Abstract
Sleeping problems are a serious public health problem, imposing a substantial burden on individuals and society. Although sleeping problems occur throughout the lifespan, and in both sexes, menopause can be considered as one important milestone of increasing occurrence in sleeping problems. However, to determine whether sleeping problems are caused by the menopause or merely occur by coincidence during the menopause is not always easy because several, particularly age-related, changes take place at the same time. The most important factors are general diseases, medications, weight changes and mood symptoms. According to women's own judgment, hormone therapy significantly improves sleep quality. Hormone therapy can thus be considered as a first-line treatment for climacteric sleeping problems. If sleeping problems are accompanied by other disorders, hormone therapy should be kept in mind as an adjuvant therapy. According to worldwide consensus on hormone therapy, the main indication of hormone therapy is alleviation of climacteric symptoms, including climacteric sleeping problems. However, when choosing hormone therapy for a patient, contraindications and possible long-term side effects should be individually considered. This review illustrates the effect of menopause on sleep and evaluates different treatment options, especially hormone therapy, in alleviation of symptoms.
Keywords
Menopause predisposes women for several symptoms that significantly decrease quality of life. One of these is sleeping problems, especially insomnia. The clinical picture of menopausal insomnia is similar to common insomnia, which is a nonspecific complaint of trouble in falling asleep (long sleep latency), difficulty staying asleep (excessive or prolonged awakenings) or awakening involuntary too early in the morning [1–2]. The burden of insomnia is especially drawn from the consequences, such as fatigue, sleepiness, irritability, mood disorders, memory troubles, lack of concentration or disability in daytime functioning [3], which are similar in menopausal women and the population in general.
It is uncertain whether sleeping problems in the menopausal transition are caused by declining sex hormone levels or solely by aging or by both. Age has fundamental deteriorating effects on sleep through, for example, neuronal loss and atrophy, neurotransmitter defects and decreasing cerebral blood flow [4]. These changes take place in both genders. On the other hand, sex hormone receptors, especially estrogen receptors, have been found in brain areas responsible in sleep regulation [5]. Sex hormones could thus influence sleep, for instance via several neurotransmitters [6]. Climacteric vasomotor symptoms are the most apparent consequences of decreasing sex hormone levels. Those symptoms have been clearly shown to be associated with worse subjective sleep quality [7]. Furthermore, when related to anxiety, depression, stress and tension, symptoms may cause or at least contribute to sleeping problems [8–9]. In addition to these factors, sleep-disordered breathing is regarded a primary cause of sleep disturbances during menopause [10]. To all these symptoms hormone therapy (HT) brings at least some degree of alleviation [11–12]. The best result of HT is achieved if it is started soon after menopause. At that time the side-effects are also considered to be minimized, especially regarding cardiovascular function. In long-duration use, however, evaluation of the benefits and risks of the treatment should be considered individually.
Characteristics of sleep
Sleep alternates with wakefulness in cycles. Characteristics for sleep are reduced awareness and responsiveness, as well as motor inhibition. The length of sleep varies considerably between individuals. The average need is 7.5–8.5 h daily, and thus sleep occupies approximately a third of each person's lifetime. The functions of sleep are only partly understood [13]. Sleep is considered to be important in energy conservation, thermoregulation and tissue recovery [14]. In addition, it is essential for cognitive performance, particularly memory consolidation [14–15]. The impact of good sleep on health remains generally overlooked and the importance of sleep is better understood through sleep deprivation or sleep disorders, which lead to excessive daytime sleepiness and related problems.
In sleep studies, polysomnography or actigraphy are the most common tools to measure objective sleep quality. In electrophysiological terms, wakefulness and sleep are differentiated by changes in electroencephalogram (EEG), electro-oculogram and electromyogram. Typical for sleep state is increased synchrony of the EEG accompanied by low muscle tone. Sleep consists of non-rapid eye movement (NREM) and REM sleep, which appear in cycles. NREM sleep is conventionally subdivided into four stages (stages 1, 2, 3 and 4), of which stages 3 and 4 are usually combined and known as slow-wave sleep or deep sleep [13]. Total sleep time, sleep efficiency (time spent asleep from total time in bed), proportion of different sleep stages, sleep latencies and number of awakenings are the most frequently measured variables.
Two major factors regulate sleep: the circadian factor, which facilitates falling asleep at a certain time of the day (usually in the evening), and homeostatic factor, which explains why falling asleep is easier after prolonged wakefulness [16]. Several neurotransmitters are involved in sleep–wake state regulation [13]. Breathing during sleep is crucial for sleep quality. Control of breathing is essentially different during wakefulness, NREM sleep and REM sleep. During NREM sleep, the tidal volume and the frequency of breathing are steady whereas REM sleep is characterized by irregular breathing [17].
Effect of female sex-hormone in the brain
In addition to reproductive behaviors, female sex steroids control a multitude of functions in the brain, including sleep, cognitive performance, mood, movement coordination and pain [18–20]. Steroid receptors are located in several brain areas, such as in cortex, hippocampus, hypothalamus, amygdala, basal forebrain, midbrain Raphe nuclei, pituitary gland, locus coeruleus and cerebellum [21–22].
Several neurotransmitters, such as acetylcholine, serotonin, dopamine, adrenaline, noradrenaline, γ-amino-butyric acid and opiate can trigger estrogen receptors. Furthermore, estrogen may restore CNS-related circadian hormones such as growth hormone, glucocorticoid, prolactin or melatonin [23]. The effects of estrogen on brain function have been partly solved in recent decades; however, the neurosteroid effects of progesterone are mostly unanswered. The sedative actions of progesterone are documented in both animals and humans [24]. Progesterone has also been shown to have respiratory stimulant properties [25–26].
Characteristics of climacterium
The age range for natural menopause is 45–55 years, with an average age of 51–52 years [27]. Menopause is defined as the cessation of menstrual period. Perimenopause is the time from the first signs of approaching menopause until 12 months of permanent amenorrhea [28]. Climacterium encompasses the perimenopause and the part of the post-menopausal period during which climacteric symptoms occur [29].
Vasomotor instability, hot flushes and sweating are the most typical climacteric symptoms. Peripheral vasodilatation occurs, particularly around the face and the chest, which is usually followed by sweating, causing evaporative cooling [30]. The severity and duration of the vasomotor symptoms has a wide range. They generally occur for 1–2 years, but approximately 25% of women will experience them for 5 years and 9% practically lifelong [31–32].
Other climacteric symptoms include physical symptoms such as palpitation, headache, dizziness, numbness, myalgia, vaginal dryness and urinary tract symptoms. Mental symptoms, such as anxiety, depression, decline in libido, lack of concentration and memory impairment also occur, and occasionally they exceed the severity of vasomotor symptoms. Furthermore, sleeping problems are common. They may arise as exclusive symptoms or more typically in addition to other climacteric symptoms. In menopausal women vasomotor symptoms are reported in 65–85%, mental symptoms in 50–80% and sleeping problems in 50–80% [33–34].
Sleep & climacteric symptoms
Women report sleeping problems more often than men in all age ranges [35–36]. In a questionnaire study with over 12,000 participants, both men and women, 15% of women aged over 50 years suffered from severe insomnia whereas only 5% of women aged 18–24 years reported the same. In men the corresponding numbers were 8 and 2% [36].
Several studies suggest a significant increase in sleeping problems during menopause. In a study by Baker and colleagues, perimenopausal women reported more frequent and longer arousals than premenopausal women, resulting in less sleep. Mood symptoms were more common in perimenopausal women and associated with sleep disturbance [37]. In another study with 100 symptomatic menopausal women, nearly 80% complained of insomnia and over 90% suffered from fatigue. The typical complaints included too early morning awakenings or intermittent sleep [38]. According to a French survey with 1000 responders, the odds ratio (OR) for sleeping problems after controlling for age was 1.5 in postmenopausal women compared with menstruating women [39]. In another study with over 1200 participants in UK, the risk for sleep disturbance was even higher: 1.5 in perimenopausal women and 3.4 in postmenopausal women compared with premenopausal women [40]. Furthermore, in a multicenter Survey of Women's Health Across the Nation, ORs for trouble sleeping were 1.6 for postmenopausal and 1.3 for perimenopausal compared with premenopausal women [41].
Several studies support an association between self-reported sleeping problems and climacteric symptoms [7,34,41–42]. Oldenhave and colleagues reported a definite correlation between insomnia and vasomotor symptoms in a study with over 5000 women [34]. In a more resent study with 12,600 women the OR for sleeping problems in women with climacteric symptoms was 2.0 compared with asymptomatic women [41].
There are only a few studies that have aimed to evaluate the relationship between objectively measured sleep quality and climacteric symptoms [7,43–47]. Presumably due to the different research techniques, particularly in recording climacteric symptoms, the results have been inconsistent. In Shaver and colleagues' study, symptomatic women had longer time in bed and longer REM latency compared with asymptomatic women [43]. Erlik and colleagues found an association between hot flushes and arousals [45]. Three other studies could not characterize any specific abnormalities in polysomnography in connection with climacteric symptoms [7,44,46]. Freedman and colleagues reported that hot flushes disclosed by increased sternal skin conductance did not correlate with objective sleep quality [47]. Recent data suggest that electrophysiological changes in polysomnography may already take place several years before menopause, not during the menopausal transition [Kalleinen and colleagues, Unpublished data], which could explain the neutral outcomes between objectively measured sleep quality and climacteric symptoms.
Effect of hormone therapy on sleep quality
As climacteric symptoms, especially vasomotor symptoms, are derived from a lack of female sex hormones, HT is a first-line treatment to control climacteric symptoms [33,48]. HT has also been found to have a preventive action in a osteoporosis [49]. Owing to the side-effects and complications associated particularly with long-duration HT in older women [50], recommendations for HT use include only alleviation of climacteric symptoms with short duration of the treatment. With regard to control of climacteric sleeping problems HT has been found effective [33,48,50–51]. In one study 4 weeks treatment with HT significantly decreased sleeping problems [48]. In that study, alleviation of the vasomotor symptoms strongly associated with improvement in sleep quality. In a recent large, randomized, placebo-controlled trial of the Women's Health Initiative, HT also marginally alleviated sleeping problems, although enrolment of the participants excluded women with moderate to high climacteric symptoms [50].
In previous literature, evaluation of the effect of HT on subjective sleep quality has been only one minor part of the study and trials designed merely to evaluate sleep quality are limited [44,51]. In a study with postmenopausal women with or without vasomotor symptoms, estrogen facilitated falling asleep and decreased nocturnal restlessness and awakenings, as well as decreasing tiredness in the morning and during the daytime. Although sleep quality improved more in symptomatic women than in asymptomatic women and although the degree of improvement in vasomotor symptoms was an important predictor for the degree of improvement in sleep disturbance, the subgroup of asymptomatic women with at least some degree of insomnia also reported improved sleep quality during HT [51].
The beneficial effects of HT on subjective sleep quality are easily explained in symptomatic women, in whom sleeping problems are obviously secondary to vasomotor symptoms. In the case of asymptomatic women, there are two possible explanations. First, women may underestimate or not recognize their climacteric symptoms and thus alleviation of the vasomotor symptoms again plays an important role in improving sleep quality. Second, decreased sex hormone levels associated with menopause may interfere with sleep regulation in the CNS, causing sleeping disturbance. Using HT, premenopausal hormone levels can at least partly be achieved.
Though HT is beyond dispute shown to alleviate subjective sleeping problems and is thus regarded as a first-line treatment for sleeping symptoms during menopause, the results in HT studies evaluating objective sleep quality are inconsistent. An increase in REM sleep and a reduction of awakenings have been reported, as well as a decrease in nocturnal wakefulness during the entire night or in the first sleep circles. Moreover, a shortening of sleep latency, an improvement in sleep efficiency and a reduction of the rate of cyclic alternating patterns of sleep have also been reported. In some studies no improvement whatsoever has been found [11,12,23]. In a very large study, postmenopausal women with HT had worse sleep quality compared with those without HT, as they had a smaller amount of slow-wave sleep, more stage 1 sleep and their sleep was more fragmented [44].
Conclusions regarding the effect HT on objective sleep quality are difficult, almost impossible, to draw due to fundamental differences in study designs, subject enrolments and administration of the treatments (form, dose and duration). Some researches have recruited perimenopausal women instead of postmenopausal women or both naturally and surgically menopausal women to the same sleep study. All prospective studies have been short in duration, from 4 weeks to 7 months, leaving possible long-term effects of HT unanswered. [23] In observational studies, which have also gained conflicting results, self-chosen use of HT might have influenced the outcomes [44,52,53]. The differences have been quite small, although statistically significant, which calls into question their real clinical significance. The exception to this is probably reduction of awakenings during HT [45,54–56]. All in all, even though women in general feel definite improvement in their sleep quality during HT, the data obtained from polysomnographic sleep studies do not unambiguously support this improvement. However, in clinical practice subjective sleep complaints are often sufficient for diagnosis and thus large and expensive polysomnographic sleep studies are necessary only on selected occasions.
Sleep-disordered breathing during menopause
Sleep-disordered breathing should not be dismissed as a possible reason for sleeping problems. Even young subjects may have a clinically significant nocturnal breathing disorder (apnea–hypopnea index >5). According to a large epidemiological study, the level of clinically important nocturnal breathing in women was 9% (age range 30–60 years), while in men it was 24% in the same age ranges. In women aged 30–39 years the level was as high as 6.5%, gradually increasing to 8.7% for 40–49 year-olds and to 16% in those aged 50–60 years. [57] The obstructive sleep apnea syndrome (OSAS) is clinically the most important breathing abnormality with an incidence of 1–2% in the general population. It is characterized by repeated episodes of upper-airway obstruction, leading either to an apnea or a marked airflow limitation (hypopnea). It is often accompanied by severe snoring and the diagnosis requires occurrence of daytime sleepiness. In women, especially in menopausal transition and after menopause, partial upper-airway obstruction seems to be a more common nocturnal breathing disorder than OSAS. Typical for partial upper-airway obstruction are hypoventilation and carbon dioxide retention. Symptoms resemble those of sleep apnea, for example, heavy snoring, excessive sleepiness, an irresistible tendency to fall asleep, sweating, morning headache, lack of energy, low initiation capacity, difficulties in concentration, poor memory and low mental tolerance, and may even be interpreted as climacteric symptoms [11–13,23].
In a large study with 1000 women from the general population, 3.9% of postmenopausal women had OSAS [10]. In a study with 63 healthy postmenopausal women a significant partial upper-airway obstruction was found in 17% of the study population [58]. Sleep-disordered breathing is caused by structural or physiological factors. After menopause a redistribution of body fat, especially an increase in the waist: hip ratio and neck circumference has been shown to occur [57]. Another proposed mechanism is the decrease in female sex hormones, particularly progesterone [59], which is known to have respiratory stimulant properties [25–26].
Based on this physiological hypothesis, some studies have attempted to evaluate whether HT would be effective as the main or at least the adjuvant treatment of nocturnal breathing problems after menopause. HT has been shown to reduce apnea–hypopnea-index [10,60]. In contrast, others could not verify the above findings, possibly as they included women using various HT preparations, including unopposed estrogen [61]. In a placebo-controlled crossover study with unopposed estrogen, little improvement in nocturnal breathing problems was found [58]. By contrast, Keefe and colleagues showed in their pilot study that both estrogen alone and combined HT decreased apnea–hypopnea index [62]. The different outcomes in these two studies may be explained by the differences in subject recruitment: Polo-Kantola and colleagues studied originally healthy women [58], whereas Keefe and colleagues enrolled exclusively OSAS patients [62].
High doses of medroxyprogesterone acetate (30 mg/day) have been shown to decrease duration of apneas [63]. Also, Saaresranta and colleagues reported significantly improved ventilation in postmenopausal women during high-dose medroxyprogesterone acetate therapy (60 mg/day) [64]. Whether HT or progesterone alone is a real treatment option remains an open question, since the studies available are small in number and mainly conducted without placebo groups. Therefore, nasal continuous positive airway pressure is the treatment of choice in sleep-disordered breathing in women.
Mood, menopause & sleep
Women report more mood symptoms, especially depression and anxiety, than men [65–66]. The influence of fluctuations in female sex hormones is plausible, since mood symptoms occur in some women connected to female reproductive cycle, especially premenstrually, after delivery or during menopause. A total of 70–90% of climacterically symptomatic women report depressive symptoms [38]. Lack of estrogen reduces brain serotonin concentration, a critical neurotransmitter for mood. Adrenergic and dopaminergic activity can also be altered, suggesting a potency of estrogen to act as an anti-depressive agent. Previous studies have suggested an association between mood symptoms and sleep disturbance in peri- and post-menopausal women [37,67]. Sleep is sensitive for mood disturbances, being frequently the first symptom of mental distress. Several sleep EEG findings have been shown in patients with depression, such as increased sleep latency or number of awakenings and arousals, as well as decreased sleep efficiency, slow-wave sleep or REM latency [68].
Special treatment for depression and connected sleeping problems in the menopausal transition are subtle. If the diagnosis is depressive disorder, the treatment of choice is standard psychiatric treatment, either pharmacological or psychological, or both. In women with vasomotor symptoms HT as an adjuvant therapy certainly improves compliance and probably recovery. If the diagnosis is solely depressive mood or mood swings, no specific drug should be initiated before treatment response to HT has been estimated. In case depression persists for 3 months despite successful HT, the indications for hormone should be re-evaluated and other forms of therapy initiated.
Other sleep disorders during menopause
Other important reasons for sleeping problems comprise stress and behavioral factors, as well as restless legs (RLS) and periodic limb movement syndromes (PLMS). Although these conditions are generally independent causes for sleeping problems, and probably just coincide with the menopause, menopause may further exacerbate the symptoms [13].
RLS and PLMS are movement disorders, which may lead to sleeping problems. The two conditions are distinguished in timing: RLS occurs during wakefulness, whereas PLMS occurs while asleep. Characteristics in RLS are unpleasant sensations, with an urge to move. In PLMS, abrupt and repetitive movements last 0.5–5 s with short intervals. RLS occurs in 5–15% of the adult population and PLMS in 30%. The prevalence increases with age and is equal in both genders. Etiologies include idiopathic (with highly genetic basis) to iron, magnesium or folate deficiency, renal failure, peripheral neuropathies and use of drugs, especially CNS stimulants and dopamine antagonists [13]. Estrogen has a dopaminergic action and thus could benefit these syndromes. Yet only one randomized, placebo-controlled study has evaluated the effect of estrogen, with no change on the frequency of PLMS during sleep [69].
Treatment of sleep disturbances during menopause
Whatever the origin, sleeping problems always demand to be cured or alleviated. In women with climacteric vasomotor symptoms, the first-line therapy for insomnia is HT. In addition, climacteric women whose insomnia is essentially related to mood swings often benefit from HT [51]. A subset of vasomotorically asymptomatic women may also benefit from HT [51], but careful screening of other underlying reasons for sleeping problems is essential. The doses used should be as low as possible, although high enough to abolish symptoms. In women aged over 60 years the vascular side effects of HT may exceed the favorable effects on sleep and caution should be exercized in starting HT [50]. In long-term HT, the increased risk for breast cancer should be considered and yearly breast examination accompanied with mammography every second year should be performed.
Other treatment options, that may alleviate vasomotor symptoms and connected sleeping problems, for example selective serotonin-reuptake inhibitors, gabapentin, dietary iso-favones and soy foods, could be considered, although the long-term effectiveness and safety are still uncertain [70–73]. Appropriate sleep hygiene is crucial, including a dark, quiet room with comfortable (often low) temperature and bed. Avoiding daytime napping, especially long ones, may also improve sleep. Intake of refreshing drugs (tea, coffee, some soft drinks and herbal drinks), smoking and alcohol may interfere with sleep and should thus be avoided, especially in the evening [74–75]. Relaxation training with different forms (progressive muscle relaxation, hypnosis, yoga or biofeedback) or stimulus–control therapy, especially before bedtime, should be included in primary interventions. Cognitive–behavioral or sleep restriction and sleep consolidation therapies may be of benefit for some women [75]. In severe or persistent insomnia, pharmacological options, for example, hypnotics, anxiolytics or antidepressants are often necessary [75].
Future perspective
Sleep problems are increasing. They are not only a source of unpleasant symptoms and daytime sleepiness, but are also connected with increase in systemic diseases, mental morbidity and accidents. In women menopause, especially when connected with other climacteric symptoms, often deteriorates sleep quality. After recent large HT trials, Women's Health Initiative [76], Heart and Estrogen/Progestin Replacement Study [77–78] and the Million Women Study [79], attitudes towards HT have become more negative. Although unfavorable effects of HT are mostly consequences of a long-term use, women with even severe climacteric symptoms try to cope without HT. This leads them to seek other treatment options, which in fact may be more harmful. Definite increase in use of hypnotic drugs, antidepressants and anxiolytics has been observed in several countries. In climacteric symptoms these seldom cover all symptoms, may be addictive and in a long run may even deteriorate sleep quality. In future the development of low-estrogen HT options would be warranted, because they offer more fluctuation in individual dosage, often provide sufficient efficacy and have fewer side effects.
Executive summary
Sleeping problems are common and their prevalence is increasing.
Women report sleeping problems and seek assistance for them more often than men in all age ranges.
Menopause, and especially menopause-associated symptoms, such as nocturnal hot flushes and sweating, often causes or worsens sleep disturbances.
After menopause other conditions predisposing for sleeping problems, such as nocturnal breathing problems or mood symptoms, also increase.
Today, use of hypnotic drugs, sedatives and antidepressants are increasing. One important indication for use is sleeping problems.
Hormone treatment (HT) significantly improves sleep quality and can thus be considered as a first-line therapy for menopausal sleeping problems, especially if other climacteric symptoms are also present. Compared with CNS-affecting medications, HT offers a more physiological and even natural treatment option for climacteric sleeping problems.
Careful follow-up of relief in symptoms during HT is necessary, in order to control compliance and possible side effects of the treatment. In case no respond for symptoms have been achieved within 2–3 months, other treatment options should be considered.
Some sleep disturbances just coincide with the the menopausal period, are more age-related, and thus cannot be cured by HT.
