Do Sleep Spindles Depend on the Brain's Wiring?
Sleep spindles help lock in memories and fade before dementia. Do they depend on the brain's long-distance wiring? Mostly not, with one interesting exception.
In one paragraph
Sleep spindles help consolidate memories and decline on the path to dementia. This study of 67 older adults with memory concerns combined an overnight sleep study with brain MRI. People with mild cognitive impairment had fewer fast spindles, but after correction for multiple tests, spindle density and slow-wave power were not linked to connectivity of three white matter tracts. In an exploratory analysis, better connectivity went with bigger spindles in mild cognitive impairment only.
How did researchers link sleep spindles to brain wiring?
Sleep spindles are bursts of activity in non-REM sleep, generated in the thalamus and spread across the cortex through white matter, the brain's long-distance fibre bundles. White matter wears down with age, especially in people at risk of dementia.
At a memory clinic in Sydney, 67 older adults (mean age 68.1) took part: 21 with subjective cognitive decline and 46 with mild cognitive impairment. Each had an overnight in-lab sleep study and a 3T MRI within three months.
The team measured slow and fast spindle density, spindle amplitude and slow oscillation power. Tractography estimated fibre bundle capacity, how much information a tract can carry, for the anterior thalamic radiation and two parts of the superior longitudinal fasciculus. Regressions were adjusted for age and sex, with correction for multiple tests.
Were sleep spindles linked to white matter connectivity?
People with mild cognitive impairment had lower frontal and central fast spindle density than those with subjective cognitive decline, while total sleep time and sleep apnoea did not differ.
After false discovery rate correction, slow spindle density, fast spindle density and slow oscillation power were not significantly linked to connectivity of any tract. A link between central fast spindles and SLF-III explained 8% of variance but did not survive correction.
In an exploratory analysis, greater SLF-II connectivity went with higher spindle amplitude in mild cognitive impairment, explaining 7% (frontal) and 11% (central) of variance, but not in subjective cognitive decline. Removing two outliers eliminated the frontal result.
Why does this mostly null result matter?
Spindle loss is a candidate early marker of cognitive decline, so it matters what drives it. This study suggests that connectivity of these major white matter tracts is not a strong driver of spindle numbers or slow-wave power in older adults at risk.
The authors note that spindle density may reflect how often the thalamus initiates spindles, while amplitude reflects how much of the cortex joins in, which could explain why wiring related to size rather than number in mild cognitive impairment.
A fair reading is that the exploratory amplitude link is a lead for longer studies, not a conclusion. This study doesn't change practice on its own and is not medical advice.
What are the limits of this sleep and white matter study?
Read before you quote it
The study has clear limits:
- Small and cross-sectional. 67 people, measured once, so direction cannot be determined.
- Modest imaging. Diffusion MRI used 32 directions, relatively low for tractography.
- No biomarkers. Alzheimer's pathology was not measured.
- Uncontrolled factors. Sleep apnoea and depressive symptoms were not adjusted for, and one author reports consulting fees from Eisai and Roche.
The study does not show that white matter damage causes spindles to fade, or the reverse. It also does not show that improving sleep protects white matter or memory, or whether other sleep features, such as the coupling of spindles with slow waves, would show different links.
- It does not show that white matter damage causes spindle loss, or the reverse.
- It does not show that improving sleep protects white matter or memory.
- The sample was small, especially within groups, and the design was cross-sectional.
- Diffusion MRI used 32 directions, a relatively low angular resolution for tractography.
- Biomarkers of Alzheimer's pathology were not measured, and sleep apnoea and depressive symptoms were not controlled for.
- Excluding two outliers removed the frontal spindle amplitude result in the mild cognitive impairment group.
Takeaways
- saysPeople with mild cognitive impairment had fewer fast sleep spindles than those with only subjective memory complaints.
- saysAfter correcting for multiple tests, spindle numbers and slow-wave power weren't linked to the capacity of three key white matter cables.
- saysIn an exploratory look, better wiring went with bigger spindles only in mild cognitive impairment, and part of that disappeared without two outliers.
- explainsSleep spindles are short bursts of brain activity during non-REM sleep that help consolidate memories.
- interpretsA fair reading is that spindle loss in early cognitive decline is probably not explained by these major white matter tracts alone.
Ask the paper
In PaperSays you can ask this paper anything, by text or by saying “Hey Aiden”. Three questions listeners asked:
What are sleep spindles and why do they matter?
Bursts of brain activity in non-REM sleep that help memory; they are reduced in mild cognitive impairment. Hear the full answer in the app →
Is brain white matter linked to sleep spindles?
In this study, spindle numbers were not robustly linked to three major tracts after correction for multiple tests. Ask in the app →
What was the exception in this study?
In mild cognitive impairment, better connectivity of one tract went with bigger spindles, in an exploratory analysis. Ask in the app →
Briefing reviewed by an editor before publication. The paper belongs to its authors; this page summarises it and links to the original. Paper licence: CC BY 4.0. Corrections: support@papersays.com.