© Geriatric Times. All rights reserved.
For the Clinician


Insomnia in Peri-Menopausal and Post-Menopausal Women

by Andrew D. Krystal, M.D., M.S.

Geriatric Times May/June 2001 Vol. II Issue 3

As women age, there is a gradual decrease in ovarian function that culminates in the onset of menopause. Menopause is defined as the time of the final menstrual period and occurs at a mean age of 51 years (Krystal et al., 1998). During the peri-menopausal period of diminished ovarian production of estrogen and progesterone occurring prior to menopause, women frequently begin to experience a number of associated symptoms. These symptoms include insomnia, night sweats, hot flashes, vaginal dryness, urinary frequency, palpitations, headaches, vertigo, anxiety, dysphoria, irritability, forgetfulness and an inability to concentrate. These symptoms may persist after menopause (Chakravarti et al., 1977; Gambrell, 1986; Sharma and Saxena, 1981). The centrality of insomnia is evident from data suggesting that it is experienced by 30% to 63% of peri- and post-menopausal women. In some studies it is the most common symptom experienced (Krystal et al., 1998; von Muhlen et al., 1995).

Despite the frequency of insomnia in peri- and post-menopausal women, there is a surprising lack of research studies in this area (Krystal et al., 1998). In this brief article, the available literature on the pathophysiology, diagnosis and treatment of insomnia in peri- and post-menopausal women will be reviewed.

It is first necessary to determine whether there is evidence for an insomnia syndrome that is specifically linked to the changes of the peri-/post-menopausal period. Attribution of insomnia to peri-/post-menopausal changes is clouded by the fact that the incidence of insomnia complaints among all women over 30 is 26% to 45% (Krystal et al., 1998). One important contributing factor is that a number of sleep disorders that can cause insomnia (restless legs syndrome, periodic movements of sleep and sleep apnea) increase in frequency with advancing age (Ancoli-Israel et al., 1981; Fry, 1987).

Despite these considerations, there does appear to be evidence that in many women, insomnia is specifically linked to peri-/post-menopausal changes. A number of studies indicate an association between nighttime vasomotor changes (often manifested in night sweats) caused by diminished ovarian hormone production and sleep disruption (Campbell and Whitehead, 1977; Coope, 1996; Woodward and Freedman, 1994).

Hormone replacement therapy has also been found to lead to improvement in both polysomnographic evidence of insomnia and vasomotor symptoms (Campbell, 1976; Regestein et al., 1981). In contrast, subjective complaints of insomnia not associated with polysomnographic evidence of increased frequency of arousals tend not to respond to hormonal therapy, as is also the case for non-vasomotor symptoms. Therefore, these may not be due to diminished hormone production (Campbell, 1976; Fry, 1987; Regestein et al., 1981). This non-hormonally related insomnia is of unclear origin, and there are a number of possible etiologies, including: 1) a period of insomnia due to vasomotor events that led to behavioral conditioning causing a persistent psychophysiologic insomnia independent of the vasomotor events (Krystal et al., 1998); 2) unresolved grief associated with menopausal changes (Thomson and Oswald, 1977; U.S. Department of Health and Human Services [HHS], 1993); or 3) insomnia not related to menopausal status (Krystal et al., 1998).

Proper management of a peri-/post-menopausal woman with insomnia is highly dependent upon determining the correct etiology for the insomnia, since the optimal treatment differs greatly with different etiologies (Krystal et al., 1998). In order to sort out this differential diagnosis, it is crucial to take a careful history.

While this may appear straightforward, insomnia may sometimes be the first or only presenting symptom. It should also be noted that individuals suffering from insomnia often do not mention this to their health care practitioners. Because of the very high incidence of insomnia in peri-/post-menopausal women, it seems justified to ask patients about their sleep or even consider implementing routine screening for insomnia using an instrument such as the Insomnia Symptom Questionnaire (Krystal et al., 1998).

A careful history will also rule out an underlying sleep disorder. When restless legs syndrome is the cause of the insomnia, the patient will have a history of trouble falling asleep due to an uncomfortable, crawling feeling in the legs associated with the need to walk (Fry, 1987). Similarly, periodic movements of sleep are typically associated with a history of the bed partner noticing twitching movements throughout the night. The reports of the bed partner are also helpful in identifying sleep apnea. They generally will provide a history of loud snoring and episodes of stopping breathing at night. A history of morning headaches and dry mouth and daytime sleepiness is also commonly present.

Despite efforts to identify these disorders by taking a careful history, some individuals -- roughly 10% in the elderly -- may have clinically significant periodic movements of sleep or sleep apnea when there is nothing in the history to suggest their presence (McCall et al., 1992). Because such cases can only be picked up with polysomnography, it is prudent to refer patients for polysomnography when all the usual treatments fail.

The history should also help to identify if the onset of insomnia symptoms coincided with the onset of vasomotor symptoms. In this case, a referral for hormone replacement therapy is appropriate (Krystal et al., 1998).

If hormone replacement does not work, behavioral conditioning should be considered. Behavioral sleep therapy will break the association of anxiety and alertness with the attempts to sleep, alter the expectations of poor sleep and correct aberrant sleep hygiene practices. In circumstances where the insomnia is associated with peri-menopausal symptoms but vasomotor symptoms are absent, or the onset of the insomnia occurred at a different time than clear vasomotor symptoms, other etiologies should be explored (Krystal et al., 1998).

In addition, it is very important to explore the patient's reactions to their hormonal changes to determine if unresolved grief may be a contributing factor (HHS, 1993). A referral for grief-oriented psychotherapy should be considered in these cases (Krystal et al., 1998).

In some cases, sedative/hypnotic medications may be very helpful. These should generally be reserved for cases when all other attempts to treat the underlying cause of the insomnia have failed or because hormone replacement therapy is not possible (e.g., there is a high risk of cancer) (Krystal et al., 1998).

Because the course of medication is likely to be relatively long-term, sedative/hypnotic medications -- such as the benzodiazepines and chloral hydrate -- that lead to significant tolerance and withdrawal should be avoided. Zaleplon (Sonata) and zolpidem (Ambien) are better in this regard than the older benzodiazepines and could be considered. The sedating antidepressants have excellent tolerance and withdrawal profiles, but have significantly more side effects and a greater likelihood of next-day sedation and, therefore, may not be tolerated by many patients. There is one report of the use of trazodone (Desyrel) in this setting (Pansini et al., 1995), but there has been no systematic study of treatment with a sedative/hypnotic medication in this population. On the basis of the above considerations, I use the protocol listed in the Table in order to assess and treat women with insomnia who may be peri-/post-menopausal.

There are substantial gaps in the available information as to how to best manage insomnia in women of peri-/post-menopausal age. For example, a way to determine definitively if peri-menopausal hormonal changes are present is glaringly absent. Leutenizing hormone levels appear to increase during vasomotor events, however, this occurs so sporadically that it has not proven useful as a screening test (Krystal et al., 1998).

Additional work is also needed to determine if, and in what conditions, polysomnography, including skin conductivity and temperature recording, may be useful to determine if nighttime sleep disruptions are due to vasomotor events and whether hormone replacement therapy actually eliminates nighttime vasomotor events (Krystal et al., 1998). Additional studies would also be helpful to determine if behavioral sleep therapy following unsuccessful hormone replacement therapy results in resolution of symptoms. Finally, further research on sedative/hypnotic medication is needed.

In summary, some progress has been made about how to manage insomnia in peri-/post-menopausal women, however, a review of the literature illustrates that a great deal more research is needed in order to optimize diagnosis and treatment.

Dr. Krystal is professor of psychiatry and behavioral sciences at Duke University Medical Center. He is also director of the Duke Clinic Sleep Disorders Research Laboratory.

References

Ancoli-Israel S, Kripke DF, Mason W, Messin S (1981), Sleep apnea and nocturnal myoclonus in a senior population. Sleep 4(4):349-358.

Campbell S (1976), Double blind psychometric studies on the effects of natural estrogens on post-menopausal women. In: The Management of the Menopause and Post-Menopausal Years, Campbell S, ed. Baltimore: University Park Press, pp149-158.

Campbell S, Whitehead M (1977), Oestrogen therapy and the menopausal syndrome. Clin Obstet Gynaecol 4(1):31-47.

Chakravarti S, Collins WP, Newton JR et al. (1977), Endocrine changes and symptomatology after oöphorectomy in premenopausual women. Br J Obstet Gynaecol 84(10):769-775.

Coope J (1996), Hormonal and non-hormonal interventions for menopausal symptoms. Maturitas 23(2):159-168.

Fry JM (1987), Sleep disorders. Med Clin North Am 71(1):95-110.

Gambrell RD Jr (1986), The menopause. Invest Radiol 21(4):369-378.

Krystal AD, Edinger J, Wohlgemuth W, Marsh GR (1998), Sleep in peri-menopausal and post-menopausal women. Sleep Medicine Reviews 2(4):243-253.

McCall WV, Erwin CW, Edinger JD et al. (1992), Ambulatory polysomnography: technical aspects and normative values. J Clin Neurophysiol 9(1):68-77.

Pansini F, Albertazzi P, Bonaccorsi G et al. (1995), Trazodone: a non-hormonal alternative for neurovegetative climacteric symptoms. Clin Exp Obstet Gynecol 22(4):341-344.

Regestein QR, Schiff I, Tulchinsky D, Ryan KJ (1981), Relationship among estrogen-induced psychophysiological changes in hypogonadal women. Psychosom Med 43(2):147-155.

Sharma VK, Saxena MS (1981), Climacteric symptoms: a study in the Indian context. Maturitas 3(1):11-20.

Thomson J, Oswald I (1977), Effect of oestrogen on the sleep, mood, and anxiety of menopausal women. Br Med J 2(6098):1317-1319.

U.S. Department of Health and Human Services (1993), Depression Guideline Panel. Depression in Primary Care. Vol. 1. Detection and Diagnosis Clinical Practice Guideline. Rockville, Md.: Health and Human Services. AHCPR Pub. No. 93-0550.

von Muhlen DG, Kritz-Silverstein D, Barrett-Connor E (1995), A community-based study of menopause symptoms and estrogen replacement in older women. Maturitas 22(2):71-78.

Woodward S, Freedman RR (1994), The thermoregulatory effects of menopausal hot flashes on sleep. Sleep 17(6):497-501.