Why are estrogens rarely used for the treatment of depression in women?

The short answer to this question could be that they do not work, but that is not true. It is much more likely that a turf war is developing between psychiatrists and gynaecologists/endocrinologists for this common disorder. This is not surprising as we are all products of our training, with hormones and vaginal bleeding an unchartered country for some. As depressed women would normally gravitate to their family doctor and to psychiatrists, it is usual that psychiatric intervention including anti-depressants would be the first line of therapy. However, there are depressed women with particular features in their history which make them very suitable for estrogen therapy as an excellent clinical response can be expected. These are currently being neglected.

Depression is more common in women than men, whether these data come from hospital admissions, community studies, suicide attempts or a prescription of anti-depressants [1]. There will always be debate whether this is genetic, environmental or hormonal with psychiatrists, and no doubt feminists, supporting the view that it is the lifestyle, domestic problems and career limitations which is an essential factor in this excess. Nobody would doubt the social aspects of depression, but the role of hormones is inadequately recognised in spite of good evidence that this is an aetiological factor in many women.

The increase of depression in women occurs at times of hormonal fluctuation. It begins at puberty and adolescence, and the difference is no longer present five years after the menopause. Depression occurs in the pre-menstrual phase, sometimes lasting for as long as 14 days each month, is rare during pregnancy when there are stable and high levels of estradiol, only to occur post-partum. Such post-partum depression occurs in up to 15% of women, may last for years, and becomes more cyclical when the periods recur.

Climacteric depression is at its worst in the 2–3 years before the periods cease. Indeed, if the 45-year old depressed woman is asked when she was last well, the frequent reply is that it was during the last pregnancy, several years previously. Then post-natal depression occurred, followed by cyclical pre-menstrual depression which then became more constant as the menopause approached. Such patients respond well to moderately high doses of transdermal estrogens. Unfortunately, the association of depression with these reproductive events is rarely sought in a history taking by psychiatrists. Similarly, the enquiry about how many good days a month a woman experiences, reveals not only the severity of her distress with the premenstrual syndrome, menstrual headaches and heavy, painful periods, but also reinforces the cyclical and hormonal nature of the problem. This is another very appropriate question for psychiatrists (and gynaecologists!) to ask.

Thus, this triad of hormone-related depression – premenstrual depression, post-natal depression and peri-menopausal depression often occur in the same vulnerable woman. Perhaps the correct name should be “reproductive depression”.

The earliest double-blind study by Klaiber [2] using huge daily doses of 5–25 mg of Premarin in a mixed group of patients with severe recurrent depression showed an impressive significant response to the active therapy. However, this somewhat bizarre but valuable high-dose estrogen study has not been replicated. Montgomery et al. [3], using 100 mg estradiol implants, which would produce plasma estradiol levels of approximately 600 p.mols/l, had a significant beneficial effect on depression on peri-menopausal women but not on those women after the menopause. This improvement was not transient, lasting for the 23 months' duration of the study.

Schmidt and colleagues [4] have repeatedly shown the association of premenstrual depression and climacteric depression, and recorded the improvement which occurs with treatment by transdermal estrogens [5]. This is not the “domino-effect” of relief of vasomotor symptoms as the patients studied were without hot flushes and night sweats. Also Soares [6], in several studies has reported the beneficial effects of 100 mcg transdermal estrogens on peri-menopausal depression regardless of the precise DSM IV diagnosis.

Transdermal estrogens in a dose of 200 mcg twice weekly have been shown to be more effective than placebo in post-natal depression [7]. It is incomprehensible that this study of 1996 has not been repeated. Either it was perfect and didn't need confirmation, or psychiatrists who deal with post-natal depression prefer their own Mother and Baby Units, psychoanalysis or anti-depressant therapy rather than consider the logical causative factor of decreasing estrogens. The hormonal aetiology of post-natal depression has been supported by an experiment by Bloch et al. [8] who created a pseudo-pregnancy by hormonal manipulation in women with and without a history of post-natal depression. 5 out of 8 of those women with a history of post-natal depression developed depression when estrogens were withdrawn, compared with no patients who did not have a history of post-natal depression.

The common condition of premenstrual syndrome (PMDD), with its psychiatric problems of depression, irritability and irrational behaviour, together with the somatic symptoms of mastalgia and bloating are clearly related to undetermined hormonal changes which follow ovulation. It follows that the essential tenant of treatment is to suppress ovulation and the cyclical hormonal changes responsible for the cyclical symptoms of PMDD. This can be done by GnRH analogues [9] creating a temporary medical menopause or by transdermal estrogens in a dose of 200 mcg twice-weekly has been shown in a cross-over trial to improve every cluster of Moos symptoms against placebo [10]. This treatment is as effective as it is logical, but not used by psychiatrists. This treatment, based upon suppression of ovulation and cyclical symptoms, is safe, simple and successful. It would be, in the view of many gynaecologists, first choice therapy if one was looking for a pharmacological treatment. Once again, this Lancet study has not been repeated. Is it perfection or fear of the findings?

Most “menopausologists” see many women with recurrent, cyclical or climacteric depression who have been on the full gamut of psychiatric interventions, when estrogen therapy would have been a more logical and more effective therapy. These women are still having electro-convulsive therapy, the majority are having SSRI medication with its unfortunate effect upon weight and libido.

It is true that the beneficial effects of estradiol on depression is not universally accepted and advisory bodies will state that estrogens are useful for depressed mood, but not depression [11]. Indeed, it will be difficult for a gynaecologist to claim to make a precise diagnosis of depression in scientific papers, and it is for this reason that all of the studies referenced above included psychiatrists, expert in their field, to assess the clinical condition and response. But still there remains doubt in the minds of many practitioners or advisory bodies as to whether estrogens are effective for depression.

It would seem appropriate to clarify this by further studies of estrogen, preferably by the transdermal route, in the common problems of premenstrual depression and peri-menopausal depression. Although such studies are desperately required, it is virtually impossible to obtain funding for such an important study. On a personal level, I have requested help from pharmaceutical companies, even offering to complete the study without any cost to the companies. Placebo patches were all that was required. These, incomprehensibly, were not available. Of course, such non-availability is nonsense but on a commercial level there is no reason why a company selling profitable anti-depressants would wish to create competition from small-profit out-of-patent estradiol patches.

In the meantime, women suffer from inappropriate therapy for their depression which is as misguided as a surgical treatment of the 19th century [12].