Objective
Menopause-associated insomnia is commonly associated with other symptoms (hot flashes, depression, anxiety). Given frequent symptom cooccurrence, therapies targeting sleep may provide an important approach to treatment during midlife.
Study Design
Peri/postmenopausal women (40-65 years old) with sleep-onset and/or sleep-maintenance insomnia cooccurring with hot flashes and depressive and/or anxiety symptoms were randomized to eszopiclone 3 mg orally or placebo in a double-blinded, crossover 11 week trial. Changes in the Insomnia Severity Index (ISI) scale and secondary outcomes (diary-based sleep parameters, depression/anxiety, hot flashes, quality of life) were analyzed using repeated-measure linear models.
Results
Of 59 women, 46 (78%) completed the study. Eszopiclone reduced ISI scores by 8.7 ± 1.4 more points than placebo ( P < .0001). Eszopiclone improved ( P < .05) all sleep parameters, depressive symptoms, anxiety symptoms, quality of life, and nighttime but not daytime hot flashes.
Conclusion
Eszopiclone treats insomnia and cooccurring menopause-related symptoms. Our results provide evidence that hypnotic therapies may improve multiple domains of well-being during midlife.
Sleep disturbance is a core symptom of the menopause transition, with poor sleep quality reported more commonly by perimenopausal and postmenopausal women than by older premenopausal women. The prevalence of an insomnia disorder increases from 13% in older premenopausal women to 26% in peri- and postmenopausal women. Hot flashes are the primary symptom of the menopause transition, and women with nocturnal hot flashes report repeated awakenings. Women with hot flashes report worse sleep quality, and are more likely to meet criteria for insomnia, than those without hot flashes.
Insomnia involves severe and persistent sleep disturbance that either induces marked distress or has a deleterious effect on daytime function or well-being, which commonly includes anxiety and depressive symptoms during the menopause transition. The risk for anxiety and depressive symptoms is increased during the perimenopause and postmenopause.
Menopausal symptoms of hot flashes, insomnia, depression, and anxiety are all associated with worse quality of life among midlife women. Given that insomnia commonly cooccurs with these other menopause-associated symptoms, therapies targeting sleep disturbance may improve quality of life overall. We have previously shown that the nonbenzodiazepine sedative-hypnotic eszopiclone is an effective treatment of sleep-onset insomnia in peri- and early postmenopausal women.
In the current study, we conducted a double-blind, placebo-controlled, crossover trial to examine the efficacy of eszopiclone in peri- and postmenopausal women with sleep-onset and/or sleep-maintenance insomnia. Participants also presented with cooccurring hot flashes, depressive and/or anxiety symptoms. We hypothesized that eszopiclone would treat sleep-onset and/or sleep-maintenance problems in this population and also improve cooccurring symptoms of nocturnal hot flashes and depressive and anxiety symptoms, thereby improving quality of life.
Materials and Methods
Subjects
Sixty-five perimenopausal and postmenopausal women 40-65 years old were enrolled in an 11 week randomized, double-blind, placebo-controlled, crossover study, which was approved by the Partners Healthcare Systems Institutional Review Board, with written informed consent obtained from all subjects.
Subjects were either perimenopausal (early or late menopausal transition) or postmenopausal (amenorrhea >12 months) or had bilateral oophorectomy. Women who had undergone a hysterectomy with ovaries preservation were included if their follicle-stimulating hormone level was greater than 20 IU/L.
Women were eligible if they met insomnia criteria because of difficulty initiating (>30 minutes) and/or maintaining sleep (wake time after sleep onset >30 minutes) 3 or more nights per week for 1 month or longer prior to enrollment, confirmed prospectively with a sleep diary during a 7 day run-in period. Hot flashes were assessed daily during the run-in but not required for eligibility.
Eligible participants had evidence of impaired daytime function or well-being as a result of insomnia, based on a clinical interview conducted by study psychiatrists, as well as mild depressive and/or anxiety symptoms, defined by a Montgomery-Åsberg Depression Rating Scale (MADRS) score of 10-18 and a Beck Anxiety Inventory (BAI) score of 10-18.
The MADRS is a widely used clinician-rated questionnaire of depression symptoms over the past 7 days (score of 0-60, with scores <10 suggesting no depression and scores >18 implying clinically significant depression). The BAI is a widely used self-report measure of anxiety over the past 7 days (score of 0-63, with a similar interpretation of scores).
The Mini-International Neuropsychiatric Interview structured psychiatric interview was administered by study psychiatrists to exclude women with psychiatric disorders (major depression, dysthymia, panic disorder, or posttraumatic stress disorder in past 3 months) or who had current suicidal ideation, homicidal ideation, psychotic symptoms, or a suicide attempts or substance-use disorder within the past 5 years. Women who developed significant depressive symptoms (MADRS ≥18) or anxiety symptoms (BAI ≥18) during the study were withdrawn.
Other exclusion criteria were previously diagnosed sleep apnea, periodic leg movement syndrome, or medical conditions affecting sleep. Use of hypnotic agents in the 4 weeks before enrollment was prohibited, but antidepressants and hormone therapy were allowed if the dose remained stable 8 weeks or longer before study entry and throughout the study.
Procedures
The overall study design involved an 11 week randomized, double-blind, crossover period to fixed-dose eszopiclone 3 mg nightly (Food and Dug Administration–approved adult dose) and identical matching placebo patches ( Figure 1 ), which were provided by the manufacturer. After eligibility was confirmed, subjects were randomized in a 1-to-1 ratio to 4 weeks of treatment with eszopiclone first (es→plb) or placebo first (plb→esz), followed by a 2 week washout during which no medications were administered, and then crossed over to the other treatment for another 4 weeks.
Treatment effects on sleep parameters were assessed with the Insomnia Severity Index (ISI) and a sleep diary. The ISI is a widely used, 7 item self-report questionnaire addressing insomnia symptoms (scores 0-7 indicating no clinically significant insomnia; 8-14, subthreshold insomnia; 15-28, clinical insomnia) with good reliability and validity for insomnia. Sleep parameters calculated from the sleep diary were wake time after sleep onset (WASO), sleep efficiency (percent of time between bedtime and wake-up time spent asleep), sleep-onset latency (minutes), total sleep time (TST), and number of awakenings. Average daily scores were calculated for each 7 day period.
Secondary measures included the following: (1) number of hot flashes/night sweats, (2) depressive symptoms, (3) anxiety symptoms, (4) menopause-related quality of life, and (5) functional impairment. The frequency of daytime and nighttime hot flashes was tabulated separately throughout the study using a diary. Menopause-related quality of life was assessed with the Menopause Quality of Life Scale (MENQOL), an instrument that assesses quality of life over the prior month (score 0-8, with higher scores worse). The MADRS was used to measure changes in depressive symptoms, and the BAI was used to measure changes in anxiety symptoms. The Sheehan Disability Scale (SDS, range 0–30, higher score indicates greater disability) measures functional impairment related to current symptoms.
Statistical analysis
The primary analysis was conducted using an intent-to-treat, repeated-measure, linear mixed model, with change in the ISI score from the beginning to end of each treatment period as the dependent measure, treatment and time as main effects, and a time-by-treatment interaction term to account for potential carryover effects between treatment periods. The study was powered (90%) based on the ISI, assuming a true difference between groups of 3.6 and an SD of the difference of 8.3 (2 sided α = 0.05).
The same analytic approach was used to evaluate the efficacy of eszopiclone on secondary outcome measures, including changes in sleep parameters (WASO, sleep efficiency, sleep-onset latency, TST) and changes in depressive symptoms (MADRS), anxiety symptoms (BAI), daytime and nighttime hot flashes, quality-of-life (MENQOL), and functional impairment (SDS).
All regression models included a time-by-treatment interaction term (significant if P < .10). There was no significant interaction between time and treatment for the primary outcome measure. For those secondary outcome measures with significant interactions (TST, WASO, sleep efficiency, MADRS scores, nighttime hot flashes), separate analyses for each treatment period were conducted using Student t tests.
Mixed-effect linear regression models were built to determine the association between improvement in sleep parameters and other menopause-related symptoms after adjusting for treatment assignment, treatment period, and their interaction.
Analyses were conducted using SAS statistical software version 9.1 (SAS Institute Inc, Cary, NC) with statistical significance at the 2-sided α = 0.05 level.
Results
Fifty-nine of 65 women who completed the run-in (91%) were eligible to be randomized and 6 (9%) were excluded ( Figure 1 ), with 30 (50.8%) assigned to the esz→plb group and 29 (49.2%) to the plb→esz group. Overall, 46 of 59 women initiating treatment (78%) completed the study (90% of the esz→plb group and 65.5% of the plb→esz group).
More women in the plb→esz group than the esz→plb group withdrew from the study (Fisher’s exact, P = .03), although there was no difference in the proportion withdrawing early because of side effects. For the esz→plb group, study noncompleters withdrew because of side effects (n = 2) and logistical reasons (n = 1). Reasons for early withdrawal in the plb→esz group were side effects (n = 1), logistical reasons (n = 3), initiation of disallowed medications (n = 2), loss to follow-up (n = 3), and increase in depressive symptoms (n = 1).
Subject characteristics
Demographic and menopause characteristics did not differ between groups ( Table 1 ). The mean (±SD) age overall was 52.2 ± 4.6 years and the mean body mass index was 27.5 ± 6.6 kg/m 2 . The majority were white (71.2%) and naturally postmenopausal (52.5%), 40.7% were married, and 40.7% had at least a college education. Only a small minority of participants was using hormonal therapy (8.5%) or antidepressants (15.3%). Whereas hot flashes were not an eligibility criterion, all women reported them during the run-in (2.4 ± 2.0 daytime hot flashes, 2.1 ± 1.5 nighttime hot flashes).
| Demographic | Eszopiclone then placebo (n = 30) | Placebo then eszopiclone (n = 29) |
|---|---|---|
| Age, mean ± SD | 51.8 ± 4.7 | 52.6 ± 4.5 |
| White, n (%) | 20 (66.7) | 22 (75.9) |
| College degree, n (%) | 9 (30.0) | 15 (51.7) |
| Marital status, n (%) | ||
| Never married/single | 8 (26.7) | 5 (17.2) |
| Married/living with partner | 14 (47.6) | 10 (34.5) |
| Separated/divorced/widowed | 8 (26.7) | 14 (48.3) |
| Body mass index, kg/m 2 | 27.4 ± 6.8 | 27.8 ± 7.0 |
| Menopause status, n (%) | ||
| Perimenopausal | 8 (26.60) | 10 (34.5) |
| Natural postmenopause | 17 (56.7) | 14 (48.3) |
| Surgical postmenopause | 0 (0.0) | 1 (3.5) |
| Hysterectomy only | 5 (16.7) | 4 (13.8) |
| FSH levels, mean ± SD | 67.3 ± 41.5 | 63.4 ± 35.9 |
| Concurrent medication use | ||
| Antidepressant | 6 (20.0) | 3 (10.3) |
| Hormonal therapy b | 4 (13.3) | 1 (3.4) |
| Sleep problems meeting eligibility criteria, n (%) | ||
| Difficulty initiating sleep | 6 (20.0) | 2 (6.9) |
| Difficulty maintaining sleep | 5 (16.7) | 6 (20.7) |
| Difficulty initiating and maintaining sleep | 19 (63.3) | 21 (72.4) |
| Vasomotor symptoms, mean ± SD | ||
| Number of nighttime hot flashes | 2.0 ± 1.5 | 2.2 ± 1.6 |
| Number of daytime hot flashes | 2.68 ± 2.2 | 2.25 ± 1.8 |
| Mood characteristics meeting eligibility criteria | ||
| Significant depressive symptoms only, n (%) | 18 (60.0) | 13 (44.8) |
| MADRS score for the subgroup | 11.8 ± 2.3 | 11.7 ± 2.4 |
| Significant anxiety symptoms only, n (%) | 2 (6.7) | 2 (6.9) |
| BAI score for the subgroup | 11.0 ± 1.1 | 14.5 ± 4.9 |
| Both depressive and anxiety symptoms, n (%) | 10 (33.3) | 14 (48.3) |
| MADRS score for the subgroup | 13.3 ± 3.3 | 13.6 ± 2.6 |
| BAI score for the subgroup | 12.9 ± 2.9 | 14.6 ± 2.7 |
| Baseline measures prior to period 1, mean ± SD | ||
| Sleep measures | ||
| ISI | 15.6 ± 3.4 | 15.3 ± 5.0 |
| Wake time after sleep onset, min | 68.8 ± 48.7 | 88.3 ± 45.5 |
| Number of awakenings | 2.2 ± 1.4 | 2.5 ± 1.3 |
| Sleep latency, min | 56.9 ± 45.7 | 48.0 ± 36.6 |
| Sleep efficiency, % | 74.3 ± 14.6 | 70.0 ± 12.3 |
| Total sleep time, min | 356.8 ± 70.6 | 325.8 ± 75.0 |
| Mood measures | ||
| MADRS score | 11.4 ± 3.9 | 10.6 ± 3.8 |
| BAI score | 6.6 ± 4.5 | 7.3 ± 4.6 |
| Vasomotor symptoms | ||
| Number of nighttime hot flashes | 2.0 ± 1.5 | 2.2 ± 1.6 |
| Number of daytime hot flashes | 2.68 ± 2.2 | 2.25 ± 1.8 |
| Quality of life and functional disability | ||
| Functional disability (SDS) | 9.6 ± 5.7 | 8.4 ± 4.8 |
| Quality of life (MENQOL) | 4.4 ± 1.3 | 4.1 ± 1.4 |
| Baseline measures prior to period 2, mean ± SD | ||
| Sleep measures | ||
| ISI | 12.2 ± 4.6 | 14.2 ± 5.2 |
| Wake time after sleep onset, min | 43.5 ± 46.0 | 60.6 ± 42.3 |
| Number of awakenings | 1.4 ± 1.3 | 1.7 ± 1.4 |
| Sleep latency, min | 48.8 ± 49.7 | 47.4 ± 38.7 |
| Sleep efficiency, % | 82.0 ± 14.5 | 76.0 ± 15.5 |
| Total sleep time, min | 385.9 ± 81.2 | 345.0 ± 73.6 |
| Mood measures | ||
| MADRS score | 8.3 ± 4.2 | 10.5 ± 6.1 |
| BAI score | 4.1 ± 2.8 | 6.2 ± 4.5 |
| Vasomotor symptoms | ||
| Number of nighttime hot flashes | 1.8 ± 1.7 | 1.9 ± 1.8 |
| Number of daytime hot flashes | 2.4 ± 2.1 | 1.6 ± 1.8 |
| Quality of life and functional disability | ||
| Functional disability (SDS) | 7.3 ± 6.9 | 7.5 ± 6.7 |
| Quality of life (MENQOL) | 3.6 ± 1.0 | 3.3 ± 1.3 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
