The use of repetitive transcranial magnetic stimulation for treatment of obsessive-compulsive disorder: a scoping review

Purpose This paper aims to explore the relevant literature available regarding the use of repetitive transcranial magnetic stimulation (rTMS) as a mode of treatment for obsessive-compulsive disorder (OCD); to evaluate the evidence to support the use of rTMS as a treatment option for OCD. Design/methodology/approach The authors electronically conducted data search in five research databases (MEDLINE, CINAHL, Psych INFO, SCOPUS and EMBASE) using all identified keywords and index terms across all the databases to identify empirical studies and randomized controlled trials. The authors included articles published with randomized control designs, which aimed at the treatment of OCD with rTMS. Only full-text published articles written in English were reviewed. Review articles on treatment for conditions other than OCD were excluded. The Covidence software was used to manage and streamline the review. Findings Despite the inconsistencies in the published literature, the application of rTMS over the supplementary motor area and the orbitofrontal cortex has proven to be promising in efficacy and tolerability compared with other target regions such as the prefrontal cortex for the treatment of OCD. Despite the diversity in terms of the outcomes and clinical variability of the studies under review, rTMS appears to be a promising treatment intervention for OCD. Research limitations/implications The authors of this scoping review acknowledge several limitations. First, the search strategy considered only studies published in English and the results are up to date as the last day of the electronic data search of December 10, 2020. Though every effort was made to identify all relevant studies for the purposes of this review per the eligibility criteria, the authors still may have missed some relevant studies, especially those published in other languages. Originality/value This review brought to bare the varying literature on the application of rTMS and what is considered gaps in the knowledge in this area in an attempt to evaluate and provide information on the potential therapeutic effects of rTMS for OCD.


Introduction
Transcranial magnetic stimulation (TMS) is a non-invasive neuromodulatory intervention, which affects neural activity through rapidly alternating magnetic fields. The stimulation operates through Faraday's law of electromagnetic induction, where the rapidly alternating electric current in the stimulating coil placed over the scalp generates a magnetic field that moves across the skull and produces electric currents in the neural tissue beneath (Wagner et al., 2009). This magnetic field has the capacity to penetrate the bone of the skull to stimulate cortical activity. Pulses can be delivered in a repeated manner to induce long-term changes in neural activity (Dhaliwal et al., 2015) as an increase or a decrease in cortical excitability through relatively high (>5 Hz) or low frequency (1 Hz) stimulation (Rossi et al., 2009;Wassermann et al., 1996). Repetitive transcranial magnetic stimulation (rTMS) is very flexible and, depending on the site and frequency, it can inhibit or induce local and remote brain activity (Liu et al., 2007). Typical rTMS comprises a train of repetitive pulses with similar stimulus intervals (Dhaliwal et al., 2015;Sandrini et al., 2011). Barker (1985) originally introduced TMS as a safe, and painless non-invasive means of applying focal brain stimulation, to stimulate the motor cortex and to assess human central motor pathways (Barker et al., 1985). rTMS has become an integral research tool in psychiatric treatment as method to exert explicit effects on a range of measures of brain function (Hallett, 2000;Rossini and Rossi, 2007). rTMS has been evaluated quite extensively as a therapeutic tool for several psychiatric disorders and is accepted as a brain-system-based, neuromodulation treatment for impacting direct targets involved in the neural circuitry of these disorders (Nahas et al., 2001).
Generally, rTMS treatments are comparatively simple and relatively easy to administer, are non-invasive and are typically well-tolerated by patients (Pink et al., 2021). A major benefit of rTMS is its relative safety being devoid of any major adverse side-effects (Machii et al., 2006). It is a highly cost-effective alternative to other more expensive treatment methods such as electroconvulsive therapy (Coles et al., 2018). The most frequent negative effect noticed by patients is temporary pain in the scalp, although with a moderate increase in the intensity of rTMS, it should be normalized (Perera et al., 2016). Vasovagal syncope may also manifest at the initial stages of the treatment and caution is taken to not avoid having the patient stand up, in addition, earplugs can help reduce the clicking sound experienced during rTMS administration (Tringali et al., 2012). rTMS was approved in Canada in 2002 and in the USA in 2008 (Kennedy et al., 2009;Höflich et al., 1993). In 2015, it was also approved by the National Institute for Health and Care Excellence for treatment-resistant depression in the UK (Fregni et al., 2005;Hara et al., 2016).
The large literature on superficial brain stimulation for mental disorders is based on rTMS for major depressive disorder (Cristancho et al., 2013a). Based on its versatility and efficacy, rTMS use has now been investigated in other psychiatric conditions including bipolar disorders, psychotic disorders, anxiety disorders, obsessive-compulsive disorder (OCD) and post traumatic stress disorders (PTSD) (Cristancho et al., 2013b). Evidence-based guidelines for the therapeutic use of rTMS (Lefaucheur et al., 2014) drew attention to the analgesic effect of high frequency (HF) rTMS of the motor cortex and the antidepressant effect of HF rTMS of the DLFPC. Similar encouraging outcomes have been reported for neuropsychiatric conditions such as schizophrenia and motor stroke. It has also been revealed that rTMS is capable of regulating cortical plasticity and brain network movements. The outcome depends on the selected cortical section and the different stimulating parameters such as the frequency, design and the potency of stimulations (Lefaucheur, 2008;Lefaucheur, 2012). Many studies including a metaanalysis confirm the antidepressant effects of rTMS of the DLFPC (Burt et al., 2002;Couturier, 2005), but there seems to be conflicting outcomes in relation to anxiety disorders (Herwig et al., 2007;O'Reardon et al., 2007).
Although antidepressants or psychotherapy help the symptoms of patients with OCD, this condition can be very debilitating and presents with a greater degree of non-response to conventional treatments (Ressler and Mayberg, 2007). Despite the wide use of rTMS for the management of mental disorders and the continuous interest in research for newer treatments for OCD, the therapeutic use of rTMS is still focused in the domain of depression (Schoenfeldt-Lecuona et al., 2010), and much less is known and evaluated for its use in the management of OCD.
In view of the above considerations, the clinical effectiveness of rTMS should be assessed in relation to its potential to provide OCD patients with safe, and lasting improvement in quality of life (Machado et al., 2012;Grant and Booth, 2009). This scoping review aims to identify what we know and to consider gaps in our knowledge in this area in an attempt to evaluate and provide information on the potential therapeutic effects of rTMS for OCD.

Methods
We developed an operationalized search strategy, which was applied to an electronically conducted data search in five research databases (MEDLINE, CINAHL, Psych INFO, SCOPUS and EMBASE) using relevant keywords and index terms across all the databases to identify empirical studies and randomized controlled trials (RCTs).
Key terms included: rTMS, OCD, Post-traumatic stress disorder, Bipolar disorders and Treatment. This was a larger search strategy involving results for the use of rTMS for the treatment of three major mental disorders (OCD, PTSD and Bipolar Disorders). This paper reports only on and discusses the results specifically for OCD. Table 1 shows a sample of the search strategy, for Medline.
Two independent reviewers (Medard Adu and Ejemai Eboreime) conducted the title and abstract screening, as well as the full text screening and came out with relevant articles that conformed to the objectives of the scoping review. Thematic classifications were done by the first reviewer (MA), with decisions analyzed by the second reviewer (EE). Where conflicts in classification arose, the articles in question were scrutinized and consensus was reached between the two reviewers.

Inclusion and exclusion criteria
Inclusion criteria included studies involving a completed RCT of rTMS as a treatment intervention for OCD. Open label trials on OCD using rTMS as a treatment intervention were also included. The review only covered full text articles and studies published in English. Studies involving rTMS as a form of treatment for PTSD, Bipolar disorders, OCD with comorbidities or studies involving any other conditions other than OCD, as well as those examining rTMS as a combined therapy with pharmacotherapy or any other interventions were excluded. Systematic reviews, meta-analysis and study protocols and experiments with rTMS that were not designed for treatment for OCD were not included.

Results
Through the search strategy and the use of the Covidence software, we identified a total of 2,373 studies from the electronic databases searched. The Covidence software automatically screened and removed 872 studies as duplicates. The remaining items (1,501) were screened against the eligibility criteria set by the authors based on the title and abstract only, yielding 182 remaining records for full text screening. In total, 154 studies were excluded in the full text screening phase, leaving a final pool of 28 studies that were eligible for inclusion in this scoping review ( Figure 1).
Many of the studies examined rTMS as a stand-alone treatment intervention for OCD with most of them comparing the use and efficacy of rTMS to sham treatment. Relevant and detailed methodological information was extracted and summarized from the various studies and presented in Table 2.
We examined the geographical distribution of studies conducted on rTMS treatment for OCD globally, as presented in Figure 2. Out of the total of 28 studies included in our review, 12 (43%) were conducted in Asia, North America and South America had 4 (14%) and 2 (7%) studies, respectively, Europe had 5 (18%), Africa had 3 (11%) and Australia had 2 (7%) studies. This indicates that research on rTMS in OCD is being conducted across all continents, but the quantity and scope vary widely across geographical jurisdictions. Table 2 summarizes the main findings for these included studies.
Study designs vary widely, including 18 RCTs, 4 open-label trials, 4 retrospective analysis, 1 brief report and 1 case report. All these studies sought to evaluate the efficacy and effectiveness of rTMS for the treatment of OCD. Sample sizes ranged from 10 to 100 subjects across included studies with a mean sample size of 31.68. The studies were heterogeneous in terms of features of clinical variability such as the severity of OCD symptoms, duration of sickness and rate of resistance to pharmacotherapy. Location of rTMS stimulation, varied among studies, as did treatment duration and stimulus intensity. Of the 28 studies included, 19 used 70 mm figure-ofeight shaped coils because of their ability to induce more focal current compared to circular coils. The remaining studies     variously used the 9 cm circular coil, DB-80 butterfly doublecone coil and the H-shaped coil design. Duration of treatment varied across studies, from two weeks to seven weeks. In total, 19 studies applied rTMS with a low frequency and eight applied HF ranging from 10 Hz to 20 Hz, the one remaining study of the 28 compared effects of low and HF treatment protocols.
In total, 19 studies (68%) reported significant positive outcomes and the other 9 studies reported no significant symptom improvement. In each of the included studies, rTMS application was reported as well-tolerated with no significant side-effects, although there were a few reports of mild sideeffects such as mild headache, dizziness and scalp pain, across the studies.

Discussion
The 28 studies under review suggest that rTMS has potential as a safe and clinically efficacious treatment intervention for OCD. Despite the diverse outcome measures included in this selection of studies, there were some consistent significant OCD symptom improvements.
Many factors may have accounted for the varying effectiveness of the application of rTMS across the studies and major domains of outcomes. For instance, rTMS treatment protocols and stimulation parameters vary greatly across studies, with poorly defined intervention protocols. Another factor is that different measuring tools are used to evaluate similar outcomes across studies, making a comparative evaluation of results difficult. It also makes it difficult to understand which rTMS parameters lead to the most significant outcomes and treatment response.
However, due to the diverse nature and presentation of mental conditions, it may seem unrealistic to think uniquely of an optimal or even a standardized rTMS protocol that will work across studies of the different conditions even if they target similar symptoms. One important aspect of rTMS, as identified in this review is its versatility, which allows for the development and adaption of protocols addressing similar symptoms from different conditions with potentially positive outcomes.

Targeted brain regions of repetitive transcranial magnetic stimulation
The pathophysiology of OCD according to structural and functional neuroimaging studies is linked with the dysfunction of the orbitofronto-striato-pallido-thalamic circuitry, which includes the orbitofrontal cortex (OFC), Dorsolateral prefrontal cortex (DLPFC) and the medial PFC, as well as the thalamus (Saxena and Rauch, 2000;Van Den Heuvel et al., 2005). Modulation of this circuitry by neurosurgical mechanisms and by means of deep brain stimulation has proven to be effective in reducing symptoms of OCD (Mallet et al., 2008). Bearing in mind the possibility of rTMS in modulating cortical and subcortical structures of the brain, the possible therapeutic effects of rTMS have been extensively studied and evaluated in literature in the quest to normalizing hyper-or hypoactive brain regions by targeting dysfunctional cortico-subcortical circuits in people with OCD.
For many of the studies extracted, the locus of rTMS stimulation was at either the left-DLPFC or the right-DLPFC and with high or low frequency rTMS. The overall accepted rationale is that the DLPFC could be a possible starting point for the induction of remote stimulation in the corticosubcortical circuits connected. For most of the trials, the leftdorsolateral prefrontal cortex (LDLPFC) and rightdorsolateral prefrontal cortex (RDLPFC) were stimulated with the "5 cm method" where the figure-of-eight coil was centered on a point at 5 cm rostral to and in the same sagittal line as the optimal area for activating the right or left abductor pollicis brevis muscles during motor threshold (MT) assessment (Sachdev et al., 2007;Praško et al., 2006;Greenberg et al., 1997). As prefrontal mechanisms are implicated in OCD, Greenberg et al. (1997) undertook a non-sham-controlled, single-blind rTMS study on the evidence of PFC hypermetabolism and hyperperfusion in untreated OCD patients. The preliminary results suggest that DLPFC rTMS had modest, lateralized effects on compulsions but not obsessions.
From the data extracted, another brain region studied for the administration of rTMS is the OFC. As indicated earlier, the OFC performs a very important function in the pathophysiology of OCD and because obsessions and compulsions are deemed to be mediated at least in part by the hyperactivity in the orbitofrontal-subcortical circuits and the increase in functional activity in the OFC. Inspired by the fact that OFC rTMS may seem OCD-specific, a randomized, single blind sham-controlled study was conducted by Ruffini et al. (2009). The researchers evaluated the efficacy of LF-rTMS over the left OFC with a low frequency (1 Hz) rTMS at 80% RMT for 3 weeks. There was a significant reduction in Yele-Brown obsessive compulsive scale (YBOCS) scores for the active group after the 3 rd and 10 th weeks compared to sham treatment.
The supplementary motor area (SMA) is one of the most recent brain targets used for the application of rTMS and evidence suggests that the motor and premotor cortex are hyperexcitable in OCD. An open-label trial conducted by

ConƟnental distribuƟon of data on rTMS for OCD
NORTH AMERICA SOUTH AMERICA ASIA AFRICA AUSTRALIA EUROPE Mantovani et al. (2006) sought to evaluate whether lowfrequency rTMS to the SMA could normalize overactive motor cortical regions and thereby improve symptoms of patients with OCD. There was clinical improvement at the end of the first week of the treatment with rTMS and by the second week, there was a statistically significant improvement in the reductions seen in Yele-Brown obsessive compulsive scale (YBOCS), Clinical Global Impression, Beck depression inventory (BDI), Hamilton depression rating scale (HDRS), Hamilton anxiety rating scale (HARS) and Symptom Checklist-90. Following the publication of this study, many of the most recent trials on rTMS application for the treatment of drug resistant OCD focused on the SMA (Lee et al., 2017;Arumugham et al., 2018;Singh et al., 2019;Pelissolo et al., 2016;Talaei et al., 2009;Hegde et al., 2016;. Results suggest that 1 Hz rTMS over the SMA could be an efficient and safe add-on therapeutic method in treatmentresistant patients with OCD.

Treatment modality and stimulation frequencies
In regard to differences in low and HFs of rTMS, results from the extracted studies suggest that, administration of HF (10 Hz) rTMS at either 100% or 110% MT over the RDLPFC did not differ from sham rTMS in terms of efficacy in relieving symptoms, reducing clinical severity or improving responses in treatmentresistant OCD (Mansur et al., 2011;Elbeh et al., 2016). By contrast, another study indicated that low frequency (1 HZ) rTMS delivered to the RDLPFC appeared to be superior to sham rTMS for relieving OCD symptoms and depression, in patients with treatment-resistant OCD. Based on the results from the selected studies in this review, there is no evidence for a statistically significant difference between low or HF rTMS over RDLPFC and LDLPFC for the treatment of OCD. The different study designs did not contribute to any differences in the outcomes for treatment between the sham and active subjects. A study conducted (Sachdev et al., 2007;Praško et al., 2006) using the double-blind, randomized, shamcontrolled trial with the application of low or HF rTMS over the left or right PFC presented with a significant reduction in YBOCS scores in both sham and active subjects with no significant statistical difference in the two groups at the end of the treatment intervention. The results also failed to depict any meaningful therapeutic efficacy in treatment non-responder OCD patients from either of the groups (Kang et al., 2009). Sachdev et al. (2001) compared effects of active HF-RDLPFC rTMS to active HF-LDLPFC rTMS. The evaluation yielded notable improvement in the symptoms of the OCD in study subjects. Notwithstanding the significant improvement in YBOCS scores for the two arms of the study, it is possible that the positive results were because of the smaller sample size (N = 12) and also the absence of a control group. These same researchers six years later conducted a similar study that confirmed the assertion of a smaller sample size and the lack of a sham control. Sachdev et al. (2007) in their study with a larger sample size (N = 18) revealed that the active and sham arms of the study did not show any difference in the reduction in OCD symptoms after the treatment. These conflicting results indicate that prefrontal high or low frequency rTMS may probably not be effective in the treatment of OCD symptoms.
In contrast to the contradictory results from other studies, most of the trials that presented with major clinically insignificant improvements in OCD symptoms were the studies with the targeted brain regions over the SMA with low frequency rTMS (Donse et al., 2017;Lee et al., 2017;Arumugham et al., 2018;Singh et al., 2019;Pelissolo et al., 2016;Talaei et al., 2009;Hegde et al., 2016;Gomes et al., 2012; and also the left OFC with LF-rTMS (Kumar et al., 2018;Singh et al., 2019;Ruffini et al., 2009). These studies suggest that rTMS had a specific and significant clinically effective influence on OCD symptoms: specifically in relation to the SMA stimulation site.
Poor study outcomes as witnessed in most of the studies could be partly attributed to differences in stimulation parameters, shorter treatment durations (as many used two weeks), the levels of frequencies used and, in some cases, the use of the circular coil, which typically induces less focal current compared to the figureof-eight shape coil. Differences may also be attributed to the choice of whether left or right prefrontal cortices of targets for stimulations and the severity of the drug resistance of the subjects used for the purposes of the studies.
Other factors affecting therapeutic outcomes Many factors may have accounted for the varied effectiveness of the application of rTMS across the studies and major domains of outcomes. For instance, rTMS treatment protocols and stimulation parameters vary greatly across studies, with poorly defined intervention protocols. Another factor is the different measurement tools used for the evaluation of similar outcomes across studies, and therefore, making comparison and evaluation of results difficult. These inconsistencies also make it difficult to understand which rTMS parameters lead to the most significant outcomes and treatment responses. It remains possible that positive outcomes may also be attributed partially to the therapeutic contributions of concurrent medications taken by the subjects although most of the subjects have been on these medications for a long time without yielding improvements in their OCD symptoms.
Additionally, the varied clinical significance and effectiveness of rTMS across studies can also be partly attributed to factors such as, variations in coil type and coil positions, the different cortical targets and the variations in motor thresholds. In the case of the application of rTMS for the treatment of OCD, a majority of the studies applied rTMS to normalize frontal dysfunction associated with OCD symptoms, choosing to stimulate the left/right DLPFC or the SMA. For example, in the case of the cortical target, the SMA was consistently used to relieve subjects of their OCD symptoms with consistent and clinically significant treatment responses noted. Thus, from the data gathered with respect to rTMS in OCD, it seems that the SMA may be a promising target region for the application of rTMS to treat the symptoms of OCD in contrast to either left or right DLPFC.
Furthermore, an important factor noticed is the evaluation of the longevity and time course effects of rTMS. The majority of studies reviewed evaluated the treatment outcomes of the various interventions immediately after the last session of rTMS with a few months of follow-up. Considering the chronic, debilitating and high prevalent nature of mental conditions, evaluating the long-term therapeutic effects of rTMS intervention is of great importance. Therefore, it would be of high clinical significance and research value to estimate the sustainability of treatment effects, and specifically, maintenance strategies following response or remission with rTMS.

Limitations
The authors of this scoping review acknowledge several limitations. First, our search strategy considered only studies published in English and the results are up to date as the last day of the electronic data search of December 10, 2020. Though every effort was made to identify all relevant studies for the purposes of this review per our eligibility criteria, we still may have missed some relevant studies, especially those published in other languages.

Conclusion
Many of the studies included in this scoping review presented with conflicting and inconsistent outcomes on the efficacy and utilization of rTMS as a treatment intervention for OCD. This makes it difficult to make definitive conclusions on the clinical usefulness and the appropriate technique for rTMS treatment interventions for OCD. Larger sample sizes for sufficiently powered and preferably multi-centered sham-controlled trials with the appropriate coil and stimulation parameters, welldefined stimulation targets and a longer treatment duration would be required to bring clarity to the therapeutic effect of rTMS in the treatment of resistant OCD.
Despite the inconsistencies in the published literature, the application of rTMS over the SMA and the OFC has proven to be promising in efficacy and tolerability compared with other target regions such as PFC for the treatment of OCD. Despite the diversity in terms of the outcomes and clinical variability of the studies under review, rTMS appears to be a promising treatment intervention for OCD.