antidepressants in bipolar patients

Page 1
Am J Psychiatry 161:9, September 2004
1537
Reviews and Overviews
Antidepressants for Bipolar Depression:
A Systematic Review of Randomized, Controlled Trials
Harm J. Gijsman, Ph.D.,
M.R.C.Psych.
John R. Geddes, M.D.,
F.R.C.Psych.
Jennifer M. Rendell, M.A.
Willem A. Nolen, Ph.D.
Guy M. Goodwin, D.Phil.,
F.R.C.Psych.
Objective: This study reviewed the evi-
dence from randomized, controlled trials
on the efficacy and safety of antidepres-
sants in the short-term treatment of bipo-
lar depression.
Method: The authors performed a sys-
tematic review and meta-analysis of ran-
domized, controlled trials. They searched
the Cochrane Collaboration Depression,
Anxiety, and Neurosis Controlled Trials
Register, incorporating results of searches
of MEDLINE, EMBASE, CINAHL, PsycLIT,
PSYNDEX, and LILACS. The main outcome
measures were the proportion of patients
who clinically responded to treatment
and the rate of switching to mania.
Results: Twelve randomized trials were
included, with a total of 1,088 randomly
assigned patients. Five trials compared
one or more antidepressants with pla-
cebo: 75% of these patients were receiv-
ing a concurrent mood stabilizer or an
atypical antipsychotic. Antidepressants
were more effective than placebo. Antide-
pressants did not induce more switching
to mania (the event rate for antidepres-
sants was 3.8% and for placebo, it was
4.7%). Six trials allowed comparison be-
tween two antidepressants. The rate of
switching for tricyclic antidepressants was
10%, and for all other antidepressants
combined, it was 3.2%.
Conclusions: Antidepressants are effec-
tive in the short-term treatment of bipo-
lar depression. The trial data do not sug-
gest that switching is a common early
complication of treatment with antide-
pressants. It may be prudent to use a se-
lective serotonin reuptake inhibitor or a
monoamine oxidase inhibitor rather than
a tricyclic antidepressant as first-line treat-
ment. Given the limited evidence, there is
a compelling need for further studies with
longer follow-up periods and careful defi-
nition and follow-up of emerging mania
and partial remission.
(Am J Psychiatry 2004; 161:1537–1547)
The treatment of depressive episodes in patients with
bipolar disorder presents a puzzling paradox. In perhaps
most countries in the world where treatment is available,
patients are offered antidepressants with little delibera-
tion. By contrast, in all treatment guidelines, experts ap-
pear to agonize over whether antidepressants should be
offered at all, at least as monotherapy. This division of
opinion between the real world of treatment and the more
refined world of evidence is one of the major current chal-
lenges in psychiatric management. It reflects genuine un-
certainty about the benefit and harm of antidepressants in
treating the 50 million people who may have bipolar disor-
der worldwide. This is a substantial therapeutic challenge
because in bipolar disorder, depression—either recurrent
or chronic—causes more disability than any other mani-
festation of the illness (1–3).
Clinically, the issue is an apparently straightforward bal-
ance between benefit and risk. Do antidepressants work in
acute bipolar depression as well as they do in unipolar de-
pression? If they do, is the potential cost—the risk of in-
duction of a manic episode or mood instability—too high
to justify their use? And should they be prescribed only in
combination with mood stabilizers? These questions
clearly concern both the short-term use of antidepres-
sants and also the longer-term use for relapse prevention.
Depression is often chronic and highly recurrent in bipo-
lar patients, and it carries a significant risk of suicide (4, 5).
In unipolar patients, there is strong evidence that long-
term treatment with antidepressants prevents relapse of
depression for at least 3 years and perhaps indefinitely (6).
By contrast, the long-term use of antidepressants in bipo-
lar patients may be associated with manic relapse (7, 8)
and sometimes an increased frequency of bipolar epi-
sodes (rapid cycling) (9). Spontaneous switching from
depression to mania is associated with poor long-term
outcome (10).
The precautionary view is that to prescribe antidepres-
sants puts bipolar patients at risk of iatrogenic episodes of
mania, mixed states, or rapid cycling. Thus, antidepres-
sants cannot stabilize mood because they may worsen the
outcome of the opposite (elated) pole of the illness. All
major reviews and guidelines for bipolar depression over
the past 10 years have advised the use of a mood stabilizer
(usually lithium or valproate) rather than an antidepres-
sant as the first-line treatment for bipolar depression.
However, the available evidence favoring lithium, for ex-
ample, would not be sufficient to register it as an antide-
pressant in the current regulatory climate because there

Page 2
1538
Am J Psychiatry 161:9, September 2004
ANTIDEPRESSANTS FOR BIPOLAR DEPRESSION
are no trials showing superiority versus placebo or an ac-
tive comparator that meets current standards of trial de-
sign. Almost all were small crossover studies of very short
duration (11). Antidepressants are recommended only as
second-line treatment and then always with a concurrent
mood stabilizer to prevent switching to mania. However,
even in North America, where this policy is most strongly
advocated, in clinical practice, bipolar patients are still fre-
quently prescribed antidepressants as monotherapy (12).
The gap between practice and theory should make us all
uncomfortable. There is a current need to make the best
sense of the evidence, and we believe the key concerns are
as follows:
1. Previous reviews have all been essentially qualitative
in approach. Accordingly, they have run the well-rec-
ognized risk of bias and overinterpretation of favored
studies at the expense of more reliable average ef-
fects. Systematic reviews are less vulnerable to bias
and more likely to highlight true areas of uncertainty
(13, 14).
2. Should we really be skeptical about the efficacy of an-
tidepressants? How does the totality of the random-
ized evidence stack up?
3. Can we be confident that the association between
mood switches to mania and the use of antidepres-
sants is usually causal? The natural course of bipolar
disorder is to switch between the two poles of depres-
sion and mania. Can we quantify the increased risk
from the existing data?
4. Are some antidepressants either more effective than
others or less likely to produce a manic switch?
5. Conservative use of antidepressants complicates de-
cision making in bipolar depression. When do we de-
cide that a mood stabilizer alone has failed? It is easy
to leave patients with mild but chronic depressive
symptoms.
The objective of this study was systematically to assess
the evidence from randomized, controlled trials of antide-
pressants in short-term treatment of bipolar depression
using the rigorous methods developed by the Cochrane
Collaboration (15). Our aim was to produce the best avail-
able quantitative estimates of the risks and benefits of an-
tidepressant treatment in bipolar disorder.
Method
Studies Included
We included randomized, controlled double-blind trials, pub-
lished in any language, that compared antidepressants with pla-
cebo or alternative drug treatments. Antidepressants included all
tricyclic antidepressants, selective serotonin reuptake inhibitors
(SSRIs), noradrenergic reuptake inhibitors, reversible mono-
amine oxidase inhibitors (MAOIs), bupropion, St. John’s wort, and
tryptophan but not mood stabilizers, anticonvulsants, lamotri-
gine, sulpiride, benzodiazepines, or ECT. Alternative comparator
treatments included mood stabilizers, anticonvulsants, and other
antidepressants. The inclusion criteria for the review were trials
with patients with a current depressive or mixed depressive/
manic episode with or without psychotic symptoms who had had
at least one previous episode of mania or hypomania, including
antidepressant-induced mania or hypomania. Trials that re-
cruited nonbipolar patients as well as bipolar patients were in-
cluded if they randomly assigned the bipolar patients separately
or if the majority of the patients were bipolar. The primary out-
come measures for the review were clinical response and remis-
sion rates (derived from observer-rated symptom reductions), in-
duction of mania or hypomania, and the overall dropout rate as a
proxy measure of the acceptability of treatment.
Search Strategy
We searched the Cochrane Collaboration Depression, Anxiety,
and Neurosis Controlled Trials Register (December 2002), incorpo-
rating results of group searches of MEDLINE (1966 to the present),
EMBASE (1980 to the present), CINAHL (1982 to the present),
PsycLIT (1974 to the present), PSYNDEX (1977 to the present), and
LILACS (1982–1999). These were searched by using the following
terms for diagnosis: bipolar III disorder, unipolar mania, rapid-
cycling disorder, affective disorders, affective psychosis, bipolar,
bipolar disorder, bipolar I disorder, bipolar II disorder, cyclothy-
mic disorder, depression, depressive psychosis, excited psycho-
sis, hypomania, mania, manic depressive, manic disorder, manic
episode, melancholia, mixed depression, mood disorders, bipo-
lar affective disorder, bipolar not otherwise specified, dysphoric
mania, manic episode, manic symptoms, schizoaffective disor-
der, psychoses, psychotic disorders, puerperal psychosis, and re-
active depressive psychosis. The terms searched for intervention
and antidepressive agents were monoamine oxidase inhibitors,
selective serotonin reuptake inhibitors, and tricyclic drugs. The
following drug names were also searched: acetylcarnitine,
alaproclate, amesergide, amiflamine, amineptine, amitriptyline,
amoxapine, befloxatone, benactyzine, brofaromine, bupropion,
butriptyline, caroxazone, chlorpoxiten, cilosamine, cimoxatone,
citalopram, clomipramine, clorgyline, clorimipramine, clovoxa-
mine, deanol, demexiptiline, deprenyl, desipramine, dibenzepin,
diclofensine, dothiepin, doxepin, duloxetine, etoperidone, femoxe-
tine, fluotracen, fluoxetine, fluparoxan, fluvoxamine, idazoxan,
imipramine, iprindole, iproniazid, isocarboxazid, litoxetine,
lofepramine, maprotiline, medifoxamine, melitracen, metapra-
mine, mianserin, milnacipran, minaprine, mirtazapine, moclobe-
mide, nefazodone, nialamide, nomifensine, nortriptyline, noxipti-
line, opipramol, oxaflozane, oxaprotiline, pargyline, paroxetine,
phenelzine, piribedil, pirlindole, pivagabine, prosulpride, pro-
triptyline, quinupramine, reboxetine, rolipram, sertraline, setipti-
line, teniloxine, tetrindole, thiazesim, thozalinone, tianeptine,
toloxatone, tomoxetine, tranylcypromine, trazodone, trimipra-
mine, venlafaxine, viloxazine, viqualine, and zimeldine.
The Cochrane Library was searched by using the same terms as
the Cochrane Collaboration Depression, Anxiety, and Neurosis
Controlled Trials Register excluding references that came from
that database. The reference lists of selected studies were in-
spected for more published reports and citations of unpublished
research. In addition, other relevant papers and major textbooks
that cover affective disorder were checked.
Methods of the Review
Two of us (H.J.G., J.M.R.) independently checked studies gener-
ated by the search strategy to ensure that they met the inclusion
criteria. Any disagreement was resolved by consensus discussion
with another coauthor (J.R.G.). One reviewer (H.J.G.) assessed the
methodological quality of the included studies according to the re-
porting of the randomization procedure, especially allocation con-
cealment (15), blinding, and the reporting of withdrawals. Infor-
mation about participant characteristics, intervention details, and
outcome measures was extracted independently by two reviewers.

Page 3
Am J Psychiatry 161:9, September 2004
1539
GIJSMAN, GEDDES, RENDELL, ET AL.
Data Analysis
Data were entered twice into Revman 4.2, a program developed
by the Cochrane Collaboration for systematic reviews. For binary
efficacy outcomes, a pooled relative risk (with 95% confidence in-
tervals [CIs]) was calculated by using both fixed- and random-
effects models to investigate the sensitivity of results to the choice
of statistical method. Heterogeneity between studies was as-
sessed by using the Q statistic (16). When significant heterogene-
ity was identified, sources were investigated by visual inspection
of the forest plots for outlying trials. Two sensitivity analyses were
conducted, one excluding two trials that included less than 100%
bipolar patients and another including a trial with bipolar pa-
tients who were not randomized separately from unipolar pa-
tients. Treatment effects are also presented as the number needed
to treat, which is the reciprocal of the absolute difference between
the response rates of the compared treatments. The number
needed to treat can assist clinical interpretation of the results of
trials and expresses the number of patients that must be treated
with the drug of interest to achieve one positive outcome more
than would be achieved with a comparator or placebo (17).
Results
Description of Studies
Twelve trials were included in the primary analysis: five
comparing one or more antidepressants with placebo (to-
tal participants, N=779), four studies comparing two differ-
ent antidepressants only (N=236), and three trials compar-
ing an antidepressant with another type of drug (N=73).
We had originally identified 23 trials in which bipolar
depressed patients received an antidepressant, but we had
to exclude 11 trials. Eight of these studies had included
only a low proportion of bipolar patients, who were not
randomized separately (18–25). Two excluded trials com-
pared other treatments that were added to antidepressant
treatment (26, 27). One trial compared two antidepres-
sants but was not blinded (28).
Design Characteristics
The characteristics of the included trials are presented
in Table 1. All trials were described as randomized and
used a parallel group design. Study duration varied be-
tween 4 and 10 weeks. All trials excluded patients with a
diagnosis of serious physical illness or substance abuse.
Participants were adult patients up to 70 years old. In most
studies, the majority (approximately 60%–70%) of partici-
pants were women.
Outcomes
Dichotomous outcomes are presented here. With regard
to continuous outcomes, nine studies reported Hamilton
Depression Rating Scale or Montgomery-Åsberg Depres-
sion Rating Scale scores, but only some reported baseline
scores of all randomly assigned patients. Thus, meta-anal-
ysis of continuous data was not feasible.
Antidepressants Versus Placebo
Five trials were available for this comparison. About
75% of the patients were taking a concurrent mood stabi-
lizer or an atypical antipsychotic. One recent trial com-
pared three groups: olanzapine plus fluoxetine (N=86),
olanzapine alone (N=370), and placebo (N=377) (29). We
only included the former two groups in this analysis as,
respectively, the fluoxetine and placebo groups.
Clinical Response
One of these five trials only reported remission and was
therefore excluded from this comparison (30). One other
study did report responders but not how this was defined
and was included in this analysis (33). As a result, four tri-
als with a total of 662 randomly assigned patients were
available for this comparison, 213 in the experimental
group and 449 in the placebo group. Patients treated with
an antidepressant were more likely to respond by the end
of the trial (risk ratio=1.86, 95% CI=1.49–2.30) (Figure 1).
The number needed to treat with antidepressants was 4.2
(95% CI=3.2–6.4). The treatment effect remained statisti-
cally significant after taking account of the heterogeneity
by using a random-effect model (risk ratio=2.29, 95% CI=
1.29–4.04). The direction of the effect favored antidepres-
sants in all the trials, which implies that the heterogeneity
was quantitative rather than qualitative. We could not dis-
tinguish response rates for patients taking and not taking
concurrent medication within individual trials.
Clinical Remission
This was defined in only two studies, as a Hamilton de-
pression scale score ≤7 (39) and a Montgomery-Åsberg
Depression Rating Scale score ≤12 (29). However, these are
the two largest studies, with a total of 160 patients in the
experimental group and 413 patients in the comparison
group. All of these patients were taking a concurrent mood
stabilizer or an atypical antipsychotic. Patients treated
with an antidepressant (paroxetine, imipramine, or flu-
oxetine) were more likely to reach remission than those
who were not taking an antidepressant (risk ratio=1.41,
95% CI=1.11–1.80) (Figure 2). The number needed to treat
was 8.4 (95% CI=4.8–33).
Switching to Mania
Two studies had predefined criteria for switching to ma-
nia, as DSM-III-R mania or hypomania (30) or a score ≥15
on the Young Mania Rating Scale (29). One study did not
specifically describe how they monitored the emergence
of mania (31). One study reported the absence of emer-
gence of manic states (32). One small study did not ad-
dress or report switches to mania but was included in the
analysis on the inference that no mania did occur (33).
There was no evidence of an increased risk of switching
to a manic episode in the trials (risk ratio=1.00, 95% CI=
0.47–2.13) (Figure 3). However, there were very few manic
events; hence, of course, this would have limited the
power to detect a difference between antidepressant and
placebo. The risk of switching for antidepressants was
3.8%, and for placebo, it was 4.7% (difference=0.9%, 95%
CI=–2.0 to 3.8).

Page 4
1540
Am J Psychiatry 161:9, September 2004
ANTIDEPRESSANTS FOR BIPOLAR DEPRESSION
Overall Acceptability
The dropout rate was 32% in the antidepressant group,
49% in the placebo group, and 43% in total. There were
significantly fewer study withdrawals while subjects were
taking antidepressants (fixed-effect risk ratio=0.71, 95%
CI=0.58–0.88; χ2=1.82, df=3, p=0.61, for heterogeneity).
Sensitivity Analysis
We performed a sensitivity analysis excluding the two
studies with less than 100% bipolar patients (33, 36). This
did not materially alter the estimate of treatment effect
(fixed-effect risk ratio=1.61, 95% CI=1.27–2.04) and re-
duced heterogeneity (χ2=2.71, df=1, p=0.10). The risk of
withdrawal also remained unchanged (fixed-effect risk ra-
tio=0.74, 95% CI=0.60–0.92).
Tricyclic Antidepressants
Versus Other Antidepressants
This comparison includes 134 patients from the anti-
depressant arms of two trials in which two antidepres-
sants were compared with placebo (31, 30) and 236 pa-
tients from four trials with two antidepressant arms and
no placebo group (34–37). In total, 201 of the 370 patients
(54%) distributed throughout all of these studies were re-
ceiving co-therapy with mood stabilizers.
Clinical Response
A dichotomous measure of clinical response was
reported in five trials. This was based on scores on the
Hamilton depression scale in five trials and on the Clinical
Global Impression scale in one trial. One trial did not report
such a measure and was therefore excluded from this anal-
ysis (30). Tricyclic antidepressants may be less effective
than other antidepressants, but this did not reach statistical
significance (risk ratio=0.84, 95% CI=0.67–1.06) (Figure 4).
Clinical Remission
There was no difference in remission rates between
imipramine and paroxetine based on one trial only (risk
ratio=0.84, 95% CI=0.47–1.48) (30).
TABLE 1. Characteristics of Randomized, Controlled Trials of Antidepressants for Bipolar Depression
Comparison and Study
Year
N
Country
of Study
Drug Comparison
Concurrent Medication
Antidepressant versus
placebo
Tohen et al. (29)
2003
456
13 countries Fluoxetine, placebo
100% olanzapine
Nemeroff et al. (30)
2001
117
United States Paroxetine, imipramine,
placebo
100% lithium
Cohn et al. (31)
1989
89
United States Fluoxetine, imipramine,
placebo
25% lithium
Himmelhoch et al. (32)
1982
59
(29 with bipolar
disorder)
United States Tranylcypromine, placebo
None
Mendlewicz and Youdim
(33)
1980
58
(34 with bipolar
disorder)
Belgium
Deprenyl, placebo
None
Antidepressant versus
antidepressant
Silverstone (34)
2001
156
Australia,
Africa,
Europe
Imipramine, moclobemide
40%–45% lithium,
6% carbamazepine,
1% sodium valproate
Sachs et al. (35)
1994
15
United States Desipramine, bupropion
100% lithium, sodium
valproate, or carbamazepine
Himmelhoch et al. (36)
1991
56
United States Imipramine,
tranylcypromine
None
De Wilde and Doogan (37)
1982
9
Belgium
Fluvoxamine, clomipramine None
Antidepressant versus other
medicine
Young et al. (38)
2000
27
Canada
Paroxetine, mood stabilizer 100% lithium or sodium valproate
Grossman et al. (39)
1999
16
United States Idazoxan, bupropion
>0% lithium
Bocchetta et al. (40)
1993
30
Italy
Amitriptyline, sulpiride
100% lithium

Page 5
Am J Psychiatry 161:9, September 2004
1541
GIJSMAN, GEDDES, RENDELL, ET AL.
Switching to Mania
Three out of six trials had predefined criteria for switch-
ing to mania, as DSM-III-R mania or hypomania (30, 35) or
a score higher than 5 on the Raskin Mania Scale (36). One
trial used the Young Mania Rating Scale for follow-up but
simply used “withdrawal from the study because of mania”
as a clinical description (T. Silverstone, personal communi-
cation). One study did not specifically describe how the au-
thors monitored the emergence of mania (31). Two studies
reported the absence of emergence of manic states (32, 37).
On average, tricyclic antidepressants caused more
switching to mania than other antidepressants (risk ratio=
2.92, 95% CI=1.28–6.71) (Figure 5). The rate of switching
for tricyclic antidepressants was 10%, and for other anti-
depressants combined, it was 3.2%. This gives an absolute
risk difference of 6.8% (95% CI=1.7%–11.9%).
Overall Acceptability
The dropout risk in the tricyclic antidepressant group
was 37%; in the other groups, it was 30%; and in total, it
was 34%. The tricyclic antidepressants did not cause sig-
nificantly more withdrawals than other antidepressants
(risk ratio=1.23, 95% CI=0.92–1.64).
Subgroups
In the subgroups in Figure 4 and Figure 5, there are pos-
sible differences between treatments, although none was
statistically significant. First, SSRIs may have greater effi-
cacy than tricyclic antidepressants. Second, MAOIs may in-
duce less switching to mania than tricyclic antidepressants.
Sensitivity Analysis
We performed a sensitivity analysis including 33 bipolar
patients from a unipolar randomized, controlled trial of
381 patients comparing moclobemide and imipramine
(25). This trial had been included in several previous
reviews on bipolar depression. We excluded it from the
primary analysis because, although the results for these
patients were presented separately, they were not random-
ized separately. The sensitivity analysis did not alter the
overall pattern of clinical response seen in our primary
analysis.
Duration
(weeks)
Outpatients
(%)
Bipolar Diagnosis
Depression Threshold
Rapid
Cycling
Clinical Response Criterion
8
87
DSM-IV bipolar I plus one
previous mixed/manic
episode requiring a mood
stabilizer or antipsychotic
Montgomery-Åsberg
Depression Rating Scale
score ≥20
39%
≥50% improvement in Montgomery-
Åsberg Depression Rating Scale score;
remission: Montgomery-Åsberg
Depression Rating Scale score ≤12
10
100
DSM-III-R bipolar I plus
affective episode in
previous 5 years
17-item Hamilton depression
scale score >15
Excluded Remission: Hamilton depression scale
score <8
6
100
DSM-III bipolar plus manic
episode in previous 5 years
31-item Hamilton depression
scale score >19 plus Raskin
Depression Scale score >7
>50% reduction in Hamilton
depression scale score
6
100
DSM-III bipolar I (N=10) or
bipolar II (N=19)
Raskin Depression Scale score
>6 plus Schedule for
Affective Disorders and
Schizophrenia rating plus
RDC rating
Marked or moderate improvement in
Clinical Global Impression scale score
5
0
Bipolar disorder with
Feighner criteria
Feighner criteria
Not defined
8
75
DSM-III-R bipolar disorder
17-item Hamilton depression
scale score >15
>50% reduction in Hamilton depression
scale score or Hamilton depression
scale score <10
8
100
DSM-III-R bipolar disorder
31-item Hamilton depression
scale score >19
>50% reduction in Hamilton depression
scale score
6
100
DSM-III bipolar disorder, RDC
bipolar I (N=24) or bipolar II
(N=32)
17-item Hamilton depression
scale score >15 plus item 8
>2 plus anergia plus atypical
Excluded Marked or moderate improvement in
Clinical Global Impression scale score
4
0
One previous manic episode
17-item Hamilton depression
scale score >16 plus
Feighner criteria
>50% reduction in Hamilton depression
scale score
6
100
DSM-IV bipolar I (N=11) or II
(N=16)
17-item Hamilton depression
scale >15
Excluded Not reported
6
100
DSM-IV bipolar I
17-item Hamilton depression
scale >15
Not defined
4
100
DSM-III-R bipolar disorder
17-item Hamilton depression
scale >17
>50% reduction in Hamilton depression
scale score

Page 6
1542
Am J Psychiatry 161:9, September 2004
ANTIDEPRESSANTS FOR BIPOLAR DEPRESSION
Antidepressants Versus Other Medications
Only three small trials were identified comparing anti-
depressants with different types of agents, and therefore,
no meta-analyses were performed. One study (N=30)
compared an antidepressant with a second mood stabi-
lizer but was unclear about which drugs were specifically
used by the different groups of patients and did not report
dichotomous response data (38). The second study (N=27)
compared a low dose of amitriptyline (average 62 mg/day)
with a low dose of sulpiride (average 55 mg/day) and re-
FIGURE 1. Fixed-Effect Model of Clinical Response in Randomized, Controlled Trials of Antidepressants Versus Placebo for
the Treatment of Bipolar Depression
a Significance test for heterogeneity (χ2=10.51, df=3, p=0.01; I2=71.4%). Significance test for overall effect (z=5.60, p<0.00001).
FIGURE 2. Fixed-Effect Model of Clinical Remission in Randomized, Controlled Trials of Antidepressants Versus Placebo for
the Treatment of Bipolar Depression
a Significance test for heterogeneity (χ2=0.54, df=1, p=0.46; I2=0%). Significance test for overall effect (z=2.79, p=0.005).
FIGURE 3. Fixed-Effect Model of Switching to Mania in Randomized, Controlled Trials of Antidepressants Versus Placebo for
the Treatment of Bipolar Depression
a Significance test for heterogeneity (χ2=0.18, df=2, p=0.91; I2=0%). Significance test for overall effect (z=0.01, p=0.99).
0.1
0.2
0.5
1.0
Favors antidepressant
Favors placebo
Risk Ratio (fixed)
±95% CI
Placebo
N of Subgroup/
Total N
Antidepressant
N of Subgroup/
Total N
Study/
Subcategory
Risk
Ratio
(fixed)
Weight
(%)
2.0
5.0
10.0
1.88
5.54
2.90
1.44
13.14
Mendlewicz et al. 1980 (33)
27/39
7/19
20/28
4/31
30/60
5/29
46/86
137/370
5.30
9.41
72.14
1.86
1.01–3.51
2.15–14.23
1.26–6.69
1.14–1.83
1.49–2.30
100.00
Himmelhoch et al. 1982 (32)
Cohn et al. 1989 (31)
Tohen et al. 2004 (29)
213
449
123
153
Total events
Total (95% CI)a
95% CI
0.1
0.2
0.5
1.0
Risk Ratio (fixed)
±95% CI
Placebo
N of Subgroup/
Total N
Antidepressant
N of Subgroup/
Total N
Study/
Subcategory
2.0
5.0
10.0
1.20
1.50
29/74
14/43
40/86
115/370
28.99
71.01
1.41
100.00
0.72–2.02
1.14–1.97
1.11–1.80
Nemeroff et al. 2001 (30)
Tohen et al. 2004 (29)
160
413
69
129
Total events
Total (95% CI)a
Favors antidepressant
Favors placebo
Risk
Ratio
(fixed)
Weight
(%)
95% CI
0.1
0.2
0.5
1.0
Risk Ratio (Fixed)
±95% CI
Placebo
N of Subgroup/
Total N
Antidepressant
N of Subgroup/
Total N
Study/
Subcategory
2.0
5.0
10.0
Not estimable
0.97
0.77
1.13
Himmelhoch et al. 1982 (32)
0/28
0/31
Not estimable
Mendlewicz et al. 1980 (33)
0/39
0/19
2/60
1/29
4/74
3/43
5/86
19/370
10.95
30.83
58.22
1.00
0.09–10.23
0.18–3.30
0.43–2.95
0.47–2.13
100.00
Cohn et al. 1989 (31)
Nemeroff et al. 2001 (30)
Tohen et al. 2004 (29)
287
492
11
23
Total events
Total (95% CI)a
Favors antidepressant
Favors placebo
Risk
Ratio
(fixed)
Weight
(%)
95% CI

Page 7
Am J Psychiatry 161:9, September 2004
1543
GIJSMAN, GEDDES, RENDELL, ET AL.
ported response rates of 80% or higher in both groups (40).
A third study (N=16) compared bupropion with idazoxan,
a selective α2 antagonist, but only two of nine and three of
seven patients in the respective groups responded (39).
Discussion
We raised five current concerns in the introduction, and
we will address them as follows.
Comparison With Previous Reviews
This review demonstrates the advantages of a systematic
search when reviewing the literature. Thus, we included
five trials that were not included in previous reviews; two
small studies compared antidepressants with other drugs
not in wide use (39, 40). Another consisted of nine bipolar
patients who were separately randomly assigned within a
larger study (37). Finally, we included two placebo-con-
trolled studies with fewer than 100% bipolar patients. One
of these was included because the other 50% of patients
suffered from anergic depression, which is very similar to
typical bipolar depression (32). The other included 65%
patients with bipolar depression, which is substantially
more than any other trial with mixed inclusion (33).
We did not include several trials that have been cited in
recent reviews on bipolar depression. One is a trial with no
bipolar patients at all, since the patients are described as
having manic-depressive psychosis, depressed type, the
ICD-9 description for unipolar depression (21). This study
was erroneously cited in several reviews on bipolar de-
pression (41–43). In two other trials comparing bupropion
with placebo in unipolar depression (22, 23), the majority
of patients had a diagnosis of DSM-II manic-depressive,
depressed, which is the DSM-II description for unipolar
depression (“manic-depressive, circular” is used for bipo-
lar patients). These two trials have been used to advocate
the use of bupropion as a first antidepressant choice in bi-
FIGURE 4. Fixed-Effect Model of Clinical Response in Randomized, Controlled Trials of Tricyclic Antidepressants Versus
Other Antidepressants for the Treatment of Bipolar Depression
a Significance test for overall effect (z=1.96, p=0.05). Significance test for heterogeneity (χ2=0.89, df=1, p=0.35; I2=0%).
b Significance test for overall effect (z=2.69, p=0.007).
c Significance test for overall effect (z=0.95, p=0.34).
d Significance test for overall effect (z=0.37, p=0.71).
e Significance test for overall effect (z=1.46, p=0.14). Significance test for heterogeneity (χ2=11.26, df=4, p=0.02; I2=64.5%).
0.1
0.2
0.5
1.0
Favors tricyclic
antidepressant
Favors other
antidepressant
Risk Ratio (fixed)
±95% CI
Other
Antidepressant
N of Subgroup/
Total N
Tricyclic
Antidepressant
N of Subgroup/
Total N
Study/
Subcategory
Tricyclics versus SSRIs
Tricyclics versus MAOIs
Tricyclics versus reversible inhibitors of MAO-A
Tricyclics versus bupropion
2.0
5.0
10.0
0.27
0.67
1/5
3/4
12/30
18/30
4.04
21.80
0.60
25.83
De Wilde and Dogan 1982 (37)
Cohn et al. 1989 (31)
35
34
13
21
Total events
Subtotal (95% CI)a
0.48
10/28
21/28
25.43
0.48
25.43
Himmelhoch et al. 1991 (36)
28
28
10
21
Total events
Subtotal (95% CI)b
1.17
40/75
37/81
43.08
1.17
43.08
Silverstone 2001 (34)
75
81
40
37
Total events
Subtotal (95% CI)c
1.14
5/7
5/8
5.65
1.14
5.65
Sachs et al. 2001 (35)
7
8
5
5
Total events
Subtotal (95% CI)d
0.84
0.04–1.68
0.39–1.13
0.36–1.00
0.28–0.82
0.28–0.82
0.85–1.60
0.85–1.60
0.56–2.33
0.56–2.33
0.67–1.06
100.00
145
151
68
84
Total events
Total (95% CI)e
Risk
Ratio
(fixed)
Weight
(%)
95% CI

Page 8
1544
Am J Psychiatry 161:9, September 2004
ANTIDEPRESSANTS FOR BIPOLAR DEPRESSION
polar depression (42, 44). Bupropion also takes a promi-
nent place in recent expert guidelines (45). The actual evi-
dence for the use of bupropion is very limited. Apart from
some open studies, there is only one small study, which
does not substantiate its advocated lower risk of switching
to mania (35).
Antidepressant Efficacy
A quantitative review of the existing randomized data
confirms that the numbers of bipolar depressed patients
entered into clinical trials is around 1% of the figures for
unipolar depression. However, in acceptance of that limi-
tation, the data nevertheless strongly support an average
positive efficacy for antidepressants versus placebo in tri-
als up to 10 weeks. There is no coherent basis for doubting
that conventional antidepressants have efficacy in bipolar
depression on the basis of the existing evidence. The size
of the antidepressant effect was comparable to that in uni-
polar depression (46), suggesting that antidepressants
may be of comparable efficacy in unipolar and bipolar de-
pression. Withdrawal rates also tended to be higher from
placebo, compared with active treatment arms. This sug-
gests a greater acceptability of active treatment, probably
due to lower efficacy for placebo.
It is the advantage of a systematic review that individual
negative trials are appropriately weighted. In a qualitative
review, a trial that fails to achieve significant results may
receive unwarranted attention at the expense of an overall
pattern of positive findings (or vice versa).
Switching to Mania
The rates of switching to mania in the short time spans
of 4 to 10 weeks covered by these studies were low and lent
no support to the belief that switching is a common early
complication of treatment with antidepressants. However,
the numbers were small, and larger studies, especially
FIGURE 5. Fixed-Effect Model of Switching to Mania in Randomized, Controlled Trials of Tricyclic Antidepressants Versus
Other Antidepressants for the Treatment of Bipolar Depression
a Significance test for overall effect (z=1.78, p=0.08). Significance test for heterogeneity (χ2=0.05, df=1, p=0.82; I2=0%).
b Significance test for overall effect (z=0.75, p=0.45).
c Significance test for overall effect (z=1.47, p=0.14).
d Significance test for overall effect (z=0.74, p=0.46).
e Significance test for overall effect (z=2.53, p=0.01). Significance test for heterogeneity (χ2=1.35, df=4, p=0.85; I2=0%).
0.1
0.2
0.5
1.0
Favors other
antidepressant
Favors tricyclic
antidepressant
Risk Ratio (fixed)
±95% CI
Other
Antidepressant
N of Subgroup/
Total N
Tricyclic
Antidepressant
N of Subgroup/
Total N
Study/
Subcategory
2.0
5.0
10.0
5.00
8.10
2/30
0/30
4/39
0/35
7.26
7.65
6.59
14.91
Cohn et al. 1989 (31)
Nemeroff et al. 2001 (30)
Not estimable
0/5
0/4
De Wilde and Dogan 1982 (37)
74
69
6
0
Total events
Subtotal (95% CI)a
1.67
5/28
3/28
43.59
1.67
43.59
Himmelhoch et al. 1991 (36)
28
28
5
3
Total events
Subtotal (95% CI)b
3.24
6/75
2/81
27.94
3.24
27.94
Silverstone 2001 (34)
75
81
6
2
Total events
Subtotal (95% CI)c
2.29
2/7
1/8
13.56
2.29
13.56
Sachs et al. 1994 (35)
7
8
2
1
Total events
Subtotal (95% CI)d
2.92
100.00
0.25–99.95
0.45–145.29
0.83–52.54
0.44–6.31
0.44–6.31
0.67–15.56
0.67–15.56
0.26–20.13
0.26–20.13
1.28–6.71
184
186
19
6
Total events
Total (95% CI)e
Tricyclics versus SSRIs
Tricyclics versus MAOIs
Tricyclics versus reversible inhibitors of MAO-A
Tricyclics versus bupropion
Risk
Ratio
(fixed)
Weight
(%)
95% CI

Page 9
Am J Psychiatry 161:9, September 2004
1545
GIJSMAN, GEDDES, RENDELL, ET AL.
with longer follow-up and systematic monitoring of manic
symptoms, could change this conclusion. However, cur-
rent guidelines for treatment of bipolar depression with
antidepressants tend to stress use in the short term and
early discontinuation. The present findings are relevant to
that pattern of use.
We could not show a difference between an antidepres-
sant as monotherapy compared with when added to a
mood stabilizer. To our knowledge, no trial has addressed
this directly, and the scope for comparison between trials
was limited by different definitions of switching and,
probably, by different baseline rates of switching.
Interpretation of the data could have been improved by
better definition and reporting of switching to mania, es-
pecially when it occurred in patients who had responded
to treatment. Future studies also will have to pay more at-
tention to the definition, follow-up, and reporting of
manic symptoms.
Differences Between Antidepressants
At present, the data are inadequate to definitively favor
one medicine over another within the several generic cat-
egories of the antidepressants. Our review, however, tends
to support the findings of nonrandomized studies in bipo-
lar patients and comparative studies in unipolar patients
that have suggested a higher risk of switching to mania for
tricyclic antidepressants (47, 48). Our results suggest that
tricyclic antidepressants cause more switching to mania,
and tricyclic antidepressants are not more, and may even
be less, effective. This suggests that antidepressant re-
sponse and mania may not be correlated, and indeed, the
pharmacology of tricyclic antidepressants may make
them less suitable for bipolar patients in general.
We did not include lamotrigine in this review since it is
pharmacologically unlike existing antidepressants. How-
ever, data from the one published study in acute bipolar
depression suggest that the effect size for the efficacy of
antidepressants against placebo is comparable to that of
lamotrigine (49). Lamotrigine, 200 mg/day, induced ma-
nia in 3% of patients (two of 63), which is no different from
the risks we report for both nontricyclic antidepressants
and placebo. Lamotrigine, 50 mg/day, was ineffective.
Therefore, while there is no clear short-term benefit for
lamotrigine over antidepressants for bipolar depression, it
may be an alternative option for treatment.
The Antidepressant Effect of Mood Stabilizers
Only one small study in patients already taking a mood
stabilizer compared the addition of an antidepressant
with the addition of a second mood stabilizer. No firm
conclusions can be drawn from that study because of its
size and the absence of information on clinical response
(38). Future large studies will need to address the efficacy
of mood stabilizers alone in more detail. These studies
should look at the long-term outcome of depression, qual-
ity of life, and the focus on duration of remission as well as
response.
Conclusions
We will compare our current conclusions with the re-
cent APA Practice Guideline for the Treatment of Patients
With Bipolar Disorder (50) on two key points. First, there is
no strong reason to avoid antidepressants for patients
with bipolar depression. This is at odds with the recom-
mendation to use lithium or lamotrigine as a first-line of
treatment for bipolar depression. For patients already tak-
ing a mood stabilizer, we advise adding an antidepressant
as a first-line treatment. For patients not taking a mood
stabilizer but with a history of mania, the current consen-
sus is to use antidepressants in combination with an anti-
manic agent or a mood stabilizer (51). Much of the data re-
viewed here reflect that practice.
Whether nontricyclic antidepressants can safely be
used as monotherapy, especially in bipolar II depression,
is not excluded by the existing data. Current APA guide-
lines suggest that the rate of switching is less of a worry in
the acute treatment of bipolar II than in bipolar I depres-
sion and that, therefore, antidepressants can be added
earlier in treatment. This is neither supported nor contra-
dicted by the evidence from randomized, controlled trials.
Consensus views are heavily influenced by clinical experi-
ence and formal audit, not randomized, controlled trials.
Second, it may be prudent to use an SSRI or an MAOI
rather than a tricyclic antidepressant or bupropion as a
first-line treatment. This also differs from the guidelines,
which specifically recommend paroxetine and bupropion
as antidepressants. We see only very limited evidence for
bupropion and only limited evidence for a special place
for paroxetine among the SSRIs. To prefer either is to move
beyond the evidence.
Obviously, our conclusions are based on the results of
randomized, controlled trials only, and we have not con-
sidered evidence from naturalistic and other nonrandom-
ized studies. However, there is no strong signal that anti-
depressants (other than tricyclic antidepressants) cause
mania or even rapid cycling from these studies either (52,
53). On the basis of current evidence, we believe that it is
overcautious and potentially not in the best interest of pa-
tients to discourage the use of antidepressants for bipolar
depression. We appreciate that the existing APA guidelines
do recommend the use of specific antidepressants for se-
vere depression. However, in practice, we have seen cases
in which patients have not been treated with antidepres-
sants and have been left chronically and significantly de-
pressed for very long periods of time. This is almost cer-
tainly one consequence of an emphasis on the first-line
use of mood stabilizers such as lithium and valproate for
bipolar depression, despite the inadequate evidence that
they actually w
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: