Disruptions in surrogates of glymphatic clearance and coupled sleep rhythms are jointly associated with AD-related cognitive decline, and combining these metrics effectively predicted disease progression.
Key Findings
Results
Individuals with AD had reduced DTI-ALPS index and BOLD-CSF coupling compared to cognitively normal controls.
Study analyzed data from 75 individuals: 54 with AD and 21 cognitively normal (CN) controls.
Both DTI-ALPS index and BOLD-CSF coupling were significantly reduced in AD (p < 0.05).
DTI-ALPS reflects diffusivity along perivascular spaces as a surrogate of glymphatic function.
BOLD-CSF coupling captures blood oxygen level-dependent signal coupled to CSF signal as a putative glymphatic metric.
Results
Individuals with AD showed disrupted slow oscillation (SO)-spindle coupling compared to cognitively normal controls.
SO-spindle coupling was significantly disrupted in the AD group (p = 0.029).
Coupled sleep rhythms were assessed via slow oscillation (SO)-theta and SO-spindle couplings using sleep EEG.
SO-spindle coupling is a coordinated sleep rhythm thought to support memory consolidation and brain clearance.
Results
Lower global BOLD-CSF coupling was correlated with misaligned SO-theta burst coupling across all participants.
Correlation coefficient r = 0.311, p = 0.018 across all 75 participants.
This association links a putative glymphatic metric with a specific coupled sleep oscillation pattern.
Analysis was performed using correlation analyses across the full sample.
Results
Reduced DTI-ALPS index was associated with misaligned SO-spindle coupling across all participants and within the AD group alone.
Across all participants: r = 0.370, p = 0.008.
Within the AD group specifically: r = 0.376, p = 0.028.
This finding suggests a specific relationship between perivascular space diffusivity and spindle-coupled slow oscillations that persists within the AD population.
Results
Mediation analysis revealed that SO-spindle misalignment contributed to cognitive decline through its effect on DTI-ALPS.
SO-spindle misalignment mediated the relationship between sleep rhythm disruption and cognitive decline via its effect on the DTI-ALPS index.
This suggests a pathway by which disrupted sleep oscillations may impair glymphatic clearance surrogates, leading to cognitive decline.
Cognitive assessments were conducted longitudinally over two years.
Mediation analysis was one of the primary analytical approaches used alongside correlation and LASSO regression.
Results
Combining putative glymphatic metrics and sleep EEG metrics effectively predicted AD progression.
LASSO (least absolute shrinkage and selection operator) regression was used to predict AD progression.
The combined model incorporating both MRI-derived glymphatic surrogates and coupled sleep EEG metrics outperformed individual metrics alone.
Glymphatic metrics included CP volume, perivascular spaces (PVSs), DTI-ALPS index, and BOLD-CSF coupling.
Two-year longitudinal cognitive assessments were used as the outcome measure for AD progression.
Methods
Putative glymphatic function was assessed using four MRI-derived metrics: choroid plexus volume, perivascular spaces, DTI-ALPS index, and BOLD-CSF coupling.
All four metrics serve as surrogate markers of glymphatic clearance rather than direct measures.
DTI-ALPS uses diffusion tensor imaging along the perivascular space.
BOLD-CSF coupling captures blood oxygen level-dependent signal coupled to cerebrospinal fluid signal.
These metrics were correlated with CSF AD biomarkers and sleep EEG data.
What This Means
This research suggests that two types of brain health indicators — measures related to the brain's waste-clearance system (called the glymphatic system) and the coordination of brain wave patterns during sleep — are both disrupted in people with Alzheimer's disease (AD), and that these disruptions are linked to each other and to cognitive decline. The study examined 75 people (54 with AD and 21 healthy controls) using brain scans, sleep brain wave recordings, spinal fluid tests, and cognitive tests tracked over two years. People with AD showed weaker signals on two brain imaging measures of glymphatic activity and had disrupted coordination between specific sleep brain waves called slow oscillations and sleep spindles.
The study also found that poorer glymphatic measures were statistically associated with worse coordination of these coupled sleep rhythms. Importantly, the disruption of sleep spindle coupling appeared to contribute to cognitive decline by affecting the brain's glymphatic activity — suggesting a chain of events where poor sleep rhythms impair the brain's waste-clearance capacity, which in turn worsens cognition. When researchers combined the sleep EEG and brain imaging metrics together in a statistical model, they were better able to predict how much a person's cognition would decline over two years compared to using either type of measure alone.
This research suggests that monitoring both the brain's waste-clearance system and the coordination of sleep brain waves together may offer a more complete picture of Alzheimer's disease progression than either approach alone. These findings point toward the importance of sleep quality — specifically the coordination of specific sleep rhythms — as potentially relevant to the brain processes involved in Alzheimer's disease, and suggest that these combined metrics could serve as useful tools for tracking disease progression in research settings.
Liu X, Wei T, Zhao B, Zhou S, Liu L, Tang Y. (2026). Surrogates of glymphatic metrics decline and coupled sleep rhythms disruption in Alzheimer's disease.. Alzheimer's research & therapy. https://doi.org/10.1186/s13195-026-01962-4