Endometriosis and Infertility: What the Evidence Supports

No. Plenty of women with endometriosis conceive without any treatment, and the 2022 ESHRE guideline reserves assisted reproduction for specific situations — where tubal function is compromised, where there is a male factor, or where the post-surgical prognosis score is low. It is a condition that makes conceiving harder on average, not a condition that removes the possibility.

The same guideline is blunt about the opposite error: treating endometriosis harder does not reliably make you more fertile. Hormone treatments that control pain do not improve fertility, and prevent pregnancy while you take them. Surgery helps some women and costs others ovarian reserve.

The honest starting position: endometriosis is common, it is under-diagnosed, and the single most useful thing you can leave a consultation with is a number for your own prognosis rather than a stage for your disease.

Three mechanisms are generally accepted. They do not all apply to every woman:

• Distorted anatomy. Adhesions — bands of scar tissue — can tether the ovary, tube or bowel so that the tube cannot collect an egg. This is why ESHRE singles out compromised tubal function as the situation where assisted reproduction is specifically indicated.
• Endometriomas and ovarian reserve. An endometriosis cyst on the ovary, and surgery to remove it, both reduce the pool of eggs that ovary can produce — ESHRE states that removing an endometrioma before IVF is likely to have a negative impact on ovarian reserve.
• Pelvic inflammation. The lesions provoke an inflammatory response. This is the mechanism most often invoked and least well quantified; see the section on what the evidence does not establish.

None of it is predicted by how much pain you are in. Severe pain occurs with minimal disease, and severe disease is found in women with no pain at all.

Sometimes, and less certainly than it is usually presented. The Cochrane review of laparoscopic surgery for endometriosis, updated in 2020, pooled 14 randomised trials in 1,563 women. Comparing laparoscopic treatment with diagnostic laparoscopy alone, it found moderate-quality evidence that treatment raises the rate of a viable intrauterine pregnancy — odds ratio 1.89, 95% CI 1.25 to 2.86, from 3 trials and 528 participants.

The sentence that usually gets left out: the same review states that no trial in any of its comparisons measured live birth. So the evidence supports a better chance of a pregnancy on a scan. It does not yet tell you what happens to the chance of taking a baby home, because nobody has measured it.

ESHRE's position follows from that. Operative laparoscopy “could be offered” for endometriosis-associated infertility in rASRM stage I/II disease — a weak recommendation. Before IVF the guideline runs the other way, with strong recommendations against routine surgery in stage I/II disease and against routine surgery on an endometrioma, because neither has been shown to improve live birth and the second is likely to reduce ovarian reserve. That decision has its own page: endometriosis and IVF — should you have surgery first.

ESHRE also says the decision should be driven by pain symptoms, your age, previous surgery, other infertility factors, ovarian reserve and your own preference — not by the stage alone.

The staging system most patients are given — rASRM stage I to IV — was built to describe what a surgeon sees, not to predict pregnancy. The Endometriosis Fertility Index was built for the second job. It scores 0 to 10 from your age, how long you have been trying, whether you have been pregnant before, and above all the “least function score”: how well the tube, fimbria and ovary work on each side once the surgeon has finished.

In the external validation of 233 women attempting to conceive without assisted reproduction straight after endometriosis surgery, cumulative pregnancy at 12 months was 45.5% overall. Split by score, cumulative pregnancy ran from 16.7% at EFI 0-3 to 62.6% at EFI 9-10, and in that cohort each extra point raised the chance of conceiving by 31% (hazard ratio 1.31).

In that analysis the least function score — post-surgical tube and ovary function — was the strongest contributor, ahead of the rASRM lesion score and the rASRM total. Which is the practical argument for asking about the EFI: it sorts women by prognosis in a way the stage does not. ESHRE's guideline says the EFI should be used to identify who may benefit from assisted reproduction after surgery, taking the partner's semen analysis into account as well.

Its limits are real and the authors state them: the score's performance was moderate, with an R² of 13% and a concordance index of 0.629. It separates groups. It does not tell an individual what will happen.

No, and ESHRE says so at its strongest level of confidence. Three recommendations are relevant if you are trying to get pregnant:

• In infertile women with endometriosis, clinicians should not prescribe ovarian suppression treatment to improve fertility. Strong recommendation.
• Women seeking pregnancy should not be prescribed post-operative hormone suppression with the sole purpose of enhancing future pregnancy rates. Strong recommendation.
• Clinicians should not prescribe letrozole outside ovulation induction to improve natural pregnancy rates. Strong recommendation.

This is not an argument against hormonal treatment. Suppression is a reasonable, evidence-backed way to control endometriosis pain, and ESHRE recommends it for that. It is an argument against being told that six months on a GnRH agonist will make you more fertile afterwards. It will not, and you cannot conceive during it.

Less than you have probably been told, once other diagnoses are separated out. The largest analysis available covers 347,185 IVF cycles using the woman's own eggs in the American SART registry between 2008 and 2010. Endometriosis appeared in 11% of 347,185 cycles, and in 64% of those endometriosis cycles a second diagnosis was recorded as well: male factor in 42% of the endometriosis cycles (n = 347,185 overall), tubal factor in 29%, diminished ovarian reserve in 22%.

The result that matters for counselling is the split. Women with isolated endometriosis had similar or higher live birth rates than the other diagnostic groups, including unexplained infertility and tubal factor. Women who had endometriosis plus another diagnosis had lower implantation and lower live birth rates. Endometriosis was associated with a lower number of eggs collected in both groups.

Worth flagging, because it is how this gets misquoted: the paper's own summary conclusion states that endometriosis is associated with lower oocyte yield, implantation and pregnancy rates after IVF. Its results are narrower than that, and the authors themselves note the association is confounded by the other diagnoses. Fewer eggs with an unchanged chance per transfer can mean more cycles for the same outcome — a different problem from IVF working less well on you.

Not necessarily, and this has changed. ESHRE's 2022 guideline recommends imaging — ultrasound or MRI — in the work-up, with a warning attached: a negative scan does not exclude endometriosis, particularly superficial peritoneal disease. Laparoscopy is for where imaging is negative or empirical treatment has not worked. It is no longer the mandatory first step it used to be.

One thing the guideline rules out. Clinicians should not use biomarkers measured in blood, endometrial tissue, menstrual or uterine fluid to diagnose endometriosis — a strong recommendation, and it covers CA-125. If a blood test has been offered to you as a way to confirm or exclude endometriosis, that is the line to ask about.

WHO puts the average time to diagnosis at 4 to 12 years, which is why the condition is worth naming out loud in a fertility consultation rather than waiting to be asked about it.

On conception specifically, no. ESHRE's 2022 conclusion is that there is no clear evidence any non-medical intervention will benefit the chance of pregnancy in women with endometriosis, and no recommendation can be made for nutrition, Chinese medicine, electrotherapy, acupuncture, physiotherapy, exercise or psychological interventions to increase fertility. The potential benefits and harms are unclear.

The guideline does recommend discussing non-medical strategies for quality of life and psychological well-being. So an anti-inflammatory diet, turmeric, yoga or acupuncture may be worth your time for how you feel. Presented as a way to get pregnant, they are being sold beyond the evidence, and this page would rather say so than fill a section.

Endometriosis care in India costs money before IVF is even discussed. In an ICMR study across five tertiary centres, endometriosis had the highest annual out-of-pocket spend of any infertility diagnosis — a median of ₹15,084 (IQR ₹8,114-₹20,758) for diagnosis, management and IUI. A single IVF cycle in the same programme averaged ₹1,10,104 in public hospitals and ₹2,30,668 in private ones. So make the first consultation count. Things worth asking, in roughly this order:

1. If I have had surgery: what is my Endometriosis Fertility Index, and what was the least function score on each side?
2. Are my tubes working? That is the finding ESHRE ties directly to needing assisted reproduction.
3. If surgery is being proposed: is it for pain or for fertility, and what happens to my ovarian reserve?
4. Has my partner had a semen analysis? ESHRE says the EFI should be read alongside it, and in the SART registry (n = 347,185 cycles) male factor was present in 42% of the endometriosis cycles.
5. What is my realistic chance per attempt, with a denominator — per cycle or per transfer, at my age?

Related on this site: common causes of female infertility, understanding thin endometrium, unexplained infertility and ovarian hyperstimulation syndrome.

The gaps here are large enough that anyone claiming certainty is overselling:

• No trial has measured live birth after surgery. Cochrane 2020 searched to April 2020 across 14 randomised trials and found no study reporting live birth for any comparison. The pregnancy benefit is real; the birth benefit is unmeasured.
• The EFI predicts groups, not people. In external validation in a cohort of 233 women its R² was 13% and its concordance index 0.629 — better than nothing, a long way from a forecast.
• Stage is not prognosis. In the same validation, the rASRM lesion and total scores were weaker contributors than post-surgical tube and ovary function. A stage IV label does not by itself tell you your chance of conceiving.
• The IVF comparison is observational. The SART finding that isolated endometriosis does as well or better comes from registry data across 347,185 cycles, not a randomised comparison, and the authors say the association is confounded by co-existing diagnoses.
• How much the inflammation itself matters is unquantified. It is a plausible and widely cited mechanism. This page has not found a figure for its independent effect on conception, so it does not give one.
• There is no cure. WHO states that treatment aims to control symptoms and limit long-term impacts. Nothing on this page is a cure, and anything sold to you as one is not.
• No Indian age-stratified outcome data. The IVF figures here are from a US registry. The cost figures are Indian; the outcome figures are not, because no comparable published Indian series was found.