Testosterone Improves Sexual Function in Postmenopausal Women, While Route of Delivery Changes the Risk Profile
A large systematic review and meta-analysis has clarified a long-contested question in women’s health: testosterone can improve sexual function in postmenopausal women when low desire is driving distress, but the benefits are not biologically uniform across formulations. Across 46 reports from 36 randomized controlled trials, involving 8,480 participants, testosterone was associated with better sexual function on multiple measures, including sexual desire, pleasure, arousal, orgasm, responsiveness, self-image, and the frequency of satisfactory sexual events. The signal was clinically consistent rather than narrow, suggesting a real treatment effect on the domains that matter most to women seeking help for diminished sexual wellbeing.
The trials were blinded, randomized, and at least 12 weeks in duration, giving the analysis a stronger evidentiary base than the smaller fragmented studies that have often shaped practice. Investigators also searched drug registration submissions to both the European Medicines Agency and the US Food and Drug Administration, a step that helps reduce publication bias and capture data that never reaches a journal page. The protocol was registered in PROSPERO under CRD42018104073.
One sentence captures the central divide: oral testosterone worsened the lipid profile, while non-oral delivery did not.
Sexual Benefit Came Through Across Several Domains of Function
The meta-analysis did more than confirm a generic improvement in libido. Women receiving testosterone experienced higher satisfactory sexual event frequency, with a mean difference of 0.85, and stronger sexual desire, with a standardized mean difference of 0.36. Pleasure improved as well, alongside arousal, orgasm, responsiveness, and self-image. Distress fell, with a standardized mean difference of -0.27, and sexual concerns declined by a mean difference of 8.99.
Those changes matter because postmenopausal sexual dysfunction is rarely a single-variable problem. Desire, arousal, orgasm, and emotional response often move together in a self-reinforcing loop, and distress can intensify avoidance, anticipatory anxiety, and relationship strain. A treatment that improves several domains simultaneously is more plausible as a bedside intervention than one that shifts a single questionnaire score without changing lived experience. In that sense, the observed pattern fits the biology of androgen action in women, where testosterone likely influences central sexual motivation as well as peripheral sexual response.
The evidence is still bounded by the population studied. The benefit was identified in postmenopausal women with low sexual desire causing distress, not as a general recommendation for all women with sexual complaints. The distinction matters for prescribing, because sexual problems in women can arise from endocrine, relational, psychiatric, medication-related, and genitourinary causes, each requiring different treatment priorities.
Oral Testosterone Raised LDL, While Transdermal and Other Non-Oral Routes Looked Cleaner
Safety findings turned heavily on route of administration. Oral testosterone increased LDL cholesterol and reduced total cholesterol, HDL cholesterol, and triglycerides. Non-oral formulations, such as transdermal patch or cream, did not show that lipid penalty. An overall increase in weight also appeared with testosterone treatment.
The route effect is biologically credible. Oral androgen exposure undergoes first-pass hepatic metabolism, which can distort lipoprotein handling in a way that does not occur to the same extent with transdermal delivery. For clinicians, that means the same active hormone can produce a meaningfully different cardiometabolic footprint depending on how it reaches the circulation. A therapy aimed at improving sexual wellbeing should not quietly create avoidable hepatic and lipid-related concerns when an alternative route can preserve the benefit signal without the same lipid disruption.
No serious adverse events were recorded across the trial set, but the absence of serious events in these studies should not be mistaken for proof of long-term cardiovascular safety. The follow-up windows were limited, and the number of women contributing data for body composition, musculoskeletal outcomes, and cognitive measures was small. Such limitations leave important questions unresolved, especially for outcomes that emerge slowly or require larger samples to detect.
Acne, Hair Growth, and the Boundaries of the Evidence Base
Androgenic skin effects were more common. Testosterone was associated with a greater likelihood of acne and hair growth, findings that align with its known peripheral effects on pilosebaceous units. These adverse effects are usually not life-threatening, but they can influence adherence, satisfaction, and whether a patient sees treatment as acceptable in daily life.
No clear effects emerged for body composition, musculoskeletal variables, or cognitive measures. That absence is not trivial. Women and clinicians often hope that testosterone might improve energy, strength, lean mass, or cognition, yet the available randomized data did not support those expectations. The number of women contributing to those analyses was small, which limits statistical confidence and leaves open the possibility that clinically relevant effects were simply underpowered rather than absent.
For practice, the most defensible interpretation is narrow but useful: testosterone has a role for postmenopausal women with low sexual desire causing distress, and non-oral administration is preferred because it avoids the adverse lipid pattern seen with oral treatment. The review does not justify broad use for vitality, muscle preservation, cognition, or general wellbeing. Those uses remain outside the evidentiary core of the analysis.
Long-Term Safety Still Needs a Larger and Longer Trial Program
The central clinical question has shifted from whether testosterone can improve sexual function to how to use it without importing avoidable harm and how long benefit persists. The current randomized evidence supports efficacy on sexual outcomes, but long-term cardiovascular safety, durability of benefit, and effects on bone, muscle, and cognition remain incompletely defined. A treatment that changes desire and distress in the short term still requires scrutiny over years, not weeks, before it can be regarded as fully characterized.
Until those data arrive, route of administration should remain part of the prescribing decision, not an afterthought. Oral testosterone produced the least favorable metabolic signal in this analysis, while transdermal and other non-oral approaches preserved the sexual benefit profile without the same lipid shift. The evidence base now supports a more disciplined approach: target the right symptom cluster, choose the safer route, and avoid extending the indication beyond what the trials actually tested.
The study was funded by the Australian National Health and Medical Research Council, and its PubMed identifier is 31353194.
- • PubMed PMID 31353194: https://pubmed.ncbi.nlm.nih.gov/31353194/