Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 96–101

Contents lists available at ScienceDirect

Progress in Neuro-Psychopharmacology & Biological

Psychiatry
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p n p

Transcranial direct current stimulation (tDCS) in unipolar vs. bipolar

depressive disorder
A.R. Brunoni a,f,1, R. Ferrucci a,b,1, M. Bortolomasi c, M. Vergari a,d, L. Tadini a,d, P.S. Boggio e,
M. Giacopuzzi c, S. Barbieri d, A. Priori a,c,⁎
a
Centro Clinico per la Neurostimolazione, le Neurotecnologie ed i Disordini del Movimento, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy
b
Università degli Studi di Milano, Dipartimento di Scienze Neurologiche, Milan, Italy
c
Unità Operativa di Psichiatria, Ospedale Villa Santa Chiara, Verona, Italy
d
Unità Operativa di Neurofisiologia Clinica, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy
e
Cognitive Neuroscience Laboratory and Developmental Disorders Program, Center for Health and Biological Sciences, Mackenzie Presbyterian University, Sao Paulo, Brazil
f
Department of Neurosciences and Behavior, Institute of Psychology, University of Sao Paulo, Sao Paulo, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Transcranial direct current stimulation (tDCS) is a non-invasive method for brain stimulation. Although pilot
Received 29 May 2010 trials have shown that tDCS yields promising results for major depressive disorder (MDD), its efficacy for
Received in revised form 9 September 2010 bipolar depressive disorder (BDD), a condition with high prevalence and poor treatment outcomes, is
Accepted 11 September 2010
unknown. In a previous study we explored the effectiveness of tDCS for MDD. Here, we expanded our
Available online 18 September 2010
research, recruiting patients with MDD and BDD. We enrolled 31 hospitalized patients (24 women) aged 30–
70 years 17 with MDD and 14 with BDD (n = 14). All patients received stable drug regimens for at least two
Keywords:
Bipolar disorder
weeks before enrollment and drug dosages remained unchanged throughout the study. We applied tDCS over
Clinical trial the dorsolateral prefrontal cortex (anodal electrode on the left and cathodal on the right) using a 2 mA-
Major depressive disorder current for 20 min, twice-daily, for 5 consecutive days. Depression was measured at baseline, after 5 tDCS
Non-invasive brain stimulation sessions, one week later, and one month after treatment onset. We used the scales of Beck (BDI) and
Transcranial direct current stimulation Hamilton-21 items (HDRS). All patients tolerated treatment well without adverse effects. After the fifth tDCS
session, depressive symptoms in both study groups diminished, and the beneficial effect persisted at one
week and one month. In conclusion, our preliminary study suggests that tDCS is a promising treatment for
patients with MDD and BDD.2
© 2010 Elsevier Inc. All rights reserved.

1. Introduction
Bipolar disorder is a psychiatric condition with high prevalence
and poor treatment outcomes (Soares-Weiser et al., 2007). Treatment
is associated with cognitive impairment (Henin et al., 2009) and
metabolic changes, leading to non-adherence and treatment discon-
tinuation (Velligan et al., 2009). Research seeking alternative or
adjunctive therapies has led to studies using transcranial direct
current stimulation (tDCS), a neurostimulation technique in which a
Abbreviations: DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, 4th
edition; MDD, major depressive disorder; BDD, bipolar depressive disorder; tDCS,
transcranial direct current stimulation; rTMS, repetitive transcranial magnetic
stimulation; BDI, Beck Depression Inventory; HDRS, Hamilton Depression Rating
Scale 21 items; ANOVA, analysis of variance.
⁎ Corresponding author. Centro Clinico per la Neurostimolazione, le Neurotecnologie
ed i Disordini del Movimento, Fondazione IRCCS Ca' GrandaOspedale Maggiore
Policlinico, Dipartimento di Scienze Neurologiche, Università degli Studi di Milano,
Via F. Sforza 35, Milano, 20122 Italy. Tel.: + 39 2 50320439.
E-mail address: alberto.priori@unimi.it (A. Priori).
1
The first two authors are in alphabetical order.
weak direct current applied on the scalp, reaches the brain and
induces shifts in membrane resting potentials — thereby depolarizing
or hyper-polarizing brain neurons (Nitsche et al., 2003).
Pilot trials investigating the potential therapeutic effects of tDCS,
particularly for treating major depressive disorder (MDD), have raised
considerable interest. For instance, in a study enrolling 10 patients in a
sham-controlled design, Fregni et al. (2006) reported that tDCS
induced antidepressant effects after five sessions. In a larger study,
Boggio et al. (2008) enrolled 40 patients in a similar design, and also
reported positive findings. In an open-label trial enrolling 14 patients
with severe depression, Ferrucci et al. (2009b) also found that
depression improved after tDCS. Similarly, Rigonatti et al. (2008), who
compared a subgroup of Boggio et al. (2008) with 20 patients taking
fluoxetine, found that both groups had similar efficacy after 6 weeks
of treatment, but the tDCS group showed a fast, sustained response at
week-2. Finally, double-blinded, controlled study recruiting 40
patients Loo et al. (2009) found that tDCS had no efficacy after
5 days of stimulation; but achieved positive results after 10 days of
stimulation during the open-label phase. This observation also
suggests that a longer observation period is necessary to distinguish

0278-5846/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.pnpbp.2010.09.010
 A.R. Brunoni et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 96–101 97

a true from a sham tDCS response, as observed in pharmacotherapy Table 1

trials. Clinical and demographic characteristics of the sample.

No studies have investigated tDCS for patients with bipolar de- Unipolar Bipolar Significance
pressive disorder (BDD) and the few treating these patients with depressive depressive (p)
another brain stimulation therapy, repetitive transcranial magnetic disorder disorder

stimulation (rTMS), have reported mixed results (Brunoni et al., n 17 14

2010). In an early double-blinded, sham-controlled trial in 23 patients Age in years — mean (SD) 54.41 (11.16) 44.21 (12.94) p = 0.03
Gender (F/M) 12/5 11/3 p = 0.6
Nahas et al. (2003) failed to show positive results. More recently,
Months of disease — mean (SD) 17 (12.42) 19.29 (9.73) p = 0.58
Dell'Osso et al. (2009) demonstrated TMS efficacy in a open-label Prior suicidal attempts 2 9 p b 0.01
study with 11 patients; and Cohen et al. (2010) in a naturalistic, Prior ECT use 7 6 p = 0.92
follow-up study with 56 patients. Severe/moderate depression 8/9 11/3 p = 0.17
To expand current information on brain stimulation as an emerging Treatment-resistant depression 14 11 p = 0.64
Baseline BDI scores — 26.76 (11.62) 34.21 (8.33) p = 0.06
treatment for patients with depressive disorders, in this exploratory
mean (SD) T1
proof-of-concept study we investigated the effectiveness of tDCS in BDI scores at T2 18.29 (9.69) 27.79 (11.07) p = 0.02
patients hospitalized for acute MDD and acute BDD. BDI scores at T3 (RI) 19.96 (9.17) 18.42 (11.77) p = 0.68
BDI scores at T4 (RI) 20.3 (11.06) 17.25 (11.84) p = 0.46
BDI scores at T4 (LOCF) 20.71 (9.85) 14.43 (8.38) p = 0.07
2. Methods
Baseline HDRS scores — 21.53 (6.83) 30.77 (8.96) p b 0.01
mean (SD) T1
2.1. Patients HDRS scores at T2 16.65 (7.3) 25.23 (10.43) p = 0.01
HDRS scores at T3 (RI) 21.31 (9.22) 19.94 (10.82) p = 0.71
We enrolled 31 hospitalized patients aged 30–70 years, 17 with HDRS scores at T4 (RI) 20.35 (7.68) 15.39 (8.07) p = 0.1
HDRS scores at T4 (LOCF) 20.65 (10.38) 16.86 (9.65) p = 0.3
MDD and 14 with BDD diagnosed according to DSM-IV criteria.
Response (BDI) 3 9 p = 0.01
Patients with schizophrenia, substance use disorders, eating dis- Response (HDRS) 2 4 p = 0.22
orders, personality disorders, mental retardation and other severe Remission (BDI) 4 5 p = 0.36
medical conditions were excluded. All patients were using psychiatric Remission (HDRS) 1 3 p = 0.30
drugs at stable doses for at least two weeks before enrollment and HDRS = Hamilton Depression Rating Scale; BDI = Beck Depression Inventory; SD =
continued the same treatment throughout the study — no interven- standard deviation; RI = regression imputation method; LOCF = Last Observation
tions were started or modified. The MDD sample consisted mainly of Carried Forward.

patients from our previous study (Ferrucci et al., 2009b) investigating

tDCS for MDD.
The study was performed in accordance with the Declaration of
Helsinki. Written, informed consent was obtained from all partici-
pants before inclusion in the study, and the procedures were
approved by the local institutional review board. The patients were
evaluated and enrolled for treatment within 48 h of hospitalization.
2.2. Depression measurement
We used the Hamilton Rating Depression Scale (HDRS) (Hamilton,
1960) and the Beck Depression Inventory (BDI) (Beck et al., 1961) to
evaluate depression. HDRS was scored by a trained experienced
psychologist who was not blinded to the patients' treatment, while
the BDI was self-applied. We defined clinical response as an
improvement of N50% in HDRS; and remission as an HDRS of ≤7 or
a BDI score of ≤10. Finally, because no consensual definition exists, we
defined refractory BDI according to Gitlin (2006), as no symptom
remission after two or more mood stabilizer trials.
2.3. Study design
Initially, we designed a double-blind study, and collected data for
8 patients (3 in the active arm and 5 in the sham arm). Later, when we
noticed that patients on sham intervention could be clearly dis-
tinguished from the others because their symptoms (Table 1)
including suicidal ideation persisted we decided for an open-label
study based on the following issues: (1) the low sham response
observed in previous studies and also in ours (Table 1); (2) the
remarkably more severe disease in our study sample than in usual
phase III trials (Wisniewski et al., 2009) and (3) the ethical concern of
maintaining suicidal patients in sham conditions.
All patients received five stimulation sessions of 20 min, twice a
day at least 4 h apart. Treatment lasting five consecutive days was
chosen because other initial studies using tDCS for MDD (Fregni et al.,
2006) and rTMS for MDD (Pascual-Leone et al., 1996) also opted for
this approach. Moreover, an rTMS studies study applying stimulation
twice a day achieved positive results (Loo et al., 2007).
The patients were evaluated at baseline (T1), immediately after
five tDCS sessions (T2), one week after (T3), and one month after (T4)
treatment onset.
2.4. tDCS
Direct current generated by an electrical stimulator was bilaterally
delivered through a pair of saline-soaked surface sponge electrodes.
Stimulation was applied at an intensity of 2 mA (current density:
0.06 mA/cm2) for 20 min, twice-daily. The anodal electrode was
placed over the left dorsolateral prefrontal cortex (DLPFC) and
cathode electrode over the right DLPFC, 5 cm laterally and 5 cm
ventrally from the center of the scalp (where the sagittal and coronal
planes cross). The electrodes were thick (0.3 cm), rectangular saline-
soaked synthetic sponges (surface area of 35 cm2). The total charge
applied was 0.086 C/cm2.
2.5. Statistical analysis
We used SPSS statistical software (version 16, Chicago, Illinois,
USA) and Microsoft Excel for data analysis. Baseline demographic and
clinical data were compared using the Fisher's exact test (for
categorical variables) or the unpaired t-test (for continuous vari-
ables). All repeated-measures analyses of variation (ANOVA) used
linear assumptions and Mauchly's test was used to assess sphericity.
Two repeated-measures ANOVA were used to address the acute
and long-lasting effects of tDCS. In the “acute” model the dependent
variable (time) therefore had two levels (T1, T2) and in the “long-
lasting” model it had four levels (T1, T2, T3 and T4). The independent
variable was group (bipolar vs. unipolar depression). We analyzed the
main effects of time and group and the interaction effects of
time × group for BDI and HDRS for each model. In addition, we
constructed further mixed models controlling for gender (male vs.
female), age (continuous), and baseline depression severity (moder-
ate vs. severe). Post hoc tests were run when the model was significant
at a p level of ≤0.05. We did exploratory analyses only in the BDD
98 A.R. Brunoni et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 96–101

Table 2
Baseline characteristics of the 31 patients in our sample.

Patient Gender Age (years) Suicide attempt Previous ECT use Bipolar disorder Antidepressant Mood stabilizer Baseline BDI Baseline HDRS

1 Female 59 No No No SNRI + TC – 33 32
2 Female 27 Yes No Yes SNRI LMT 39 34
3 Male 32 No No No SSRI + TC – 16 25
4 Female 71 Yes Yes Yes SNRI OLZ 23 14
5 Male 67 No Yes No TC – 18 15
6 Female 56 no no Yes – OLZ + DVP 41 25
7 Female 48 No Yes Yes SNRI + SSRI QTP 26 22
8 Male 32 Yes No Yes SSRI LMT 36 38
9 Female 52 No No No SSRI + TC – 31 14
10 Female 53 No Yes Yes – Li 38 32
11 Male 34 No Yes No NDRI + TC – 17 16
12 Female 54 No No No TC – 17 19
13 Female 40 No No Yes SSRI QTP 24 28
14 Male 26 Yes No Yes TC Li 35 28
15 Female 58 Yes No No SSRI + NaSSA – 18 18
16 Female 46 No Yes No SNRI + TC – 52 34
17 Female 68 No Yes No NaSSA – 23 21
18 Female 33 Yes Yes Yes TC + SNRI QTP 36 35
19 Female 68 No No No SNRI + SSRI – 29 15
20 Female 67 No No No SSRI + NaSSA – 30 24
21 Female 48 Yes Yes Yes SSRI ARP 43 45
22 Female 61 No Yes No NaSSA – 18 20
23 Female 37 Yes No Yes SSRI + TC DVP 42 43
24 Female 58 No No No SSRI + SNRI – 21 28
25 Male 51 No No No NaSSA – 15 12
26 Male 50 No No Yes SNRI + TC LMT 17 20
27 Male 40 No No No SSRI + TC + NaSSA – 29 20
28 Female 54 Yes Yes No SSRI + NaSSA – 52 33
29 Female 56 No Yes No SSRI + TC – 36 20
30 Female 40 Yes Yes Yes SNRI QTP 43 35
31 Female 58 Yes No Yes TC OLZ 36 26

HDRS = Hamilton Depression Rating Scale; BDI = Beck Depression Inventory. SSRI = Selective Serotonin Reuptake Inhibitors; SNRI = Serotonin and Noradrenalin Reuptake
Inhibitors; TC = Tricyclic Antidepressants; NaSSA = Noradrenergic and Specific Serotoninerg Antidepressants; NDRI = Noradrenaline and Dopamine Reuptake Inhibitors; LMT =
lamotrigine; DVP = sodium divalproate; QTP = quetiapine; OLZ = olanzapine; ARP = aripiprazole; Li = Lithium.

group because the MDD group had already been studied (Ferrucci
et al., 2009b).
Although no patients dropped out, for technical reasons we had
about 5–10% of missing data at T2 and T3. The missing data were
considered at random and missing values were substituted with
regression imputation methods.
3. Results
tDCS treatment was well tolerated, and none of the patients
manifested treatment-emergent mania or other previously reported
adverse effects (Arul-Anandam et al., 2010; Baccaro et al., 2010).
Although both groups were comparable regarding gender, prior use of
ECT, severity, refractoriness of depression, and duration of disease
they differed in age and prior suicidal attempts (Table 2). Also,
baseline HDRS depression scores were higher in BDD.
All patients except two were receiving antidepressants (Table 3),
the most used being venlafaxine, mirtazapine, sertraline, and
escitalopram. All patients with BDD were taking mood stabilizers,
either antipsychotics (three patients were taking olanzapine, four
quetiapine, and one aripiprazole), anticonvulsants (two patients were
taking sodium divalproate, and three lamotrigine) or lithium (two
patients).
BDI and HDRS scores showed a significant time effect as soon as
the fifth tDCS session (Table 4), showing that tDCS was an effective
treatment. When treatment began, the time × group interaction was
not significant but at T4 it became significant for both BDI (p = 0.01)
and HDRS scores (p = 0.03). Post hoc analyses at each time point
showed no differences, however, between groups (for BDI — Level 1
vs. Level 2, p = 0.63; Level 2 vs. Level 3, p = 0.06; Level 3 vs. Level 4,
p = 0.74 and for HDRS — Level 1 vs. Level 2, p = 0.8; Level 2 vs. Level 3,
p = 0.09; Level 3 vs. Level 4, p = 0.69). Finally, the “acute” and “long-
lasting” analyses, controlling for baseline severity, were significant for
BDI (p = 0.01) and HDRS (p b 0.01) whereas gender was significant
only for BDI (p = 0.01) and age only for HDRS (p = 0.02). This
difference suggests that depression severity was positively related
with depression improvement. Finally, the separate analysis in the
BDD group disclosed a general improvement over time, but failed to
identify specific predictors presumably because the wide range of
therapeutic interventions having diverse mechanisms of action
allowed us to explore only the variables age, gender and baseline
severity and not medication.

Table 3
The table shows clinical and demographical data from 5 patients that received sham tDCS treatment. The final BDI/HDRS scores were collected at different periods for each patient
(1–3 weeks), when we decided to stop the double-blind study (see the main text for further details). F = female; M = male. BDI = Beck Depression Inventory; HDRS = Hamilton
Depression Rating Scale.

Patient Gender Age Suicide attempt Drug regimen Baseline BDI/HDRS Final BDI/HDRS

Unipolar MDD F 47 Yes Fluvoxamine 400 mg/day 46/35 46/35

Unipolar MDD M 69 No Sertraline 200 mg/day + Imipramine 50 mg/day 43/44 41/44
Unipolar MDD F 43 Yes Venlafaxine 300 mg/day + Reboxetin 8 mg/day 42/35 40/35
Bipolar Depression M 37 Yes Imipramine 300 mg/day + Valproic Acid 1200 mg/day 34/39 34/39
Unipolar MDD F 32 No Paroxetine 40 mg/day + Imipramine 75 mg/day 38/39 37/36
 A.R. Brunoni et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 96–101 99

Table 4
The table shows the results of repeated-measures ANOVA testing for the main effects of group, i.e., MDD (major depressive disorder) vs. BDD (bipolar depressive disorder), and time
(on the left, baseline and after the fifth tDCS session; on the right, baseline, after the fifth tDCS session, one week after and one month after) and time × group interaction effects,
which tests whether groups performed differently throughout the treatment. We also performed univariate exploratory analyses for the variables Age, Gender and Baseline Severity
used mixed models to address the effects of these variables on the main and interaction effects. All analyses used linear assumptions and respected the assumption of sphericity
according to the Mauchly's sphericity test. Results were considered significant at a p level b0.05 and are in bold in the table.

Variables Acute treatment period (T1–T2) Treatment and Follow-up period (T1–T4)

Scale MDD vs. BD Time effects Time × group MDD vs. BD Time effects Time × group
interaction interaction

F p F p F p F p F p F p

BDI 0.23 0.6 12.1 0.02 0.22 0.63 1.64 0.21 9.88 b 0.01 4.14 0.01

Covariate adjustment for BDI

Age 0.26 0.61 0.02 0.89 0.28 0.6 0.34 0.56 1.9 0.18 0.22 0.86
Gender 0.08 0.77 0.8 0.37 1.95 0.17 0.3 0.5 4 0.01 2.1 0.1
Baseline severity 1.64 0.21 1.99 0.17 0.6 0.44 0.4 0.5 6.72 0.01 1.41 0.24
HDRS 10.3 0.03 16.1 0.01 0.64 0.8 3.77 0.06 4.72 0.04 5.68 0.03

Covariate adjustment for HDRS

Age 1.02 0.32 0.43 0.51 0.98 0.33 0.82 0.37 5.6 0.02 0.32 0.81
Gender 0.8 0.78 0.4 0.85 2.48 0.13 0.36 0.55 1.66 0.18 1.07 0.36
Baseline severity 1.08 0.31 0.6 0.48 0.16 0.69 0.34 0.56 26.5 b 0.01 2.3 0.14

MDD = major depressive disorder; BDD = bipolar depressive disorder; BDI = Beck Depression Inventory; HDRS = Hamilton Depression Rating Scale.

The mean BDI (Fig. 1) and HDRS scores for clinical depression
(Fig. 2) improved over time.
4. Discussion
Our results in this exploratory, proof-of-concept study show that
tDCS effectively improves depressive symptoms in patients with BDD
(Table 5). In our patients, the beneficial effects persisted for at least
four weeks after treatment onset. Our study therefore extends
previous reports describing tDCS as a promising emerging therapy
for major depression (Boggio et al., 2008; Ferrucci et al., 2009a,b;
Rigonatti et al., 2008) also to patients with BDD.
40.00
Mean Beck Depression Inventory
In our study, we were unable to identify whether gender, age or
baseline severity modify the acute and long-term effects of tDCS for
BDD, because our study had a small sample size and was underpow-
ered. Moreover, the patients were taking medications that could have
altered the effectiveness of tDCS. A previous study (Nitsche et al.,
2003) showed that carbamazepine, a drug related to other anti-
convulsants such as lamotrigine and divalproate that many of our
patients were taking, abolished the excitability enhancement induced
by anodal stimulation. Future trials should therefore replicate our
study in drug-free patients.
When we compared the clinical effects of tDCS in patients with
MDD and BDD, the two-way repeated-measures ANOVA suggested a
time × group interaction, implying that improvement differed in the
two groups. Post hoc analyses at each time point failed, however, to
identify a difference. In addition, the BDD group had a higher
depression score (Table 2) associated with a greater improvement
in depression than the MDD group (Table 4). These findings might
explain why depression improved between T1 and T4 solely in

35.00 patients with BDD (Figs. 1 and 2). Because patients with BDD had
higher scores at baseline, they presumably had more “room for
improvement”. Conversely, because patients with MDD had lower
scores at baseline their scores might have improved less than those of
30.00
patients with BDD between T1 and T4 (despite a sustained response)
owing to a “ceiling effect”. In addition, the initial BDD improvement
might be partially explained by the statistical effect of regression to
25.00 the mean due to the high baseline score. Another point to take into
account, given that antidepressant medications differed in the two
groups, is that pharmacotherapy might have augmented the tDCS-
induced improvement in depression. Even though all our patients
20.00 maintained stable drug regimens for at least two weeks before
enrollment, this time is too short to exclude the adverse antidepres-
sant effects that can appear up to six to eight weeks after treatment
(Nakajima et al., 2009). Another possibility is that mood stabilizers
15.00
(used only in the BDD group) played a role in maintaining the initial
tDCS-induced improvement; whereas antidepressants (used in both
Baseline(T1) T2 T3 T4 groups) did not. This hypothesis also acknowledges shorter-lasting
Time tDCS-induced effects in patients with MDD than in those with BDD.
Because all the patients in this study were hospitalized, we were
Fig. 1. The mean Beck Depression Inventory (BDI) scores in patients with bipolar able to apply tDCS twice daily and thus establish that this therapeutic
depression and unipolar depression at 4 evaluation times: Baseline (T1), after the fifth approach is effective and safe, as a previous trial has demonstrated for
transcranial direct current stimulation (tDCS) session (T2), one week later (T3) and one
month later (T4). Error bars represent 95% confidence intervals. The continuous line
rTMS (Loo et al., 2007). We chose a 4-hour interval to elapse between
represents the major depressive disorder (MDD) group; the dashed line represents the stimulations in our previous study (Ferrucci et al., 2009b). This time
bipolar depressive disorder (BDD) group. lapse also fitted in well with the ward's routine. A recent study
100 A.R. Brunoni et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 96–101

(Boutron et al., 2007). Another study limitation, besides being a non-

35.00 controlled study, is that raters were aware of the assigned treatment, i.e.,
they were not blinded. Based on previous experiences, Boggio et al.
(2008) showed a placebo response of 20% whereas in the study
conducted by Fregni et al. (2006) none of the five placebo treatment
30.00
group responded. In the study by Loo et al. (2009) the scores changed by
about 25% in the sham group and active group. Unlike our study,
however, these trials studied unipolar depression because the predictors
of a placebo response for bipolar depression are less well characterized
Mean HDRS

25.00
than those for unipolar depression (Mendlewicz et al., 2010). Hence an
important limitation is that we cannot fully address the amount of
placebo response in our sample. Although we hypothesize that our
20.00 study yielded a low sham response given the severity of our sample,
future studies should use sham-controlled designs to clarify this matter.
Our study had a modest number of missing data that we had to
handle with the proper statistical models. Analyses with and without
15.00 imputation data nonetheless yielded similar results. We also ran several
exploratory analyses. Although these were appropriate for a preliminary
study, false-positive results might have occurred by chance.
In summary, although further controlled studies with larger sam-
10.00
ples are warranted tDCS seems a promising treatment for patients
with unipolar and bipolar depression.
Baseline(T1) T2 T3 T4

Time
Acknowledgements
Fig. 2. The mean Hamilton Depression Rating Scale (HDRS) scores in patients with
bipolar depression and unipolar depression at 4 evaluation times: Baseline (T1), after We are grateful to the three anonymous reviewers for their
the fifth transcranial direct current stimulation (tDCS) session (T2), one week after valuable comments that we believe improved the manuscript. We
stimulation (T3) and one month after stimulation (T4). Error bars represent 95% would also like to thank Alice Crossman for linguistic revision of the
confidence intervals. The continuous line represents the major depressive disorder
paper.
(MDD) group; the dashed line represents the bipolar depressive disorder (BDD) group.

References
(Monte-Silva et al.) nevertheless suggested that for twice-daily tDCS, Arul-Anandam AP, Loo C, Mitchell P. Induction of hypomanic episode with transcranial
the second stimulation should be optimally timed to coincide with the direct current stimulation. J ECT 2010;1:68–9.
Baccaro A, Brunoni AR, Bensenor IM, Fregni F. Hypomanic episode in unipolar
aftereffects of the first but not when homeostatic brain regulation depression during transcranial direct current stimulation. Acta Neuropsychiatrica
succeeds, a period that might be of limited neuroplasticity. Hence 2010;22:316–8.
different times elapsing between tDCS might result in different Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring
depression. Arch Gen Psychiatry 1961;4:561–71.
outcomes, an issue that should be addressed in future studies.
Boggio PS, Rigonatti SP, Ribeiro RB, Myczkowski ML, Nitsche MA, Pascual-Leone A, et al.
An important point in our study design is that although we A randomized, double-blind clinical trial on the efficacy of cortical direct current
followed-up our patients for one month, we gave tDCS for only five stimulation for the treatment of major depression. Int J Neuropsychopharmacol
days. In rTMS studies, trials that stimulated patients for longer periods 2008;11:249–54.
Boutron I, Guittet L, Estellat C, Moher D, Hrobjartsson A, Ravaud P. Reporting methods
achieved better results (Gross et al., 2007). Future tDCS trials for MDD of blinding in randomized trials assessing nonpharmacological treatments. PLoS
and BDD should therefore also explore whether the effectiveness of Med 2007;4:e61.
treatment increases as the number of days on stimulation lengthens. Brunoni AR, Lopes M, Kaptchuk TJ, Fregni F. Placebo response of non-pharmacological
and pharmacological trials in major depression: a systematic review and meta-
A limitation of our study is that we changed our initial study design analysis. PLoS ONE 2009;4:e4824.
from a double-blind to an open-label trial because five patients receiving Brunoni AR, Teng CT, Correa C, Imamura M, Brasil-Neto JP, Boechat R, Rosa M, Caramelli
sham tDCS showed a very low response. Low sham responses may P, Cohen R, Del Porto JA, Boggio PS, Fregni F. Neuromodulation approaches for the
treatment of major depression: challenges and recommendations from a working
depend on depression severity and refractoriness (Brunoni et al., 2009) group meeting. Arq Neuropsiquiatr 2010;3:433–51.
and also on some degree of blinding breaking because hospitalized Cohen C, Brunoni AR, Boggio PS, Fregni F. Clinical predictors associated with duration of
patients could share information regarding their treatment and try to rTMS treatment in patients with bipolar depression: a naturalistic study. J Nerv
Ment Dis 2010;9:679–81.
guess the assigned treatment, even tDCS having only mild effects. Hence Dell'Osso B, Mundo E, D'Urso N, Pozzoli S, Buoli M, Ciabatti M, et al. Augmentative
contact between participants in different arms should be avoided repetitive navigated transcranial magnetic stimulation (rTMS) in drug-resistant
bipolar depression. Bipolar Disord 2009;11:76–81.
Ferrucci R, Bortolomasi M, Brunoni AR, Vergares M, Tadini L, Giacopuzzi M, et al.
Comparative benefits of transcranial direct current stimulation (tDCS) treatment in
patients with mild/moderate vs. severe depression. Clin Neuropsychiatry 2009a;6:
Table 5 246–51.
The table shows exploratory analyses in bipolar patients. We used mixed ANCOVA Ferrucci R, Bortolomasi M, Vergari M, Tadini L, Salvoro B, Giacopuzzi M, et al.
models in which the dependent variable (columns) was the scores of BDI or HDRS after Transcranial direct current stimulation in severe, drug-resistant major depression.
5 sessions of tDCS (T2) or one month after (T4). The rows represent univariate analysis J Affect Disord 2009b;118:215–9.
of gender, age, and baseline severity (severe vs. moderate). Fregni F, Boggio PS, Nitsche MA, Marcolin MA, Rigonatti SP, Pascual-Leone A. Treatment
of major depression with transcranial direct current stimulation. Bipolar Disord
Variables BDI (T2) BDI (T4) HDRS (T2) HDRS (T4) 2006;8:203–4.
Gitlin M. Treatment-resistant bipolar disorder. Mol Psychiatry 2006;11:227–40.
F p F p F p F p
Gross M, Nakamura L, Pascual-Leone A, Fregni F. Has repetitive transcranial magnetic
Age 0.6 0.81 5.2 0.05 0.09 0.77 0.54 0.48 stimulation (rTMS) treatment for depression improved? A systematic review and
Gender 2.64 0.13 0.89 0.44 1.74 0.22 0.57 0.47 meta-analysis comparing the recent vs. the earlier rTMS studies. Acta Psychiatr
Baseline severity 0.8 0.9 0.95 0.41 0.9 0.37 0.08 0.79 Scand 2007;116:165–73.
Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:
BDI = Beck Depression Inventory. HDRS = Hamilton Depression Rating Scale. 56–62.
 A.R. Brunoni et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 35 (2011) 96–101
Henin A, Mick E, Biederman J, Fried R, Hirshfeld-Becker DR, Micco JA, et al. Is
psychopharmacologic treatment associated with neuropsychological deficits in
bipolar youth? J Clin Psychiatry 2009;70:1178–85.
Loo CK, Mitchell PB, Mcfarquhar TF, Malhi GS, Sachdev PS. A sham-controlled trial of the
efficacy and safety of twice-daily rTMS in major depression. Psychol Med 2007;37:
341–9.
Loo CK, Sachdev P, Martin D, Pigot M, Alonzo A, Malhi GS, et al. A double-blind, sham-
controlled trial of transcranial direct current stimulation for the treatment of
depression. Int J Neuropsychopharmacol 2009:1–9.
Mendlewicz J, Massat I, Linotte S, Kasper S, Konstantinidis A, Lecrubier Y, Montgomery
S, Serretti A, Zohar J, Souery D. Identification of clinical factors associated with
resistance to antidepressants in bipolar depression: results from an European
Multicentre Study. Int Clin Psychopharmacol 2010;25:297–301.
Monte-Silva KK, Kuo MF, Liebetanz D, Paulus W, Nitsche MA. Shaping the optimal
repetition interval for cathodal transcranial direct current stimulation (tDCS). J
Neurophysiol 2010 Jan 27 [Electronic publication ahead of print].
Nahas Z, Kozel FA, Li X, Anderson B, George MS. Left prefrontal transcranial magnetic
stimulation (TMS) treatment of depression in bipolar affective disorder: a pilot
study of acute safety and efficacy. Bipolar Disord 2003;5:40–7.
Nakajima S, Suzuki T, Watanabe K, Kashima H, Uchida H. Accelerating response to
antidepressant treatment in depression: a review and clinical suggestions. Prog
Neuropsychopharmacol Biol Psychiatry 2010;2:259–64.
101
Nitsche MA, Fricke K, Henschke U, Schlitterlau A, Liebetanz D, Lang N, et al.
Pharmacological modulation of cortical excitability shifts induced by transcranial
direct current stimulation in humans. J Physiol 2003;553:293–301.
Pascual-Leone A, Rubio B, Pallardo F, Catala MD. Rapid-rate transcranial magnetic
stimulation of left dorsolateral prefrontal cortex in drug-resistant depression.
Lancet 1996;348:233–7.
Rigonatti SP, Boggio PS, Myczkowski ML, Otta E, Fiquer JT, Ribeiro RB, et al. Transcranial
direct stimulation and fluoxetine for the treatment of depression. Eur Psychiatry
2008;23:74–6.
Soares-Weiser K, Bravo Vergel Y, Beynon S, Dunn G, Barbieri M, Duffy S, et al. A
systematic review and economic model of the clinical effectiveness and cost-
effectiveness of interventions for preventing relapse in people with bipolar
disorder. Health Technol Assess 2007;11 iii–iv, ix-206.
Velligan DI, Weiden PJ, Sajatovic M, Scott J, Carpenter D, Ross R, et al. The expert
consensus guideline series: adherence problems in patients with serious and
persistent mental illness. J Clin Psychiatry 2009;70(Suppl 4):1-46 quiz 47-8.
Wisniewski SR, Rush AJ, Nierenberg AA, Gaynes BN, Warden D, Luther JF, McGrath PJ,
Lavori PW, Thase ME, Fava M, Trivedi MH. Can phase III trial results of
antidepressant medications be generalized to clinical practice? A STAR*D report.
Am J Psychiatry 2009;4:599–607.