Abstract
RATIONALE: Cerebrovascular regulation is impaired in patients with moderate to severe obstructive sleep apnea; however, it is unknown whether this impairment exists in individuals with less severe sleep-disordered breathing. OBJECTIVES: To test the hypothesis that cerebrovascular responses to hypercapnia are attenuated in a nonclinical population-based cohort. METHODS: A rebreathing test that raised end-tidal CO₂ tension by 10 mm Hg was performed during wakefulness in 373 participants of the Wisconsin Sleep Cohort. MEASUREMENTS AND MAIN RESULTS: We measured cerebral flow velocity (transcranial Doppler ultrasound); heart rate (electrocardiogram); blood pressure (photoplethysmograph); ventilation (pneumotachograph); and end-tidal CO₂ (expired gas analysis). Cerebrovascular CO₂ responsiveness was quantified as the slope of the linear relationship between flow velocity and end-tidal CO₂ during rebreathing. Linear regression analysis was performed using cerebrovascular CO₂ responsiveness as the outcome variable. Main independent variables were the apnea-hypopnea index and the mean level of arterial oxygen saturation during sleep. We observed a positive correlation between cerebrovascular CO₂ responsiveness and the mean level of oxygen saturation during sleep that was statistically significant in unadjusted analysis and after adjustment for known confounders and the increase in arterial pressure during rebreathing. Each 5% decrease in Sa(O₂) during sleep predicted a decrease in cerebrovascular reactivity of 0.4 ± 0.2 cm/second/mm Hg P(ET)CO₂. In contrast, the negative correlation between cerebrovascular CO₂ responsiveness and apnea-hypopnea index was statistically significant only in the unadjusted analysis. CONCLUSIONS: Hypercapnic vasodilation in the cerebral circulation is blunted in individuals with sleep-disordered breathing. This impairment is correlated with hypoxemia during sleep.