Bipolar depression: Clinical correlates of receiving antidepressants
Juan Undurragaa,b, Ross J. Baldessarinib,c, Marc Valentía, Isabella Pacchiarottia,
Leonardo Tondob,c,d, Gustavo Vázquezb,e, Eduard Vietaa,b,⁎
aBipolar Disorders Program, Institute of Neuroscience, Hospital Clínic Barcelona, IDIBAPS, CIBERSAM, University of Barcelona, Barcelona, Catalonia, Spain
bInternational Consortium for Bipolar Disorder Research, McLean Hospital, Belmont, MA, United States
cDepartment of Psychiatry, Harvard Medical School, Boston, MA, United States
dLucio Bini Mood Disorders Center, Cagliari, Sardinia, Italy
eDepartment of Neuroscience, Palermo University, Buenos Aires, Argentina
a r t i c l ei n f o a b s t r a c t
Received 29 August 2011
Received in revised form 23 January 2012
Accepted 23 January 2012
Available online 9 March 2012
Background: The efficacy and tolerability of antidepressants (ADs) to treat or avoid episodes of
depression in bipolar disorder (BPD) patients as well as reasons for using them remain unre-
Methods: We analyzed patient-characteristics and outcomes of episodes of acute major depres-
sion among 290 adult, DSM-IV BPD patients (71% type-I, 52% women) at the Hospital Clinic of
Barcelona; 80% were given an AD and 20% were not; 80% of both groups also received mood-
stabilizers. We evaluated factors associated with AD-treatment using bivariate analyses and
multiple logistic-regression modeling.
Results: Factors associated with AD-use by multivariate modeling ranked: [a] more years ill, [b]
depressive first-lifetime episode, [c] more depressions/year, [d] melancholic index episode,
and [e] less affective illness in first-degree relatives. Within 8 weeks, depression improved
by ≥50%, less often among BPD patients given an AD (64.4%; 38.6% without switching into
hypo/mania) than not (82.1%; 78.6% without switching).
Conclusions: Use of ADs to treat acute BP-depression was very common and associated with a
more severe clinical history. Mood-switching was prevalent with AD-treatment even with
© 2012 Elsevier B.V. All rights reserved.
The value and tolerability of antidepressant drugs (ADs)
for the treatment or prevention of depressive episodes of bi-
polar disorder (BPD) patients remain surprisingly uncertain
(Baldessarini et al., 2010). Recent reviews of controlled trials
of antidepressant treatment focusing on acute depression
have drawn inconsistent conclusions, finding either support
of efficacy with low risk of switching into elevated mood-
states (Gijsman et al., 2004), or lack of consistent efficacy
(Sidor and MacQueen, 2011). However, only five of nine trials
included in those reviews or reported more recently involved
only cases of BP-depression with response rates. Two of these
found superior efficacy of an AD over placebo (Cohn et al.,
1989; Tohen et al., 2003) and three did not (Nemeroff et al.,
2001; Sachs et al., 2007; McElroy et al., 2010). Our prelimi-
nary, random-effects meta-analysis of the five trials yielded
a nonsignificant pooled drug/placebo response rate ratio
(RR: 1.26 (95% confidence-interval [CI]: 0.90–1.78; z=1.44,
p=0.15). Conversely, however, a recent meta-analysis of 10
controlled trials directly comparing BP and unipolar (UP) de-
pressive cases found little difference between pooled AD re-
sponses by diagnosis (Vázquez et al., 2011).
The preceding findings leave the possible clinical value of
ADs for acute BP-depression uncertain. Their inconsistencies
may reflect technical differences among trials or involve po-
tential confounds, such as conflation of clinical nonresponse
Journal of Affective Disorders 139 (2012) 89–93
⁎ Corresponding author at: Bipolar Disorders Program, Clinical Institute of
Neuroscience, Hospital Clínic Barcelona, Villarroel 170, 08036, Barcelona,
Spain. Tel.: +34 93 2275400; fax: +34 93 2275795.
E-mail address: firstname.lastname@example.org (E. Vieta).
0165-0327/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
Contents lists available at SciVerse ScienceDirect
Journal of Affective Disorders
journal homepage: www.elsevier.com/locate/jad
with switching into mixed-dysphoric mood-states, and ef-
fects of co-treatment with mood-stabilizing drugs (Tondo et
al., 2010; Rybakowski, 2012). Given these uncertainties, addi-
tional studies of AD effects in BP-depression are needed.
Moreover, since antidepressants are the most commonly
used and long-sustained treatments for BPD patients, regard-
less of current phase of illness (Baldessarini et al., 2007,
2008), reasons for this practice, and characteristics of BPD pa-
tients given ADs or not remain uncertain.
Accordingly, colleagues at the present study-site recently
evaluated factors associated with responsiveness to AD-
treatment in BPD patients (Pacchiarotti et al., 2011a), those
associated with manic-switching
(Valentí et al., 2011), use of AD in mixed states (Valentí et
al., in press), comparisons of AD alone or combined with
mood-stabilizers (Pacchiarotti et al., 2011b), and factors
that explain why clinicians maintain AD-treatment even dur-
ing acute manic episodes (Rosa et al., 2010). We now report
on an independent study involving 290 adult, currently de-
pressed BPD patients at the University of Barcelona to distin-
guish characteristics of those who did or did not receive AD
treatment during acute episodes of major depression.
This prospective, naturalistic, cohort study involved 290
consenting adult patients in the BPD Program of the Hospital
Clínic of the University of Barcelona, treated clinically with
systematic follow-up. All patients provided written, informed
consent for anonymous and aggregate analysis and reporting
of their clinical findings, following review and approval of the
study protocol by the Hospital ethics committee. All patients
currently attending the study clinic with a DSM-IV diagnosis
of BPD, type-I (BP-I) or type-II (BP-II), in a current DSM-IV
major depressive episode were included and treated clinical-
ly. Diagnoses were backed with the Structured Clinical Inter-
view for DSM-IV, I and II (SCID-I and -II; First et al., 1997),
and supported by a 17-item Hamilton Depression Rating
Scale (HDRS-17 (Hamilton, 1967)) total score of ≥20 at in-
take and Young Mania Rating Scale (YMRS (Young et al.,
1978)) scoreb7, using validated Spanish-language versions
of the scales (Colom et al., 2002; Bobes et al., 2003). Treat-
ments were determined clinically by each patient's treating
psychiatrist, and included antidepressants (ADs), lithium, an-
ticonvulsants, antipsychotics, and electroconvulsive treat-
ment (ECT). Patients with major medical co-morbidities
were excluded but psychiatric co-morbidity was allowed.
Systematic, prospective follow-up involved weekly visits for
the first two months, and biweekly to six months.
2.2. Clinical assessments
Structured interviews with patients and family members,
medical records, and systematic follow-up assessments pro-
vided socio-demographic data and clinical details of illness
history, including sex, approximate age at onset and years
to diagnosis of BPD, first-episode type, diagnosis (BP-I or
BP-II), years of illness, polarity, count and annual rate of re-
currences and their predominant polarity (defined at ≥2:1
excess of depressive or mania-like [hypo/mania, mixed] epi-
sodes (Colom et al., 2006; Baldessarini et al., 2012)), season-
ality of recurrences, occurrence of ≥4 major recurrences
within any 12-month period (rapid-cycling), lifetime ECT
courses given, number and annual rate of hospitalizations,
presence of current, and timing of past psychotic symptoms
and suicidal behaviors, as well as current outcome. We also
recorded information about lifetime treatment, including oc-
currences, doses, duration and outcomes of previous AD-
trials, and use of mood-stabilizers in the past and during
treatment of the index depressive episode, as well as history
of mood-switching during AD-treatment. We considered
some of the second generation antipsychotics as mood-
stabilizer agents according to the CANMAT guidelines
(Yatham et al., 2009).
At intake as well as each follow-up visit, patients were
assessed clinically by clinical mood-disorder experts and
rated with the HDRS-17 and YMRS scales. To define specific
outcomes of the index depressive episode and to characterize
individual illness history, we applied operational definitions
of symptomatic response (reduction of initial HDRS score by
≥50%), symptomatic remission (final HDRS ≤7), recovery (re-
mission sustained ≥8 weeks), subsyndromal depression (clini-
cally symptomatic but not meeting DSM-criteria for a current
major depressive episode), relapse (new DSM-IV episode of
major depression within 12 weeks of remission), recurrence
(new depressive episode >12 weeks following remission)
and treatment-emergent mood-switch (new-onset of DSM-IV
hypomania, mania, or mixed-episode following initial remis-
sion), all according to recently recommended guidelines
(Pacchiarotti et al., 2011b; Tohen et al., 2009). The rate of
switching during treatment and rates of response adjusted
for switching (as response–switch) were recorded.
2.3. Data analysis
Preliminary bivariate analyses compared factors associat-
ed with AD-treatment, using ANOVA (F) for continuous mea-
sures, and contingency tables (χ2) for categorical measures.
Factors associated with AD-treatment in these preliminary
bivariate analyses, or considered clinically relevant were test-
ed for independent and significant association with AD-
treatment using stepwise, logistic multiple regression model-
ing to obtain Odds Ratios (OR) and their 95% CIs. Averages are
means±standard deviation (SD). Statistics involving multiple
comparisons were Bonferroni-corrected (p≤0.05/23≤0.002).
Analyses employed commercial statistical programs (PASW
Statistics 18.0, SPSS Corp., Chicago, IL; Stata.8, StataCorp, Col-
lege Station, TX; Statview.5, SAS Institute, Cary, NC).
Patient-subjects were 290 adults (150 women and 140
men), of mean age 48.0±14.0 years, diagnosed with BP-I
(71.4%) or BP-II (28.6%) disorder, in index episodes of DSM-
IV major depression; 233 patients (80.3%) received ADs
with other medicines, and 57 (19.7%) were treated with
other medicines only. Of patients given ADs in the index de-
pressive episode, 60.9%receiveda serotonin-reuptake
J. Undurraga et al. / Journal of Affective Disorders 139 (2012) 89–93
inhibitor, 27.8% a mixed serotonin–norepinephrine reuptake
inhibitor, and 16.5% a tricyclic, including 3.86% given more
than one AD. Mood-stabilizers were given to 79.6% of all
BPD patients, including 79.8% of those given an AD and
78.6% of those not given an AD (χ2=0.044, p=0.8), includ-
ing similar proportions of BP-I (84.4%) and BP-II cases (78.8%;
3.2. Bivariate analyses
Preliminary bivariate analyses (Table 1) indicated that
AD-treatment was associated with the following factors be-
fore the index treatment-trial (ranked by relative risk):
[a] 18.6-times more predominant (≥2:1 episodes/year) de-
pressive polarity of previous episodes (42.1% of the subjects
could be classified by predominant polarity); [b] 8.23-times
more melancholia in the index episode; [c] 3.57-times longer
latency to diagnosis of BPD; [d] 3.40-times less BP-I diagnosis;
[e] 3.08-more previous ECT use; [f] 2.85-times greater likeli-
hood of a depressive first-lifetime episode; [g] 2.02-times
more depressions/year; [h] 52% more total illness; [i] 46% less
family history of mood disorders in first-degree relatives;
[j] 27% less likelihood of psychotic features in previous or
index episodes; and [k] 20% greater current age. With respect
to responses to treatment of the index depressive episode,
AD-use was associated with: [l] 7.37-times more switching
into elevated mood-states; [m] 60% less clinical recovery;
and [n] 27% less treatment response (≥50% improvement).
Switching rate between different mood stabilizing agents
was 19% for lithium, 23% for anticonvulsivants and 12% for
antipsychotics (χ2=1.36, p=0.51).
3.3. Multivariate modeling
In multivariate logistic regression modeling, factors that
remained significantly and independently associated with AD-
treatment ranked by significance: [a] more years ill; [b] more
depressions/year; [c] depressive first-lifetime episode, [d] mel-
ancholic current episode, and [e] less affective illness among
first-degree relatives. Inaddition, ≥2:1 predominant depressive
Characteristics of 290 depressed bipolar disorder patients given antidepressants (AD) or not.
Factor Total AD givenNo AD
F or χ2
Sex (n [%])
Family history (n [%])
First episode type (n [%])
Mania, hypomania, mixed
Bipolar diagnosis (n [%])
Years to BPD diagnosis
Total years ill
Living independently (n [%])
Predominant polarity (n [%])
Ever rapid-cycling (n [%])
Suicidal (n [%])
Lifetime psychosis (n [%])
Lifetime anxiety disorder (n [%])
Mood-stabilizer given (n [%])
Lifetime ECT (n [%])
Current melancholia (n [%])
Outcomes (n [%])
Response (≤8 weeks)
Switched ≤8 weeks (n [%])
Significance with Bonferroni correction (0.05/28) requires ≤0.002.
J. Undurraga et al. / Journal of Affective Disorders 139 (2012) 89–93
polarity (versus hypomania, mania, or mixed-states) was
strongly associated with current AD-treatment (Table 1), but
was excluded from logistic modeling as it conflicted with
other indices of depressive illness (including depressions/
year). Factors not significantly associated with AD-treatment
included: sex; age at index depression; diagnostic type (I or
II); latency from onset to bipolar diagnosis; lifetime counts or
annual rates of total episodes, manias–hypomanias or mixed-
states; ever rapid-cycling; seasonality; suicidal thoughts or
acts; lifetime ECT, ever being psychotic, and independent
We examined factors associated with AD-use in an acute
index episode of major depression among types I and II BPD
patients, and considered clinical outcomes. Most patients
(80.3%) received AD-treatment, despite the conflicting evi-
dence of its efficacy considered above. The overall rate of re-
sponse within eight weeks of various treatments (67.8%) was
similar to findings in treatment trials for unipolar major de-
pression (Undurraga and Baldessarini, 2012), and similar
among BP-I and BP-II cases, but 1.27-times lower with AD-
treatment (64.4%) than without it (82.1%).
with AD-treatment included: [a] more years ill, [b] more
depressions/year, [c] depressive first-lifetime episode, [d] mel-
ancholic features in current depressive episode, and [e] lower
risk of affective illness among first-degree relatives (Table 2).
These factors and others indicated by initial bivariate compari-
sons (Table 1) suggest that AD treatment was chosen for BPD
depressed patients with a more severe past history of depres-
sive illness, and avoided in patients with a BP-I diagnosis or a
positive family history. With AD-treatment, short-term clinical
responses were less favorable (notably, clinical recovery was
only half as likely), and risk of switching into mania-like states
was 7.4-times (25.8%/3.5%) higher (Table 1). The lower re-
sponse rates with ADs may reflect their observed association
with greater illness-severity, itself probably reflecting the na-
ture of the study-site as a referral center (Vieta, 2011).
Limitations of this study include lack of randomization, in-
clusion of uncontrolled treatment regimens not matched for
use of specific mood-stabilizing or antimanic agents or
doses, and sampling at a prominent university referral center
that may not generalize to other sites. Nevertheless, such nat-
uralistic experiences may serve as useful representations of
current clinical practices and results.
The prolonged latency to the diagnosis of BPD in the pre-
sent sample (averaging 6.44 years) is consistent with other
international findings (Baldessarini et al., 2003). It may be
that prominent depressive morbidity, especially in early
years of BPD illness, was accompanied by limited expression
of elevated mood. Nevertheless, such delay in recognizing
BPD clinically underscores the need for earlier recognition
and introduction of appropriate treatments, particularly of
In summary, the present study found that AD-treatment
was very common among depressed BPD patients at an aca-
demic referral center (80.3%), and that several differences be-
tween depressed BPD patients who were treated with ADs or
not were striking, including strong evidence of more promi-
nent and intensive lifetime depressive morbidity. Treatment-
responses were less favorable with AD-use than without it,
perhaps owing to more severe and depression-prone lifetime
illnesses. Response-rates with ADs (usually added to mood-
stabilizing treatments) were similar in types I and II BPD. Re-
sponses with AD-treatment were similar to those that have
been reported for unipolar depressive episodes; however, the
question of relative responsiveness to ADs in bipolar versus
unipolar depression requires further study in randomized
trials. The rate of switching into elevated mood-states was
high with AD-use (25.8%), despite use of mood-stabilizers,
and was seven-times greater than the risk of spontaneous
switching in the depressed BPD patients treated with mood-
stabilizing or antimanic agents but not ADs. This finding does
not seem to support a major protective effect of mood-
stabilizers with ADs, but such reactions were associated with
relatively severe illnesses and the comparison was not based
on randomized treatment-assignment. These conclusions indi-
cate that additional effort is required to identify safe and effec-
tive methods of treating BP-depression, of recognizing and
appropriately treating BPD earlier, to clarify the protective
effect of specific mood-stabilizers and doses against mood-
switching with ADs, and to resolve the question of possible
differences in AD-responses in specific types of mood disorders.
Role of funding source
This study was supported by a Research Fellowship from the University
of Barcelona and the Spanish Foundation of Psychiatry & Mental Health, and
a Josep Font Research Grant from the Hospital Clínic of Barcelona (to JU), by
the Instituto de Salud Carlos III through the Centro para la Investigación Bio-
médica en Red de Salud Mental (CIBERSAM), the Generalitat de Catalunya
(2009-SGR-1022) (to EV), and by a grant from the Bruce J. Anderson founda-
tion and the McLean Private Donors Bipolar Disorder Research Fund (to RJB).
Conflict of interest
Dr. Vieta is a consultant or grant recipient with: Almirall, Astra-Zeneca,
Bristol-Myers-Squibb, Eli Lilly, Forest Research Institute, Gedeon Richter,
Glaxo-Smith-Kline, Janssen-Cilag, Jazz, Lundbeck, Merck, Novartis, Otsuka,
Pfizer, Sanofi, Servier, Schering-Plough, Takeda, and United Biosource Corpo-
rations. No other authors or immediate family members have financial rela-
tionships with pharmaceutical or biomedical corporations that might
represent potential conflicts of interest in this work.
We acknowledge the support of the CIBERSAM.
Baldessarini, R.J., Henk, H.J., Sklar, A.R., Chang, J., Leahy, L.F., 2008. Psychotro-
pic medications for bipolar disorder patients in the United States: poly-
therapy and adherence. Psychiatr. Serv. 59, 1175–1183.
Baldessarini, R.J., Leahy, L.F., Arcona, S., Gause, D., Zhang, W., Hennen, J.,
2007. Prescribing patterns of psychotropic medicines in the United
Multivariate logistic regression model of factors associated with antidepres-
Factor OR [95% CI]
More years ill
Depressive first episode
Melancholic index episode
Lack of familial affective illness
Factors are ranked by significance.
J. Undurraga et al. / Journal of Affective Disorders 139 (2012) 89–93
States for patients diagnosed with bipolar disorders. Psychiatr. Serv. 58, Download full-text
Baldessarini, R.J., Undurraga, J., Vázquez, G.H., Tondo, L., Salvatore, P., Ha, K.,
Khalsa, H.-M.K., Lepri, B., Ha, T.H., Chang, J.S., Tohen, M., Vieta, E., 2012.
Predominant recurrence polarity among 928 adult international bipolar-I
disorder patients. Acta Psychiatr. Scand. 125, 293–302.
Baldessarini, R.J., Tondo, L., Hennen, J., 2003. Treatment-latency and previous
episodes: relationships to pretreatment morbidity and response to
maintenance treatment in bipolar I and II disorders. Bipolar Disord. 5,
Baldessarini, R.J., Vieta, E., Calabrese, J.R., Tohen, M., Bowden, C.L., 2010. Bi-
polar depression: overview and commentary. Harv. Rev. Psychiatry 18,
Bobes, J., Bulbena, A., Luque, A., Dal Re, R., Ballesteros, J., Ibarra, N., 2003.
Comparative psychometric study of the Spanish versions of the Hamil-
ton Depression Rating Scale, with 6, 17, and 21 items. Med. Clin. 120,
Cohn, J.B., Collins, G., Ashbrook, E., Wernicke, J.F., 1989. Comparison of fluox-
etine, imipramine, and placebo in patients with bipolar depressive dis-
order. Int. Clin. Psychopharmacol. 4, 313–322.
Colom, F., Vieta, E., Martinez-Aran, A., Garcia-Garcia, M., Reinares, M.,
Torrent, C., Goikolea, J.M., Banús, S., Salamero, M., 2002. Spanish version
of a scale for the assessment of mania: validity and reliability of the
Young Mania Rating Scale. Med. Clin. 119, 366–371 (Barc.)..
Colom, F., Vieta, E., Daban, C., Pacchiarotti, I., Sánchez-Moreno, J., 2006. Clin-
ical and therapeutic implications of predominant polarity in bipolar dis-
order. J. Affect. Disord. 93, 13–17.
First, M.B., Spitzer, R., Gibbon, M., 1997. Structured Clinical Interview for
DSM-IV Axis-I and Axis-II Disorders. American Psychiatric Press Inc,
Gijsman, H.J., Geddes, J.R., Rendell, J.M., Nolen, W.A., Goodwin, G.M., 2004.
Antidepressants for bipolar depression: systematic review of random-
ized, controlled trials. Am. J. Psychiatry 161, 1537–1547.
Hamilton, M., 1967. Development of a rating scale for primary depressive ill-
ness. Br. J. Soc. Clin. Psychol. 6, 278–296.
McElroy, S.L., Weisler, R.H., Chang, W., Olausson, B., Paulsson, B., Brecher, M.,
Agambaram, V., Merideth, C., Nordenhem, A., Young, A.H., 2010. Double-
blind, placebo-controlled study of quetiapine and paroxetine as mono-
therapy in adults with bipolar depression (EMBOLDEN II). J. Clin. Psychi-
atry 71, 163–174.
Nemeroff, C.B., Evans, D.L., Gyulai, L., Sachs, G.S., Bowden, C.L., Gergel, I.P.,
Oakes, R., Pitts, C.D., 2001. Double-blind, placebo-controlled comparison
of imipramine and paroxetine in the treatment of bipolar depression.
Am. J. Psychiatry 158, 906–912.
Pacchiarotti, I., Valentí, M., Bonnin, C.M., Rosa, A.R., Murru, A., Kotzalidis,
G.D., Nivoli, A.M., Sánchez-Moreno, J., Vieta, E., Colom, F., 2011a. Factors
associated with initial treatment response with antidepressants in bipo-
lar disorder. Eur. Neuropsychopharmacol. 21, 362–369.
Pacchiarotti, I., Valentí, M., Colom, F., Rosa, A.R., Nivoli, A.M., Murru, A.,
Sánchez-Moreno, J., Vieta, E., 2011b. Differential outcome of bipolar pa-
tients receiving antidepressant monotherapy versus combination with
an antimanic drug. J. Affect. Disord. 129, 321–326.
Rosa, A.R., Cruz, N., Franco, C., Haro, J.M., Bertsch, J., Reed, C., Aarre, T.F.,
Sanchez-Moreno, J., Vieta, E., 2010. Why do clinicians maintain
antidepressants in some patients with acute mania? Hints from the Eu-
ropean Mania in Bipolar Longitudinal Evaluation of Medication (EM-
BLEM), a large naturalistic study. J. Clin. Psychiatry 71, 1000–1006.
Rybakowski, J.K., 2012. Bipolarity and inadequate response to antidepressant
drugs: clinical and psychopharmacological perspective. J. Affect. Disord.
Sachs, G.S., Nierenberg, A.A., Calabrese, J.R., Marangell, L.B., Wisniewski, S.R.,
Gyulai, L., Friedman, E.S., Bowden, C.L., Fossey, M.D., Ostacher, M.J.,
Ketter, T.A., Patel, J., Hauser, P., Rapport, D., Martinez, J.M., Allen, M.H.,
Miklowitz, D.J., Otto, M.W., Dennehy, E.B., Thase, M.E., 2007. Effective-
ness of adjunctive antidepressant treatment for bipolar depression.
New Engl. J. Med. 356, 1711–1733.
Sidor, M.M., MacQueen, G.M., 2011. Antidepressants for the acute treatment
of bipolar depression: systematic review and meta-analysis. J. Clin. Psy-
chiatry 72, 156–167.
Tohen, M., Frank, E., Bowden, C.L., Colom, F., Ghaemi, S.N., Yatham, L.N.,
Malhi, G.S., Calabrese, J.R., Nolen, W.A., Vieta, E., Kapczinski, F.,
Goodwin, G.M., Suppes, T., Sachs, G.S., Chengappa, K.R., Grunze, H.,
Mitchell, P.B., Kanba, S., Berk, M., 2009. The International Society for Bi-
polar Disorders (ISBD) Task Force report on the nomenclature of course
and outcome in bipolar disorders. Bipolar Disord. 11, 453–473.
Tohen, M., Vieta, E., Calabrese, J.R., Ketter, T.A., Sachs, G.S., Bowden, C.L.,
Mitchell, P.B., Centorrino, F., Risser, R., Baker, R.W., Evans, A.R., Beymer,
K., Dubé, S., Tollefson, G.D., Breier, A., 2003. Efficacy of olanzapine and
olanzapine–fluoxetine combination in the treatment of bipolar-I depres-
sion. Arch. Gen. Psychiatry 60, 1079–1088.
Tondo, L., Vázquez, G., Baldessarini, R.J., 2010. Mania associated with antide-
pressant treatment: comprehensive meta-analytic review. Acta Psy-
chiatr. Scand. 121, 404–414.
Undurraga, J., Baldessarini, R.J., 2012. Randomized, placebo-controlled trials
of antidepressants for acute major depression: thirty-year meta-analytic
review. Neuropsychopharmacology 37, 851–864.
Valentí, M., Pacchiarotti, I., Bonnín, C.M., Rosa, A., Popovic, D., Nivoli, A., Goiko-
antidepressant-related switch to mania. J. Clin. Psychiatry.
Valentí, M., Pacchiarotti, I., Rosa, A.R., Bonnín, C.M., Popovic, D., Nivoli, A.M.,
Murru, A., Grande, I., Colom, F., Vieta, E., 2011. Bipolar mixed episodes
and antidepressants: a cohort study of bipolar I disorder patients. Bipo-
lar Disord. 13, 145–154.
Vázquez, G., Tondo, L., Baldessarini, R.J., 2011. Comparison of antidepressant
responses in patients with bipolar vs. unipolar depression: meta-
analytic review. Pharmacopsychiatry 44, 21–26.
Vieta, E., 2011. Bipolar units and programmes: are they really needed?
World Psychiatry. 10, 152.
Yatham, L.N., Kennedy, S.H., Schaffer, A., Parikh, S.V., Beaulieu, S., O'Donovan,
C., MacQueen, G., McIntyre, R.S., Sharma, V., Ravindran, A., Young, L.T.,
Young, A.H., Alda, M., Milev, R., Vieta, E., Calabrese, J.R., Berk, M., Ha, K.,
Kapczinski, F., 2009. Canadian Network for Mood and Anxiety Treat-
ments (CANMAT) and International Society for Bipolar Disorders
(ISBD) collaborative update of CANMAT guidelines for the management
of patients with bipolar disorder: update 2009. Bipolar Disord. 11,
Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. Rating scale for mania:
reliability, validity and sensitivity. Br. J. Psychiatry 133, 429–435.
J. Undurraga et al. / Journal of Affective Disorders 139 (2012) 89–93