Repetitive transcranial magnetic stimulation (rTMS) for obsessive-compulsive disorder (OCD): an exploratory meta-analysis of randomized and sham-controlled trials

Objective: Randomized and sham-controlled trials (RCTs) on repetitive transcranial magnetic stimulation (rTMS) for treating obsessive-compulsive disorder (OCD) have yielded conflicting results that may be due to limited statistical power among individual studies. We pursued the present systematic review and meta-analysis to assess the efficacy of rTMS for OCD and to generate hypotheses for more robustly powered RCTs.

Method: We searched the literature for RCTs on rTMS for OCD from 1995 through December 2012 using MEDLINE, EMBASE, PsycINFO, Cochrane Central Register of Controlled Trials, and SCOPUS. We then performed an exploratory random-effects meta-analysis with the main outcome measures as pre-post changes in Yale-Brown Obsessive Compulsive Scale (Y-BOCS) scores, response to treatment and overall dropout rates at study end.

Results: Data were obtained from 10 RCTs, totaling 282 subjects with OCD. The pooled Hedges' g for pre-post Y-BOCS scores was 0.59 (z = 2.73, p = 0.006), indicating a significant and medium-sized difference in outcome favoring active rTMS. Furthermore, response rates were 35% and 13% for patients receiving active and sham rTMS, respectively (OR = 3.4, p = 0.002). Sub-group analyses indicated that LF-rTMS and rTMS protocols targeting non-DLPFC regions (i.e., orbitofrontal cortex or supplementary motor area) seem to be the most promising for reducing OCD-related symptoms. No differences on baseline depression scores or dropout rates at study end were observed between active and sham rTMS groups, although OCD severity at baseline was higher in the active group.

Conclusions: Our exploratory analyses show that active rTMS seems to be efficacious for treating OCD. Moreover, LF-rTMS and protocols targeting the orbitofrontal cortex or the supplementary motor area seem to be the most promising. Nevertheless, future RCTs on rTMS for OCD should include larger sample sizes and be more homogeneous in terms of demographic/clinical variables as well as stimulation parameters and brain targets.