Psychoendocrinology of the Climacteric
Table Of Contents
Lorraine Dennerstein, AO, PhD, MBBS, DPM, FRANZCP
MEANING OF MENOPAUSE|
MOOD AND MENOPAUSE
ROLE OF HORMONE THERAPY
This woman’s presentation exemplifies several issues of concern and debate for women and their doctors.
|MEANING OF MENOPAUSE|
Menopause literally means cessation of menstruation. Scientifically, menopause refers to the moment of last menstruation. This chapter uses the definitions of reproductive aging agreed on at the 2001 STRAW Workshop.1 Natural menopause is recognized as having occurred when there have been 12 consecutive months of amenorrhea, generally in women over 40 years of age, for which there is no other cause, pathologic or physiologic. The endocrine changes underlying menopause take place over a number of years. This period of time, both before and after cessation of menses, is often termed perimenopause or the climacteric. Most women are aware of changes in menstrual flow and frequency for a variable number of years preceding the last menses. This phase is described as menopausal transition. The term early menopausal transition is used when the middle-aged woman notices changes in her menstrual cycle length. The term late menopausal transition is used when there are two or more skipped cycles (corresponding to 90 days or more of amenorrhea but 12 months or less of amenorrhea). In this chapter, the term late reproductive phase refers to the phase in which middle-aged women experience regular cycles and have not yet entered menopausal transition. Induced menopause is the cessation of menses that occurs after either surgical removal of both ovaries (with or without hysterectomy) or other iatrogenic ablation of ovarian function (chemotherapy, radiation, etc.). The term postmenopause refers to the time after the final menstrual period.
The criteria women use to self-define their menopausal status have not been fully elucidated. In a population study of 2000 Australian women between 45 and 55 years of age, a question was posed about where each woman placed herself with regard to menopause.2 The responses were the following: lack of any sign; just beginning; in the middle; near the end; or completed. There were discrepancies between women’s self-reported menopausal status and menopausal status defined based on menstrual status.2,3 Interestingly, self-reported menopausal status (i.e., a woman’s own concept of where she is in menopausal transition) was more strongly related to symptom experience scores than was menstrually determined menopausal status.2 In a later analysis based on prospectively acquired menstrual diaries in the Melbourne Women’s Midlife Health Project, self-rated menopausal status increased the likelihood of the woman reaching final menstrual period in a shorter time interval.4 Further research needs to explore the basis women use for determining their menopausal status.
|MOOD AND MENOPAUSE|
The pursuit of happiness is considered among the most important of human rights. Happiness is the simple emotion underlying well-being, just as the simple emotion underlying depression is sadness.5 Despite universal acceptance of the World Health Organization’s definition of health as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity,” few studies positively measure health and well-being. Most medical research focuses on ill health or morbidity and mortality indices. Menopause research has, for the most part, failed to assess psychological well-being during the menopausal transition. During the baseline cross-sectional phase of the Melbourne Women’s Midlife Health Project, subjects completed a 20- to 25-minute telephone interview.2 There was a 71% response rate. A validated measure of psychological well-being was used: the Affectometer.5,6 The Affectometer provides a measure of positive mood or affect, a measure of negative mood, and an overall measure of well-being, which was the difference between the positive and negative mood scales.6 Each scale contains 10 adjectives. Women were asked about their mood in the past week, and whether they had felt a particular way most of the time, often, sometimes, or hardly ever (Table 1).5
It is evident that the majority of these middle-aged women reported experiencing positive feelings most of the time. The only positive feelings not experienced by the majority of women most of the time were “relaxed” and “enthusiastic.”
Positive mood was significantly associated with the following: being married or living with a partner; having few symptoms; no history of premenstrual complaints; better self-rated health; low interpersonal stress; exercise; and fewer worries about aging or menopause. 7
There was no relation between positive mood and menopausal status, a finding confirmed by 8 years of longitudinal follow-up of the Australian women who were still menstruating at baseline.8 This longitudinal study found that positive mood was stable over time and not related to menopausal transition, age, or education. In the early phase of menopausal transition, positive mood was adversely affected by baseline interpersonal stress and negative attitudes to aging. The most important predictor of positive mood in the late menopausal transition/postmenopause was the woman’s positive mood level at baseline. Other factors negatively affecting mood in late menopausal transition/postmenopause were: increase in dysphoric symptoms; major life events; increase in daily stressors; loss of a partner; and decreased work satisfaction.8
The middle-aged Australian women were also found to overwhelmingly endorse positive responses to life satisfaction questions when these were administered in the sixth year of follow-up.9 Life satisfaction was highly correlated with mood scales and satisfaction with work and daily living, but was not related to menopausal status, hormone levels, or age. Life satisfaction levels were predicted by earlier attitudes toward menopause and aging, and were positively associated with feelings for the partner and exercising regularly and negatively associated with daily stressors, interpersonal stress, dysphoric symptoms, and current smoking habit.9
Negative or Depressed Moods
Research on menopause has primarily focused on pathology and medical treatments, and menopausal women have been portrayed in much of the Western medical literature as depressed, anxious, and with low self-esteem.10 Ballinger11 attributes these negative viewpoints to the experience of gynecologists with patients who seek their help. Clinic attendance is determined by a range of personal, social, and cultural factors. Several studies12,13 have confirmed that women who seek medical help for menopausal symptoms differ on many variables from women in the general population of the same age and menopausal status and are likely to report more distress.14
Surveys that have drawn their samples from general populations, rather than menopause clinics, contradict assertions that menopause has a pronounced negative effect on mental health.7 Population studies have attempted to examine the relation of mood symptoms to chronological age and menopausal status. The Melbourne Women’s Midlife Health Project2 found that only a small number of women reported experiencing negative feelings most of the time (see Table 1).5 In the Melbourne study,7 depressed moods at baseline were significantly related to the following: not living with a partner; presence of more symptoms; worse self-rated health; an increase in interpersonal stress; current smoking habit; lack of exercise; increased worries about aging and menopause.
There is some evidence that minor psychological changes (such as nervousness, irritability, headaches, depression, and decreased social adaptation) do occur in greater frequency during perimenopause.15 Sociocultural factors also play a role. In a comparison of North American and Japanese cross-sectional surveys, Avis and associates16 found low rates of depression in Japan and higher rates in North America. The Massachusetts Women’s Health Study17 found that experiencing a long menopausal transition phase (at least 27 months) was associated with increased risk of depressed mood, which appears to be transitory. Increased menopausal symptoms contributed significantly to the occurrence of depression at follow-up. Menstrual problems were more predictive of depression than were vasomotor symptoms.
Using data from 6 years of follow-up from the Melbourne Women’s Midlife Health project, longitudinal analysis found that depressed mood decreased over time and was not related to natural menopausal transition, follicle-stimulating hormone (FSH), estradiol, age, or education.18 The magnitude of depressed mood was predicted by baseline reporting of premenstrual complaints, negative attitudes toward aging and menopause, and parity of one. During follow-up, the magnitude of depressed mood was significantly adversely affected by prior levels of depressed mood, bothersome symptoms, poor self-rated health, negative feelings for the partner, no partner, current smoking habit, low exercise levels, daily stressors, and high stress. Menopausal transition had an indirect effect on depressed mood by amplifying the effect of stress associated with becoming unemployed, developing poor health, and increased daily stressors.18
Although some links have been found between minor psychological symptoms (negative moods) and perimenopause, multivariate analyses of population sample data have found social and family factors to be more important in the etiology of psychological morbidity than physiological changes.19,20
In the Melbourne study those respondents concerned about reduced physical attractiveness, reduced income, and the deaths of themselves or people close to them scored higher on the negative affect scale than those who were most concerned about children moving away or being too old to have children.7 Women who believed that depression or irritability are associated with menopause or women worried about losing their minds during menopause scored highest on negative affect.7 Those who believed that women with many interests hardly noticed menopause scored lowest on negative affect.7 Depression scores were associated with surgical menopause.20 Data derived from the Massachusetts Women’s Health Study,17 a 5-year longitudinal study of a cohort of 2565 women 45 to 55 years old at baseline, found that prior depression is the variable most predictive of subsequent depressed mood, as measured by the Center for Epidemiological Studies–Depression (CES-D) scale. Health status problems, such as physical symptoms and multiple causes of worry, particularly those involving adolescent children, ailing husbands, and aging parents or parents-in-law, also predicted depressed mood. Kaufert and colleagues19 reported similar results. They found that the risk of depressed mood was increased by poor general health and by current stress, particularly in relationships with partner and children. Avis and McKinlay21 also found that psychological and physical symptom reporting is highly related to negative attitudes toward menopause. The Melbourne data reported above agrees with these findings.18
Holte and Mikkelsen,22 in a Norwegian study, reported that earlier “menstrual coping style” was related to the experience of psychological symptoms in the midlife years. The Melbourne Women’s Midlife Health Project7,18 found that a history of premenstrual complaints was associated with negative moods. It is unclear whether these findings indicate that women who suffer from premenstrual mood states are in some way vulnerable to phases of hormonal change or whether experiencing premenstrual complaints is indicative of other risk factors, such as habitual ways of dealing with stressors. Certainly studies indicate a link between the experiencing of premenstrual complaints and postpartum depression.23 Again, the mechanism is unclear.
Clinical Depression (Major Depressive Disorder)
It is important that clinicians and researchers make a distinction between depressed mood and the psychiatric disorder of major depression. Depressed mood is the feeling of sadness familiar to everyone. Less common, but more serious, is major depression. Major depression is a psychiatric syndrome in which there is a persistent lowering of mood as the primary feature. Nearly half of the patients with depression report complaints suggestive of physical illness including fatigue; decreased energy; appetite disturbance; weight loss or gain; constipation; bodily aches and pains; headaches; difficulty breathing; dry mouth; and unusual sensations in the abdomen, chest, or head. In addition to feelings of sadness, depressed patients often experience anxiety, loss of interest, difficulty concentrating, feelings of worthlessness or inappropriate guilt, and thoughts of suicide. Behavior may be agitated or lethargic.
The diagnosis of major depression is made on the basis of persistent and prominent lowering of mood or diminished interest or pleasure in usual activities, combined with the presence of at least four accompanying symptoms every day for at least 2 weeks, which must represent a change from previous functioning.24 The symptoms must cause clinically significant distress or impairment in social, occupational, or other important areas of functioning. It is important to note that symptoms of depression may occur in other physical or psychiatric illnesses (e.g., schizophrenia), and the presence of such illnesses must be excluded.
No increase in the incidence of major depression associated with menopause has been demonstrated.14,25 Studies using large random samples, precise definitions of menopausal status, and standardized methods of case detection have been carried out in four countries: Sweden, England, the United States, and Canada. No excess of major depression associated with menopause was found in any of the studies.25 Winokur26 studied psychiatric admissions and found that the risk of developing depression at menopause was the same as the risk of developing depression at other stages of life.
Studies of menopause clinic populations report a high prevalence of sexual difficulties and marital problems.27 A critical question for clinicians is the possible relation of any deterioration in sexual functioning with changing endocrine status during the menopausal transition.
Relatively few of the population studies of menopausal transition in middle-aged women have inquired about sexual functioning. A cross-sectional study of 800 Swedish women 38, 46, 50, and 54 years of age who were randomly selected from the general population, found a decline in sexual interest, capacity for orgasm, and coital frequency associated with menopausal status rather than with aging.28 Decline in sexual interest was also associated with low social class and psychiatric disorder.29 Osborn and coworkers30 studied 436 women ages 35 to 59 years who were registered with two Oxford general practices. No association between sexual dysfunction and menopause appeared. A Danish postal questionnaire study of 474 women also found that menopausal status did not predict decreased or infrequent sexual desire at age 51 years.31 However, the question regarding changed sexual desire covered a long time frame (11 years) and may have been subject to recall bias. Other epidemiologic studies, such as that of Van Keep and Kellerhals,32 have also shown a significant association between social factors (e.g., social background, marital status, and educational level) and sexual behavior.
The Melbourne Women’s Midlife Health Project33 questioned women at baseline about changes in sexual interest in the past 12 months, reasons for any changes, occurrence of sexual intercourse, and unusual pain on intercourse. Logistic regression was used to identify explanatory variables for change in sexual interest. The majority of women (62%) reported no change in sexual interest; 31% reported a decrease. Decline in sexual interest was significantly and adversely associated with natural menopause (p < .01); decreased wellbeing (p < .001); decreasing employment (p < .01); and symptomatology (vasomotor, p < .05; cardiopulmonary, p < .001; and skeletal, p < .01). Eleven to 12 years of education was associated with a lowered risk of decreased sexual functioning (p < .01). Heterogeneous results were reported by users of hormone therapies. The results of the baseline phase of this randomly derived population study of Australian-born women strongly suggested that the sexual functioning of some women is adversely affected by natural menopause transition.
Eight years of follow-up data were available from participants in the longitudinal phase of the Melbourne Women’s Midlife Health Project.34 They completed a detailed sexuality questionnaire annually. Passing through the menopausal transition was associated with a significant decline in sexual functioning as measured by the total score of the Short Personal Experiences Questionnaire. There was a significant decline in sexual responsiveness, libido, frequency of sexual activities, and a significant increase in vaginal dyspareunia and partner’s sexual problems. Only sexual responsivity was shown to decline with the length of the study indicating that the other factors were affected more by menopausal transition than by aging.34
|ROLE OF HORMONE THERAPY|
Gynecologists have concentrated on determining which symptoms can truly be attributed to an estrogen-deficient state and are thus amenable to hormone replacement therapy. The most reliable way of determining response to hormones is via the randomized, double-blind clinical trial. To date, most trials that have assessed effects on mood and sexual functioning have had relatively small sample sizes. The stage of menopausal transition of the women participants has not always been clarified. There are also some difficulties in extrapolating results from studies of women who have undergone hysterectomy and bilateral oophorectomy to women who have retained their ovaries. Many of the trials failed to use validated and reliable assessments of mood and sexual functioning.
A review of six earlier double-blind studies found that all but one study reported that compared with placebo, there was a decrease in mood-related complaints, such as irritability, fatigue, insomnia, anxiety, and depression.35
Two large double-blind, placebo-controlled, crossover studies reported improvements in patients who received estrogen therapy rather than placebo for a variety of psychological symptoms. In a short-term crossover study involving 64 patients who had severe symptoms, Campbell and Whitehead36 compared the results of those who received 2 months of conjugated estrogen therapy (1.25 mg daily) with those who received placebo. Estrogen was found to be significantly more effective than placebo in alleviating symptoms (hot flashes, insomnia, vaginal dryness, irritability, poor memory, anxiety, worry about age, headaches, worry about self, urinary frequency) and produced increased optimism, good spirits, and coital satisfaction. To determine whether this improvement reflected relief of hot flashes, an analysis was made of the 20 patients who had not reported hot flashes. Vaginal dryness, poor memory, anxiety, and worry about age and self continued to be significantly improved by estrogens, indicating a smaller, but direct, positive effect of estrogen on mental status.
It was not clear whether these 20 women were endocrinologically as postmenopausal as the other women in the study or if a large number of their symptoms reflected psychological or sociologic causes that would be unlikely to respond to estrogens. In a longer term study of 62 patients with less severe symptoms, estrogen and placebo were each given for 6 months. Five physical and psychological symptoms (hot flashes, vaginal dryness, insomnia, urinary frequency, poor memory) were significantly relieved by conjugated estrogen compared with placebo. Once again, it is unclear in which ways, if any, the sample of 62 differed from the 64 patients with more severe symptoms.
In another large double-blind study37 we also found a beneficial effect of estrogen on mood. All 49 women subjects had undergone hysterectomy with bilateral oophorectomy for benign diseases. The study used a crossover design so that every woman received 3 months each of the following medications, the order of which was randomly allocated: ethinyl estradiol, 50 μg/day; levonorgestrel, 250 μg/day; Nordiol, the combination of these two compounds; and placebo. Drug taking was continuous; there were no drug-free periods between medications. Ethinyl estradiol given alone was found to have the most beneficial effect on mood as measured by Hamilton Depression Rating Scale scores and interview ratings of general well-being, depression, fatigue, anxiety, irritability, and insomnia. The combination Nordiol was next for beneficial effect, with norgestrel scoring slightly better than placebo. An analysis of covariance of Hamilton Depression Rating Scale scores found that at least some of the effects on mood were related to the effects on hot flashes, but a significant effect of the hormones on the depression scores remained, suggesting a direct beneficial psychotropic effect of estrogen.
This study also found that ethinyl estradiol had a beneficial effect on female sexual desire, enjoyment, and vaginal lubrication (all measured by ordinal scales) and on orgasmic frequency (recorded daily).38 The combination pill was less beneficial than estrogen alone, but norgestrel was found to be more inhibitory.38 In a double-blind, crossover study, Paterson39 studied 23 postmenopausal women who also had had a hysterectomy. The women received 3 months each of either graded sequential mestranol and norethisterone, or placebo. Insomnia was slightly improved, and energy and confidence were increased. No alteration in depression, anxiety, memory, tension, libido, or vaginal dryness was noted. These findings, in part, confirm those of the previous study37,38 that indicated when a progestin was added to estrogen, beneficial effects on mood and sexual functioning were reduced. These results were also confirmed in a study by Hammarback and colleagues.40 They studied 22 symptomatic postmenopausal women who were treated with either estrogen only (estradiol cream percutaneously for 3 of 4 weeks) or estrogen with the addition of a progestin (lynestrenol 5 mg/day) in the last 11 days of the cycle. It was found that when progestin was added to the estrogen medication, there was a significant cyclicity in mood and physical signs (such as breast tenderness and swelling), with a maximum symptom score noted during the final days of progestin treatment. Those receiving only estrogen did not show any deterioration of mood or physical signs during the treatment. Sexual feeling was the only unchanged parameter in this study.
To summarize, most double-blind studies have found that estrogens have beneficial effects on mood and sexual functioning. The addition of a progestin leads to less favorable results. Factors involved in determining the acceptability of progestins were reviewed.41 Personality variables, dosage, and type of progestins used, and individual patient vulnerability may be important in determining the response to treatment.
Double-blind clinical trials have found supraphysiologic doses of androgen to be associated with increased sexual interest and improved mood.42 The majority of these studies have been carried out on women who have had oophorectomies. These women have lost the important contribution of ovarian production to the total androgen pool. They have also lost ovarian production of estrogens and progesterone. The incremental effect of adding androgen to estrogen replacement can then be assessed. Sherwin and colleagues42 found that women who had bilateral oophorectomies who received androgen had a higher sense of well-being, higher energy, sexual arousal, and sexual fantasy levels than women receiving estrogen or placebo. Whereas earlier studies used doses often above physiologic ranges, later trials used doses of hormones similar to the upper end of laboratory ranges. The study reported by Shifren and colleagues43 involved 75 women who had had hysterectomies and oophorectomies, all of whom received conjugated equine estrogens of at least 0.625 mg daily and were randomly assigned either to testosterone, 150 μg, or 300 μg or placebo. Only at the higher dose of testosterone were there increased scores over placebo with regard to frequency of sexual activity, masturbation, sexual fantasies, pleasure-orgasm, and improved well-being. There have been relatively few negative trials suggesting that whatever the role of androgens physiologically on mood and sexual functioning, androgen administration in women who have had oophorectomies can have a powerful pharmacologic effect. Further studies are needed in intact women.
These findings support an interactive biopsychosocial model of the complaints experienced in the climacteric years and have important implications for management. Biologic, psychological, and sociologic factors are probably all important in these complaints. Underlying endocrinologic changes may, in themselves, be sufficient to trigger distress. In other women, these changes may lead to complaints only when other factors are present, such as social stressors. Personality factors, including habitual patterns of coping with stress, may also play a role.
The clinician is concerned not with what proportion of the climacteric population experiences symptoms, but rather with the individual who comes to the doctor seeking relief for her complaints. She may be in the late reproductive phase, in menopausal transition, or postmenopausal or have had a hysterectomy with or without oophorectomy. Despite the advent of menopause treatment clinics in many countries, most women continue to consult their own doctor.
The first major step for the clinician is to establish the diagnosis of the symptoms presented. With regard to mood and sexual complaints, the doctor needs to have complete details of the presenting symptoms, their duration, and their association with the menopausal transition. The doctor should ask about other menopausal complaints, such as hot flashes, night sweats, and vaginal dryness or dyspareunia. When taking the history, symptoms suggesting other major disorders should be sought. These include psychiatric disorders, especially major depression and generalized anxiety disorder; marital problems or sexual dysfunction; and midlife crisis. A personal developmental history will give some idea of the patient’s coping style, personality, and vulnerability. Details of the woman’s current environment, stressors, and support within this framework should be examined. Her attitudes toward and expectations of menopause are also important. Information about factors that may affect her response to hormones, such as her experience with the oral contraceptive pill, postpartum depression, or premenstrual complaints should be sought. Organic aspects that may limit the use of hormones must be determined.
Many authors have drawn attention to possible problems in the doctor-patient relationship that may produce barriers to free communication. Doctors and patients may have qualitatively different ways of looking at the nature, context, and management of menopause. Differences in gender and social background may produce a social distance that may make women less willing to share personal problems with their doctors. This is especially likely if the patient suspects that her doctor accepts the stereotype of the “difficult” or overanxious menopausal woman whose problems are basically physical or “merely” psychosomatic. Care must therefore be taken by the doctor to establish an empathic relationship with the patient.
An integrated approach to management is needed, aimed at reducing contributory hormonal, psychological, and social stress, and at promoting a positive adaptation to this life phase.
EXPLANATION, REASSURANCE, AND COUNSELING.
The interview provides the opportunity for the woman to express her attitudes, feelings, and concerns. The doctor can use the interview situation to assist the woman in understanding the context and causes of her complaints; in correcting any erroneous beliefs she may have about menopause; and in encouraging positive attitudes and lifestyles. Because most studies indicate that psychological and social factors affect menopausal experiences, an assessment must be made of the contributing factors for each woman. Counseling may then be aimed at reducing the effect of such factors: marital counseling to reduce relationship difficulties and problem-solving counseling to increase ability to deal with stressors.
Cognitive therapy strategies may be used to help the patient deal with stress. These are aimed at helping the woman change her personal perception of events in her life. Such techniques may also help her to improve the quality of other social support relationships in her environment. Other counseling may be needed; for example, the woman with sexual problems frequently has a partner who is also dysfunctional. Individual and couple counseling are often needed because once dysfunction has been present for some time, negative expectations and sexual anxiety may prevent response to hormone therapy alone.
The aim of such therapy should be to provide an optimal hormonal background. It is clear that women who have had oophorectomies will be estrogen-deficient and that estrogen treatment in these women is likely to improve psychological and sexual functioning. Clinical trials suggest that these women may also benefit from the administration of testosterone. The role of estrogen for women who have retained their ovaries remains uncertain. A significant improvement in the Beck Depression scores was found in a randomized double-blind study that assessed the effects of conjugated equine estrogens on the psychological functioning of 36 asymptomatic women who had had hysterectomies.44 Hormonal prescription may be more warranted when there are other coexisting symptoms, such as hot flashes, night sweats, or vaginal dryness. A progestin must be added for those women with an intact uterus. The lowest dosage needed to reverse endometrial changes is recommended because of possible adverse mood and sexual effects. There may be fewer side effects if natural progesterone or chemically similar progestins are used.
When a psychiatric disorder such as major depression is diagnosed, the woman is best treated with appropriate specific therapy, such as antidepressants and supportive psychotherapy. Because many classes of antidepressants may have deleterious effects on sexual functioning, an antidepressant with the least adverse side effects on sexuality should be chosen.
5. Dennerstein L: In pursuit of happiness: Well-being during the menopausal transition. In Berg G, Hammar M (eds): The Modern Management of the Menopause. A Perspective for the 21st Century. pp 151, 160 Lancaster, Parthenon Publishing Group, 1994
9. Dennerstein L, Dudley E, Guthrie J, et al: Life satisfaction, symptoms and the menopausal transition. Medscape Women’s Health [serial online] 5: 2000 Available at: http://www.medscape.com/viewarticle/408924
38. Dennerstein L, Burrows GD, Wood C, et al: Hormones and sexuality: Effect of estrogen and progesterone. Obstet Gynecol 56:316, 1980
40. Hammarback Hammarbäck S, Backstrom T, Holst J, et al: Cyclical mood changes as in the premenstrual syndrome during sequential estrogen: Progestagen postmenopausal replacement therapy. Acta Obstet Gynaecol Scand 64:393, 1985
42. Sherwin BB, Gelfand MM, Brender W: Androgen enhances sexual motivation in females: A prospective cross-over study of sex steroid administration in the surgical menopause. Psychosom Med 7:339, 1985
43. Shifren JL, Braunstein GD, Simon JA, et al: Transdermal testosterone treatment in owmen with impaired sexual function after oophorectomy. N Engl J Med 343:682, 2000