Age‐stratified, sex‐specific differences in cognitive performance based on risk of obstructive sleep apnea and systemic inflammation: A cross‐sectional analysis of the Canadian Longitudinal Study of Aging

Abstract

AbstractBackgroundGrowing evidence supports associations between cognitive impairment and obstructive sleep apnea (OSA). Our study aimed at determining the age and sex‐specific, independent relationship between risk of having OSA and cognitive performance, and the influence of systemic inflammation on this relationship.MethodOur sample included 25,712 participants from the Canadian Longitudinal Study of Aging (CLSA) comprehensive cohort aged 45 to 85 years (M = 63 years; 51% women; 78% with post‐secondary degree). Exclusion criteria included Alzheimer’s and Parkinson’s disease and history of stroke and of traumatic brain injury. Participants were classified as high‐ or low‐risk of having OSA based on the STOP questionnaire (high‐risk = ≥ 2/4), considering signs and symptoms associated with OSA including snoring (S), tiredness (T), observation of apneas (O), and high blood pressure (P). Neuropsychological tests assessed memory, executive functioning and psychomotor speed. Levels of high‐sensitivity C‐reactive protein (hs‐CRP) were obtained through blood samples. We conducted sex‐ and age‐specific (45‐59; 60‐69; ≥70) General Linear Models between cognitive scores and risk of OSA, and moderation and mediation analyses with levels of hs‐CRP. Analyses were adjusted for socio‐demographic, lifestyle and medical comorbidities.ResultIn the fully adjusted model, 45‐59 years old women at high‐risk of OSA presented poorer short‐term memory and learning (Rey Auditory Verbal Learning Test, immediate recall, p=0.040), and executive functioning (Animal Fluency test, p=0.011; Mental Alternation test, p=0.006; Stroop Test, inhibition condition, p=0.047; low‐interference index, p=0.027 and high‐interference index, p=0.022) compared to those at low‐risk of OSA. These associations were not found in women aged ≥60 and in men of all age groups. The association between OSA risk and cognition was not moderated by education nor consistently moderated or mediated by the level of hs‐CRP.ConclusionThese findings suggest that OSA might represent a modifiable risk factor of early cognitive impairment in women aged 45‐59. We suggest clinicians systematically screen for cognitive impairment when patients of this group present suspected or diagnosed OSA. Furthermore, it is important to track changes in cognitive profile of women aged 45‐59 in the CLSA cohort to identify those at risk of long‐term cognitive decline.