Vaginal estrogen: a potentially life-saving therapy

In any given year, about 11% of women experience at least one uncomplicated urinary tract infection (UTI), and 37% of them have at least one UTI during their lifetime.¹ A first UTI often leads to more: 14.5% of women with a first infection go on to suffer recurrent bouts of UTI,² and one in six women who meet criteria for recurrent infection go on to have at least two further infections within the next year.³ 

Most people think of UTIs as little more than a painful but transient nuisance. And in many younger women, they are—in the short term. But in addition to near-term recurrences, UTIs at a younger age are also the strongest and most consistent risk factor for UTIs in menopause⁴—and in older women, UTIs can be deadly. From 1999 to 2023, over a million American adults aged 25 and older died from UTI-related causes, with women being twice as likely to die of this cause as men.⁵ Relative to women aged 18–50, women aged 51–65 who go to the hospital with an UTI are over three times as likely to die from it (odds ratio (OR) 3.17, 95% CI 2.12–4.73), rising to over 8 times as likely for women over 85.⁶

What is driving this terrible disparity—and more important, what can a woman do to keep herself from becoming a victim?

The shearing of Rapunzel’s hair

UTIs happen most frequently in young, sexually active women,⁷ but menopausal women are more biologically vulnerable to UTIs. The reason for this is the withdrawal of estrogen from the vulva, vagina, and urinary tract. The genitourinary syndrome of menopause (GSM—formerly known as “vulvovaginal atrophy” or “atrophic vaginitis”) refers to the ensuing structural changes in these organs and the associated symptoms, such as vaginal itchiness or dryness, pain during penetrative sex, and urinary symptoms that make it feel like a woman has an UTI when she doesn’t.⁸

In addition to these symptoms, GSM makes a woman more vulnerable to actual UTIs.⁸ When healthy, the cells lining the lower urinary tract form the most impermeable membrane in the mammalian body, which allows it to keep pathogens from getting into the blood, where they can cause sepsis.⁹ Estrogen helps keep these cells replacing themselves¹⁰ and stimulates them to join together seamlessly and express peptides that directly destroy microbes and help the immune system fight them off.¹¹ The withdrawal of estrogen in menopause thins the lining of the urinary tract, reduces its blood supply, and alters the composition of its connective tissue. These effects raise the risk of UTI by weakening the body’s ability to close off the opening where urine exits the body and impairing the barrier function of the urinary tract lining.¹²

Meanwhile, the absence of estrogen causes the cells in the vaginal wall to produce less glycogen, which is needed to maintain the beneficial Lactobacillus bacteria in the vaginal microbiome. These bacteria produce lactic acid that makes the vagina inhospitable to potentially pathogenic microbes such as E. coli, the most common culprit in UTIs.^13,14 Menopause is associated with decreased vaginal Lactobacilli, and an abnormal increase in the overall diversity of the vaginal microbiome, which in this case reflects the expansion of species that are normally kept in check by the beneficial Lactobacillus bacteria.¹³ 

One common solution to recurrent UTIs is continuous low-dose antibiotics. While effective, this approach runs the risk of driving antibiotic resistance, as well as side effects such as oral and vaginal candidiasis (”Candida”), C. difficile, organ toxicities, and gastrointestinal problems.^15,16,17

Get back to where you once belonged

Fortunately, there is a more physiologic and obvious solution to this problem: replace vaginal estrogen to restore the normal functions of the urinary tract and vagina. However, there is little evidence to suggest that conventional, systemic menopausal hormone therapy (MHT) (pills, patches, or transdermal creams) provide meaningful help on this front.¹⁸ 

Instead, the best evidence for hormone replacement for GSM and associated UTIs is local treatment of the vagina itself, via topically applied estrogen creams and gels and vaginal tablets, gel caps, and rings. This therapy is recommended by the American Urological Association’s guideline for postmenopausal women with recurrent UTIs.¹⁵ There’s also more limited evidence for topical DHEA (generally labeled as “prasterone” for historical reasons, and originally approved as Intrarosa^®), which converts locally into estrogen and testosterone and is approved for GSM.¹⁹

While its benefits for the other effects of GSM were proven in larger and earlier trials, the protective effect of local estrogen therapy against UTIs was first demonstrated in a 1993 trial published in the New England Journal of Medicine.²⁰  Raz and Stamm demonstrated that eight months’ treatment of menopausal women with recurrent UTIs with intravaginal estriol cream (estriol is a form of estrogen) slashed UTI recurrence by almost twelvefold relative to placebo (0.5 versus 5.9 UTIs per person per year, p<0.001).²⁰ 

This trial also gave us some insight into how the treatment provides this protection. Before they started treatment, there were no Lactobacillus bacteria in any woman’s vaginal cultures; Lactobacilli emerged after one month of treatment in 22 of the 36 estriol-treated women, while none of the 24 placebo-treated women improved. Similarly, estriol recipients’ vaginal pH declined from 5.5 to a more physiologically appropriate 3.8 (p < 0.001), reflecting increased lactic acid production, whereas there was no significant change in the placebo group.²⁰

Later trials with other products²¹ and observational studies²² also demonstrated vaginal estrogen’s protective effects, though the magnitude of effect was not as dramatic. In one trial, 108 menopausal women with recurrent UTIs were randomly assigned to either a vaginal ring that delivers estradiol (the main form of estrogen) locally to the vagina or a placebo ring with no estrogen for 36 weeks.²³ By the end of the trial, the estradiol ring had cut the rate of new UTIs from 45% in the placebo group to 20% in the estrogen ring group—and even in the women who did suffer a recurrence, the ring delayed the time it took for the first recurrence to occur.²³

And these results might have underestimated the full power of the ring (pace Tolkien), since women in the placebo group were taken out of the analysis after their first recurrence. The estrogen ring also lowered vaginal pH, and supported more mature cells in the vaginal lining and the lining of the urethra (the tube that carries urine from the bladder to the exit from the body),²³ giving more clues into how vaginal estrogen delivers protection.

While trials suggest product-to-product variations, every trial with every form of vaginal estrogen tested found that they reduce risk of UTIs in women with recurrent UTIs,^21,24 with a meta-analysis reporting a pooled 60% reduction in risk (RR= 0.4, 95% CI 0.33-0.59). If anything, this risk estimate understates the effectiveness of vaginal estrogen, since the overall effect in the meta-analysis was dragged down by one trial of an “ultra-low-dose” 0.005% estriol vaginal gel that only reduced risk by 26%.²⁵ And all this is in addition to what these medications were designed to do in the first place: restore a woman’s comfort and wellbeing in the most private parts of her body, and make intimacy with a partner free of pain.

The lives-saved count

No one has yet done the large, multi-year trial that would be required to prove that vaginal estrogen’s ability to prevent UTIs translates into reduced rates of mortality from this cause. However, a recent observational study²⁶ gives us a window of insight, albeit one where we need to rest cautiously on the sill. Researchers from the Albany Medical College and others used an electronic health record database to identify women with at least 2 separate coded UTIs within a 6-month period, and looked to see whether these women had or had not received a vaginal estrogen prescription within 2 months of their second UTI. They then looked forward to see what happened to them over the next 8 years.

Across all age groups, vaginal estrogen was associated with greatly reduced risk of all UTI-related harms. Depending on age group, women using vaginal estrogen were between 40% and two-thirds as likely to be hospitalized with a recurrent UTI as nonusers in the same age group (relative risks (RR) 0.39–0.66); in the key age group of 55–69, users were half as likely to be hospitalized (RR 0.48, 95% CI 0.44–0.51).²⁶ That association, if entirely causal, would mean that one hospitalization would be prevented for every 10 women treated for 8 years. If that were a clinical trial result, it would be a remarkable “number needed to treat” (NNT) statistic.

The results were similar for sepsis, the fearsome, life-threatening condition in which the body’s own response to infection causes a person’s organs to shut down. Depending on age group, women using vaginal estrogen were between one quarter and half as likely to suffer sepsis after recurrent UTI as nonusers in the same age group (RRs 0.27–0.48); amongst women aged 55–69, users were only a quarter as likely to suffer sepsis as nonusers (RR 0.27; 95% CI 0.26–0.28). If those numbers reflect causality—and again, all this study shows is an association— it would mean that just 6 women would need to use vaginal estrogen to prevent a case of sepsis (NNT 6).²⁶

And a similarly strong association held between vaginal estrogen use and survival in the years after an UTI. Mortality increased monotonically with age among users and nonusers alike (as indeed would occur with age even in people without UTIs), but women using vaginal estrogen were only one- to two-fifths as likely to die in the 8 years following recurrent UTI as nonusers in the same age group. In the prime 55–69 age group, users were only one fifth as likely to die as nonusers (RR 0.21; 95% CI 0.19–0.22). Again, that’s an observational finding, not the result of a randomized trial, but if the effect were fully causal, it would imply an NNT of 18—still impressive population-level effectiveness.²⁶

We hasten to add that these numbers can’t be taken at face value. This is an observational study, not a clinical trial, and there are doubtless important differences between women who began using vaginal estrogen after suffering repeated UTIs and those who did not. Women in the 55–69 age group who went on vaginal estrogen were less likely to have diabetes than nonusers (38.1% vs. 48.9%) and less likely to have kidney disease (27.5% vs. 37.0%)—though vaginal estrogen was also associated with greatly reduced risk of all severe outcomes in women in the 20–39 age group, in whom these and other risk factors were more likely to be present among users than nonusers.^*

Beyond the clinical characteristics captured by their electronic health records, it is reasonable to think that women who began vaginal estrogen after recurrent UTIs are likely to differ from those who didn’t in other ways that are hard to measure but impactful, such as being more health-conscious, having higher health literacy, having greater ability to navigate the healthcare system, or having a more knowledgeable and proactive doctor—the kinds of thing that cluster under the idea of “healthy user bias.”

Still, every clinical trial that has tested the question has found that vaginal estrogen slashes the rate of new UTIs in women with recurrent infections—in all but one case (the ultra-low-dose formulation), by more than half.²⁴,²¹ And we have human and other evidence that it reinforces the barrier function of the urinary tract,^11,23 which would be expected to be protective against UTIs progressing into blood-borne infections.²³ Shoring up this vulnerability could be key, as blood-borne infections are the starting point for the plurality of cases of sepsis.²⁷

So while the true ability of vaginal estrogen to prevent sepsis and death may not be as dramatic as this observational study suggests, the human trials and mechanistic evidence make it reasonable to think that the reduction in risk is real. Where antibiotics can treat or prevent UTIs by killing bacteria, topical estrogen helps restore the health and integrity of the vagina and urinary tract themselves, which delivers multiple health and quality of life benefits, including shoring up a woman’s own defenses against infection—and there’s good reason to think that it even prevent sepsis and death related to UTIs. For menopausal women with GSM or UTIs, the value of vaginal estrogen is clear.

Our research team spends hundreds of hours each month vetting studies and distilling dense literature to deliver evidence-informed insights on health and longevity. If you find value in our work, consider becoming a premium member and supporting our mission.

Join Now

Grossly underutilized

In addition to benefitting the personal health, comfort, and even survival of individual women, a modeling study from researchers at Georgetown University (including Drive guest Dr. Rachel Rubin) found that wider use of vaginal estrogen would save the healthcare system money. Using the published cost of treating UTIs per Medicare beneficiary, the likelihood of complicated and resistant infections, the established effects of vaginal estrogen, and the cost of the therapy itself, these researchers found that the annual cost savings for topical estrogen in postmenopausal women with recurrent UTIs were $1,226 to $4,888 per patient.²⁸

And yet, studies consistently find that only a tiny fraction of women who could benefit from these medications receive them. In the electronic health record observational study we discussed in the last section, just 5.1% of the nearly 2 million women with recurrent UTIs received vaginal estrogen. While that number is skewed by the many women in the study who were so young that that few would benefit from this therapy,^* usage was low (only 21%) even among women in the prime demographic (ages 55–69).²⁶

Similarly, a study of US commercial insurance claims for women 50 years and older found that the age-standardized annual prevalence of a vaginal estrogen prescription claims peaked in 2011 at 4.2% of women, and then declined to 3.5% by 2015.²⁹ Even among female Medicare beneficiaries with GSM-related diagnoses, only about 9% fill a vaginal estrogen prescription.³⁰

Some of this underutilization is simple lack of education among doctors and awareness among women themselves of the benefits of vaginal estrogen. But some of it is unjustified fear—a hangover from the hasty alarmism that attached itself to systemic MHT almost a quarter century ago. This long shadow has been the subject of many an episode of The Drive, such as those with Drs. Carol Tavris, JoAnn Manson, and Rachel Rubin.

That fear is out of proportion to the actual risks, even for systemic MHT—and the risks are lower still for vaginal estrogen. Although the product labels for vaginal estrogen list some scary side effects, including jaundice and blood clots,^31,32 these were not reported in the clinical trials for these products,^20,21,32,33 and appear to have been imputed from older oral MHT products. The side effects reported in trials include increased vaginal bleeding or spotting, breast tenderness or pain, increased vaginal secretions, vulvovaginal skin rash, vaginal discharge, and headaches—and in some cases these outcomes did not differ between the estrogen product and placebo.^20,21,32,33 A Cochrane pooled analysis of trials that used biopsy to assess the risk of dangerous overgrowth of the lining of the uterus (a theoretical risk of these products) found a nonsignificant 2% incidence of simple increased growth when a vaginal ring was compared with cream, and a 4% incidence of increased growth when an older conjugated equine estrogen cream was compared with an estradiol tablet.³⁴ Such overgrowth of the lining of the uterus from these products—and its worst consequence, endometrial cancer—appears to be rare, and was nonexistent in several newer studies.³⁵

Studies consistently show that vaginal estrogen has minimal effect on estrogen levels outside of the vagina and urinary tract.^36,37 Accordingly, a recent systematic review found that breast cancer survivors who used vaginal estrogen were not at greater risk of breast cancer recurrence or breast cancer-specific mortality than nonusers, with the only question mark being among women who were being treated with aromatase inhibitors (a class of estrogen-inhibiting therapies).³⁸ And breast cancer survivors are more likely to need treatment for GSM compared to similarly-situated women free of breast cancer. Seventy percent of postmenopausal survivors are afflicted with GSM, versus half of other postmenopausal women;³⁹ among premenopausal survivors, 23.4% have (at least) one specific GSM symptom (vaginal dryness), whereas it is rare in the general premenopausal population.⁴⁰ So women who have suffered breast cancer are at even greater need of a safe therapy for their symptoms and for protection against UTIs.

A recent observational study in Denmark has also addressed the worry—again, not grounded in trial evidence—that vaginal estrogen might cause blood clots and lead to a stroke. Among 34,274 postmenopausal women aged 45 and older (but with a median age of 75) who had experienced a first ischemic stroke, the study found that there was no greater risk of a second stroke during followup among women who were current, recent, or long-past users of vaginal estrogen than those who were not, adjusting for age.⁴¹ There was also no signal of risk in comparing high- versus low-dose users. The lack of additional stroke risk even in these very high-risk women—and the lack of trial evidence—suggests that the concern about stroke risk is unfounded.

The bottom line

For postmenopausal women—and women with premature estrogen deprivation for other reasons—vaginal estrogen is a safe and highly effective therapy to alleviate the genitourinary syndrome of menopause, prevent recurrent UTIs, and likely to save lives. Although no doctor caring for an individual patient should robotically follow the guidelines, this is one case where the guidelines have it right.¹⁵ Yet it is grossly underutilized. If you are a woman in this situation, you don’t have to suffer; and if your wife, mother, sister, or friend is in this situation, don’t let them.

* That isn’t to say that vaginal estrogen is never indicated for women too young to experience age-related menopause. Cases do crop up in which a young woman suffers GSM due to premature estrogen deprivation. This can result from hormone deprivation for the treatment of breast cancer, primary ovarian insufficiency, and (less often) breastfeeding or the use of hormonal contraception. These underlying conditions might explain why vaginal estrogen users in the younger age group were sicker than nonusers—the reverse of what’s seen with conventional menopause.

For a list of all previous weekly emails, click here. 

podcast | website | ama