Skip to main content
NCBI home page
Sleep logoLink to Sleep
. 2019 Dec 21;43(6):zsz298. doi: 10.1093/sleep/zsz298

Sleep characteristics and white matter hyperintensities among midlife women

Rebecca C Thurston 1,2,3,✉, Minjie Wu 1, Howard J Aizenstein 1, Yuefang Chang 4, Emma Barinas Mitchell 2, Carol A Derby 5, Pauline M Maki 6
PMCID: PMC7294405  PMID: 31863110

Abstract

Study Objectives

Sleep disturbance is common among midlife women. Poor self-reported sleep characteristics have been linked to cerebrovascular disease and dementia risk. However, little work has considered the relation of objectively assessed sleep characteristics and white matter hyperintensities (WMHs), a marker of small vessel disease in the brain. Among 122 midlife women, we tested whether women with short or disrupted sleep would have greater WMH, adjusting for cardiovascular disease (CVD) risk factors, estradiol, and physiologically assessed sleep hot flashes.

Methods

We recruited 122 women (mean age = 58 years) without a history of stroke or dementia who underwent 72 h of actigraphy to quantify sleep, 24 h of physiologic monitoring to quantify hot flashes; magnetic resonance imaging to assess WMH; phlebotomy, questionnaires, and physical measures (blood pressure, height, and weight). Associations between actigraphy-assessed sleep (wake after sleep onset and total sleep time) and WMH were tested in linear regression models. Covariates included demographics, CVD risk factors (blood pressure, lipids, and diabetes), estradiol, mood, and sleep hot flashes.

Results

Greater actigraphy-assessed waking after sleep onset was associated with more WMH [B(SE) = .008 (.002), p = 0.002], adjusting for demographics, CVD risk factors, and sleep hot flashes. Findings persisted adjusting for estradiol and mood. Neither total sleep time nor subjective sleep quality was related to WMH.

Conclusions

Greater actigraphy-assessed waking after sleep onset but not subjective sleep was related to greater brain WMH among midlife women. Poor sleep may be associated with brain small vessel disease at midlife, which can increase the risk for brain disorders.

Keywords: white matter hyperintensities, brain, sleep, actigraphy

Statement of Significance.

Disrupted sleep is common for midlife and older women. An increasing literature links poorer self-reported sleep to dementia and stroke risk. However, little is known about how sleep is related to cerebrovascular disease risk using actigraphy assessments of sleep. In this investigation of 122 midlife women, greater actigraphy-assessed wake after sleep onset was associated with greater white matter hyperintensities in the brain, an indicator of cerebrovascular disease. Associations persisted adjusting for CVD risk factors, estradiol, mood, and sleep hot flashes. Poor sleep may be associated with brain small vessel disease at midlife, which can increase the risk for future brain disorders. Future work should consider any cerebrovascular benefit of treating women’s sleep problems as they age.

Introduction

Sleep disturbance is common among women in midlife and older ages. In fact, some estimates indicate that a third to 60% of peri- and postmenopausal women in the United States report poor sleep [1–5]. These sleep problems have been attributed to chronologic aging as well as to the menopause transition and symptoms such as hot flashes during sleep, which most women experience during this transition [6]. Poor sleep among midlife women can persist well into their later postmenopausal years [7] and cause considerable distress, functional impairment, and poor quality of life [8].

Sleep problems can also have implications for women’s physical health as they age. For example, a growing body of literature links poor sleep to cardiovascular disease (CVD) risk [9, 10]. However, more limited research has examined the relation between sleep and cerebrovascular health, and particularly lesions in the white matter (WM) that appear as white matter hyperintensities (WMH) on T2-weighted magnetic resonance imaging (MRI) scans. WMH are a subclinical marker of cerebral small vessel damage thought to develop due to small vessel disease in the brain [11]. WMH are linked to later stroke, cognitive decline and dementia, and mortality [12]. As women rarely have overt cerebrovascular disease or dementia at midlife [13, 14], WMH can serve as an early risk marker for the development of these brain disorders and a target for future intervention studies.

The limited work that has examined relationships between sleep and WMH have relied on self-reported indices of sleep. For example, a study of middle aged adults has found short self-reported sleep duration associated with greater WMH volume in the parietal lobe [15]. Other research with older adults has found long self-reported sleep duration or poor subjective sleep quality linked to more global WMH [16, 17]. Other work did not find significant associations between self-reported sleep and WMH [18]. While self-reported sleep does have clinical utility, precise reporting on sleep characteristics can be challenging for participants, and objectively assessed sleep features (e.g. via actigraphy and polysomnography) can yield important information on an individual’s sleep characteristics beyond subjective reports [19]. Moreover, most studies of sleep and WMH have been conducted on older adults, with the literature on midlife sleep and WMH limited. Notably, an important consideration for midlife women is hot flashes. The majority of midlife women experience menopausal hot flashes, many of which occur during sleep [6] and can disrupt sleep [20]. We have previously linked sleep hot flashes to WMH [21]; thus, for midlife women, sleep hot flashes should be taken into account when exploring sleep–WMH relationships.

To address these gaps in the literature, the goal of the present study was to determine whether objectively assessed sleep is linked to greater WMH among midlife women after accounting for the occurrence of hot flashes. We examined the relationship between sleep and WMH among a sample of 122 women aged 45–66 who were free of clinical CVD, stroke, or dementia. Women underwent actigraphic assessments of their sleep and brain MRI to assess WMH, as well as objective monitoring of physiologic hot flashes using ambulatory skin conductance monitoring [22, 23]. We tested the hypothesis that short or disrupted sleep would be associated with greater WMH, after adjustment for sleep hot flashes and a range of other potential confounders.

Materials and Methods

Sample

Participants were recruited from a cohort of non-smoking late perimenopausal and postmenopausal midlife women who had participated in a study on menopausal hot flashes and cardiovascular health (MsHeart) between 2012 and 2015. MsHeart exclusion criteria included: current smoking; reported history of CVD/stroke/cerebrovascular accident; insulin-dependent diabetes; Parkinson’s disease; hysterectomy and/or bilateral oophorectomy; current pregnancy; and use of hormone therapy (oral or transdermal estrogen and/or progesterone), select cardiovascular medications (beta blockers, calcium channel blockers, and alpha-2 adrenergic agonists), selective estrogen receptor modulators, aromatase inhibitors, selective serotonin reuptake inhibitors, or serotonin norepinephrine reuptake inhibitors. For the present study, 126 of the MsHeart participants were recruited between 2017 and 2019 to participate in the MsBrain Study, a study of menopause and brain aging. MsBrain exclusion criteria included: a reported history of stroke/cerebrovascular accident; dementia; seizure disorder; brain tumor; Parkinson’s Disease; a history of head trauma with loss of consciousness; contraindications to MRI (e.g. metal in the body); current chemotherapy; active substance use; pregnancy; and current use of medications including hormone therapy (oral or transdermal estrogen and/or progesterone), selective estrogen receptor modulators, aromatase inhibitors, selective serotonin reuptake inhibitors, or serotonin norepinephrine reuptake inhibitors.

Of the 126 women, one woman was excluded due to brain tumor, two women were excluded due to suspected stroke, one woman was excluded due to seizure disorder, yielding a sample of 122 women with actigraphy and MRI data. Due to missing values, an additional five women were excluded from multivariable models which included low-density lipoprotein (LDL) cholesterol (N = 3) and physiologic hot flashes (N = 2).

Participants underwent telephone and in-person screening procedures, physical measurements and questionnaire completion; 3 days of ambulatory monitoring, including sleep measurement by actigraphy and hot flashes by skin conductance; and MRI brain imaging. Procedures were approved by the University of Pittsburgh Institutional Review Board. Participants provided written, informed consent.

Sleep

Women wore an Actiwatch 2 wrist actigraph unit on the wrist of the non-dominant hand (Respironics, Inc., Murrysville, PA) [19] and completed a sleep diary [24] for 3 consecutive days. Actigraphy data were collected in 1-min epochs and analyzed with Philips Actiware v6.0.0 software, with a wake threshold of 40 and number of epochs of sleep/wake for sleep onset/offset of 10. Bedtime (time tried to go to sleep) and rise time (final wake time) were determined via sleep diary reports. Wake after sleep onset (WASO; minutes of wakefulness between actigraphy-defined sleep onset time and actigraphy-defined final wake time) was our primary sleep outcome given its specific relevance to menopause, and to circulating markers of endothelial function in our prior work [25]. Total sleep time [(difference between actigraphy-defined sleep onset and actigraphy-defined final wake time) − (actigraphy-defined WASO)] was also considered. Total sleep time was considered primarily as a continuous variable given that there were few long sleepers (e.g. only two women had sleep times ≥9 h/night) in the sample, but it was also considered categorized according to its distribution (<6 h, 6–7 h, and >7 h) in additional analyses. Women completed the Pittsburgh Sleep Quality Index (PSQI), a widely used and well-validated measure of subjective sleep quality [26]. Women also reported a history of sleep apnea and completed the Berlin Questionnaire, a validated inventory assessing sleep apnea symptoms [27].

WMH

MRI scanning was performed at the MR Research Center of the University of Pittsburgh. MRI scanning was performed an average of 12 days (standard deviation = 4.9; range 4–30) from the sleep measurements. A 3T Siemens Prisma MR scanner was used, with a Siemens 64-channel head coil. Two series of MR images were analyzed for the current study: A magnetization-prepared rapid gradient echo (MPRAGE) T1-weighted sequence and T2-weighted (T2w) fluid-attenuated inversion recovery (FLAIR) sequence. MPRAGE images were acquired in the sagittal plane using the following parameters: TR = 2400 ms; TE = 2.22 ms; TI = 1000 ms; flip angle = 8°; FOV = 240*256 mm; slice thickness = 0.8 mm; voxel size = 0.8 mm*0.8 mm; matrix size = 300*320; and number of slices = 208. FLAIR images were acquired in the axial plane using the following parameters: TR = 9690 or 10000 ms; TE = 91 ms; TI = 2500 ms; flip angle = 135°; FOV = 256 × 256 mm; matrix = 320 × 320; slice thickness = 1.6 mm; voxel size = 0.8 mm*0.8 mm; and number of slices = 104. The small change in TR from 9699 to 1000 was performed 1 year into the study to meet Specific Absorption Rate human safety guidelines for participants with a higher body mass index (BMI). This change slightly increased the time of acquisition but had minimal effect on image contrast.

An automated pipeline was used to segment WMH on the T2w FLAIR images using previously documented methods [28]. Cerebral and cerebellar WM were segmented in individual T2w FLAIR image space using SPM12 (Welcome Trust Centre for Neuroimaging, http://www.fil.ion.ucl.ac.uk/spm/). Given that there were very few lesions in the cerebellum in our participants, the mean and standard deviation of the cerebellar WM on the FLAIR image were used to Z-transform the FLAIR image (Z-T2w FLAIR). On the Z-transformed FLAIR images, voxels greater than or equal to 2 and within the cerebral WM mask were identified as WMH. This method uses individual mean and standard deviation from normal cerebellar WM to standardize individual FLAIR images, which avoids systematic bias in seed selection between participants with significant cerebral WMH versus those with few WMH. Z-transformation also reduces variations in FLAIR images. The total WMH volume (in cubic centimeters) was normalized by total gray matter (GM) and WM volumes [nWMH = WMH/(GM + WM)] and log transformed for analysis.

Covariates

Height was measured via fixed stadiometer and weight via Detecto Apex scale, and BMI was calculated [weight (kg)/height2 (m)]. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were the average of three seated measurements taken via a Dinamap v100. Demographics, medical history, medication use, and health behaviors were assessed by questionnaires and interview. Race/ethnicity was self-reported. Educational attainment was assessed as years of completed education and classified as less than or greater than a college degree for analysis. Depressive symptoms were assessed via the Center for Epidemiologic Studies Depression Survey [29]. Consistent with prior work, physiologic hot flashes were assessed via the VU-AMS monitor (Amsterdam, The Netherlands), a portable device that measures sternal skin conductance sampled at 1 Hz from the sternum via a 0.5 V constant voltage circuit passed between two Ag/AgCl electrodes (UFI) filled with 0.05 M KCl Velvachol/glycol paste [30]. Physiologic hot flashes were classified via standard methods, with skin conductance rise of 2 µmho in 30 s [22] flagged by UFI software (DPSv3.6; Morro Bay, CA), and edited for artifact [23]. Physiologic hot flashes were classified as occurring during sleep or wake hours as defined by the sleep diary, with sleep physiologic hot flashes considered here.

Participants provided a morning fasting blood sample for the assessment of glucose, insulin, lipids, and estradiol. Glucose, total cholesterol, high-density lipoprotein cholesterol, and triglycerides were determined using an enzymatic assay; insulin was determined by immunoturbidimetric assay in serum (ACE Axcel, Alfa Wasserman; West Caldwell, NJ). LDL was calculated using the Friedewald equation [31]. Homeostatic model assessment, an index reflecting insulin resistance, was calculated [(fasting insulin*fasting glucose)/22.5] [32]. Estradiol was measured at the University of Pittsburgh’s Small Biomarker Core using liquid chromatography–tandem mass spectrometry, the reference method for the measurement of estradiol as it can quantify the low levels of estradiol observed in postmenopausal women with high precision and accuracy [33, 34]. This assay employs liquid–liquid extraction, derivatization, and detection with a triple quad mass spectrometer [35]. The lower limit of quantitation was 1.0 pg/mL. Intraday statistics showed errors below 8.1% and relative standard deviations (RSDs) below 10.4%; interday statistics showed errors below 5.0% with RSD below 7.4%. Standards, blanks, calibrators, and control pools were run simultaneously with all samples.

Statistical analysis

Variables were examined for distributions, outliers, and cell sizes. WMH and estradiol were log transformed to conform to model assumptions of normality. One observation was excluded as an outlier: a WASO value that was greater than 6 standard deviations from the mean. Bivariate relations between study variables and WMH were examined via Pearson and Spearman correlation coefficients. We tested relations between sleep variables and WMH in separate linear regression models. Age, BMI, and sleep apnea (current sleep apnea, which was the sleep apnea most strongly related to WMH) were included in models as covariates a priori. All other covariates were included based upon their relationship with the outcome at p < 0.10. The main analysis included three models. The first model was unadjusted for covariates. The second model included age, BMI, race/ethnicity, education, sleep apnea (current sleep apnea), sleep medication, diabetes, DBP, and LDL cholesterol. The third model also included physiologic hot flashes. In additional analyses, we adjusted for Berlin scores, estradiol, and depressive symptoms; excluded a nightshift worker; considered PSQI subscales; and considered total sleep time as a categorical variable. All tests were two tailed with an alpha set to 0.05. Analyses were conducted using SAS v9.4 (SAS Institute, Cary, NC).

Results

Women were on average 58.8 years old, overweight, and normotensive (Table 1). Over a quarter of the women in the sample (27%) were black, with the remainder non-Hispanic white. Half of the women had PSQI scores indicating poor subjective sleep quality. The average actigraphy sleep time was 6.5 h, and the average minutes of wakening after sleep onset detected on actigraphy were 42 min. Factors associated with increased WMH at p < 0.10 were non-white race/ethnicity (vs. white: r = −0.29, p = 0.001), lower education (<college vs. ≥college, r = −0.26, p = 0.004), higher DBP (r = 0.24, p = 0.007), and higher LDL cholesterol (r = 0.19, p = 0.04), and lower sleep physiologic hot flashes (r = 0.16, p = 0.09).

Table 1.

Sample characteristics

N 122
Age, M (SD) 58.86 (4.14)
Race/ethnicity, N (%)
 White 89 (72.95)
 Black 33 (27.05)
Education, N (%)
 High school/some college/vocational 36 (29.51)
 College or higher 86 (70.49)
BMI, M (SD) 29.58 (6.51)
SBP, mmHg, M (SD) 117.34 (13.21)
DBP, mmHg, M (SD) 67.93 (8.78)
LDL cholesterol, mg/dL, M (SD) 113.77 (34.05)
HDL cholesterol, mg/dL, M (SD) 61.90 (16.80)
Estradiol, pg/mL, median (IQR) 2 (2, 6)
Triglycerides, mg/dL, median (IQR) 96.00 (70.00, 123.00)
Homeostatic model assessment, median (IQR) 4.49 (2.94, 5.36)
Women reporting hot flashes, N (%) 65 (53.28)
Physiologic overnight hot flashes, number, median (IQR)* 1 (0, 3)
Sleep medication use, N (%) 2 (1.64)
Antihypertensive medication use, N (%) 30 (24.59)
Depressive symptoms (CESD), median (IQR) 6.00 (3.00, 10.00)
Subjective sleep quality (PSQI), M (SD) 5.88 (3.06)
Total sleep time (actigraphy), min, M (SD) 395.59 (62.08)
WASO (actigraphy), min, M (SD) 42.73 (24.50)
Open in a new tab

*Number divided by sleep monitoring time and standardized to 7-h sleep time.

In analyses of the relation between actigraphic sleep and WMH, greater WASO was associated with greater WMH (Figure 1). Associations persisted when covarying for a range of covariates, including sleep physiologic hot flashes, DBP and LDL cholesterol (Table 2). Neither total sleep time nor subjective sleep quality [PSQI: B(SE)= −02 (.02), p = 0.26, age, race, education, BMI, sleep medications, sleep apnea, diabetes, DBP, LDL cholesterol, and physiologic hot flashes] was related to WMH.

Figure 1.

Figure 1.

Open in a new tab

Scatterplot showing (A) the raw association between WASO and WMH (B) representative T2 FLAIR images from individuals with low, medium, and high WMH (red dots in A).

Table 2.

Relationship of actigraphy-assessed WASO and sleep time to WMH

WMH
Model 1 Model 2 Model 3
B(SE) B(SE) B(SE)
WASO .007 (.003)** .008 (.002)** .008 (.002)**
Total sleep time −.03 (.05) .03 (.05) .03 (.05)
Open in a new tab

**p < 0.01; WMH log transformed.

Model 1: unadjusted; Model 2: adjusted for age, race/ethnicity, education, BMI, sleep apnea, sleep medication, diabetes, DBP, and LDL cholesterol; Model 3: Model 2 covariates + physiologic hot flashes.

In additional models, although estradiol was not significantly related to WMH (r = 0.03, p = 0.75), we considered estradiol as a covariate in models and findings were unchanged (data not shown). We also considered PSQI subscales in relation to WMH, with all subscales nonsignificant with the exception of sleep efficiency, which was marginally related to fewer WMH [B(SE)= −.09(.05), p = 0.08, adjusted for age, race, education, BMI, sleep medications, sleep apnea, diabetes, DBP, LDL cholesterol, and physiologic hot flashes]. We also considered total sleep time as a categorical variable (<6 h, 6–7 h, and >7 h), and conclusions were unchanged (data not shown). We also considered models with Berlin scores instead of reported sleep apnea, with conclusions unchanged (data not shown). Further, we excluded the one shift worker in the sample, with findings unchanged (data not shown). Moreover, although depressive symptoms were not significantly related to WMH, we considered models adjusting for these symptoms and findings were unchanged (data not shown).

Discussion

In this study of midlife and older women, we found more awakening during the night related to greater brain WMH. These associations persisted adjusting for a range of covariates, including standard risk factors for cerebrovascular disease as well as monitor-assessed sleep hot flashes. These findings indicate that more awakening during the night is associated with indicators of poorer cerebrovascular health at midlife.

A growing literature points to the importance of sleep for cerebrovascular health. A more limited body of research links poorer self-reported sleep to greater WMH [15–17]. However, we found objective (actigraphy) sleep measures associated with WMH. Notably, objective indices can be particularly useful in investigating sleep as they avoid the need for individuals to precisely report on their sleep characteristics. A small literature has linked actigraphic indices of poor sleep to diffusion tensor imaging indices of WM microstructure [36, 37], which can reflect a range of processes, such as changes in myelination, changes in the alignment of fibers, or disease states [38–40]. Further, the present findings complement our prior work which finds that poorer actigaphically assessed sleep was associated with greater carotid atherosclerosis among midlife women [10]. The present work represents an important contribution to the literature on sleep and cerebrovascular health, indicating more objectively assessed nighttime awakening is associated with more WMH as early as midlife in women.

Midlife and early old age are particularly important times to study how sleep related to the brain health of women. Poor sleep is prevalent in women during midlife and early old age [41]. For women, midlife includes the menopause transition, when poor sleep and hot flashes are common; these symptoms can persist well into the later postmenopausal years [3, 6]. In prior work, we found a relationship between hot flashes accompanied by awakening during the night [20] and greater subclinical CVD [42]. We have also found sleep hot flashes linked to greater WMH [21]. However, state-of-the-art-measured hot flashes did not account for the present relationships between sleep and WMH. Notably, poor sleep is common among midlife women even in the absence of hot flashes [43]. Further, midlife and early old age are important windows of opportunity to lower the risk of cerebrovascular disease [13], and sleep may be a modifiable risk factor for adverse brain health.

Associations with WMH were observed for wakening during the night rather than for sleep duration or subjective sleep quality. Notably, the limited prior research on sleep duration and WMH, which relied exclusively on self-reported sleep, has produced highly mixed findings [15, 16]. Our work is consistent with our prior work linking more nighttime wakening to a more proinflammatory/coagulant profile [25]. We did not observe associations between global subjective sleep quality and WMH, with the exception of the PSQI subscale measuring poorer subjective sleep continuity. These findings further suggest the potential specificity of poorer sleep continuity, rather than sleep duration or global sleep quality, related to greater WMH among midlife and older women.

Multiple mechanisms can underlie associations between poor sleep and WMH. We considered a range of standard CVD risk factors in these associations, such as blood pressure, obesity, diabetes, and lipids, and findings persisted. We controlled for several indices of sleep apnea, and associations remained. We carefully considered the role of overnight hot flashes, assessed via physiologic monitoring, important given difficulties in accurately estimating hot flashes occurring during sleep [44]. Although hot flashes did not explain the observed associations here, their role is important for future work to consider when examining relations between sleep and WMH during the menopause transition. Future work should also consider other potential mechanisms such as genetic [45] or epigenetic [46] processes that may underlie links between sleep and WMH.

Study findings should be interpreted in light of several limitations. First, in this cross-sectional study we cannot make conclusions about the causal nature of or directionality of findings. It is plausible that the WMH cause poor or disrupted sleep [47]. Further, while actigraphy provides a strong proxy measure sleep, we did not use the more burdensome and costly polysomnography indices which include more direct measures of sleep/wake state. We conducted 3 days of actigraphy monitoring, which is less ideal than more extended monitoring durations for obtaining estimates of typical sleep patterns (e.g. weekends vs. weekdays, intra–inter day stability) [19]. We had several self-reported indices of sleep apnea, and findings persisted even when controlling for these factors, yet we did not have direct measures of sleep disordered breathing. Moreover, while WMH are commonly interpreted as reflecting components of small vessel disease, they can reflect a range of pathophysiologic processes, including demyelination, mild gliosis, and axonal loss [12]. Finally, the study sample was comprised of non-Hispanic white and black women, and conclusions may not apply to other racial/ethnic groups of women or to men.

This study has several strengths. This analysis was based on a well-characterized sample of midlife and older women. Women underwent detailed, actigraphic assessments of sleep and brain MRI indices of WMH. Multiple covariates were considered, including CVD risk factors, estradiol, and sleep hot flashes measured via state-of-the-art means.

In conclusion, this study indicated that midlife women with more awakening during the night had greater brain WMH, a marker of small vessel disease associated with risk for future stroke and dementia. Findings indicate the importance of poor sleep continuity not only to quality of life, but also to brain health. These findings can also point to sleep as an important modifiable risk factor, that if causally related to WMH, can be targeted to improve brain health as women age.

Acknowledgments

This research was supported by the National Institutes of Health (NIH), National Institute on Aging (RF1AG053504 to Thurston and Maki) and the National Institutes of Health (NIH), National Heart, Lung, and Blood Institute (2K24HL123565 to Thurston). This work was also supported by the University of Pittsburgh Clinical and Translational Science Institute (NIH Grant UL1TR000005). This project used the University of Pittsburgh Small Molecule Biomarker Core (NIH Grant S10RR023461-01).

Conflict of interest statement: R Thurston: Astellas, Pfizer, and Procter and Gamble (consulting). No other authors have financial disclosures to declare.

References

  • 1. NIH. State-of-the-Science Conference statement. Management of menopause-related symptoms. Ann Intern Med. 2005;142:1003–1013. [PubMed] [Google Scholar]
  • 2. Kravitz HM, et al. Sleep difficulty in women at midlife: a community survey of sleep and the menopausal transition. Menopause. 2003;10(1):19–28. [DOI] [PubMed] [Google Scholar]
  • 3. Kravitz HM, et al. Sleep disturbance during the menopausal transition in a multi-ethnic community sample of women. Sleep. 2008;31(7):979–990. [PMC free article] [PubMed] [Google Scholar]
  • 4. Kravitz HM, et al. Sleep during the perimenopause: a SWAN story. Obstet Gynecol Clin North Am. 2011;38(3):567–586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Hall MH, et al. Race and financial strain are independent correlates of sleep in midlife women: the SWAN sleep study. Sleep. 2009;32(1):73–82. [PMC free article] [PubMed] [Google Scholar]
  • 6. Gold EB, et al. Longitudinal analysis of the association between vasomotor symptoms and race/ethnicity across the menopausal transition: study of women’s health across the nation. Am J Public Health. 2006;96(7):1226–1235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Krystal AD, et al. Sleep in peri-menopausal and post-menopausal women. Sleep Med Rev. 1998;2(4):243–253. [DOI] [PubMed] [Google Scholar]
  • 8. Bolge SC, et al. Burden associated with chronic sleep maintenance insomnia characterized by nighttime awakenings among women with menopausal symptoms. Menopause. 2010;17(1):80–86. [DOI] [PubMed] [Google Scholar]
  • 9. Cappuccio FP, et al. Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies. Eur Heart J. 2011;32(12):1484–1492. [DOI] [PubMed] [Google Scholar]
  • 10. Thurston RC, et al. Sleep characteristics and carotid atherosclerosis among midlife women. Sleep. 2017;40(2). doi:10.1093/sleep/zsw052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Avis NE, et al. ; Study of Women’s Health Across the Nation. Duration of menopausal vasomotor symptoms over the menopause transition. JAMA Intern Med. 2015;175(4):531–539. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Debette S, et al. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. BMJ. 2010;341:c3666. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Benjamin EJ, et al. ; American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics—2019 update: a report from the American Heart Association. Circulation. 2019;139(10):e56–e528. [DOI] [PubMed] [Google Scholar]
  • 14. Lambert MA, et al. Estimating the burden of early onset dementia; systematic review of disease prevalence. Eur J Neurol. 2014;21(4):563–569. [DOI] [PubMed] [Google Scholar]
  • 15. Yaffe K, et al. Sleep duration and white matter quality in middle-aged adults. Sleep. 2016;39(9):1743–1747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Ramos AR, et al. Sleep duration is associated with white matter hyperintensity volume in older adults: the Northern Manhattan Study. J Sleep Res. 2014;23(5):524–530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Del Brutto OH, et al. Poor sleep quality and silent markers of cerebral small vessel disease: a population-based study in community-dwelling older adults (The Atahualpa Project). Sleep Med. 2015;16(3):428–431. [DOI] [PubMed] [Google Scholar]
  • 18. Sexton CE, et al. Associations between self-reported sleep quality and white matter in community-dwelling older adults: a prospective cohort study. Hum Brain Mapp. 2017;38(11):5465–5473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Ancoli-Israel S, et al. The role of actigraphy in the study of sleep and circadian rhythms. Sleep. 2003;26(3):342–392. [DOI] [PubMed] [Google Scholar]
  • 20. Thurston RC, et al. Hot flashes and awakenings among midlife women. Sleep 2019. [Epub ahead of print]. doi:10.1093/sleep/zsz131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Thurston RC, et al. Menopausal hot flashes and white matter hyperintensities. Menopause. 2016;23(1):27–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Freedman RR. Laboratory and ambulatory monitoring of menopausal hot flashes. Psychophysiology. 1989;26(5):573–579. [DOI] [PubMed] [Google Scholar]
  • 23. Carpenter JS, et al. Feasibility and psychometrics of an ambulatory hot flash monitoring device. Menopause. 1999;6(3):209–215. [DOI] [PubMed] [Google Scholar]
  • 24. Monk TH, et al. The Pittsburgh Sleep Diary. J Sleep Res. 1994;3(2):111–120. [PubMed] [Google Scholar]
  • 25. Nowakowski S, et al. Sleep characteristics and inflammatory biomarkers among midlife women. Sleep. 2018;41(5). doi:10.1093/sleep/zsy049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Buysse DJ, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213. [DOI] [PubMed] [Google Scholar]
  • 27. Netzer NC, et al. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med. 1999;131(7):485–491. [DOI] [PubMed] [Google Scholar]
  • 28. Wu M, et al. ; Neuropsychology Working Group of the Multicenter AIDS Cohort Study. HIV disease and diabetes interact to affect brain white matter hyperintensities and cognition. AIDS. 2018;32(13):1803–1810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401. [Google Scholar]
  • 30. Dormire SL, et al. An alternative to Unibase/glycol as an effective nonhydrating electrolyte medium for the measurement of electrodermal activity. Psychophysiology. 2002;39(4):423–426. [DOI] [PubMed] [Google Scholar]
  • 31. Friedewald WT, et al. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499–502. [PubMed] [Google Scholar]
  • 32. Matthews DR, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412–419. [DOI] [PubMed] [Google Scholar]
  • 33. Demers LM. Testosterone and estradiol assays: current and future trends. Steroids. 2008;73(13):1333–1338. [DOI] [PubMed] [Google Scholar]
  • 34. Nelson RE, et al. Liquid chromatography-tandem mass spectrometry assay for simultaneous measurement of estradiol and estrone in human plasma. Clin Chem. 2004;50(2):373–384. [DOI] [PubMed] [Google Scholar]
  • 35. Santen RJ, et al. Potential role of ultra-sensitive estradiol assays in estimating the risk of breast cancer and fractures. Steroids. 2008;73(13):1318–1321. [DOI] [PubMed] [Google Scholar]
  • 36. Kocevska D, et al. The prospective association of objectively measured sleep and cerebral white matter microstructure in middle-aged and older persons. Sleep. 2019;42(10). doi:10.1093/sleep/zsz140 [DOI] [PubMed] [Google Scholar]
  • 37. Khalsa S, et al. Habitual sleep durations and subjective sleep quality predict white matter differences in the human brain. Neurobiol Sleep Circadian Rhythms. 2017;3: 17–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Alexander AL, et al. Diffusion tensor imaging of the brain. Neurotherapeutics. 2007;4(3):316–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Jones DK, et al. White matter integrity, fiber count, and other fallacies: the do’s and don’ts of diffusion MRI. Neuroimage. 2013;73:239–254. [DOI] [PubMed] [Google Scholar]
  • 40. Korgaonkar MS, et al. Loss of white matter integrity in major depressive disorder: evidence using tract-based spatial statistical analysis of diffusion tensor imaging. Hum Brain Mapp. 2011;32(12):2161–2171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Sowers MF, et al. Sex steroid hormone profiles are related to sleep measures from polysomnography and the Pittsburgh Sleep Quality Index. Sleep. 2008;31(10):1339–1349. [PMC free article] [PubMed] [Google Scholar]
  • 42. Thurston RC, et al. Menopausal hot flashes and carotid intima media thickness among midlife women. Stroke. 2016;47(12):2910–2915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Freeman EW, et al. Poor sleep in relation to natural menopause: a population-based 14-year follow-up of midlife women. Menopause. 2015;22(7):719–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Thurston RC, et al. Are vasomotor symptoms associated with sleep characteristics among symptomatic midlife women? Comparisons of self-report and objective measures. Menopause. 2012;19(7):742–748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Goel N. Genetic markers of sleep and sleepiness. Sleep Med Clin. 2017;12(3):289–299. [DOI] [PubMed] [Google Scholar]
  • 46. Carroll JE, et al. Epigenetic aging and immune senescence in women with insomnia symptoms: findings from the Women’s Health Initiative Study. Biol Psychiatry. 2017;81(2):136–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Alosco ML, et al. Reduced cerebral blood flow and white matter hyperintensities predict poor sleep in heart failure. Behav Brain Funct. 2013;9:42. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES