Background We studied the effect of one session of high-frequency repetitive transcranial magnetic stimulation (rTMS) on brainstem stem excitability in patients with incomplete SCI. Methods The study was a randomized, double-blind, sham-controlled trial. We recruited 27 patients with SCI to be randomly distributed in two study groups: active or sham rTMS. We recorded the blink reflex (BR) induced with electrical stimulation at 15 times sensory threshold (ST) to the supraorbital nerve (SON). A prepulse electrical stimulus on BR (PP-BR), was delivered with ring electrodes to the right index finger at 2 times ST and applied 100 ms before the SON stimulus. We also examined the excitability recovery curve of the blink reflex (RC-BR), with pairs of stimuli at interstimuli intervals (ISI) of 160, 300, 500, and 1000 ms. Patients were examined at baseline, and after high-frequency vertex rTMS (20 Hz; 40 pulses/burst, 1800 pulses total over 20 min). Results Fifteen patients received active rTMS and 12 sham rTMS. The age between groups was not significant. At baseline, R2 area of BR and percentage changes (%) in RC-BR at all intervals were similar in both groups. Active but not sham rTMS significantly reduced the area of R2 in BR and decreased the % inhibition with respect to baseline (p ⩽ 0.05), RC-BR did not change with any rTMS stimulation at any time. Conclusions High frequency-rTMS over motor cortex induces a decrease in brainstem reflex excitability and reduces inhibitory brainstem activity in patients with SCI.