Long sleep duration, short sleep combined with higher stress, and greater sleep problems, as well as sleep changes over 2 years are associated with smaller volumes of hippocampal subfields, according study findings published in the journal Sleep.
Aging is associated with a reduction in hippocampal volume, but the underlying causes and mechanisms are still poorly understood. In people with sleep disorders, there are negative associations between hippocampal volume and sleep deprivation. Decreased hippocampal volume and dysfunction has also been tied to chronic stress and raised glucocorticoid levels. Yet, the evidence is mixed on the relationship between poor sleep and hippocampal volume and between chronic stress and hippocampal volume.
The objective of the current study was to investigate the relationship between sleep and stress on the hippocampus to better understand the pathological mechanisms in age-related hippocampal atrophy.
Investigators evaluated cross-sectional and 2-year follow-up relationships regarding sleep and stress and total hippocampal volume and hippocampal subfield volumes in older adults. The cross-sectional analyses included data from wave 3 of the Irish Longitudinal Study on Ageing (TILDA). Two-year follow-up analyses were based on data from wave 2 (2012 to 2013) and wave 3 (2014 to 2015).
Participants had a mean (SD) age of 68.8 (7.3) years, 54% were female, and 45% had a tertiary level of education. Those with short sleep duration (≤6 hours/night) were older, more likely to live in an urban area, have 1 or more long-term diseases, and have a history of smoking vs those with normal sleep. Participants with longer sleep (≥9-10 hours/night) were older vs those with normal sleep.
In the cross-sectional analyses, no association was observed regarding hippocampal volume and sleep duration, sleep problems, or stress. The interaction between sleep duration and stress was not significant (P >.05).
Fully adjusted models showed a significant interaction between sleep duration and hippocampal subfield for hippocampal tail (P <.0001) and molecular layer (P =.004), as smaller volumes were found for long sleep vs normal sleep. Greater sleep problems were associated with a smaller volume of cornu ammonis-1 (CA1) (P =.006), hippocampal tail (P =.0001), and molecular layer (P =.002).
Short sleep with higher perceived stress was associated with a smaller volume of CA1 (P <.0001), hippocampal tail (P <.0001), molecular layer (P <.0001), and subiculum (P <.001). The interaction between sleep problems and perceived stress was negatively associated with CA1 (P <.001) volume.
In the 2-year follow-up analyses, no association was found among sleep duration change, sleep problems change, or perceived stress change between waves 2 and 3, as well as total hippocampal volume at wave 3 (P >.05).
The fully adjusted model showed that patients who had long sleep at waves 2 and 3 had smaller hippocampal tail (P <.0001), molecular layer (P <.0001), presibiculum (P =.001), and subiculum (P =.0004) volume vs those with normal-stable sleep. Participants with long sleep at wave 2 but shorter sleep at wave 3 had smaller hippocampal tail volume compared with those who had normal-stable sleep (P <.0001).
An increase in sleep problems from wave 2 to wave 3 was associated with smaller volumes of CA1 (P =.001), hippocampal tail (P =.001), and molecular layer (P =.001). In addition, an increase in stress from wave 2 to wave 3 was marginally associated with a smaller volume of hippocampal tail (P =.01).
Study limitations included the researchers’ reliance on participants’ self-reported measures of perceived stress, sleep duration, and sleep problems, with no additional objective measures. Also, TILDA did not collect information on sleep disorders such as sleep apnea or treatments for other disorders. Furthermore, the study focused on the hippocampal structure and no other brain regions.
“Our findings highlight the importance of concurrently assessing suboptimal sleep and stress in neuropsychiatric research for the development of targeted interventions aiming at reducing the risk of late-life hippocampal atrophy,” the researchers concluded.
De Looze C, Feeney JC, Scarlett S, et al. Sleep duration, sleep problems, and perceived stress are associated with hippocampal subfield volumes in later life: findings from The Irish Longitudinal Study on Ageing. Sleep. Published online September 24, 2021. doi: 10.1093/sleep/zsab241