Pilot open-label application of high-frequency repetitive transcranial magnetic stimulation (rTMS) with H-coil in Parkinson's Disease (PD) have shown promising results. To evaluate safety and efficacy of high-frequency rTMS with H-coil in PD in a double-blind, placebo-controlled, randomized study. Sixty patients with PD were randomized into 3 groups: M1-PFC (real stimulation on primary motor-M1 and pre-frontal cortices-PFC), M1 (real rTMS on M1, sham on PFC), Sham (apparent stimulation). Primary outcome was baseline-normalized percent improvement in UPDRS part III OFF-therapy at the end of treatment (12 rTMS sessions, 4 weeks). Secondary outcomes were improvement in UPDRS part III sub-scores, timed tests, and neuropsychological tests. Statistical analysis compared improvement following real and sham stimulation at the end of the protocol using either a -test or a Mann-Whitney test. All patients tolerated the treatment and concluded the study. One patient from M1-PFC group was excluded from the analysis due to newly discovered uncontrolled diabetes mellitus. No serious adverse effect was recorded. At the end of treatment, patients receiving real rTMS (M1-PFC and M1 combined) showed significantly greater improvement compared to sham in UPDRS part III total score ( = 0.007), tremor subscore ( = 0.011), and lateralized sub-scores ( = 0.042 for the more affected side; = 0.012 for the less affected side). No significant differences have been oserved in safety and efficacy outcomes between the two real rTMS groups. Notably, mild, not-distressing and transient dyskinesias occurred in 3 patients after real rTMS in OFF state. The present findings suggest that high-frequency rTMS with H-coil is a safe and potentially effective procedure and prompt larger studies for validation as add-on treatment in PD.