This study included 115 participants without cognitive impairment, but at higher risk for developing AD. They were part of a larger prospective cohort, the Wisconsin Alzheimer’s Disease Research Center (ADRC) Clinical Core, and 81 members of the cohort had a polysomnography (PSG) with high-density EEG (hdEEG), multiple cognitive assessments related to memory, attention, executive function, etc., markers of the degree of hypoxemia during sleep, as well as measurement of their verbal memory assessed by the Rey Auditory Verbal Learning Test (RAVLT). Participants in this sub-study, termed the Predicting Alzheimer’s from Metabolic Markers and Sleep (PAMMS), were at higher risk for developing AD; 32.5% were APOE4 carriers, 69.1% had a parental history of AD, and 26.3% had both genetic and parental risk. The average age was 61.7, with a range from 44 to 88. Almost half of the cohort had obstructive sleep apnea (OSA), defined as an apnea–hypopnea index (AHI, the number of apneas and hypopneas per hour) of ≥ 5/h, while 16.05% had moderate or severe OSA (AHI ≥ 15/h).
The study's results indicated an association between OSA severity during REM sleep and worse word list learning delayed memory recall (recalling a set of 15 words immediately after being read the words and recalling them 1 [short-term delay] and 20 minutes [long-term delay] after being read a set of 15 interference words). More drops in blood oxygen levels (more oxyhemoglobin desaturations) during REM sleep were also associated with reducing learning performance, especially among participants over 60 as well as APOE4 carriers. APOE4 carriers had poorer recall performance with more respiratory events and arousals. Notably, these associations seemed to be limited to REM sleep, not NREM sleep, and most pronounced in individuals with both a parental history of AD and who also carried an APOE4 allele. The authors conclude that these results support the hypothesis that OSA, particularly during REM sleep, has important consequences on the memory of older adults with other risk factors for AD.
The authors point to evidence that REM sleep may be especially important to memory, such as selectively pruning or strengthening newly formed synapses (needed for memory and learning consolidation) along with the neocortical activation of hippocampal circuits which occurs during REM sleep, also important for memory consolidation. They suggest that OSA during REM may degrade memory both by transient disruptions in cerebral glucose metabolism as well as long-term memory degeneration from repeated bouts of hypoxia. OSA severity has previously been shown to be a significant predictor of Aβ plaque burden in the hippocampus, along with cerebrospinal fluid amyloid β levels among “cognitively normal” older adults. REM sleep is known to have a greater neurometabolic demand in regions impacted by early AD compared to NREM, and OSA during REM has consequences on memory consolidation independent of total sleep duration. OSA during REM has also been linked with other specific pathologies, such as hypertension, suggesting that reduced regional blood flow may be a potential mechanism for the associated pathologies. Oxidative stress, neuroinflammation, blood brain barrier breakdown, and/or endothelial dysfunction were all suggested by the authors as possible contributors.
The authors of the recent publication described above also point out that deficits in verbal memory may be one of the most sensitive markers of the early cognitive changes associated with AD. For example, verbal memory tests have been shown to be quite valuable in distinguishing normal aging from mild cognitive impairment, as well as predicting the likelihood of progression from mild cognitive impairment to AD.
Collectively this suggests that REM may be particularly important for some brain functions, such as memory, and that OSA during REM has significant detrimental consequences, especially among those most at risk for AD. A systematic review published in 2020 in the journal Sleep Medicine Reviews reported that the incidence of OSA among older adults is between 30-80%. They suggest possible variations in the effect of OSA depending on age; among younger adults, OSA is linked to cognitive impairment, while among older adults the results are more variable. Yet in older populations, OSA is more prevalent among people with AD or dementia compared to those with normal cognitive function, and is often associated with the subsequent development of mild cognitive impairment or AD. They also found that CPAP treatment of OSA in both middle-aged and older adults improved sleep parameters as well as cognitive function, an important observation. The authors of the most recent study emphasize the importance of a thorough OSA screening, one capable of differentiating OSA severity by sleep stage, as many cases of REM-specific OSA would otherwise be missed, overlooking the heightened risk for cognitive decline.