Young et al

A Double-Blind, Placebo-Controlled Study

of Quetiapine and Lithium Monotherapy in Adults
in the Acute Phase of Bipolar Depression (EMBOLDEN I)
Allan H. Young, FRCPsych; Susan L. McElroy, MD;
Michael Bauer, MD, PhD; Nabil Philips, MD; William Chang, PhD;
Bengt Olausson, MD; Björn Paulsson, MD; and Martin Brecher, MD;
for the EMBOLDEN I (Trial 001) Investigators

Submitted: December 9, 2008; accepted October 27, 2009.

Online ahead of print: January 26, 2010 (doi:10.4088/JCP.08m04995gre).
Corresponding author: Allan H. Young, FRCPsych, Department of
Objective: The aim of this study was to compare Psychiatry, University of British Columbia, Ste 430–5950 University Blvd,
the efficacy and tolerability of quetiapine and lith- Vancouver, British Columbia BC V6T 1Z3, Canada ([email protected]).
ium monotherapy with that of placebo for a major

T
depressive episode in bipolar disorder.
Method: 802 patients with DSM-IV–defined
bipolar disorder (499 bipolar I, 303 bipolar II) he impact of mental health–related disorders on the
were randomly allocated to quetiapine 300 mg/d global burden of disease is substantial. Over a quar-
(n = 265), quetiapine 600 mg/d (n = 268), lithium ter of the disease burden worldwide is attributed to mental
600 to 1800 mg/d (n = 136), or placebo (n = 133) disorders and other neuropsychiatric conditions compared
for 8 weeks. Primary endpoint was the change with 22% for cardiovascular disease and 11% for cancer.1
in Montgomery-Asberg Depression Rating Scale
(MADRS) total score. The study was conducted Bipolar disorder, and in particular the depressive phase,
from August 2005 to May 2007. ranks high among the major contributors to the greater
Results: Mean MADRS total score change from disability associated with mental disorders.1
baseline at week 8 was –15.4 for quetiapine 300 Depression dominates the course of bipolar disorder and
mg/d, –16.1 for quetiapine 600 mg/d, –13.6 for is associated with significant morbidity and mortality.2,3
lithium, and –11.8 for placebo (P < .001 for both
quetiapine doses, P = .123 for lithium, vs placebo). Nonetheless, the treatment of acute depressive episodes
Quetiapine 600 mg/d was significantly more ef- in bipolar disorder remains understudied and controversial.
fective than lithium in improving MADRS total Most treatment guidelines for bipolar disorder advocate
score at week 8 (P = .013). Quetiapine-treated (both first-line monotherapy with conventional “mood stabiliz-
doses), but not lithium-treated, patients showed ers,” especially lithium, for mild to moderate episodes of
significant improvements (P < .05) in MADRS
response and remission rates, Hamilton Depression depression.3–8 However, the empirical evidence supporting
Rating Scale (HDRS), Clinical Global Impressions- the antidepressant efficacy of lithium is currently limited.
Bipolar-Severity of Illness and -Change, and Although historically lithium has demonstrated puta-
Hamilton Anxiety Rating Scale (HARS) scores at tive antidepressant efficacy in older studies, in more recent
week 8 versus placebo. Both quetiapine doses were stringent investigations, the response to lithium of patients
more effective than lithium at week 8 on the HDRS
and HARS. The most common adverse events were with bipolar depression has been more modest.9,10 More-
somnolence, dry mouth, and dizziness with quetia- over, achievement of an antidepressant effect with lithium
pine (both doses) and nausea with lithium. may take several weeks,11 which is a notable disadvantage
Conclusions: Quetiapine (300 or 600 mg/d) in the acute treatment setting. Traditional antidepressants
was more effective than placebo for the treatment (such as monoamine oxidase inhibitors, selective sero-
of episodes of acute depression in bipolar disorder.
Lithium did not significantly differ from placebo tonin reuptake inhibitors, and tricyclic antidepressants),
on the main measures of efficacy. Both treatments well-established treatments for major depression, are also
were generally well tolerated. used alone or in combination with mood stabilizers for the
Trial Registration: clinicaltrials.gov Identifier: treatment of bipolar depression. The use of antidepressants,
NCT00206141 however, has been the focus of controversy because of the
J Clin Psychiatry 2010;71(2):150–162
© Copyright 2010 Physicians Postgraduate Press, Inc.
potential risk of switching to mania or hypomania during
treatment.12–14 Alternative treatments such as lamotrigine
have also been investigated. The efficacy of lamotrigine as
See companion EMBOLDEN II article on page 163. maintenance treatment for bipolar disorder, particularly
for delaying depressive episodes, is well established, but

150 J Clin Psychiatry 71:2, February 2010

 Quetiapine vs Lithium vs Placebo in Bipolar Depression

its efficacy in the acute treatment of bipolar depression is conducted at 110 centers throughout Europe, Canada, and
less clear.15 Asia, conformed to the current amendment of the Dec-
More recently, second-generation antipsychotics have laration of Helsinki and the International Conference on
been shown to be effective in acute bipolar disorder. The Harmonization/Good Clinical Practice guidelines. Approval
olanzapine-fluoxetine combination (OFC) and quetiapine for the study was obtained from either a central institutional
have each received registrations in the United States for review board or a review board at the study site. All patients
acute bipolar depression. The efficacy of OFC or olanzapine provided written informed consent prior to participation in
monotherapy versus placebo was investigated in an 8-week the study. The first patient was enrolled in August 2005, and
study of 833 patients with acute bipolar I depression.16 the last patient completed the study in May 2007.
Change from baseline in Montgomery-Asberg Depression Once enrolled, patients underwent a washout period
Rating Scale (MADRS)17 total score indicated significantly of at least 5 to 28 days, during which prior psychotropic
greater symptom improvements with olanzapine and OFC medications were discontinued. Patients were subsequently
compared with placebo at study end point. Additionally, randomly assigned to receive acute treatment with quetiapine
OFC demonstrated greater reductions in MADRS total (300 or 600 mg/d), lithium (600–1,800 mg/d), or placebo
score than olanzapine from week 4 onward. The efficacy for 8 weeks. Eligible patients, those with both MADRS and
of quetiapine monotherapy (300 mg/d and 600 mg/d) in Young Mania Rating Scale (YMRS)22 total scores of ≤ 12
bipolar depression was initially demonstrated in 2 large, at the end of the acute treatment phase, could then enter a
8-week studies (BipOLar DEpRession [BOLDER] I and 26- to 52-week continuation treatment phase with quetia-
II).18,19 In these studies, quetiapine (both doses) signifi- pine (300 or 600 mg/d) versus placebo. Continuation-phase
cantly improved symptoms of depression compared with data will be published separately.
placebo at week 8, as demonstrated by a greater reduction
in MADRS total score from baseline. Furthermore, a sig- Patient Population
nificantly higher proportion of quetiapine-treated patients Adult outpatients (aged 18–65 years) meeting
met criteria for response and remission than patients in the DSM-IV criteria for bipolar I or II disorder (with or with-
placebo group. Notably, quetiapine was not associated out rapid cycling; ≥ 4 episodes/year to ≤ 8 per year) and
with an increased risk of treatment-emergent mania or who were experiencing a recent major depressive episode
hypomania compared with placebo. Recent guideline (duration ≤ 1 year and onset ≥ 4 weeks) prior to entry were
recommendations for first- and second-line treatment of eligible for inclusion in the study. Patients were also re-
acute bipolar depression reflect these clinical trial data and quired to demonstrate a Hamilton Depression Rating Scale
registrations.20,21 (HDRS)23 score ≥ 20 and an HDRS item 1 (depressed mood)
Given the limited evidence for effective treatments for score ≥ 2.
bipolar depression, emerging results from monotherapy tri- Key exclusion criteria were (1) active Axis I disorders
als such as BOLDER are promising. However, head-to-head requiring treatment within 6 months of study entry; (2)
comparisons between established and newer treatments are a YMRS total score > 12; (3) a history of nonresponse to
lacking, which hinders the identification of a gold-standard an adequate treatment period (6 weeks) with ≥ 2 classes
treatment for acute bipolar depression. The present study of antidepressants during the current episode; (4) known
(Efficacy of Monotherapy Seroquel in BipOLar DEpressioN nonresponse to quetiapine or lithium, as judged by the in-
I [EMBOLDEN I]) is one of 2 large similarly designed stud- vestigator; (5) substance dependence (DSM-IV) or abuse;
ies (a total of 1,542 patients randomized) that compared (6) a current serious suicidal or homicidal risk (as judged
the efficacy and tolerability of quetiapine monotherapy with by the investigator); and (7) a clinically relevant medical
that of placebo for the acute treatment (8 weeks) of bipolar illness.
I and II disorder (most recent episode depression) followed Treatment randomization, using an interactive re-
by a 26- to 52-week continuation treatment phase. The stud- sponse system, was stratified by bipolar diagnosis (I or II)
ies also included lithium (EMBOLDEN I) and paroxetine in order to achieve an approximate balance between the
(EMBOLDEN II; see companion article46) comparator arms. 2 bipolar subtypes. Within each stratum, patients enter-
Results from the acute treatment phase of EMBOLDEN I ing the acute treatment phase were randomly assigned in
are presented here. a 2:2:1:1 ratio to receive quetiapine 300 mg/d, quetiapine
600 mg/d, lithium 600 to 1,800 mg/d, or placebo. The ran-
METHOD domization was centralized, and randomization numbers
were not sequential within a site. No member of the inves-
This study was an 8-week, multicenter, randomized, tigational team had access to the randomization scheme
double-blind, parallel-group, placebo-controlled, fixed- during the conduct of the study. To ensure that study par-
dose trial assessing the efficacy and tolerability of quetiapine ticipants and study investigators were blinded to treatment
and lithium monotherapy for episodes of major depres- allocation, all medication packaging was identical, with ac-
sion in patients with bipolar I and II disorder. The study, tive tablets identical in size, color, smell, and taste to the

J Clin Psychiatry 71:2, February 2010 151

Young et al

placebo tablets. A double-dummy method was employed, and examination, and vital signs. In addition, the proportion
the number of tablets dispensed was identical across all of patients with treatment-emergent mania or hypomania
treatment arms. (defined as a YMRS total score ≥ 16 on 2 consecutive
assessments or at final assessment, or an AE report of treatment-
Study Treatments emergent mania or hypomania) or treatment-emergent
Quetiapine or matching placebo was administered orally suicidal ideation (incidence of patients with HDRS item 3
once a day at bedtime. Quetiapine was initiated at a dose of [suicide] score ≥ 3 or an AE of suicidality, suicidal ideation,
50 mg/d and then increased to achieve a target dose of 300 suicide attempts, or suicide completion) was assessed.
mg/d by day 4 or 600 mg/d by day 8. Lithium or matching
placebo was administered twice daily (morning and bed- Statistical Analysis
time). Lithium was initiated at 600 mg/d and subsequently Efficacy analyses were conducted in the intent-to-treat
increased to 900 mg/d from day 4 until day 8. Lithium was (ITT) population (patients who received at least 1 dose of
dosed thereafter in a blinded manner at 600 to 1,800 mg/d study medication and had at least 1 postbaseline efficacy
to maintain a serum lithium concentration between 0.6 and assessment) using last-observation-carried-forward meth-
1.2 mEq/L. odology to deal with missing data from patient dropout in
the efficacy analyses. Secondary analyses were conducted in
Prior and Concomitant Medication the per protocol (PP) population (patients who completed
During the study, continuation of nonpsychotropic study treatment without major protocol violations or de-
medications taken prior to study entry was permitted. viations affecting efficacy) to test the robustness of the ITT
Lorazepam (1–3 mg/d for severe anxiety) and hypnotics findings. Patients with a median serum lithium concentra-
(zolpidem tartrate up to 10 mg/d, zaleplon up to 20 mg/d, tion below 0.6 or above 1.2 mEq/L were excluded from the
zopiclone up to 7.5 mg/d, or chloral hydrate up to 1 g/d at PP population. It was estimated that sample sizes would pro-
bedtime for insomnia) were allowed during the first 3 weeks vide 87% power to detect 4 points difference between each
of treatment at the investigator’s discretion. Concomitant quetiapine dose and placebo in MADRS total score change
treatment with all other psychotropic drugs was prohibited from baseline to week 8 with a pooled standard deviation
during the study. of 10 using a 2-sided test at an α level of .025 (Bonferroni
corrected). No power calculations were made for a non-
Efficacy Assessments inferiority comparison between quetiapine and lithium
The primary efficacy outcome measure was the change treatment groups.
from baseline to week 8 in MADRS total score. Secondary The primary efficacy outcome measure of change from
efficacy outcome measures included response (≥ 50% reduc- baseline in MADRS total score tested the efficacy of que-
tion in MADRS total score) and remission (MADRS total tiapine 300 or 600 mg/d versus placebo using analysis of
score ≤ 12) rates and Clinical Global Impressions-Bipolar- covariance (ANCOVA) with baseline MADRS total score
Change24 response at week 8. Additional efficacy measures as the covariate, treatment and diagnosis strata as fixed ef-
were change from baseline to week 8 in MADRS individual fects, and country as a random effect in the model. A similar
items, MADRS item 10 (suicidal thoughts), HDRS total ANCOVA model was also used for other continuous efficacy
score, HDRS item 1 (depressed mood), Clinical Global measures. MADRS response and remission were analyzed
Impressions-Bipolar-Severity of Illness (CGI-BP-S),24 and using Cochran-Mantel-Haenszel test stratified by bipolar
the Hamilton Anxiety Rating Scale (HARS).25 Patient- diagnosis. The number needed to treat (NNT) in order to
reported changes from baseline in the Sheehan Disability achieve response and remission outcomes was also calculat-
Scale (SDS)26 and the Medical Outcomes Study Cognitive ed for each active treatment group versus placebo. Statistical
Scale (MOS-Cog)27 were used to evaluate treatment effects analysis of safety end points was not planned in the study
on social, occupational, and cognitive functioning. Assess- protocol. Analyses of safety variables, including AEs, are
ments of efficacy were performed at baseline, at weeks 1 and presented descriptively. The safety population comprised pa-
2, and then every 2 weeks until week 8, except for SDS and tients who received at least 1 dose of the study medication.
MOS-Cog, which were assessed at baseline and at weeks 4
and 8 only. RESULTS

Safety and Tolerability Assessments Patients

Safety and tolerability assessments included adverse A total of 802 patients with bipolar I or II disorder expe-
events (AEs; incidence and severity), withdrawals due to riencing an episode of depression were randomly allocated
AEs, and extrapyramidal symptoms (EPS; measured using to treatment with quetiapine 300 mg/d (n = 265), quetiapine
the Simpson Angus Scale [SAS]28 and the Barnes Akathisia 600 mg/d (n = 268), lithium 600 to 1,800 mg/d (n = 136),
Rating Scale [BARS]29), as well as laboratory tests, weight or placebo (n = 133) for 8 weeks (Figure 1). Of these, 794
and body mass index, electrocardiogram (ECG), physical patients were included in the safety population, and 783

152 J Clin Psychiatry 71:2, February 2010

 Quetiapine vs Lithium vs Placebo in Bipolar Depression

Figure 1. Patient Disposition in the Acute Treatment Phase

1042 Patients screened

802 Patients randomized

Quetiapine Quetiapine
Placebo Lithium
300 mg/d 600 mg/d
(n = 133) (n = 136)
(n = 265) (n = 268)

Discontinued 65 Discontinued 63 Discontinued 37 Discontinued 34

Not eligible 4 Not eligible 1 Not eligible 2 Not eligible 1
Adverse event 26 Adverse event 35 Adverse event 11 Adverse event 12
Lost to follow-up 1 Lost to follow-up 3 Lost to follow-up 0 Lost to follow-up 0
Protocol noncompliance 0 Protocol noncompliance 4 Protocol noncompliance 3 Protocol noncompliance 2
Informed consent 20 Informed consent 17 Informed consent 13 Informed consent 9
withdrawn withdrawn withdrawn withdrawn
Lack of efficacy 11 Lack of efficacy 2 Lack of efficacy 7 Lack of efficacy 10
Other 3 Other 1 Other 1 Other 0

200 Patients 205 Patients 96 Patients 102 Patients

completed study completed study completed study completed study

Table 1. Baseline Patient Demographics and Disease

Similar proportions of patients completed the acute
Characteristics (ITT; acute treatment phase) treatment phase of the study (75.5%, 76.5%, and 75.0% in
Quetiapine Quetiapine the quetiapine 300 mg/d, quetiapine 600 mg/d, and lithium
300 mg/d 600 mg/d Placebo Lithium groups, respectively, vs 72.2% with placebo). Patient dispo-
Characteristic (n = 255) (n = 263) (n = 129) (n = 136)
sition is shown in Figure 1.
Sex, %
Men 42.7 36.5 45.7 40.4
Women 57.3 63.5 54.3 59.6 Efficacy
Age, mean, y 42.3 42.8 41.5 41.4 MADRS. Starting at week 1 (first assessment) and con-
Weight, mean, kg 75.8 74.4 76.0 77.0
Bipolar disorder type, % tinuing through week 8, both doses of quetiapine (300
Bipolar I 62.7 61.6 60.5 64.0 and 600 mg/d) were significantly better than placebo at
Bipolar II 37.3 38.4 39.5 36.0 improving MADRS total scores in the ITT population
Rapid-cycling course 6.3 6.1 3.9 5.9
  (≥ 4 mood episodes (P < .05; Figure 2A, Table 2). Lithium-treated patients dem-
  in past year), % onstrated numerically greater but not statistically significant
Non-rapid-cycling 93.7 93.9 96.1 94.1 improvement in MADRS total score compared with pla-
course, %
MADRS total score, 28.1 28.3 28.5 28.3 cebo throughout 8 weeks of treatment (Figure 2A; at week
mean 8, –13.6; P = .123). Further post hoc analysis of the lithium
HDRS total score, mean 24.2 24.3 24.4 24.1 treatment group, in which patients with median serum lith-
YMRS total score, mean 3.1 3.3 3.3 3.4
HARS total score, mean 18.3 18.2 18.3 18.0 ium concentrations ≥ 0.8 mEq/L were analyzed separately
CGI-BP-S score, mean 4.4 4.3 4.3 4.4 (n = 34; ITT population), revealed no difference from place-
Abbreviations: CGI-BP-S = Clinical Global Impressions-Bipolar- bo in terms of the improvement in MADRS total score from
Severity of Illness, HARS = Hamilton Anxiety Rating Scale, baseline (difference of –2.76 points at week 8; P = .128).
HDRS = Hamilton Depression Rating Scale, ITT = intent-to-
treat, MADRS = Montgomery-Asberg Depression Rating Scale, Quetiapine 600 mg/d (but not 300 mg/d) significantly
YMRS = Young Mania Rating Scale. improved MADRS total score compared with lithium
from day 8 through week 8 (difference of –2.49 points at
and 704 patients were included in the efficacy ITT and PP week 8, P = .013).
populations, respectively. Baseline characteristics did not Similar results were observed in the PP population. In the
markedly differ between treatment groups (Table 1). In the PP population, throughout 8 weeks of treatment, the magni-
overall population, the mean age was 42.2 years, and the tude of improvements in MADRS total score for quetiapine
majority of patients were women (59.3%). The mean me- 300 mg/d (–15.7; P < .001) and 600 mg/d (–16.5; P < .001)
dian daily dose and serum concentration of lithium in the was greater compared with placebo (–12.1). Improvements
ITT population were 981 mg and 0.61 mEq/L, respectively. in MADRS total scores for lithium-treated patients (n = 87)
In total, 83 (64.4%) patients in the lithium group achieved in the PP population (mean median lithium concentration
median serum concentrations of lithium that were in the 0.7 mEq/L) did not differ significantly from placebo-treated
target serum concentration range of 0.6 to 1.2 mEq/L, while patients (n = 126) from day 8 through study end (at week 8,
45 (34.9%) patients had median serum concentrations of –13.3; P = .349). Significant improvements in MADRS total
lithium below 0.6 mEq/L. scores were observed for quetiapine 300 mg/d (n = 241) and

J Clin Psychiatry 71:2, February 2010 153

Young et al

Figure 2. Mean Change From Baseline in MADRS Total Score

600 mg/d (n = 250) when compared with lithium at week
in Patients With Bipolar I or II Disorder (ITT; LOCF; acute 8 (300 mg/d: difference of –2.35 points, 95% CI, –4.66 to
treatment phase) –0.04; 600 mg/d: difference of –3.16 points, 95% CI, –5.46
to –0.86).
In patients with bipolar I disorder, quetiapine treatment
A. All Patients
(both doses) significantly improved MADRS total scores
Study Week compared with placebo (difference of –3.61 points at week
0 1 2 4 6 8 8 for quetiapine 300 mg/d [n = 160], P = .006; difference of
0 Quetiapine 300 mg/d (n = 255) –5.28 points for quetiapine 600 mg/d [n = 162], P < .001)
Quetiapine 600 mg/d (n = 263) (Figure 2B). In patients with bipolar II disorder, patients
LS Mean Change From Baseline

Placebo (n = 129)
–5 *
Lithium (n = 136) treated with quetiapine (both doses) showed numerically
‡ greater improvements in MADRS total scores at week 8
* (300 mg/d [n = 95]: difference of –3.19 points, P = .051; 600
–10 † † mg/d [n = 101]: difference of –2.43 points, P = .131) (Figure
‡ * 2C). Patients without a rapid-cycling disease course showed
–15 † ‡ significantly greater improvement over placebo (n = 128)
‡ at week 8 in MADRS score when treated with quetiapine
300 mg/d (difference, –3.80; P < .001 [n = 239]) or quetia-
–20
pine 600 mg/d (difference, –4.56; P < .001 [n = 247]), but not
with lithium (difference, –2.15; P = .066 [n = 124]). For the
B. Bipolar I relatively few patients with a rapid-cycling disease course
Study Week
(n = 45), the difference in change from baseline to week
8 in MADRS total score compared with placebo was 2.82
0 1 2 4 6 8
(P = .607) for quetiapine 300 mg/d, 0.76 (P = .889) for que-
0 Quetiapine 300 mg/d (n = 160) tiapine 600 mg/d, and 6.92 (P = .275) for lithium.
Quetiapine 600 mg/d (n = 162)
LS Mean Change From Baseline

Placebo (n = 78) At end point (week 8), significantly more patients treat-
Lithium (n = 87)
–5 ed with quetiapine met response criteria than those who
*
received placebo: 68.6% for quetiapine 300 mg/d (P < .05)
–10
and 69.6% for quetiapine 600 mg/d (P < .01) versus 55.8%
* for placebo, with NNTs of 8 and 7 for quetiapine 300 mg/d
* and quetiapine 600 mg/d, respectively. Patients treated
*
–15
* with lithium demonstrated a numerically greater response
* (62.5%) than those receiving placebo (NNT = 15), but the
–20 difference was not statistically significant (P = .279). At end
point (week 8), significantly more patients met remission
criteria in the quetiapine 300 mg/d (69.8%) and quetia-
C. Bipolar II pine 600 mg/d (70.3%) groups than in the placebo group
Study Week (55.0%; P < .01 both doses), while 62.5% of patients in the
0 1 2 4 6 8 lithium group met remission criteria (P = .228 vs placebo).
0 The corresponding NNTs were 7, 7, and 13 for quetiapine
Quetiapine 300 mg/d (n = 95)
Quetiapine 600 mg/d (n = 101) 300 mg/d, quetiapine 600 mg/d, and lithium, respectively.
LS Mean Change From Baseline

*
*
Placebo (n = 51)
Lithium (n = 49)
Of those patients meeting remission criteria, the following
–5
proportions had MADRS total scores < 24 at baseline: 28.2%
of the placebo group; 27.5% and 23.2% of the quetiapine
*
–10 300 mg/d and 600 mg/d groups, respectively; and 24.7% of
* the lithium group.
–15
By week 8, there were significant improvements com-
pared with placebo in 7 of 10 individual MADRS items with
quetiapine 300 mg/d (P < .05) and 9 of 10 items with que-
–20 tiapine 600 mg/d (P < .05); only 2 items (inner tension and
reduced sleep) were significantly improved following treat-
ment with lithium (P < .05) (Figure 3). At week 8, quetiapine
*P < .05, †P < .01, ‡P < .001 vs placebo.
Abbreviations: ITT = intent-to-treat, LOCF = last observation carried 600 mg/d was significantly better than lithium in improv-
forward, LS = least squares, MADRS = Montgomery-Asberg Depression ing the following individual MADRS item scores: apparent
Rating Scale. sadness (difference –0.42 points, P = .003), reported sadness

154 J Clin Psychiatry 71:2, February 2010

 Quetiapine vs Lithium vs Placebo in Bipolar Depression

Table 2. Mean Change From Baseline in Primary and Secondary Outcome Measures (ITT population; LOCF; acute treatment phase)
Analysis
Change in Score at End (comparison with placebo)
Baseline Score of Acute Treatment Phase, ANCOVA,
Measure and Treatment Mean SE LS Mean (SE) LS Mean (SE) P Value
Montgomery-Asberg Depression Rating Scale
Quetiapine 300 mg/d, n = 255 28.1 0.39 −15.36 (0.93) −3.55 (1.02) < .001
Quetiapine 600 mg/d, n = 263 28.3 0.40 −16.10 (0.92) −4.29 (1.02) < .001
Placebo, n = 129 28.5 0.54 −11.81 (1.10)
Lithium, n = 136 28.3 0.48 −13.60 (1.08) −1.79 (1.16) .123
Hamilton Depression Rating Scale
Quetiapine 300 mg/d, n = 255 24.2 0.22 −13.98 (0.78) −3.26 (0.83) < .001
Quetiapine 600 mg/d, n = 263 24.3 0.21 −14.17 (0.77) −3.45 (0.82) < .001
Placebo, n = 129 24.4 0.28 −10.72 (0.91)
Lithium, n = 136 24.1 0.28 −12.36 (0.90) −1.64 (0.94) .082
Hamilton Depression Rating Scale item 1 (depressed mood)
Quetiapine 300 mg/d, n = 255 2.9 0.03 −1.52 (0.10) −0.26 (0.11) .023
Quetiapine 600 mg/d, n = 263 3.0 0.04 −1.62 (0.10) −0.36 (0.11) .001
Placebo, n = 129 3.0 0.05 −1.26 (0.12)
Lithium, n = 136 2.9 0.05 −1.36 (0.12) −0.10 (0.13) .438
Clinical Global Impressions-Bipolar-Severity of Illness
Quetiapine 300 mg/d, n = 255 4.4 0.05 −1.51 (0.13) −0.37 (0.14) .008
Quetiapine 600 mg/d, n = 263 4.3 0.04 −1.57 (0.13) −0.43 (0.14) .002
Placebo, n = 129 4.3 0.07 −1.14 (0.15)
Lithium, n = 135 4.4 0.07 −1.40 (0.15) −0.26 (0.16) .098
Hamilton Anxiety Rating Scale
Quetiapine 300 mg/d, n = 255 18.3 0.40 −9.14 (0.63) −2.60 (0.71) < .001
Quetiapine 600 mg/d, n = 263 18.2 0.37 −9.29 (0.63) −2.75 (0.70) < .001
Placebo, n = 129 18.3 0.51 −6.55 (0.75)
Lithium, n = 136 18.0 0.47 −7.72 (0.74) −1.17 (0.80) .144
Sheehan Disability Scale
Quetiapine 300 mg/d, n = 244 19.3 0.33 −6.90 (0.78) −1.57 (0.77) .041
Quetiapine 600 mg/d, n = 248 18.7 0.35 −7.54 (0.78) −2.21 (0.76) .004
Placebo, n = 124 18.3 0.51 −5.33 (0.89)
Lithium, n = 131 19.4 0.45 −7.00 (0.88) −1.67 (0.87) .056
Medical Outcomes Study Cognitive scale
Quetiapine 300 mg/d, n = 244 18.8 0.31 5.67 (0.55) 1.03 (0.66) .120
Quetiapine 600 mg/d, n = 248 18.7 0.33 6.34 (0.55) 1.70 (0.66) .010
Placebo, n = 124 19.3 0.46 4.64 (0.66)
Lithium, n = 131 19.5 0.49 5.98 (0.65) 1.34 (0.75) .075
Abbreviations: ANCOVA = analysis of covariance, ITT = intent-to-treat, LOCF = last observation carried forward, LS = least squares.

(difference, –0.37; P = .008), reduced sleep (difference, –0.42; placebo (P < .05) and lithium at weeks 4, 6, and 8 (difference,
P = .004), and inability to feel (difference, –0.34; P = .011); –0.26; P = .018 at end point). Quetiapine 300 mg/d but not
quetiapine 300 mg/d was significantly better than lithium lithium was significantly better than placebo at improving
in improving MADRS item 4 (reduced sleep) score (differ- the depressed mood item at week 8 (P < .05).
ence, –0.48; P = .001). Improvement in suicidal thoughts was CGI-BP. At week 8, compared with patients receiving
significantly greater with quetiapine 600 mg/d (P < .05) than placebo, quetiapine-treated patients demonstrated signifi-
placebo from week 4 to week 8; quetiapine 300 mg/d and cantly greater improvement in mean CGI-BP-S total score,
lithium showed numerical but not statistical improvement whereas lithium-treated patients showed nonsignificant im-
over placebo during the 8 weeks of treatment. provement (Table 2). At week 8, significantly more patients
HDRS. Patients receiving quetiapine (both doses) treated with quetiapine 300 or 600 mg/d were “much im-
showed significantly greater improvements in HDRS total proved” or “very much improved” compared with placebo
score compared with placebo at every assessment (from (64.7% and 61.6%, respectively, vs 48.1%; P < .05 both doses);
week 1−week 8; P < .001 at end point) (Table 2). Lithium- lithium treatment resulted in numerical but not significant
treated patients did not show significant improvements improvement (51.1%; P = .631) compared with placebo.
versus placebo at any time during the acute treatment phase. HARS. Both doses of quetiapine were associated with
At week 8, both doses of quetiapine were significantly better significant (P < .05) improvement in HARS scores compared
than lithium in improving the HDRS total score (difference, with placebo from week 1 through week 8, while lithium
–1.62; P = .047 for quetiapine 300 mg/d; difference, –1.81; was associated with only numerical improvements in this
P = .026 for quetiapine 600 mg/d). Patients receiving quetia- measure (Table 2).
pine 600 mg/d also had significantly greater improvements Functional improvement: SDS and MOS-Cog. At
in the HDRS item 1 (depressed mood) score compared with week 8, quetiapine 600 mg/d demonstrated a significant

J Clin Psychiatry 71:2, February 2010 155

Young et al

Figure 3. Mean Difference From Placebo in Change From

The incidence of treatment-emergent mania was low in
Baseline to Week 8 in MADRS Individual Item Scores in all treatment groups, but higher in the quetiapine and lithi-
Patients With Bipolar I or II Disorder (ITT; LOCF; acute um groups compared with placebo: 4.2% for quetiapine 300
treatment phase) mg/d, 2.2% for quetiapine 600 mg/d, and 2.2% for lithium,
Quetiapine 300 mg Quetiapine 600 mg Lithium (n = 136) versus 0.8% for placebo. A small proportion of patients in
(n = 255) (n = 263)
each treatment group demonstrated treatment-emergent
suicidal ideation (incidence of patients with HDRS item 3
Apparent Sadness ‡
* [suicide] score ≥ 3 or an AE of suicidality, suicidal ideation,
suicide attempt, or suicide completion), which was highest
Reported Sadness ‡
* in the placebo group (2.3%) compared with the quetia-
* pine 300 and 600 mg/d (1.9% and 1.1%, respectively) and
Inner Tension ‡
‡ lithium (0.7%) groups. Adverse events potentially related
* to EPS (including the MedDRA terms akathisia, hypokine-
Reduced Sleep ‡
‡ sia, restlessness, tremor, dyskinesia, extrapyramidal disorder,
psychomotor hyperactivity, hyperkinesia, hypertonia, muscle
Reduced Appetite †
† rigidity, and nuchal rigidity) were reported by 5.0%, 7.5%,
3.8%, and 8.1% of patients treated with quetiapine 300 mg/d,
Concentration Difficulties
quetiapine 600 mg/d, placebo, and lithium, respectively.
The mean change in SAS total score from baseline to
Lassitude *
week 8 was similar between treatment groups (–0.1 for
quetiapine 300 mg/d, 0.0 for quetiapine 600 mg/d, –0.2 for
Inability to Feel ‡
† placebo, and –0.1 for lithium). There were minimal mean
changes in BARS scores from baseline at week 8 in all treat-
Pessimistic Thoughts †
† ment groups (0 for both quetiapine 300 and 600 mg/d, –0.1
for placebo, and 0 for lithium).
Suicidal Thoughts *
Laboratory results and vital signs. There were no
0 10 20 30 40 50 60 70 80 90 100
clinically relevant differences between treatments in ECG
Percent (%) Improvement measurements or vital signs. Mean changes from baseline in
*P < .05, †P < .01, ‡P < .001 vs placebo. Values based on change from weight and clinical laboratory measures (glucose, lipid, and
baseline LOCF analyses. prolactin parameters) and the proportion of patients dem-
Abbreviations: ITT = intent-to-treat, LOCF = last observation carried onstrating clinically relevant changes in lipid and glucose
forward, MADRS = Montgomery-Asberg Depression Rating Scale.
variables were generally similar across the treatment groups
(Table 4), except that the treatment groups differed in terms
of effects on triglyceride levels. Triglycerides increased with
improvement over placebo in the SDS total score (P < .01) quetiapine 300 and 600 mg/d, but decreased with placebo
and the MOS-Cog total score (P = .01) (Table 2). Quetia- and with lithium. In addition, all treatments were associ-
pine 300 mg/d demonstrated a significant improvement ated with a decrease in prolactin levels at week 8 (Table 4).
in SDS total score (P < .05), but only a numerical improve- However, a higher proportion of patients treated with que-
ment in MOS-Cog total score at week 8. Lithium treatment tiapine 300 mg/d (10.2%) demonstrated clinically relevant
was associated with numerical but statistically insignifi- increases in prolactin levels (> 20 µg/L in men or > 30 µg/L
cant improvements in SDS and MOS-Cog total scores than in women) than patients treated with quetiapine 600 mg/d
placebo. (3.3%), placebo (4.0%), or lithium (2.8%).

Safety and Tolerability DISCUSSION

Adverse events. The proportion of patients discontinu-
ing due to AEs was 10.4% and 13.9% in the quetiapine 300 This study (EMBOLDEN I) and the similarly designed
and 600 mg/d groups, 8.4% in the placebo group, and 8.8% EMBOLDEN II study represent 2 of the largest placebo-
in the lithium group. The incidence of serious AEs was low controlled studies to date of acute treatment in patients
and similarly distributed across the treatment groups (3.8% with bipolar I and II disorder. EMBOLDEN I replicates the
and 2.6% for quetiapine 300 and 600 mg/d, 2.3% for place- antidepressant efficacy of quetiapine 300 and 600 mg/d, as
bo, and 2.2% for lithium). The majority of AEs experienced initially demonstrated in the BOLDER studies.18,19 Quetia-
by patients were mild to moderate in severity, with the most pine treatment was associated with significant improvements
common AEs (occurring in > 10% of patients) reported as in primary and secondary measures of efficacy compared
somnolence, dry mouth, and dizziness with quetiapine with placebo. Improvements in symptoms reached signifi-
(both doses) and nausea with lithium (Table 3). cance from week 1 of treatment and were maintained up to

156 J Clin Psychiatry 71:2, February 2010

 Quetiapine vs Lithium vs Placebo in Bipolar Depression

Table 3. Adverse Events (≥5% in any group; safety population; acute treatment phase)a
Quetiapine 300 mg/d (n = 260) Quetiapine 600 mg/d (n = 267) Lithium (n = 136) Placebo
P Value P Value P Value (n = 131)
Adverse Event n % (vs Placebo) n % (vs Placebo) n % (vs Placebo) n %
Somnolence 47 18.1 < .001 47 17.6 < .001 12 8.8 .132 5 3.8
Dry mouth 37 14.2 < .001 40 15.0 < .001 10 7.4 .035 2 1.5
Dizziness 25 9.6 .174 30 11.2 .066 6 4.4 .782 7 5.3
Headache 19 7.3 .045 23 8.6 .118 13 9.6 .341 18 13.7
Sedation 16 6.2 .042 14 5.2 .103 1 0.7 .617 2 1.5
Constipation 12 4.6 .403 21 7.9 .026 4 2.9 1.000 3 2.3
Nausea 10 3.8 .143 15 5.6 .510 23 16.9 .025 10 7.6
Diarrhea 6 2.3 .518 7 2.6 .540 9 6.6 .412 5 3.8
Insomnia 6 2.3 .138 3 1.1 .017 12 8.8 .343 7 5.3
Tremor 2 0.8 1.000 9 3.4 .176 8 5.9 .036 1 0.8
a
Patients with multiple events in the same category are counted only once in that category.

week 8. Quetiapine monotherapy (600 mg/d) also demon- from the study before the first sampling time point, or may
strated significant efficacy on the majority of the individual not have had an opportunity for dose adjustment. It should
MADRS items including the core symptoms of depression also be noted that the lithium level quoted is the median
(apparent sadness, reported sadness, lassitude, and suicidal concentration, and as a result, some patients may in fact
thoughts). It is worth noting that the response and remission have had concentrations above the lower limit of the tar-
rates associated with quetiapine in this study were reported geted range. The lack of significance versus placebo in terms
in the context of placebo response and remission rates of of the primary efficacy measure observed in the lithium-
55.8% and 55.0%, respectively. Although high, these findings treated ITT population was echoed by the results from the
are largely consistent with those from previous studies in PP population, which only included patients with a median
bipolar depression, in which, for example, placebo response serum lithium concentration within the targeted range. A
rates of 29%–50% have been reported for lamotrigine.15 separate analysis of patients whose median serum lithium
Clinical studies in major depression also report wide vari- concentrations were ≥ 0.8 mEq/L10 yielded similar results
ability, with placebo response rates of up to 70%.30,31 The to that with the whole lithium group and did not reveal
statistical differentiation of quetiapine from placebo, in spite significant improvements in MADRS total score versus pla-
of these rates of placebo response, demonstrates the robust cebo, although the low patient numbers probably preclude
antidepressant efficacy of quetiapine. obtaining any statistically significant difference between
The breadth of the therapeutic profile of quetiapine, groups. In addition, only 2 of the 10 MADRS items (inner
previously shown in the BOLDER studies,18,19 was also tension and reduced sleep) significantly improved following
evident in this study. Symptomatic improvements were ob- lithium treatment compared with significant improvements
served with quetiapine treatment in the subpopulations of in the majority of these items with quetiapine treatment.
patients with bipolar I and II disorder. Similar to the Importantly, for some efficacy measures (MADRS and
results of the BOLDER I study,18 butnot BOLDER II,19 im- HDRS), quetiapine 600 mg/d demonstrated significantly
provements in MADRS total score for the smaller group of greater improvements over lithium while quetiapine 300
quetiapine-treated patients with bipolar II disorder did not mg/d showed numerical improvements over lithium. These
reach statistical significance. In patients with and with- findings are consistent with the suggestion that lithium has
out rapid cycling, improvements in symptoms (change in limited efficacy in managing symptoms of depression.11
MADRS total score) following quetiapine treatment were Persistent depressive symptoms may have detrimental
statistically significant only in the latter subgroup. In the effects on functioning and quality of life in patients with bi-
BOLDER studies, MADRS total score reductions with polar disorder.32 In addition to improvements in symptoms
quetiapine achieved significance versus placebo in both of depression and anxiety, both quetiapine doses showed
rapid- and non-rapid-cycling subgroups (P ≤ .01),18,19 pos- greater improvements than placebo in social and occupa-
sibly due to the larger sample population of rapid-cycling tional functioning according to the SDS, while quetiapine
patients available for analysis when compared with the cur- 600 mg/d showed significantly greater improvement than
rent study (143 in BOLDER II vs 45 in this study). placebo in cognitive function, as assessed by the patient on
In this study, lithium treatment was associated with the MOS-Cog scale.
numerically greater, but not statistically significant, improve- Patients with bipolar depression are at high risk of sui-
ments in most efficacy measures compared with placebo. cide,3,33 and, interestingly, lithium may have antisuicidal
It should be noted, however, that the mean of the median properties.34,35 No suicides occurred in the acute treatment
serum lithium concentrations was at the lower end of the phase of this study, and the incidence of treatment-emergent
targeted range. A number of factors may have accounted for suicidality was low across all treatment groups. Furthermore,
this finding; for example, some patients may have withdrawn quetiapine and lithium were associated with improvements

J Clin Psychiatry 71:2, February 2010 157

Young et al

Table 4. Weight and Clinical Laboratory Measures (safety population; acute treatment phase)
Change Proportion of Patients
Baseline From Baseline P Value With Clinically Relevant
Parameter and Treatment Mean SE Mean SE (vs placebo) nb Changes,a n (%)
Weight, kg
Quetiapine 300 mg/d 75.9 0.95 0.6 0.17 < .001 240 11 (4.6)
Quetiapine 600 mg/d 75.2 1.05 0.8 0.17 < .001 240 20 (8.3)
Placebo 75.3 1.28 −0.7 0.23 … 123 4 (3.3)
Lithium 77.5 1.54 0.2 0.19 .006 127 3 (2.4)
BMI, kg/m2
Quetiapine 300 mg/d 26.5 0.28 0.2 0.06 < .001 … …
Quetiapine 600 mg/d 26.5 0.33 0.3 0.06 < .001 … …
Placebo 26.4 0.44 −0.2 0.08 … … …
Lithium 27.2 0.49 0.1 0.07 .005 … …
HbA1c, %
Quetiapine 300 mg/d 5.52 0.04 0.02 0.02 .068 224 1 (0.4)
Quetiapine 600 mg/d 5.49 0.03 0.02 0.02 .066 237 1 (0.4)
Placebo 5.46 0.04 −0.04 0.03 … 116 0 (0.0)
Lithium 5.48 0.05 −0.12 0.03 .045 115 0 (0.0)
Glucose, mg/dL
Quetiapine 300 mg/d 95.51 0.94 2.19 1.20 .558 193 5 (2.6)
Quetiapine 600 mg/d 98.69 1.37 0.42 1.16 .684 186 8 (4.3)
Placebo 95.29 1.43 1.23 1.78 … 94 4 (4.3)
Lithium 97.54 1.54 2.40 1.70 .289 92 5 (5.4)
Insulin, pmol/L
Quetiapine 300 mg/d 86.43 8.01 26.14 11.65 .312 … …
Quetiapine 600 mg/d 88.69 6.94 21.51 7.75 .452 … …
Placebo 89.27 9.34 10.52 10.51 … … …
Lithium 94.25 9.59 −3.58 11.99 .478 … …
Triglycerides, mg/dL
Quetiapine 300 mg/d 148.47 5.77 3.46 5.81 .335 181 18 (9.9)
Quetiapine 600 mg/d 151.37 6.76 9.46 5.41 .086 183 27 (14.8)
Placebo 164.35 13.57 −11.33 10.48 … 88 8 (9.1)
Lithium 147.30 9.04 −2.17 6.95 .791 92 9 (9.8)
Total cholesterol, mg/dL
Quetiapine 300 mg/d 208.23 3.11 −5.56 2.29 .275 172 14 (8.1)
Quetiapine 600 mg/d 207.81 3.13 −2.24 2.18 .047 179 16 (8.9)
Placebo 205.64 4.26 −8.79 2.94 … 89 9 (10.1)
Lithium 206.69 4.93 −5.31 3.31 .365 86 4 (4.7)
HDL cholesterol, mg/dL
Quetiapine 300 mg/d 55.21 0.95 −1.10 0.74 .547 189 15 (7.9)
Quetiapine 600 mg/d 57.11 1.04 −1.45 0.64 .592 209 18 (8.6)
Placebo 57.04 1.62 −0.81 0.92 … 98 9 (9.2)
Lithium 57.30 1.51 −0.55 1.10 .813 103 3 (2.9)
LDL cholesterol, mg/dL
Quetiapine 300 mg/d 123.02 2.79 −4.56 2.00 .332 188 16 (8.5)
Quetiapine 600 mg/d 119.88 2.62 −2.61 1.99 .199 195 12 (6.2)
Placebo 116.16 3.45 −5.44 2.74 … 100 3 (3.0)
Lithium 119.88 3.94 −4.99 2.87 .647 95 5 (5.3)
Prolactin, µg/L
Quetiapine 300 mg/d 19.22 1.72 −4.12 2.38 .567 M = 80, F = 107 M = 6 (7.5), F = 13 (12.1)
Quetiapine 600 mg/d 16.62 1.92 −3.87 1.95 .905 M = 80, F = 129 M = 4 (5.0), F = 3 (2.3)
Placebo 21.95 3.37 −7.70 2.66 … M = 47, F = 53 M = 2 (4.3), F = 2 (3.8)
Lithium 15.40 2.17 −0.82 2.56 .534 M = 46, F = 61 M = 1 (2.2), F = 2 (3.3)
a
At end of treatment. Clinically relevant changes defined as weight ≥ 7% increase from baseline; HbA1c > 7.5%; glucose (fasting) ≥ 126 mg/dL;
triglycerides ≥ 200 mg/dL; total cholesterol ≥ 240 mg/dL; HDL cholesterol ≤ 40 mg/dL; LDL cholesterol ≥ 160 mg/dL; prolactin > 20 µg/L (men) or > 30
µg/L (women).
b
Number of patients with value below the threshold for potential clinical significance at baseline.
Abbreviations: F = female, HbA1c = glycosylated hemoglobin, HDL = high-density lipoprotein, LDL = low-density lipoprotein, M = male.

in suicidal thoughts as measured by item 10 on the MADRS whether quetiapine will eventually be shown to have anti-
scale, which reached significance for quetiapine 600 mg/d suicidal properties similar to those suggested for lithium.
versus placebo. This suggests that quetiapine is unlikely to Treatment-emergent mania remains another clinical
increase the risk of suicidality in acute bipolar depression concern during treatment of acute bipolar depression.12,13
in adults. However, patients at increased suicidal risk were In this study, the rate of treatment-emergent mania was low
excluded, and, as a result, this study is unable to provide across all treatment groups, although rates were higher in
conclusive evidence of treatment benefits in patients who the quetiapine 300 and 600 mg/d (4.2% and 2.2%, respec-
were currently suicidal and it remains to be determined tively) and lithium (2.2%) groups compared with placebo

158 J Clin Psychiatry 71:2, February 2010

 Quetiapine vs Lithium vs Placebo in Bipolar Depression

(0.8%). In BOLDER I and II, treatment-emergent mania (SDS and MOS-Cog) was similar for quetiapine and lithium,
rates with quetiapine were similar or lower when compared but the smaller sample population size may account for the
with placebo, which reached significance for quetiapine 300 nonsignificant results in lithium-treated patients.
mg/d in BOLDER II (1.8% vs 6.6% for placebo, P = .027).18,19 In addition to quetiapine monotherapy, the atypical anti-
However, it is notable that the incidence of treatment- psychotic olanzapine (monotherapy or in combination with
emergent mania in the placebo group in this study is low- fluoxetine) has demonstrated efficacy in the treatment of
er than that previously reported.18,19 Indeed, the rates of bipolar depression.16 In contrast, patients with bipolar I dis-
treatment-emergent mania for placebo groups in the order experiencing a major depressive episode treated with
BOLDER studies (BOLDER I, 3.9%; BOLDER II, 6.6%) were aripiprazole monotherapy had no significant improvements
similar or higher than the rates observed for quetiapine- over placebo treatment in 2 randomized, placebo-controlled
treated patients in this study.18,19 studies.39 It remains to be determined whether additional
Quetiapine treatment was generally well tolerated dur- atypical antipsychotics are similarly efficacious as mono-
ing acute treatment. Discontinuations due to AEs were therapy in bipolar depression, although variability in the
similar across the placebo, lithium, and quetiapine 300 mg/d pharmacologic profiles of these agents implies differences in
treatment groups (8.3%–9.8%) and slightly higher in the their clinical effectiveness. Indeed, it has been shown that the
quetiapine 600 mg/d group (13.1%). Although quetiapine magnitude of effect (compared with placebo) for quetiapine
monotherapy was associated with small mean increases in monotherapy exceeds that for olanzapine monotherapy.19,40
weight during the study and more quetiapine- than placebo- Studies of other treatments for bipolar depression have
treated patients experienced a ≥ 7% increase in body weight, not shown promising results. Patients in a depressive phase
none of the patients discontinued the study as a result of of bipolar I or II disorder who were receiving a mood stabi-
weight gain, an important finding given the negative im- lizer were randomly assigned to treatment with risperidone,
pact of weight gain on treatment compliance among patients paroxetine, or a combination of both. The results indicated
with bipolar disorder.36 Although mean changes in prolactin that there were no differences between each of the 3 treat-
levels associated with either dose of quetiapine were lower ment groups and that the depressive effects of treatment were
than those associated with placebo, a higher proportion of only modest.41 Lamotrigine has also been shown not to be
patients in the quetiapine 300 mg/d group, but not in the effective for the acute treatment of bipolar depression.15 In
600 mg group, had clinically relevant increases in prolactin contrast, the beneficial antidepressant effects of quetiapine
levels, which differs from previous studies that did not report may reflect a unique mechanism of action that involves
any clinically significant differences in prolactin parameters direct and indirect pharmacologic actions mediated by que-
with either dose of quetiapine and placebo.18,19 tiapine and its active metabolite, norquetiapine. The affinity
Adverse events experienced by patients in all treatment of norquetiapine for the norepinephrine transporter (NET)
groups were mainly mild to moderate in intensity. Adverse and the consequent inhibition of norepinephrine uptake is
events potentially related to EPS up to week 8 were similar one of the potential mechanisms that are postulated to un-
across the quetiapine treatment groups, but were marginally derlie the observed antidepressant properties of quetiapine
higher in the lithium-treated group and lower in the placebo in patients with bipolar depression.42 Currently, quetiapine
group. Mean change from baseline in SAS and BARS scores is the only atypical antipsychotic approved as monotherapy
at week 8 were minimal in all treatment groups. The overall by the US Food and Drug Administration for both phases
safety results did not alter the known risk-benefit profile of of bipolar disorder.43 In addition to its acute efficacy, 2 large
quetiapine.37 clinical studies have also shown long-term maintenance of ef-
Given the observations in this study and the necessity for ficacy (up to 104 weeks) with quetiapine in combination with
the monitoring of potential toxicity associated with lithium lithium or valproate in patients with bipolar I disorder.44,45
treatment,38 these findings suggest that quetiapine treatment The burden of symptoms of depression on the clinical
in patients with bipolar depression may generally provide outcome of patients with bipolar disorder is a mounting
greater benefits than lithium, an older, more conventional concern. The results from the acute treatment phase of this
treatment. large-scale study and the similarly designed EMBOLDEN
Differences in patient numbers within the sample popu- II and BOLDER I and II studies have demonstrated the effi-
lations may have influenced the results for some outcome cacy of quetiapine monotherapy for the treatment of bipolar
measures in this study. For example, although treatment depression. The improvements in symptoms of depression
differences for MADRS total score between placebo and que- were evident within the first week of treatment and confirm
tiapine treatment in patients with bipolar I and II disorder previous findings from the BOLDER studies. Quetiapine
were similar, this difference only reached significance for pa- was also associated with greater symptomatic improvements
tients with bipolar I disorder, which may be explained by the than lithium, one of the oldest and most established mood
fact that the number of patients with bipolar I disorder was stabilizers. The findings from this study have considerable
almost twice as high as the number of patients with bipolar II importance in view of the clear need for effective treatments
disorder. Similarly, the magnitude of functional improvement for bipolar depression.

J Clin Psychiatry 71:2, February 2010 159

Young et al

Drug names: aripiprazole (Abilify), fluoxetine (Prozac and others), Unpublic Centre of Health Care, Leszno, Poland; Won-Myong Bahk,
lamotrigine (Lamictal and others), lithium (Eskalith, Lithobid, and Catholic University of St Mary’s Hospital, Seoul, Republic of Korea;
others), lorazepam (Ativan and others), olanzapine (Zyprexa), Yeon-Ho Joo, Asian Medical Centre Psychiatry, Seoul, Republic of Korea;
olanzapine-fluoxetine combination (Symbyax), paroxetine (Paxil, Kyoo-Soeb Ha, Seoul National University Bundang, Gyeonggi-do,
Pexeva, and others), quetiapine (Seroquel), zaleplon (Sonata and Republic of Korea; Yakhin Kausar, City Psychoneurological Clinical
others), zolpidem (Ambien, Edluar, and others), zopiclone (Lunesta). Hospital. Kazan, Republic of Russia; Pavel Sidorov, Arkhangelsk
Author affiliations: Institute of Mental Health, Department of Regional Psychoneurological Dispensary, Arkhangelsk, Republic of
Psychiatry, The University of British Columbia, Vancouver, Canada Russia; Alexander Okhapkin, Regional Clinical Hospital, Smolensk,
(Dr Young); Lindner Center of HOPE, Mason, Ohio, and University Republic of Russia; Lala Kasimova, City Psychiatric Clinical Hospital,
of Cincinnati College of Medicine, Ohio (Dr McElroy); Department of Republic of Russia; Andrey Gribanov, Lipetsk Regional Psychiatric and
Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Neurologic Hospital #1 Dispensary, Lipetsk, Republic of Russia; Igor
Dresden, Germany (Dr Bauer); Anxiety and Mood Disorder Center, Boyev, Clinic of Borderline Mental Disorders, Stavropol, Republic of
Mississauga, Canada (Dr Philips); AstraZeneca Pharmaceuticals LP, Russia; Elena Grigorieva, Yaroslavl Regional Psychiatry Clinical Hospital,
Wilmington, Delaware (Drs Chang and Brecher); and AstraZeneca, Yaroslavl, Republic of Russia; Felix Torubarov, Clinical Hospital #6,
Södertälje, Sweden (Drs Olausson and Paulsson). Moscow, Republic of Russia; Galina Panteleeva, Mental Health Research
EMBOLDEN I (Trial 001) Study Investigators: Angelo Fallu, Clinique Centre, Moscow, Republic of Russia; Yuri Suchkov, City Psychiatric
Woodward, Sherbrooke, Canada; Jean-Guy Gagnon, Edwards Building, Clinical Hospital #1, Nizhyny Novgorod, Republic of Russia; Alexander
Sudbury, Canada; Carlos Galarraga-Carrero, ERB Center, Waterloo, Bukhanovsky, Rehabilitology Scientific Center, Rostov-on-Don, Republic
Canada; Paul Latimer, Kelowna, Canada; Serge Lessard, Introspect of Russia; Nikolay Ivanets, Korsakov Psychiatric Clinic, Moscow,
Clinic, Orleans, Canada; Nabil Philips, Credit Valley Medical Arts, Republic of Russia; Mikhail Burdukovsky, 4-th Psychiatric Hospital,
Mississauga, Canada; Francisco Pinero-Medina, Sherbrooke, Canada; Saint Petersburg, Republic of Russia; Nikolay Neznanov, Krestovsky
Javed Ali, Global Psychiatric Research, Sydney, Canada; Rustom Sethna, Island Medical Institute, Saint Petersburg, Republic of Russia; Valentina
Markham-Stouffville Health Centre, Markham, Canada; Smadar Rasnuk, City Clinical Hospital #2, Saratov, Republic of Russia; Natalia
Tourjman, Centre Medical Rene Laennac, Montreal, Canada; Vera Maxsimova, Regional Psychoneurological Hospital, Tver, Republic of
Folnegovic-Smalc and Bacic Davor, Psychiatric Hospital Vrapce, Zagreb, Russia; Vitaly Tadtaev, Psychoneuropatology Dispensary #10, Saint
Croatia; Pavo Filakovic, University Hospital Osijek, Osijek, Croatia; Petersburg, Republic of Russia; Maria Andrusenko, Clinic Mental
Goran Dodig, Psychiatric Hospital Split, Split, Croatia; Neven Health, Moscow, Republic of Russia; Vladimir Paunovic, Clinical Centre
Henigsberg, Policlinic Neuron, Zagreb, Croatia; Miro Jakovljevic, of Serbia Institute for Psychiatry, Belgrade, Serbia and Montenegro;
Clinical Hospital Centre Rebro, Zagreb, Croatia; Dragica Kozaric- Ivana Timotijevic, Institute for Mental Health, Belgrade, Serbia and
Kovacic, University Hospital Dubravna, Zagreb, Croatia; Branka Restek Montenegro; Goran Mihajlovic, Clinical Hospital Centre Kragujevac,
Petrovic, Psychiatric Hospital Jankomir, Zagreb, Croatia; Gilic Ante, Kragujevac, Serbia and Montenegro; Vladimir Diligensk, Clinical Centre
Department of Psychiatry, Zadar, Croatia; Georgi Belotserkovski, Ahtme Dedinje, Belgrade, Serbia and Montenegro; Jelena Martinović, Clinical
Mental Hospital, Kohtle-Jarve, Estonia; Kairi Magi, Tartu University Centre Zvezdara, Belgrade, Serbia and Montenegro; Željko Špirić,
Clinics, Tartu, Estonia; Katrin Noorkoiv and Petra Poolamets, North- Military Medical Academy Clinic for Psychiatry, Belgrade, Serbia and
Estonia Regional Hospital, Tallinn, Estonia; Michael Bauer, University Montenegro; Grozdanko Grbeša, Clinical Hospital Centre Niš, Niš,
Clinic for Psychiatry and Psychotherapy, Berlin, Germany; Martin Klein, Serbia and Montenegro; Tomislav Gajić, Health Centre Valjevo
Medical Studycentra, Würzburg, Germany; Alexander Schulze, Praxis Psychiatric Service, Valjevo, Serbia and Montenegro; Mirjana Šojić,
Dr Schulze, Berlin, Germany; Jana Thomsen, Praxis Dr Jana Thomsen, Health Centre Dr Dragiša Mišović, Čačak, Serbia and Montenegro;
Berlin, Germany; Jürgen Ribbschlaeger, Praxis Dr Margit and Jürgen Milanka Cvetković, Health Counselling for Neuro Disorders, Belgrade,
Ribbschlaeger, Berlin, Germany; Nurmiati Amir, Dr Cipto Serbia and Montenegro; Tung-Ping Su, Tapei Veterans General Hospital,
Mangunkusumo General Hospital, Jakarta, Indonesia; H Aris Sudiyanto, Tapei, Taiwan; Nan-Ying Chin, Changhau Christain Hospital, Changau,
Surakarta Mental Health Hospital, Solo, Indonesia; M A Widyastuti Taiwan; Chih-Jen Ho and Shen-Ing Liu, Mackay Memorial Hospital,
Darmohusodo, Dr Radjiman W State Mental Hospital, Lawang, Tapei, Taiwan; Bohdan Suvalo, Lviv Regional State Clinical Psychiatric
Indonesia; Biruta Kupca, Mental Health State Agency, Riga, Latvia; Ilona Hospital, Lviv, Ukraine; Petro Vlasovych Voloshyn, Institute of
Paegle, Outpatient Clinic, Riga, Latvia; Olga Sokolova, Daugavpils Neurology Psychiatry and Narcology of the Academy of Medical
Psychoneurologica, Daugavpils, Latvia; Valentinas Maciulis, Republican Sciences of Ukraine, Kharkiv, Ukraine; Oleksandr Konstyantynovych
Psychiatric Hospital, Vilnuis, Lithuania; Gintautas Daubaras, Antakalnio Napryeyenko, Kiev City Clinical Psychoneurologic Hospital No. 1, Kiev,
Psychiatric Consultation Centre, Vilnuis, Lithuania; Petras Janulionis, Ukraine; Oleh Sozontovych Chaban, Scientific Research Institute of
Vilnuis Mental Health Centre, Vilnuis, Lithuania; Daiva Deltuviene, Forensic, Social Psychiatry and Narcology, Kiev, Ukraine; Gennadiy
Medical Centre Neuromeda, Kaunas, Lithuania; Eugenijus Mikaliunnas, Vasliovych Zilberblat, Kiev Regional Specialized Psycho-Narcological
Sauliai Mental Hospital, Sauliai, Lithuania; Sonata Rudzianskiene, Medical Centre, Glevaha, Ukraine; Natallya Grygoruivna Pshuk,
Kaunas Silainiai Outpatient Clinic, Kaunas, Lithuania; Laisve Regional Clinical Psycho-Neurological Hospital, Vinnytsya, Ukraine;
Dembinskiene, Vilnuis, Lithuania; Ahmad Sulaiman, University Malaya Ellina Vitalyivna Melnyk, Odessa Regional Psychiatry Hospital No. 1,
Medical Centre, Petaling Jaya, Malaysia; Teck Hoe Yen, Chinese Odessa, Ukraine; Vladyslav Andriyovych Demchenko, Psychoneurologic
Maternity Hospital, Kuala Lumpur, Malaysia; Abdul Kadir Abu Bakar, Hospital No. 2, Kiev, Ukraine; Anatoliy Ipatov, Ukrainian Research
Hospital Sentosa, Kuching, Malaysia; Paul Stronegger, Hospital Østfold Institute of Disability Problems, Dnipropetrovsk, Ukraine; Irina
HF Division, Mysen, Norway; Erik Øfjord, Centre for Clinical Studies, Dmitriyevna Spirina, Dnipropetrovsk Regional Clinical Psychiatric
Paraida, Norway; Ole Johan Høyberg, Doctors Office Brattvåg, Brattvåg, Hospital, Dnipropetrovsk Ukraine; Tetyana Leonidovna Ryapolova,
Norway; Hans Jørgan Nyrerød, Drammen Psychiatric Centre, Drammen, Donetsk Medical Psycho-Neurological Centre, Donetsk, Ukraine; Valeriy
Norway; Dag Norum, Dag Norum, Fredrikstad, Norway; Efren Reyes N Kuznetsov, Kiev City Clinical Psychiatric and Neurological Hospital
and Juan Villacorta, National Center for Mental Health, Mandaluyong No. 1, Kiev, Ukraine; Svitlana Yehv Kazakova, Lugansk State Medical
City, Philippines; Victor Amantillo, West Visayas State University University, Lugansk, Ukraine; Svitlana Moroz, Dnipropetrovsk Regional
Medical Centre, Iloilo City, Philippines; Agnes Padilla, Davao Mental Clinical Hospital, Dnipropetrovsk Ukraine; Vitaliy Pischel, Ukrainian
Hospital, Davao City, Philippines; Michal Kujawski and Agata Szulc, Scientific Institute of Social and Forensic Psychiatry and Drug Abuse,
Unpublic Centre of Health Care, Zabrze, Poland; Aleksander Kiev, Ukraine; and Viktoriya Verbenko and Mykola Verbenko, Crimea
Araszkiewicz, Cathedral and Psychiatric Clinic of Medical Academy, Republic Clinical Hospital, Simferopol, Ukraine.
Bydgoszcz, Poland; Piotr Baranowski, Medical Psychiatric Outpatient Potential conflicts of interest: Dr Young has received honoraria from
Clinic, Wroclaw, Poland; Wlodzimierz Chrzanowski, Private Medical pharmaceutical companies, including AstraZeneca, for lecturing on this
Offices Promedicus, Bialystok, Poland; Leszek Bidzan, Unpublic Centre topic and has also received grant support from AstraZeneca. Dr McElroy
of Health Care, Gdynia, Poland; Zbigniew Wawrzyniak, Specialist is a consultant to or member of the scientific advisory boards of Abbott,
Psychiatric Complex of Health Care, Lodz, Poland; Przemyslaw Bogacki, AstraZeneca, Eli Lilly, GlaxoSmithKline, Janssen, Jazz, Ortho-McNeil,
Private Medical Office, Skorzewo, Poland; Jaroslaw Bialek, Complex of and Wyeth-Ayerst. She is a principal or co-investigator on studies spon-
Psychiatric Clinics, Wloclawek, Poland; Magdelena Kielkiewicz, sored by the above companies and Bristol-Myers Squibb, Eisai, Forest,

160 J Clin Psychiatry 71:2, February 2010

 Quetiapine vs Lithium vs Placebo in Bipolar Depression

National Institute of Mental Health, OREXIGEN Therapeutics, Pfizer, 12. Gijsman HJ, Geddes JR, Rendell JM, et al. Antidepressants for bipolar
Sanofi-Synthelabo, Somaxon, Stanley Medical Research Institute, and depression: a systematic review of randomized, controlled trials. Am J
Takeda and is an inventor on United States Patent No. 6,323,236 B2, Use Psychiatry. 2004;161(9):1537–1547.PubMed doi:10.76/apj953
of Sulfamate Derivatives for Treating Impulse Control Disorders, and, 13. Leverich GS, Altshuler LL, Frye MA, et al. Risk of switch in mood polar-
along with the patient’s assignee, University of Cincinnati, Cincinnati, ity to hypomania or mania in patients with bipolar depression during
Ohio, has received payments from Johnson & Johnson, which has ex- acute and continuation trials of venlafaxine, sertraline, and bupropion
clusive rights under the patent. Dr Bauer is a consultant to or member as adjuncts to mood stabilizers. Am J Psychiatry. 2006;163(2):232–239.PubMed doi:10.76/apj32
of the scientific advisory boards of and has received honoraria from Eli 14. Altshuler LL, Suppes T, Black DO, et al. Lower switch rate in depressed
Lilly, Wyeth, Servier, GlaxoSmithKline, AstraZeneca, and Bristol-Myers patients with bipolar II than bipolar I disorder treated adjunctively
Squibb. Drs Chang and Brecher are former employees of AstraZeneca. with second-generation antidepressants. Am J Psychiatry. 2006;163(2):
Drs Olausson and Paulsson are employees of AstraZeneca. Dr Philips 313–315.PubMed doi:10.76/apj32
reports no additional personal financial or other relationship relevant to 15. Calabrese JR, Huffman RF, White RL, et al. Lamotrigine in the acute
the subject of this article. treatment of bipolar depression: results of five double-blind, placebo-
Funding/support: Supported by AstraZeneca Pharmaceuticals (Study controlled clinical trials. Bipolar Disord. 2008;10(2):323–333.PubMed doi:10./j39-568270.x
D1447C00001). 16. Tohen M, Vieta E, Calabrese J, et al. Efficacy of olanzapine and
Previous presentation: Previously presented as posters at the follow- olanzapine-fluoxetine combination in the treatment of bipolar I depres-
ing scientific conferences: 3rd Biennial Conference of the International sion. Arch Gen Psychiatry. 2003;60(11):1079–1088.PubMed doi:10./archpsy679
Society for Bipolar Disorders, January 27–30, 2008, Delhi and Agra, 17. Montgomery SA, Asberg M. A new depression scale designed to be
India; European Congress of Psychiatry, April 5–9, 2008, Nice, sensitive to change. Br J Psychiatry. 1979;134(4):382–389.PubMed doi:10.92/bjp348
France; International Review of Bipolar Disorders, April 14–16, 2008, 18. Calabrese JR, Keck PE Jr, Macfadden W, et al; for the BOLDER Study
Copenhagen, Denmark; Society of Biological Psychiatry, May 1–3, 2008, Group. A randomized, double-blind, placebo-controlled trial of que-
Washington, DC; American Psychiatric Association, May 3–8, 2008, tiapine in the treatment of bipolar I or II depression. Am J Psychiatry.
Washington, DC; European College of Neuropsychopharmacology, 2005a;162(7):1351–1360.PubMed doi:10.76/apj2351
August 30–September 3, 2008, Barcelona, Spain; and the International 19. Thase ME, Macfadden W, Weisler RH, et al. BOLDER II Study Group.
Forum on Mood and Anxiety Disorders, November 2008, Vienna, Efficacy of quetiapine monotherapy in bipolar I and II depression: a
Austria. double-blind, placebo-controlled study (the BOLDER II study). J Clin
Acknowledgment: We thank Eleanor Bull, PhD, from PAREXEL, who Psychopharmacol. 2006;26(6):600–609.PubMed doi:10.97/jcp248631.b7
provided medical writing support funded by AstraZeneca. 20. O’Dowd A. NICE issues new guidance to improve the treatment
of bipolar disorder. BMJ. 2006;333(7561):220.PubMed doi:10.36/bmj752
21. Yatham LN, Kennedy SH, O’Donovan C, et al. Guidelines Group,
REFERENCES CANMAT. Canadian Network for Mood and Anxiety Treatments
(CANMAT) guidelines for the management of patients with bipolar
  1. Prince M, Patel V, Saxena S, et al. No health without mental health. disorder: update 2007. Bipolar Disord. 2006;8(6):721–739.PubMed doi:10./j39-568204.x
Lancet. 2007;370(9590):859–877.PubMed doi:10.6/S4-73()1280 22. Young RC, Biggs JT, Ziegler VE, et al. A rating scale for mania:
  2. Dilsaver SC, Chen YW, Swann AC, et al. Suicidality, panic disorder and reliability, validity and sensitivity. Br J Psychiatry. 1978;133(5):429–435.PubMed doi:10.92/bjp354
psychosis in bipolar depression, depressive-mania and pure-mania. 23. Hamilton M. A rating scale for depression. J Neurol Neurosurg
Psychiatry Res. 1997;73(1–2):47–56.PubMed doi:10.6/S5-78(9)10 Psychiatry. 1960;23(1):56–62.PubMed doi:10.36/jnp25
  3. American Psychiatric Association. Practice Guideline for the Treatment 24. Spearing MK, Post RM, Leverich GS, et al. Modification of the Clinical
of Patients With Bipolar Disorder (Revision). Am J Psychiatry. 2002; Global Impressions (CGI) Scale for use in bipolar illness (BP):
159(suppl 4):1–50. the CGI-BP. Psychiatry Res. 1997;73(3):159–171.PubMed doi:10.6/S5-78(9)123
  4. Goodwin GM; Consensus Group of the British Association for 25. Hamilton M. The assessment of anxiety states by rating. Br J Med
Psychopharmacology. Evidence-based guidelines for treating bi- Psychol. 1959;32(1):50–55.PubMed
polar disorder: recommendations from the British Association for 26. Sheehan DV. The Anxiety Disease. New York, NY: Scribner’s; 1983.
Psychopharmacology. J Psychopharmacol. 2003;17(2):149–173, 27. Stewart AL, Ware JE, Sherbourne CD, et al. Psychological Distress/
discussion 147.PubMed doi:10.7/269830 Well-Being and Cognitive Functioning Measures. Measuring Functioning
  5. Grunze H, Kasper S, Goodwin G, et al. World Federation of Societies of and Well-Being: the Medical Outcomes Study Approach. Durham, NC:
Biological Psychiatry Task Force on Treatment Guidelines for Bipolar Duke University Press; 1992:102–142.
Disorders. World Federation of Societies of Biological Psychiatry 28. Simpson GM, Angus JW. A rating scale for extrapyramidal side effects.
(WFSBP) guidelines for biological treatment of bipolar disorders, Acta Psychiatr Scand suppl. 1970;45(S212):11–19. doi:10./j6-479tb02.x
part I: treatment of bipolar depression. World J Biol Psychiatry. 29. Barnes TR. A rating scale for drug-induced akathisia. Br J Psychiatry.
2002;3(3):115–124.PubMed doi:10.39/5627 1 1989;154(5):672–676.PubMed doi:10.92/bjp5467
  6. Royal Australian and New Zealand College of Psychiatrists Clinical 30. Walsh BT, Seidman SN, Sysko R, et al. Placebo response in stud-
Practice Guidelines Team for Bipolar Disorder. Australian and New ies of major depression: variable, substantial, and growing. JAMA.
Zealand clinical practice guidelines for the treatment of bipolar disorder. 2002;287(14):1840–1847.PubMed doi:10./jam2874
Aust N Z J Psychiatry. 2004;38(5):280–305.PubMed doi:10./j4-620135.x 31. Schatzberg AF, Kraemer HC. Use of placebo control groups in evaluat-
  7. Suppes T, Dennehy EB, Hirschfeld RM, et al; Texas Consensus ing efficacy of treatment of unipolar major depression. Biol Psychiatry.
Conference Panel on Medication Treatment of Bipolar Disorder. 2000;47(8):736–744.PubMed doi:10.6/S-32()845
The Texas implementation of medication algorithms: update to the 32. Piccinni A, Catena M, Del Debbio A, et al. Health-related quality of life
algorithms for treatment of bipolar I disorder. J Clin Psychiatry. and functioning in remitted bipolar I outpatients. Compr Psychiatry.
2005;66(7):870–886.PubMed 2007;48(4):323–328.PubMed doi:10.6/jcmpsyh21.07
  8. Yatham LN, Kennedy SH, O’Donovan C, et al; Canadian Network 33. Valtonen HM, Suominen K, Mantere O, et al. Prospective study of risk
for Mood and Anxiety Treatments. Canadian Network for Mood and factors for attempted suicide among patients with bipolar disorder.
Anxiety Treatments (CANMAT) guidelines for the management of Bipolar Disord. 2006;8(5, Pt 2):576–585.PubMed doi:10./j39-5682041.x
patients with bipolar disorder: consensus and controversies. Bipolar 34. Baldessarini RJ, Tondo L, Davis P, et al. Decreased risk of suicides and
Disord. 2005;7(suppl 3):5–69.PubMed attempts during long-term lithium treatment: a meta-analytic review.
  9. Hlastala SA, Frank E, Mallinger AG, et al. Bipolar depression: an under- Bipolar Disord. 2006;8(5, Pt 2):625–639.PubMed [Erratum in: Bipolar Disord.
estimated treatment challenge. Depress Anxiety. 1997;5(2):73–83.PubMed doi:10.2/(SIC)5-639417:2<AID3>.0CO;2-6 2007;9:314]. doi:10./j39-568204.x
10. Nemeroff CB, Evans DL, Gyulai L, et al. Double-blind, placebo- 35. Cipriani A, Pretty H, Hawton K, et al. Lithium in the prevention of
controlled comparison of imipramine and paroxetine in the treatment suicidal behavior and all-cause mortality in patients with mood dis-
of bipolar depression. Am J Psychiatry. 2001;158(6):906–912.PubMed doi:10.76/apj589 orders: a systematic review of randomized trials. Am J Psychiatry.
11. Zornberg GL, Pope HG Jr. Treatment of depression in bipolar disorder: 2005;162(10):1805–1819.PubMed doi:10.76/apj2805
new directions for research. J Clin Psychopharmacol. 1993;13(6): 36. Johnson FR, Ozdemir S, Manjunath R, et al. Factors that affect adher-
397–408.PubMed doi:10.97/4-320 5 ence to bipolar disorder treatments: a stated-preference approach.

J Clin Psychiatry 71:2, February 2010 161

Young et al

Med Care. 2007;45(6):545–552.PubMed doi:10.97/MLRb3e840ad9 Presented at the 20th European College of Neuropsychopharmacology;
37. Keating GM, Robinson DM. Quetiapine: a review of its use in the October 2007; Vienna, Austria
treatment of bipolar depression. Drugs. 2007;67(7):1077–1095.PubMed doi:10.265/349-70 8 43. Seroquel [package insert]. Wilmington, DE: AstraZeneca
38. Delva NJ, Hawken ER. Preventing lithium intoxication. Guide for Pharmaceuticals; 2008
physicians. Can Fam Physician. 2001;47:1595–1600.PubMed 44. Vieta E, Suppes T, Eggens I, et al. Efficacy and safety of quetiapine
39. Thase ME, Jonas A, Khan A, et al. Aripiprazole monotherapy in nonpsy- in combination with lithium or divalproex for maintenance of pa-
chotic bipolar I depression: results of 2 randomized, placebo-controlled tients with bipolar I disorder (international trial 126). J Affect Disord.
studies. J Clin Psychopharmacol. 2008;28(1):13–20.PubMed doi:10.97/jcpb3e8614 2008;109(3):251–263.PubMed doi:10.6/ja28
40. Calabrese JR, Elhaj O, Gajwani P, et al. Clinical highlights in bipo- 45. Suppes T, Vieta E, Liu S, et al. Trial 127 Investigators. Maintenance treat-
lar depression: focus on atypical antipsychotics. J Clin Psychiatry. ment for patients with bipolar I disorder: results from a North American
2005b;66(suppl 5):26–33.PubMed study of quetiapine in combination with lithium or divalproex (trial
41. Shelton RC, Stahl SM. Risperidone and paroxetine given singly and 127). Am J Psychiatry. 2009;166(4):476–488.PubMed doi:10.76/apj28019
in combination for bipolar depression. J Clin Psychiatry. 2004;65(12): 46. McElroy SL, Weisler RH, Chang W, et al. A double-blind, placebo-
1715–1719.PubMed controlled study of quetiapine and paroxetine as monotherapy in
42. Goldstein JM, Christoph G, Grimm S, et al. Unique mechanism for adults with bipolar depression (EMBOLDEN II). J Clin Psychiatry.
the antidepressant properties of the atypical antipsychotic quetiapine. 2010;71(2):163–174.

162 J Clin Psychiatry 71:2, February 2010