Greater cognitive deficits with sleep-disordered breathing among individuals with genetic susceptibility to alzheimer disease: The multi-ethnic study of atherosclerosis

Abstract

Rationale: There are conflicting findings regarding the link between sleep apnea and cognitive dysfunction. Objectives: Investigate associations between indicators of sleepdisordered breathing (SDB) and cognitive function in the Multi- Ethnic Study of Atherosclerosis and assess effect modification by the apolipoprotein e-4 (APOE-e4) allele. Methods: A diverse population (N = 1,752) underwent type 2 in-home polysomnography, which included measurement of percentage sleep time less than 90% oxyhemoglobin saturation (%Sat,90%) and apnea-hypopnea index (AHI). Epworth Sleepiness Scale score (ESS) and sleep apnea syndrome (SAS; AHI>5 and ESS.10) were also analyzed. Cognitive outcomes included the Cognitive Abilities Screening Instrument; Digit Symbol Coding (DSC) test; and Digit Span Tests (DST) Forward and Backward. Results: Participants were 45.4% men, aged 68.1 years (SD, 9.1 yr) with a median AHI of 9.0 and mean ESS of 6.0. Approximately 9.7% had SAS, and 26.8% had at least one copy of the APOE-e4 allele. In adjusted analyses, a 1-SD increase in %Sat,90% and ESS score were associated with a poorer attention and memory assessed by the DST Forward score (b =20.12 [SE, 0.06] and b =20.13 [SE, 0.06], respectively; P<0.05). SAS and higher ESS scoreswere also associated with poorer attention and processing speed as measured by the DSC (b =20.69 [SE, 0.35] andb =21.42 [SE, 0.35], respectively;P,0.05). The presence of APOE-e4 allelemodified the associations of %Sat ,90% with DST forward and of ESS with DSC (Pinteraction<0.05). Conclusions: Overnight hypoxemia and sleepiness were associated with cognition. The average effect estimates were small, similar to effect estimates for several other individual dementia risk factors. Associations were strongest in APOE-e4 risk allele carriers. Our results (1) suggest that SDB be considered among a group of modifiable dementia risk factors, and (2) highlight the potential vulnerability of APOE-e4 risk allele carriers with SDB.