989. Efficacy and tol-
erability of moclobemide compared with imipramine in depressive disorder
(DSM—III): An Austrian double-blind, multicentre study. Br. J. Psychiatry 155 (Suppl
6), 78–83.
Cohn, JB, Collins, G, Ashbrook, E, Wernicke, JF, 1989 Oct Oct. A comparison of fluoxetine
imipramine and placebo in patients with bipolar depressive disorder. Int. Clin.
Psychopharmacol. 4 (4), 313–322.
Himmelhoch, JM, Thase, ME, Mallinger, AG, Houck, P, 1991 Jul Jul. Tranylcypromine
versus imipramine in anergic bipolar depression. Am. J. Psychiatry 148 (7), 910–916.
Thase, ME, Mallinger, AG, McKnight, D, Himmelhoch, JM, 1992 Feb Feb. Treatment of
imipramine-resistant recurrent depression, IV: a double-blind crossover study of
tranylcypromine for anergic bipolar depression. Am. J. Psychiatry. 149 (2), 195–198.
Himmelhoch, JM, Fuchs, CZ, Symons, BJ, 1982 Oct Oct. A double-blind study of tra-
nylcypromine treatment of major anergic depression. J. Nerv. Ment. Dis. 170 (10),
628–634.
Altshuler, LL, Sugar, CA, McElroy, SL, Calimlim, B, Gitlin, M, Keck, PEJ, et al., 2017 Mar
1. Switch rates during acute treatment for bipolar ii depression with lithium, ser-
traline, or the two combined: a randomized double-blind comparison. Am. J.
Psychiatry 174 (3), 266–276.
De Wilde, JE, Doogan, DP, 1982 Sep Sep. Fluvoxamine and chlorimipramine in en-
dogenous depression. J. Affect. Disord. 4 (3), 249–259.
Grossman, F, Potter, WZ, Brown, EA, Maislin, G, 1999 Dec Dec. A double-blind study
comparing idazoxan and bupropion in bipolar depressed patients. J. Affect. Disord.
56 (2–3), 237–243.
Davis, LL, Bartolucci, A, Petty, F, 2005 Apr Apr. Divalproex in the treatment of bipolar
depression: a placebo-controlled study. J. Affect. Disord. 85 (3), 259–266.
Ghaemi, SN, Gilmer, WS, Goldberg, JF, Zablotsky, B, Kemp, DE, Kelley, ME, et al., 2007
Dec Dec. Divalproex in the treatment of acute bipolar depression: a preliminary
double-blind, randomized, placebo-controlled pilot study. J. Clin. Psychiatry 68 (12),
1840–1844.
184
Journal of Affective Disorders 269 (2020) 154–184
Muzina, DJ, Gao, K, Kemp, DE, Khalife, S, Ganocy, SJ, Chan, PK, et al., 2011 Jun Jun.
Acute efficacy of divalproex sodium versus placebo in mood stabilizer-naive bipolar I
or II depression: a double-blind, randomized, placebo-controlled trial. J. Clin.
Psychiatry 72 (6), 813–819.
Sachs, GS, Yan, LJ, Swann, AC, Allen, MH, 2001. Integration of suicide prevention into
outpatient management of bipolar disorder. J. Clin. Psychiatry 62 (Suppl 25), 3–11.
Brown, EB, McElroy, SL, Keck, PE, Deldar, A, Adams, DH, Tohen, M, et al., 2006 Jul Jul. A
7-week, randomized, double-blind trial of olanzapine/fluoxetine combination versus
lamotrigine in the treatment of bipolar I depression. J. Clin. Psychiatry 67 (7),
1025–1033.
Calabrese, JR, Bowden, CL, Sachs, GS, Ascher, JA, Monaghan, E, Rudd, GD, 1999. A
double-blind placebo-controlled study of lamotrigine monotherapy in outpatients
with bipolar I depression. J. Clin. Psychiatry 60 (2), 79–88.
Frye, MA, Ketter, TA, Leverich, GS, Huggins, T, Lantz, C, Denicoff, KD, et al., 2000. The
increasing use of polypharmacotherapy for refractory mood disorders: 22 years of
study. J. Clin. Psychiatry 61 (1), 9–15.
Stokes, Peter E, Shamoian, Charles A, Stoll, Peter M, Patton, Michael J, 1971. efficacy of
lithium as acute treatment of manic-depressive illness. The Lancet 297 (7713),
1319–1325.
Detke, HC, DelBello, MP, Landry, J, Usher, RW, 2015 Mar Mar. Olanzapine/Fluoxetine
combination in children and adolescents with bipolar I depression: a randomized,
double-blind, placebo-controlled trial. J. Am. Acad. Child Adolesc. Psychiatry 54 (3),
217–224.
Schulz, KF, Altman, DG, Moher, D, 2010 Mar 24. CONSORT 2010 Statement: updated
guidelines for reporting parallel group randomised trials. BMJ 340, c332.
Dias, S, Welton, NJ, Caldwell, DM, Ades, AE, 2010 Mar 30. Checking consistency in mixed
treatment comparison meta-analysis. Stat. Med. 29 (7–8), 932–944.
PEJr, Keck, SA, Corya, Altshuler, LL, Ketter, TA, McElroy, SL, Case, M, et al., 2005.
Analyses of Treatment-Emergent Mania With Olanzapine/Fluoxetine Combination in
the Treatment of Bipolar Depression. Boerlin B Cohn, Goodwin, Joffe, Kessing,
MacQueen, Peet, Post, Rouillon, Sachs, Spearing, Stoll, Tohen, Tohen, Tohen, Vieta,
Wehr, Young, editor.. J. Clin. Psychiatry 66 (5), 611–616.
Henry, C, Sorbara, F, Lacoste, J, Gindre, C, Leboyer, M, 2001 Apr Apr. Antidepressant-
induced mania in bipolar patients: identification of risk factors. J. Clin. Psychiatry 62
(4), 249–255.
Prien, RF, Kupfer, DJ, Mansky, PA, Small, JG, Tuason, VB, Voss, CB, et al., 1984 Nov Nov.
Drug therapy in the prevention of recurrences in unipolar and bipolar affective dis-
orders. Report of the nimh collaborative study group comparing lithium carbonate,
imipramine, and a lithium carbonate-imipramine combination. Arch. Gen. Psychiatry
41 (11), 1096–1104.
Frye, MA, Helleman, G, McElroy, SL, Altshuler, LL, Black, DO, Keck, PE, et al., 2009 Feb
Feb. Correlates of treatment-emergent mania associated with antidepressant treat-
ment in bipolar depression. Am. J. Psychiatry 166 (2), 164–172.
Lattanzi, L, Dell'Osso, L, Cassano, P, Pini, S, Rucci, P, Houck, PR, et al., 2002 Oct Oct.
Pramipexole in treatment-resistant depression: a 16-week naturalistic study. Bipolar
Disord. 4 (5), 307–314.
Ostacher, MJ, Iosifescu, DV, Hay, A, Blumenthal, SR, Sklar, P, Perlis, RH, 2014 Mar Mar.
Pilot investigation of isradipine in the treatment of bipolar depression motivated by
genome-wide association. Bipolar Disord. 16 (2), 199–203.
Steinert, T., 2009 Jul Jul. [Why are guidelines more irrational than metaanalyses].
Psychiatry Prax. 36 (5), 238–242.
Berkey, CS, Hoaglin, DC, Antczak‐Bouckoms, A, Mosteller, F, Colditz, GA, 1998. Meta-
analysis of multiple outcomes by regression with random effects. Stat. Med. 17 (22),
2537–2550.
López-López, JA, Van den Noortgate, W, Tanner-Smith, EE, Wilson, SJ, Lipsey, MW, 2017
Dec Dec. Assessing meta-regression methods for examining moderator relationships
with dependent effect sizes: a Monte Carlo simulation. Res. Synth. Methods 8 (4),
435–450.
Murasaki, M, Koyama, T, Kanba, S, Takeuchi, M, Shimizu, Y, Arita, E, et al., 2018 Oct Oct.
Multi-center, randomized, double-blind, placebo-controlled study of quetiapine ex-
tended-release formulation in Japanese patients with bipolar depression.
Psychopharmacology (Berl) 235 (10), 2859–2869.
Cipriani, A, Reid, K, Young, AH, Macritchie, K, Geddes, J, 2013 Oct 17. Valproic acid,
valproate and divalproex in the maintenance treatment of bipolar disorder. Cochrane
Database Syst. Rev.(10), CD003196.