What Is Premarin Made From? Horse Urine Explained

Premarin is made from the urine of pregnant mares. The name itself is a portmanteau: Pregnant mare urine. The drug contains a complex mixture of at least ten different estrogens extracted and purified from that urine, and it has been prescribed to women for menopausal symptoms since the 1940s.¹ That a blockbuster pharmaceutical still depends on a biological supply chain involving thousands of horses surprises many people, and the story behind it touches on pharmacology, animal welfare, and the ongoing search for alternatives.

What Is Actually in the Urine

Pregnant mares produce unusually large quantities of estrogens during the second half of gestation. Their urine contains a cocktail of estrogen compounds, many of which are not found in meaningful amounts in humans. The manufacturer has reported that Premarin contains a mixture of ten estrogens, though the complete composition and the relative contribution of each component have never been fully disclosed.¹ The major components include estrone (the most abundant), equilin, and equilenin, along with their various metabolites. Estrone is also a human estrogen, but equilin and equilenin are equine-specific, meaning they are naturally produced by horses and not by people.

These horse-specific estrogens are not inert in the human body. They bind to human estrogen receptors and trigger biological activity, though they do so differently from the human body’s own estradiol. Research on receptor binding has shown that most of these ring B unsaturated estrogens (the chemical class that includes equilin and equilenin) bind to estrogen receptors with lower affinity than estradiol, but they show a preference for one receptor subtype over another. Equilenin, for instance, was the most active of all tested estrogens at the beta estrogen receptor, exceeding even estradiol’s activity at that receptor.² This means Premarin does not act like a single drug with a single mechanism. It is a blend of compounds with overlapping but distinct biological effects, which makes it pharmacologically unusual and difficult to replicate exactly.

Why Pregnant Mares Produce So Much Estrogen

The high estrogen output during equine pregnancy traces back to the fetus itself. Research in the 1970s showed that when both fetal gonads were surgically removed from pregnant mares, urinary estrogen levels dropped rapidly to very low levels. When only one fetal gonad was removed, estrogen output fell to roughly half of what it had been before.³ The fetal gonads play a central role in driving estrogen production during pregnancy, which is why the urine of non-pregnant mares or mares in early pregnancy would not work for manufacturing Premarin. Collection takes place during a specific window of gestation when estrogen output is at its peak.

How Premarin Became One of the Most Prescribed Drugs in History

Premarin was originally developed by the Canadian pharmaceutical firm Ayerst, McKenna and Harrison, which later became part of Wyeth (now owned by Pfizer).¹ Approved in the 1940s for treating menopausal symptoms, it grew steadily for decades and became the single most prescribed drug in the United States by 1992.⁴ At its peak, millions of women were taking it, not only for hot flashes and vaginal dryness but also with the expectation that it would protect against heart disease and keep bones strong well into old age.

That trajectory reversed sharply in 2002 when the Women’s Health Initiative, a large government-funded clinical trial, published results showing that the combination of conjugated equine estrogens plus a progestin raised the risk of breast cancer, stroke, and blood clots. Prescriptions fell dramatically: estrogen therapy use dropped by about 45%, and combined estrogen-progestin therapy use dropped by about 22%.⁴ The story since then has been about nuance. A separate arm of the WHI that tested conjugated equine estrogens alone (without a progestin, in women who had already had a hysterectomy) found a more balanced picture: increased stroke risk but reduced hip fractures and a borderline reduction in breast cancer.⁵ That finding pointed squarely at the progestin component as a major contributor to the breast cancer signal, not the estrogen alone.

What the WHI Told Us About Age and Timing

One of the most important insights to come out of the WHI data was that the age at which a woman starts hormone therapy matters a great deal. A reanalysis focused on women aged 50 to 59 who took conjugated equine estrogens alone found a favorable profile: for roughly every 76 women treated with CEE instead of placebo, one fewer had a heart attack, and for roughly every 37 women treated, one fewer experienced a serious adverse outcome across a composite of heart disease, cancer, stroke, and other conditions.⁶ For older women who started therapy well past menopause, the risks were higher. This “timing hypothesis” has reshaped clinical practice: most guidelines now suggest that hormone therapy, when appropriate, should be started relatively close to the onset of menopause rather than decades later.

Long-term follow-up of the WHI confirmed this nuanced picture. For conjugated equine estrogens alone, most risks and benefits seen during the treatment period faded after women stopped taking the drug, and a reduction in breast cancer actually became statistically clearer during the post-treatment follow-up period.⁷ For the combination of CEE plus a progestin, however, the breast cancer risk persisted even after women stopped treatment.

What Premarin Does Clinically

Premarin has been used primarily for two purposes: relieving menopausal symptoms and preventing bone loss. For vasomotor symptoms like hot flashes, the standard dose (0.625 mg per day) combined with a progestin performed similarly to lower doses in clinical trials, though the standard dose alone was more effective than lower doses of CEE without a progestin.⁸

For bone health, a prospective study found that conjugated estrogens at 0.625 mg slowed bone loss to roughly a third of the rate seen in postmenopausal women not taking estrogen, and the protective effect held across a wide age range from 51 to 80.⁹ Newer formulations have paired conjugated estrogens with bazedoxifene (a selective estrogen receptor modulator) instead of a progestin, aiming to protect the uterine lining without the drawbacks associated with progestins.¹⁰

The Endometrial Safety Question

If you still have a uterus, taking any estrogen without something to oppose its effects on the uterine lining raises the risk of endometrial overgrowth. A Cochrane review found that unopposed estrogen therapy roughly six- to ninefold increased the odds of endometrial hyperplasia compared to placebo, depending on the duration of use.¹¹ This is why Premarin prescribed to women with an intact uterus is almost always paired with a progestin or with bazedoxifene. Vaginal formulations of conjugated estrogens, used at low doses for localized symptoms like dryness, generally pose less systemic risk, though even in one small study of vaginal CEE cream used at higher doses, a few cases of hyperplasia were detected.¹²

Animal Welfare and How the Urine Is Collected

The collection of pregnant mare urine has drawn persistent criticism from animal welfare organizations. The traditional method, used extensively on farms in Canada and North Dakota, involves housing mares in tether stalls for much of their pregnancy so that urine can be captured via a collection device attached to the mare. Critics have argued that the confinement restricts natural movement and social behavior.

An alternative collection procedure developed in Australia allows mares to be loose-housed in indoor stables or outdoors in paddocks rather than tethered in stalls. A study evaluating the welfare effects of this alternative method on pregnant mares found it to be a viable approach, though the research focused on whether the collection device itself posed welfare risks to loose-housed animals.¹³ The existence of such alternatives suggests that the supply chain does not inherently require the most restrictive housing conditions, though adoption of improved methods across the industry has been gradual.

Beyond the pregnant mares themselves, there is the question of what happens to the foals. Mares on PMU farms are bred annually to maintain pregnancy, and the resulting foals have historically entered a range of outcomes: some are kept for breeding, some are sold as riding horses, and some have ended up in the slaughter pipeline. The scale of the operation has shrunk since the post-WHI drop in prescriptions reduced demand for the urine, meaning fewer farms and fewer mares are involved than at the peak in the late 1990s.

Bioidentical and Synthetic Alternatives

One of the most common follow-up questions people have after learning what Premarin is made from is whether there are alternatives that do not come from horses. The answer is yes, and the alternatives fall into two broad categories: bioidentical hormones and synthetic conjugated estrogens.

Bioidentical estradiol (usually written as 17β-estradiol or simply E2) is chemically identical to the estrogen the human body produces. It is manufactured from plant precursors, typically soy or yam, through chemical synthesis. Combined with micronized progesterone rather than a synthetic progestin, it has become increasingly popular. There is growing evidence that this combination may carry a somewhat different risk profile than Premarin plus medroxyprogesterone acetate. A study of breast tissue found that while both conjugated equine estrogens and oral estradiol increased breast cell proliferation, the increase was substantially larger with estradiol, which might sound alarming at first glance, but the picture is more complex: CEE raised proliferation by about 75%, while oral E2 raised it by roughly 260 to 330%.¹⁴ However, when researchers looked at breast cancer-related gene expression in a randomized trial comparing CEE plus MPA against estradiol plus micronized progesterone, the combination with the synthetic progestin (MPA) activated far more mammary tumor development genes than the bioidentical combination did.¹⁵ The progestin component, in other words, appears to matter as much as or more than the estrogen source.

Blood clot risk also appears to differ. A real-world analysis comparing oral estradiol plus micronized progesterone to oral CEE plus MPA found that the bioidentical combination was associated with about 30% lower rates of venous blood clots.¹⁶ These findings have driven a shift in prescribing patterns, with many clinicians now preferring bioidentical formulations, particularly for women concerned about the horse-derived origin of Premarin or about the specific risks tied to the older drug combination.

Why a Synthetic Copy Is Harder Than It Sounds

Given that Premarin is a mixture of at least ten compounds in specific proportions, you might assume that a chemist could simply synthesize each component and blend them together. Attempts have been made. A formulation of nine synthetic plant-derived conjugated estrogens was shown to match Premarin’s neuroprotective effects in laboratory studies on cultured brain cells, protecting hippocampal neurons against several types of toxicity at levels comparable to the natural equine mixture.¹⁷

But neuroprotection in a dish is not the same as clinical equivalence in a human body. When researchers compared the pharmacokinetics of a synthetic conjugated estrogen formulation (marketed as C.E.S.) to Premarin in postmenopausal women, the two were not bioequivalent. The way the body absorbed, distributed, and metabolized the synthetic version differed enough that the researchers concluded it would be risky to assume the long-term benefits established for Premarin would apply to the synthetic copy.¹⁸ This is partly because Premarin’s complexity goes beyond a list of named ingredients. The relative ratios of its components, and trace constituents that have not been fully characterized, are difficult to reproduce exactly from scratch. Premarin occupies an unusual place in pharmacology: a drug so complex that even replicating it chemically has proven unreliable.

Environmental Concerns Around Estrogen-Rich Waste

Any discussion of an industry that handles large volumes of hormone-rich biological material raises questions about environmental contamination. Estrogens, whether from human wastewater, agricultural runoff, or pharmaceutical production, are increasingly detected in waterways near treatment facilities and in groundwater. At polluting levels, estrogens have been linked to disrupted reproductive development in fish and other aquatic organisms, as well as concerns about effects on human health at chronic low-level exposure. The environmental footprint of PMU farming specifically is not well-studied in the peer-reviewed literature, but the broader issue of estrogenic contamination from both human and animal sources is well established.

How Equine Estrogens Act Differently in the Body

The pharmacological story of Premarin gets interesting when you look at how its horse-specific estrogens interact with human biology. Human estrogen receptors come in two main subtypes, alpha and beta, which are distributed differently across tissues. The alpha subtype predominates in breast and uterine tissue, while the beta subtype is more common in bone, brain, and the cardiovascular system. Research has shown that equine estrogens like equilin and equilenin, while weaker than estradiol at the alpha receptor, are relatively more active at the beta receptor. Equilenin triggered beta receptor activity at nearly three times the level of estradiol in laboratory assays.²

This receptor preference has practical implications. Because the beta receptor is less involved in stimulating breast and uterine tissue growth, the equine estrogens in Premarin may exert some of their beneficial effects (on bone, for instance) through a pathway that carries less proliferative risk to those tissues. The researchers who characterized these receptor interactions suggested it might eventually be possible to develop hormone therapy using selected equine-type estrogens targeted at tissues where the beta receptor predominates.² That idea has not yet made it into a marketed product, but it illustrates how the horse-derived origin of Premarin is not just a quirky manufacturing story. The equine estrogens are biologically distinct molecules with properties that human-identical estradiol does not share.

Whether those properties are ultimately advantages or disadvantages is still being sorted out. Some of Premarin’s equine components have longer half-lives in the body than estradiol, meaning they linger and continue signaling for longer periods. For a woman taking the drug daily, this produces a steadier estrogenic background, which some clinicians view favorably for symptom control. Others see it as a reason for concern: a compound you cannot clear quickly is one that is harder to dial back if side effects emerge. The debate over whether horse-derived estrogens are meaningfully better or worse than bioidentical estradiol for any given patient remains one of the more contested questions in menopause medicine, and the answer likely depends on individual circumstances including which symptoms are being treated, personal risk factors, and how a woman’s body metabolizes each formulation.