Dryness, Thrush and Unexpected Bleeding: Vaginal Symptoms on a GLP-1

Four complaints, one embarrassing heading

Somewhere in the second or third month, a woman on a GLP-1 notices something she has no comfortable way to describe. It might be dryness. It might be itching, or a discharge that has changed, or a smell that has. It might be a smear of blood at a point in the month where blood does not belong.

All four get typed into a search bar as roughly the same question, and all four come back with roughly the same answer, which is a page about "Ozempic vulva" — a phrase that had its viral moment and then collapsed, and which we have written about as a media event rather than a diagnosis. This piece is the other half: the four complaints taken separately, because they have different causes, different tests, and very different levels of urgency.

The labels do not mention any of it

Start where a reporter starts, with the primary document, and re-read it rather than remembering it.

We downloaded the current prescribing information for Wegovy and for Zepbound and counted. The words vaginal, vulvovaginal, candidiasis and mycotic appear zero times in the Wegovy label¹ and zero times in the Zepbound label². These are not short documents that skip the small stuff. The Wegovy label characterises hair loss down to the fraction of a percent and splits it by sex. The Zepbound label reports hair loss in 7.1% of female patients against 1.3% on placebo. A drug-attributable vulvar or vaginal effect of any frequency would have a line.

So nothing below is a known side effect of the medication. All of it is a plausible consequence of losing weight fast, in a woman who is often also somewhere in the perimenopausal transition, sometimes on a second drug, and occasionally simply unlucky.

Dryness and irritation: usually oestrogen, not the pen

Dryness is the most common of the four and the one most likely to be misattributed.

The condition clinicians recognise here is the genitourinary syndrome of menopause — the vulvovaginal, urinary and sexual symptoms that follow falling oestrogen and androgen levels through the menopausal transition. The 2025 joint guideline from the American Urological Association, SUFU and AUGS is explicit that GSM is diagnosed from symptoms, with or without physical findings, and only after ruling out other causes or co-occurring conditions³. That sentence cuts both ways. It means dryness deserves a real assessment rather than a shrug. It also means it should not be pinned on the newest thing in your bathroom cabinet just because that is the newest thing in your bathroom cabinet.

For a woman in her forties, the medication and the transition frequently arrive in the same eighteen months. Untangling them is a clinical job, and it is worth doing, because the treatment is different depending on the answer. The appetite half of that same transition is its own piece.

Thrush and BV: what the weight data actually shows

Itching, an unusual discharge or a change in odour points toward infection rather than tissue change — and infection is tested, not guessed at.

What weight loss does to that risk is genuinely unsettled, and it is worth saying so plainly. For bacterial vaginosis, a cross-sectional study of 5,918 women in the Contraceptive CHOICE Project found BV in 21.3% of lean women, 30.4% of overweight women and 34.5% of women with obesity, with Nugent scores rising alongside body mass index⁴. A separate NHANES analysis of 5,428 women pointed the same way, at an odds ratio of 1.03 per unit of BMI⁵. Read straightforwardly, that suggests losing weight ought to lower BV risk, not raise it — but neither study followed anyone through weight loss, so it is an inference, not a finding.

The one large dataset that has looked at vulvovaginal outcomes on these drugs specifically runs against the panic too. A phenome-wide analysis of 17,267 adults with type 2 diabetes in the All of Us programme screened 974 diagnoses after a prescription; set against an SGLT2 inhibitor, semaglutide was associated with less candidiasis of the vulva and vagina, at a per-protocol hazard ratio of 0.31 with a confidence interval of 0.17 to 0.55⁶. The comparator there is a drug class notorious for genital infection, which flatters the result. But it is the only large body of evidence aimed at this part of the body, and it does not point where the headlines did.

Nobody has studied vulvar tissue on a GLP-1. That is not reassurance. It is an absence, and it is the honest thing to say about it.

The bleeding that is never "probably nothing"

Three of these symptoms can reasonably be watched for a fortnight. One cannot.

Bleeding that is not a period — after sex, between periods, or at any point after menopause — is not something to sit on while you decide whether the medication caused it. Clinical practice recommendations for investigating postmenopausal uterine bleeding treat it as requiring evaluation on presentation, because the differential includes endometrial hyperplasia and endometrial cancer, and because that risk is elevated in women with obesity⁷. It is also the symptom most likely to be quietly rationalised as hormonal noise, particularly by a woman whose cycle has been unpredictable anyway.

Changes to the period itself — heavier, longer, more painful, or arriving at a new interval — are a separate story with a separate mechanism, and they are covered here. This section is only about blood that is not menstrual. If you cannot tell which you are looking at, that uncertainty is itself the reason to be seen.

What actually treats it

The treatment picture is better than the evidence picture, which is a pleasant change.

The most rigorous synthesis to date pooled 46 randomised trials of hormonal treatments and vaginal moisturisers for GSM. Vaginal oestrogen may improve dryness, painful sex, a woman's own most bothersome symptom and her satisfaction with treatment. Vaginal DHEA and oral ospemifene may improve dryness and dyspareunia, and vaginal moisturisers may improve dryness. Vaginal testosterone, systemic DHEA and oral raloxifene showed no benefit or uncertain effects⁸. The authors are careful about the certainty of all of it — most trials ran twelve weeks or less — and the AUA guideline's own conclusion is that low-dose local vaginal oestrogen has by far the most robust evidence behind it³.

Lubricants and moisturisers are not the same product and not interchangeable: one is used at the time, the other on a schedule to change the tissue. That distinction is worth getting right before concluding that "nothing works".

The reason none of this gets said out loud

The last obstacle is not clinical.

The Women's EMPOWER survey put the question to 1,858 American women aged 45 and over who had vulvar and vaginal symptoms. Of the respondents who had never used any treatment, 72% had never discussed their symptoms with a health professional at all — most often because they had decided the symptoms were a natural part of ageing and something to live with. Where a conversation did happen, the woman started it 85% of the time. Overall, clinicians had recommended vaginal oestrogen therapy to 23% of these women, and 81% had not known that what they had was a recognised medical condition⁹.

That silence predates the medication by decades, and a joke-shaped nickname stapling a drug brand to a body part has not made it easier to break. But the asymmetry in that survey is also the instruction. If the conversation happens, it is overwhelmingly because a woman decided to start it. Start it with the person who wrote the prescription — and if the bleeding is the symptom, start it this week.