What is the central question of this study? Can repetitive transcranial magnetic stimulation (rTMS) of the genioglossus enhance the beneficial effects observed with transcranial magnetic stimulation single twitches on upper airway mechanical properties? What is the main finding and its importance? We found that both inspiratory and expiratory rTMS protocols induce a different activation pattern of upper airway muscles, with evidence for an increase in genioglossus corticomotor excitability in response to rTMS. This is of major importance because it might open the door for rTMS protocols with the goal of increasing corticomotor excitability and, thus, possibly increasing the tonic genioglossus activity, which is known to be diminished during sleep in subjects with sleep apnoea. Stimulation of upper airway (UA) muscles during sleep by isolated transcranial magnetic stimulation (TMS) twitch can improve airflow dynamics without arousal, but the effect of repetitive TMS (rTMS) on UA dynamics is unknown. Phrenic nerve magnetic stimulation (PNMS) can be used to produce painless experimental twitch-induced flow limitation during wakefulness. The aim of this study was to quantify the effects of rTMS applied during wakefulness on UA mechanical properties using PNMS in subjects with obstructive sleep apnoea (OSA). Phrenic nerve magnetic stimulation was applied to 10 subjects, with and without simultaneous rTMS, during inspiration and expiration. Flow-limitation characteristics and UA obstruction level were determined [maximal (V̇I,max)and minimal inspiratory airflow (V̇I,min),V̇I,max-V̇I,min flow drop (ΔV̇I),oropharyngeal (POro,peak ) and velopharyngeal peak pressures, oropharyngeal k1 /k2 ratios with k1 and k2 determined by the polynomial regression model between instantaneous flow and pharyngeal pressure and UA resistance]. Both genioglossus and diaphragm root mean squares and motor-evoked potential amplitudes (geniolossus, GGAmp ) and latencies were computed. A flow-limitation pattern always occurred after PNMS. A decrease in V̇I,max and an increase in ΔV̇I occurred following rTMS applied during inspiration, and POro,peak values were more negative with both inspiratory and expiratory rTMS. The GGAmp also increased significantly from the second to the last rTMS expiratory train twitch. All other parameters remained unchanged. These results suggest the following conclusions: (i)rTMS does not improve UA mechanical properties in awake subjects with OSA; (ii)the activation pattern of UA muscles differs following isolated twitch and repetitive cortical stimulation of the genioglossus; and (iii)rTMS applied during expiration induces corticomotor facilitation.