The lines between nodes represent direct comparisons in various trials, and the size of each circle is proportional to the size of the population involved in each specific treatment. The thickness of the lines is proportional to the number of trials connected to the network. a-tDCS indicates anodal transcranial direct current stimulation; cTBS, continuous theta-burst stimulation; c-tDCS, cathodal tDCS; dTMS, deep transcranial magnetic stimulation; F3, left dorsolateral prefrontal cortex (DLPFC); F4, right DLPFC; Fp2, right supraorbital area; HF, high-frequency; iTBS, intermittent TBS; LF, low-frequency; rTMS, repetitive TMS; T3, left auditory cortex; T4, right auditory cortex; and tRNS, transcranial random noise stimulation.
When the effect size is less than zero, it indicated the specified treatment was associated with higher improvement in severity of tinnitus than controls. a-tDCS indicates anodal transcranial direct current stimulation; cTBS, continuous theta-burst stimulation; c-tDCS, cathodal tDCS; dTMS, deep transcranial magnetic stimulation; F3, over the left dorsolateral prefrontal cortex (DLPFC); F4, over the right DLPFC; F3F4, over the bilateral DLPFC; Fp2, over the right supraorbital area; HF, high-frequency; iTBS, intermittent TBS; LF, low-frequency; rTMS, repetitive TMS; SMD, standardized mean difference; T3, over the left auditory cortex; T4, over the right auditory cortex; T3T4, over the bilateral auditory cortices; and tRNS, transcranial random noise stimulation.
eMethods. Study Identification and Data Analysis
eTable 1. PRISMA Extension Checklist of Current Meta-analysis
eTable 2. Excluded Studies and Reasons
eTable 3. Characteristics of the Included Studies
eTable 4. Node Definition
eTable 5. League Table of the Improvement of Severity of Tinnitus
eTable 6. SUCRA of Patient Improvement, Response Rate, and Dropout Rate
eTable 7. Meta-regression Results
eTable 8. League Tables of Improvement of Quality of Life, Response Rate, and Dropout Rate
eTable 9. Inconsistency of Different Intervention
eTable 10. Estimated Between-Studies Standard Deviation of Different Outcomes
eFigure 1. Network Structure of NMA of Changes of Quality of Life, Response Rate, and Dropout Rate
eFigure 2. Forest Plots of NMA Outcomes
eFigure 3. Overview of and Detailed Risk of Bias
eFigure 4. Funnel Plots of Outcomes, Response Rate, and Dropout Rate
Customize your JAMA Network experience by selecting one or more topics from the list below.
Chen J, Zeng B, Wu C, et al. Association of Central Noninvasive Brain Stimulation Interventions With Efficacy and Safety in Tinnitus Management: A Meta-analysis. JAMA Otolaryngol Head Neck Surg. 2020;146(9):801–809. doi:10.1001/jamaoto.2020.1497
Which noninvasive brain stimulation treatment was associated with the best efficacy and acceptability in tinnitus management?
In this meta-analysis of 32 unique studies including 1458 unique participants, the cathodal transcranial direct current stimulation over the left dorsolateral prefrontal cortex combined with transcranial random noise stimulation over the bilateral auditory cortex was associated with the greatest improvement in both tinnitus severity and quality of life. Continuous theta-burst stimulation over both auditory cortices ranked more favorably than that over the left auditory cortex only.
Regarding the efficacy and acceptability for tinnitus treatment, these findings suggest that the cathodal transcranial direct current stimulation over the left dorsolateral prefrontal cortex combined with transcranial random noise stimulation over the bilateral auditory cortex is preferable.
Tinnitus has a prevalence of 10% to 25% and is frequently associated with numerous complications, such as neuropsychiatric disease. Traditional treatments have failed to meet the needs of patients with tinnitus. Noninvasive brain stimulation (NIBS) can focally modify cortical functioning and has been proposed as a strategy for reducing tinnitus severity. However, the results have been inconclusive.
To evaluate the association between different central NIBS therapies and efficacy and acceptability for treatment of tinnitus.
ClinicalKey, Cochrane CENTRAL, Embase, ProQuest, PubMed, ScienceDirect, and Web of Science databases were searched from inception to August 4, 2019. No language restriction was applied. Manual searches were performed for potentially eligible articles selected from the reference lists of review articles and pairwise meta-analyses.
Randomized clinical trials (RCTs) examining the central NIBS method used in patients with unilateral or bilateral tinnitus were included in the current network meta-analysis. The central NIBS method was compared with sham, waiting list, or active controls. Studies that were not clinical trials or RCTs and did not report the outcome of interest were excluded.
Data Extraction and Synthesis
Two authors independently screened the studies, extracted the relevant information, and evaluated the risk of bias in the included studies. In cases of discrepancy, a third author became involved. If manuscript data were not available, the corresponding authors or coauthors were approached to obtain the original data. This network meta-analysis was based on the frequentist model.
Main Outcomes and Measures
The primary outcome was change in the severity of tinnitus. Secondary outcomes were changes in quality of life and the response rate related to the NIBS method in patients with tinnitus.
Overall, 32 unique RCTs were included with 1458 unique participants (mean female proportion, 34.4% [range, 0%-81.2%]; mean age, 49.6 [range, 40.0-62.8] years; median age, 49.8 [interquartile range, 48.1-52.4] years). The results of the network meta-analysis revealed that cathodal transcranial direct current stimulation over the left dorsolateral prefrontal cortex combined with transcranial random noise stimulation over the bilateral auditory cortex was associated with the greatest improvement in tinnitus severity (standardized mean difference [SMD], –1.89; 95% CI, –3.00 to –0.78) and quality of life (SMD, –1.24; 95% CI, –2.02 to –0.45) compared with the controls. Improvement in tinnitus severity ranked more favorably for continuous theta-burst stimulation (cTBS) over both auditory cortices (SMD, −0.79; 95% CI = −1.57 to −0.01) than cTBS over only the left auditory cortex (SMD, −0.30; 95% CI, −0.87 to 0.28), compared with controls. Repetitive transcranial magnetic stimulation with priming had a superior beneficial association with tinnitus severity compared with the strategies without priming. None of the investigated NIBS types had a significantly different dropout rate compared with that of the control group.
Conclusions and Relevance
This network meta-analysis suggests a potential role of NIBS interventions in tinnitus management. Future large-scale RCTs focusing on longer follow-up and different priming procedure NIBS are warranted to confirm these findings.
Quiz Ref IDIn the adult population, tinnitus has a prevalence of 10% to 25%,1,2 and 6% to 25% of these persons suggest that these symptoms are severely debilitating and have an adverse effect on quality of life.3 Tinnitus is recognized as a difficult disease to identify and manage because of controversy regarding its definition and treatment. Tinnitus usually results from heterogeneous causes: (1) the auditory system (usually peripheral, rarely central), (2) the somatosensory system (head and neck), or (3) a combination of these. The condition results when the interactivity of the auditory and somatosensory systems exceeds the individual’s tinnitus threshold. Treatments to reduce tinnitus severity and tinnitus-related distress include cognitive behavioral therapy, acoustic stimulation, and educational counseling. Although somatic treatments can be effective in cases of tinnitus with a specific origin (eg, palatal myoclonus, deafferentation of the auditory system, loss of cochlear hair cells, and ototoxic drugs), no specific interventions have been proven to be effective in treating tinnitus without a specific cause.2
With the help of functional brain imaging studies, abnormal hyperactivity has been detected in the whole brain area, especially in both auditory cortices,4 the anterior cingulate cortex,5 and the insula.5Quiz Ref ID Therefore, suppression of abnormal brain hyperactivity through central noninvasive brain stimulation (NIBS) has been proposed as a tinnitus management strategy.6 The central NIBS techniques include repetitive transcranial magnetic stimulation (rTMS),7 the rTMS variant theta-burst stimulation (TBS),8,9 and transcranial electrical stimulation, such as transcranial direct current stimulation (tDCS)10 and transcranial random noise stimulation (tRNS).11 Based on the frequency applied, rTMS can induce different changes in brain activity in patients with tinnitus. For example, high-frequency rTMS induces higher brain activity, whereas low-frequency rTMS suppresses hyperactivity in the cerebral cortex.12 Transcranial DCS, in which a weak direct electrical current is passed through the brain cortex, has a similar suppressing or enhancing effect on brain activity. Similarly, tDCS has been found to have a modulating effect on the stimulated brain cortex, although the suppressing or enhancing effect due to the polarity of tDCS is still debated.8,13 The suppression of hyperactivity in the temporoparietal cortex (ie, the primary or secondary auditory cortex) through the use of rTMS or tDCS was considered a reasonable NIBS method in the management of tinnitus.14
Pairwise meta-analyses have indicated significant efficacy of tDCS15 and rTMS16 vs sham treatment in improving the severity of tinnitus. However, these meta-analyses were based on only 2 to 4 randomized clinical trials (RCTs), resulting in poor evidence of efficacy. In addition, different combination protocols of NIBS interventions have been developed for tinnitus management, including different brain stimulation targets and stimulation parameters. Specifically, additive stimulation over the prefrontal cortex (often dorsolateral prefrontal cortex [DLPFC]), also termed the priming procedure, has been reported to have an enhancing effect on tinnitus management.7,17 However, the most recent meta-analyses have not provided conclusive results relevant to clinical practice.15,16,18 In addition, traditional pairwise meta-analyses cannot provide further information about the relative efficacy of interventions that have not been directly compared in head-to-head trials, which is an essential aspect when judging the therapeutic value of an intervention. Considering these issues, we conducted a network meta-analysis of the currently published RCTs to estimate the association between different central NIBS interventions and relative efficacy and acceptability in patients with tinnitus.19
The present study did not receive any ethics committee approval or informed consent from the participants because we did not approach any specific participants or report any detailed information of specific participants. Our previous project was approved by the institutional review board of the Tri-Service General Hospital, National Defense Medical Center. This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline (eTable 1 in the Supplement).
Detailed information regarding the methods and materials is presented in the eMethods in the Supplement. We searched the ClinicalKey, Cochrane CENTRAL, Embase, ProQuest, PubMed, ScienceDirect, and Web of Science databases from inception to August 4, 2019. No language restriction was applied. Manual searches were performed for potentially eligible articles selected from the reference lists of review articles and pairwise meta-analyses. We included RCTs with sham-controlled, waiting list–controlled, or active-controlled design conducted in patients with tinnitus. The detailed categorization of the targets of comparison arms were listed in the node definition section of the eMethods in the Supplement.
Following the flowchart used in previous network meta-analyses,20-26 we extracted the relevant information from the RCTs and evaluated the risk of bias in the included studies. Two authors (J.J.C. and B.S.Z.) independently screened the studies, extracted the relevant information from the manuscripts, and evaluated the risk of bias in the included studies. In cases of discrepancy, a third author (P.T.T.) became involved. If manuscript data were not available, the corresponding authors or coauthors were approached to obtain the original data. We only extracted data on central NIBS and not peripheral stimulation.
The primary outcome was change in the severity of tinnitus after NIBS in patients with tinnitus, which could be rated using a different tinnitus questionnaire (described in the Results section) (outcomes in eMethods in the Supplement). The secondary outcomes were change in quality of life and response rate related to the NIBS method in patients with tinnitus. The detailed definition of response rate and quality of life had been presented in the outcomes in eMethods in the Supplement. Finally, the safety profile was calculated using the dropout rate, which was defined as the percentage of patients leaving the study before its conclusion for any reason.
The risk of bias was evaluated according to the Cochrane risk-of-bias tool.27 The current network meta-analysis was conducted under the frequentist model and generalized linear mixed models to make direct and indirect comparisons.28 In our analysis, the mvmeta command was applied in the STATA program, version 14.0.29 We estimated the standardized mean difference (SMD) with 95% CI for continuous variables (ie, the primary outcome of tinnitus severity and the secondary outcome of quality of life). We evaluated categorical values with the rate ratio and 95% CI (ie, the secondary outcome of response and safety of dropout) and applied a 0.5 zero-cell correction during the meta-analysis procedure. Heterogeneity among the included studies was evaluated using the τ value, which is the estimated SD of the association across the included studies.
To provide additional information for clinical applications, we calculated the surface under the cumulative ranking curve (SUCRA), which indicates the relative ranking probabilities of the treatment effects for the target outcomes.30 We conducted meta-regression to determine the associations between change in tinnitus severity and participant characteristics, such as mean age and the sex distribution. Finally, we evaluated the potential inconsistencies between the direct and indirect evidence within the network by using the loop-specific approach and identified local inconsistencies by using the node-splitting method. The design-by-treatment model was used to evaluate global inconsistencies across the entire network meta-analysis.31
After the initial screening procedure, 104 articles were considered for full-text review (Figure 1). However, 72 were excluded for various reasons (eTable 2 in the Supplement). Finally, 32 articles7-11,17,32-57 were included in the current study (eTable 3 in the Supplement). Figure 2 depicts the entire geometric distribution of the treatment arms. The detailed categorization of the treatment arms is provided in eTable 4 in the Supplement.
A total of 1458 participants were included. The mean age of the participants was 49.6 years (range, 40.0-62.8 years; median, 49.8 [interquartile range (IQR), 48.1-52.4] years); the mean female proportion was 34.4% (range, 0%-81.2%; median, 30.3% [IQR, 24.7%-40.9%]), and the mean male proportion was 65.6% (range, 18.8%-100%; median, 69.7% [IQR, 59.1%-75.3%]). The mean duration of central NIBS treatment was 16.9 weeks (range, 2-54 weeks; median, 12 [IQR, 4-24] weeks). The baseline characteristics of the included participants are summarized in eTable 3 in the Supplement. The definition of response varied among the recruited studies as patient self-rated global impression, Tinnitus Handicap Inventory score reduction of greater than 7 or 10 points, Tinnitus Handicap Inventory score reduction of greater than 20% of the baseline score, and Tinnitus Questionnaire score reduction of greater than 5 or 10 points.
The network meta-analysis revealed that only the cathodal tDCS over the left DLPFC (F3) plus anodal tDCS over the right DLPFC (F4) plus tRNS of the left auditory cortex (T3) combination (SMD, −1.89 [95% CI, −3.00 to −0.78]), continuous TBS (cTBS) over the bilateral auditory cortices (T3T4) (SMD, −0.79 [95% CI, −1.57 to −0.01]), and high-frequency rTMS-F3 plus low-frequency rTMS-T3T4 combination (SMD, −0.70 [95% CI, −1.38 to −0.02]) were associated with significant improvement in the severity of tinnitus compared with the control (eTable 5 in the Supplement and Figure 3). The associations between an NIBS method and the change in tinnitus severity were ranked according to the SUCRA, where lower values indicate superior outcomes of tinnitus severity. In brief, the combination of cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3 was associated with the largest improvement, followed by deep TMS bilateral medial frontal cortex plus low-frequency rTMS-T3 and cTBS-T3T4. In addition, rTMS with a priming procedure (ie, rTMS over the frontal lobe followed by rTMS over the auditory cortex, such as high-frequency rTMS-F3 plus low-frequency rTMS-T3T4 (SUCRA, 36.6), high-frequency rTMS-F3 plus low-frequency rTMS-T3 (SUCRA, 38.8), low-frequency rTMS-F3 plus low-frequency rTMS-T3 (SUCRA, 48.3), and low-frequency rTMS-F4 plus low-frequency rTMS-T3 (SUCRA, 38.1) was ranked more highly than rTMS without priming (ie, rTMS over the auditory cortex alone, such as low-frequency rTMS-T3 (SUCRA, 62.9), low-frequency rTMS-T3T4 (SUCRA, 54.0), and high-frequency rTMS-T3T4; SUCRA, 56.5). Finally, bilateral cTBS (ie, cTBS-T3T4; SUCRA, 32.7) was ranked more highly than unilateral cTBS (ie, cTBS-T3; SUCRA, 64.1) and intermittent TBS (iTBS) (ie, iTBS-T3; SUCRA, 53.4) (eTable 6A in the Supplement). A meta-regression using the restricted maximum likelihood estimator was performed to examine the potential association of age and sex with the change in tinnitus severity. The results reveal a nonsignificant association with the change in tinnitus severity when using the moderating variables age and sex (eTable 7A in the Supplement).
The network meta-analysis revealed that the cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3 combination (SMD, −1.24 [95% CI, −2.02 to −0.45]), low-frequency rTMS-T3T4 (SMD, −0.52 [95% CI, −0.83 to −0.20]), and high-frequency rTMS-T3 (SMD, −0.49 [95% CI, −0.93 to −0.04]) were associated with significant improvements in the quality of life of patients with tinnitus in comparison to the sham control (eTable 8A, eFigure 1A, and eFigure 2A in the Supplement). The associations between NIBS method and change in quality of life were ranked according to the SUCRA. In brief, the combination cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3 was associated with the largest improvement in quality of life (SUCRA, 5.0), followed by cTBS-T3 (SUCRA, 28.8) and low-frequency rTMS-T3T4 (SUCRA, 33.4) (eTable 6B in the Supplement). The results of this meta-regression revealed a nonsignificant association with change in quality of life when using the moderating variables age and sex (eTable 7B in the Supplement).
The network meta-analysis revealed that none of the investigated NIBS methods were associated with significantly better response rates than the sham control (eTable 8B, eFigure 1B, and eFigure 2B in the Supplement). The associations between the NIBS methods and change in quality of life were ranked according to the SUCRA. In brief, high-frequency rTMS-T3 was associated with the highest response rate (SUCRA, 27.5) (eTable 6C in the Supplement). The results of this meta-regression revealed a nonsignificant association with the response rate when using the moderating variables age and sex (eTable 7C in the Supplement).
In the network meta-analysis, none of the investigated NIBSs were associated with significantly different dropout rates when compared with the sham control (eTable 6D, eTable 8C, eFigure 1C, and eFigure 2C in the Supplement). The results of this meta-regression reveal a nonsignificant association with the dropout rate when using the moderating variables age and sex (eTable 7D in the Supplement).
Among the included studies, we found that 134 of 224 items (59.8%) had a low risk of bias; 69 of 224 items (30.8%), an unclear risk of bias; and 21 of 224 items (9.4%), a high risk of bias. Unclear reporting of the allocation procedure or blinding of the studies further contributed to the risk of bias (eFigure 3A-B in the Supplement).
Funnel plots of the publication bias (eFigure 4A-H in the Supplement) revealed general symmetry, and the results of the Egger test indicated no significant publication bias among the articles included in the network meta-analysis. In general, the analysis did not demonstrate inconsistencies in terms of local inconsistencies, as assessed using the loop-specific approach and node-splitting method, or global inconsistencies, as determined using the design-by-treatment method except for the situation mentioned below. Overall inconsistencies were detected in the outcomes of severity of tinnitus and quality of life (eTables 9 and 10 in the Supplement).
To our knowledge, the current study is the first comprehensive network meta-analysis performed to investigate the association between central NIBS interventions and efficacy and acceptability in patients with tinnitus. Quiz Ref IDEvidence from this network meta-analysis revealed that cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3 was associated with significantly greater improvement in both tinnitus severity and quality of life than the sham control and the largest improvement in both tinnitus severity and quality of life. In addition, cTBS-T3T4 was ranked more highly than cTBS-T3 or iTBS. Repetitive TMS with a priming procedure may be better at improving the severity of tinnitus. Noninvasive brain stimulation using high-frequency rTMS as the priming intervention (ie, high-frequency rTMS-F3 plus low-frequency rTMS-T3T4) was associated with significantly greater improvement in tinnitus severity than the sham control. Finally, most of the investigated NIBS methods were suggested by the dropout rate to be well tolerated.
The first main finding of this study was that the cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3 combination was associated with significantly greater improvement in both tinnitus severity and quality of life than the sham control and was also associated with the greatest improvement in both the severity of tinnitus and quality of life. The cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3 combination involved cathodal tDCS-F3 and tRNS-T3. Transcranial RNS is a modification of transcranial alternating current stimulation with random oscillations (ranging from 0.1-640.0 Hz).58 In 1 RCT,10 the effect of cathodal tDCS-F3 on tinnitus intensity was demonstrated. However, in a head-to-head trial,59 tRNS was proven to be superior to tDCS in suppressing tinnitus intensity and decreasing distress after a single session. Furthermore, the findings of the current network meta-analysis supported the superiority of continuous sessions of a combination of cathodal tDCS-F3, anodal tDCS-F4, and tRNS-T3 for tinnitus intensity. Based on the findings of hyperactivity detected in both auditory cortices,4 the anterior cingulate cortex,5 and the insula5 in functional brain imaging studies (ie, magnetoencephalography, functional magnetic resonance imaging, or brain positron emission tomography) for patients with tinnitus, the rationale of a combination of stimulation types over these sites is a reasonable strategy.11 Furthermore, the hypothesis of the preconditioning phenomenon can be a valid explanation of the importance of sequence of stimulation.60 According to this hypothesis, the beneficial effect of the second stimulation targeting another region (ie, the auditory cortex) of the tinnitus network is enhanced by the priming stimulation (ie, over the frontal region).61 However, only 1 RCT with waiting list controls has reported the additive effect of tRNS in patients with tinnitus recruited for this network meta-analysis.11 Although significant inconsistency was not detected within the comparison of treatment arms of cathodal tDCS-F3 plus anodal tDCS-F4, cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3, and controls, which were treatment arms applied in that RCT,11 according to the side-splitting inconsistency model (side-splitting inconsistency model in eTable 7 in the Supplement), the clinicians should pay special attention when applying this result in their clinical practice. Further large-scale RCTs are required to support or refute the results of this study.
Another remarkable finding of the present network meta-analysis was that cTBS-T3T4 resulted in superior outcomes to cTBS-T3 or iTBS-T3 only. Moreover, the potential benefit of cTBS regarding severity of tinnitus has been proven in many trials, both in a single session62,63 and multiple sessions.8,52 Continuous TBS, which is reported to better suppress both pure-tone tinnitus and white noise than tonic TMS,63 had a more powerful effect on tinnitus relief than high-frequency rTMS.8 In addition, according to reports, cTBS was suggested to be able to modulate both the extralemniscal and lemniscal systems, the systems that mainly manage the sensory input to the central nervous system, whereas tonic TMS modulates only the lemniscal system.63,64 Furthermore, another RCT8 demonstrated that bilateral cTBS was more effective than unilateral stimulation. However, because bilateral cTBS (cTBS-T3T4) in patients with tinnitus was reported only in 2 RCTs included in the current network meta-analysis,8,52 future large-scale RCTs are required to support or refute this study’s results.
Finally, the present network meta-analysis identified rTMS with priming procedure as more beneficial to tinnitus severity than such stimulation without priming. Specifically, rTMS in combination with stimulation over the frontal lobe and then over the auditory cortex was superior compared with rTMS over the auditory cortex only, either in high frequency or low frequency. Furthermore, NIBS using high-frequency rTMS as the priming intervention (ie, high-frequency rTMS-F3 plus low-frequency rTMS-T3T4) was associated with significantly greater improvement in tinnitus severity than in the sham control. These findings correspond with the results of clinical brain imaging studies in which patients with tinnitus were demonstrated to have hyperactivity in multiple brain regions, including both the auditory cortex and DLPFC.4,5 In addition, the hypothesis of the preconditioning phenomenon supports multiple site interventions applied in a sequence.60 This hypothesis of the priming effect is also supported by previous clinical reports demonstrating a superior prolonged beneficial effect on tinnitus severity when a priming rTMS protocol was used in such patients for long-term follow-up compared with rTMS without priming, especially for stimulation at higher frequency.7,45,65,66 Therefore, the present network meta-analysis result can be considered as further essential evidence of the preconditioning phenomenon hypothesis. However, because of limited RCTs addressing the potential benefit of a priming procedure, future large-scale RCTs are required to support or refute the results of the current network meta-analysis.
Several potential limitations should be considered for this network meta-analysis. Quiz Ref IDFirst, this analysis may have been underpowered owing to the heterogeneity of the participants (eg, comorbidities, mood disorder, baseline severity of tinnitus, history of tinnitus onset, commercial machine used in each study, and follow-up duration), variety in the definition of response, and variety in tinnitus severity or quality-of-life rating scales. Although meta-regression analyses were performed to reduce the heterogeneity, some differences did exist between the included RCTs, which were attributed to other unknown factors. Second, although most of the RCTs included a sham control in their study design, the blindness of those RCTs may not have been complete because of the limitation of the commercial machine used. Third, given the relatively small number of patients and RCTs, the main results of this network meta-analysis should perhaps be conservatively applied in clinical practice. Specifically, the potential effect of additive tRNS, the priming procedure, and bilateral cTBS should be carefully interpreted because only a few RCTs reported the results of these NIBS methods (1 trial for additive tRNS,11 7 trials for priming procedure,7,17,40,41,44,45,47 and 2 trials for bilateral cTBS8,52). In addition, the relatively small number of patients and RCTs would limit the potential benefit of NIBS interventions in some outcomes. For example, although most of the NIBS interventions were associated with relatively better response than the sham or control group, the variation and CIs ranged widely, which would result in an insignificant outcome (eFigure 2B in the Supplement). Future larger-scale RCTs are warranted to support or refute the result of the present network meta-analysis. Quiz Ref IDFinally, we detected significant inconsistency in some of the outcomes (ie, severity of tinnitus and quality of life). Clinicians should pay attention when applying these results in their clinical practice.
This study showed that the cathodal tDCS-F3 plus anodal tDCS-F4 plus tRNS-T3 combination was associated with the greatest improvement in tinnitus severity and quality of life. A specific central NIBS protocol (ie, bilateral cTBS and priming with high-frequency rTMS or tDCS) was also associated with superior improvement in tinnitus severity. All central NIBS methods had similar tolerability in terms of the dropout rate compared with the sham control. However, because some of the intervention comparisons were based on only a few RCTs, clinicians should select specific treatments with caution and avoid the one-size-fits-all treatment for all clinical conditions.
Accepted for Publication: May 5, 2020.
Corresponding Authors: Ping-Tao Tseng, MD, Prospect Clinic for Otorhinolaryngology and Neurology, No. 252, Nanzixin Road, Nanzi District, Kaohsiung City 81166, Taiwan (email@example.com), and Cheng-Ta Li, MD, PhD, Division of Community and Rehabilitation Psychiatry, Department of Psychiatry, Taipei Veterans General Hospital, Taiwan, No. 201, Section 2, Shipai Road, Beitou District, Taipei City 11267, Taiwan (firstname.lastname@example.org).
Published Online: July 9, 2020. doi:10.1001/jamaoto.2020.1497
Author Contributions: Drs J.-J. Chen and Zeng contributed equally as co–first authors, Drs Y-W Chen and Tseng had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: J.-J. Chen, Zeng, C.-N. Wu, Su, Tu, T.-Y. Chen, Liang, C.-W. Hsu, S.-P. Hsu, Kuo, Y.-W. Chen, Tseng.
Acquisition, analysis, or interpretation of data: J.-J. Chen, Zeng, Stubbs, Carvalho, Brunoni, Tu, Y.-C. Wu, T.-Y. Chen, Lin, Liang, Tseng, Li.
Drafting of the manuscript: J.-J. Chen, Zeng, Stubbs, Carvalho, Tu.
Critical revision of the manuscript for important intellectual content: C.-N. Wu, Stubbs, Carvalho, Brunoni, Su, Tu, Y.-C. Wu, T.-Y. Chen, Lin, Liang, C.-W. Hsu, S.-P. Hsu, Kuo, Y.-W. Chen, Tseng, Li.
Statistical analysis: J.-J. Chen, Carvalho, Tu, Liang, Y.-W. Chen, Tseng, Li.
Administrative, technical, or material support: J.-J. Chen, Zeng, Stubbs, T.-Y. Chen, Kuo, Tseng, Li.
Supervision: Stubbs, Brunoni, Su, Tu, Y.-C. Wu, Tseng, Li.
Conflict of Interest Disclosures: Dr Brunoni reported receiving academic support in the form of financial fees, outside the submitted work, from the University of Sao Paulo Medical School and the National Council for Scientific and Technological Development, and being a medical adviser for Flow Neuroscience with a small equity of the company (he receives no personal fees from the company). No other disclosures were reported.
Funding/Support: This study was supported by Clinical Lectureship ICA-CL-2017-03-001 jointly funded by Health Education England and the National Institute for Health Research (NIHR) (Dr Stubbs); the NIHR Biomedical Research Centre at South London and Maudsley NHS (National Health Service) Foundation Trust (Dr Stubbs); Maudsley Charity, King’s College London and the NIHR South London Collaboration for Leadership in Applied Health Research and Care funding (Dr Stubbs); grants MOST 106-2314-B-039-027-MY3, 107-2314-B-039-005, 108-2320-B-039-048, and 108-2314-B-039-016 from the Ministry of Science and Technology, Taiwan (Dr Su); grants DMR-107-091, DRM-108-091, CRS-108-048, CMU108-SR-106, DMR-108-216, CMRC-CMA-3, and DMR-109-102 from the Chinese Medicine Research Center from the China Medical University, Taiwan (Dr Su); grants MOST 106-2314-B-182A-085-MY2 and MOST 105-2314-B-182A-057 from the Ministry of Science and Technology, Taiwan (Dr Lin); grants CMRPG8F1371 and CMRPG8E1061F from Kaohsiung Chang Gung Memorial Hospital, Taiwan (Dr Lin); grant 106-2314-B-002-098-MY3 from the Ministry of Science and Technology, Taiwan (Dr Tu); grant MOST 108-2321-B-075-004-MY2) from the Ministry of Science and Technology; and grant 108BRC-B502 from the Brain Research Center within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan (National Yang-Ming University from The Featured Areas Research Center Program).
Role of the Funder/Sponsor: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed in this publication are those of the authors and not necessarily those of the acknowledged institutions.
Additional Contributions: This article was edited by Wallace Academic Editing. The work was not funded by any source of funding.