Can Uterine Polyps Cause Infertility? What the Trials Show

Sometimes, probably - but the evidence is weaker than the confidence with which it is usually stated. Polyps are found more often in women having trouble conceiving than in women who are not, and the mechanism people describe (a growth in the cavity making the lining a worse place for an embryo to settle) is plausible. Association and plausibility are not the same as proof that taking it out helps.

What exists is one randomised trial, of 215 women, in one setting: polyp removal before intrauterine insemination. When Cochrane reviewed the entire question in 2018 - hysteroscopic treatment of polyps, fibroids, septa and adhesions in subfertility - only two trials in the whole field met its inclusion criteria. Its verdict on the polyp evidence was "low-quality", and no trial in it reported live birth.

So the sentence a clinic website will give you - polyps cause infertility and removing them fixes it - is running well ahead of what has been measured. That does not mean refusing the procedure. It means the decision is a judgement call rather than a settled one, and you are entitled to see it framed that way.

In Madrid, 215 women with a polyp seen on ultrasound who were about to start IUI were randomly assigned either to have the polyp removed at hysteroscopy, or to have a diagnostic hysteroscopy with a biopsy and nothing removed. They then had up to four IUI cycles.

Ninety-three pregnancies happened: 64 in the group that had the polyp removed, 29 in the group that did not. That is roughly twice the chance of pregnancy, and it was not a marginal result. One detail is striking - about two thirds of the pregnancies in the polypectomy group occurred before the first insemination, which suggests the removal itself, rather than the IUI that followed, was doing the work.

Three honest limitations. The outcome was pregnancy, not live birth - the trial did not report how many of those pregnancies became babies. It was a single centre, published in 2005, and has not been replicated in a second randomised trial in twenty years. And Cochrane, restating the same data, put it as: if 28 women in 100 conceive without removal, the evidence suggests about 63 would after it - a range it gives as 45 to 89, which is how uncertain a 204-woman comparison actually is.

There is no randomised answer. Cochrane looked and reported finding no studies of any suspected cavity abnormality before IVF or ICSI. A 2017 review in the Journal of Obstetrics and Gynaecology of India put it the same way: when a polyp is suspected during stimulation for IVF, there is insufficient evidence to recommend one option over another, and a properly designed randomised trial is needed.

In practice most units remove a polyp before transferring an embryo, and that is a defensible position - an embryo transfer is expensive, cannot easily be repeated, and nobody wants to put one into a cavity with a question mark in it. What it is not is an evidence-based recommendation in the strict sense. The cost argument has been modelled: office polypectomy before IVF or ICSI came out cost-saving in a 2017 decision model, on the assumption that removal raises pregnancy rates by the amount the IUI trial suggested.

The timing question - found before stimulation, found during it, found after a failed transfer - is a different decision with different answers, and is covered separately in IVF with endometrial polyps: does it need removing first?.

It might. The AAGL practice guideline reports that up to a quarter of endometrial polyps regress without any treatment, and that this is more likely for small ones - under about 10 mm.

This is the fact most often left out, and it matters for two reasons. It is the reason a small, symptomless polyp found incidentally can reasonably be watched with a repeat scan rather than operated on. And it is the reason that a polyp which disappeared after some intervention does not prove the intervention did anything.

This is usually the real question under the surface, and it has a reasonably precise answer. Pooled across 51 studies of 35,345 women who had polyps removed, the prevalence of cancer in the polyp was 2.73%. By menopausal status it was 1.12% premenopausal versus 4.93% postmenopausal; by symptoms, 1.89% in women without abnormal bleeding versus 5.14% in women with it.

For most women reading this - premenopausal, trying to conceive, polyp found on a fertility scan rather than because of bleeding - that puts the risk at roughly one in a hundred or lower. Low, and not zero. The only way to know which one it is, is histology, which requires the polyp to be removed. That is a legitimate reason to remove one on its own, separate from any fertility argument, and it is worth separating the two in your head when you are weighing the procedure.

Postmenopausal bleeding is a different situation and is not what this page is about; the AAGL guideline is explicit that symptomatic postmenopausal polyps should be excised for histological assessment.

A transvaginal ultrasound raises the suspicion. Hysteroscopy - a thin telescope passed through the cervix - confirms it, and is also how it is removed, usually in the same sitting. That distinction is not pedantry: in the one randomised trial, every woman in both arms had a hysteroscopy, and the only difference between the groups was whether the polyp came out. A shadow on a scan is a suspicion, not a diagnosis.

Two points from the AAGL guideline about the procedure itself. Outcomes do not differ between the methods of hysteroscopic removal, so there is no need to chase a particular technique. And blind removal - curettage without seeing what you are doing - is not indicated where equipment for guided removal exists, because it misses polyps. If you are offered a D&C for a polyp, that is a fair thing to question.

What a hysteroscopy involves as part of a fertility assessment, and when it is worth doing at all, is covered in hysteroscopy in the infertility workup.

No, and this page is not an argument for refusing it. Hysteroscopic polypectomy is a short day-case procedure whose surgical risks the AAGL guideline describes as low, it gives a histological answer nothing else can give, and the one randomised trial in the field found a real benefit before IUI. Declining on the basis of an article is not a good trade.

What changes with better information is the conversation. Worth asking: how big is it and was it seen on more than one scan; is removal being advised for fertility, for bleeding, or for histology, because those are three different reasons; what happens to my treatment timeline if I have it; and if it is small and causing no symptoms, is a repeat scan in a few months a reasonable alternative.

It is worth naming that being told about a polyp usually lands as one more thing wrong, at a point when the list already feels long. In practice this is one of the more tractable findings on that list - it is visible, it is removable, and it is dealt with in a morning.

Set out plainly:

• That polypectomy increases live births. No randomised trial retrieved for this article reported live birth for polyp removal in subfertility, before IUI or before IVF.
• That polypectomy helps before IVF or ICSI. Cochrane found no randomised studies of any cavity abnormality before IVF or ICSI, and the practice is based on reasoning, not trial data.
• That polypectomy helps in unexplained subfertility without planned treatment. Cochrane found no randomised studies in women with endometrial polyps for that comparison either.
• That polyps block the tubes or the cervix. No retrieved source describes either mechanism for an endometrial polyp; this claim circulates widely and was in the previous version of this page.
• That polyps can be prevented, shrunk with medication, or influenced by diet, weight or stress. No source retrieved for this article tested any of these. Increasing age is the strongest risk factor the AAGL guideline names, and it is not modifiable.
• That a polyp found on ultrasound is definitely a polyp. Both arms of the randomised trial had hysteroscopy precisely because ultrasound suspicion needs confirming.

Sometimes the answer is to remove it because it might be something, and sometimes it is to watch a 6 mm finding for three months while the rest of the assessment is finished. Both are defensible, which is the actual state of the evidence.