Should women receive androgen replacement therapy, and if so, how?


  • Susan R. Davis

    1. Professor of Women’s Health and NHMRC Principal Research Fellow, Women’s Health Program, Department of Medicine, Central & Eastern Clinical School, Monash Medical School, Alfred Hospital, Prahran, Australia
    Search for more papers by this author

Susan R. Davis, Professor of Women’s Health and NHMRC Principal Research Fellow, Women’s Health Program, Department of Medicine, Central Clinical School, Monash Medical School, Alfred Hospital, Commercial Road, Prahran VIC 3181, Australia. Tel.: +61 3 9903 0827; Fax: +61 3 9903 0828;


The available clinical evidence supports efficacy of testosterone therapy for the treatment of postmenopausal women with hypoactive sexual desire disorder (HSDD) who have undergone a comprehensive clinical evaluation. Although few preparations designed to deliver an appropriate dose of testosterone for women are available, use of testosterone by women for the management of HSDD is widespread. Issues that continue to simulate debate in this therapeutic area include whether HSDD is a condition that merits pharmacotherapy, how effective is such treatment and whether testosterone therapy is safe. Hence the question, should women receive androgen replacement therapy, and if so, how?


Circulating levels of testosterone and the major pre-androgens, dehydroepiandrosterone (DHEA), DHEA sulfate (DHEAS) and androstenedione decline with age in women, with the maximal rate of decline occurring in the premenopausal years (Fig. 1).1 However, there is no diagnostic lower limit for any of these circulating steroids which can be used to classify a woman as androgen deficient.2 Thus the use of testosterone therapy for women is not based on an established link between symptoms and biochemistry, but rather clinical evidence that exogenous testosterone improves specific parameters of sexual function in women.

Figure 1.

 Relationship between age and individual androgens for the reference group.

The most commonly reported sexual problems in women relate to sexual desire and interest, pleasure and global satisfaction.3–5 The Diagnostic and Statistical Manual for Mental Disorders, Fourth Edition provides major classifications for female sexual dysfunction (FSD): hypoactive sexual desire disorder (HSDD), sexual arousal disorder, orgasmic disorder, dyspareunia and other (such as aversion).6 HSDD is diagnosed when a woman presents with loss of sexual desire in association with personal distress.6 The prevalence of HSDD amongst postmenopausal women is in the order of 9 to 14%, with no differences between natural in surgically menopausal women.7,8 Most studies evaluating the efficacy of testosterone for the treatment of FSD have required women to fulfil the diagnostic criteria of HSDD. The controversies regarding the use of testosterone to treat HSDD appear to be firstly whether this is a condition that merits pharmacological intervention and secondly whether testosterone as a pharmacological intervention is sufficiently effective and safe. The views expressed in this manuscript differ from the more conservative recommendations by the Endocrine Society Clinical practice guidelines published in 2006 which stated that ‘Although there is evidence for short-term efficacy of testosterone in selected populations, such as surgically menopausal women, we recommend against the generalized use of testosterone by women because the indications are inadequate and evidence of safety in long-term studies is lacking’9 Since that time several large studies of the efficacy and safety of testosterone and DHEA use in women which have been of longer duration have been published providing more outcome data.10–13

Is there evidence of a clinical need?

Sexual satisfaction is associated with better general health.14 More than 80% of women over the age of 30 years believe that an active sex life is important for one’s sense of well-being, with higher levels of physical pleasure in sex significantly associated with higher levels of emotional satisfaction.15 At the other end of the spectrum, HSDD is associated with significant decreases in health-related quality of life in both naturally and surgically menopausal women.16,17 The largest significant, negative effects in women with HSDD were observed in the mental health, vitality, social function and bodily pain subscales. Affected women were more likely to be depressed, be dissatisfied with their home life and with the emotional and physical relationships with their sexual partner. The overall effect of HSDD on quality of life in women with HSDD is similar in magnitude to that seen in adults with other common chronic conditions such as diabetes and back pain.16 HSDD also adversely affects relationship health. Both men and women reporting a discrepancy between their own and their partner’s sexual desire have lower relationship satisfaction18 and individuals in sexually inactive marriages report less marital happiness.19 Sensitive to the importance of sexual intimacy in the relationship, women commonly continue to engage in sexual activity despite experiencing dyspareunia and/or little sexual desire or pleasure.20–22

In summary, loss of sexual desire for many women is a significant cause of loss of personal well-being, personal distress and relationship disharmony, often with an unavoidable negative ripple effect from the impact of relationship dynamics, impacting other household members. Just as individuals seek treatment for other factors that impact negatively on their quality of life such as depression, many women experiencing HSDD also seek and merit treatment.

Does androgen therapy improve sexual function of women with HSDD?

Large studies have now shown that the administration of testosterone to naturally or surgically postmenopausal women improves sexual desire, arousal, orgasm frequency, pleasure and sexual satisfaction, reduces personal distress associated with HSDD and increases the number of occasions on which a woman experiences a satisfactory sexual event.23

There has been a consistency across published studies of the testosterone patch that at baseline on average women enrolled in the various studies have reported 50% of sexual events were not satisfying events.11,24–26 With testosterone treatment, women report on average 80% of sexual events to be satisfactory, which translates into 1–2 extra satisfactory sexual events per month above that seen with placebo, and for women treated with testosterone without concurrent oestrogen, more than a 115% increase in reported orgasms compared with placebo (38% increase).11

In postmenopausal oestrogenized women, the effects appear to be a consequence of direct androgen action as opposed to the aromatization of testosterone to oestrogen.27 Two small studies have also shown that exogenous testosterone improves sexual function in older premenopausal women presenting with loss of libido.10,28 Interestingly, when administered transdermally at doses that restore levels of free testosterone to those seen in healthy young women, consistently the effects of testosterone on sexual function does not become manifest until about 6–8 weeks of treatment.

Appropriate candidates for testosterone therapy

Overall, women who present with HSDD from the late reproductive years and beyond are potential candidates for androgen therapy. Physicians should specifically consider women who have experienced premature ovarian failure or surgical menopause, or who have adrenal insufficiency or hypopituitarism, to have a high likelihood of having HSDD as these are conditions characterized by low androgen production. The use of androgen therapy in women with hypopituitism and adrenal insufficiency has recently been reviewed elsewhere.29,30 Importantly, the desire to experience sexual well-being is not limited to women in relationships.21 Although most of the published studies of testosterone have required participants to be in established monogamous relationships, this should not be a requirement for consideration of testosterone therapy in clinical practice.

An array of physical, psychological, cultural and relationship factors influence sexual well-being and sexual function. Hence women presenting with the complaint of diminished sexual interest with or without impaired sexual responsiveness need a general psychological, physical and social health assessment. Commonly contributing factors include stress, fatigue, relationship issues, depression and medication side-effects. Although identification of such factors does not exclude a woman from consideration for treatment with testosterone, such factors need to be concurrently addressed. For example, relationship issues are not uncommonly a consequence of a woman’s diminished desire to engage in sexual activity rather than the primary cause. Thus HSDD may be primary, secondary or a concurrent issue.

Loss of sexual desire is not uncommon amongst women using the combined oral contraceptive pill (COCP), particularly those containing an anti-androgenic progestin. The COCP suppresses ovarian testosterone production and increases sex hormone binding globulin (SHBG) thus reducing free testosterone. Switching from a COCP to another form of contraception will often improve sexual well-being in younger women.

Antidepressant use may be associated with FSD reflecting either inadequate treatment of depression or a drug side-effect.31 Women who present primarily with arousal disorder and inability to achieve orgasm, but no significant loss of libido, in association with antidepressant therapy may respond well to phosphodiesterase type 5 inhibitor therapy.32

As there is no level of testosterone, total or free, below which a woman can be diagnosed as being androgen deficient, the measurement of testosterone should not be used as an indicator as to which women merit treatment. However, a testosterone level should be measured prior to therapy primarily to exclude women who may be at risk of side-effects if treated (see below). Total testosterone is notoriously difficult to measure at lower levels, as seen in women with sensitivity and precision.33 Total testosterone and SHBG should be measured and from these, free testosterone can be reliably estimated using the Sodergard equation.34 The measurement of free testosterone in women by direct ‘kit’ assays is generally completely unreliable.35 In postmenopausal women, transdermal testosterone is likely to be ineffective if the SHBG level is high (>160 pmol/L)24 or if used by women taking conjugated equine oestrogens.36

Contraindications to testosterone therapy are shown in Table 1. Women with a low SHBG level are likely to have a more rapid clearance of administered testosterone, and thus are much more likely to experience androgenic side-effects. Similarly women with a free testosterone level in the high normal or above normal range for healthy young premenopausal women1 are more likely to achieve supraphysiological levels with treatment, and hence experience side-effects.

Table 1.   Clinical contraindications to testosterone therapy
Pregnancy and lactation
Current use of anti-androgen therapy
Troublesome acne or hirsutism
Hormone dependent malignancy current or past
Sex hormone binding globulin level below the lower limit of normal
Free testosterone level in the mid-normal range for young women and above

Safety of exogenous testosterone

The links between postmenopausal oestrogen-progestin use and both breast cancer and cardiovascular disease have created a level of concern regarding any form of hormone use in postmenopausal women.37 Testosterone has been widely used by women as an unapproved therapy for decades with an estimated 2 million prescriptions written for women in the US in 2006 and 2007.38 Excessive therapy will clearly result in undesirable androgenic effects such as hirsutism and acne, although this is not common when treatment is aimed at achieving testosterone levels in the female range.24–26 About 20% of women report increased hair growth, as opposed to hirsutism, and this rarely results in the decision to discontinue therapy.10,11 There is no evidence from studies of premenopausal women and postmenopausal women that systemic transdermal testosterone is associated with a change in the risk of invasive breast cancer39–41 or increased cardiovascular morbidity or mortality.11,42 Oral methyltestosterone is associated with a reduction in HDL-cholesterol, an effect not seen with transdermal therapy.23 However, long-term safety data for the use of testosterone in women are limited.40,41 Only two small studies have provided safety data for premenopausal women.10,28 There is uncertainty as to the consequences of restoring testosterone levels to those of premenopausal women in women who are many years past menopause.

Treatment options

Presently there is a lack of approved preparations of testosterone specifically suitable for use in women. Use of preparations designed to deliver a dose of testosterone to men cannot be condoned.

In several countries, testosterone therapy is commonly initiated with testosterone pellets implanted under local anaesthetic subcutaneously. Most commonly a dose of 50 mg is used.43 These implants remain effective for periods of 4 to 6 months. Repeat implantation should not be undertaken without confirmation that total testosterone corrected for SHBG, or free testosterone has fallen back into the lower quartile of the normal female range.

Testosterone transdermal patches have been shown to be effective and have a good short-term safety profile when used by naturally or surgically postmenopausal women with and without concurrent oestrogen therapy.11,42 The Intrinsa® patch which delivers 300 μg of testosterone daily has been approved for use by surgically menopausal women using concurrent systemic oestrogen therapy in European Union countries. A transdermal testosterone cream for women, marketed as Androfeme1%®, is available in Australia28 and testosterone gels for women and a transdermal spray are in development.

Various pharmacists prepare testosterone for buccal administration in the form of troches, or as creams, but there are no published pharmacokinetic or safety data or efficacy studies to validate this method of administration.

Some clinicians undertake a clinical trial of intramuscular injections of testosterone esters 50 to 100 mg. This may or may not result in a clinical response over 1–2 weeks or more. A positive response supports the initiation of longer term therapy. However, as peak levels are supraphysiological, testosterone esters should not be considered a long-term treatment option.

Another agent which can be conceived as having androgenic effects (in addition to having properties as an oestrogen and a progestin) is tibolone. In a dose of 2·5 mg daily, it improves sexual function in postmenopausal women.44 DHEA therapy has not been covered in this article. There are little data to support the use of DHEA for the treatment of sexual dysfunction and safety data are limited.12,13,30

Evaluating efficacy

The time between commencement of therapy and improvement in symptoms varies according to the mode of administration of testosterone which most probably reflects the serum levels achieved with different delivery modes. Intramuscular injections of testosterone esters result in an immediate high serum level and women commonly report a surge in libido within 1 to 2 days. The decline in libido with such therapy is also rapid and women treated in this manner often report experiencing an unpleasant roller-coaster effect. Testosterone implants provide an initial supraphysiological rise in serum testosterone which lasts a few days after which levels fall to the upper level of the normal female range with a 50 mg testosterone implant. Women consistently report experiencing an effect about 2 weeks after insertion of the implant. By contrast, when testosterone is administered transdermally as a cream, patch or skin spray at a dose that brings free testosterone into the range of that of young women, an effect is consistently experienced after 6–8 weeks.10,11,24,28 It is critical women are made aware of this when such therapies are prescribed. Improvement is unlikely to occur beyond 16 weeks of therapy and if no improvement has been experienced by 26 weeks, the woman should be considered a nonresponder. In general, about 60% of women treated with active therapy in clinical trials are responders to testosterone in the setting of strict participant inclusion criteria.10


Taken together, the available clinical evidence supports the efficacy of testosterone therapy for the treatment of postmenopausal women with HSDD who have undergone a comprehensive clinical evaluation. Data for use in women in their late reproductive years remain limited. Available data do not demonstrate serious safety concerns; however, further studies are required to determine the long-term safety of testosterone in women.