There is no standard procedure to evaluated the peak of reflex cough flow (PCF-reflex) in the literature, which is important assessment in subjects with acquired brain injury and tracheostomy cannula. The present study aims to investigate the PCF of the reflex cough in a broad sample of healthy controls and, furthermore, the presence and the strength of voluntary and reflex cough in subjects with ABI with tracheostomy cannula.We recruited 147 participants including the healthy subjects (n = 105) and acquired brain injury subjects (n = 43), who underwent respiratory assessment: the Tidal Volume, Forced Vital Capacity, PCF of voluntary cough (PCF-voluntary) and PCF-reflex (using a spirometer connected with a nebulizer by a bidirectional).The PCF-reflex of controls and subjects was significant lower than the PCF-voluntary (P 0.01). The PCF-voluntary was not assessed in 26 (60.5 %) subjects due to severe cognitive deficit. In subjects without cognitive deficits (n = 17; 39.5 %), it was significantly lower than in healthy controls (p 0.01). In contrast, the PCF-reflex was completed in all subjects and it was not significantly different from healthy controls. Furthermore, the strength of the PCF-reflex decreased with increasing inhalation numbers of nebulised air.Reflex cough behaviour differs largely from voluntary cough and the PCF results reflect this great discrepancy. PCF-reflex could be useful parameter for assessing the airway protection whereas PCF-voluntary for measuring airway clearance.