A systematic review and meta-analysis of deep brain stimulation in treatment-resistant depression

Accepted Manuscript A systematic review and meta-analysis of deep brain stimulation in treatment-resistant depression

Chanjuan Zhou, Hanping Zhang, Yinhua Qin, Tian Tian, Bing Xu, Jianjun Chen, Xinyu Zhou, Li Zeng, Liang Fang, Xunzhong Qi, Bin Lian, Haiyang Wang, Zicheng Hu, Peng Xie PII: DOI: Reference:

S0278-5846(17)30307-X doi:10.1016/j.pnpbp.2017.11.012 PNP 9278

To appear in:

Progress in Neuropsychopharmacology & Biological Psychiatry

Received date: Revised date: Accepted date:

20 April 2017 9 November 2017 9 November 2017

Please cite this article as: Chanjuan Zhou, Hanping Zhang, Yinhua Qin, Tian Tian, Bing Xu, Jianjun Chen, Xinyu Zhou, Li Zeng, Liang Fang, Xunzhong Qi, Bin Lian, Haiyang Wang, Zicheng Hu, Peng Xie , A systematic review and meta-analysis of deep brain stimulation in treatment-resistant depression. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Pnp(2017), doi:10.1016/j.pnpbp.2017.11.012

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.

ACCEPTED MANUSCRIPT A Systematic Review and Meta-Analysis of Deep Brain Stimulation in Treatment-Resistant Depression ChanjuanZhou1,2,3,6* , HanpingZhang2,3,4* , YinhuaQin2,3* , TianTian2,3,4* , Bing Xu2,3,4* , JianjunChen2,3,5,6 , XinyuZhou2,3,7 , Li Zeng2,3,4 , Liang Fang1 , XunzhongQi2,3,4 , Bin

1

PT

Lian2,3 , HaiyangWang2,3,6 , ZichengHu2,3 , PengXie1,2,3,4,6 Department of Neurology, Yongchuan Hospital of Chongqing Medical University,

RI

Chongqing, China

Institute of Neuroscience, Chongqing Medical University, China

3

Chongqing Key Laboratory of Neurobiology, Chongqing Medical University, China

4

Department of Neurology, the First Affiliated Hospital of Chongqing Medical

NU

SC

2

University, Chongqing, China

Institute of Life Sciences, Chongqing Medical University, Chongqing, China

6

Institute of Neuroscience and the Collaborative Innovation Center for Brain Science,

MA

5

Department of Neurology and Psychiatry, the First Affiliated Hospital of Chongqing

EP T

7

ED

Chongqing Medical University, China;

Medical University, Chongqing, China

*

AC C

Short title: Antidepressant effect of DBS in TRD: a meta-analysis These authors contributed equally in this work

Direct correspondence to: Professor Peng Xie Department of Neurology, The First Affiliated Hospital of Chongqing Medical University, 1 Youyi Road, Yuzhong District, Chongqing, P.R.C. 400016 Tel: +86-23-68485490; Fax: +86-23-68485111; E-mail: [email protected]

1

ACCEPTED MANUSCRIPT Abstract BACKGROUND:

Deep

brain

stimulation

(DBS)

has

been

applied

in

treatment-resistant depression (TRD) as a putative intervention targeting different brain regions. However, the antidepressant effects of DBS for TRD in recent clinical

PT

trials remain controversial.

RI

METHODS: We searched Scopus, EMBASE, the Cochrane Library, PubMed, and

SC

PsycINFO for all published studies investigating the efficacy of DBS in TRD up to Feb 2017. Hamilton depression rating scale (HDRS) scores and Montgomery–Asberg

NU

depression rating scale (MARDS) scores were compared between baseline levels and

MA

those after DBS using the standardized mean difference (SMD) with 95% confidence intervals (CIs). The pooled response and remission rates were described using Risk

ED

Difference with 95% CIs.

RESULTS: We identified 14 studies of DBS in TRD targeting the subcallosal

EP T

cingulate gyrus (SCG), ventral capsule/ventral striatum (VC/VS), medial forebrain bundle (MFB), and nucleus accumbens (NAcc). The overall effect sizes showed a

AC C

significant reduction in HDRS after DBS stimulation in these four regions, with a standardized mean difference of −3.02 (95% CI = −4.28 to −1.77, p <0.00001) for SCG, −1.64 (95% CI = −2.80 to −0.49, p = 0.005) for VC/VS, −2.43 (95% CI = −3.66 to −1.19, p =0.0001) for MFB, and −1.30 (95% CI = −2.16 to −0.44, p = 0.003) for NAcc. DBS was effective, with high response rates at 1, 3, 6, and 12 months. Some adverse events (AEs), especially some specific AEs related to targeting regions, occurred during the DBS treatment. 2

ACCEPTED MANUSCRIPT CONCLUSIONS: DBS significantly alleviates depressive symptoms in TRD patients by targeting the SCG, VC/VS, MFB, and NAcc. Several adverse events might occur during DBS therapy, although it is uncertain whether some AEs can be linked to DBS

PT

treatment. Further confirmatory trials are required involving larger sample sizes.

RI

Keywords: Deep brain stimulation (DBS); Treatment-resistant depression (TRD);

AC C

EP T

ED

MA

NU

SC

Safety; Antidepressant effect; Meta-analysis

3

ACCEPTED MANUSCRIPT 1. Introduction Depression is the most common of all mental disorders, affecting more than 300 million people of all ages globally, and ranks among the top causes of disability (1). Although depression can be effectively treated in the majority of patients by

PT

pharmacotherapy such as selective serotonin reuptake inhibitors (SSRIs) or

RI

serotonin noradrenaline reuptake inhibitors (SNRIs), psychotherapy such as

SC

cognitive-behavior therapy (CBT) or interpersonal psychotherapy (IPT) and electroconvulsive therapy (2-4), almost 30% of patients fail to respond to inte rventions

(5,6).

For this

subpopulation

NU

adequate

diagnosed

with

MA

treatment-resistant depression (TRD), it is important to explore alternative treatment for them (7, 8).

ED

Recently, deep brain stimulation (DBS) has been used as a potential

EP T

neurosurgical treatment in cooperation with antidepressant medicine for TRD, because of its adjustable and reversible electrical stimulation. The first clinical trial proved the efficacy of DBS treatment in TRD patients by stimulating the subcallosal

AC C

cingulate gyrus (SCG) (9). Later, DBS for TRD was studied in a number of open- label trials targeting different brain regions, including the SCG, ventral capsule/ventral striatum (VC/VS), nucleus accumbens (NAcc), and the medial forebrain bundle (MFB). And there were also data of DBS targeting VC/VS from randomized controlled trials (RCTs) in TRD patients (10, 11). Although some of these studies confirmed the treatment efficacy and safety of short- or long-term active DBS in TRD, there were inconsistent results. For instance, a RCT of DBS targeting the VC/VS in 4

ACCEPTED MANUSCRIPT TRD reported no differences between the active and control groups at the end of the 16-week controlled phase (10). By contrast, a significant decrease in depressive symptoms was identified in TRD patients with active DBS in the ventral anterior limb of the internal capsule, compared with sham stimulation (11). The contrasting

PT

antidepressant effects of DBS in TRD treatment may be due to the difference in

RI

clinical trial design, DBS treatment duration, the heterogeneity in disease

SC

pathology and limited sample size. Therefore, the optimal target brain regions and treatment stimulation for DBS remain undetermined. However, few specific

NU

meta-analyses focus on these impact factors for DBS in TRD patients. A review of

MA

clinical outcomes demonstrated that NAcc-DBS and VC/VS-DBS benefit patients with MDD (12), and a previous meta-analysis only focused on SGC-DBS trials

ED

finding promising outcomes in TRD patients (13). Owing to the lack of evidence and

EP T

practical information on DBS treatment among TRD patients, we performed herein an updated systematic review and meta-analyses to detail the effects of DBS in TRD. 2. Methods

AC C

2.1.Search Strategy

We systematically searched electronic databases including Scopus, EMBASE, the Cochrane Library, PubMed, and PsycINFO for all published studies examining the efficacy of DBS for TRD up to February 2017. The searched MeSH terms were: (“deep brain stimulation” OR DBS) AND (“treatment-resistant depression” OR TRD). 2.2.Study Selection Strategy Three independent authors (ZCJ, TT, and CJJ) screened and selected eligible studies 5

ACCEPTED MANUSCRIPT for analysis, and the procedure is shown in Figure 1. All patients recruited for the clinical trials were diagnosed with TRD. We included research that investigated DBS treatment for improvement of depressive symptoms that were evaluated using either the Hamilton depression rating scale (HDRS) or Montgomery–Asberg depression

PT

rating scale (MARDS) in TRD patients. We also profiled the response rate and

RI

adverse effects of DBS during therapy in each included study. Studies with new

SC

antidepressants added or changed for TRD during the DBS trial were excluded. Abstracts, case studies, reviews, and duplicate cohorts were excluded.

NU

2.3.Data Extraction and Outcome Measures

MA

Four authors (ZCJ, ZHP, QYH and XB) independently extracted data to avoid extraction errors, and discrepancies were resolved through discussion. The following

ED

parameters were extracted from each eligible article: first author, publication year,

EP T

country of origin, mean age, DBS stimulation parameters, number of patients, measurement outcomes (HDRS scores, MARDS scores), side effects, and treatment regions. The primary outcomes were HDRS and MARDS scores in TRD patients

AC C

from baseline to post-treatment. We extracted the data for short- and long-term DBS treatment at different time points (1, 3, 6, 12 months) in each study, and the last time point within the range was considered if a study reported data for more than one time within our pre-defined open-label optimization phase. 2.4.Statistical Methods Statistical analyses were performed using statistical software Cochrane Review Manager (Rev Man 5.1.1) and STATA 12.0 (Stata Corp, College station, TX). 6

ACCEPTED MANUSCRIPT Improvements in depressive symptoms before and after DBS stimulation were described using HDRS or MDRS, and were assessed by standardized mean differences (SMDs) with 95% confidence intervals (CIs) for each study. To evaluate the response and remission rates related to DBS, the pooled response and remission

PT

rates were described by the Risk Difference with 95% CIs. All p-values were

RI

two-sided with a p< .05 being considered statistically significant. A chi-squared-based

SC

Q-statistic test was used to detect heterogeneity among studies. We used a random-effects model since we assumed that the true treatment effects varied between

NU

the included studies. To evaluate possible biases, a sensitivity analysis was conducted

MA

by the leave-one-out method to assess the contribution of each individual dataset to the pooled SMDs. Finally, we estimated the publication bias using Egger’s test, with a

ED

p < .05 being considered statistically significant. For adverse events (AEs), the

EP T

number of patients that reported any AE (N) was used as the denominator. The number of patients in each AE (n) was used as the numerator (AEs% = n/N100%). 3. RESULTS

AC C

We initially identified 60 potentially relevant records according to pre-defined MeSH terms. After screening titles and abstracts, 30 articles were full-text reviewed, and 14 studies were ultimately included in this meta-analysis based on our inclusion criteria (10, 11, 14-25). The study characteristics are displayed in Table 1. The other 16 studies were excluded for the following reasons: (i) two studies described protocol design and eligibility for DBS in TRD (26,27); (ii) six trials were considered likely to use duplicated patient recruitments and clinical outcomes (9, 28-32); (iii) two studies 7

ACCEPTED MANUSCRIPT were case reports (33,34); (iv) three studies were reviews and a meta-analysis of DBS (13,35-36); (v) one study only reported the bias towards negative words rather than depressive symptom reduction (37); (vi) one study was a single lab perspective (Emory University, Atlanta GA, USA) from 8 years of studies of SCG DBS for TRD

PT

(38); (vii) one study defined the obstacles of DBS surgery for stimulation of the

RI

lateral habenular (LH) complex rather than the antidepressant effects of DBS in TRD

SC

(39). 3.1.Meta-analyses of DBS in TRD

NU

The 14 included trials mainly targeted four brain regions: seven for the SCG, three for

MA

the VC/VS, two for the MFB, and two for the NAcc. 3.1.1. SCG DBS in TRD

ED

Seven studies were included in our meta-analysis of SCG DBS, with a total of 77 subjects with TRD. Of the seven studies, all evaluated clinical efficacy using the

EP T

HDRS, while three also used the MADRS. The pooled Hedges’ g effect sizes showed a significant decrease in HDRS scores (SMD −3.02; 95% CI = −4.28 to −1.77, Z =

AC C

4.71, p< .00001; Figure 2), and MADRS scores (SMD −1.32; 95% CI = −2.10 to −0.53, Z = 3.29, p = .001; Figure 3) between baseline and post-DBS stimulation for the entire group. I2 was 83% in the meta-analysis of HDRS, suggesting strong heterogeneity, while no heterogeneity was found in the MADRS analysis (I2 = 0%). The pooled Hedges’ g effect sizes at all four time points showed a significant and large HDRS reduction in depressive symptoms (Figure 4), with an SMD of −3.31 (95% 8

ACCEPTED MANUSCRIPT CI = −4.92 to −1.70, Z = 4.04, p< .0001) at 1 month, −2.52 (95% CI = −3.22 to −1.81, Z = 7.00, p < .00001) at 3 months, −3.71 (95% CI = −5.36 to −2.06, Z = 4.40, p< .0001) at 6 months, and −3.76 (95% CI = −5.32 to −2.21, Z = 4.75, p< .00001) at 12 months. A strong heterogeneity remained between the studies (I2 = 83%).

PT

3.1.2. VC/VS DBS in TRD

RI

Patients from three studies were treated with DBS bilateral to the VC/VS. All three

SC

studies used MADRS and two also used HDRS to evaluate clinical efficacy. There was a significant decrease in MADRS scores in 55 patients (SMD −1.19; 95% CI =

NU

−1.81 to −0.56, Z = 3.74, p= .0002; Figure 5). Although there were a limited number

MA

of studies using HDRS, the HDRS scores were also significantly improved by DBS (SMD −1.64; 95% CI = −2.80 to −0.49, Z = 2.78, p = .005; Figure 6). I2 was 77% in

ED

the meta-analysis of HDRS scores, suggesting strong heterogeneity, while a slight heterogeneity was found in the MADRS score analysis (I2 = 54%).

EP T

The pooled Hedges’ g effect sizes of the MADRS scores were SMD −1.48 (95% CI = −2.82 to −0.14, Z = 2.17, p = .03) at 3 months, and −1.40 (95% CI = −2.38 to −0.42,

AC C

Z = 2.80, p = .005) at 12 months, suggesting an improvement after both short- and long term DBS stimulation (Figure 7). 3.1.3. MFB DBS in TRD Two studies were included in the meta-analysis of MFB-DBS in TRD, and all studies used both HDRS and MADRS for evaluating clinical efficacy. There was a significant reduction in depressive symptom severity after DBS stimulation for both the HDRS scores (SMD −2.43; 95% CI = −3.66 to −1.19, Z = 3.85, p= .0001; Figure 8) and 9

ACCEPTED MANUSCRIPT MADRS scores (SMD −1.91; 95% CI = −3.01 to −0.81, Z = 3.40, p< .0007; Figure 9). There was no significant heterogeneity based on HDRS or MADRS (I2 = 0%). 3.1.4. NAcc DBS in TRD Only two studies investigated the antidepressant effects of DBS targeting the NAcc in

PT

TRD using HDRS, and both were included in our meta-analysis. The pooled Hedges’

RI

g effect sizes showed significant antidepressant effects for MFB-DBS (SMD −1.30;

SC

95% CI = −2.16 to −0.44, Z = 2.97, p = .003; Figure 10).There was no significant

3.1.5. Sensitivity and Publication Bias

NU

heterogeneity based on HDRS (I2 = 0%).

MA

The results of the sensitivity analysis confirmed that no single study had influenced the pooled SMDs. Furthermore, no strong statistical evidence for publication bias was

ED

observed in the Egger’s test (all p > .05).

EP T

3.2. Response and Remission Rate of DBS in TRD The response and remission rates in patients were directly extracted and calculated from each study (Supplemental Table S1). The percentage of patients who achieved a

AC C

50% reduction in the severity of depression as measured by MADRS or HDRS was defined as a ‘response’, while those who achieved an HDRS-17 score <8, HDRS-28 score <10, or a >75% reduction in MADRS were defined as in clinical ‘remission’. The pooled response rates and remission rates of the overall analysis are shown in Table 2. Over 1–12 months of DBS stimulation, the pooled response rates were 37% at 1 month, 36% at 3 months, 50% at 6 months, and 48% at 12 months, while the pooled remission rates were 10% at 1 month, 25% at 3 months, 25% at 6 months, and 10

ACCEPTED MANUSCRIPT 30% at 12 months. There was a moderate heterogeneity (I2 = 53%) in response rates and a low heterogeneity (I2 = 14%) in remission rates between studies. 3.3.AEs Associated with DBS in TRD In the 14 studies meeting our inclusion data, 13 reported clinical AEs. We described

PT

and categorized all AEs as surgery-related, device-related, or psychiatric and somatic

RI

side effects during the optimization phase related to DBS stimulation. No cognitive

SC

side effects were observed in all included trials during DBS treatment. AEs reported by patients are described in Supplemental Table S2. AEs related to the surgical

NU

procedure were found in 8 trials for several days or within 1 month after the surgical

MA

procedure for the treatment of TRD (Table 3). Swollen eye (7.69%) was the most common AE, followed by headache (5.77%), pain (5.13%), and infection (5.13%).

ED

Some AEs, such as aconuresis, occurred due to anesthesia. The AEs related to device

EP T

use are shown in Table 4; transient pain (2.56%) in four patients was the major device-related AE.

During the stimulation phase, AEs were classified as somatic events, psychiatric

AC C

events, and other events. As shown in Table 5, the most common somatic side effects were headaches (13.46%), nausea/vomiting/diarrhea (8.33%), balance/dizziness (7.69%) and oculomotor (blurred vision/double vision) (5.77%). The adverse psychiatric events are shown in Table 6. The most common psychiatric side effect was worsening depression (7.69%) and agitation (7.69%),f ollowed by sleep disturbances (6.41%), suicide attempt (6.41%), anxiety (5.77%), disinhibition (5.13%), and suicide ideation (4.49%). Suicide completion occurred in 2.56% of TRD patients 11

ACCEPTED MANUSCRIPT during the optimization phase. Other adverse events that were unknown, some of which

were

not

associated

with

device/surgery

or

changes

in

DBS

stimulation/parameters, are shown in Table 7. Skin problems (7.05%) such as erythema were the most common events.

PT

Specific AEs related to target region were classified, and no specific AEs were

RI

observed from the included SCG-DBS trials. AEs including hypomania (2.56%),

SC

mania (1.92%), and disinhibition (5.13%) are specific to VC/VS-DBS therapy. AEs such as vision/eye movement disorder (5.13%) and oculomotor (5.13%) had relatively

NU

high occurrence during MFB-DBS trials. The NAcc DBS treatment led to aberrant

MA

behaviors, including eating problems (1.28%) and sexuality changes (0.64%). 4. DISCUSSION

ED

Our study systematically reviewed the antidepressant efficacy and safety of DBS

EP T

targeting different eligible brain regions for TRD. The meta-analysis results confirmed and extended previous observations that DBS provides significant antidepressant effects when targeting the SCG, VC/VS (ALIC), MFB, and NAcc, with

AC C

a large reduction in HDRS or MADRS. We also found a significant reduction in depressive symptoms following either short-term or long-term DBS, with response rates of 37% at 1 month, 36% at 3 months, 50% at 6 months, and 48% at 12 months in patients with TRD. By comparison, response rates increased to 50% after 6 months, which was sustained over the last 6–12 months. Furthermore, the responders maintained increased remission especially from 6 months (25%) to 12 months (30%). Indeed, patients with TRD had markedly improved health and showed long- lasting 12

ACCEPTED MANUSCRIPT benefits from DBS. In a follow- up study by Kennedy et al., the response rates of DBS in TRD patients were 62.5% after 1 year, 46.2% after 2 years, 75.0% after 3 years, and 64.3% at the last follow- up visit (31). Combined with previous reports, these data suggest that DBS remains an effective short-term and long-term treatment for TRD.

PT

It remains unclear why some patients with TRD respond well to DBS, while others

RI

do not. In the 14 included studies in the present analysis, the mean stimulation

SC

parameters were: amplitude 3.5–5 V or current 4 mA, pulse width 90 µs, and frequency 130 Hz. If no improvement occurred, the stimulation intensity was adjusted

NU

and increased to provide optimization of the electrical stimulation parameters. Based

MA

on a small number of patients, the DBS stimulation para meters of TRD subjects were quite different over the included studies, despite the fact that the same brain region

ED

was being targeted. For instance, the parameters were amplitude 3.5 V, pulse width 90

EP T

µs, and frequency 130 Hz for SCG-DBS in 20 patients from Lozano et al. (14), while an amplitude maximum of 10 V (range, 2.5–10 V) was reported by Merkl et al. (18). A similar phenomenon was seen for DBS applied to the VC/VS (ALIC) region, with

AC C

an amplitude of 3.0 V, pulse width of 60 µs, and frequency of 130 Hz used for 15 patients (10), but an amplitude of 2.5–7 V, pulse width of 90/210 µs, and frequency of 100–130 Hz was used in a different group of 25 patients (21).Thus, differences in the various DBS stimulation parameters may alter its efficacy with respect to depressive symptoms, while different subjects may also have different responses to the reported averaged parameters. At present, there is no evidence-based guideline for the selection of optimal electrical parameters for TRD. The stimulation settings for individual 13

ACCEPTED MANUSCRIPT patients are often guided by the clinician’s previous experience in order to maximize therapeutic efficacy and minimize adverse effects. More rigorous evaluation of different stimulus parameters is required in future studies to determine the optimal settings.

PT

Positron emission tomography (PET) studies have also reported that DBS can

RI

influence and modulate the activity of brain regions that are distributed along

SC

downstream targets of the SCG (9). Furthermore, Bewernick et al. found changes in metabolic activity across cortical and subcortical areas following NAcc-DBS (24).

NU

However, very minimal and insignificant changes were seen in TRD patients

MA

receiving DBS to the MFB (22). It is also possible that individual neuroanatomical variability may contribute to the rate of clinical improvement following DBS. In this

ED

context, the combination of functional neuroimaging monitoring of regional brain

EP T

activity and metabolism may help to identify and evaluate anticipated biological responses to DBS in TRD.

In the present study, AEs were infrequent and mostly transient in DBS for TRD

AC C

treatment. Headaches were the most commonly occurred AEs during both the surgical procedure and the stimulation phase. Of particular note, there was 2.56% incidence of suicidality in three studies which included trials targeting the SCG, VC/VS, and NAcc respectively. The AEs of both suicide attempt (6.41%) and suicide ideation (4.49%) are also relatively high. However, patients with TRD have a higher risk of attempting suicide than patients with MDD in general (40). Meanwhile, there is still a lack of accurate comparisons of these events pre- and post- DBS treatment. Besides, the 14

ACCEPTED MANUSCRIPT completed suicide did not only occur during the process of DBS therapy, but also occurred during the cessation of intervention and the long te rm follow up. A non-responder who was taken off stimulation committed suicide (10), and two patients who received DBS committed suicide during depressive relapses in the 3 to 6

PT

year follow-up (31). Thus suicidality should be carefully monitored and recorded to

RI

establish whether DBS stimulation may increase the associated risk.

SC

There are several limitations to our findings. First, a small number of TRD patients were enrolled in each clinical trial. Furthermore, the number of DBS studies targeting

NU

the MFB and NAcc were limited, although we were able to perform a relevant

MA

meta-analysis. Secondly, strong heterogeneity was found in the HDRS meta-analysis, but not for the MADRS, which is likely related to different type of HDRS scales

ED

being used in the included studies. In addition, HDRS functions as a multi-domain

EP T

depression scale while MADRS does not. Thirdly, we analyzed the response and remission rates with pooled outcomes, without targeting specific brain regions, because of the limited number of studies. Finally, although we recorded several AEs,

AC C

most of the associations between these side effects and DBS remain uncertain as to whether they are truly linked to DBS treatment, especially with regards to suicidality. Some AEs such as flight of ideas and hallucination were also considered to be rare or unrelated to DBS, and were not specifically described in our study. Thus, it remains possible that such AEs may occur in DBS for TRD. 5. Conclusion In general, our systematic meta-analyses of 14 clinical trials related to DBS in TRD 15

ACCEPTED MANUSCRIPT suggest that depressive symptoms were effectively and significantly alleviated following either short-term or long-term (range, 1–12 months) DBS stimulation in various brain regions, including the SCG, VC/VS (ALIC), MFB, and NAcc. Although several adverse events occurred during DBS therapy, most of these associations

PT

remain uncertain. Future sham-control or open- label clinical trials on DBS for TRD

RI

should include larger sample sizes, targeting differential and optimal reported brain

SC

regions, which will provide further therapeutic understanding on the use of DBS in TRD.

NU

ACKNOWLEDGMENTS

FINANCIAL DISCLOSURES

MA

None

ED

This work was supported by The National Key Research and Development Programm

EP T

of China (Grant No.2017YFA0505700), and National Natural Science Foundation of China (Grant No.81601207). Ethical Statement

AC C

The protocols of clinical experimentation were approved by the Ethics Committee of Chongqing Medical University (ECCU, Chongqing, China)

COMPETING INTERESTS The authors have declared that no competing interests exist.

REFERENCES 1.

WHO

2017:

Depression.

online: http://www.who.int/mediacentre/factsheets/fs369/en/ 16

Available

ACCEPTED MANUSCRIPT 2. Lisanby SH. (2007): Electroconvulsive therapy for depression. New England Journal of Medicine 357(19): 1939-1945. 3. Grunhaus L, Dannon PN, Schreiber S, Dolberg OH, Amiaz R, Ziv R, et al. (2000): Repetitive transcranial magnetic stimulation is as effective as electroconvulsive

PT

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

RI

Biological psychiatry 47(4): 314-324.

SC

4. Chen JJ, Zhao LB, Liu YY, Fan SH, Xie P. (2017): Comparative efficacy and acceptability of electroconvulsive therapy versus repetitive transcranial magnetic

NU

stimulation for major depression: A systematic review and multiple-treatments

MA

meta-analysis. Behavioural brain research 320: 30-36. 5. Keller MB, Lavori PW, Mueller TI, Endicott J, Coryell W, Hirschfeld RM, et al.

ED

(1992): Time to recovery, chronicity, and levels of psychopathology in major

EP T

depression: a 5- year prospective follow-up of 431 subjects. Archives of General Psychiatry 49(10): 809-816.

6. Fava M. (2003): Diagnosis and definition of treatment-resistant depression.

AC C

Biological psychiatry 53(8): 649-659. 7. Fava M, Davidson KG. (1996): Definition and epidemiology of treatment-resistant depression. Psychiatric Clinics of North America 19(2): 179-200. 8. Zhou X, Keitner GI, Qin B, Ravindran AV, Bauer M, Del Giovane C, et al. (2015): Atypical antipsychotic augmentation for treatment-resistant depression: a systematic review

and

network

meta-analysis.

Neuropsychopharmacology, pyv060. 17

International

Journal

of

ACCEPTED MANUSCRIPT 9. Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, et al. (2005): Deep brain stimulation for treatment-resistant depression. Neuron 45(5): 651-660. 10. Dougherty DD, Rezai AR, Carpenter LL, Howland RH, Bhati MT, O'Reardon JP,

PT

et al. (2015): A randomized sham-controlled trial of deep brain stimulation of the

RI

ventral capsule/ventral striatum for chronic treatment-resistant depression. Biological

SC

Psychiatry 78(4): 240-248.

11. Bergfeld IO, Mantione M, Hoogendoorn ML, Ruhé HG, Notten P, van Laarhoven

NU

J, et al. (2016): Deep brain stimulation of the ventral anterior limb of the internal

MA

capsule for treatment-resistant depression: a randomized clinical trial. JAM A psychiatry 73(5): 456-464.

ED

12. Morishita T, Fayad SM, Higuchi M, Nestor KA, Foote KD. (2014). Deep brain

EP T

stimulation for treatment-resistant depression: systematic review of clinical outcomes. Neurotherapeutics the Journal of the American Society for Experimental Neurotherapeutics 11(3): 475.

AC C

13. Berlim MT, McGirr A, Van den Eynde F, Fleck MP, Giacobbe P. (2014): Effectiveness and acceptability of deep brain stimulation (DBS) of the subgenual cingulate cortex for treatment-resistant depression: A systematic review and exploratory meta-analysis. Journal of affective disorders 159: 31-38. 14. Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH. (2008): Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biological psychiatry 64(6): 461-467. 18

ACCEPTED MANUSCRIPT 15. Holtzheimer PE, Kelley ME, Gross RE, Filkowski MM, Garlow SJ, Barrocas A, et al. (2012): Subcallosal cingulate deep brain stimulation for treatment-resistant unipolar and bipolar depression. Archives of general psychiatry 69(2): 150-158. 16. Lozano AM, Giacobbe P, Hamani C, RizviSJ, Kennedy SH, Kolivakis TT, et al.

PT

(2012). A multicenter pilot study of subcallosal cingulate area deep brain stimulation

RI

for treatment-resistant depression: clinical article. Journal of neurosurgery 116(2):

SC

315-322.

17. Puigdemont D, Pérez- Egea R, Portella MJ, Molet J, de Diego-Adeliño J, Gironell

in

treatment-resistant

major

depression.

International Journal of

MA

evidence

NU

A, et al. (2012): Deep brain stimulation of the subcallosal cingulate gyrus: further

Neuropsychopharmacology 15(1): 121-133.

ED

18. Merkl A, Schneider GH, Schönecker T, Aust S, Kühl KP, Kupsch A, et al. (2013):

EP T

Antidepressant effects after short-term and chronic stimulation of the subgenual cingulate gyrus in treatment-resistant depression. Experimental neurology 249: 160-168.

AC C

19. Ramasubbu R, Anderson S, Swati Chavda MBBS M, Kiss ZH. (2013): Double-blind optimization of subcallosal cingulate deep brain stimulation for treatment-resistant depression: a pilot study. Journal of psychiatry & neuroscience: JPN 38(5): 325. 20. AccollaEA, Aust S, Merkl A, Schneider GH, Kühn AA, Bajbouj M, et al. (2016): Deep brain stimulation of the posterior gyrus rectus region for treatment resistant depression. Journal of affective disorders 194: 33-37. 19

ACCEPTED MANUSCRIPT 21. Malone DA, Dougherty DD, Rezai AR, Carpenter LL, Friehs GM, Eskandar EN, et al. (2009): Deep brain stimulation of the ventral capsule/ventral striatum for treatment-resistant depression. Biological psychiatry 65(4): 267-275. 22. Fenoy AJ, Schulz P, Selvaraj S, Burrows C, Spiker D, Cao B, et al. (2016): Deep

RI

depression. Journal of affective disorders 203: 143-151.

PT

brain stimulation of the medial forebrain bundle: Distinctive responses in resistant

(2017):

Deep

brain

stimulation

to

the

SC

23. Bewernick BH, Kayser S, Gippert SM, Switala C, Coenen VA, Schlaepfer T E medial

forebrain

bundle

for

NU

depression-long-term outcomes and a novel data analysis strategy. Brain Stimulation.

MA

24. Bewernick BH, Hurlemann R, Matusch A, Kayser S, Grubert C, Hadrysiewicz B, et al. (2010): Nucleus accumbens deep brain stimulation decreases ratings of

ED

depression and anxiety in treatment-resistant depression. Biological psychiatry 67(2):

EP T

110-116.

25. Millet B, Jaafari N, Polosan M, Baup N, Giordana B, Haegelen C, et al. (2014): Limbic versus cognitive target for deep brain stimulation in treatment-resistant accumbens

AC C

depression:

more

promising

than

caudate.

European

Neuropsychopharmacology 24(8): 1229-1239. 26. Koek RJ, Langevin JP, Krahl SE, Kosoyan HJ, Schwartz HN, Chen JW, et al. (2014): Deep brain stimulation of the basolateral amygdala for treatment-refractory combat post-traumatic stress disorder (PTSD): study protocol for a pilot randomized controlled trial with blinded, staggered onset of stimulation. Trials 15(1): 356. 27. Filkowski MM, Mayberg HS, Holtzheimer PE. (2016): Considering eligibility for 20

ACCEPTED MANUSCRIPT studies of deep brain stimulation for treatment-resistant depression: insights from a clinical trial in unipolar and bipolar depression. The journal of ECT 32(2): 122-126. 28. Kubu CS, Brelje T, Butters MA, Deckersbach T, Malloy P, Moberg P, et al. (2016): Cognitive

outcome

after

ventral

capsule/ventral

striatum

stimulation

for

PT

treatment-resistant major depression. Journal of Neurology, Neurosurgery &

RI

Psychiatry, jnnp-2016.

SC

29. Schlaepfer TE, Bewernick BH, Kayser S, Mädler B, Coenen VA. (2013): Rapid effects of deep brain stimulation for treatment-resistant major depression. Biological

NU

psychiatry 73(12): 1204-1212.

MA

30. Puigdemont D, Portella MJ, Pérez-Egea R, Molet J, Gironell A, de Diego-Adeliño J, et al. (2015): A randomized double-blind crossover trial of deep brain stimulation of

ED

the subcallosal cingulate gyrus in patients with treatment-resistant depression: a pilot

EP T

study of relapse prevention. Journal of psychiatry & neuroscience: JPN 40(4): 224. 31. Kennedy SH, Giacobbe P, Rizvi SJ, Placenza FM, Nishikawa Y, Mayberg HS, et al. (2011): Deep brain stimulation for treatment-resistant depression: follow-up after 3

AC C

to 6 years. American Journal of Psychiatry 168(5): 502-510. 32. Schlaepfer TE, Cohen MX, Frick C, Kosel M, Brodesser D, Axmacher N, et al. (2008): Deep brain stimulation to reward circuitry alleviates anhedonia in refractory major depression. Neuropsychopharmacology 33(2): 368-377. 33. Sartorius A, Kiening KL, Kirsch P, von Gall CC, Haberkorn U, Unterberg AW, et al. (2010): Remission of major depression under deep brain stimulation of the lateral habenula in a therapy-refractory patient. Biological psychiatry 67(2): e9-e11. 21

ACCEPTED MANUSCRIPT 34. Torres CV, Ezquiaga E, Navas M, Pallero MAG, Sola RG. (2016): Long-term Results

of

Deep

Brain

Stimulation

of

the

Subcallosal

Cingulate

for

Medication-Resistant Bipolar I Depression and Rapid Cycling Bipolar II Depression. Biological Psychiatry.

PT

35. Appleby BS, Duggan PS, Regenberg A, Rabins PV. (2007): Psychiatric and

RI

neuropsychiatric adverse events associated with deep brain stimulation: A meta‐

SC

analysis of ten years' experience. Movement Disorders 22(12): 1722-1728. 36. Gálvez JF, Keser Z, Mwangi B, Ghouse AA, Fenoy AJ, Schulz PE, et al. (2015):

NU

The medial forebrain bundle as a deep brain stimulation target for treatment resistant

Biological Psychiatry 58: 59-70.

MA

depression: a review of published data. Progress in Neuro-Psychopharmacology and

ED

37. Hilimire MR, Mayberg HS, Holtzheimer PE, Broadway JM, Parks NA, DeVylder

EP T

JE, et al. (2015): Effects of subcallosal cingulate deep brain stimulation on negative self-bias in patients with treatment-resistant depression. Brain stimulation 8(2): 185-191.

AC C

38. Crowell AL, Garlow SJ, Riva-Posse P, Mayberg HS. (2015): Characterizing the therapeutic response to deep brain stimulation for treatment-resistant depression: a single center long-term perspective. Frontiers in integrative neuroscience 9. 39. Schneider TM, Beynon C, Sartorius A, Unterberg AW, Kiening KL. (2013): Deep brain stimulation of the lateral habenular complex in treatment-resistant depression: traps and pitfalls of trajectory choice. Neurosurgery 72:ons184-ons193. 40. Amital D, Fostick L, Silberman A, Beckman M, Spivak B. (2008). Serious life 22

ACCEPTED MANUSCRIPT events among resistant and non-resistant mdd patients. Journal of Affective Disorders

AC C

EP T

ED

MA

NU

SC

RI

PT

110(3): 260-264.

23

ACCEPTED MANUSCRIPT Table Legends

Table 1.Demographic and Clinical Characteristics of the Included Studies

PT

Table 2.Meta-Analysis of Response and Remission Rates on Deep Brain Stimulation

SC

RI

(DBS) for Treatment-Resistant Depression (TRD) at Different Time Points

NU

Table 3.Surgery-Related Adverse Events (n = 163)

MA

Table 4. Device-Related Adverse Events (n = 163)

EP T

ED

Table 5. Somatic Adverse Events (n = 163)

Table 6. Psychiatric Adverse Events (n = 163)

AC C

Table 7. Other Adverse Events During Stimulation (n = 163)

Supplementary Tables Legends Supplemental Table S1. Response and remission of patients receiving deep brain stimulation (DBS) in the included studies.

Supplemental Table S2. Adverse effects of DBS in the included studies. 24

ACCEPTED MANUSCRIPT Figure Legends

Figure 1.Flowchart of study selection.

to

the subcallosal cingulate

gyrus (SCG)

in

RI

stimulation (DBS) applied

PT

Figure 2.Meta-analysis of alterations in HDRS scores at baseline and post-deep brain

SC

treatment-resistant depression (TRD).

NU

Figure 3.Meta-analysis of alterations in MADRS at baseline and post-DBS applied to

MA

the SCG in TRD.

ED

Figure 4.Subgroup meta-analysis of alterations in HDRS applied to the SCG at

EP T

different time points of DBS in TRD. Forest plots show the summary effect sizes for HDRS at 1, 3, 6, and 12 months.

AC C

Figure 5.Meta-analysis of alteration in MADRS at baseline and post-DBS applied to the VC/VS in TRD.

Figure 6.Meta-analysis of alterations in HDRS at baseline and post-DBS applied to the VC/VS in TRD.

Figure 7.Subgroup meta-analysis of alterations in MADRS applied to the ALIC at 25

ACCEPTED MANUSCRIPT different time points of DBS in TRD. Forest plots show the summary effect sizes for HDRS at 3 and 12 months.

Figure 8.Meta-analysis of alterations in HDRS at baseline and post-DBS applied to

RI

PT

the MFB in TRD.

SC

Figure 9.Meta-analysis of alterations in MADRS at baseline and post-DBS applied to

NU

the MFB in TRD.

MA

Figure 10.Meta-analysis of alterations in HDRS at baseline and post-DBS applied to

AC C

EP T

ED

the nucleus accumbens (NAcc) in TRD.

26

ACCEPTED MANUSCRIPT Table 1.Demographic and Clinical Characteristicsof Included Studies Y S tudy

ea

Coun

No

try

.

Age

DBS stimulation

Outc

Basel

parameters

omes

ine

r

~1

~3

~6

~12

mont

mont

mont

mont

h

h

h

h

S CG region

20

eimer

12

Lozan

20

o

12

Puigde

20

mont

12

USA Cana da Spain

20

Germ

13

any

Ramas

20

Cana

ubbu

13

da

M erkl

17

21

8

5

4

3.5Volts, 90µs,

HDR

24.4±

15.4±

12.5±

11.8±

12.6±

10.4

130Hz

S-17

3.5

5.7

7.2

5.9

6.3

42.0±

6-10mA,91µs,130H

HDR

23.9±

17.9±

13.1±

13.6±

8.9

z

S-17

0.7

1.5

2.1

47.3 ±

4.2-5.2mA,91-100.5

HDR

27.6±

6.1

µs,128-130.5Hz

S-17

4.5

47.4 ±

3.5-5V,120-210μs,1

HDR

21.3±

11.3

35Hz

S-17

2.4

M AD

28.5±

10.8±

RS

6.3

11.3

/2

a

01

Germ any

6 NAcc region 20

Germ

nick

10

any

M illet

10

20

Franc

14

e

4

15.6±

16.4±

2.9

3.4

3.1

3.5

15.1±

7.9±1

8.9±2

2.5

.5

.9

M AD

33.6±

27±1

z

RS

4.3

0.1

50.25

3.25V,225

HDR

30.8±

19.2±

18.5±

±4.19

µs,130Hz

S-17

2.9

4.5

6.6

M AD

37.8±

26±8.

26.5±

RS

3.9

2

10.4

40.8±

2.5-10V,90µs,130H

HDR

32.4±

24.9±

12.2

z

S-24

5.2

8.5

HDR

32.5±

20.8±

S-28

5.3

12.6

M AD

30.6±

20.3±

RS

4.1

10.2

HDR

25.5±

24.3±

14.5±

12±8.

S-17

3.1

1.7

10.8

5

48.6±

4 V,90µs,130Hz

11.7

AC C

Bewer

7

15.2±

2.5-10V,90µs,130H

ED

Accoll

17.5±

9.2

EP T

13

0.9

50.7 ±

20 M erkl/

PT

Holtzh

da

47.4±

RI

08

20

SC

o

Cana

NU

20

MA

Lozan

55.5±

4-8V,60µs,130Hz

9.56

VC/VS (ALIC) region M alon e

20 09

Dough

20

erty

15

Bergfe

20

ld

16

USA

USA

15

15

Nethe rlands

25

46.3±

2.5-7 V, 100-130Hz,

HDR

33.1±

22.2±

16.6±

17.5±

18.5±

10.8

90/210µs

S-24

5.5

6.2

7.8

8.2

6.8

M AD

34.8±

21.6±

16.1±

17.9±

18.5±

RS

7.3

8.0

9.2

7.9

8.8

47.7

3.0 V, 60 µs, 130

M AD

37.7±

29.7±

±12.0

Hz

RS

4.4

12.6

53.2±

2-6 V, 90 us, 130 or

HDR

22.2±

15.9±

8.4

180 Hz

S-24

4.9

6.2

27

ACCEPTED MANUSCRIPT M AD

34.0±

23.8±

RS

5.8

13.7

MFB region

nick

20 17

3

Germ any

8

46.0 ±

3.0 Volts, 60µs, 130

HDR

40±2.

16.7±

10.81

Hz

S-29

6

12.7

M AD

33±2.

13.3±

RS

6

14.5

41.9 ±

4 Volts, 90µs,

HDR

28.13

11.88

13.88

12.25

10.25

8.7

130Hz

S-28

±4.76

±6.88

±7.92

±7.36

±8.14

M AD

30±7.

11.38

10.5±

RS

39

±7.95

10.39

PT

Bewer

16

USA

11.75

±8.71

11±9

RI

20 Fenoy

MADRS : Montgomery–Asberg depression rating scale.

SC

TRD: treatment-resistant depression; S CG: subgenual cingulated; HDRS : Hamilton depression rating scale;

VC/VS : Ventral Capsule/Ventral; ALIC: Anterior Limb of the Internal Capsule; NAcc: Nucleus

AC C

EP T

ED

MA

NU

Accumbens; MFB: Medial forebrain bundle.

28

ACCEPTED MANUSCRIPT Table 2. Meta-analysis of response and re mission rates on DBS for TRD at different time points

1 month

12 months

1 month

3 months

6 mo

0.37

0.36

0.50

0.48

0.10

0.25

0.25

0.21-0.54 ＜0.0001 52%

0.23-0.48 ＜0.00001 7%

0.33-0.68 ＜0.00001 70%

0.36-0.60 ＜0.00001 45%

0.01- 0.20

0.13-0.37 ＜0.0001 0%

0.15 ＜0. 0%

PT

Risk Difference 95%CIs P（Effect size） I2 index

Remission

Response 3 months 6 months

0.03 8%

AC C

EP T

ED

MA

NU

SC

RI

CIs: Confidence Intervals Clinical response: decrease HDRS score >50%. Clinical remission: absolute HDRS17 score <8. HDRS28 score < 10

29

ACCEPTED MANUSCRIPT

Table 3. Surgery-related adverse events (n=156) No. of AE’s % of AE’s

Intracranial bleeding Headache Seizure Infection Pain around incisions Partial hemiparesis Dysphagia Dysarthria Swollen Eye Pain Nausea Aconuresis Postoperative delirium Allergic reaction Irritation around extensions Nonspecific somatic symptoms

2 9 1 8 7 1 3 1 12 8 2 1 1 1 1 2

AC C

EP T

ED

MA

NU

RI

SC

1.28 5.77 0.64 5.13 4.49 0.64 1.92 0.64 7.69 5.13 1.28 0.64 0.64 0.64 0.64 1.28

PT

Adverse event

30

ACCEPTED MANUSCRIPT Table 4. Device-related adverse events (n=156) No. of AE’s % of AE’s

Pain Erosion Reinsertion/ Misplacement Lead fracture Extension break Contact malfunction Battery failure Hypertension System dislodged Vibrations around neurostimulator Spontaneous interruption of stimulation

4 2 1 1 3 1 3 1 2 3 1

AC C

EP T

ED

MA

NU

SC

RI

2.56 1.28 0.64 0.64 1.92 0.64 1.92 0.64 1.28 1.92 0.64

PT

Adverse event

31

ACCEPTED MANUSCRIPT

Table 5. Somatic adverse events (n=156) Adverse event

No. of AE’s % of AE’s

Pain Nausea/vomiting/diarrhea Headache Balance/ dizziness Syncope Sweating Paresthesia Oculomotor (blurred vision/ double vision) Dysphagia Dyskinesia Vision/eye movement disorder Weight gain Hand numbness Arm weakness Buzzing in ears Nocturia Pollakiuria Increased libido Change in taste Restlessness Internal unrest Intraocular pressure Memory problems

6 13 21 12 3 8 3 9 1 2 8 2 2 1 1 1 1 2 3 6 1 2 1

PT

RI SC

NU

MA

ED

EP T

AC C 32

3.85 8.33 13.46 7.69 1.92 5.13 1.92 5.77 0.64 1.28 5.13 1.28 1.28 0.64 0.64 0.64 0.64 1.28 1.92 3.85 0.64 1.28 0.64

ACCEPTED MANUSCRIPT

Table 6. Psychiatric adverse events (n=156) No. of AE’s

% of AE’s

Worsen depression Suicide completion Suicide ideation Suicide attempt Disinhibition Hypomania Mania Agitation Psychosis Mixed bipolar state Anxiety Sleep disturbances Irritability Hallucinations Flight of ideas

12 4 7 10 8 6 3 12 1 1 9 10 5 1 1

7.69 2.56 4.49 6.41 5.13 3.85 1.92 7.69 0.64 0.64 5.77 6.41 3.21 0.64 0.64

AC C

EP T

ED

MA

NU

SC

RI

PT

Adverse event

33

ACCEPTED MANUSCRIPT

Table 7.Other adverse events during stimulation (n=156) No. of AE’s

% of AE’s

Cancer Lung calcification Shortness of breath Musculoskeletal Chest pain Anemia Skin Infections Polyuria Circulation problems Pneumonia Hypothyroidism Gastritis Leg Fracture Herniated Disk

2 1 1 6 2 1 11 3 2 1 1 1 4 2 1

1.28 0.64 0.64 3.85 1.28 0.64 7.05 1.92 1.28 0.64 0.64 0.64 2.56 1.28 0.64

AC C

EP T

ED

MA

NU

SC

RI

PT

Adverse event

34

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

35

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

36

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

37

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

38

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

39

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

40

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

41

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

42

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

43

AC C

EP T

ED

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

44

ACCEPTED MANUSCRIPT Highlights

PT RI SC NU MA ED EP T



This is a systematic meta-analysis of DBS targeting all four reported brain regions in TRD Adverse effects of DBS during TRD therapy were statistically analyzed and reported in this work.

AC C



45