New Danish research has identified a small, but significant link between oral menopausal hormone therapy and thrombotic outcomes.
Oral menopausal hormone therapy is associated with an increased likelihood of venous thromboembolism, ischaemic stroke, and myocardial infarction, irrespective of dose and treatment duration.
However, the real-world impact of this association is very small, causing no more than nine additional events per 10,000 women per year, according to a new Danish study published in the British Medical Journal.
“Oral menopausal oestrogen therapy has been associated with increased risk of venous thromboembolism and stroke, but evidence for an increased risk of myocardial infarction remains inconsistent, [while] evidence suggests that no increased thrombosis risk exists with use of transdermal menopausal hormone therapy,” the researchers wrote.
“Considering the increasing life expectancy and the rising number of postmenopausal women, updated evidence is needed to guide safe and evidence based clinical decision making.”
Danish researchers used three national registries in their nationwide, nested case-control study: the Danish Civil Registration System (containing data on birth, sex, deaths, and migration status), the Danish National Patient Registry (which contains data on all diagnoses made in a hospital since 1977, and all diagnoses made in the emergency department and outpatient clinics since 1995), and the Danish National Prescription Registry (which is a record of all prescriptions that have been redeemed in the country since 1995). Linkage between registries was possible due to each Danish resident having their own unique identification number.
The researchers focused on women aged 50-69 years who experienced their first thrombotic event between 2003 and 2021 as cases, while women who did not experience a thrombotic event were used as controls.
Women with a history of arterial or venous thrombosis, cancer (apart from non-melanoma skin cancer), liver disease, thrombophilia, oophorectomy, infertility treatment, endometriosis, or polyendocrine metabolic ovarian syndrome (formerly known as polycystic ovary syndrome) were excluded.
This age bracket was selected to capture postmenopausal women. All women were followed from 1 January 2003 or their 50th birthday (whichever was later) until 30 June 2021, their 70th birthday, emigration, death, a thrombotic event, or the occurrence of an exclusion criterion.
The use of systemic menopausal hormone therapy in the 90 days leading up to the index date (the start of the trial or their 50th birthday) was the treatment of interest. An International Classification of Diseases, 10th Revision-confirmed diagnosis of venous thromboembolism and pulmonary embolism, ischaemic stroke, and myocardial infarction were the three thrombotic outcomes of interest.
The researchers identified 1,060,082 eligible women who were followed for a median of nine years.
Among these women, there were 9807 incident cases of VTE, 18,460 cases of ischaemic stroke, and 11,974 cases of myocardial infarction who were matched with 49,035, 92,300, and 59,870 women without thrombotic disease, respectively. The median age across cases and controls for each of the three thrombotic outcomes was 62 years.
For each outcome, a greater proportion of cases had pre-existing comorbidities (e.g., hypertension, diabetes, thyroid or kidney disease, COPD, migraine, etc.).
Systemic menopause hormone therapy use was more common among cases than controls for each thrombotic outcome: 7.3% versus 4.7% for VTE, 6.2% versus 4.7% for ischaemic stroke, and 5.8% versus 4.8% for myocardial infarction.
Across all current users of systemic menopause hormone therapy – some 12,000 women – most women used oral therapy (88%), that was combined oestrogen-progestin-based (73%) and had continuous progestin use (72%). Over 96% of hormone therapy users taking oestrogen took oestradiol specifically.
After accounting for a range of potential confounding factors, including yearly income, education status, a medical history of a variety of comorbidities, and the concomitant use of certain drugs (e.g., anticoagulants, aspirin, etc.), the current use of systemic oral menopause hormone therapy was associated with an increased likelihood of VTE (adjusted hazard ratio and 95% confidence interval 1.6, 1.5-1.8) compared to non-users. This corresponds to an absolute risk increase per year of 0.09%, a number needed to harm for one year of 1055, and nine additional events per 10,000 women per year.
When considering the specific treatment characteristics of oral menopause hormone therapies, oral oestrogen-only (1.4, 1.2-1.7), oral combined continuous (1.5, 1.3-1.7), and oral combined cyclic (1.9, 1.6-2.4) regimens were all associated with an increased likelihood of VTE compared to non-users, as were women who used therapies where the active ingredient was oral oestradiol alone (1.4, 1.2-1.7), oral oestradiol and norethisterone acetate (1.6, 1.4-1.8), and oral oestradiol and medroxyprogesterone acetate (1.9, 1.4-2.6).
There was also an association with VTE regardless of whether women used a low oral dose (≤1mg oestradiol/day; 1.3, 1.1-1.6) or a high oral dose (>1mg oestradiol/day; 1.8, 1.6-2.0). There was no association between transdermal use and VTE, regardless of regimen, the active ingredients, or the daily dose.
Oral menopause hormone therapy use was associated with an increased likelihood of ischaemic stroke (1.3, 1.2-1.4), which corresponds to an absolute risk increase of 0.06% per year, a number needed to harm for one year of 1642, and six additional events per 10,000 women per year.
This association remained for different oral regimens (oestrogen-only [1.4, 1.2-1.6], combined continuous [1.2, 1.1-1.4], and combined cyclic [1.3, 1.1-1.6]), different active ingredients (oral oestradiol only [1.5, 1.3-1.7], oral oestradiol and norethisterone acetate [1.3, 1.2-1.4], and oral oestradiol and medroxyprogesterone acetate [1.3, 1.0-1.7]), and for women taking high daily oral doses (1.5, 1.4-1.6). As with VTE, there was no association between transdermal menopause hormone therapy use and ischaemic stroke.
Oral menopause hormone therapy was also associated with an increased likelihood of myocardial infarction (1.2, 1.1-1.3), corresponding to an absolute risk increase of 0.03% per year, a number needed to harm for one year of 3846, and three additional events per 10,000 women per year.
This association remained for different oral regimens (oestrogen-only [1.4, 1.2-1.6], combined continuous [1.2, 1.1-1.4], and combined cyclic [1.3, 1.1-1.6]) as well as for transdermal combined cyclic regimens (2.1, 1.1-4.1). There was only one association found based on the therapy’s active ingredient (oestradiol and norethisterone acetate; 1.2, 1.0-1.3) and one for the dose (high-dose daily oral therapy; 1.4, 1.2-1.6).
The three thrombotic outcomes were also linked to how long the women had been using the particular menopause hormone therapy. There was an increased likelihood among women using low-dose oral hormone therapy for between one and five years (1.5, 1.1-2.0), high-dose oral therapy for less than a year (3.2, 2.3-4.3), high-dose oral therapy for between one and five years (1.4, 1.1-1.9), and high-dose oral therapy for more than five years (1.9, 1.4-2.5).
Ischaemic stroke was associated with using high-dose oral therapy for between one and five years (1.4, 1.1-1.7), high-dose oral therapy for more than five years (1.8, 1.4-2.2), and low-dose transdermal (≤50µg oestradiol/day) therapy for between one and five years (0.4, 0.2-0.9), while myocardial infarction was associated with using high-dose oral therapy for more than five years (1.8, 1.4-2.4).
“These findings do not support describing menopausal hormone therapy as simply safe or unsafe; the results instead suggest that there is no single risk profile for hormone therapy. Risk may depend on the therapy prescribed, the method of use, the dosage, and duration of treatment,” said Dr Amani Meaidi of the Department of Gynaecology and Obstetrics at Copenhagen University Hospital of North Zealand in Denmark, in an accompanying opinion piece. Dr Meaidi was the senior author on the new research.
“We already know that hormone therapy can be highly effective for many women, but we are yet to elucidate the variation in benefits and harms between individuals. Who is most likely to experience substantial benefit? For whom might the balance of benefits and harms be less favourable? And how do age, comorbidities, and underlying risk modify these effects?
“These are the questions menopause researchers need to answer next, and doing so requires research designed to capture heterogeneity rather than exploring average treatment effects alone.”
The findings of the new Danish study were consistent to those from previous research, according to Associate Professor Michelle Wise, a specialist gynaecologist from the University of Auckland.
“Most women on MHT today are taking transdermal estrogen – it is reassuring that with this more ‘modern’ method of taking MHT, there does not seem to be an associated risk of blood clot, stroke or heart attack,” she said.
“[However,] we still don’t have enough research on MHT in younger peri-menopausal women or in women taking body-identical progesterone.”
Related
Professor Martha Hickey, a professor of Obstetrics and Gynaecology at the University of Melbourne, said the findings had some relevance for Australia.
“[A] 1mg [dose] of oestrogen is considered ‘low dose’ by the Australasian Menopause Society, where many clinicians seek advice about MHT doses,” the head of Menopause Services at The Women’s Hospital Victoria said in a statement.
“Since there is no good evidence that doses over 1mg are needed to manage symptoms, clinicians should consider whether doses of >1mg of oral oestrogen are required.
“Since transdermal preparations are equally effective and may be safer, clinicians should consider whether to prescribe oral MHT. Concerningly, recent UK data suggest that women from more deprived areas are more likely to receive oral (vs transdermal) MHT, a potential health inequality [for] MHT.”
Body mass index data and smoking status were not available in the national registries used as part of this study, which the researchers highlighted as an important limitation.
“This may have caused an underestimation of the associations between menopausal hormone therapy use and thrombosis risk, since clinicians may be more reluctant to prescribe menopausal hormone therapy to women with increased cardiovascular risk factors,” the researchers noted.
“However, we believe that this potential bias is small; we considered socioeconomic and educational status in the study, which is highly correlated with body mass index and smoking status in Denmark.”
A further limitation was the fact that researchers did not have access to information about the age of onset of menopause for participants.
“This information gap is particularly relevant for the higher observed risk among women starting menopausal hormone therapy before the age of 50 and those with long term, high dose use, which may partly reflect confounding by indication and differences in baseline risk related to menopausal history,” the researchers said.
Dr Meaidi concluded her opinion piece by saying that further research was needed to better understand the potential risks and benefits of the same menopause treatment for different women.
“The goal should not be to return to an era in which women were unnecessarily frightened away from effective treatment, but neither should we enter an era in which enthusiasm for menopause care makes uncertainty disappear from the conversation,” she wrote.
“Women deserve to know when treatment is likely to help, what risks may be relevant to them, and where evidence remains uncertain. Clinicians need the same information, and researchers need to ask more fundamental questions about menopause itself.”
BMJ, 23 September 2026 (research)
BMJ, 23 September 2026 (opinion)
