We have recently observed obstructive apneas during nasal intermittent positive-pressure ventilation (nIPPV) and suggested that they were due to hypocapnia-induced glottic closure. To confirm this hypothesis, we studied seven healthy subjects and submitted them to nIPPV while their glottis was continuously monitored through a fiber-optic bronchoscope. During wakefulness, we measured breath by breath the widest inspiratory angle formed by the vocal cords at the anterior commissure along with several other indexes. Mechanical ventilation was progressively increased up to 30 l/min. In the absence of diaphragmatic activity, increases in delivered minute ventilation resulted in progressive narrowing of the vocal cords, with an increase in inspiratory resistance and a progressive reduction in the percentage of the delivered tidal volume effectively reaching the lungs. Adding CO2 to the inspired gas led to partial widening of the glottis in two of three subjects. Moreover, activation of the diaphragmatic muscle was always associated with a significant inspiratory abduction of the vocal cords. Sporadically, complete adduction of the vocal cords was directly responsible for obstructive laryngeal apneas and cyclic changes in the glottic aperture resulted in waxing and waning of tidal volume. We conclude that in awake humans passive ventilation with nIPPV results in vocal cord adduction that depends partly on hypocapnia, but our results suggest that other factors may also influence glottic width.