Rationale: We have previously shown that in subjects with obstructive sleep apnea, repetitive hypoxia is associated with long-term facilitation as manifested by decreased upper airway resistance (R_UA). Objective: To study the influence of long-term facilitation on upper airway collapsibility as measured by the critical closing pressure model and to determine if changes in upper airway resistance correlated with changes in collapsibility. Methods: We studied 13 subjects (10 men, 3 women) with a mean apnea-hypopnea index of 43.9±24.0 events/hr. In the first protocol with 11 subjects, we measured collapsibility using a critical closing (Pcrit) protocol before and after episodic hypoxia. Brief (3 minute) isocapnic hypoxia (FIO₂=8%) followed by 5 minutes of room air was induced ten times. A sham study without hypoxia was performed on 8 subjects. Measurements: ventilatory parameters, R_UA and Pcrit before and after episodic hypoxia. Results: At 20 minutes of recovery, there was no change in minute ventilation but there was a significant decrease in R_UA compared to the control period (control, 8.6±4.8 cmH₂O/l/s v. recovery, 5.9±3.8 cmH₂O/l/s, p<0.05). However, there was no change in Pcrit between the control (2.3±1.9 cmH₂O) and recovery (2.7±3.2 cmH₂O) periods. No changes in R_UA or Pcrit were observed in the sham protocol. Conclusions: Long-term facilitation of upper airway dilators is not associated with changes in upper airway collapsibility in subjects with obstructive sleep apnea. These results corroborate previous evidence that changes in upper airway resistance and caliber can be dissociated from changes in upper airway collapsibility.