Efficacy of Repetitive Transcranial Magnetic Stimulation (rTMS) Combined with Psychological Interventions: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

(1) Background: Psychological interventions are effective in alleviating neuropsychiatric symptoms, though results can vary between patients. Repetitive transcranial magnetic stimulation (rTMS) has been proven to improve clinical symptoms and cognition. It remains unclear whether rTMS can augment the efficacy of psychological interventions. (2) Methods: We examined the effects of rTMS combined with psychological interventions on clinical, functional, and cognitive outcomes from randomized controlled trials conducted in healthy and clinical populations. We searched PubMed, EMBASE, Cochrane Library, and PsycINFO databases up to April 2023. (3) Results: Twenty-seven studies were ultimately included. Compared to sham rTMS combined with psychological interventions, active rTMS combined with psychological interventions significantly improved overall clinical symptoms (k = 16, SMD  =  0.31, CIs 0.08 to 0.54, p  <  0.01). We found that 10 or more sessions of rTMS combined with cognitive behavioural therapy significantly improved clinical outcomes overall (k = 3, SMD  =  0.21, CIs  0.05 to 0.36, Z = 2.49, p  <  0.01). RTMS combined with cognitive training (CT) significantly improved cognition overall compared to sham rTMS combined with CT (k = 13, SMD  =  0.28, CIs 0.15 to 0.42, p  <  0.01), with a significant effect on global cognition (k = 11, SMD  =  0.45, CIs 0.21 to 0.68, p  < 0.01), but not on the other cognitive domains. (4) Conclusion: The current results provide preliminary support for the augmentation effects of active rTMS on clinical and cognitive outcomes across diverse populations. Future clinical trials are required to confirm these augmentation effects for specific psychological interventions in specific clinical populations.


Introduction
Psychological interventions, including cognitive behavioural therapy (CBT), mindfulnessbased cognitive therapy, cognitive remediation, etc., aim to promote one's ability to adapt to a given situation, leading to improved functioning [1].Despite the increasing clinical use of psychological interventions, their efficacy can be limited.A systematic review of 419 randomized controlled trials (RCTs) from clinical and non-clinical populations found that psychological interventions produced only small-to-moderate effect sizes on mental well-being and/or indicators of illness [2].Efficacy is often also lower in patients with more severe symptoms and in those who have failed other therapies [3].The long duration (e.g., 3-12 months) required for a psychological intervention to take effect [4] is a further consideration.For these reasons, the combination of psychological therapies with biological therapies (e.g., pharmacotherapy) is often recommended in clinical practice guidelines [5,6].While the efficacy of the combination of different psychological interventions with pharmacotherapies has been commonly studied [7,8], combining psychological Brain Sci.2023, 13, 1665 2 of 17 therapies with non-invasive brain stimulation (NIBS) treatments is a relatively new area of research interest with the potential to improve therapeutic outcomes.
Transcranial magnetic stimulation (TMS) is a form of NIBS that has been extensively utilized for research and clinical applications [9,10].TMS exerts its effects by delivering magnetic pulses that go through the scalp and skull, which induce electrical currents that can depolarize neurons in the brain [11,12].Repetitive transcranial magnetic stimulation (rTMS), which delivers multiple repeated magnetic pulses, causes long-lasting effects on cortical excitability beyond the stimulation period [13].RTMS can also improve cognitive functions in various domains in both healthy [14] and clinical populations [15][16][17].RTMS parameters can be adjusted to have different effects: high frequency (≥5 Hz) rTMS and intermittent theta-burst stimulation (iTBS) facilitate neural activity, while low frequency (≤1 Hz) rTMS and continuous theta-burst stimulation (cTBS) inhibit it [9,18].Theta burst stimulation (TBS), a newer form of rTMS, involves delivering patterned stimulation to the brain at specific frequencies (typically at a theta frequency with gamma frequency "bursts") and has a shorter treatment duration than standard rTMS [19].Evidence from several meta-analyses has shown that compared to sham stimulation, active rTMS can alleviate symptoms of depression [20] and improve clinical outcomes in other neuropsychiatric illnesses [21][22][23][24][25][26].Research has shown that, typically, at least 20-30 rTMS sessions administered over consecutive weekdays are necessary for optimal clinical effects [27].
In vivo and in vitro studies have shown that rTMS can affect neurotransmission, influencing adaptive synaptic plasticity and producing long-lasting effects [24].It can also prolong the time window for the neural interactions subserving behavioural adaptation [28].Psychological interventions such as CBT and cognitive training (CT) can also facilitate synaptic plasticity [29].Based on the above, it is possible that rTMS could potentially enhance the effectiveness of psychological interventions by maximizing plasticity.Neuroplasticity is considered to be a critical treatment target for clinical improvement and enhancing cognitive and functional abilities in various neuropsychiatric diseases [30,31].Existing studies that have paired rTMS with psychological interventions, including exposure therapies [32], CT [33], and CBT [34] have provided preliminary evidence supporting the augmentation effect of rTMS on clinical outcomes across a broad range of neuropsychiatric disorders.While meta-analyses of RCTs of rTMS alone or psychological interventions alone for individual disorders are available, there is currently no systematic examination specifically focusing on the use of rTMS to augment different types of psychological interventions.
In the current study, we aimed to investigate the potential augmentation effects of rTMS on psychological interventions in healthy and clinical populations for different outcomes (i.e., clinical, functional, and cognitive outcomes) to identify common trends and patterns.A secondary aim was to explore which type of psychological interventions may exhibit greater augmentation effects when combined with rTMS.To address these aims, we conducted a systematic review and meta-analysis of RCTs of active rTMS combined with psychological interventions compared to sham rTMS combined with psychological interventions.

Materials and Methods
This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting of Systematic Reviews and Meta-Analyses (PRISMA) statement and was registered on the International Prospective Register of Systematic Reviews, PROS-PERO [35].The PRISMA checklist for this systematic review and meta-analysis is provided in the Supplementary Materials [36].(3) reporting pre-intervention and post-intervention assessment of clinical outcomes (i.e., symptom severity, functioning, and quality of life) or cognitive outcomes; (4) RCTs published in peer-reviewed English journals.After removing duplicates in the original search results, the titles and abstracts were screened independently against the selection criteria by two authors.The full texts were then reviewed and eligible studies were included in the meta-analysis.Any disagreements were resolved by discussion until a consensus was reached or by consulting a third author.

Data Extraction and Outcome Measures
All data were extracted by one author and checked by another author.We also contacted the study authors to request data related to the meta-analysis that was not accessible from the original publications.The following data were extracted: author, year of publication, characteristics of the study population (health condition, sample sizes, age, and gender), study design, rTMS parameters (the number of sessions, stimulation sites, frequency, intensity, total pulses, and timing of stimulation), type of psychological interventions, and pre-intervention and post-intervention outcomes.The outcome measures of interest assessed clinical symptoms and cognitive functions, the former including disease symptom severity (Supplementary Table S2) and functional outcomes (Supplementary Table S3) measured by standardised questionnaires, and the latter from standardised cognitive tasks or questionnaires (Supplementary Table S4).The means, standard deviations (SD), and sample sizes for outcome measures in each group were extracted for the pooled analysis.We also used a Web-based program called WebPlotDigitizer (WebPlotDigitizer, Austin, TX, USA, A. Rohatgi, 2018) to estimate data from figures.If the standard error (SE) was reported, we calculated SD using the equation SD = SE × √ N.

Statistical Analysis
Statistical analysis was conducted using R version 4.2.1 (R Core Team, 2022 [37]), RStudio software version 2022.12.0.353Posit Team, 2022 [38]) and 'meta', 'metafor' packages.The effect of active or sham rTMS combined with psychological interventions was examined using standardized mean differences (SMD) with 95% confidence intervals (CIs) for each outcome measure as part of a random-effects model.Meta-analyses were conducted when outcomes were available from at least three studies.There were three outcome measures of interest: clinical symptoms, functional outcomes, and cognition.For outcomes where higher scores were associated with poorer performance or more severe disease functional outcomes, these scores were recoded to represent positive effect sizes in favour of the active condition.Cognitive tasks and questionnaires were categorized according to six cognitive domains as defined in the DSM-5, including perceptual-motor function, language, learning and memory, complex attention, social cognition, and executive function [39].Executive function included tasks which involved updating, shifting, or inhibition [40].Furthermore, we included two additional cognitive domains: global cognition and working memory.Global cognition was added because a global cognitive score rather than scores within cognitive domains was reported in some of the studies which were included.Similarly, working memory was analysed because it was a commonly assessed cognitive outcome.If multiple cognitive outcomes from the same cognitive task or questionnaire were reported in an individual study, the most commonly used outcome measure as defined by the authors was selected [41] (Supplementary Table S4).If the primary outcome measure was not specified for a particular task, we included the most relevant measure based on our predefined cognitive domains.When studies assessed multiple cognitive tasks within the same domain, outcomes from different tasks were averaged to generate domain-specific aggregate effect sizes [14].Where multiple questionnaires measured the symptom severity for the Brain Sci.2023, 13, 1665 4 of 17 same disease in each study, the primary outcome measure as defined by the original authors was extracted and analysed [41].Heterogeneity between studies was assessed using the I 2 test.I 2 values > 25%, 50%, and 75% indicate low, moderate, and high heterogeneity, respectively [42].Secondary subgroup analyses were conducted according to different types of psychological interventions (i.e., rTMS + CBT, rTMS + CT, rTMS + exposure therapy, and rTMS + Mindfulness-Based Stress Reduction (MBSR)), study populations (healthy and clinical populations), and cognitive domains where sufficient data was available for meta-analyses.Given that patients exhibit faster improvement or recovery when subjected to a higher number of psychotherapy sessions [43], we additionally performed subgroup analyses on studies involving 10 or more sessions of combined interventions to examine the effects on clinical symptoms, functional outcomes, and cognition.The revised Cochrane risk of bias assessment tool for RCTs (Risk of Bias tool, RoB 2) was used to independently assess the quality of included studies by two authors.All discrepancies were resolved by consulting a third author.The tool was used to assess bias across five domains: the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result [44].Each study was judged as having a low risk of bias, some concerns, or a high risk of bias.Publication bias was assessed using funnel plots and Egger's test [45] for the outcomes in which 10 or more available studies were included.

Overview
A total of 3969 articles were identified initially, and 27 studies were ultimately included in the meta-analysis (see Figure 1).The characteristics of the included studies are summarized in Table 1.
All studies utilized a between-subject design, other than studies from Gy et al. [46] and Osuch et al. [47], which used a within-subject design.In the study conducted by Gy et al. [46], participants were randomized to receive 30 sessions of active or sham rTMS + cognitive stimulation in the first phase and then crossed over to receive the opposite type of rTMS with the same cognitive stimulation in the second phase after a 4-week washout period; we only included the pre-and post-data of the first phase prior to crossover.Osuch et al. [47] reported change scores between baseline and endpoint for each condition, which were then used to calculate the SMD.In studies with multiple treatment arms [48][49][50][51][52][53][54][55][56], only the active/sham rTMS + psychological intervention arms were analysed.Of note, 4 studies had two active arms [50,52,57,58].In these cases, the sample size of the control group was halved to avoid calculating the same control group twice in the meta-analysis [59].In summary, 31 treatment arms were included.There were 72 healthy participants and 1060 participants from a clinical population included in the meta-analysis.Clinical populations included patients with low cognitive restructuring ability (n = 46), post-traumatic stress disorder (PTSD) (n = 180), anxiety disorder (n = 83), smokers (n = 156), alcohol-dependent patients (n = 119), Alzheimer's disease (AD) (n = 167), obsessive compulsive disorder (OCD) (n = 30), attention-deficit/hyperactivity disorder (ADHD) (n = 62), cognitive impairment (n = 88), attention dysfunction (n = 58), post-stroke depression (n = 47), and major depressive disorder (MDD) (n = 24).For healthy participants, data were only available to assess the effect on clinical symptoms and cognition in two RCTs.Four treatment arms involved low-frequency (1 Hz) rTMS, three involved iTBS, and twenty-four involved high-frequency (≥5 Hz) rTMS.Most of the studies (77.8%) stimulated one brain region (especially the dorsolateral prefrontal cortex (DLPFC): 59.3% of total studies), and only six studies stimulated multiple sites.In the included studies, only one utilized a single-session intervention, while the majority employed multiple sessions.Notably, 85.2% of the studies involved 10 or more sessions.treatment arms involved low-frequency (1 Hz) rTMS, three involved iTBS, and twent four involved high-frequency (≥5 Hz) rTMS.Most of the studies (77.8%) stimulated on brain region (especially the dorsolateral prefrontal cortex (DLPFC): 59.3% of total studies and only six studies stimulated multiple sites.In the included studies, only one utilized single-session intervention, while the majority employed multiple sessions.Notabl 85.2% of the studies involved 10 or more sessions.

Clinical Outcomes in Healthy and Clinical Populations
Sixteen studies (19 treatment arms) reported the effect of rTMS combined with psychological interventions on clinical outcomes involving 751 participants: five studies used rTMS + CBT [52,[60][61][62][63], three studies used rTMS + CT [55, 57,70], one study used rTMS + mindfulness-based stress reduction [51], and seven studies used rTMS + exposure therapy [47,48,50,53,[71][72][73]. Of note, data for healthy participants were only available to assess the effect on clinical outcomes in one study using rTMS + CBT [62].Across the entire sample, significantly greater improvements in clinical symptoms were found after active difference between active rTMS + exposure therapy and sham rTMS + exposure therapy on clinical symptoms in patients with PTSD (Supplementary Figure S1) or smokers (Supplementary Figure S2).Subgroup analysis in studies with 10 or more sessions of combined interventions yielded a significant improvement in clinical symptoms following rTMS + psychological interventions (SMD = 0.34, CIs 0.07 to 0.61, Z = 2.49, p = 0.01, I 2 = 94%).Of note, the results showed that the augmentation effects of rTMS became significant for CBT when pooling studies with 10 or more sessions (SMD = 0.21, CIs 0.05 to 0.36, Z = 2.49, p < 0.01, I 2 = 78%) (Supplementary Figure S3).0.01, I 2 = 94%).Subgroup analyses according to different types of psychological interventions were further explored.The results showed that neither rTMS + CBT (SMD = 0.14, CIs −0.03 to 0.32, Z = 1.58, p = 0.11), rTMS + CT (SMD = 0.59, CIs −0.15 to 1.32, Z = 1.55, p = 0.12), nor rTMS + exposure therapy (SMD = 0.35, CIs −0.06 to 0.75, Z = 1.66, p = 0.10) had a statistically significant greater benefit for improving clinical outcomes (Figure 2).Moreover, data were only available to do subgroup analyses of rTMS + exposure therapy on clinical symptoms in patients with PTSD and smokers.Results showed no significant difference between active rTMS + exposure therapy and sham rTMS + exposure therapy on clinical symptoms in patients with PTSD (Supplementary Figure S1) or smokers (Supplementary Figure S2).Subgroup analysis in studies with 10 or more sessions of combined interventions yielded a significant improvement in clinical symptoms following rTMS + psychological interventions (SMD = 0.34, CIs 0.07 to 0.61, Z = 2.49, p = 0.01, I 2 = 94%).Of note, the results showed that the augmentation effects of rTMS became significant for CBT when pooling studies with 10 or more sessions (SMD = 0.21, CIs 0.05 to 0.36, Z = 2.49, p < 0.01, I 2 = 78%) (Supplementary Figure S3).The meta-analysis showed that active rTMS combined with psychological interventions did not produce a significantly greater improvement in depressive symptoms compared to sham rTMS + psychological intervention in the analysed clinical populations (SMD = 0.06, CIs − 0.16 to 0.29, Z = 0.55, p = 0.58, I 2 = 86%; Figure 3).A further subgroup analysis in patients with PTSD showed no difference in depressive symptoms between active or sham rTMS when combined with exposure therapy (Supplementary Figure S4).

Cognitive Outcomes in Healthy and Clinical Populations
We investigated the effect of active/sham rTMS + CT on cognitive function from 13 studies (14 treatment arms) for 741 participants across 7 cognitive domains: perceptualmotor function, language, executive function, learning and memory, complex attention, working memory, and global cognition.Data for the effect on cognition were only available in one study in healthy participants (n = 27).When data from these ten studies were pooled, there was an overall significantly greater effect with active rTMS + CT compared to sham rTMS + CT (SMD = 0.28, CIs 0.15 to 0.42, Z = 4.03, p < 0.01).However, the results were highly heterogeneous (I 2 = 88%).Due to insufficient data, subgroup analyses were limited to the cognitive domains of complex attention, executive function, learning and memory, and global cognition.Results showed that the combination of rTMS + CT had a small effect on global cognition (SMD = 0.45, CIs 0.21 to 0.68, Z = 3.75, p < 0.01), with significant heterogeneity (I 2 = 89%).When subgroup analysis in clinical populations was conducted, the results remained significant on overall cognition when all domains were collapsed (SMD = 0.33, CIs 0.18 to 0.49, Z = 4.22, p < 0.01, I 2 = 88%) and for global cognition (SMD = 0.47, CIs 0.22 to 0.72, Z = 3.70, p < 0.01, I 2 = 89%).Furthermore,

Cognitive Outcomes in Healthy and Clinical Populations
We investigated the effect of active/sham rTMS + CT on cognitive function from 13 studies (14 treatment arms) for 741 participants across 7 cognitive domains: perceptualmotor function, language, executive function, learning and memory, complex attention, working memory, and global cognition.Data for the effect on cognition were only available in one study in healthy participants (n = 27).When data from these ten studies were pooled, there was an overall significantly greater effect with active rTMS + CT compared to sham rTMS + CT (SMD = 0.28, CIs 0.15 to 0.42, Z = 4.03, p < 0.01).However, the results were highly heterogeneous (I 2 = 88%).Due to insufficient data, subgroup analyses were limited to the cognitive domains of complex attention, executive function, learning and memory, and global cognition.Results showed that the combination of rTMS + CT had a small effect on global cognition (SMD = 0.45, CIs 0.21 to 0.68, Z = 3.75, p < 0.01), with significant heterogeneity (I 2 = 89%).When subgroup analysis in clinical populations was conducted, the results remained significant on overall cognition when all domains were collapsed (SMD = 0.33, CIs 0.18 to 0.49, Z = 4.22, p < 0.01, I 2 = 88%) and for global cognition (SMD = 0.47, CIs 0.22 to 0.72, Z = 3.70, p < 0.01, I 2 = 89%).Furthermore, the results of subgroup analysis in patients with AD also remained significant for overall cognition (SMD = 0.43, CIs 0.20 to 0.66, Z = 3.6, p < 0.01; Supplementary Figure S7) and for global cognition (SMD = 0.41, CIs 0.08 to 0.75, Z = 2.41, p = 0.02; Supplementary Figure S7).There was no significant difference between conditions for the complex attention, learning and memory, and executive function domains (Figure 4).

Publication Bias
The funnel plots for the effect sizes of clinical symptoms, depressive symptoms, functional outcomes, overall cognitive effects, and global cognition showed symmetry, indicating no evidence of publication bias (Supplementary Figures S8-S11).This was supported by the results of the Egger's tests (p > 0.05).

Risk of Bias Assessment and Sensitivity Analyses
Two studies (7.4%) were assessed as having a high risk of bias.Herrmann et al. [73] had a high risk of bias arising from the randomization process and Osuch et al. [47] had a high risk of bias due to deviations from the intended intervention (Supplementary Figure S12).Results from the ROB2 tool are shown in Supplementary Table S5.After removing these two studies, the effect of active rTMS combined with psychological interventions was still significantly greater for clinical outcomes compared to sham rTMS combined with psychological interventions (SMD = 0.32, CIs 0.06 to 0.57, Z = 2.41, p = 0.02, I 2 = 95%).When pooling studies with 10 or more sessions of combined interventions, subgroup analysis showed rTMS had augmentation effects on psychological interventions for overall clinical symptoms with a small-sized effect (SMD = 0.36, CIs 0.07 to 0.64, Z = 2.47, p < 0.01, I 2 = 95%) after excluding one study with a high risk of bias [47].Functional and cognitive the results of subgroup analysis in patients with AD also remained significant for overall cognition (SMD = 0.43, CIs 0.20 to 0.66, Z = 3.6, p < 0.01; Supplementary Figure S7) and for global cognition (SMD = 0.41, CIs 0.08 to 0.75, Z = 2.41, p = 0.02; Supplementary Figure S7).There was no significant difference between conditions for the complex attention, learning and memory, and executive function domains (Figure 4).

Publication Bias
The funnel plots for the effect sizes of clinical symptoms, depressive symptoms, functional outcomes, overall cognitive effects, and global cognition showed symmetry, indicating no evidence of publication bias (Supplementary Figures S8-S11).This was supported by the results of the Egger's tests (p > 0.05).

Risk of Bias Assessment and Sensitivity Analyses
Two studies (7.4%) were assessed as having a high risk of bias.Herrmann et al. [73] had a high risk of bias arising from the randomization process and Osuch et al. [47] had a high risk of bias due to deviations from the intended intervention (Supplementary Figure S12).Results from the ROB2 tool are shown in Supplementary Table S5.After removing these two studies, the effect of active rTMS combined with psychological interventions was still significantly greater for clinical outcomes compared to sham rTMS combined with psychological interventions (SMD = 0.32, CIs 0.06 to 0.57, Z = 2.41, p = 0.02, I 2 = 95%).When pooling studies with 10 or more sessions of combined interventions, subgroup analysis showed rTMS had augmentation effects on psychological interventions for

Discussion
To the best of our knowledge, this is the first systematic review and meta-analysis to investigate whether rTMS can potentially augment the efficacy of psychological interventions.In summary, the results provided preliminary evidence that: (1) rTMS could potentially augment the effects of psychological interventions on overall clinical symptoms with a small-sized effect across a broad range of health conditions; and (2) active rTMS could potentially produce small-sized cognitive-enhancing benefits on CT for overall cognition in studies involving healthy and clinical populations.
While the combined effects of rTMS and psychological interventions are increasingly being investigated, it remains unclear whether rTMS can augment clinical outcomes.Overall, the results showed that there may be a small-sized beneficial effect of rTMS on psychological interventions for clinical outcomes.More intervention sessions seemed to slightly increase effect size when conducted subgroup analysis including studies with 10 or more sessions of combined interventions.Heterogeneity between studies, however, was high, and subgroup analyses according to different treatment strategies (CBT, CT, and exposure therapy) failed to reach statistical significance, likely due to reduced statistical power available in subgroup analyses.Interestingly, rTMS showed a potential augmentation effect on CBT in studies which included 10 or more sessions, suggesting rTMS may facilitate more rapid clinical effects of CBT with a higher number of sessions.In the subgroup analyses of patients with PTSD or smokers, no beneficial effect was observed following rTMS combined with exposure therapy; however, there were insufficient studies to do subgroup analyses on other health conditions.Of note, only three studies were available for the subgroup analyses of patients with PTSD and smokers, respectively, and the rTMS protocol utilized in each study was unique, which limited statistical power to detect a beneficial effect and so suggests the need for further research.Nevertheless, the current findings provide preliminary and promising evidence suggesting a potential augmentation effect of rTMS on psychological interventions for clinical outcome measures in a broad context.Further research, particularly with larger sample sizes within specific disorders and rTMS protocols with sufficient therapeutic parameters, is warranted to clarify and validate these preliminary findings.
A recent review provided preliminary evidence that greater improvement in depressive symptoms could be achieved by rTMS combined with psychological interventions, such as CBT and cognitive-emotional reactivation, but no meta-analysis was conducted [3].Active rTMS combined with psychological interventions was not significantly more beneficial for improving depressive symptoms in clinical populations in this analysis.The result may be due to the large variability in the types of psychological interventions used in this analysis: three studies used rTMS + exposure therapy, four studies used rTMS + CT, two studies used rTMS + CBT, whereas only one study used rTMS + MBSR to investigate the effect on depressive symptoms.A subgroup analysis revealed that patients with PTSD who received rTMS + exposure therapy did not significantly improve their depressive symptoms, which might be partially attributed to no improvement in the clinical symptoms mentioned above.There is a parallel association between improvement in PTSD symptoms and improvements in depression [74].Depressive symptom outcomes were obtained from different clinical populations, which may also have contributed to the non-significant results.It is also important to note that there were only two studies investigating the effect on depressive symptoms in patients with depression, which limits any conclusions about the use of rTMS to boost the effects of psychological interventions in this population.The majority of patients in the included studies were treated with a minimum of 10 sessions of combined interventions, with either high frequency rTMS applied to the left PFC or low frequency rTMS applied to the right PFC.A large naturalistic study found that rTMS combined with psychotherapy resulted in a 66% response and a 56% remission rate after treatment, with no difference found between high-frequency and low-frequency rTMS [75].No standard dosage of rTMS was applied, which may also explain the lack of augmentation effects of rTMS on psychological interventions for depressive symptoms.Except for one study, that utilized 2880 pulses per session, the TMS dose ranged from 1400-2000 pulses per session, which was much lower than the FDA-approved rTMS protocol of 3000 pulses per session [75].Future randomized trials with a larger number of sessions of rTMS and psychological interventions in patients with MDD or comorbid depression are needed to determine the efficacy of this combined intervention for improving depressive symptoms.
We did not find an additional benefit of rTMS on psychological interventions for functional outcomes in clinical populations or in sub-samples treated with 10 or more sessions.However, it is important to note that only a small number of studies (N = 8) reported functional outcomes, meaning that a lack of effect cannot be concluded from this preliminary analysis, which included diverse functional outcomes measures (e.g., quality of life, vocational and social functioning, and activities of daily living).Deficits in social and vocational functioning, as well as compromised quality of life, are critical and shared features across psychiatric disorders [76,77].Given the significance of these functional outcomes for patients, it is critical that future research investigates the augmentation effects of rTMS on psychological interventions to assess these outcomes.
Previous meta-analyses have found greater improvement in cognition with active rTMS combined with CT compared to rTMS alone [78,79].In the present study, we extended these past findings by examining whether active rTMS combined with CT was more beneficial compared to sham rTMS combined with CT in a sample of studies involving healthy and clinical populations.The results showed that overall cognitive functions were potentially improved following active rTMS combined with CT, albeit with a small effect size and significant heterogeneity.Additionally, we also observed a greater improvement in overall cognition following active rTMS combined with CT compared to sham rTMS combined with CT in a subgroup analysis of patients with AD.In addition, high-frequency rTMS applied to the prefrontal cortex, particularly the DLPFC, was used in most of the included studies assessing the effects of rTMS + CT.Existing evidence has demonstrated that high-frequency rTMS targeted over the prefrontal cortex has beneficial effects on cognitive functioning [17,78].In this study, we provide further evidence that multiple sessions of high-frequency rTMS sessions administered to this region could have additional cognitive-enhancing effects on CT.
Interestingly, following a subgroup analysis of different cognitive domains, rTMS + CT was only significantly beneficial for improving global cognition with a small effect size.Global cognition measures a wide range of cognitive skills, including orientation, attention, memory, and visuospatial and constructional abilities [80].A recent systematic review and meta-analysis of neuropsychiatric symptoms across various diagnoses also found rTMS to the left DLPFC alone had a small-sized effect not only for global cognition, but also for declarative memory, working memory, and cognitive control [80].The current study extends this finding by showing that rTMS might augment the effects of CT in improving global cognition.In this subgroup analysis, nine out of eleven studies examined the effect on global cognition in older populations, who were more likely to be cognitively impaired at baseline.Moreover, a subgroup analysis showed the augmentation effect on global cognition remained significant in patients with AD, who were characterized by significant cognitive impairment and probably benefited most from these combined interventions [81].The failure to find cognitive-enhancing effects for the executive function, learning and memory, and complex attention domains was likely due to the small number of studies included for these domains.Therefore, we cannot rule out that there are no additional beneficial effects of rTMS on CT for these outcomes at this stage.
This study has several limitations.First, when a reasonable lower limit of 5 studies is recommended for subgroup analyses [82], the limited number of studies in some subgroups hampers our comprehensive assessment of the effects of certain treatment strategies (e.g., rTMS + MBSR), outcome measures, or in certain disorders (e.g., panic disorders, ADHD).This limited statistical power to detect potential beneficial effects of rTMS on psychological interventions, suggesting that more studies across diverse categories are warranted to validate and further extend upon these preliminary findings.Secondly, the included studies exhibited substantial clinical and methodological heterogeneity.We attempted to conduct subgroup and sensitivity analyses to identify potential sources of heterogeneity.However, the heterogeneity remained high.The current study examined the effects of four different treatment strategies (various combinations of rTMS and psychological intervention parameters) in participants with twelve health conditions on three outcome measures grouped into different domains.Thus, aggregating effects from different conditions likely contributed to heterogeneity.The high heterogeneity made it challenging to draw meaningful and generalizable conclusions from our current meta-analysis.Future research would benefit from including larger sample sizes in different clinical populations and further examination of the specific efficacy of rTMS with a standardized protocol (e.g., sufficient number of sessions) on psychological interventions in improving clinical, functional, and cognitive outcomes.Third, the combination of rTMS with psychological interventions is a relatively new and emerging area of research, so the number of included studies in the respective analyses was limited, which limited the statistical power to detect bias.Fourth, a high proportion of included studies were assessed as having some concern of risk of bias, while two studies exhibited high risk.Nevertheless, sensitivity analyses showed that the results were not changed on overall cognition after the removal of these two studies.

Conclusions
In conclusion, this study provides preliminary evidence that active rTMS may augment clinical outcomes when combined with psychological interventions.Further, rTMS might produce a small-sized cognitive-enhancing effect on CT, warranting further research into this combined intervention.These preliminary results, however, must be interpreted with caution due to high heterogeneity and, therefore, require confirmation and replication in future research.In particular, studies with large sample sizes are needed to better understand and optimize the combined effect of rTMS and psychological interventions for specific clinical conditions.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/brainsci13121665/s1,Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist.Table S1: Search Strategies; Table S2: Clinical symptoms variables included in the analyses; Table S3: Functional outcomes variables included in the analyses; Table S4: Cognitive variables included in the analyses; Table S5: Risk of bias assessment;

Figure 1 .Figure 1 .
Figure 1.Flowchart of study searching and selection.Figure 1. Flowchart of study searching and selection.
Brain Sci.2023, 13, 1665 10 of 17 outcomes were not affected by the removal of these two studies as they only examined clinical outcomes.Brain Sci.2023, 13, x FOR PEER REVIEW 12 of 20
Fifth, we only included studies comparing active or sham rTMS combined with active psychological interventions.Future research involving sham psychological interventions is needed to Brain Sci.2023, 13, 1665 13 of 17 determine whether improved clinical outcomes with rTMS combined with psychological interventions are due to synergistic or additive mechanisms.
Figure S1: Forest plots of the effect of rTMS + exposure therapy on clinical symptoms in patients with PTSD; Figure S2: Forest plots of the effect of rTMS + exposure therapy on clinical symptoms in smokers; Figure S3.Forest plots of the effect of rTMS + psychological interventions with 10 or more sessions on clinical symptoms; Figure S4: Forest plots of the effect of rTMS + exposure therapy on depressive symptoms in patients with PTSD; Figure S5: Forest plots of the effect of rTMS + psychological interventions on functional outcomes; Figure S6.Forest plots of the effect of rTMS + psychological interventions with 10 or more sessions on functional outcomes; Figure S7: Forest plots of the effect of rTMS + CT for different cognitive domains in patients with AD; Figure S8: Funnel plot for the effect of clinical symptoms; Figure S9: Funnel plot for the effect of depressive symptoms; Figure S10: Funnel plot for the effect of functional outcomes; Figure S11: Funnel plot for the effect of global cognition; Figure S12: Risk of bias assessment.
Library, and PsycINFO (Ovid) up to 22 April 2023.The search strategies are shown in Supplementary Table S1.Studies were included if they fulfilled the following criteria: (1) healthy or clinical populations ≥ 18 years; (2) the comparison of rTMS combined with psychological interventions (e.g., CBT, exposure therapy, and CT) to sham rTMS combined with psychological interventions;