ORIGINAL STUDY
Low-Frequency Repetitive Transcranial Magnetic Stimulation
in the Right Prefrontal Cortex Combined With Partial Sleep
Deprivation in Treatment-Resistant Depression
A Randomized Sham-Controlled Trial
Jelena Krstić, MD, PhD,*† Ivana Buzadžić, MS,†‡ Slađan D. Milanović, PhD,§ Nela V. Ilić, MD, Mr Sci,||
Sanja Pajić, MS,† and Tihomir V. Ilić, MD, PhD†¶
Introduction: Sham-controlled low-frequency repetitive transcranial
magnetic stimulation (rTMS) was used in patients with pharmacoresistant
major depression as an added treatment along with partial sleep deprivation
(PSD). In addition, the potential predictive role of brain-derived neurotrophic
factor genetic polymorphism on treatment response was analyzed.
Methods: We recruited 19 female patients (48.3 ± 8.6 years old) with
treatment-resistant unipolar major depression (Hamilton Depression
Rating Scale [HDRS] score ≥20) who were on a stable antidepressant
treatment. They received either 1-Hz rTMS or sham stimulation over
the right dorsolateral prefrontal cortex (intensity of 110% of the thresh-
old; 3000 stimuli per protocol; and 10 daily sessions). Additionally, PSD
was applied once per week during the treatment. The patients were eval-
uated (HDRS and Clinical Global Impression Scale) by a blind rater at
baseline (B) and after 2 and 3 weeks (W2 and W3) of treatment for
short-term outcome. Long-term evaluations were performed after 12 (W12)
and 24 weeks (W24) for patients who received active stimulation.
Results: Eleven patients in the active group showed a significant
HDRS score reduction from 30.09 ± 3.53 (B) to 16.73 ± 5.71 (W3)
compared to the lack of therapeutic response in the sham-treated
patients. The long-term follow-up for the active group included 64%
of the responders at W12 and 55% at W24. Full remission (HDRS
≤10) was achieved in 5 of 11 patients. Four of these 5 patients with
long-term sustained remission expressed the Val66Val genotype.
Conclusion: Our study suggests a clinically relevant response, persisting
for up to 6 months, from 1-Hz rTMS over the right dorsolateral prefrontal
cortex and PSD in patients with pharmacoresistant major depression. The
brain-derived neurotrophic factor Val66Val homozygous genotype may be
related to a better treatment outcome.
Key Words: major depression, transcranial magnetic stimulation, sleep
deprivation, BDNF
(J ECT 2014;30: 325–331)
From the *Psychiatric Hospital, Clinical Hospital Center “Dr Dragisa Misovic”,
Belgrade, Serbia; †Faculty of Biology, University of Belgrade, Belgrade,
Serbia; ‡Department of Human Genetics and Prenatal Diagnostics, Clinical
Hospital Center “Zvezdara”, Belgrade, Serbia; §Institute for Medical Research,
University of Belgrade, Belgrade, Serbia; ║School of Medicine, University of
Belgrade, Belgrade, Serbia; and ¶Department of Clinical Neurophysiology,
Medical Faculty of Military Medical Academy, University of Defense,
Belgrade, Serbia.
Received for publication July 21, 2013; accepted November 12, 2013.
Reprints: Tihomir V. Ilić, MD, PhD, Department of Clinical
Neurophysiology, 4th Floor, Room 52, Medical Faculty of Military
Medical Academy, Crnotravska 17, 11000 Belgrade, Serbia
(e‐mail: tihoilic@gmail.com).
The authors have no conflicts of interest or financial disclosures to report. This
project was partially supported by grants from the Ministry of Education and
Science of the Republic of Serbia (Project No. 41014and 175012) and the
Ministry of Defense of the Republic of Serbia (Project MFVMA/12/13-15).
Copyright © 2014 by Lippincott Williams & Wilkins
DOI: 10.1097/YCT.0000000000000099
The Journal of ECT • Volume 30, Number 4, December 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
D uring the past decade, repetitive transcranial magnetic stim-
ulation (rTMS) has emerged as a new and promising thera-
peutic tool for pharmacoresistant major depression (MD).
Despite numerous clinical trials in this field, there are still
doubts about the most effective protocol paradigms. After early
clinical trials assessing the influence of the coil position on
treatment efficacy in depression, high-frequency (HF) rTMS
of the left dorsolateral prefrontal cortex (DLPFC) became the
most frequent target in subsequent trials,1 although some re-
searchers have suggested that low-frequency (LF) rTMS of the
right DLPFC could also be effective while being better tolerated
and safer.2 However, in addition to rTMS, there are other biolog-
ical methods with antidepressant effect as it is sleep deprivation
(SD), for which the PET study shows that Hamilton Depression
Rating Scale (HDRS) score reduction after SD correlates with in-
creased relative cerebral glucose metabolism in specific brain
areas (ie, the left medial prefrontal cortex, left frontal pole, and
right lateral prefrontal cortex),3 similar to changes observed after
rTMS treatment in depressive patients.4 In view of these findings,
the 2 studies using 10-Hz rTMS over the left DLPFC combined
with simultaneous SD have shown antidepressant efficacy.5,6
The present study was designed to evaluate the possible
synergistic effects of combined use of LF rTMS to the right
DLPFC and partial SD in patients with treatment-resistant MD,
as well as long-term effectiveness of these procedures.
Furthermore, there is accumulating evidence that brain-
derived neurotrophic factor (BDNF), a protein involved in the
regulation of synaptic transmission and plasticity, determines
the susceptibility to long-term potentiation-like plasticity in-
duced at the system level by rTMS.7
It has been suggested recently that BDNF functional poly-
morphism could represents predictor of rTMS antidepressant
response.8 In this study, we sought to provide additional evi-
dence to confirm the hypothesis.
MATERIALS AND METHODS
Subjects
Nineteen female outpatients (mean age, 48.3 ± 8.6 years)
from the Psychiatric Hospital “Dr. Dragisa Misovic” in Belgrade
participated in this study. The study was approved by the local
ethics committee according to the Declaration of Helsinki (1975).
After an explanation of the treatment procedures, all patients
signed a written informed consent. Diagnosis of nonpsychotic
MD was established by an experienced psychiatrist after a
semistructured interview in accordance with Diagnostic and
Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV )
and Mini-international neuropsychiatric interview criteria.9 Owing
www.ectjournal.com 325
Krstić et al
to the specific ward from which the patients were recruited, our
sample consisted of female patients.
Patients were included in study if they met the following
criteria: older than 18 years; a DSM-IV diagnosis of treatment-
resistant MD; a modified 24-item HDRS score of 20 or greater10,11;
nonpsychotic and moderately to markedly depressed; no personal
history of seizure or brain surgery; and with no medical, neurolog-
ical, or neurosurgical disorder that might preclude the administra-
tion of rTMS. No patient met the DSM-IV criteria for any other
axis I disorder. Any patients with personality disorder, substance
or alcohol abuse, or a history of suicide attempt in the past
6 months were also excluded.
The patients were referred for rTMS after failing to re-
spond to 2 different antidepressant trials of adequate dose and
duration, and they had undergone stable drug treatment for at
least 4 weeks12; thus, rTMS was an add-on treatment. The anti-
depressant medication included the following: 7 patients (37%)
on sertraline, 150–200 mg/d; 4 patients (21%) on venlafaxine,
150–225 mg/d; 4 patients (21%) on escitalopram, 10–20 mg/day;
3 patients (16%) on fluoxetine,40 mg/d; and one patient (5%)
was prescribed clomipramine,125 mg/d. The antidepressant phar-
macotherapy was kept unchanged during the study. Any patient
requiring a change in pharmacotherapy during the study was
excluded.
Efficacy Measures
The primary efficacy outcome was the change in the 24-item
HDRS total score from baseline across the 6-month follow-up
period. Depending on the clinical outcome, the patients were di-
vided into 4 categories: complete clinical remission (HDRS ≤10),
responders (≥50% reduction from baseline), partial responders
(≥25% reduction), and nonresponders (patients with less than
25% HDRS reduction).
For further analysis, the effects on the 4 principal factors of
HDRS (1, depression; 2, loss of motivated behavior; 4, anxiety;
and 5, sleep disturbances) were examined separately.13 Factor 3,
psychosis, was not analyzed because psychotic symptoms were
among the exclusion criteria.
Another important secondary efficacy measure was the
change on the Clinical Global Impression Scale—Severity of
Illness (CGI-S).
The treatment effects were assessed by a blind experienced
rater at baseline (B), immediately after 2 weeks of stimulation
(W2), and in the follow-up period: 3 (W3), 12 (W12), and
24 weeks (W24) after treatment.
Magnetic Stimulation
Transcranial magnetic stimulation was delivered by a
Magstim Stimulator with a figure 8–shaped coil. The resting
motor threshold (RMT) was determined in accordance with in-
ternational standards. Transcranial magnetic stimulation was
delivered at an intensity of 110% of the RMT over the frontal
area overlying the right DLPFC, localized 5 cm anterior to the
hotspot of the abductor pollicis brevis muscle.1
To maintain an auditory click in the sham rTMS condition,
the coil was placed in the same position, perpendicular to the
scalp, with a reduced stimulation intensity of 60% of the RMT.
Experimental Design
We conducted a blind randomized study of active and sham
right prefrontal rTMS in patients with treatment-resistant MD.
After a baseline evaluation, the patients were randomly
assigned to receive 10 sessions over a 2-week period (starting
on a Monday with a pause for the weekend, 5 days/wk) of active
326 www.ectjournal.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Journal of ECT • Volume 30, Number 4, December 2014
or sham rTMS over the right DLPFC. Each daily session of
rTMS consisted of 5 trains of 60 stimuli (300 stimuli/d), with
intertrain pauses of 3 minutes; each session lasted approxi-
mately 20 minutes (a total of 3000 stimuli were presented to
each subject).
Both the patients and the rater (T.V.I.) were blind to the
treatment, although the physician who applied the rTMS (J.K.)
was aware of the methods of stimulation; the rater and the
treating physicians were different persons. Late partial SD was
applied twice during the 2-week period. We chose late partial
SD instead of total SD because of its proven efficiency,14 and
the abbreviated procedure seemed more likely to be acceptable
in patients’ compliance.
Late partial SD was applied twice during the 2-week period.
The patients went to bed as usual, were awakened at approxi-
mately 1:30 AM, and remained awake for the next 20 hours. Fam-
ily members were involved in assuring patients’ compliance.
All of the patients were naive to both TMS and partial SD;
and thus, it was not likely that they could recognize the treat-
ment modality. The sham-treated patients were excluded from
further follow-up after W3; because of ethical reasons, it would
not have been acceptable to continue with unchanged treatment.
BDNF Genotyping
Genomic DNA was extracted from whole blood using a
standard phenol–phenol–chloroform method, and the DNA qual-
ity and quantity was assessed using a spectrophotometer. The
Val66Met single nucleotide polymorphism in the BDNF gene
was typed using polymerase chain reaction (PCR) in a total vol-
ume of 25 μL containing approximately 50 ng of genomic tem-
plate, 1 μm of each primer, 200 μM of each deoxyribonucleotide
triphosphate (dNTP mix), and 25 μL of 10 reaction or PCR
buffer. The primer sequences were SBDNF1-AAA GAA GCA
AAC ATC CGA GGA CAA G and SBDNF2-ATT CCT CCA
GCA GAA AGA GAA GAG G, resulting in a 274-base pair
PCR product. A PerkinElmer 9700 thermal cycler was used for
DNA amplification.
Data Analysis
A 2-way repeated-measures analysis of variance (ANOVA;
time-intervention interaction) was used. Both a between-group
patient factor (group) and a within-patient factor (time) were
considered for HDRS, a subset of HDRS factors and CGI-S score.
For post hoc analysis, a one-way repeated-measure ANOVA with
a 2-tailed Tukey mean honestly significant difference (HSD) com-
parison test was performed on the change in the HDRS scores
from baseline. A χ2 test with Yates’ correction was performed
when appropriate. The results are presented as the mean ± SD,
and the significance level was set at P < 0.05.
RESULTS
Nineteen patients (11 patients treated with rTMS and
8 sham-treated patients) completed the 2-week treatment proto-
col (W2) and were available for short follow-up assessment
(W3). As it already stated in “Materials and Methods,” sham-
treated patients were excluded from long-term follow-up (W12
and W24), because in case of rTMS treatment failure, further
treatments were used. Only one rTMS-treated patient dropped out
of the study after W3 because of treatment failure. Hence, long-
term follow-up was related only to 10 patients treated with rTMS.
The baseline demographic and illness characteristics of the
entire sample are shown in Table 1. There were no significant
© 2014 Lippincott Williams & Wilkins
The Journal of ECT • Volume 30, Number 4, December 2014 Low-Frequency rTMS With Sleep Deprivation

differences in age, duration of current depressive episode, or
HDRS and CGI-S score at baseline between the 2 groups.
Furthermore, the specific clinical features of patients with
regard to the number of previous episodes, and antidepressants
dosages during current depressive episode, immediately before
study entry, are shown in Table 2.
All patients treated with LF rTMS combined with partial
SD improved over time (Fig. 1A). In the real stimulation (RS)
group, only one patient showed treatment failure (a <10% de-
crease in HDRS) and was subsequently excluded from further
analysis. At the group level, the treatment response differed sig-
nificantly at W2, and these changes were maintained throughout
the entire follow-up period.
A 2-way repeated-measures ANOVA was conducted to com-
pare the effect of rTMS with partial SD (intervention) on depres-
sive symptoms throughout the follow-up period (time). There was
a significant reduction in the HDRS score in the RS group across
all 3 time points [F(1,18) = 12.01; P = 0.00011]. Post hoc
comparisons using the Tukey HSD test indicated that the mean
scores in the RS were significantly different from the SS scores
at W2 (t = 4.822; P < 0.05) and W3 (t = 6.367; P < 0.01).
However, when the 2-way repeated-measures ANOVA
analysis was expanded to include individual factors within the
HDRS scale, the results showed a significant reduction of the
factor 1 subscore (depression) [F(1,18) = 5.61; P = 0.007], with
a post hoc Tukey HSD test showing significance at W2 (t = 5.244;
P < 0.01) and W3 (t = 5.142; P < 0.01). Additionally, the factor 2
subscore (motivation) had a between-subject effect [F(1,18) =10.86;
P = 0.004], with a post hoc Tukey HSD test showing significance
at W2 (t = 4.319; P < 0.05) and W3 (t = 4.497; P < 0.05). In con-
trast, a subset data analysis for factor 4 (anxiety) [F(1,18) = 1.33;
P = 0.263] and factor 5 (sleep) [F(1,18) = 0.05, P = 0.826] showed
a lack of statistical significance (Fig. 1B-E).
To determine the duration of the antidepressant effect, fur-
ther analysis was performed using a one-way within-subjects
ANOVA (factor: time) for patients in the RS that showed a sig-
nificant reduction in the full HDRS score [F(1,18) = 51.88;
P = 0.0001], with a post hoc Tukey HSD test indicating differences
between the initial baseline score against all subsequent mea-
surements (P < 0.01; Fig. 1A). The same analysis applied to the
individual factors of the HDRS indicated significant declines in
the score at W2, W3, W12, and W24 compared to the baseline
values, although with no difference between each individual: fac-
tor 1 (depression) [F(1,18) = 38.95; P = 0.0001], post hoc Tukey
HSD, t =1.88; P < 0.01; factor 2 (motivation) [F(1,18) = 22.35;
P = 0.0001], post hoc Tukey HSD, t = 1.5; P < 0.01; factor 4
(anxiety) [F(1,18) = 22.29; P = 0.0001], post hoc Tukey HSD,
t = 1.36; P < 0.01 and factor 5 (sleep) [F(1,18) = 12.47; P =
0.0001], post hoc Tukey HSD, t = 1.26; P < 0.01 (Fig. 1B-E).
Mean CGI-S scores changes were consistent with other
parameters presented. The data are shown in Table 3.
The mean ± SD HDRS scores during the baseline were:
30.1 ± 3.5 for the RS and 28 ± 2.7 in the SS; on W2, RS, 17.5 ±
5.6 and SS, 23.9 ± 3.8; on W3, RS, 16.7 ± 5.7 vs SS, 25.2 ±
4.5; on W12, RS, 13.9 ± 3.8; and on W24, RS,14.8 ± 5.6.
The mean ± SD HDRS reduction from baseline at W2 in the
RS was 39% ± 19% and 17% ± 10% in the SS, similar to
the changes at W3 when a significant 45% ± 17% decrease in
the score (relative to the baseline) was sustained in the RS; in the
SS, the decrease was 17% ± 12%.
Four patients in the RS at W2 were classified as responders,
with an additional 5 patients as partial responders, compared to
only 2 partial responders in the SS (χ2 = 4.9, df = 1, P = 0.0134).
After W3, the patients from the sham group were also
monitored by the treating physician; however, before W12, all
of the patients required changes in pharmacological treatment
and were excluded from further follow-up study.
In addition to the primary purpose of the study, we analyzed
the individual treatment effects in association with polymorphism
in the BDNF gene and found a slightly better, but not significant,
response for a group of Val66Val homozygotes [F(1,10) = 0.05].
However, only one of the 5 patients with Val66Met, who were
followed up to W24, met the criteria for complete clinical

TABLE 1. Demographic and Baseline Clinical Characteristics of 19 Patients With Unipolar Depression

Total Sample (N = 19) Active Treatment (n = 11) Sham Treatment (n = 8) P

Age, mean (SD) 48.3 (8.6) 50.7 (7.3) 46.1 (8.5) n.s.
Range 32–60 41–60 32–57
Current episode duration, mos 6.9 (1.3) 6.8 (1.2) 6.9 (1.5) n.s.
BDNF genotyping
Val66Val 12 (63) 5 (55) 7 (77.8) n.s.
Val66Met 7 (37) 6 (45) 1 (22.2) —
Baseline evaluation, mean (SD)
HDRS 29.45 (3.20) 30.09 (3.53) 28.00 (2.74) n.s.
Range 24–37 26–37 24–34
HDRS—factor 1 9.40 (2.14) 9.36 (2.42) 9.44 (1.88) n.s.
Range 6–13 6–13 6–12
HDRS—factor 2 6.20 (0.89) 6.18 (0.98) 6.00 (0.83) n.s.
Range 4–8 4–7 5–8
HDRS—factor 4 7.05 (0.88) 7.27 (1.00) 6.78 (0.67) n.s.
Range 5–8 5–8 6–8
HDRS—factor 5 3.70 (1.09) 4.00 (1.09) 3.13 (1.00) n.s.
Range 2–6 2–6 2–5
CGI-S 4.50 (0.61) 4.64 (0.50) 4.33 (0.71) n.s.
Range 3–5 3–5 4–5

© 2014 Lippincott Williams & Wilkins www.ectjournal.com 327

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 328
Krstić et al

www.ectjournal.com
TABLE 2. Clinical Characteristics Related to Previous Treatment and BDNF Genetic Polymorphism of Patients

Depressive Episode No. Total Depressive Daily Duration of Antidepressant

Patients Years rTMS Duration (Months) Episodes Lifetime Antidepressants Dosage* Treatment Before Study (Weeks) BDNF Val66Met
1 43 1 Hz 4.5 5 Fluoxetine 60 6 val/met
2 41 1 Hz 7 3 Escitalopram 20 7 val/met
3 57 1 Hz 6 4 Sertraline 250 7 val/met
4 60 1 Hz 6.6 6 Venlafaxine 150 8 val/met
5 59 1 Hz 5.4 6 Clomipramine 150 6 val/met
6 59 1 Hz 6.5 5 Sertraline 200 7 val/met
7 42 1 Hz 8 4 Venlafaxine 225 8 val/val
8 55 1 Hz 7 5 Escitalopram 20 6 val/val
9 49 1 Hz 7.5 3 Sertraline 200 6 val/val
10 56 1 Hz 8.5 4 Escitalopram 20 7 val/val
11 48 1 Hz 8 4 Fluoxetine 40 7 val/val
12 57 Sham 8.5 4 Sertraline 250 6 val/val
13 49 Sham 6.5 5 Sertraline 150 7 val/val
14 33 Sham 4 3 Venlafaxine 225 7 val/val
15 38 Sham 7.2 4 Sertraline 200 6 val/val
16 46 Sham 7.6 5 Venlafaxine 225 8 val/val
17 53 Sham 9 6 Sertraline 200 7 val/val
18 44 Sham 5.5 3 Fluoxetine 40 7 val/val
19 32 Sham 7 5 Escitalopram 20 7 val/val
*During a period of not less than 4 weeks before the start of the study.

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
© 2014 Lippincott Williams & Wilkins
The Journal of ECT • Volume 30, Number 4, December 2014
The Journal of ECT • Volume 30, Number 4, December 2014 Low-Frequency rTMS With Sleep Deprivation

FIGURE 1. A, Effects of LF rTMS combined with partial SD on the HDRS score between the RS (empty squares) and SS (filled circles)
groups of patients with pharmacoresistant depression. All data are the mean values of each group of subjects; the error bars are ± 1 SD.
The times of the HDRS tests are denoted on the x-axis. B–E, The panels represent the HDRS subscores for factor 1, depression;
factor 2, motivation; factor 4, anxiety; and factor 5, sleep. A significant improvement in the HDRS score of the RS group versus the
SS group was achieved at W2 and W3. The asterisks indicate significant differences (Friedmann test, followed by the Tukey mean HSD,
post hoc analysis with asterisk [*] and double asterisk [**] for P < 0.05 and P < 0.01, respectively).

remission (HRDS ≤10); in contrast, 4 of the 5 homozygous
Val66Val patients met the full response criteria (Fig. 2).
DISCUSSION
The most relevant finding of this study is the antidepres-
sant effectiveness, sustained over a 6-month follow-up period,
of LF rTMS combined with partial SD in patients with pharma-
coresistant unipolar depression.
Several studies in recent years have explored the efficacy
of 1-Hz stimulation over the right DLPFC using a double-
blind sham-controlled design.15–18 However, there was a lack
of homogeneity in the patient selection, TMS parameters,
periodicity of clinical assessments, and duration of follow-up
in these studies. Considering the sham rTMS technique with
perpendicular coil orientation and subthreshold intensity (60%
RMT) that we have implemented in our protocol, we have cho-
sen the method that substantially reduced the induced magnetic
field, and has already been successfully applied in a number of
studies.19
In the first double-blind study on the efficacy of LF rTMS,
the mean improvement in the HDRS score was reported to be
47%, with a responder rate of 49% after the 2-week treatment
(1200 magnetic pulses at 110% RMT).17 These results are
slightly better than the results in our study, showing an mean
improvement of 39% at W2 and a responder rate of 36%, with

© 2014 Lippincott Williams & Wilkins www.ectjournal.com 329

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Krstić et al The Journal of ECT • Volume 30, Number 4, December 2014

TABLE 3. Mean Scores of CGI-S From Baseline to W2, W3, and W24

Active Treatment Sham Group P of Independent Sample t test

Baseline CGI-S 4.64 (0.50) 4.33 (0.71) P = 0.139
W2 CGI-S 3.00 (0.89) 3.78 (0.67) P = 0.022
W3 CGI-S 2.55 (0.93) 3.89 (0.78) P = 0.0014
W24 CGI-S 1.91 (0.94) N/A NC
NC indicates not calculated; N/A, not available.

an additional 45% of patients who had only a partial response
(HDRS improvement ≥25%). However, the key differences be-
tween these 2 studies are related to the initial patient character-
istics: in Klein’s study, the patients were not treatment resistant,
and the follow-up period was only one month.
Patient selection criteria similar to ours were used in 2 further
studies, although they included patients resistant to previous phar-
macological treatment.15,16 Hoppner et al15 reported a modest re-
duction in HDRS of 11% on average, with only one patient
classified as a responder. In contrast, the study of Januel et al16 in-
cluded patients who were antidepressant-free after a period of
4 weeks before the rTMS application. This study reported a high
improvement rate with a 1-Hz rTMS protocol, with an mean re-
duction in the HDRS score of 54%, and 63% of the patients
achieved clinical remission. However, an improvement of 26%
in HDRS was observed in the sham-treated group.
Finally, Pallanti et al18 published the results of a 3-group,
double-blind, randomized controlled trial (1-Hz rTMS vs sequential
bilateral stimulation vs sham stimulation for 3 weeks) in patients with
unipolar depression. In this study, LF rTMS showed the best results,
where clinical remission was achieved in 6 of the 20 patients.
In summary, our study results are comparable with the treat-
ment effects observed in several double-blind randomized clinical
trials (RCTs) that considered the effectiveness of a 1-Hz proce-
dure over the right DLPFC.
However, the longest reported period of follow-up used in
our study indicates a longer duration of treatment effect than
any RCTs published to date in this category.
Having regard to the treatment of a variety of antidepressants
among our patients before the study treatment, we believe that
this factor did not significantly affect the outcome of the study,
as patients were on stable therapy for at least 4 weeks before inclu-
sion in the protocol. This is a common study design in pharmaco-
logical trials where a new agent is used as add-on therapy.
Therefore, it is a rational assumption that the combined use
of rTMS and partial SD affects the same cortical regions with a
possibly synergistic effect.
Our results show that in addition to a reduction of the
HDRS total score after LF rTMS and partial SD, significant
changes were also found in the HDRS subscore 1 (depression)
and subscore 2 (loss of motivated behavior), but not subscore 4
(anxiety) and subscore 5 (sleep), compared to the sham-treated
patients. However, during the long-term follow-up (6-month pe-
riod), the effects of the 1-Hz protocol were maintained for
6 months with a significant improvement compared with the base-
line HDRS subscores 1, 2, 4, and 5.
A similar, but relatively weak, trend was shown in a study
with an application of 20-Hz rTMS, where there was an im-
provement trend in all of the subscores with the exception of
the psychosis subscore.20

FIGURE 2. Individual treatment effects in RS and SS groups of patients with pharmacoresistant depression. Patients with val/met
BDNF genetic polymorphism were marked with squares, and patients with val/val by circles. Different filling patterns of squares and
circles, in accordance with the symbols below figure, indicate the individual effects of treatment through follow-up period.

330 www.ectjournal.com © 2014 Lippincott Williams & Wilkins

Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Journal of ECT • Volume 30, Number 4, December 2014
Finally, this is the only RCT where stimulation of 1-Hz
rTMS over the right DLPFC was applied together with an as-
sessment of the treatment effectiveness in relation to BDNF genetic
polymorphism. A number of recent studies and meta-analyses21
support the neurotrophic theory of depression and suggest that im-
provement in MD is related to BDNF neuroplasticity, where
changes could be the final common pathway for different antide-
pressant therapies. To our knowledge, only one short rTMS study
to date has investigated the influence of the BDNF Val66Met
polymorphism on rTMS treatment in depression (1 or 17 Hz)
and found a significantly better response to rTMS treatment in
subjects with BDNF Val66Val, a genetic variant associated with
increased protein activity.8
The obvious drawback of our genetic study is the small
number of patients in the active treatment arm, with only 5 patients
who were homozygous for the BDNF Val66Val variant. Regard-
less, it is interesting to note that the criteria of a full response
(HDRS improvement ≥50%) were already met in 3 of 5 patients
after W2; and by the end of the follow-up period, a full response
was achieved in all of the patients with BDNF Val66Val. In con-
trast, after 2 weeks of treatment, only 1 of 6 patients with BDNF
Val66Met achieved the criteria.
After W2, one of the patients in the Val66Met group was
excluded from further follow-up owing to a lack of response
(HDRS improvement <10%). However, no less significant
was the finding that full remission was achieved in 4 of the
5 Val66Val homozygotes in our study, in contrast to only one
heterozygous patient (Vall66Met), indicating the possible sig-
nificance of this genetic polymorphism.
The main limitation in the study design was the use of par-
tial SD in the sham stimulation group; otherwise, it would have
been a 3-arm, double-blind, controlled study, with active rTMS
only as one arm. We decided to use partial SD in both groups to
maintain the same procedures for all of the patients, whereas it
would have be difficult to recognize the treatment modality
without the use of partial SD in the sham group.
Regarding partial SD effectiveness, the clinical outcome
measures could be assessed immediately after SD and one day
thereafter (although the HDRS should be modified to omit the
items related to sleep and diurnal variation because they could
not be meaningfully assessed on the day after partial SD; self-
rating scales could be used). A further limitation was the sample
size, preventing randomization based on BDNF genotype.
CONCLUSION
Our study reveals that a clinically relevant response, lasting
up to 6 months, of 1-Hz rTMS applied over the right DLPFC to-
gether with partial SD in patients with pharmacoresistant de-
pression. Moreover, there is a possible role for the BDNF
gene polymorphism as a predictor of the magnitude of the treat-
ment response.
REFERENCES
1. Pascual-Leone A, Rubio B, Pallardo F, et al. Rapid-rate transcranial
magnetic stimulation of left dorsolateral prefrontal cortex in
drug-resistant depression. Lancet. 1996;348:233–237.
2. Menkes DL, Bodnar P, Ballesteros RA, et al. Right frontal lobe slow
frequency repetitive transcranial magnetic stimulation (SF r-TMS) is an
effective treatment for depression: a case-control pilot study of safety
and efficacy. J Neurol Neurosurg Psychiatry. 1999;67:113–115.
3. Wu JC, Gillin JC, Buchsbaum MS, et al. Sleep deprivation PET
correlations of Hamilton symptom improvement ratings with changes in
relative glucose metabolism in patients with depression. J Affect Disord.
2008;107:181–186.
© 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Low-Frequency rTMS With Sleep Deprivation
4. Speer AM, Benson BE, Kimbrell TK, et al. Opposite effects of
high and low frequency rTMS on mood in depressed patients:
relationship to baseline cerebral activity on PET. J Affect Disord.
2009;115:386–394.
5. Kreuzer PM, Langguth B, Schecklmann M, et al. Can repetitive
transcranial magnetic stimulation prolong the antidepressant effects of
sleep deprivation? Brain Stimul. 2012;5:141–147.
6. Padberg F, Schule C, Zwanzger P, et al. Relation between responses to
repetitive transcranial magnetic stimulation and partial sleep deprivation
in major depression. J Psychiatr Res. 2002;36:131–135.
7. Cheeran B, Talelli P, Mori F, et al. A common polymorphism in the
brain-derived neurotrophic factor gene (BDNF) modulates human
cortical plasticity and the response to rTMS. J Physiol 2008;586:
5717–5725.
8. Bocchio-Chiavetto L, Miniussi C, Zanardini R, et al. 5-HTTLPR and
BDNF Val66Met polymorphisms and response to rTMS treatment in
drug resistant depression. Neurosci Lett. 2008;437:130–134.
9. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini-International
Neuropsychiatric Interview (M.I.N.I.): the development and validation
of a structured diagnostic psychiatric interview for DSM-IV and ICD-10.
J Clin Psychiatry. 1998;59(suppl 20):22–33.
10. Hamilton M. A rating scale for depression. J Neurol Neurosurg
Psychiatry. 1960;23:56–62. 56–62.
11. Miller IW, Bishop S, Norman WH, et al. The Modified Hamilton
Rating Scale for Depression: reliability and validity. Psychiatry Res.
1985;14:131–142.
12. Thase ME, Rush AJ. When at first you don’t succeed: sequential
strategies for antidepressant nonresponders. J Clin Psychiatry. 1997;
58(suppl 13):23–29.
13. Milak MS, Parsey RV, Keilp J, et al. Neuroanatomic correlates of
psychopathologic components of major depressive disorder. Arch Gen
Psychiatry. 2005;62:397–408.
14. Giedke H, Schwarzler F. Therapeutic use of sleep deprivation in
depression. Sleep Med Rev. 2002;6:361–377.
15. Hoppner J, Schulz M, Irmisch G, et al. Antidepressant efficacy of two
different rTMS procedures. High frequency over left versus low
frequency over right prefrontal cortex compared with sham stimulation.
Eur Arch Psychiatry Clin Neurosci. 2003;253:103–109.
16. Januel D, Dumortier G, Verdon CM, et al. A double-blind sham
controlled study of right prefrontal repetitive transcranial magnetic
stimulation (rTMS): therapeutic and cognitive effect in medication free
unipolar depression during 4 weeks. Prog Neuropsychopharmacol Biol
Psychiatry. 2006;30:126–130.
17. Klein E, Kreinin I, Chistyakov A, et al. Therapeutic efficacy of right
prefrontal slow repetitive transcranial magnetic stimulation in major
depression: a double-blind controlled study. Arch Gen Psychiatry.
1999;56:315–320.
18. Pallanti S, Bernardi S, Di RA, et al. Unilateral low frequency versus
sequential bilateral repetitive transcranial magnetic stimulation: is
simpler better for treatment of resistant depression? Neuroscience.
2010;167:323–328.
19. Ruohonen J, Ilmoniemi RJ. Basic physics and design of transcranial
magnetic stimulation devices and coils. In: Hallett M, Chokroverty S,
eds. Magnetic Stimulation in Clinical Neurophysiology. 2nd Ed. 2005:
Philadelphia, PA: Butterworth-Heinemann; 2005:17–30.
20. Brakemeier EL, Luborzewski A, Danker-Hopfe H, et al. Positive
predictors for antidepressive response to prefrontal repetitive
transcranial magnetic stimulation (rTMS). J Psychiatr Res.
2007;41:395–403.
21. Brunoni AR, Lopes M, Fregni F. A systematic review and meta-analysis
of clinical studies on major depression and BDNF levels: implications
for the role of neuroplasticity in depression. Int J
Neuropsychopharmacol. 2008;11:1169–1180.
www.ectjournal.com 331