Objective: Obstructive sleep apnea (OSA) is a stroke risk factor and is increasingly recognized as a risk factor for cognitive impairment. Altered cerebral autoregulation may play a role in these relationships. We measured the association between OSA and two forms of cerebral autoregulation: (1) dynamic cerebral autoregulation (DCA), which plays a homeostatic role; and (2) vasomotor reactivity (VMR), which is a measure of cerebrovascular reserve. We hypothesized that both VMR and DCA would be impaired in subjects with OSA.
Methods: We recruited subjects with untreated OSA. VMR and DCA were measured with continuous transcranial Doppler (TCD) of the middle cerebral arteries (MCA). DCA was measured with phase shift analysis where lower degrees of phase shift indicate greater impairment; values <24 degrees are abnormal. VMR was measured as % change in MCA velocity in response to 5% CO2 inhalation; values <2% change are abnormal. We assessed the relationship between apnea-hypopnea index (AHI) and autoregulation using bivariate correlations (Pearson coefficient). We also assessed the association between moderate to severe OSA (AHI≥15) and abnormal autoregulation (Fisher’s exact test).
Results: Twelve subjects were enrolled; 11 had TCD data. Mean age was 53 (SD 11) and the majority had moderate to severe OSA (median AHI 27, IQR 16-37). Mean VMR (% change in MCA velocity) was 3.1 (SD 0.7); mean phase shift was 34 degrees (SD 15). There was a moderate association between AHI and phase shift (r=-0.40); the correlation with VMR was weaker (r=-0.25). The proportion of subjects with abnormal DCA was greater among those with moderate-severe OSA compared to those with mild OSA (66.7% vs. 0%, p=0.2). No enrolled subjects had abnormal VMR.
Conclusion: Moderate to severe OSA is associated with abnormal dynamic cerebral autoregulation and normal vasomotor reactivity. The mechanism underlying this dissociation may involve OSA-mediated inflammation and endothelial dysfunction. Further study may clarify how this dissociation relates to increased risk of cerebral ischemia among patients with OSA.