CPAP is the gold standard treatment for OSA and was conceived to be applied through a nasal interface. This study was designed to determine the acute effects of changing the nasal CPAP route to oronasal and oral in upper airway patency during sleep in patients with OSA. We hypothesized that the oronasal route may compromise CPAP's effectiveness in treating OSA. Eighteen patients (mean ± SD age, 44 ± 9 years; BMI, 33.8 ± 4.7kg/m2; apnea-hypopnea index, 49.0 ± 39.1 events/hour) slept with a customized oronasal mask with nasal and oral sealed compartments connected to a multidirectional valve. Sleep was monitored byusing full polysomnography and induced by low doses of midazolam. Nasal CPAP was titrated up to holding pressure. Flow route was acutely changed to the oronasal (n= 18) and oral route (n= 16) during sleep. Retroglossal area was continuously observed by using nasoendoscopy. Nasal CPAP (14.8 ± 4.1cm H2O) was able to stabilize breathing in all patients. Incontrast, CPAP delivered by the oronasal and oral routes promoted obstructive events in 12 (66.7%) and 14 (87.5%) patients, respectively. Compared with stable breathing during thenasal route, there was a significant and progressive reduction in the distance between theepiglottis and tongue base and the retroglossal area when CPAP was delivered by the oronasal and oral routes. CPAP delivered through the oronasal route may compromise CPAP's effectiveness in treating OSA.