An updated meta-analysis: Short-term therapeutic effects of repeated transcranial magnetic stimulation in treating obsessive-compulsive disorder

Highlights

• Targeting the right DLPFC seems produce larger therapeutic effects than targeting other regions;

• Low frequency and high frequency rTMS produce similar and significant treatment effects.

• Stimulation at 100% of the RMT appear to provide the best improvement.

• Studies using tilted coils as control yielded significantly larger effects than sham coils.

Abstract

Background

This study was conducted to evaluate the short-term therapeutic effects of using repeated transcranial magnetic stimulation (rTMS) to treat obsessive-compulsive disorder (OCD) and to examine potential influencing factors.

Method

We searched the PubMed, EMBASE, CENTRAL, Wanfang, CNKI, and Sinomed databases on September 18, 2016 and reviewed the references of previous meta-analyses. Sham-controlled, randomized clinical trials using rTMS to treat OCD were included. Hedge's g was calculated for the effect size. Subgroup analyses and univariate meta-regressions were conducted.

Results

Twenty studies with 791 patients were included. A large effect size (g=0.71; 95%CI, 0.55–0.87; P<0.001) was found for the therapeutic effect. Targeting the supplementary motor area (SMA) (g=0.56; 95%CI, 0.12–1.01; P<0.001), left dorsolateral prefrontal cortex (DLPFC) (g=0.47; 95%CI, 0.02–0.93; P=0.02), bilateral DLPFC (g=0.65; 95%CI, 0.38–0.92; P<0.001) and right DLPFC (g=0.93; 95%CI, 0.70–1.15; P<0.001), excluding the orbitofrontal cortex (OFC) (g=0.56; 95%CI, −0.05–1.18; P=0.07), showed significant improvements over sham treatments. Both low-frequency (g=0.73; 95%CI, 0.50–0.96; P<0.001) and high-frequency (g=0.70; 95%CI, 0.51–0.89; P<0.001) treatments were significantly better than sham treatments, with no significant differences between the effects of the two frequencies. The subgroup analyses indicated that patients who were non-treatment resistant, lacked concurrent major depressive disorder (MDD) and received threshold-intensity rTMS showed larger therapeutic effects than the corresponding subgroups. The subgroup analysis according to sham strategy showed that tilted coils yielded larger effects than sham coils. Meta-regression analyses revealed that none of the continuous variables were significantly associated with the therapeutic effects.

Limitations

Only short-term therapeutic effects were assessed in this study.

Conclusions

Based on this study, the short-term therapeutic effects of rTMS are superior to those of sham treatments. The site of stimulation, stimulation frequency and intensity and sham condition were identified as potential factors modulating short-term therapeutic effects. The findings of this study may inspire future research.

Introduction

Approximately 1–3% of the global population suffers from obsessive-compulsive disorder (OCD) (Horwath and Weissman, 2000). Pathological obsessions and compulsions can lead to significant distress and functional impairment. In addition, approximately 40–60% of OCD patients remain resistant to current first-line therapies (Pallanti and Quercioli, 2006).

Several randomized control trials using repeated transcranial magnetic stimulation (rTMS) to treat OCD have been published since 1997, but their results are inconclusive. The differences in results may be due to their use of different rTMS protocols or the inclusion of patients with different characteristics. Three meta-analyses evaluating the efficacy of rTMS for treating OCD have been conducted (Berlim et al., 2013, Ma and Shi, 2014, Trevizol et al., 2016). Berlim et al. (2013) included 10 RCTs with 282 subjects and identified a significant and medium effect size in favor of active rTMS (g=0.59), and the subgroup analyses in their study (which were only performed according to stimulation frequency and target) indicated that there was no significant difference in the high-frequency (HF) and dorsolateral prefrontal cortex (DLPFC) subgroups. A second meta-analysis (Ma and Shi, 2014) was limited to SSRI-resistant OCD patients and included 9 RCTs with 290 subjects. It found that rTMS can be an effective addition to SSRI therapy and subgroup analyses were conducted only for the weeks of rTMS treatment. In the third meta-analysis (Trevizol et al., 2016), 15 studies (n=483) were included, and a medium effect size (g=0.45) was found. The meta-regression identified no significant variables. Recently, several RCTs using rTMS to treat OCD have been published (Elbeh et al., 2016, Hawken et al., 2016, Pelissolo et al., 2016, Seo et al., 2016). Thus, it is necessary to perform an updated meta-analysis to explore other important factors which may be associated with the efficacy of rTMS for treating OCD.

Before conducting this meta-analysis, we made several assumptions. OCD symptoms are correlated with hyperactivity in the cortico-striato-thalamo-cortical circuits (Anticevic et al., 2014, Milad and Rauch, 2012), and we assumed that the inhibitory effect of LF stimulation would be more effective than the excitatory effect of HF stimulation (Speer et al., 2009). The DLPFC, which is connected to the striatum, the anterior cingulate cortex, and the thalamus (Barbas, 2000, Paus et al., 2001, Petrides and Pandya, 1999), is the most common target for rTMS. Stimulating the DLPFC can also affect connected areas, some of which are associated with OCD symptoms. Therefore, we assumed that stimulating the DLPFC could result in effective treatment.