PSQI global score
At baseline, PSQI global score did not differ significantly among treatment groups (Table 2). Overall, 24% of participants (22% PBO, 21% t-E2, and 28% o-CEE) had poor sleep quality (global scores >8) at study entry. Compared with baseline values, each of the two HT groups and the PBO group showed a reduction in average PSQI global scores over 4 years of treatment (each P < 0.001). The average reduction in the global score was similar between the two hormone-treated groups (average change of −1.27 [o-CEE] and −1.32 [t-E2] points), and both were significantly greater than that in the PBO group (−0.60 points; P = 0.001 [o-CEE vs PBO] and P = 0.002 [t-E2 vs PBO]; Table 2). Similarly, the percentage of women with poor sleep quality (PSQI global score >8) decreased with t-E2 (from 21% to 9%; P < 0.001) and o-CEE (from 28% to 16%; P < 0.001), with a smaller decline with PBO (from 22% to 17%; P = 0.06). When compared with PBO, the reduction in the percentage of women with poor sleep quality (ie, the number of women with improved sleep quality) was significantly greater in the t-E2 group (P = 0.003) and modestly but not significantly greater in the o-CEE group (P = 0.07).
PSQI individual domains score
Compared with baseline scores, significant improvements in all six domain scores were observed during treatment within each group, with the exception of sleep latency in the PBO group (P = 0.06; Table 2). The average improvement in score differed between treatments for three out of the six domains (sleep satisfaction, sleep latency, and sleep disturbances), with each of these three domains showing significantly greater improvement in one or both of the HT groups when compared with PBO ( Fig. 2 ). Among the three domains, only sleep disturbances showed a statistically significant difference between the two HT formulations (P = 0.029), with t-E2 improving more during follow-up than o-CEE (Table 2). There were no significant differences between groups for the changes in domain scores pertaining to sleep efficiency (P = 0.45), sleep duration (P = 0.38), or daytime dysfunction (P = 0.26) across follow-up.
Association with VMS
A total of 662 women providing responses of VMS severity during the course of treatment were retained in the corresponding treatment comparisons. As reported previously for the entire KEEPS cohort,27 the average severity scores of hot flashes and night sweats were significantly reduced by both HT formulations when compared with PBO (P < 0.001 for both), with no difference between the o-CEE and t-E2 groups (P = 0.343 and P = 0.919, respectively; Table 2). For the 646 participants for whom both PSQI and VMS data were available, there were positive associations between the average change in global PSQI score and the average changes in severity of hot flashes (r s = 0.170, P < 0.001) and night sweats (r s = 0.177, P < 0.001; Table 3). Among the correlations assessed between changes in the individual sleep domain scores and changes in scores for hot flashes, all domains except sleep latency and sleep efficiency correlated positively with change in severity. In addition to the domains of sleep latency and sleep efficiency, the domain of sleep duration also did not correlate with changes in night sweats.
In multivariable analysis to examine the effect of treatment on change in sleep quality while controlling for changes in VMS, the difference in average improvement in PSQI global score between treatments, though attenuated, remained significant after adjustment for average improvement in each symptom (P = 0.020 adjusting for changes in hot flashes; P = 0.004 adjusting for changes in night sweats). Controlling for the effects of treatment, the associations between the change in each VMS and change in sleep remained significant as well (P = 0.002 for hot flashes and P = 0.029 for night sweats).
Post hoc VMS subgroup analysis
At baseline, 226 women reported having moderate to severe night sweats, and 279 women (some overlap with those having moderate to severe night sweats) reported having moderate to severe hot flashes. In these respective subgroups of women, there were significant and positive correlations of changes in the global PSQI score, with changes in severity of night sweats (r s = 0.181, P = 0.002) and with changes in severity of hot flashes (r s = 0.207, P = 0.002; Table 3). In contrast to the results obtained on the overall set of participants, improvement in sleep latency in these women with moderate/severe symptoms correlated significantly with reduced severity of hot flashes (r s = 0.137, P = 0.022) and night sweats (r s = 0.147, P = 0.028); in addition, sleep duration correlated more strongly with reduced severity of hot flashes (r s = 0.239, P < 0.001) and night sweats (r s = 0.169, P = 0.011). In models adjusting for HT, the association between average changes in VMS and the average change in PSQI global score remained significant in both VMS subgroups (both P < 0.001). In contrast, the association of treatment with average change in PSQI global score was attenuated and no longer significant in both subgroups after adjustment for the average change in the corresponding VMS (P = 0.525 and 0.128 from partial tests of treatment effects in the subgroup with baseline hot flashes and with baseline night flashes, respectively).
In a population of a majority of white recently menopausal women, improvements in sleep quality were observed with the use of low-dose HT (oral and transdermal) over a 4-year period. The global sleep score indicative of sleep quality in women at baseline is consistent with what has been reported for population-based studies.2 The average change in the global sleep score during treatment was a reduction of about 1.3 points in both groups randomized to HT. The magnitude of change in sleep score with HT was about twice as great as that reported by the PBO group (0.06), which is consistent with effects of HT on sleep reported in other studies.12
A second finding of the present study is that changes in sleep quality correlated with changes in VMS (hot flashes and night sweats)—a finding that persisted after controlling for treatment assignment to HT and consistent with other studies.10 However, a multivariable analysis on the overall set of women demonstrated significant partial effects of both treatment and change in VMS, indicating the alleviation of VMS does not fully account for the improved sleep outcomes among those assigned to HT and suggesting that HT affects other mechanisms associated with sleep. A causal relationship between these factors is hard to establish, as there is a bidirectional relationship of perceived sleep quality and VMS, in that poor sleep quality is both a consequence of VMS and also an influence on the extent to which VMS are perceived as bothersome. Some insight into why HT may affect sleep through mechanisms other than alleviation of VMS is provided by the subanalysis of the relationship between symptom relief and sleep domains in women reporting moderate/severe VMS at baseline. In these women, unlike the entire set of women, the association between the change in symptom severity and the change in PSQI score was attenuated and no longer significant after adjustment for treatment. These results suggest that in women with moderate to severe symptoms, but not in those with none to mild, the effects of HT on sleep are mediated through symptom relief—a finding consistent with conclusions of the recent meta-analysis of other studies of sleep, VMS, and HT.12
A third finding of the present study is the direct comparison between two formulations and doses of HT, which are commonly used in clinical practice, on sleep domains. The majority of previous sleep studies evaluated o-CEE at 0.625 mg/d.12 In the present study, following clinical guidelines that followed the cessation of the Women's Health Initiative in 2002 for use of lower doses of HT,32 0.425 mg/d of o-CEE was used in KEEPS when it was designed in 2004. Although overall sleep quality was improved with HT, not all domains of sleep showed significant change averaged across the treatment period. The finding that t-E2 was more efficacious than o-CEE in alleviating sleep disturbances may be related to the pharmacokinetics of these two formulations. Transdermal E2 is likely to provide more consistent 24-hour estradiol dosing, whereas o-CEE may engender daytime peaks and night-time troughs if women took their o-CEE in the morning, thus, leading to less relief of sleep disturbances. In the entire group, sleep efficiency and sleep duration were not affected by either HT, reflecting, perhaps, that other factors such as life circumstances may impact these domains, especially in women who do not report moderate to severe symptoms.
Unlike what has been reported in other studies,12 daytime dysfunction was not improved by HT in the KEEPS cohort. This difference may be explained, in part, by a lower prevalence of women with poor sleep quality (global score >8) in the present study sample (24% at baseline) compared with other studies.33,34 In addition, the majority of participants in the KEEPS were white and recently menopausal as confirmed by strict criteria, whereas in other studies, women were more often of mixed ethnicity and were perimenopausal and also postmenopausal. The possibility that doses or formulations of HT directly influence specific domains of sleep apart from changes in VMS in women of different ethnicities and ages will require additional study.
There are difficulties in comparing self-reported sleep outcomes among clinical trials due to heterogeneity among the available questionnaires and the absence of objective testing.12 For example, self-reported sleep tools may include validated and nonvalidated scales, single-item or multiple-items measures, visual analog scales, and diaries such that clinical applicability of results in various reports to clinical management is uncertain.35 Yet, the use of sleep architecture measurements (polysomnography, wrist actigraphy) does not always correlate with perceived sleep quality,36,37 and due to costs and accessibility, these tests have limited applicability in large population settings. Therefore, self-report instruments remain crucial in the clinical assessment of outcomes after interventions to improve sleep. In studies of menopause, self-report instruments may allow clinicians to evaluate who might benefit from HT for sleep and who might benefit from additional clinical assessment, such as testing for sleep apnea. Although the PSQI contains questions regarding symptoms of snoring and stopping breathing during the night, the responses to these questions are included in an overall score for sleep disturbances and may not provide accurate information regarding sleep apnea for those individuals who sleep alone. However, the answers to these questions may provide information for the individual physician who examines the questionnaire for their individual patient.
This present study has a number of strengths. Evaluating the recently menopausal women enrolled in KEEPS allowed many of the deficiencies of the prior literature to be addressed. First, KEEPS was a randomized clinical trial enrolling a large number of well-characterized and otherwise healthy, recently menopausal women meeting the standard and stringent clinical and biochemical criteria for menopause. In addition, the design allowed for a direct comparison between the two modalities of menopausal HT commonly used in current clinical practice—o-CEE and t-E2. The differences in pharmacokinetics and dynamics between the products may help to direct clinical decisions related to which formulation best meets the women's needs. A second strength of the study is the use of a validated tool for self-reporting global sleep quality and domains of sleep. The PQSI provided important qualitative information on the domains of sleep that cannot be obtained by laboratory analysis with important implications for women whose major sleep complaints relate to their ability to fall asleep, sleep disturbances, and overall sleep satisfaction. Because data regarding VMS was by self-report, responses of symptom severity are susceptible to misclassification bias and may under or over-represent the symptoms. However, the associations between sleep and symptom severity are, to our knowledge, the first reported with use of low doses and two different routes of two different estrogen products in the same study. A limitation of the present study is that other stressors related to sleep quality such as marital, employment and socioeconomic status, allergies, caffeine intake, and numbers of children in the home were not considered. Although the study could be criticized for eliminating the domain of sleep medication, the use of such medications was low at baseline, limiting statistical power, and their use is not well-established with health outcomes.15 Additionally, no clinical assessments were obtained regarding other potential confounders of sleep such as emotional stress, obstructive sleep apnea, or restless leg syndrome. As with other studies, it is important not to generalize the results of this study to other groups not defined by the inclusion and exclusion criteria.
In recently menopausal women, the overall sleep quality was improved by both HT regimens compared with PBO, with the transdermal estrogen formulation performing modestly better than the oral formulation. Of the studied sleep domains, sleep satisfaction, disturbances, and duration were improved, with sleep disturbances more improved by t-E2 than by o-CEE. Domains of sleep latency, efficiency, and daytime dysfunction were not affected by the HT regimes used in this study. Alleviation of VMS was associated with improvements in overall sleep quality. This result in not an unexpected outcome as it is common to use amelioration of VMS as a clinical guide to treatment. In a subset of women reporting moderate to severe VSM at baseline, reduced symptoms also correlated with improvement in the domain of sleep latency. These findings suggest that at least one way to approach the use of HT for sleep complaints is to assess the severity of VMS, and perhaps explore additional underlying problems affecting sleep, for example, obstructive sleep apnea.
Sleep disorders in midlife women warrant evaluation because treatment can lead to substantial improvements in quality of life and health outcomes.1,17 These results from a sufficiently powered, randomized clinical trial suggest that fostering a conversation about sleep quality, and sleep domains during clinical encounters may be a more appropriate guide for a patient-centered approach for achieving optimal sleep health.
The study would not be possible without the dedicated volunteers participating in this study; and collaborators and co-workers at each study center who include the following: Albert Einstein College of Medicine: Ruth Freeman, Hussein Amin, Barbara Isaacs, Maureen Magnani; Brigham and Women's Hospital/Harvard Medical School: Maria Bueche, Marie Gerhard-Herman, Kate Kalan, Jan Lieson, Kathryn M. Rexrode, and Frank Rybicki; Columbia College of Physicians and Surgeons: Luz Sanabria, Maria Soto, Michelle P. Warren, and Ralf C. Zimmerman; Kronos Longevity Research Institute: Mary Dunn; Mayo Clinic: Philip A. Araoz, Rebecca Beck, Dalene Bott-Kitslaar, Sharon L. Mulvagh, Teresa G. Zais (deceased); University of California, Los Angeles, CAC Reading Center: Chris Dailing, Yanlin Gao, Angel Solano; University of California, San Francisco: Nancy Jancar, Grechen Good; Statistical Reading Center: Lisa Palermo; University of Southern California, Atherosclerosis Research Unit: Yanjie Li; University of Utah School of Medicine: M. Nazeem Nanjee, Paul N. Hopkins, Kirtly Jones, Timothy Beals, Stacey Larrinaga-Shum; VA Puget Sound Health Care System and University of Washington School of Medicine: Pamela Asberry, SueAnn Brickle, Colleen Carney, Molly Carr, Monica Kletke, and Lynna C. Smith; Yale University, School of Medicine: Kathryn Czarkowski, Linda MacDonald, Mary Jane Minkin, Lubna Pal, Diane Wall, and Erin Wolff from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and the NIH Roadmap for Medical Research. Contents of this paper are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH. Information on NCRR is available at http://www.ncrr.nih.gov/. Information on Reengineering the Clinical Research Enterprise can be obtained from http://nihroadmap.nih.gov.
1. Luyster FS, Strollo PJ Jr, Zee PC, Walsh JK. Boards of Directors of the American Academy of Sleep M, the Sleep Research S. Sleep: a health imperative. Sleep
2. Moline ML, Broch L, Zak R, Gross V. Sleep in women across the life cycle from adulthood through menopause. Sleep Med Rev
3. Im EO, Ko Y, Chee E, Chee W, Mao JJ. Midlife women's cardiovascular symptoms: a cluster analysis. Healthcare Women Int
4. Biglia N, Cagnacci A, Gambacciani M, Lello S, Maffei S, Nappi RE. Vasomotor symptoms
in menopause: a biomarker of cardiovascular disease risk and other chronic diseases? Climacteric
5. Keefe DL, Watson R, Naftolin F. Hormone replacement therapy may alleviate sleep apnea in menopausal women: a pilot study. Menopause
6. Stuenkel CA, Davis SR, Gompel A, et al. Treatment of symptoms of the menopause: an Endocrine Society Clinical Practice guideline. J Clin Endocrinol Metab
7. de Villiers TJ, Hall JE, Pinkerton JV, et al. Revised global consensus statement on menopausal hormone therapy. Maturitas
8. Bianchi MT, Kim S, Galvan T, White DP, Joffe H. Nocturnal hot flashes
: relationship to objective awakenings and sleep stage transitions. J Clin Sleep Med
9. Rapkin AJ. Vasomotor symptoms
in menopause: physiologic condition and central nervous system approaches to treatment. Am J Obstet Gynecol
10. Shaver JL, Woods NF. Sleep and menopause: a narrative review. Menopause
11. Freeman EW, Sammel MD, Gross SA, Pien GW. Poor sleep in relation to natural menopause: a population-based 14-year follow-up of midlife women. Menopause
12. Cintron D, Lipford M, Larrea-Mantilla L, et al. Efficacy of menopausal hormone therapy on sleep quality: systematic review and meta-analysis. Endocrine
13. Tao MF, Sun DM, Shao HF, Li CB, Teng YC. Poor sleep in middle-aged women is not associated with menopause per se. Braz J Med Biol Res
14. Jehan S, Masters-Isarilov A, Salifu I, et al. Sleep disorders in postmenopausal women. J Sleep Disord Ther
2015; 4: pii:1000212.
15. Buysse DJ. Sleep health: can we define it? Does it matter? Sleep
16. Xu Q, Lang CP. Examining the relationship between subjective sleep disturbance and menopause: a systematic review and meta-analysis. Menopause
17. Joffe H, Massler A, Sharkey KM. Evaluation and management of sleep disturbance during the menopause transition. Semin Reprod Med
18. Archer DF, Sturdee DW, Baber R, et al. Menopausal hot flushes and night sweats
: where are we now? Climacteric
19. Fu P, Matthews KA, Thurston RC. How well do different measurement modalities estimate the number of vasomotor symptoms
? Findings from the Study of Women's Health Across the Nation FLASHES Study. Menopause
20. Pimenta F, Leal I, Maroco J, Ramos C. Menopause Symptoms’ Severity Inventory (MSSI-38): assessing the frequency and intensity of symptoms. Climacteric
21. Harman SM, Brinton EA, Cedars M, et al. KEEPS: The Kronos Early Estrogen Prevention Study. Climacteric
22. Wharton W, Gleason CE, Miller VM, Asthana S. Rationale and design of the Kronos Early Estrogen Prevention Study (KEEPS) and the KEEPS cognitive and affective sub study (KEEPS Cog). Brain Res
23. Gleason CE, Dowling NM, Wharton W, et al. Effects of hormone therapy on cognition and mood in recently postmenopausal women: findings from the randomized, controlled KEEPS-Cognitive and Affective Study. PLoS Med
24. Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index
: a new instrument for psychiatric practice and research. Psychiatry Res
25. Mollayeva T, Thurairajah P, Burton K, Mollayeva S, Shapiro CM, Colantonio A. The Pittsburgh sleep quality index
as a screening tool for sleep dysfunction in clinical and non-clinical samples: a systematic review and meta-analysis. Sleep Med Rev
26. Ensrud KE, Guthrie KA, Hohensee C, et al. Effects of estradiol
and venlafaxine on insomnia symptoms and sleep quality in women with hot flashes
27. Santoro N, Allshouse A, Neal-Perry G, et al. Longitudinal changes in menopausal symptoms comparing women randomized to low-dose oral conjugated estrogens or transdermal estradiol
plus micronized progesterone versus placebo: the Kronos Early Estrogen Prevention Study. Menopause
28. Ensrud KE, Joffe H, Guthrie KA, et al. Effect of escitalopram on insomnia symptoms and subjective sleep quality in healthy perimenopausal and postmenopausal women with hot flashes
: a randomized controlled trial. Menopause
29. Miller VM, Black DM, Brinton EA, et al. Using basic science to design a clinical trial: baseline characteristics of women enrolled in the Kronos Early Estrogen Prevention Study (KEEPS). J Cardiovasc Transl Res
30. Miller VM, Petterson TM, Jeavons EN, et al. Genetic polymorphisms associated carotid artery intima-media thickness and coronary artery calcification in women of the Kronos Early Estrogen Prevention Study. Physiol Genomics
31. Harman SM, Black DM, Naftolin F, et al. Arterial imaging outcomes and cardiovascular risk factors in recently menopausal women: a randomized trial. Ann Intern Med
32. North American Menopause Society. Estrogen and progestogen use in peri- and postmenopausal women: March 2007 position statement of The North American Menopause Society. Menopause
33. Kravitz HM, Joffe H. Sleep during the perimenopause: a SWAN story. Obstet Gynecol Clin North Am
34. Roth T. Insomnia: definition, prevalence, etiology, and consequences. J Clin Sleep Med
2007; 3 (5 Suppl):S7–S10.
35. Carpenter JS, Andrykowski MA. Psychometric evaluation of the Pittsburgh Sleep Quality Index
. J Psychosom Res
36. Ameratunga D, Goldin J, Hickey M. Sleep disturbance in menopause. Intern Med J
37. Kay DB, Buysse DJ, Germain A, Hall M, Monk TH. Subjective-objective sleep discrepancy among older adults: associations with insomnia diagnosis and insomnia treatment. J Sleep Res
Keywords:© 2018 by The North American Menopause Society.
Conjugated equine estrogens; Estradiol; Hot flashes; Night sweats; Pittsburgh Sleep Quality Index; Vasomotor symptoms