We wished to assess the role of increased vagal tone and arterial oxygen saturation (SaO2) as determinants of HR response to voluntary respiratory maneuvers in OSAS. The changes in HR and SaO2 during breath-hold (B), Valsalva (V) and Mueller (M) maneuvers were determined in nine male subjects with OSAS while breathing RA or O2. Oxygen saturation was significantly lower breathing RA than O2 at the end of B (92.6 +/- 1.6 vs 97.2 +/- 0.8 percent), V (92.9 +/- 1.3 vs 95.2 +/- 1.7 percent), and M (92.7 +/- 1.2 vs 95.3 +/- 1.9 percent). Despite this, there was no significant difference between the HR change while breathing RA and O2 during B (12 +/- 18 vs 7 +/- 15 beats/minute), V (-2 +/- 12 vs -5 +/- 17 beats/minute), and M (5 +/- 16 vs 1 +/- 8 beats/minute). The change in HR was not related to the duration of B, V, or M or to the mouth pressure generated during V and M. In order to determine if awake HR response to the maneuvers reflected HR response to obstructive apnea, we examined the relationship between the HR response to B, V, and M during wakefulness and the response to obstructive apnea of similar duration while asleep. A significant correlation was found between the HR response to obstructive sleep apnea during sleep and the response to awake B (r = 0.67, p less than 0.001), V (r = 0.51, p less than 0.05), and M (r = 0.75, p less than 0.001). We conclude that in OSAS, increased vagal tone is a major determinant of HR response to voluntary respiratory maneuvers, that bradycardia can occur in the absence of hypoxemia, and that HR response to these maneuvers, especially to M, during wakefulness predicts HR response to obstructive apnea while asleep.