Mild cognitive impairment in obstructive sleep apnea

Abstract

INTRODUCTION Obstructive sleep apnea (OSA) causes sleep disruption and intermittent nocturnal hypoxemia, which can lead to daytime sleepiness and cognitive deficits. Cognitive deficits reported in individuals suffering from OSA can be severe enough to warrant a diagnosis of mild cognitive impairment (MCI). However, the frequency of MCI in individuals with OSA is unknown and factors associated with co-morbid MCI and OSA need to be investigated. Objectives: Determine the frequency and subtypes of MCI in subjects with OSA and identify demographic and clinical variables associated with co-morbid MCI and OSA. Materials and methods Fifty-one subjects with OSA (apnea- hypopnea index 10; mean age: 60.13 9.48 yrs) and 51 healthy controls (mean age: 61.82 9.49 yrs), matched for sex and education, underwent an overnight polysomnography and a comprehensive neuropsychological assessment. MCI was defined as: 1) objective evidence of cognitive decline; and 2) no major impact of cognitive deficits in activities of daily living. MCI subtypes were categorized as nonamnestic single domain, amnestic single domain, nonamnestic multiple domains and amnestic multiple domains. Group difference in the proportion of MCI was assessed using a ÷ test. Subjects with co-morbid OSA and MCI were compared to subjects with OSA alone on respiratory, demographic, questionnaire, and polysomnographic variables using Student t-tests. Results MCI was found in 37% (19/51) of OSA subjects. In contrast, only 13% (7/51) of controls had MCI ( ÷ = 7.43, dl = 1, p 0.01). Eight (42%) of the 19 patients with comorbid OSA and MCI met the criteria for nonamnestic single domain (6 with impaired attention and executive functions and 2 with impaired visuospatial abilities), two (11%) for amnestic single domain, four (21%) for nonamnestic multiple domains and five (26%) for amnestic multiple domains. Subjects with co-morbid OSA and MCI showed a lower education level (t (52) = 5.17, p 0.01) and a higher score on the vascular burden index (t (23) = -2.07, p 0.05) compared to patients with OSA alone. No difference was found between these two groups for respiratory and polysomnographic variables. Conclusion Results showed that OSA subjects have higher risk of developing MCI compared to control subjects. Nonamnestic MCI subtypes with attention/executive dysfunctions were predominant in OSA. Low education level and high vascular burden were also associated with the presence of MCI among OSA subjects. Future studies are needed to assess the effects of co- morbid MCI and OSA on the risk of converting to dementia. Acknowledgements Supported by the Canadian Institutes of Health Research and the Fonds de recherche du Quebec e Sante