Background: The literature about the relationship between obstructive sleep apnea (OSA) and stroke location is conflicting with some studies finding an association and others demonstrating no relationship. Among acute ischemic stroke patients, we sought to examine the relationship between stroke location and the prevalence of OSA; OSA severity based on apnea-hypopnea index (AHI), arousal frequency, and measure of hypoxia; and number of central and obstructive respiratory events.
Methods: Data were obtained from patients who participated in a randomized controlled trial (NCT01446913) that evaluated the effectiveness of a strategy of diagnosing and treating OSA among patients with acute ischemic stroke and transient ischemic attack. Stroke location was classified by brain imaging reports into subdivisions of lobes, subcortical areas, brainstem, cerebellum, and vascular territory. The association between acute stroke location and polysomnographic findings was evaluated using logistic regression for OSA presence and negative binomial regression for AHI.
Results: Among 73 patients with complete polysomnography and stroke location data, 58 (79%) had OSA. In unadjusted models, no stroke location variable was associated with the prevalence or severity of OSA. Similarly, in multivariable modeling, groupings of stroke location were also not associated with OSA presence.
Conclusions: These results indicate that OSA is present in the majority of stroke patients and imply that stroke location cannot be used to identify a group with higher risk of OSA. The results also suggest that OSA likely predated the stroke. Given this high overall prevalence, strong consideration should be given to obtaining polysomnography for all ischemic stroke patients.
Keywords: CT; Infarct; MRI; Sleep apnea; Stroke.
Published by Elsevier B.V.