A role of estradiol withdrawal in the onset of mood disorders in some perimenopausal women has been suggested indirectly by the clustering of depressive episodes during the late menopause transition (the stage of ovarian aging associated with maximal decline of ovarian estradiol secretion) and by estradiol's antidepressant efficacy in perimenopausal depression. Nonetheless, there is no direct evidence that estradiol withdrawal precipitates depressive episodes in those women who develop depression during the perimenopause. In these studies, first we examine the effects of estradiol withdrawal and the recent onset of hypogonadism on mood symptoms in asymptomatic premenopausal women. Second, we investigate what factors influence the development of depression during estradiol withdrawal. We administered a gonadotropin releasing hormone (GnRH) agonist for two to three months to 72 regular cycling, asymptomatic, healthy, premenopausal women. Only three women (<5.0% of the sample) reported clinically significant symptoms of depression and in only one woman were depressive symptoms sustained for two weeks. In contrast to the relative absence of depressive symptoms in these women, we did observe the significant appearance of hot flushes, disturbed sleep, and diminished libido. Virtually an identical pattern of symptom expression was observed in our previous study of asymptomatic healthy men who underwent a similar hormone manipulation. Thus, in otherwise healthy women (and men), the induction of neither hypogonadism nor hot flushes (with an accompanying sleep disturbance) uniformly precipitates depressive symptoms. In related ongoing studies, we also investigate two factors that could impact the appearance of depression during estradiol withdrawal - age and a previous episode of depression. We evaluate the effects of the acute withdrawal of estradiol therapy in older postmenopausal women with and those without a past perimenopausal depression. Estradiol withdrawal induces depressive symptoms in women with a past perimenopausal depression, but not in those without such a history. In women with a past depression during the perimenopause, estradiol withdrawal is associated with a significant increase in depressive symptoms compared with those who were maintained on estradiol therapy. Additionally, no significant depressive symptoms emerged in the women lacking a history of depression. Recruitment of only a few more controls are needed to complete this study. Our data are consistent with those from epidemiologic studies showing that, for a subgroup of women, the endocrine events of the late menopause transition represents and important trigger for mood destabilization and the onset of depression. Both the markers of this risk and the mechanisms underlying estradiol withdrawal-induced depressive symptoms remain to be identified. The results of the Womens Health Initiative (WHI) have deterred many women from using estrogen therapy, and many women and their physicians are concerned about the long-term risks of estrogen therapy. In a previous study, we demonstrated the antidepressant effects of the short-term administration of estradiol in women with perimenopausal depression. We now are examining the effects on mood and behavior of two compounds that for many women represent alternatives to traditional estrogen therapy. Specifically, selective estrogen receptor modulators (SERMs) and phytoestrogens (plant-derived estrogen-like compounds) have become available and are reported to display both tissue-specific profiles of estrogen agonist and antagonist actions and differential affinities from estradiol for the two forms of estrogen receptor. Preliminary results demonstrate that after eight weeks of either estradiol or raloxifene (but not a plant-based phytoestrogen compound), mood rating scores were significantly decreased compared with baseline scores and significantly lower than scores in women receiving placebo. Recruitment for this treatment trial is approximately 80% complete and we hope to finish this protocol this year. Equal efficacy to that of estradiol would provide an alternative therapeutic option with a more acceptable profile of long-term side effects. As phytoestrogens are already commonly recommended, and widely used for the treatment of mood symptoms, further clarification of their mood effects has significant public health implications. In future studies, we intend to investigate the behavioral relevance of estrogen receptor beta as well as a potential caged form of estradiol with brain-specific estradiol actions in humans. Specifically, we will examine the antidepressant/anxiolytic efficacy of selective estrogen receptor beta agonists (when available for human use) in women with perimenopausal depression. An additional strategy that we have employed is the use of the longitudinal design to identify predictors of the onset of depression during the menopause transition. The study now includes over 100 women, forty of whom have completed between five and 10 years of longitudinal follow-up during the menopause transition. This will be the first high-density, longitudinal study examining both menstrual cycle phase characteristics and reproductive hormones in women as they transition through the menopause. In a previous study, we demonstrated the therapeutic efficacy of the adrenal androgen DHEA in men and women with midlife-onset depression;however, clinical response was not accompanied by a specific change in adrenal or gonadal steroid hormone levels. Recently, we observed that the neurosteroid androsterone increases after DHEA treatment (with women achieving higher levels than men) and is associated with clinical response. The response pattern of androsterone differed in men and women and showed an inverted U-shaped curve, suggesting that an optimal level of androsterone is associated with antidepressant response. Other neurosteroids measured (in collaboration with Dr. Christine Marx at Duke University) showed no association with DHEAs antidepressant response. These data, (along with our prior demonstration of the physiologic relevance of androsterone secretion in male sexual function), emphasize the potential importance of this neurosteroid in human behavior. Karyotypically, normal, 46XX spontaneous (i.e., idiopathic) primary ovarian insufficiency (POI) is a condition in which ovarian failure occurs at an early age. We identified that a lifetime prevalence of depression in adult women with POI is significantly greater than that reported in both the general population, and in gynecologic or general medical outpatient settings. Additionally, first episodes of major depression preceded the date of POI diagnosis in over 70% of women. However, episodes of depression occurred after evidence of ovarian failure began in the majority of women (i.e., after the onset of menstrual cycle irregularity). Thus, we have demonstrated that depressions occur after the onset of MCI potentially secondary to the events related to this premature menopause transition (at a time similar to the onsets of depression in the normally-timed menopause transition). We now have initiated a collaboration with Drs. Maya Lodish and Laurence Nelson within NICHHD to prospectively evaluate mood changes and the prevalence of depressive illness in adolescent girls who present with POI. A better understanding of the nature of this association could identify candidate genes and physiological processes that may inform both our understanding of the relationship between depression and POI, as well as the association between depression and the menopause transition in the normally-timed menopause.
Showing the most recent 10 out of 23 publications
