HRT Side Effects: What the Latest Research Shows About Risks and Benefits
- The 2002 WHI findings that drove widespread HRT discontinuation were substantially revised in later analyses; the original trial used older women, oral estrogen, and a specific synthetic progestogen not representative of modern HRT.
- Transdermal estradiol does not increase blood clot or stroke risk, unlike oral estrogen; this is not a minor distinction.
- Estrogen-only HRT (in women without a uterus) is associated with lower breast cancer risk than placebo over 7 years in the WHI data.
- Combined estrogen-progesterone HRT shows a modest increase in breast cancer risk after 5 years; the absolute risk is smaller than commonly believed and depends heavily on the type of progestogen used.
- Cardiovascular benefit from HRT is real but timing-dependent: starting within 10 years of menopause or before age 60 is associated with reduced coronary heart disease risk.
Hormone replacement therapy has had an unusually turbulent evidence history. A single large trial published in 2002, the Women’s Health Initiative (WHI), caused millions of women to stop HRT within months, many of them abruptly, and the ripple effects on prescribing patterns lasted over two decades. The problem is that the WHI findings were substantially misinterpreted and overgeneralized, and the subsequent reanalyses tell a materially different story.
This article covers the actual evidence on HRT side effects and risks, route-of-administration differences that matter clinically, the breast cancer data in context, and a framework for weighing individual benefit versus risk.
Common Side Effects by Delivery Method
Most common side effects of HRT are dose-dependent, delivery-method dependent, and typically resolve within one to three months as the body adjusts. The table below summarizes the pattern.
| Side Effect | Oral Estrogen | Transdermal Estrogen (patch/gel/spray) | Vaginal Estrogen | Oral Progesterone/Progestogen | Progesterone IUD |
|---|---|---|---|---|---|
| Breast tenderness | Common (20-30%) | Less common (10-15%) | Rare | Common | Uncommon |
| Bloating/fluid retention | Common | Less common | Rare | Moderate | Uncommon |
| Headache/migraine | More common; variable estrogen levels | Less common; more stable levels | Rare | Less common with micronized progesterone | Uncommon |
| Mood changes/irritability | Uncommon | Uncommon | Rare | More common with synthetic progestogens (MPA); less with micronized progesterone | Variable |
| Nausea | Common (first 4-6 weeks) | Uncommon | Rare | Uncommon | Rare |
| Skin irritation | N/A | Uncommon (5-10%); rotate application sites | Rare | N/A | N/A |
| Irregular bleeding | Common in first 3-6 months | Common in first 3-6 months | Rare | More common with continuous combined | Common initially; then amenorrhea |
These are side effects, not serious risks. They affect quality of life and are often why women discontinue HRT prematurely, but they can usually be managed with dose adjustment, formulation change, or delivery method switch. Switching from a synthetic progestogen (medroxyprogesterone acetate, norethisterone) to micronized progesterone (body-identical) substantially reduces mood and bloating-related side effects for most women.
What the WHI Actually Showed, and Where It Went Wrong
The WHI enrolled women with a mean age of 63 years, most of whom were 10 or more years past menopause. Roughly 35% had metabolic syndrome. The HRT used was oral conjugated equine estrogen (CEE) plus medroxyprogesterone acetate (MPA), a synthetic progestogen not used in most contemporary HRT prescribing. The trial was halted in 2002 when interim analysis showed increased risks of coronary heart disease, stroke, blood clots, and breast cancer.
The 2013 reanalysis of WHI data, published in JAMA by Manson et al. (n=27,347), stratified results by age and time since menopause onset. Women who initiated HRT in their 50s, or within 10 years of menopause, showed net health benefits: reduced all-cause mortality, reduced coronary heart disease events, and reduced total hip fractures, outweighing the modest risks. The harms seen in the original analysis were concentrated in older late-initiator women, the majority of the WHI population, not in women using HRT at typical menopause-treatment ages.
The original 2002 publication applied the findings from 63-year-old women, many with existing cardiovascular disease, to 50-year-old women just entering menopause. These are not the same population, and the risk-benefit calculation is not the same.
Blood Clot Risk: Oral vs Transdermal Estrogen
This is one of the most clinically significant distinctions in HRT prescribing. Oral estrogen undergoes first-pass liver metabolism, which increases hepatic production of clotting factors and raises venous thromboembolism (VTE) risk by approximately two-fold compared to non-use. This is a real and consistent finding across multiple studies.
Transdermal estrogen, delivered through patches, gels, or sprays, bypasses hepatic first-pass metabolism entirely. A landmark 2003 case-control study by Scarabin et al. in The Lancet (n=271 cases, 610 controls) found that transdermal estradiol was not associated with elevated VTE risk at any dose, while oral estrogen was associated with a four-fold higher VTE risk compared to non-users at standard doses.
A 2011 meta-analysis in BMJ confirmed this finding across 17 studies. Women with personal or family history of VTE, obesity, prolonged immobility, or thrombophilia (such as Factor V Leiden) should be prescribed transdermal rather than oral estrogen. For women without these risk factors, both routes are considered acceptable, but the transdermal route eliminates the VTE concern entirely.
Stroke Risk
Oral estrogen is associated with a modest increase in ischemic stroke risk (approximately 30% relative increase over baseline) based on WHI and observational data. The absolute risk increase is small in younger women with low baseline stroke risk, but it is not zero.
Transdermal estradiol at standard doses does not appear to increase stroke risk. A 2010 cohort study in BMJ (n=15,710 women with ischemic stroke, 59,958 controls) found that current use of transdermal estradiol was not associated with elevated stroke risk, while oral estrogen use was associated with a significant increase. This is another reason to prefer transdermal delivery, particularly in women with migraine with aura (which itself elevates stroke risk).
Breast Cancer: Parsing the Nuance
Breast cancer risk is the primary concern most women and clinicians cite regarding HRT. The evidence is more nuanced than headlines suggest.
Estrogen-Only HRT
In the WHI estrogen-only arm (women who had undergone hysterectomy, using CEE alone, n=10,739), women in the HRT group had a lower incidence of breast cancer than the placebo group after 7 years of follow-up. The hazard ratio was 0.77, meaning a 23% lower relative risk of breast cancer in the HRT group. This finding is often omitted from summaries of WHI risks. Estrogen-only HRT in appropriate candidates (women without a uterus) does not increase breast cancer risk over 7 years and may reduce it.
Combined Estrogen-Progestogen HRT
The combined arm (CEE plus MPA) showed a modest increase in breast cancer risk that emerged after five years of continuous use, with a hazard ratio of approximately 1.24 (24% relative risk increase). In absolute terms, this translated to roughly 8 additional breast cancer cases per 10,000 women per year of use.
Critically, the type of progestogen matters significantly. A large 2019 study by Vinogradova et al. in BMJ (n=98,611 breast cancer cases) found that oral micronized progesterone was associated with a substantially lower breast cancer risk than synthetic progestogens (MPA, norethisterone) at equivalent durations of use. This is consistent with biological evidence that micronized progesterone does not stimulate breast cell proliferation in the way that synthetic progestogens do.
The absolute risk numbers help calibrate the clinical conversation. Breast cancer occurs in approximately 23 women per 1,000 over 5 years in the 50-59 age group without HRT. With combined HRT using synthetic progestogens for 5 years, this rises to approximately 28 per 1,000. With estrogen plus micronized progesterone, the excess risk appears to be smaller and may approach that of estrogen-only in some analyses. Obesity alone confers a comparable or larger breast cancer risk increase than combined HRT in this age group.
Cardiovascular Benefit: The Timing Window
The healthy artery hypothesis, confirmed by ELITE and KEEPS trials, holds that estrogen’s cardiovascular benefits (reduced atherosclerosis progression, improved endothelial function, favorable lipid changes) are only realized when therapy is initiated before significant arterial disease has developed. Starting estrogen in women with established atherosclerotic plaques can destabilize plaques and raise early coronary event risk, which is what the WHI observed in its older population.
The ELITE trial (n=643) found that women who initiated oral estradiol within 6 years of menopause showed significantly less carotid intima-media thickness progression (a marker of atherosclerosis) over 5 years compared to late initiators and placebo. This structural benefit in the vasculature translates to reduced long-term cardiovascular disease risk.
Current consensus from the British Menopause Society, the Menopause Society (formerly NAMS), and the European Menopause and Andropause Society is that HRT initiated in women under 60, or within 10 years of menopause, has a favorable cardiovascular risk-benefit profile and should not be withheld on cardiovascular grounds in otherwise appropriate candidates.
Gallbladder Disease
Oral estrogen increases biliary cholesterol saturation and gallbladder bile stasis, raising the risk of gallstone formation. The WHI found approximately a 40% increase in gallbladder disease (mainly gallstones requiring surgery) in the oral HRT group. Transdermal estrogen has not been consistently associated with elevated gallbladder risk, as the hepatic first-pass effect is avoided. This is another reason to prefer transdermal delivery in women with existing gallbladder disease or risk factors.
Endometrial Protection: Why Progesterone Matters
Unopposed estrogen, meaning estrogen given without progesterone, significantly increases endometrial cancer risk in women with an intact uterus. The relative risk with unopposed oral estrogen use for 10 years approaches 10-fold over baseline. This was established well before the WHI and is not contested.
Adequate progestogen or progesterone must be given to any woman with an intact uterus receiving systemic estrogen. The minimum dose and duration required to achieve full endometrial protection depends on the specific agent. For micronized progesterone 200 mg taken for 12-14 days per month (sequential regimen) or 100 mg daily (continuous regimen), endometrial protection is well-established. The levonorgestrel IUD (Mirena) provides local endometrial protection with minimal systemic absorption and is used in some women who want to avoid systemic progestogen side effects.
Who Should Avoid HRT
Absolute contraindications to systemic HRT include: active or recent breast cancer (estrogen receptor-positive), active or recent endometrial cancer, unexplained vaginal bleeding, active venous thromboembolism (though transdermal estrogen may be considered once anticoagulated), and active liver disease with impaired liver function.
Relative contraindications requiring careful individualized assessment include: history of hormone-sensitive cancer, personal history of VTE (transdermal estrogen substantially mitigates risk), thrombophilias, migraine with aura (avoid oral estrogen; transdermal may be appropriate), severe active cardiovascular disease, and liver disease with normal liver function. Personal history of breast cancer is the most common contraindication encountered; management in breast cancer survivors involves oncologist input and depends on cancer type and stage.
Monitoring Schedule
Women starting HRT should be reviewed at 3 months to assess symptom response, side effect profile, and any early concerns. Annual review thereafter covers: continued symptom control and dose appropriateness, blood pressure (oral estrogen can raise it modestly in susceptible women), breast screening per age-appropriate national guidelines (not more frequent than standard for most HRT users), and a reassessment of the ongoing benefit-risk balance.
There is no universal stopping age for HRT. Current guidance from most specialist societies is that HRT should be continued as long as the individual woman continues to benefit from it, with annual re-evaluation of risks. The practice of stopping HRT at 60 or after a fixed 5-year period is not evidence-based and is increasingly being abandoned in specialist practice.
Weighing Your Personal Risk
The decision to use HRT is individual. Key inputs are: severity of menopausal symptoms (hot flashes, sleep, mood, cognitive, sexual), age and time since menopause, personal cardiovascular risk profile, personal and family history of breast cancer, thrombotic history, bone density, and personal preference regarding treatment approach and monitoring commitment.
For most women in their late 40s and 50s who are within 10 years of menopause onset and have no contraindications, the benefit-risk balance clearly favors HRT when symptoms significantly affect quality of life. The strongest risk mitigation strategy is to use transdermal estradiol combined with body-identical micronized progesterone, which substantially reduces the VTE, stroke, gallbladder, and breast cancer concerns associated with older formulations.
For a complete guide to hormone replacement therapy including dosing, formulation selection, and monitoring protocols, see the hormone replacement therapy guide.
