Abstract
INTRODUCTION
Sleep disruption, particularly loss of slow‐wave sleep (SWS), is common in Alzheimer's disease (AD), but its neurobiological underpinnings remain unclear. We investigated whether locus coeruleus (LC) integrity relates to SWS across the AD continuum and whether sex and perivascular spaces (PVSs) modify these associations.
METHODS
In a cohort (11 controls, 30 mild cognitive impairment, 17 AD dementia) we combined overnight polysomnography with LC‐sensitive magnetic resonance imaging of the LC, basal ganglia and centrum semiovale PVS ratings, and cerebrospinal fluid noradrenaline. Multivariable linear regression adjusted for demographics, disease stage, and medication use.
RESULTS
Higher LC integrity was associated with greater slow‐wave activity and slow oscillation power, with stronger effects in females. Basal ganglia PVS burden was related to lower SWS spectral power, whereas noradrenaline levels were not associated with sleep.
DISCUSSION
LC integrity, sex, and PVS burden show associations with SWS alterations in aging and AD, supporting restorative sleep as a potential therapeutic target.
Keywords: Alzheimer's disease, locus coeruleus, neuromodulatory subcortical systems, perivascular spaces, sex, sleep
Highlights
Locus coeruleus (LC) integrity is positively associated with slow‐wave sleep (SWS) spectral power.
LC–SWS associations are stronger in females than in males.
Basal ganglia perivascular space burden relates to reduced slow oscillation power.
Noradrenergic and vascular factors may jointly contribute to sleep disruption in Alzheimer's disease.
Polysomnography and neuromelanin‐sensitive magnetic resonance imaging link LC structure to SWS in vivo.
1. BACKGROUND
Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline and memory impairment. Sleep disturbances, particularly slow‐wave sleep (SWS) disruption, are highly prevalent along the AD continuum. 1 , 2 SWS, the deepest phase of non‐rapid eye movement (NREM) sleep, is crucial for memory consolidation and overall cognitive function. 3 , 4 Reduced SWS duration and slow‐wave activity (SWA) have been consistently reported in individuals with preclinical and symptomatic AD and relate to impaired memory. 5 , 6
The locus coeruleus (LC), a brainstem noradrenergic nucleus, is pivotal in regulating arousal, attention, and the sleep–wake cycle. 7 , 8 LC degeneration is one of the earliest pathological changes in AD and may precede significant cognitive deficits. 9 , 10 , 11 Post mortem analyses show progressive LC volume loss in AD (up to 8.4% per Braak stage), with marked neuronal loss only from mid Braak stages onward. 12
The LC's extensive projections release noradrenaline (NA) throughout the brain, modulating various functions, including sleep regulation. 7 Disruption of the LC–NA system due to LC degeneration may contribute to the sleep disturbances observed in AD patients. 8 , 13 Complementary evidence from animal models focuses on SWS itself: studies using transgenic mice have demonstrated that disruptions in SWA can influence amyloid beta (Aβ) accumulation and particularly increased tau hyperphosphorylation, exacerbating AD‐like neuropathology and associated behavioral symptoms. 5 , 6
Human studies have also highlighted the association between LC integrity and sleep patterns. 8 Neuromelanin‐sensitive magnetic resonance imaging (MRI), a technique used to assess LC integrity in vivo, has shown reduced LC contrast in individuals with AD, correlating with cognitive decline and sleep disturbances. 5 , 6 , 14 Furthermore, post mortem examinations have identified early tau pathology within the LC, suggesting a mechanistic link between LC degeneration and the onset of sleep–wake alterations and neuropsychiatric symptoms in AD. 8 , 15
Despite these findings, the precise relationship between LC degeneration and SWS alterations in AD remains incompletely understood. 16 While post mortem studies and animal models have provided valuable insights, there is a paucity of in vivo data directly linking LC integrity to SWS changes in human subjects with AD. Understanding this relationship is critical, as interventions to preserve LC function or enhance SWS may offer therapeutic potential to mitigate sleep disturbances and potentially slow cognitive decline. 17 , 18
RESEARCH IN CONTEXT
Systematic review: We searched PubMed for human studies examining locus coeruleus (LC) integrity, polysomnography‐based slow‐wave sleep (SWS), and magnetic resonance imaging–visible perivascular spaces (PVSs) in Alzheimer's disease (AD). Few imaging studies linked LC structure to sleep, and none jointly assessed LC, SWS spectral markers, sex, and PVS burden across the AD continuum.
Interpretation: In cognitively unimpaired individuals and patients with biomarker‐confirmed AD, higher LC integrity was associated with greater slow‐wave activity and slow oscillation power, especially in females, whereas basal ganglia PVS burden related to reduced SWS spectral power. These findings support complementary noradrenergic and vascular–glymphatic contributions to sleep disruption in aging and AD.
Future directions: Larger, longitudinal, multicenter studies should determine whether LC–SWS alterations and PVS burden predict cognitive decline or treatment response, and test whether interventions targeting noradrenergic tone, vascular health, or SWS enhancement can preserve restorative sleep and modify AD trajectories.
In addition, sex differences in sleep physiology have been documented and may have important implications for AD. 19 , 20 In healthy adults, particularly before menopause, females generally show higher SWA and longer durations of deep NREM sleep than males. 21 , 22 , 23 These differences tend to diminish with age and hormonal changes, and may influence how LC degeneration impacts SWS and downstream neurodegenerative processes. 24 , 25 Understanding how sex modulates the relationship between LC integrity and SWS is therefore crucial for elucidating mechanisms of sleep disruption in AD.
Vascular and glymphatic mechanisms also contribute to sleep‐dependent brain homeostasis. 26 , 27 Perivascular spaces (PVSs), fluid‐filled channels surrounding cerebral vessels, are central to interstitial fluid drainage and the clearance of neurotoxic proteins such as Aβ and tau. 28 Enlargement of PVS on MRI, particularly in the basal ganglia (BG) and centrum semiovale (CSO), reflects small‐vessel dysfunction and may be associated with impaired glymphatic clearance. 29 SWS enhances convective flow along PVSs, facilitating metabolic waste removal; disruptions in SWS or PVS enlargement may therefore exacerbate protein accumulation and neurodegeneration. 30 , 31 Evaluating the interplay among sleep, LC integrity, sex, and PVS burden may clarify how vascular–glymphatic dysfunction and noradrenergic degeneration contribute to SWS alterations in aging and AD.
In summary, existing literature highlights the critical role of the LC in sleep regulation and its early degeneration in AD, as well as sex differences in SWS and the contribution of vascular–glymphatic mechanisms. 27 However, the in vivo relationships among LC integrity, sex, PVS burden, and SWS alterations remain poorly understood.
Our primary objective was to investigate whether LC integrity relates to quantitative SWS measures in aging and AD, and whether these associations differ by sex. In exploratory analyses, we further assessed how PVS burden, an index of vascular/glymphatic alterations, and cerebrospinal fluid (CSF) NA levels, a marker of the noradrenergic milieu, relate to SWS measures and LC integrity.
2. METHODS
2.1. Participants
We recruited 58 participants within an age range of 50 to 80 years old who were classified into three groups based on Clinical Dementia Rating (CDR) and biomarker positivity:
Healthy controls (n = 11): CDR 0 and performance on the neuropsychological testing within the normal range and normal phosphorylated tau (p‐tau)217 plasma levels.
Mild cognitive impairment (MCI, n = 30) due to AD: CDR 0.5, meeting the criteria of biomarker‐based AD diagnosis in agreement with the Alzheimer's Association Workgroup diagnostic criteria. 32
Dementia due to AD (n = 17): CDR 1, meeting the biomarker‐based diagnostic criteria for AD following Alzheimer's Association Workgroup guidelines. 32
All AD participants were classified based on amyloid positivity, determined by CSF biomarkers (n = 54) or amyloid positron emission tomography (PET; n = 4).
2.2. Sleep assessment
2.2.1. Polysomnography
All participants underwent overnight video‐polysomnography (PSG) performed at the Sleep Unit at the Hospital Clínic de Barcelona. These studies consisted of vertical and horizontal electrooculography, electroencephalography (O2‐A1, O1‐A2, C4‐A1, C3‐A2, F4‐A1, F3‐A1), and surface electromyography of the mentalis muscle, the right and left flexor digitorum superficialis muscles in the upper limbs, and the right and left tibialis anterior muscles in the lower limbs. Additional measurements included electrocardiography, nasal and oral airflow with oral thermistors and nasal cannula, thoracic and abdominal respiratory effort, oxygen saturation assessed by a finger pulse oximeter, microphone, and digital synchronized audiovisual monitoring. Sleep stages (% Wake, NREM N1, N2, N3, and REM) during the sleep period time (SPT) were scored following the American Academy of Sleep Medicine guidelines. 33 Apnea and hypopneas during sleep were also scored according to this guideline and the apnea–hypopnea index (AHI) calculated. NREM stages N1, N2, and N3 represent progressively deeper stages of NREM sleep.
2.2.2. SWA parameters
Sleep parameters, including power in the slow oscillation (SO, 0.5–1 Hz), delta (1–4 Hz), and SWA (0.5–4 Hz) frequency bands, were computed from deep NREM sleep stages (N2 and N3) using PSG frontal electrodes (F3 and F4).
After bandpass filtering the electroencephalography signals in each channel from 0.5 to 35 Hz, the power spectral density of each 30 second epoch classified as either N2 or N3 and free from arousals or artifacts was computed for each channel using the Fourier transform. Then, the average of all the epochs across the night was computed for each channel and finally averaged between channels F3 and F4. Last, the area under the curve of the frontal average power spectral density was computed at each frequency range.
2.3. LC integrity in MRI
2.3.1. MRI acquisition
All participants were scanned in a 3T Siemens PRISMA‐FIT System using a 20‐channel head coil, at the Magnetic Resonance Imaging Core Facility, Institut d'Investigacions Biomèdiques August Pi i Sunyer, located at the Hospital Clínic de Barcelona. The MRI protocol included the acquisition of the following sequences:
3D T1‐weighted magnetization prepared gradient echo (MPRAGE) sequence (repetition time [TR] = 2300 ms, echo time [TE] = 2.98 ms, inversion time [TI] = 900 ms, flip angle = 9°, bandwidth = 240 Hz/pixel, acquisition matrix = 256 × 256 × 240, isometric voxel size = 1 mm3).
LC‐ sensitive high resolution, T1‐weighted turbo‐spin echo (TSE) sequence: aligned perpendicularly to the plane of the respective participant's brainstem (acquisition time = 10.33 min) with the following parameters: TR = 600 ms, TE = 11 ms, flip angle = 120°, bandwidth = 180 Hz/pixel, acquisition matrix = 320 × 320 × 16, voxel size = 0.5 × 0.5 × 1.8 mm3, number of averages 7. TSE consists of 16 slices without gap, covering the pons.
2.3.2. Post‐processing
The acquired images were processed to estimate LC integrity as previously detailed by Falgàs et al. 10 , 34 Three areas of interest were identified in the Montreal Neurological Institute (MNI) template: including right and left LC as defined by the 2 standard deviation LC (2D‐LC) map, and a dorsal pontine reference region. Elastic registration between the MNI template and each participant's T1‐weighted image was performed to identify these three areas in each subject image. The integrity of LC was quantified based on the methodology described by Dahl et al., 10 that is, the ratio between maximum intensity in the 2D‐LC region and the reference area intensity averaged along the longitudinal axis.
Additionally, we analyzed the dorsal raphe nucleus (DRN) volumes (mm2) as a control region. Volumes were extracted using the subcortical atlas and methodology described by Rolls et al. 35
2.4. MRI PVSs assessment
PVSs were assessed by an experienced vascular neurologist (S.R.), blinded to clinical and sleep data, on T1‐weighted MRI images using a validated visual rating scale. 36 Separate scores were assessed for PVS. in the basal ganglia (BG‐PVS) and centrum semiovale (CSO‐PVS), according to the number of PVS. in the region of interest (0 = none, 1 = 1–10, 2 = 11–20, 3 = 21–40, and 4 = > 40 PVSs). Ratings were performed using reference images to ensure consistency and reliability. PVSs in BG and CSO were treated separately for all analyses to allow region‐specific correlations with SWS parameters.
2.5. CSF measurements
2.5.1. AD biomarkers
Lumbar punctures were performed in the morning to collect CSF samples. Levels of CSF Aβ (Aβ42), total tau (t‐tau), and p‐tau were measured using Lumipulse G enzyme‐linked immunosorbent assays, following the manufacturer's instructions (Fujirebio). Each CSF biomarker's cut‐off values for abnormality were defined based on internal controls: Aβ42 ≤ 600 pg/mL, t‐tau > 385 pg/mL, and p‐tau > 65 pg/mL.
2.5.2. CSF NA
NA was quantified using a plasma high‐performance liquid chromatography kit (Chromsystems) after the analytes were extracted from the CSF matrix through adsorption onto alumina.
2.6. Plasma biomarkers
2.6.1. Plasma p‐tau 217
Blood samples were obtained from all participants at the first visit, centrifuged to obtain plasma, aliquoted, and stored at −80°C. Plasma biomarker concentrations were measured with the plasma p‐tau217 levels, which were determined with a commercially available Lumipulse G pTau 217 (Fujirebio) assay.
2.7. Statistical analyses
Multiple group comparisons (healthy controls, MCI due to AD, and dementia due to AD) of continuous variables were performed using Kruskal–Wallis tests with post hoc Dunn tests. Categorical variables were compared using χ 2 tests. Pearson correlations and linear regression models were used to analyze the effects of LC integrity/NA on SWA, SO, delta activity, and N3 sleep %, controlling for disease stage (CDR score), age, sex, and prescribed medications. Correlations are presented as descriptive, unadjusted associations, whereas the regression models are the main hypothesis‐driven, covariate‐adjusted analyses. We additionally tested the interaction between LC integrity and sex and the effect of AHI.
In a second step, we evaluated the effect of BG‐PVS and CSO‐PVS on SWA, SO, delta activity, and N3 sleep % using reduced models that excluded non‐significant predictors of the previous model and the interaction between LC integrity and BG‐PVS. Model fit between the complete model and the reduced model was assessed by comparing Akaike's information criterion (AIC) and Bayesian information criterion (BIC), with lower values indicating better fit. All statistical analyses and figure generation were performed using Stata/IC 14.2. Statistical significance was set at p < 0.05.
3. RESULTS
3.1. Demographics, biomarkers, and MRI
The MCI and dementia due to AD groups were slightly older than healthy controls (p < 0.05), with no significant differences in age at onset (AAO) between the AD groups. The distribution of sex was balanced across all groups. Mini‐Mental State Examination scores were significantly lower in both MCI and dementia due to AD compared to healthy controls (p < 0.01), as expected. Sleep medication and antidepressant use were significantly higher in the MCI group compared to healthy controls (p < 0.05). As expected, plasma p‐tau217 levels were elevated in both AD groups (p < 0.01). CSF biomarkers for AD were similar across the AD groups. No significant differences were observed in LC integrity or CSF NA levels between the groups (Table 1).
TABLE 1.
Demographics, clinical, and neurophysiological data groups.
| Variable | Healthy controls (CDR 0, n = 11) | MCI due to AD (CDR 0.5, n = 30) | Dementia due to AD (CDR 1, n = 17) | MCI vs. HC | AD vs. HC | MCI vs. AD |
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Age | 61.55 (7.70) | 67.97 (6.16) | 70.59 (7.72) | 0.0154 | 0.0008 | 0.0640 |
| AAO | – | 65.23 (6.76) | 67.94 (7.73) | – | – | 0.1084 |
| MMSE | 29 (1.10) | 25.86 (2.74) | 21.71 (3.10) | 0.0016 | 0.0000 | 0.0004 |
| Sex (% female) | 45.5 | 46.7 | 70.6 | 0.4728 | 0.0984 | 0.0587 |
| Sleep medication (%) | 9 | 36.7 | 11.8 | 0.0350 | 0.4364 | 0.0287 |
| Antidepressants (%) | 18 | 50 | 35 | 0.0337 | 0.1851 | 0.1631 |
| AD biomarkers | ||||||
| CSF Aβ42 (pg/mL) | – | 410.43 (120.23) | 405.79 (172.90) | – | – | 0.5402 |
| CSF t‐tau (pg/mL) | – | 576.79 (196.51) | 623.57 (255.33) | – | – | 0.2573 |
| CSF p‐tau (pg/mL) | – | 105.46 (87.77) | 99.14 (42.91) | – | – | 0.5995 |
| Plasma p‐tau 217 (pg/mL) | 0.12 (0.07) (n = 11) | 0.62 (0.44) (n = 20) | 0.90 (0.53) (n = 8) | 0.0001 | 0.0001 | 0.1974 |
| Noradrenergic measures | ||||||
| LC integrity (ratio) | 0.110 (0.027) | 0.111 (0.024) | 0.108 (0.024) | 0.3387 | 0.3017 | 0.4289 |
| Noradrenaline (pg/mL) | – | 124.88 (86.88) | 114.25 (73.17) | – | – | 0.6455 |
| Perivascular spaces | ||||||
| BG PVS score | 0.471 | 0.359 | 0.354 | |||
| 1 (%) | 64 | 68 | 56 | |||
| 2 (%) | 36 | 18 | 44 | |||
| 3 (%) | 0 | 14 | 0 | |||
| CSO PVS score | 0.079 | 0.423 | 0.131 | |||
| 1 (%) | 73 | 43 | 44 | |||
| 2 (%) | 9 | 29 | 31 | |||
| 3 (%) | 9 | 18 | 19 | |||
| 4 (%) | 9 | 10 | 6 | |||
| Polysomnography | ||||||
| Wake over SPT (%) | 15.45 (10.36) | 26.94 (17.33) | 21.01 (14.20) | 0.0307 | 0.1816 | 0.1556 |
| N1 over SPT (%) | 11.75 (5.00) | 12.53 (6.67) | 16.73 (9.56) | 0.4330 | 0.1018 | 0.0771 |
| N2 over SPT (%) | 41.55 (10.51) | 35.72 (15.53) | 41.26 (14.02) | 0.1405 | 0.4339 | 0.0716 |
| N3 over SPT (%) | 15.22 (8.13) | 10.83 (8.06) | 6.84 (7.74) | 0.1012 | 0.0064 | 0.0454 |
| REM over SPT (%) | 16.04 (3.56) | 11.05 (6.11) | 15.05 (6.23) | 0.0090 | 0.2695 | 0.0249 |
| AHI | 11.11 (10.22) | 16.81 (16.68) | 24.3 (15.81) | 0.1932 | 0.0126 | 0.0323 |
| Slow wave sleep | ||||||
| SO power (log10) | 3.24 (0.27) | 3.16 (0.24) | 3.07 (0.15) | 0.3420 | 0.0331 | 0.0308 |
| Delta power (log10) | 3.15 (0.28) | 3.06 (0.23) | 2.91 (0.16) | 0.2717 | 0.0078 | 0.0088 |
| SWA power (log10) | 3.50 (0.27) | 3.42 (0.23) | 3.30 (0.14) | 0.3054 | 0.0097 | 0.0084 |
Note: Results are presented as mean (standard deviation) or percentage (%). The table includes demographic and clinical data and neurophysiological measurements by groups (healthy controls, MCI, AD). Power values for SO, delta, and SWA during NREM sleep are log‐transformed. Sleep medications included lorazepam and quetiapine; antidepressants included selective serotonin reuptake inhibitors, trazodone, mirtazapine.
Statistically significant p values are shown in bold.
Abbreviations: AAO, age at onset; Aβ, amyloid beta; AD, Alzheimer's disease; AHI, apnea–hypopnea index; BG, basal ganglia; CDR, Clinical Dementia Rating; CSF, cerebrospinal fluid; CSO, centrum semiovale; HC, healthy control; LC, locus coeruleus; MCI, mild cognitive impairment; MMSE, Mini‐Mental State Examination; NREM, non–rapid eye movement; p‐tau, phosphorylated tau; PVS, perivascular space; REM rapid eye movement; SO, slow oscillation; SPT, sleep period time; SWA, slow‐wave activity; SWS, slow‐wave sleep.
3.2. Sleep measures
PSG revealed higher wake time during SPT in MCI compared to healthy controls (p < 0.05). In dementia due to AD, there was a significantly lower N3 sleep % compared to healthy controls and MCI (p < 0.01 and p < 0.05, respectively). Also, REM sleep was lower in MCI (p < 0.05). Regarding SWS, SO, delta, and SWA power were lower in dementia due to AD compared to healthy controls (p < 0.05, p < 0.01, and p < 0.01, respectively), and MCI (p < 0.05, p < 0.01, and p < 0.01, respectively). For more details see Table 1.
3.3. Correlations between LC integrity and SWS
Correlation analyses showed a positive association between LC integrity and both SWA power (r = 0.27, p = 0.043) and SO power (r = 0.29, p = 0.028; Figure 1).
FIGURE 1.
Correlation between LC integrity and SWS measurements. Correlation analyses revealed a statistically significant positive association between LC integrity and both (A) SWA power (r = 0.27, p = 0.043) and (B) SO power (r = 0.29, p = 0.028). No statistically significant correlation was observed between LC integrity and (C) delta power (r = 0.20, p = 0.11). LC, locus coeruleus; SO, slow oscillation; SWA, slow‐wave activity; SWS, slow‐wave sleep.
3.4. Regression models testing LC × sex interactions on SWS measures
Lineal regression models including sex × LC interaction showed that LC integrity was positively associated with SO power (β = 0.632, p = 0.001) and SWA power (β = 0.532, p = 0.003), and showed a trend for delta power (β = 0.338, p = 0.064). However, significant sex × LC integrity interactions emerged for SO (β = −1.481, p = 0.008) and SWA power (β = −1.130, p = 0.039), indicating that the positive association between LC integrity and SWA/SO was stronger in females compared to males. No interaction effect was observed for delta power (p = 0.331) or N3 percentage (p = 0.132). Independent of LC integrity, age was inversely associated with delta power (β = −0.287, p = 0.026), while sex had a main effect on N3, with males showing lower N3 percentage (β = −1.201, p = 0.031). No significant associations were found for CDR or sleep medication use across the models (see Table 2 and Figure 2). The linear regression model without including interaction terms can be found in Table S1a in supporting information. Also, as a control for anatomical specificity, we repeated the SWS models using dorsal raphe (mm2) instead of LC integrity; dorsal raphe volume was not significantly associated with SO, delta, or SWA power (all p > 0.4; see Table S1b). Additional models adjusting for AHI showed no statistically significant association between AHI and SWS (Table S2 in supporting information).
TABLE 2.
Regression coefficients from linear regression models accounting for the interaction between LC integrity and sex. Regression coefficients from linear regression models assessing the effect of LC integrity levels over SWA, SO, and Delta power.
| Dependent variable | Explanatory variables | β | t | p value |
|---|---|---|---|---|
| SO power (log10) | LC integrity | 0.6320 | 3.67 | 0.001 |
| Age | −0.1174 | −0.98 | 0.333 | |
| Sex (0 = female, 1 = male) | 1.0200 | 1.96 | 0.056 | |
| Sex × LC integrity interaction | −1.4813 | −2.77 | 0.008 | |
| CDR | −0.1227 | −1.00 | 0.321 | |
| Sleep medications | 0.0900 | 0.57 | 0.568 | |
| Antidepressants | −0.1202 | −0.77 | 0.447 | |
| _cons | – | 10.22 | <0.001 | |
| Delta power (log10) | LC integrity | 0.3383 | 1.89 | 0.064 |
| Age | −0.2866 | −2.30 | 0.026 | |
| Sex (0 = female, 1 = male) | 0.1646 | 0.30 | 0.762 | |
| Sex × LC integrity interaction | −0.5442 | −0.98 | 0.331 | |
| CDR | −0.1578 | −1.24 | 0.220 | |
| Sleep medications | 0.0320 | 0.20 | 0.845 | |
| Antidepressants | −0.1549 | −0.95 | 0.346 | |
| _cons | – | 11.38 | <0.001 | |
| SWA power (log10) | LC integrity | 0.5318 | 3.09 | 0.003 |
| Age | −0.2061 | −1.72 | 0.092 | |
| Sex (0 = female, 1 = male) | 0.6781 | 1.30 | 0.199 | |
| Sex × LC integrity interaction | −1.1298 | −2.12 | 0.039 | |
| CDR | −0.1414 | −1.16 | 0.253 | |
| Sleep medications | 0.0607 | 0.39 | 0.700 | |
| Antidepressants | −0.1404 | −0.90 | 0.375 | |
| _cons | – | 12.48 | <0.001 | |
| N3 over SPT (%) | LC integrity | −0.2339 | −1.31 | 0.197 |
| Age | −0.1326 | −1.06 | 0.293 | |
| Sex (0 = female, 1 = male) | −1.2010 | −2.22 | 0.031 | |
| Sex × LC integrity interaction | 0.8505 | 1.53 | 0.132 | |
| CDR | −0.1835 | −1.44 | 0.155 | |
| Sleep medications | 0.1817 | 1.11 | 0.270 | |
| Antidepressants | −0.3293 | −2.02 | 0.049 | |
| _cons | – | 3.41 | 0.001 |
Note: Statistically significant p values are shown in bold.
Abbreviations: CDR, Clinical Dementia Rating; LC, locus coeruleus; NREM, non–rapid eye movement; p‐tau, phosphorylated tau; SO, slow oscillation; SPT, sleep period time; SWA, slow‐wave activity.
FIGURE 2.
Effect of LC integrity on SWA measurements accounting for sex. A significant sex × LC integrity interaction emerged for (A) SWA power (β = −1.130, P = 0.039) and (B) SO (β = −1.481, P = 0.008), indicating that the positive association between LC integrity and SWA/SO was stronger in females compared to males. No interaction effect was observed for (C) delta power. LC, locus coeruleus; SO, slow oscillation; SWA, slow‐wave activity.
3.5. Regression models evaluating the effect of LC integrity and SWS, accounting for MRI perivascular spaces
To evaluate the associations of MRI‐visible PVS load with SWS (SWA, SO, delta) and N3%, we first fitted full linear regression models including LC integrity, sex, age, disease stage, prescribed medications, and interaction terms (LC × sex, LC × BG‐PVS). To improve parsimony, we then fitted reduced models excluding non‐significant covariates (disease stage and prescribed medications). Both AIC and BIC consistently favored the reduced models, which were therefore used to evaluate the association between PVS scores and sleep parameters specifically (Table S3 in supporting information).
3.6. BG PVS
Using the reduced models, BG PVSs were negatively associated with SWA power (β = −1.092, p = 0.034) and SO power (β = −1.125, p = 0.030; see Table 3). Negative trends were observed for delta power (β = −0.923, p = 0.085) and N3% SPT (β = −0.995, p = 0.095), though these did not reach statistical significance. Exploratory analyses of the LC × BG‐PVS interactions showed trends for SWA and SO (contrast 2: β = 0.677, p = 0.083; β = 0.714, p = 0.069), suggesting that perivascular spaces may modulate the effect of LC integrity on SWA; however, none of these interactions reached significance.
TABLE 3.
Regression coefficients from linear regression models accounting for the MRI basal ganglia perivascular spaces. Regression models evaluating the effect of LC integrity and SWS, accounting for sex and MRI perivascular spaces.
| Dependent variable | Explanatory variables | β | t | p value |
|---|---|---|---|---|
| SO power (log10) | LC integrity | 0.417 | 2.16 | 0.036 |
| Age | −0.060 | −0.55 | 0.582 | |
| Sex (0 = female, 1 = male) | 0.657 | 1.32 | 0.192 | |
| Sex × LC integrity | −1.204 | −2.35 | 0.023 | |
| BG_PVS | −1.125 | −2.24 | 0.030 | |
| BG_PVS × LC integrity(contrast 2) | 0.714 | 1.86 | 0.069 | |
| BG_PVS × LC integrity(contrast 3) | 0.599 | 1.47 | 0.148 | |
| _cons | 9.89 | 0.000 | ||
| SWA power (log10) | LC integrity | 0.332 | 1.72 | 0.091 |
| Age | −0.153 | −1.41 | 0.164 | |
| Sex (0 = female, 1 = male) | 0.348 | 0.70 | 0.485 | |
| Sex × LC integrity | −0.881 | −1.73 | 0.091 | |
| BG_PVS | −1.092 | −2.18 | 0.034 | |
| BG_PVS × LC integrity(contrast 2) | 0.677 | 1.77 | 0.083 | |
| BG_PVS × LC integrity(contrast 3) | 0.555 | 1.37 | 0.178 | |
| _cons | – | 11.37 | 0.000 | |
| Delta power (log10) | LC integrity | 0.221 | 1.10 | 0.279 |
| Age | −0.247 | −2.17 | 0.035 | |
| Sex (0 = female, 1 = male) | −0.007 | −0.01 | 0.989 | |
| Sex × LC integrity | −0.462 | −0.86 | 0.393 | |
| BG_PVS | −0.923 | −1.76 | 0.085 | |
| BG_PVS × LC integrity(contrast 2) | 0.555 | 1.38 | 0.174 | |
| BG_PVS × LC integrity(contrast 3) | 0.430 | 1.01 | 0.319 | |
| _cons | – | 9.71 | 0.000 | |
| N3% SPT | LC integrity | −0.300 | −1.33 | 0.188 |
| Age | −0.166 | −1.31 | 0.196 | |
| Sex (0 = female, 1 = male) | −1.201 | −2.08 | 0.043 | |
| Sex × LC integrity | 0.734 | 1.23 | 0.224 | |
| BG_PVS | −0.995 | −1.71 | 0.095 | |
| BG_PVS × LC integrity(contrast 2) | 0.635 | 1.42 | 0.161 | |
| BG_PVS × LC integrity(contrast 3) | 0.835 | 1.76 | 0.085 | |
| _cons | – | 3.44 | 0.001 |
Note: Statistically significant p values are shown in bold.
Abbreviations: BG, basal ganglia; LC, locus coeruleus; MRI, magnetic resonance imaging; PVS, perivascular space; SO, slow oscillation; SPT, sleep period time; SWA, slow‐wave activity; SWS, slow‐wave sleep.
3.7. CSO PVS
In the reduced models including LC integrity, age, sex, and LC × sex interactions, CSO PVSs did not show significant associations with any sleep measures. Specifically, CSO‐PVS coefficients were small and non‐significant for SO power (β = −0.045, p = 0.699), SWA power (β = −0.006, p = 0.962), delta power (β = 0.038, p = 0.747), and N3% SPT (β = 0.024, p = 0.846; Table S4 in supporting information).
3.8. Regression models evaluating the effect of CSF NA and SWS
No significant association was found between SWA, SO, delta power, and CSF NA levels (Table S5 in supporting information).
4. DISCUSSION
This study provides novel in vivo evidence of associations between LC integrity and SWS parameters across the AD continuum, with sex‐specific effects. Using PSG and LC‐sensitive MRI in biomarker‐confirmed AD and healthy controls, we found that higher LC integrity was positively associated with SWA and, particularly, SO power, independent of disease stage, age, and medication use. These associations were more pronounced in females, suggesting potential sex‐dependent mechanisms in the relationship between LC integrity and deep sleep. Additionally, exploratory analyses showed that BG‐PVSs were negatively associated with SWA and SO, suggesting that vascular/glymphatic factors may also relate to sleep microstructure in AD.
Our results extend previous work linking LC degeneration to sleep alterations in aging and AD. 19 Post mortem studies have consistently shown early LC neuronal loss and tau pathology in AD, while animal models demonstrate that LC‐driven norepinephrine release is crucial for sleep–wake regulation and amyloid/tau dynamics. 9 , 11 , 17 In vivo human studies using neuromelanin‐sensitive MRI have associated reduced LC integrity with cognitive decline and alterations in sleep–wake cycles, such as nighttime awakenings, as measured by actigraphy or other macrostructural indices. 13 However, the direct relationship between LC integrity and PSG‐derived SWS measures has scarcely been investigated, and most mechanistic insights come from animal models rather than humans. 6 Our findings add to this literature by linking LC integrity with SWA/SO power in vivo, pointing to a potential involvement of LC‐related noradrenergic function in restorative sleep that should be further examined in longitudinal and mechanistic studies. However, supporting the anatomical specificity of these effects, similar models using dorsal raphe volume (a monoaminergic brainstem nucleus not primarily implicated in SWS regulation) as a control region did not show significant associations with SWA, SO, or delta power.
Our results further indicate that the LC–SWA association is primarily driven by SO power rather than broadband delta activity. This specificity is mechanistically plausible, as SO (<1 Hz) represents large‐scale cortical down–up state transitions that require finely tuned subcortical neuromodulation, whereas faster delta oscillations (1–4 Hz) are less dependent on ascending noradrenergic input and may be generated through more local thalamocortical dynamics. 37 , 38 The selective association of LC integrity with SO highlights its link to physiologically meaningful deep sleep, suggesting that noradrenergic dysfunction in AD selectively compromises global cortical synchronization rather than uniformly affecting all slow‐frequency activity. This distinction has important implications: first, SO are critical for coordinating sleep spindles and hippocampo–cortical communication, processes central to memory consolidation; 39 , 40 second, SO have been linked to glymphatic clearance efficiency, meaning that their disruption may exacerbate amyloid and tau accumulation. 27 , 41
Interestingly, LC integrity was not significantly associated with N3 sleep percentage, despite its robust association with spectral markers of SWS (SWA and SO), suggesting that spectral measures may more sensitively capture LC‐related modulation than conventional stage scoring. In this relatively small, single‐center sample, we also did not detect statistically significant differences in LC integrity across diagnostic groups. This contrasts with our previous study in a larger cohort, in which LC integrity was lower in patients with mild–moderate AD dementia than in healthy controls, whereas no significant differences were found between controls and MCI; the smaller size and earlier stage profile of the present sample (with a higher proportion of MCI and only mild dementia) likely reduced our power to detect such group differences in the present study. Consequently, the LC–SWS associations observed here should be interpreted as operating across cognitively normal individuals and patients along the AD continuum and, in this cohort, cannot fully explain the group‐level SWS deficits observed in AD dementia compared to controls. In this context, LC integrity is best viewed as a transdiagnostic determinant of SWS in late life. Nonetheless, converging evidence from larger neuromelanin‐sensitive MRI cohorts, including our own prior work, as well as post mortem and experimental studies, shows that LC degeneration is an early and robust feature of AD and relates to disease burden and network dysfunction; 34 within this broader context, our findings complement previous work by illustrating how individual differences in LC integrity, irrespective of clinical stage, map onto the quality of deep sleep and may render LC degeneration particularly detrimental for SWS when superimposed on aging.
We observed stronger LC–SWS associations in females than males, consistent with epidemiological data showing greater SWS and SWA in women. 21 This may reflect the presence of estrogen receptors (ERα and ERβ) in the LC, mediating neuroprotective effects on noradrenergic neurons. 25 , 42 Estrogen supports LC structure and function, helping maintain sleep architecture and cognitive health, whereas its loss during menopause may increase LC vulnerability to neurodegeneration, noradrenergic dysfunction, and sleep disturbances. 25 , 43 , 44 These factors may contribute to the higher AD risk observed in postmenopausal women, as sleep disruption and LC degeneration are implicated in AD pathogenesis. Sex differences in LC structure and hormonal regulation of sleep have been described, and our findings suggest that females may rely more on preserved LC function to sustain SWS. 45 , 46 These results underscore the importance of sex‐stratified approaches in biomarker studies and the design of interventions aimed at preserving sleep and noradrenergic function. Further work is needed to understand these mechanisms in detail.
We also found that BG‐PVSs were negatively associated with SWA and SO. Enlarged PVSs are a marker of small vessel disease, also considered linked to impaired glymphatic function, both of which are closely related to AD pathology. 47 Experimental evidence shows that SWS enhances glymphatic clearance of Aβ and tau, processes dependent on perivascular fluid dynamics. 48 Our findings suggest that LC integrity is associated with greater SWS, whereas BG‐PVS burden is linked to reduced SWS. These independent associations may have downstream relevance for clearance mechanisms and AD pathology. The regional specificity of PVS effects (BG but not CSO) may relate to differences in vascular architecture and susceptibility to small‐vessel disease in deep versus superficial brain regions. 25 Although PVS dilation caused by vascular amyloid deposition in cerebral amyloid angiopathy primarily affects the perforating arteries of the leptomeningeal circulation, the relationships with regional PVS load in AD show considerable variability. 49 Additionally, the mechanisms underlying interstitial fluid drainage remain poorly understood. Although exploratory analyses suggested possible interactions between LC integrity and BG‐PVSs in shaping SWS, these did not reach statistical significance. Future studies with larger samples are needed to clarify whether noradrenergic and vascular mechanisms converge synergistically to modulate slow‐wave dynamics.
CSF NA levels were not associated with SWS parameters. This finding aligns with our previous work, in which we observed that LC integrity was associated with neuropsychiatric symptoms, but no parallel associations emerged with CSF NA. 34 Together, these findings suggest that structural measures of LC integrity may provide a more reliable marker of noradrenergic dysfunction in AD than CSF indices. One possible explanation is that CSF NA reflects a static, extracellular snapshot that may not capture the dynamic fluctuations of noradrenergic activity across sleep–wake states. 27 In contrast, neuromelanin‐sensitive MRI provides a proxy of LC neuronal health and chronic capacity to regulate brain rhythms, which appears more directly relevant to sleep microstructure. Another possibility is that compensatory changes in NA turnover, including alterations in its metabolite 3‐methoxy‐4‐hydroxyphenylglycol, may obscure the direct relationship between CSF NA concentrations and functional outcomes, particularly in the early stages of AD. 50 ,
Taken together, our findings indicate that LC integrity is an important determinant of SWS in aging and along the AD continuum, with sex and vascular health as key modulators. This has several clinical implications. First, if future longitudinal and mechanistic work confirms a causal contribution of LC changes to SWS alterations, interventions aimed at supporting or preserving LC structure and noradrenergic signaling—through pharmacological strategies targeting the noradrenergic system or non‐invasive brain stimulation approaches—may help sustain SWS and its restorative functions. 51 Second, sex‐specific vulnerabilities should be considered in tailoring such interventions. Third, PVS burden may provide an accessible MRI proxy of small‐vessel pathology and, based on prior work, may be related to glymphatic dysfunction, although we did not directly assess glymphatic function in this study.
This study has limitations. The sample size was modest, limiting statistical power to detect small effects, differences in group comparisons, and interaction terms. In particular, analyses involving PVS burden, CSF NA, and sex‐ or PVS‐related interactions should be interpreted as exploratory and hypothesis generating, given the sample size. Our cross‐sectional design precludes causal inference; longitudinal studies are needed to determine whether LC degeneration and BG‐PVS enlargement precede and drive SWS decline, or whether impaired sleep contributes to these pathologies. Given challenges in scoring sleep in neurodegenerative disease, all PSGs were reviewed by neurologists specialized in these disorders. LC‐sensitive MRI, while validated, remains an indirect measure of LC integrity and its signal can be influenced by scanner/sequence parameters, motion, and registration/segmentation inaccuracies, which may introduce technical variability. PVS assessment in our study was performed visually, which may be less precise than automated quantitative approaches used in other studies. Finally, although we accounted for medication use, residual confounding by psychotropic drugs cannot be excluded.
In summary, we provide novel evidence that LC integrity is an important determinant of SWS in aging and AD, with stronger effects in females, and that BG‐PVS further negatively associates with SWS. These findings emphasize the interplay of noradrenergic, sex‐specific, and vascular factors in shaping sleep alterations in AD. Future longitudinal and interventional studies are warranted to test whether preserving LC function, reducing vascular burden, and enhancing SWS can mitigate neurodegenerative processes and improve clinical outcomes in aging and AD.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest. Author disclosures are available in the supporting information.
CONSENT
All participants gave written informed consent for participation. The study protocol was approved by the Hospital Clínic de Barcelona Research Ethics Committee (HCB/2021/0668, HCB/2024/0319).
Supporting information
Supporting Information
Supporting Information
Supporting Information
Supporting Information
Supporting Information
Supporting Information
ACKNOWLEDGMENTS
The authors thank the research team members and the patients for their generous contribution to science. This work was funded by the Global Brain Health Institute (GBHIALZUK‐21‐723831 to NF), Alzheimer's Association (AACSF_21_723056 to NF), and Instituto de Salud Carlos III Spain, and co‐funded by the European Union (PI25/00222 to NF, PI22/00343 to MB), Generalitat de Catalunya, AGAUR (SGR 2021‐01126) and CERCA Programme. N.F. was a recipient of Juan Rodés contract JR22/00014. G.M. was a recipient of Joan Rodés – Josep Baselga Research Contract funded by BBVA foundation. S.R. receives funding from the Instituto de Salud Carlos III through a grant for health research (JR21/00011).
Falgàs N, Tort‐Colet N, Martín‐Sobrino I, et al. Relationship between locus coeruleus and slow‐wave sleep in aging and Alzheimer's disease. Alzheimer's Dement. 2026;22:e71183. 10.1002/alz.71183
REFERENCES
- 1. Liguori C, Placidi F, Izzi F, Spanetta M, Mercuri NB, Di Pucchio A. Sleep dysregulation, memory impairment, and CSF biomarkers during different levels of neurocognitive functioning in Alzheimer's disease course. Alzheimers Res Ther. 2020;12(1):5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Rosenblum Y, Pereira M, Stange O, et al. Divergent associations of slow‐wave sleep versus rapid eye movement sleep with plasma amyloid‐beta. Ann Neurol. 2024;96(1):46‐60. [DOI] [PubMed] [Google Scholar]
- 3. Wang C, Holtzman DM. Bidirectional relationship between sleep and Alzheimer's disease: role of amyloid, tau, and other factors. Neuropsychopharmacology. 2020;45(1):104‐120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Targa ADS, Benítez ID, Dakterzada F, et al. Sleep profile predicts the cognitive decline of mild‐moderate Alzheimer's disease patients. Sleep. 2021;44(10):zsab117. [DOI] [PubMed] [Google Scholar]
- 5. Kastanenka KV, Hou SS, Shakerdge N, et al. Optogenetic restoration of disrupted slow oscillations halts amyloid deposition and restores calcium homeostasis in an animal model of Alzheimer's disease. PLoS One. 2017;12(1):e0170275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kelberman MA, Anderson CR, Chlan E, Rorabaugh JM, McCann KE, Weinshenker D. Consequences of hyperphosphorylated tau in the locus coeruleus on behavior and cognition in a rat model of Alzheimer's disease. J Alzheimers Dis. 2022;86(3):1037‐1059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Lew CH, Petersen C, Neylan TC, Grinberg LT. Tau‐driven degeneration of sleep‐ and wake‐regulating neurons in Alzheimer's disease. Sleep Med Rev. 2021;60:101541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Oh JY, Walsh CM, Ranasinghe K, et al. Subcortical neuronal correlates of sleep in neurodegenerative diseases. JAMA Neurol [Internet]. 2022;79(5):498‐508. doi:10.1001/jamaneurol.2022.0429. [cited 2022 Apr 22]; Available from:. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ehrenberg AJ, Suemoto CK, de Paula França Resende E, et al. Neuropathologic correlates of psychiatric symptoms in Alzheimer's disease. J Alzheimers Dis. 2018;66(1):115‐126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Dahl MJ, Mather M, Düzel S, et al. Rostral locus coeruleus integrity is associated with better memory performance in older adults. Nat Hum Behav. 2019;3(11):1203‐1214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Prokopiou PC, Engels‐Domínguez N, Schultz AP, et al. Association of novelty‐related locus coeruleus function with entorhinal tau deposition and memory decline in preclinical Alzheimer disease. Neurology. 2023;101(12):e1206‐17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Theofilas P, Ehrenberg AJ, Dunlop S, et al. Locus coeruleus volume and cell population changes during Alzheimer's disease progression: a stereological study in human postmortem brains with potential implication for early‐stage biomarker discovery. Alzheimers Dement. 2017;13(3):236‐246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Van Egroo M, van Someren EJW, Grinberg LT, Bennett DA, Jacobs HIL. Associations of 24‐hour rest‐activity rhythm fragmentation, cognitive decline, and postmortem locus coeruleus hypopigmentation in Alzheimer's disease. Ann Neurol. 2024;95(4):653‐664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. David M, Malhotra PA. New approaches for the quantification and targeting of noradrenergic dysfunction in Alzheimer's disease. Annals of Clinical and Translational Neurology. 2022;9(4):582‐596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Oh J, Eser RA, Ehrenberg AJ, et al. Profound degeneration of wake‐promoting neurons in Alzheimer's disease. Alzheimers Dement. 2019;15(10):1253‐1263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Mander BA. Local sleep and Alzheimer's disease pathophysiology. Front Neurosci. 2020;14:525970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Ehrenberg AJ, Kelberman MA, Liu KY, et al. Priorities for research on neuromodulatory subcortical systems in Alzheimer's disease: position paper from the NSS PIA of ISTAART. Alzheimers Dement. 2023;19(5):2182‐2196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lucey BP, Wisch J, Boerwinkle AH, et al. Sleep and longitudinal cognitive performance in preclinical and early symptomatic Alzheimer's disease. Brain. 2021;144(9):2852‐2862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Van Egroo M, Beckers E, Ashton NJ, Blennow K, Zetterberg H, Jacobs HIL. Sex differences in the relationships between 24‐h rest‐activity patterns and plasma markers of Alzheimer's disease pathology. Alzheimers Res Ther. 2024;16(1):277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Stankeviciute L, Tort‐Colet N, Fernández‐Arcos A, et al. Associations between objective sleep metrics and brain structure in cognitively unimpaired adults: interactions with sex and Alzheimer's biomarkers. Alzheimers Dement. 2025;21(6):e70353. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Latta F, Leproult R, Tasali E, Hofmann E, Van Cauter E. Sex differences in delta and alpha EEG activities in healthy older adults. Sleep. 2005;28(12):1525‐1534. [DOI] [PubMed] [Google Scholar]
- 22. Dijk DJ. Sleep of aging women and men: back to basics. Sleep. 2006;29(1):12‐13. [DOI] [PubMed] [Google Scholar]
- 23. Dijk DJ. Regulation and functional correlates of slow wave sleep. J Clin Sleep Med. 2009;5(suppl 2):S6‐S15. [PMC free article] [PubMed] [Google Scholar]
- 24. Sparks JR, Wang X. Menopause‐related changes in sleep and the associations with cardiometabolic health: a narrative review. Healthcare (Basel). 2025;13(17):2085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Luckey AM, Robertson IH, Lawlor B, Mohan A, Vanneste S. Sex differences in locus coeruleus: a heuristic approach that may explain the increased risk of Alzheimer's disease in females. J Alzheimers Dis. 2021;83(2):505‐522. [DOI] [PubMed] [Google Scholar]
- 26. Dredla BK, Del Brutto OH, Castillo PR. Sleep and perivascular spaces. Curr Neurol Neurosci Rep. 2023;23(10):607‐615. [DOI] [PubMed] [Google Scholar]
- 27. Hauglund NL, Andersen M, Tokarska K, et al. Norepinephrine‐mediated slow vasomotion drives glymphatic clearance during sleep. Cell. 2025;188(3):606‐622.e17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Nedergaard M, Goldman SA. Glymphatic failure as a final common pathway to dementia. Science. 2020;370(6512):50‐56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. van den Kerkhof M, van der Thiel MM, Postma AA, et al. Hypertension correlates with stronger blood flow pulsatility in small perforating cerebral arteries assessed with 7 Tesla Magnetic Resonance Imaging. Hypertension. 2023;80(4):802‐810. [DOI] [PubMed] [Google Scholar]
- 30. Pase MP, Pinheiro A, Rowsthorn E, et al. MRI visible perivascular spaces and the risk of incident mild cognitive impairment in a community sample. J Alzheimers Dis. 2023;96(1):103‐112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Mestre H, Kostrikov S, Mehta RI, Nedergaard M. Perivascular spaces, glymphatic dysfunction, and small vessel disease. Clin Sci (Lond). 2017;131(17):2257‐2274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Jack CR, Andrews JS, Beach TG, et al. Revised criteria for diagnosis and staging of Alzheimer's disease: Alzheimer's Association Workgroup. Alzheimers Dement. 2024;20(8):5143‐5169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Berry RB, Brooks R, Gamaldo CE, et al. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications: Version 2.6. American Academy of Sleep Medicine; 2020. [Google Scholar]
- 34. Falgàs N, Peña‐González M, Val‐Guardiola A, et al. Locus coeruleus integrity and neuropsychiatric symptoms in a cohort of early‐ and late‐onset Alzheimer's disease. Alzheimers Dement. 2024;20(9):6351‐6364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Rolls ET, Huang CC, Lin CP, Feng J, Joliot M. Automated anatomical labelling atlas 3. NeuroImage. 2020;206:116189. [DOI] [PubMed] [Google Scholar]
- 36. Potter GM, Chappell FM, Morris Z, Wardlaw JM. Cerebral perivascular spaces visible on magnetic resonance imaging: development of a qualitative rating scale and its observer reliability. Cerebrovasc Dis. 2015;39(3‐4):224‐231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Steriade M, McCormick DA, Sejnowski TJ. Thalamocortical oscillations in the sleeping and aroused brain. Science. 1993;262(5134):679‐685. [DOI] [PubMed] [Google Scholar]
- 38. Neske GT. The slow oscillation in cortical and thalamic networks: mechanisms and functions. Front Neural Circuits. 2015;9:88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Bastian L, Samanta A, Ribeiro de Paula D, et al. Spindle–slow oscillation coupling correlates with memory performance and connectivity changes in a hippocampal network after sleep. Hum Brain Mapp. 2022;43(13):3923‐3943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Staresina BP, Niediek J, Borger V, Surges R, Mormann F. How coupled slow oscillations, spindles and ripples coordinate neuronal processing and communication during human sleep. Nat Neurosci. 2023;26(8):1429‐1437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Lee YF, Gerashchenko D, Timofeev I, Bacskai BJ, Kastanenka KV. Slow wave sleep is a promising intervention target for Alzheimer's disease. Front Neurosci. 2020;14:705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Zhang J, Bai W, Wang W, et al. Mechanisms underlying alterations in norepinephrine levels in the locus coeruleus of ovariectomized rats: modulation by estradiol valerate and black cohosh. Neuroscience. 2017;354:110‐121. [DOI] [PubMed] [Google Scholar]
- 43. Harrington YA, Parisi JM, Duan D, Rojo‐Wissar DM, Holingue C, Spira AP. Sex hormones, sleep, and memory: interrelationships across the adult female lifespan. Front Aging Neurosci. 2022;14:800278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Brown AMC, Gervais NJ. Role of ovarian hormones in the modulation of sleep in females across the adult lifespan. Endocrinology. 2020;161(9):bqaa128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Mulvey B, Bhatti DL, Gyawali S, et al. Molecular and functional sex differences of noradrenergic neurons in the mouse locus coeruleus. Cell Rep. 2018;23(8):2225‐2235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Bangasser DA, Wiersielis KR, Khantsis S. Sex differences in the locus coeruleus‐norepinephrine system and its regulation by stress. Brain Res. 2016;1641(Pt B):177‐188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Duering M, Biessels GJ, Brodtmann A, et al. Neuroimaging standards for research into small vessel disease‐advances since 2013. Lancet Neurol. 2023;22(7):602‐618. [DOI] [PubMed] [Google Scholar]
- 48. Fultz NE, Bonmassar G, Setsompop K, et al. Coupled electrophysiological, hemodynamic, and cerebrospinal fluid oscillations in human sleep. Science. 2019;366(6465):628‐631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Smeijer D, Ikram MK, Hilal S. Enlarged perivascular spaces and dementia: a systematic review. J Alzheimers Dis. 2019;72(1):247‐256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Jacobs HIL, Riphagen JM, Ramakers IHGB, Verhey FRJ. Alzheimer's disease pathology: pathways between central norepinephrine activity, memory, and neuropsychiatric symptoms. Mol Psychiatry. 2021;26(3):897‐906. [DOI] [PubMed] [Google Scholar]
- 51. David MCB, Del Giovane M, Liu KY, et al. Cognitive and neuropsychiatric effects of noradrenergic treatment in Alzheimer's disease: systematic review and meta‐analysis. J Neurol Neurosurg Psychiatry. 2022;93(10):1080‐1090. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information
Supporting Information
Supporting Information
Supporting Information
Supporting Information
Supporting Information