Accepted Manuscript

Title: Comparative Efficacy and Acceptability of

Electroconvulsive Therapy versus Repetitive Transcranial
Magnetic Stimulation for Major Depression: a Systematic
Review and Multiple-Treatments Meta-Analysis

Author: Jian-jun Chen Li-bo Zhao Yi-yun Liu Song-hua Fan

Peng Xie

PII: S0166-4328(16)30805-1
DOI: http://dx.doi.org/doi:10.1016/j.bbr.2016.11.028
Reference: BBR 10568

To appear in: Behavioural Brain Research

Received date: 10-10-2016

Revised date: 11-11-2016
Accepted date: 15-11-2016

Please cite this article as: Chen Jian-jun, Zhao Li-bo, Liu Yi-yun, Fan Song-
hua, Xie Peng.Comparative Efficacy and Acceptability of Electroconvulsive Therapy
versus Repetitive Transcranial Magnetic Stimulation for Major Depression: a
Systematic Review and Multiple-Treatments Meta-Analysis.Behavioural Brain
Research http://dx.doi.org/10.1016/j.bbr.2016.11.028

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Comparative Efficacy and Acceptability of Electroconvulsive Therapy versus Repetitive Transcranial Magnetic

Stimulation for Major Depression: a Systematic Review and Multiple-Treatments Meta-Analysis

Jian-jun Chen1,2,3,4,5,6, Li-bo Zhao6,7, Yi-yun Liu1,2,6,8, Song-hua Fan1,2,6,8, Peng Xie1,2,6,7,8,#

1Institute of Neuroscience, Chongqing Medical University

2Chongqing Key Laboratory of Neurobiology, Chongqing Medical University

3Institute of Life Sciences, Chongqing Medical University

4Department of Obstetrics and Gynecology, the First Affiliated Hospital of Chongqing Medical University

5Canada - China -New Zealand Joint Laboratory of Maternal and Fetal Medicine, Chongqing Medical University

6Institute of Neuroscience and the Collaborative Innovation Center for Brain Science, Chongqing Medical University

7Department of Neurology, Yongchuan Hospital of Chongqing Medical University

8Department of Neurology, the First Affiliated Hospital of Chongqing Medical University,

#corresponding author:

Professor Peng Xie

Department of Neurology

The First Affiliated Hospital at Chongqing Medical University, 1 Yixueyuan Road, Yuzhong District, Chongqing

400016, China

Tel.: +86-23-68485490

Fax: +86-23-68485111

E-mail: xiepeng@cqmu.edu.cn

Highlights
1. ECT was the most efficacious, but least tolerated.
2. R-rTMS was the best tolerated treatment for MDD.
3. B-rTMS appears to have the most favorable balance between efficacy and acceptability.
ABSTRACT

Backgrounds: The effects of electroconvulsive therapy (ECT) and bilateral, left prefrontal, and right prefrontal

repetitive transcranial magnetic stimulation (rTMS) on major depressive disorder (MDD) have not been adequately

addressed by previous studies. Here, a multiple-treatments meta-analysis, which incorporates evidence from direct and

indirect comparisons from a network of trials, was performed to assess the efficacy and acceptability of these four

treatment modalities on MDD.

Method: The literature was searched for randomized controlled trials (RCTs) on ECT, bilateral rTMS, and unilateral

rTMS for treating MDD up to May 2016. The main outcome measures were response and drop-out rates.

Results: Data were obtained from 25 studies consisting of 1288 individuals with MDD. ECT was non-significantly

more efficacious than B-rTMS, R-rTMS, and L-rTMS. Left prefrontal rTMS was non -significantly less efficacious than

all other treatment modalities. In terms of acceptability, R-rTMS was non-significantly better tolerated than ECT, B-

rTMS, and L-rTMS. ECT was the most efficacious treatment with the cumulative probabilities of being the most

efficacious treatment being: ECT (65%), B-rTMS (25%), R-rTMS (8%), and L-rTMS (2%). R-rTMS was the best-

tolerated treatment with the cumulative probabilities of being the best-tolerated treatment being: R-rTMS (52%), B-

rTMS (17%), L-rTMS (16%), and ECT (14%). Coherence analysis detected no statistically significant incoherence in

any comparisons of direct with indirect evidence for the response rate and drop-out rate.

Conclusions: ECT was the most efficacious, but least tolerated, treatment, while R-rTMS was the best tolerated

treatment for MDD. B-rTMS appears to have the most favorable balance between efficacy and acceptability.

Key words: depression, MDD; meta-analysis; transcranial magnetic stimulation, TMS, rTMS, electroconvulsive

therapy, ECT
INTRODUCTION

Major depressive disorder (MDD, major depression) is a debilitating mental disorder affecting up to 15% of the

general population and accounting for 12.3% of the global burden of disease (1). To date, increasing evidence from

biochemical, neuropsychological, postmortem, and neuroimaging studies indicates that MDD is not likely caused by a

single brain region or neurotransmitter system, but rather is a system-level disorder affecting several integrated

pathways (2, 3).

Electroconvulsive therapy (ECT) is a well-established and effective treatment method for MDD superior to both

placebo and sham ECT (anesthesia only) (4, 5). Some researchers even consider ECT to be the most effective treatment

for MDD (6). Of MDD patients who receive ECT, approximately 70% to 80% show significant improvement (6), and

ECT is effective in half of patients with treatment-resistant depression (TRD) (7). However, ECT is complicated by a

number of side effects including cognitive impairment; so many patients are reluctant to engage in ECT treatment due

to the risks and stigma associated with cognitive side effects, which has motivated attempts at developing treatment

alternatives (8).

Repetitive Transcranial Magnetic Stimulation (rTMS) is a non-invasive method of brain stimulation for the treatment

of patients with serious neuropsychiatric disorders including MDD (9). Unlike ECT, rTMS does not require anesthesia

or induction of seizures. RTMS is divided into bilateral rTMS (B-rTMS), left prefrontal rTMS (L-rTMS), and right

prefrontal rTMS (R-rTMS) according to the stimulation location. Most studies of rTMS in MDD focus on high-

frequency (5-20 Hz) stimulation to the left dorsolateral prefrontal cortex, and L-rTMS has been shown to have positive

antidepressive effects (10, 11). Some randomized controlled trials (RCTs) have demonstrated that R-rTMS shows

significantly greater improvement in depression scores compared with sham rTMS (12, 13), and our previous research

has shown that L-rTMS and R-rTMS have a similar efficacy on MDD patients (14). Moreover, a 2012 systematic

review showed that B-rTMS is a promising treatment for MDD (15), and our previous research also found that bilateral

and unilateral rTMS had comparable efficacies on MDD patients (16).

Hitherto, ECT has been traditionally viewed as the superior treatment modality vis-a-vis rTMS (17), but this

conclusion has been primarily based on RCTs of ECT versus L-rTMS. There is still lack of quantitative data comparing

the efficacy of ECT versus B-rTMS or R-rTMS in MDD. To this end, although standard meta-analyses are an effective

tool, they can only compare two alternative treatments at a time; moreover, if no trials directly compare two

interventions, it is impossible to compare their relative efficacies (18). In contrast, multiple treatments meta-analyses

use a technique that incorporates evidence from both direct and indirect comparisons from a network of trials of

different interventions to better estimate summary treatment effects. Our group used this method to compare the

efficacy and tolerability of antidepressants for MDD in children and adolescents, and the results has been published in
Lancet in 2016 (19). Therefore, here we applied a multiple-treatments meta-analysis to compare the efficacy and

acceptability of B-rTMS, R-rTMS, L-rTMS, and ECT in the treatment of MDD.

METHODS

Study Selection

This systematic review and meta-analysis was conducted and reported according to the PRISMA statement

(http://www.prisma-statement.org/). A comprehensive literature search of RCTs comparing ECT with rTMS was

conducted up to May 2016 through the major scientific and medical databases, including international databases

(PubMed, CCTR, Web of Science, and Embase) and two Chinese databases (CBM-disc and CNKI). The key search

terms were “depression” AND (“transcranial magnetic stimulation” OR “TMS” OR “repetitive TMS” OR “rTMS”)

AND (“electroconvulsive therapy” OR “ECT”). No language or publication year limitation was imposed. To avoid

omitting relevant trials, conference summaries and reference documents listed in the obtained articles were checked.

Among the identified studies, only those meeting the following criteria were selected for subsequent analyses: (i)

RCTs comparing one treatment against another (B-rTMS, L-rTMS, R-rTMS, and ECT); (ii) assessing mood by the

Hamilton Depression Rating Scale (HDRS), Montgomery-Åsberg Depression Rating Scale (MADRS), or Clinical

Global Impression (CGI); (iii) patients over 18 years of age without metallic implants or foreign bodies, dementia,

personal or family history of epileptic seizures, severe suicidal risk, organic brain damage, severe agitation or delirium,

substance abuse, alcohol or drug dependence, and/or medically unfit for general anesthesia. Studies with pregnant

patients were excluded because rTMS and ECT have unclear fetal side effects (20).

Studies were excluded if they: (i) had no random allocation; (ii) enrolled subjects with ‘narrow’ depression diagnoses

(e.g., postpartum depression) or secondary depression diagnoses (e.g., vascular depression); (iii) used rTMS and ECT

concomitantly with a new antidepressant without wash out period; and (iv) case reports and reviews.

Data Extraction

Two reviewers independently verified all potentially suitable RCTs by the aforementioned inclusion and exclusion

criteria and the completeness of data abstraction. Any disagreement was resolved by consensus and, if needed, a third

reviewer was consulted. Data retrieved from the included RCTs were recorded in a structured fashion as follows: (i)

sample characteristics: mean age, gender, mean depression score, treatment strategy used, presence of TRD; (ii) rTMS

parameters: stimulation location, frequency, motor threshold, and duration; (iii) primary outcome measure: response

was defined as at least a 50% reduction in the absolute HDRS or MADRS score from baseline, or significant

improvement in the CGI, at the conclusion of therapy (21) with a preference for HDRS; and (iv) secondary outcome

measure: overall drop-out rates at the study's end. For data that could not be directly retrieved, good faith efforts were

applied to obtain the data by dispatching e-mails to the author, researching other studies citing the RCT in question, and
researching associated conference summaries.

Bias Risk in Individual Studies

Two reviewers independently assessed bias risk of the eligible studies according to the Cochrane handbook. We

selected the following items to assess the bias risk: (1) did the authors conduct randomization? (2) did the authors

conduct allocation concealment? (3) did the authors conduct blind treatment? and (4) were the baseline clinical

characteristics matched between two groups. Studies with three or more ‘NO’ were still excluded.

Statistical Analysis

In order to make the interpretation of current results easier for clinicians (22), the response rate (a dichotomous

primary outcome for efficacy) was used instead of a continuous symptom score. If the baseline scores, standard

deviations (SD), and endpoint means were provided instead of the dichotomous efficacy outcomes, we estimated the

number of responding patients through a validated imputation method. (23) To perform a clinically sound analysis, we

used a worst-case scenario analysis of drop-out patients, assuming all such patients failed to respond to treatment. (24)

First, with a random-effects model, we performed a meta-analysis of augmentation agents that had direct

comparisons. For each analysis, we assessed heterogeneity using the Chi-square based Q test and I squared index (I2).

(25) We performed the analyses using RevMan5.0 software (Cochrane Information Management System [IMS]).

Second, we performed multiple-treatment meta-analysis using an arm-based, random-effects model within an empirical

Bayes framework using Markov chain Monte Carlo method (26). The model allowed for estimating effect sizes for all

possible pair-wise comparisons of augmentation agents. P-values of less than 0.05 were used to assess significance. We

also computed and ranked the probabilities for each treatment's efficacy (27). The ranking of the competing treatments

was assessed with the median of the posterior distribution for the rank of each treatment. We performed this analysis

using WinBUGS (Imperial College and MRC, London, UK) and R v2.15.0 (R Development Core Team, Vienna,

Austria).

The coherence of an analyzed network – i.e., that indirect and direct evidence on the same comparisons do not

disagree beyond chance – is the key assumption behind multiple-treatments meta-analysis. Whenever indirect estimates

could be built with a single common comparator, the ratio of odds ratios for indirect vs. direct evidence is calculated to

estimate incoherence. The disagreement between direct and indirect evidence with a 95% confidence interval excluding

unity is defined as incoherent (24).

RESULTS

The electronic literature search resulted in 557 potentially relevant studies, of which 25 eligible articles were pooled

for analysis (28-52) (Figure 1). Overall, 1288 individuals were randomly assigned to one of the four treatment

modalities and were included in the multiple-treatments meta-analysis with all 1288 patients included in the efficacy
analysis (25 studies) and 820 patients included in the acceptability analysis (12 studies). The mean duration of the

studies was 3.04 weeks, and the mean sample size was 25.7 participants per group (range: 6–147). The main

characteristics of the included RCTs are described (Tables 1 and 2).

All the 25 included studies conducted the randomization, two studies did not conduct allocation concealment, seven

studies did not conduct blind treatment and the baseline clinical characteristics were matched between two groups in all

included studies (Table 2). As these studies displayed minimal or no bias risk, all of them were included in the meta-

analysis.

Direct comparisons for the four treatment modalities showed no statistically significant differences in response rates

between any two treatment modalities (Table 3). However, the odds ratio (OR) non-significantly favored ECT over L-

rTMS and R-rTMS with pooled ORs of 1.43 [95% confidence interval (CI), 0.92-2.22] and 2.19 (95% CI, 0.72-6.70),

respectively. B-rTMS was non-significantly superior to L-rTMS, but non-significantly inferior to R-rTMS, with pooled

ORs of 1.70 (95% CI, 0.74-3.92) and 0.93 (95% CI, 0.63-1.37), respectively. R-rTMS was comparably efficacious to L-

rTMS. These results obtained from 25 independent analyses without adjustment for multiple testing (i.e., about two CIs

would be expected to exclude unity by chance alone). For drop-outs, statistically significant differences did not exist

between the treatment modalities. Overall, heterogeneity was moderate. In the meta-analyses of direct comparisons, we

found no I 2 values higher than 75%.

From the multiple-treatments meta-analysis on response rates, ECT was non-significantly more efficacious than B-

rTMS, R-rTMS and L-rTMS with pooled ORs of 1.27 (95% CI, 0.58-2.61), 1.14 (95% CI, 0.63-1.94), and 1.65 (95%

CI, 0.88-2.83), respectively. L-rTMS was non-significantly less efficacious than B-rTMS and R-rTMS with pooled ORs

of 0.90 (95% CI, 0.46-1.52) and 0.96 (95% CI, 0.59-1.44), respectively. In terms of acceptability, R-rTMS was non-

significantly better tolerated than ECT, B-rTMS, and L-rTMS with pooled ORs of 0.47 (95% CI, 0.05-1.39), 0.94 (95%

CI, 0.15-2.45), and 0.57 (95% CI, 0.07-1.56), respectively. Coherence analysis detected no statistically significant

incoherence in any comparisons of direct with indirect evidence for the response rate and drop-out rate.

Figure 2 showed the distribution of probabilities of each treatment modality being ranked at each of four possible

positions. ECT was the most efficacious treatment with the cumulative probabilities of being the most efficacious

treatment being: ECT (65%), B-rTMS (25%), R-rTMS (8%), and L-rTMS (2%) (Figure 2A). R-rTMS was the best-

tolerated treatment with the cumulative probabilities of being the best-tolerated treatment being: R-rTMS (52%), B-

rTMS (17%), L-rTMS (16%), and ECT (14%) (Figure 2B).

DISCUSSION

Nowadays, metabolomics has been extensively used to identify potential biomarkers for psychiatric disorders (53,

54). Although many works has used metabolomics to identify biomarkers for MDD (55, 56), there are still no objective
methods to diagnose MDD. Besides, there are no treatment methods that could cure MDD with 100% response rate.

Here, this multiple-treatments meta-analysis was based on 25 studies consisting of 1288 individuals randomly assigned

to ECT, B-rTMS, R-rTMS or L-rTMS for MDD. We retrieved almost all relevant RCTs, and the overlooked literature

that were not indexed by international databases were likely to be of low quality and would not significantly affect the

results of this review (57). Although the results were statistically non-significant, ECT and L-rTMS were the most

efficacious and least efficacious treatments, respectively, and in terms of drop-outs, R-rTMS and ECT were the most

tolerated and least tolerated, respectively. These results suggested that the most efficacious treatment, ECT, may not be

the best in terms of overall acceptability. Although important outcomes, such as discontinuation symptoms and side-

effects, were not investigated here, B-rTMS appears to be the best choice among the four modalities, as it had the most

favorable possible balance between efficacy and acceptability. Interestingly, recent multimodal neuroimaging data

showed that B-rTMS might have synergistic therapeutic effects by reversing both the hypo-function in the right DLPFC

and the hyper-function in the left DLPFC (58, 59).

That being said, the therapeutic application of rTMS involves several parameters (e.g., frequency, resting motor

threshold, number of stimuli per day) (60); however, the optimum rTMS protocol based on these parameters has yet to

be determined. Therefore, future RCTs should seek to identify and optimize clinically relevant stimulation parameters in

order to improve the antidepressant effects of rTMS. Moreover, other subsidiary technologies, such as baseline electro-

physiological and/or neuroimaging evaluations, can be used to predict which patient subgroups can particularly benefit

from rTMS (61).

Although we did not perform a formal cost-effective analysis here, there are studies stating that ECT is more cost-

effective than rTMS in the treatment of MDD (62, 63). However, in the absence of a complete economic model, this

conclusion should not be made unequivocally because several cost components are associated with the use of ECT or

rTMS (64).

Limitations

First, as the ‘5 cm method' for locating the DLPFC has been recently criticized for its inaccuracy (9), future rTMS

studies should take advantage of neuronavigation approaches (65). Second, this review only examined efficacy at study

end, and thus our conclusion cannot be applied to medium-term or long-term outcomes. Third, patients in the selected

RCTs were over 18 years of age, so it is inappropriate to apply these findings to adolescents. Finally, several meta-

analyses have been criticized for the inclusion of poor-quality trials and for combining heterogeneous studies. However,

our comprehensive and systematic literature search combined with the use of stringent inclusion and exclusion criteria

aided in mitigating these concerns.

Conclusions
 This multiple-treatments meta-analysis comparing the efficacy and acceptability of B-rTMS, R-rTMS, L-rTMS, and

ECT in the treatment of MDD showed that: (i) ECT was the most efficacious, but least tolerated, treatment method; (ii)

R-rTMS was the best tolerated treatment method; and (iii) B-rTMS appeared to have the most favorable balance

between efficacy and acceptability. Further studies, such as well-designed, large-scale, multi-center RCTs directly

comparing B-rTMS, R-rTMS, L-rTMS, and ECT, are needed to draw more definitive conclusions.

Acknowledgements

We thank Dr N. D. Melgiri for editing and proofreading the manuscript. This work was supported by the Natural

Science Foundation Project of China (81601208, 31271189, 81200899, 31300917, and 81401140), the National Basic

Research Program of China (973 Program, grant no. 2009CB918300), the Fund for Outstanding Young Scholars in

Chongqing Medical University (CYYQ201502), and the Chongqing Science & Technology Commission

(cstc2014jcyjA10102).

Disclosure of conflicts of interest

The authors declare no financial or other conflicts of interest.

REFERENCES

1. Reynolds EH. Brain and mind: a challenge for WHO. The Lancet, 2003, 361(9373): 1924-1925.

2. Manji HK, Drevets WC, Charney DS. The cellular neurobiology of depression. Nature medicine, 2001, 7(5): 541-

547.

3. Zheng P, Zeng B, Zhou C, Liu M, Fang Z, Xu X, Jian-jun Chen, et al. Altered gut microbiome induces depressive-

like behaviors through a pathway mediated by the host's metabolism. Molecular Psychiatry 2016; 21(6):786-96.

4. Fink M. ECT has proved effective in treating depression. Nature, 2000, 403(6772): 826-826.

5. UK ECT review group. Efficacy and safety of electroconvulsive therapy in depressive disorders: a systematic review

and meta-analysis. Lancet, 2003; 361:799-808.

6. Rasmussen K G. Some considerations in choosing electroconvulsive therapy versus transcranial magnetic stimulation

for depression. The journal of ECT, 2011, 27(1): 51-54.

7. Sackeim HA, Prudic J, Devanand DP, et al. A prospective, randomized, double-blind comparison of bilateral and

right unilateral electroconvulsive therapy at different stimulus intensities. Archives of General Psychiatry, 2000, 57(5):

425-434.

8. Fitzgerald PB, Hoy KE, Herring SE, et al. Pilot study of the clinical and cognitive effects of high-frequency

magnetic seizure therapy in major depressive disorder. Depression and anxiety, 2013, 30(2): 129-136.

9. Rosa MA, Lisanby SH. Somatic treatments for mood disorders. Neuropsychopharmacology, 2012, 37(1): 102-116.

10. George MS, Lisanby SH, Avery D, et al. Daily left prefrontal transcranial magnetic stimulation therapy for major

depressive disorder. a sham-controlled randomized trial. Arch Gen Psychiatry. 2010; 67(5):507-516.

11. Schutter DJ. Antidepressant efficacy of high-frequency transcranial magnetic stimulation over the left dorsolateral

prefrontal cortex in double-blind sham-controlled designs: a meta-analysis. Psychol Med. 2009; 39(1):65-75.

12. 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. Archives of general psychiatry, 1999, 56(4): 315-320.

13. Kauffmann CD, Cheema MA, Miller BE. Slow right perfrontal transcranial magnetic stimulation as a reatment for

medication‐resistant depression: A double‐blind, placebo‐controlled study. Depression and anxiety, 2004, 19(1):

59-62.

14. Chen J, Zhou C, Wu B, et al. Left versus right repetitive transcranial magnetic stimulation in treating major

depression: a meta-analysis of randomised controlled trials. Psychiatry research, 2013, 210(3): 1260-1264.

15. Berlim MT, Van den Eynde F, Daskalakis ZJ. A systematic review and meta-analysis on the efficacy and

acceptability of bilateral repetitive transcranial magnetic stimulation (rTMS) for treating major depression. Psychol

Med, 2013, 43(11): 2245-2254.

16. Chen J, Liu Z, Zhu D, et al. Bilateral vs. unilateral repetitive transcranial magnetic stimulation in treating major

depression: a meta-analysis of randomized controlled trials. Psychiatry research, 2014, 219(1): 51-57.

17. Minichino¹ A, Bersani¹ F S, Capra¹ E, et al. ECT, rTMS, and deepTMS in pharmacoresistant drug-free patients with

unipolar depression: a comparative review. Neuropsychiatric disease and treatment, 2012, 8: 55-64.

18. Cipriani A, Barbui C, Rizzo C, Salanti G. What is a multiple treatments meta-analysis? Epidemiology and

psychiatric sciences 2012; 21(2):151-3.

19. Cipriani A, Zhou X, Del Giovane C, et al. Comparative efficacy and tolerability of antidepressants for major

depressive disorder in children and adolescents: a network meta-analysis. The Lancet, 2016; 388(10047):881-90.

20. Chabrol H, Teissedre F, Saint-Jean M, Teisseyre N, Roge B, Mullet E. Prevention and treatment of post-partum

depression: a controlled randomized study on women at risk. Psychological medicine 2002; 32(6):1039-47.

21. Hamilton M. A rating scale for depression. Journal of Neurology, Neurosurgery & Psychiatry, 1960, 23(1): 56-62.

22. Guyatt GH, Juniper EF, Walter SD, et al. Interpreting treatment effects in randomised trials. Brit Med J (BMJ).1998;

316:690–93.

23. Furukawa TA, Cipriani A, Barbui C,et al. Imputing response rates from means and standard deviations in meta-

analysis. Int Clin Psychopharm. 2005; 20:49–52.

24. Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptability of 12 new-generation

antidepressants: a multiple-treatments meta-analysis. Lancet. 2009; 373: 746-58.

25. Higgins J P T, Thompson S G, Deeks J J, et al. Measuring inconsistency in meta-analyses. Bmj, 2003, 327(7414):

557-560.

26. Ades AE, Sculpher M, Sutton A, et al. Bayesian methods for evidence synthesis in cost-eff ectiveness analysis.

Pharmacoeconomics, 2006; 24: 1–19.

27. Salanti G, Ades AE, Ioannidis JPA. Graphical methods and numerical summaries for presenting results from

multiple-treatment meta-analysis: an overview and tutorial. Journal of clinical epidemiology, 2011, 64(2): 163-171.

28. Eche J, Mondino M, Haesebaert F, Saoud M, Poulet E, Brunelin J. Low- vs High-Frequency Repetitive Transcranial

Magnetic Stimulation as an Add-On Treatment for Refractory Depression. Front Psychiatry. 2012; 3:13.

29. Fitzgerald PB, Brown TL, Marston NAU, et al. Transcranial magnetic stimulation in the treatment of depression: a

double-blind, placebo-controlled trial. Archives of General Psychiatry, 2003, 60(10): 1002-1008.

30. Fitzgerald PB, Hoy K, Daskalakis ZJ, et al. A randomized trial of the anti‐depressant effects of low‐and high‐

frequency transcranial magnetic stimulation in treatment‐resistant depression. Depression and anxiety, 2009, 26(3):

229-234.

31. Fitzgerald PB, Sritharan A, Daskalakis ZJ, et al. A functional magnetic resonance imaging study of the effects of
low frequency right prefrontal transcranial magnetic stimulation in depression. Journal of clinical psychopharmacology,

2007, 27(5): 488-492.

32. Höppner J, Schulz M, Irmisch G, et al. Antidepressant efficacy of two different rTMS procedures. European archives

of psychiatry and clinical neuroscience, 2003, 253(2): 103-109.

33.Rossini D, Lucca A, Magri L, et al. A symptom-specific analysis of the effect of high-frequency left or low-

frequency right transcranial magnetic stimulation over the dorsolateral prefrontal cortex in major depression.

Neuropsychobiology, 2010, 62(2): 91-97.

34. Stern WM, Tormos JM, Press DZ, et al. Antidepressant effects of high and low frequency repetitive transcranial

magnetic stimulation to the dorsolateral prefrontal cortex: a double-blind, randomized, placebo-controlled trial. The

Journal of neuropsychiatry and clinical neurosciences, 2007, 19(2): 179-186.

35. Triggs WJ, Ricciuti N, Ward HE, et al. Right and left dorsolateral pre-frontal rTMS treatment of refractory

depression: a randomized, sham-controlled trial. Psychiatry research, 2010, 178(3): 467-474.

36. Rosa MA, Gattaz WF, Pascual-Leone A, Fregni F, Rosa MO, et al. Comparison of repetitive transcranial

magnetic stimulation and electroconvulsive therapy in unipolar non-psychotic refractory depression: a randomized,

single-blind study. Int J Neuropsychopharmacol, 2006, 9:667–676.

37. Janicak PG, Dowd SM, Martis B, Alam D, Beedle D, et al. Repetitive Transcranial Magnetic Stimulation versus

Electroconvulsive Therapy for Major Depression: Preliminary Results of a Randomized Trial. Biol Psychiatry, 2002,

51(8): 659-67.

38. Pridmore S, Bruno R, Turnier-Shea Y, Reid P, Rybak M. Comparison of unlimited numbers of rapid

transcranial magnetic stimulation (rTMS) and ECT treatment sessions in major depressive episode. Int J

Neuropsychopharmacol, 2000, 3:129–134.

39. Grunhaus L, Dannon PN, Schreiber S, Dolberg OH, Amiaz R, et al. Repetitive transcranial magnetic stimulation is

as effective as electroconvulsive therapy in the treatment of nondelusional major depressive disorder: an open study.

Biol Psychiatry, 2000, 47:314–324.

40. Grunhaus L, Schreiber S, Dolberg OT, Polak D, Dannon PN. A randomized controlled comparison of

electroconvulsive therapy and repetitive transcranial magnetic stimulation in severe and resistant nonpsychotic

major depression. Biol Psychiatry, 2003, 53: 324–331.

41. Eranti S, Mogg A, Pluck G, Landau S, Purvis R, et al. A randomized, controlled trial with 6-month follow up of

repetitive transcranial magnetic stimulation and electroconvulsive therapy for severe depression. Am J Psychiatry, 2007,

164: 73–81.

42. Hansen PE, Ravnkilde B, Videbech P, Clemmensen K, Sturlason R, et al. Low-Frequency Repetitive Transcranial
Magnetic Stimulation Inferior to Electroconvulsive Therapy in Treating Depression. J ECT, 2011, 27(1):26-32.

43. Fitzgerald PB, Hoy KE, Singh A, et al. Equivalent beneficial effects of unilateral and bilateral prefrontal cortex

transcranial magnetic stimulation in a large randomized trial in treatment-resistant major depression. International

Journal of Neuropsychopharmacology, 2013, 16(9): 1975-1984.

44. Blumberger DM, Mulsant BH, Fitzgerald PB, Rajji TK, Ravindran AV, Young LT, Levinson AJ, Daskalakis ZJ. A

randomized double-blind sham-controlled comparison of unilateral and bilateral repetitive transcranial magnetic

stimulation for treatment-resistant major depression. World J Biol Psychiatry. 2012; 13(6): 423- 435.

45. Fitzgerald PB, Hoy KE, Herring SE, McQueen S, Peachey AV, Segrave RA, Maller J, Hall P, Daskalakis ZJ. A

double blind randomized trial of unilateral left and bilateral prefrontal cortex transcranial magnetic stimulation in

treatment resistant major depression. J Affect Disord. 2012; 139(2):193-198.

46. Conca A, Di Pauli J, Beraus W, Hausmann A, Peschina W, Schneider H, König P, Hinterhuber H. Combining high

and low frequencies in rTMS antidepressive treatment: preliminary result. Hum Psychopharmacol. 2002; 17(7):353-

356.

47. M. Rybak, R. Bruno, Y. Turnier-Shea, S. Pridmore. An Attempt to Increase the Rate and Magnitude of the

Antidepressant Effect of Transcranial Magnetic Stimulation (TMS). German Journal of Psychiatry. 2005; 8: 59-65.

48. Fitzgerald PB, Hoy K, Gunewardene R, Slack C, Ibrahim S, Bailey M, Daskalakis ZJ. A randomized trial of

unilateral and bilateral prefrontal cortex transcranial magnetic stimulation in treatment-resistant major depression.

Psychol Med. 2011; 41: 1187-1196.

49. Pallanti S, Bernardi S, Di Rollo A, Antonini S, Quercioli L. Unilateral low frequency versus sequential bilateral

repetitive transcranial magnetic stimulation: is simpler better for treatment of resistant depression? Neuroscience. 2010;

167(2): 323- 328.

50. Xiaoming Wang, Deben Yang, Yuanfeng, Hui Huang, Xiaoqiang Zhao. A controlled study of the treatment of

repetitive transcranial magnetic stimulation in patients with major depression. Chinese Journal of Clinical

Rehabilitation. 2004; 8(9):1770-1

51. Jianjuo Wan, Sanmei Hu, Meiying Chen, Xiangkun Chen, Lihua Yu, Yongmei Zhang, Qiongfang Wu. A control

study of sertraline plus RTMS in the treatment of treatment-resistant depression. J Clin Psychosom Dis, 2011, 17(3):

202-204.

52. Isenberg K, Downs D, Pierce K, et al. Low frequency rTMS stimulation of the right frontal cortex is as effective as

high frequency rTMS stimulation of the left frontal cortex for antidepressant-free, treatment-resistant depressed patients.

Ann Clin Psychiatry, 2005, 17(3):153-9.

53. Chen J, Huang H, Zhao L, et al. Sex-specific urinary biomarkers for diagnosing bipolar disorder. PloS one, 2014,
9(12): e115221.

54. Chen J, Liu Z, Fan S, et al. Combined application of NMR-and GC-MS-based metabonomics yields a superior

urinary biomarker panel for bipolar disorder. Scientific reports, 2014, 4: 5855.

55. Chen J, Zhou C, Liu Z, et al. Divergent urinary metabolic phenotypes between major depressive disorder and

bipolar disorder identified by a combined GC–MS and NMR spectroscopic metabonomic approach. Journal of

proteome research, 2015, 14(8): 3382-3389.

56. Zheng P, Wang Y, Chen L, et al. Identification and validation of urinary metabolite biomarkers for major depressive

disorder. Molecular & Cellular Proteomics, 2013, 12(1): 207-214.

57. Deeks JJ, Altman DG, & Bradburn MJ. Statistical Methods for Examining Heterogeneity and Combining Results
from Several Studies in Meta‐Analysis. Systematic Reviews in Health Care: Meta-Analysis in Context, Second
Edition, 2008; 285-312.
58. Kito S, Hasegawa T, Koga Y. Neuroanatomical correlates of therapeutic efficacy of low-frequency right prefrontal
transcranial magnetic stimulation in treatment-resistant depression. Psychiatry and Clinical Neurosciences, 2011, 65,
175–182.
59. Martinot ML, Martinot JL, Ringuenet D, Galinowski A,Gallarda T, Bellivier F, Lefaucheur JP, Lemaitre H,Artiges E.
Baseline brain metabolism in resistant depression and response to transcranial magneticstimulation.
Neuropsychopharmacology, 2011, 36, 2710–2719.
60. Xie J, Jianjun Chen, Wei Q. Repetitive transcranial magnetic stimulation versus electroconvulsive therapy for major
depression: a meta-analysis of stimulus parameter effects. Neurol Res. 2013; 35(10):1084-91.
61. Arns M, Drinkenburg W H, Fitzgerald P B, et al. Neurophysiological predictors of non-response to rTMS in
depression. Brain stimulation, 2012, 5(4): 569-576.
62. Knapp M, Romeo R, Mogg A, et al. Cost-effectiveness of transcranial magnetic stimulation vs. electroconvulsive
therapy for severe depression: a multi-centre randomised controlled trial. Journal of affective disorders, 2008, 109(3):
273-285.
63. McLoughlin D M, Mogg A, Eranti S, et al. The clinical effectiveness and cost of repetitive transcranial magnetic
stimulation versus electroconvulsive therapy in severe depression: a multicentre pragmatic randomised controlled trial
and economic analysis. Health technology assessment (Winchester, England), 2007, 11(24): 1-54.
64. Le Lay A, Despiegel N, François C, Duru G. Can discrete event simulation be of use in modelling major depression?
Cost Effectiveness and Resource Allocation, 2006; 4(1):19.
65. Schönfeldt-Lecuona C, Lefaucheur JP, Cardenas-Morales L, Wolf R, Kammer T, & Herwig U. The value of
neuronavigated rTMS for the treatment of depression. Neurophysiologie Clinique/ Clinical Neurophysiology, 2010; 40:
37-43.
Figure 1 Flow Chart of Study Selection

The electronic literature search resulted in 557 potentially relevant studies, of which 25 eligible articles were pooled for

analysis.
Figure 2 Rankings for Efficacy and Acceptability

The distribution of probabilities of each treatment modality was ranked at each of four possible positions. (A) ECT was

the most efficacious treatment with the cumulative probabilities of being the most efficacious treatment being: ECT

(65%), B-rTMS (25%), R-rTMS (8%), and L-rTMS (2%). (B) R-rTMS was the best-tolerated treatment with the

cumulative probabilities of being the best-tolerated treatment being: R-rTMS (52%), B-rTMS (17%), L-rTMS (16%),

and ECT (14%).

TABLE LEGENDS

Table 1 Demographic and Clinical characteristics of Included Subjects

Female/ Mean age, Mean MDD Primary
Study Pairs n TRD
male, n yrs (S.D.) score (S.D.) diagnosis
Blumberger et L vs. B 22 vs. 12/10 vs. 48.9 (13.4) vs. 26.0 (3.3) vs. 25.1 All MDD Yb
al., 2012 26 14/12 58.0 (12.5) (3.8)a
Fitzgerald et L vs. B 24 vs. 15/9 vs. 43.4 (12.7) vs. 23.7 (3.8) vs. 24.3 All MDD Yb
al., 2012 22 14/8 40.4 (15.5) (3.6)a
Rybak et al., L vs. B 9/9 6/3 vs. 6/3 47.0 (12.3) vs. 23.8 (2.4) vs. 23.0 17% BD, Yc
2005 53.4 (13.3) (4.0)a 83% MDD
17% BD,
Conca et al., 24 vs. 45.8 (12.5) vs. 30.8 (6.0) vs. 29.4
L vs. B 16/8 vs. 9/3 25% OP Part
2002 12 48.2 (16.1) (4.3)d
58% MDD
Fitzgerald et 71 vs. 47/24 47.4 (14.1) vs. 21.8 (2.6) vs. 21.5 15%BD,85%
R vs. B Yb
al., 2011 147 vs. 100/47 46.8 (13.7) (2.9)a MDD
Pallanti et al., R vs. B 20 vs. 12/8 vs. 51.2 (12.5) vs. 27.9 (5.9) vs. 28.7 All MDD Yb
2010 20 11/9 47.6 (12.3) (6.0)a
Fitzgerald et 91 vs. 59/32 vs. 46.7 (14.2) vs. 19.5 (4.4) vs. 19.8 22% BD,
R vs. B Yb
al., 2013 88 66/22 48.5 (15.9) (5.0)a 78% MDD
Hansen et al., R vs. 30 vs. 23/7 vs. 46.0 (N.A.) vs. 24.0 (N.A.) vs. 13% Y
2011 ECT 30 19/11 52.0 (N.A.) 24.0 (N.A.)a BD,87%
MDD
Eranti et al., L vs. 24 vs. 16/8 vs. 63.6 (17.3) vs. 23.9 (7.0) vs. 24.8 8% BD, 92% N
2007 ECT 22 16/6 68.0 (13.4) (5.0)a MDD
Rosa et al. L vs. 20 vs. 12/8 vs. 7/8 41.8 (10.2) vs. 30.1 (4.7) vs. 32.1 All MDD Yb
2006 ECT 15 46.0 (10.6) (5.0) a
Wang et al., L vs. 18 vs. N.A. 31.0 (5.0) vs. 27.8 (3.2) vs. 26.7 All MDD N
2004 ECT 18 32.0 (6.0) (2.8)a
Grunhaus et L vs. 20 vs. 14/6 vs. 57.6 (13.7) vs. 24.4 (3.9) vs. 25.5 All MDD Yc
al., 2003 ECT 20 15/5 61.4 (16.6) (5.9) a
Janicak et al., L vs. 15 vs. 42.8 (12.9) vs. 32.5(6.4)vs.33.4(9 30% BD,
11/4 vs. 6/5 N
2002 ECT 11 42.7 (14.0) .0)f 70% MDD
Grunhaus, et L vs. 20 vs. 12/8 vs. 58.4 (15.7) vs. 25.8 (6.1) vs.
All MDD Part
al., 2000 ECT 20 14/6 63.6 (15.0) 28.4 (9.3)a
Pridmore et L vs. 16 vs. N.A. 44.0 (11.9) vs. 25.3 (4.1) vs. 25.8 18% BD, Yc
al., 2000 ECT 16 41.5 (12.9) (3.6)a 82% MDD
Wan et al., L vs. 45 vs. 15/30 vs. 36.2 (18.8) vs. 26.3 (12) vs. 28.1 Y
All MDD
2011 ECT 43 15/28 35.7 (15.3) (16)d
Fitzgerald et L vs. R 15 vs. 8/7 vs. 5/6 42.4 (11.2) vs. 34.5 (4.9) vs. 33.3 All MDD Yb
al., 2007 11 39.6 (10.0) (3.8)e
Fitzgerald et 16 vs. 42.1 (9.3) vs. 33.6 (3.9) vs. 34.3
L vs. R 8/7 vs. 3/8 All MDD Yb
al., 2009 11 46.5 (11.4) (4.9)e
Rossini et al., L vs. R 32 vs. 23/9 vs. 53.6 (11.3) vs. 24.6 (4.5) vs. 24.3 54% BD, Yc
2010 42 30/12 54.5 (11.8) (4.4)d 46% MDD
Stern et al., L vs. R 20 vs. 12/8 vs. 7/3 52.7 (10.6) vs. 27.7 (3.5) vs. 27.9 All MDD Yc
2007 10 52.8 (9.5) (3.8)d
 Triggs et al., L vs. R 18 vs. 14/4 vs. 9/7 46.7 (15.3) vs. 28.2 (6.0) vs. 27.2 All MDD Yb
2010 16 48.5 (10.8) (4.8)f
Höppner et al., L vs. R 10 vs. 7/3 vs. 8/2 59.5 (6.8) vs. N.A. All MDD N
2003 10 52.0 (11.7)
Eche et al., L vs. R 6 vs. 8 2/4 vs. 6/2 50.8 (9.4) vs. 29.8 (6.9) vs. 32.0 All MDD Yc
2012 46.1 (16.3) (8.0)e
Isenberg et al., L vs. R 14 vs. 6/8 vs. 6/8 43.4 (9.7) vs. 25.1 (4.9) vs. 23.9 10% BD, Yb
2005 14 55.6 (9.7) (6.2)d 90% MDD
Fitzgerald et L vs. R 20 vs. 12/8 vs. 42.2 (9.8) vs. 36.1 (7.5) vs. 37.7 3% BD, 97% Yb
al., 2003 20 13/7 45.5 (11.5) (8.4)e MDD
a17-item Hamilton Depression Rating Scale.
bFailure to respond to 2 antidepressants in the current major depressive episode.
cFailure to respond to 1 antidepressants in the current major depressive episode.
d21-item Hamilton Depression Rating Scale.
eMontgomery–Asberg Depression Rating Scale.
f24-item Hamilton Depression Rating Scale.
Abbreviations: rTMS, repetitive transcranial magnetic stimulation; L, left rTMS; R, right rTMS; B, bilateral rTMS;
TRD, treatment-resistant depression; ECT, electroconvulsive therapy; BD, bipolar depression; MDD, major depressive
disorder; N.A., information not available; S. D., standard deviation; and OP, other psychosis
Table 2 rTMS Parameters of Included Randomized Controlled Trials
Treatment Methodology
Study Pairs Frequency % rMT Duration TPPS
strategy RD AC BT BL
Blumberger et 10 Hz vs. 1 Hz R 1450 vs.
L vs. B 110 vs. 100a 6 weeks Augmentation Y Y Y Y
al., 2012 (S)10 Hz L 1215
Fitzgerald et al., 10 Hz vs. 1 Hz R
L vs. B 120 vs. 120 3 weeks NA Augmentation Y NA Y Y
2012 (S)10 Hz L
Rybak et al., 10 Hz vs. 1 Hz R 1200 vs.
L vs. B 110 vs. 110 2 weeks Augmentation Y NA Y Y
2005 (S)10 Hz L 1200
Conca et al., 10 Hz vs. 10 Hz L 1300 vs.
L vs. B 100 vs. 100 1 weeks Augmentation Y NA NA Y
2002 (S) 1 Hz R 1300
Fitzgerald et al., 1 Hz vs. 1 Hz L (S) 900 vs. 85% augmentation,
R vs. B 110 vs. 110 4 weeks Y Y Y Y
2011 1 Hz R 900 15% monotherapy
Pallanti et al., 1 Hz vs. 10 Hz L 420 vs.
R vs. B 110 vs. 110 3 weeks Augmentation Y Y Y Y
2010 (S) 1 Hz R 1420
Fitzgerald et al., 1 Hz vs. 1 Hz R+10 900 vs. 84% augmentation,
R vs. B 110 vs. 110 4 weeks Y Y Y Y
2013 Hz L 900 16% monotherapy
Hansen et al., R vs.
1 Hz vs. right 110 3 weeks NA Augmentation Y Y N Y
2011 ECT
Eranti et al., L vs. 10 Hz vs. right or
110 2 weeks 1000 Augmentation Y Y N Y
2007 ECT bilateral
L vs. 10 Hz vs. right or
Rosa et al., 2006 100 4 weeks 2500 Monotherapy Y NA N Y
ECT bilateral
Wang et al., L vs.
20 Hz vs. N.A 70 2 weeks 500 Monotherapy Y NA NA Y
2004 ECT
Grunhaus et al., L vs. 10 Hz vs. right or
90 4 weeks 1200 Monotherapy Y N N Y
2003 ECT bilateral
Janicak et al., L vs. 9% augmentation,
10 Hz vs. bilateral 110 4 weeks 1000 Y NA NA Y
2002 ECT 91% monotherapy
Grunhaus, et al., L vs. 10 Hz vs. right or 400 or 78% augmentation,
90 4 weeks Y NA NA Y
2000 ECT bilateral 1200 22% monotherapy
Pridmore et al., L vs. 1200-
20 Hz vs. N.A 100 2 weeks Monotherapyb Y Y N Y
2000 ECT 1400
L vs.
Wan et al., 2011 10 Hz vs. N.A 100 4 weeks 1500 Augmentation Y N NA Y
ECT
Fitzgerald et al.,
L vs. R 10 Hz vs. 1 Hz 100 vs. 110 3 weeks NA Augmentation Y NA Y Y
2007
Fitzgerald et al., 64% augmentation,
L vs. R 10 Hz vs. 1 Hz 100 vs. 110 3 weeks NA Y Y Y Y
2009 37% monotherapy
Rossini et al., 600 vs.
L vs. R 15 Hz vs. 1 Hz 100 vs. 100 2 weeks Augmentation Y NA NA Y
2010 600
1 Hz or 10 Hz vs. 1
Stern et al., 2007 L vs. R 110 vs. 110 2 weeks NA Monotherapy Y NA Y Y
Hz
Triggs et al.,
L vs. R 5 Hz vs. 5 Hz 100 vs. 100 2 weeks NA Augmentation Y NA Y Y
2010
Höppner et al.,
L vs. R 20 Hz vs. 1 Hz 90 vs. 110 2 weeks NA Augmentation Y Y NA Y
2003
2000 vs.
Eche et al., 2012 L vs. R 10 Hz vs. 1 Hz 100 vs. 100 2 weeks Augmentation Y NA N Y
120
Isenberg et al.,
L vs. R 20 Hz vs. 1 Hz 80 vs. 110 4 weeks NA Monotherapy Y NA N Yc
2005
Fitzgerald et al., 1000 vs.
L vs. R 10 Hz vs. 1 Hz 100 vs. 100 4 weeks Augmentation Y Y Y Y
2003 300
a 120% of the rMT in subjects older than 60 years old.
b Medication was tapered and ceased where possible; no new medication was commenced in the two weeks before entry
into the study.
c Subjects receiving rTMS treatment on the right side were notably older.
Abbreviations: rTMS, repetitive transcranial magnetic stimulation; TPPS, total pulse per session in rTMS; RD,
randomized; AC, allocation concealment; BT, blind treatment; and BL, baseline; L, left rTMS; R, right rTMS; B,
bilateral rTMS; ECT, electroconvulsive therapy; rMT, resting motor threshold; (S), sequential; Y, yes; N, no; and NA,
not available.
Table 3 Response and dropout rates for efficacy and acceptability in meta-analyses of direct comparisons
between each pair of treatment modality
Number Number Efficacy Acceptability
of studies of patients Response rate OR (95% CI) Dropout rate OR (95% CI)
L vs.
R 8 265 53/137vs53/128 0.99(0.58,1.69) 7/46vs0/31 3.97(0.63,25.1)
B 4 150 24/81vs23/69 0.52(0.09,3.06) 4/72vs7/60 0.52(0.12,2.15)
ECT 7 262 67/135vs78/127 0.64(0.30,1.38) 4/37vs7/31 0.42(0.11,1.65)
R vs.
L 8 265 53/128vs53/137 1.01(0.59,1.73) 0/31vs7/46 0.25(0.04,1.59)
B 2 258 46/91vs81/167 1.21(0.72,2.06) 23/71vs36/147 1.48(0.79,2.76)
ECT 1 60 7/30vs12/30 0.46(0.15,1.40) 10/30vs8/30 1.38(0.45,4.17)
B vs.
L 4 150 23/69vs24/81 1.93(0.33,11.4) 7/60vs4/72 1.94(0.46,8.11)
R 2 258 81/167vs46/91 0.82(0.49,1.40) 36/147vs23/71 0.68(0.36,1.26)
ECT vs.
L 7 262 78/127vs67/135 1.55(0.72,3.34) 7/31vs4/37 2.37(0.61,9.27)
R 1 60 12/30vs7/30 2.19(0.72,6.70) 8/30vs10/30 0.73(0.24,2.21)
rTMS, repetitive transcranial magnetic stimulation; L, left rTMS; R, right rTMS; B, bilateral rTMS; ECT,
electroconvulsive therapy; OR, odds ratio; vs., versus; CI, confidence interval.