Clinical trials on low-frequency repetitive transcranial magnetic stimulation (LF-rTMS) over the right dorsolateral prefrontal cortex have yielded conflicting evidence concerning its overall efficacy for treating major depression (MD). As this may have been the result of limited statistical power of individual trials, we have carried the present systematic review and meta-analysis to examine this issue. We searched the literature for English language randomized, double-blind and sham-controlled trials (RCTs) on LF-rTMS for treating MD from 1995 through July 2012 using EMBASE, PsycINFO, Cochrane Central Register of Controlled Trials, SCOPUS, and ProQuest Dissertations & Theses, and from October 2008 until July 2012 using MEDLINE. The main outcome measures were response and remission rates as well as overall dropout rates at study end. We used a random-effects model, odds ratios (ORs) and number needed to treat (NNT). Data were obtained from eight RCTs, totaling 263 subjects with MD. After an average of 12.6±3.9 rTMS sessions, 38.2% (50/131) and 15.1% (20/132) of subjects receiving active LF-rTMS and sham rTMS were classified as responders (OR=3.35; 95% CI=1.4-8.02; p=0.007). Also, 34.6% (35/101) and 9.7% (10/103) of subjects receiving active LF-rTMS and sham rTMS were classified as remitters (OR=4.76; 95% CI=2.13-10.64; p<0.0001). The associated NNT for both response and remission rates was 5. Sensitivity analyses have shown that protocols delivering >1200 magnetic pulses in total as well as those offering rTMS as a monotherapy for MD were associated with higher rates of response to treatment. No differences on mean baseline depression scores and dropout rates for active and sham rTMS groups were found. Finally, the risk of publication bias was low. In conclusion, LF-rTMS is a promising treatment for MD, as it provides clinically meaningful benefits that are comparable to those of standard antidepressants and high-frequency rTMS. Furthermore, LF-rTMS seems to be an acceptable intervention for depressed subjects.