Depression is a major debilitating psychiatric disorder. Current antidepressant drugs are often associated with side effects or treatment resistance. The aim of this meta-analysis was to evaluate therapeutic effects of high-frequency repetitive transcranial magnetic stimulation (HF-rTMS) in major depression (MD).

The medical data bases of PubMed, Medline, Embase and Cochrane Central Register were searched for randomized controlled trials (RCTs) reporting the therapeutic effects of high-frequency rTMS for depression, which were published in English between January 1990 and June 2016. The index terms were “depress*”, “depression” and “transcranial magnetic stimulation”. Depression outcome data of different sessions (5, 10, 15, and 20 sessions of rTMS treatment) were extracted and synthesized by calculating standardized mean difference (SMD) with 95% confidence intervals (CI) by using a random-effect model. Within each session group, the subgroup analyses based on the number of pulses (≤ 1000, 1200–1500, 1600–1800, and 2000–3000) were also conducted.

Thirty RCTs with a total of 1754 subjects including 1136 in the rTMS group and 618 in the sham group were included in this meta-analysis. rTMS had a significant overall therapeutic effect on depression severity scores (SMD = −0.73, P < 0.00001). The five, 10, 15, 20 sessions of rTMS treatments yielded the significant mean effect sizes of −0.43, −0.60, −1.13, and −2.74, respectively. In the four groups (5, 10, 15, 20 sessions), the maximal mean effect size was all obtained in the subgroup of 1200–1500 pulses per day (−0.97, −1.14, −1.91, −5.47; P < 0.05).

The increasing of HF-rTMS sessions is associated with the increased efficacy of HF-rTMS in reducing depressed patients’ symptom severity. A total number of pulses of 1200–1500 per day appear to deliver the best antidepressant effects of HF-rTMS.