Endometriosis lowers IVF outcomes, and how much depends on stage: stage III-IV is associated with roughly 20% lower implantation and clinical pregnancy rates (RR 0.79 for both), while stage I-II costs only a small fertilisation penalty (RR 0.93). An endometrioma alone does not measurably reduce live birth — protocol, surgical timing and ovarian reserve decide more than the diagnosis label.
“Clinicians are not recommended to routinely perform surgery for endometrioma prior to ART to improve live birth rates as current evidence shows no benefit and surgery likely negatively impacts ovarian reserve.” — ESHRE guideline: endometriosis, recommendation T17, Human Reproduction Open, 2022
This guide synthesises the 2022 ESHRE guideline, the ASRM committee opinion and the meta-analyses behind those numbers — including where evidence runs against common clinic practice. If endometriosis is your first fertility diagnosis, read what actually happens in an IVF cycle first.
How Does Endometriosis Affect Fertility and IVF Success?
ESHRE’s 2022 guideline puts prevalence at 2-10% of women in the general population, and IVF cohorts report it in 25-40% of women investigated for infertility. It damages fertility through four mechanisms, not one — Zondervan et al., Nat Rev Dis Primers 2018.
The four mechanisms
| Mechanism | Effect | Evidence strength |
|---|---|---|
| Distorted pelvic anatomy | Tubal adhesions, impaired oocyte pickup, reduced ovarian mobility | ★★★★★ |
| Diminished ovarian reserve | Endometriomas destroy normal cortical tissue; surgery compounds the damage | ★★★★★ |
| Inflammatory peritoneal environment | Peritoneal cytokines (IL-6, TNF-α, VEGF) impair oocyte and embryo quality | ★★★★ |
| Reduced endometrial receptivity | Abnormal endometrial gene-expression profiles affect implantation | ★★★ |
Sources: ESHRE 2022; Sanchez et al., J Ovarian Res 2017. Star ratings are our editorial reading.
The fixes differ by mechanism: surgery addresses anatomy and pain, down-regulation addresses inflammation. The habits marketed as receptivity fixes — freeze-all transfer and programmed preparation — are the weakest-evidenced part of the standard script.
How much does stage change your IVF outcome?
| Finding (vs no endometriosis) | Pooled effect | 95% CI | Data source |
|---|---|---|---|
| Stage I-II: fertilisation rate | RR 0.93 | 0.87-0.99 | Harb et al., BJOG, 2013 |
| Stage III-IV: implantation rate | RR 0.79 | 0.67-0.93 | Harb et al., BJOG, 2013 |
| Stage III-IV: clinical pregnancy rate | RR 0.79 | 0.69-0.91 | Harb et al., BJOG, 2013 |
| Endometrioma: live birth rate | OR 0.98 | 0.71-1.36 | Hamdan et al., Hum Reprod Update, 2015 |
| Endometrioma: cycle cancellation | OR 2.83 | 1.32-6.06 | Hamdan et al., Hum Reprod Update, 2015 |
Harb pooled 27 studies and 8,984 women; Hamdan pooled 33 and found removing an endometrioma beforehand changed nothing (OR 0.90, 95% CI 0.63-1.28).
Read plainly: advanced disease costs about a fifth of your implantation odds, mild disease costs very little, and an endometrioma costs you cancelled cycles rather than births.
Which IVF steps does endometriosis damage most?
Endometriosis works mainly through egg number and quality, not through the transfer itself.
- Fewer eggs retrieved — endometrioma patients show a small but consistent yield loss (SMD -0.23, 95% CI -0.37 to -0.10; 5 studies, 941 cycles)
- More cancelled cycles — the same meta-analysis put cancellation odds 2.83× higher (95% CI 1.32-6.06; 491 women)
- Lower fertilisation in mild disease, lower implantation in severe disease — the stage effects in the table above
Sources: Hamdan 2015; Harb 2013; mechanisms reviewed in Sanchez 2017.
Per-cycle numbers by age: IVF success rate data guide.
Which IVF Protocol Works Best for Endometriosis?
Observational cohorts favour the GnRH agonist long protocol — one 798-patient study reports live birth in 60.1% versus 40.0% for the antagonist — but randomised evidence does not confirm it, so reserve and stage should drive the choice, not protocol branding.
Protocol options compared
| Protocol | Mechanism | Suitability for endometriosis | Key point |
|---|---|---|---|
| GnRH agonist long protocol | Down-regulation before stimulation | ★★★★ | Best observational data; no randomised confirmation |
| Ultra-long protocol | 3-6 months of GnRH agonist pretreatment | ★★ | ESHRE 2022 recommends against it before ART (rec. T16): benefit uncertain |
| Antagonist protocol | No down-regulation | ★★★ | Shorter, dose-adjustable, no down-regulation |
| Mild stimulation | Low-dose gonadotropins | ★★★ | For severely diminished reserve |
| Natural cycle | No stimulation drugs | ★★ | Very low yield; not a primary strategy |
Sources: ESHRE 2022; ASRM Practice Committee, Fertil Steril 2012. Star ratings are our editorial reading.
What does the head-to-head evidence show?
Three comparisons exist in dedicated endometriosis populations, and they disagree with each other.
| Study | Design | Patients | Agonist result | Antagonist result | Source |
|---|---|---|---|---|---|
| Wang 2023, long-acting agonist | Retrospective cohort | 798 | Live birth 60.1% | 40.0% (P = 0.032) | IJGO |
| Drakopoulos 2018, rAFS I-II | Retrospective cohort | 386 | 42.8% across beta-hCG-positive, clinical pregnancy and live birth | 26.7% (P = .07, trend only) | Gynecol Endocrinol |
| Drakopoulos 2018, rAFS III-IV | Retrospective cohort | subset of 386 | No significant difference in any pregnancy outcome | same |
Wang also reported higher implantation with the long-acting protocol (51.0% versus 44.6% and 38.7%) but no live-birth advantage over the short-acting agonist (60.1% versus 53.8%, P = 0.443).
Neither study randomised, and no trial has stratified endometriosis live birth by AMH — any AMH cut-off a clinic quotes for protocol selection is convention, not evidence. Protocol mechanics are in our IVF stimulation protocols guide.
When is the ultra-long protocol worth it?
Current guidance is negative: ESHRE 2022 recommendation T16 (strong) states that extended GnRH agonist administration before ART is “not recommended, as the benefit is uncertain”, ESHRE.
The practice came from a 2006 Cochrane review: in 3 RCTs (165 women), 3-6 months of agonist pretreatment raised clinical pregnancy about fourfold (OR 4.28, 95% CI 2.00-9.15), with the live-birth signal resting on one trial of viable pregnancy (OR 9.19, 95% CI 1.08-78.22) — Sallam et al., Cochrane 2006. Its 2019 update reproduced neither: in 8 RCTs and 640 participants clinical pregnancy was RR 1.13 (95% CI 0.91-1.41), the sole live-birth trial favoured no pretreatment (RR 0.48, 95% CI 0.26-0.87), all graded very low to low — Georgiou et al., Cochrane 2019.
Some clinics still pretreat severe or recurrent disease on mechanistic grounds. Weigh that against three months of added treatment on evidence ESHRE rates as insufficient.
When does the antagonist protocol make more sense?
Where reserve has already narrowed, the antagonist protocol wins on operational grounds: stimulation length and gonadotropin dose are adjustable, and there is no down-regulation lead-in.
No endometriosis-specific live-birth trial supports that preference: in Drakopoulos’ 386-patient series the agonist edge appeared only in stage I-II, as a non-significant trend, and no predictor of live birth survived adjustment. The long protocol has better observational support in mild disease and no proven advantage in severe disease.
Should You Have Endometrioma Surgery Before IVF?
Surgery reliably helps pain, and it can cost you eggs: pooled across 237 patients, AMH fell by a weighted mean of 1.13 ng/mL after cystectomy (95% CI -1.88 to -0.37). The decision is a reserve question before it is a cyst question.
When is surgery before IVF the right call?
| Clinical scenario | Position | Basis |
|---|---|---|
| Pain that limits daily life (dysmenorrhoea, dyspareunia) | Operate on its own merits, not for IVF | ESHRE 2022 rec. T17: guided “mainly by pain symptoms and patient preference” |
| Endometrioma before ART, no pain | Do not operate routinely | ESHRE 2022 rec. T17: “no benefit and surgery likely negatively impacts ovarian reserve” |
| Stage I-II disease with IVF planned | Surgery is not offered to improve live birth | ESHRE 2022 rec. T17 (strong) |
| Large, suspicious or infected cyst | Individualised: reserve first, then surgery | Clinical judgement; no guideline sets an ART size threshold |
| Confirmed endometriosis with no pain | Proceed directly to IVF | Rec. T17; pooled surgery shows no IVF/ICSI change (OR 0.90, 95% CI 0.63-1.28) |
Sources: ESHRE 2022; Hamdan et al., Hum Reprod Update 2015
What exactly does cystectomy cost you?
The measurable cost is ovarian reserve — and the two standard markers disagree.
- AMH falls: Raffi et al., J Clin Endocrinol Metab 2012 pooled 8 prospective studies and 237 patients — weighted mean difference -1.13 ng/mL (95% CI -1.88 to -0.37), widening to -1.52 ng/mL in women starting at AMH 3.1 or above
- AFC largely holds: Muzii et al., Human Reproduction 2014 pooled 13 studies and 597 patients — no significant AFC change after surgery (mean difference 0.10, 95% CI -1.45 to 1.65), though the operated ovary ended 1.40 follicles below its untreated partner (95% CI -2.27 to -0.52)
Muzii’s own conclusion: cystectomy “may be considered safer for the ovarian reserve than previously thought” — the older claim of a 30-50% loss does not survive the primary data. What surgery buys is histology, clearer retrieval access and lower infection risk.
How do you decide in three steps?
| Step | Condition | Decision |
|---|---|---|
| 1. Ovarian reserve | AMH above 1.5 ng/mL and AFC above 8 | Surgical risk manageable — go to step 2 |
| 1. Ovarian reserve | AMH at or below 1.0 ng/mL and AFC at or below 5 | Prioritise IVF, avoid surgery |
| 2. Cyst characteristics | 5 cm or larger, suspicious features, or unmanageable pain | Prioritise surgery |
| 2. Cyst characteristics | Under 4 cm and painless | Proceed directly to IVF |
| 3. IVF history | 0-1 failed cycles | Proceed directly to IVF |
| 3. IVF history | 2 or more failed cycles | Surgery may be considered |
These cut-offs are clinic convention, not guideline recommendations; a specialist must weigh age, AMH trend, cyst morphology and symptom burden.
How Should an Endometriosis IVF Cycle Be Managed?
Ovarian protection runs through every stage, from surgical technique to transfer policy, and pain management is individual.
Pain during stimulation has no pooled incidence in these meta-analyses; extended agonist pretreatment is not recommended (rec. T16).
What protects the ovary?
- At surgery: do not assume ablation spares the ovary — the Muzii multicentre RCT, Human Reproduction 2016 randomised stripping against partial excision plus bipolar coagulation in 51 self-controlled women: no significant AFC difference, but significantly lower ovarian volume on the coagulated side at six months; stripping remains the authors’ gold standard
- Before surgery: assess reserve and discuss banking — ESHRE recommendation T18 asks clinicians to cover fertility preservation in extensive ovarian endometriosis, noting its true benefit remains unknown
- Before stimulation: measure AMH and AFC, then choose the least aggressive protocol that meets the goal
Which transfer strategy works best?
| Strategy | Position for endometriosis | Measured basis |
|---|---|---|
| Freeze-all (elective deferred transfer) | No proven advantage | Tan et al., J Clin Med 2021, 728 women: implantation 49.7% fresh versus 49.9% deferred (P = 0.73), clinical pregnancy 40.9% versus 39.9%, miscarriage 9.4% versus 9.9% |
| Programmed (“artificial”) cycle FET preparation | Discouraged in stage III-IV and deep disease | Busnelli et al., Hum Reprod Update 2024: artificial-cycle FET carries pre-eclampsia OR 2.14 (95% CI 1.91-2.39), preterm birth OR 1.24 (1.15-1.34), caesarean OR 1.59 (1.49-1.70), post-partum haemorrhage OR 2.43 (2.11-2.81) |
| Natural or modified-natural FET cycle | Preferred where ovulation allows | Geysenbergh et al., RCT, Hum Reprod 2026, 561 ovulatory women: clinical pregnancy with heartbeat 33.1% versus 28.9% (RR 1.15, 95% CI 0.89-1.47), fewer cancellations (3.9% versus 9.4%) |
| PGT-A screening | Not indicated by the diagnosis alone | No trial has tested PGT-A for endometriosis live birth; the aneuploidy signal in endometriosis cohorts is observational |
“We strongly discourage the use of AC regimens for FET in ovulatory women with rASRM stage III-IV endometriosis or DE. We also recommend single ET at the blastocyst stage in this high-risk population.” — Busnelli et al., Human Reproduction Update, 2024
Two of those three habits point away from the data: freezing every embryo buys no measurable advantage here, and a programmed cycle adds obstetric risk. Defer transfer only for a reason: hyperstimulation, rising progesterone on trigger day, or blastocyst banking.
The logistics of a scheduled frozen transfer are in our embryo transfer aftercare guide.
Where Can You Get Endometriosis IVF Abroad?
Endometriosis care varies by national practice pattern more than by technology, and costs differ widely.
| Country | Endometriosis care profile | IVF cost reference per cycle | Advantage |
|---|---|---|---|
| United States | Mature subspecialty care; integrated surgery-plus-IVF centres | $15,000-$30,000 | Published outcome data, flexible protocols |
| Thailand | Extensive Asian experience; low agonist drug costs | $6,000-$12,000 | Value; stage III-IV volume |
| Japan | Deep experience with mild stimulation for low reserve | ¥500,000-¥1,000,000 | Microstimulation; partial insurance |
| Mainland China | Leading surgical technique; stage III-IV patients often still go abroad | ¥30,000-¥80,000 | Low surgical cost, sequenceable with overseas IVF |
Costs from publicly quoted 2025-2026 clinic fee schedules; protocols move these figures.
US outcomes are auditable through the CDC ART success rates reports and SART national summary — the main argument for the American premium. Compare accredited centres in the United States, Thailand, or the IVF hospital directory.
FAQ
Q: How much lower are IVF success rates with endometriosis?
It is stage-dependent. Pooled across 27 studies and 8,984 IVF patients, stage I-II costs a small fertilisation penalty (RR 0.93, 95% CI 0.87-0.99); stage III-IV lowers implantation and clinical pregnancy by about a fifth (RR 0.79 for both; Harb et al., BJOG 2013).
An endometrioma alone shows no live-birth difference (OR 0.98, 95% CI 0.71-1.36) — age, baseline AMH and cancellation risk predict more than the stage label.
Q: Should my endometrioma be removed before IVF?
Not to improve IVF. ESHRE recommendation T17 is explicit that routine endometrioma surgery before ART is not recommended: no benefit, and likely harm to reserve.
Pooled outcomes agree (live birth OR 0.90, 95% CI 0.63-1.28). Operate for pain, suspected malignancy, or a cyst that blocks retrieval — and discuss banking eggs or embryos first (recommendation T18).
Q: Is the long protocol or the antagonist protocol better for endometriosis?
Observational cohorts favour the agonist: in 798 patients the long-acting protocol reached live birth in 60.1% versus 40.0% on the antagonist (P = 0.032) and 53.8% on the short-acting agonist — Wang et al., IJGO 2023. No randomised trial has settled it; in 386 patients the agonist edge appeared only in stage I-II (42.8% versus 26.7%, P = .07) and vanished in stage III-IV — Drakopoulos et al. 2018.
Q: Is pregnancy higher risk after IVF for endometriosis?
Yes, and more clearly than the fertility numbers suggest. Busnelli et al., Human Reproduction Update, 2024 pooled 34 studies on placenta previa (OR 2.84, 95% CI 2.47-3.26), 43 on preterm birth (OR 1.43, 95% CI 1.32-1.56) and 29 on caesarean section (OR 1.52, 95% CI 1.41-1.63), strongest in stage III-IV and deep disease.
A Scottish cohort of 5,375 women with surgically confirmed endometriosis adds adjusted odds of 1.76 for miscarriage (95% CI 1.44-2.15), 2.70 for ectopic (95% CI 1.09-6.72) and 1.26 for preterm birth (95% CI 1.07-1.49) — Saraswat et al., BJOG 2017 — the strongest argument here for single embryo transfer.
Q: When should endometriosis patients consider egg freezing?
Before the ovary is operated on. ESHRE recommendation T18 asks clinicians to discuss oocyte or embryo banking in extensive ovarian endometriosis while stating that its true benefit remains unknown.
The measurable reason to act early is the AMH drop after cystectomy — a weighted mean fall of 1.13 ng/mL across 237 patients, larger in women who start with good reserve.
Q: Does PGT-A compensate for endometriosis-related embryo quality?
There is no trial of PGT-A in endometriosis, so screening embryos because of this diagnosis is an extrapolation. Cohorts do report observational oocyte and embryo-quality differences — Sanchez et al., J Ovarian Res 2017, which is why we grade PGT-A ★★ rather than recommending it.
What PGT-A does change is how confidently a clinic commits to single embryo transfer (see our IVF stimulation protocols guide).
The Bottom Line on Endometriosis and IVF
An endometriosis diagnosis is not an IVF failure sentence: staging, protocol selection matched to reserve, and disciplined cycle management keep live birth rates clinically meaningful.
Three priorities, in order:
- Measure ovarian reserve early — AMH plus AFC underwrites every later decision
- Put surgery behind a reserve conversation — cystectomy lowers AMH by a weighted mean of 1.13 ng/mL, leaves AFC largely unchanged, and has never been shown to raise IVF live birth
- Distrust protocol certainty — the long-acting agonist has the best observational support, no randomised confirmation, and ESHRE recommends against extended suppression
Centre volume matters as much as protocol; our IVF clinic selection guide lists seven criteria that separate marketing from evidence; you can also contact our advisory team for a protocol review.
References
- Becker CAM, Zondervan KT, et al. Hum Reprod Open, 2022;2022(2):hoac009. DOI
- Harb H, Gallos I, Chu J, et al. BJOG, 2013;120(11):1308-1320. DOI
- Hamdan M, Dunselman G, Li T, Cheong Y. Hum Reprod Update, 2015;21(6):809-825. DOI
- Raffi F, Metwally M, Amer S. J Clin Endocrinol Metab, 2012;97(9):3146-3154. DOI
- Muzii L, Di Tucci C, Di Feliciantonio M, et al. Hum Reprod, 2014;29(10):2190-2198. DOI
- Muzii L, Achilli C, Bergamini V, et al. Hum Reprod, 2016;31(2):339-344. DOI
- Sallam HN, Garcia-Velasco JA, Dias S, Raine-Fenning N. Cochrane Database Syst Rev, 2006;(1):CD004635.pub2. DOI
- Georgiou A, Melo P, Baker P, Konstantinidis D. Cochrane Database Syst Rev, 2019;(11):CD013240.pub2. DOI
- Wang X, Li W, Chen X, et al. Int J Gynaecol Obstet, 2023;162(1):325-332. DOI
- Drakopoulos P, Rosetti J, Pluchino N, et al. Gynecol Endocrinol, 2018;34(10):884-889. DOI
- Tan J, Cerrillo M, Cruz M, et al. J Clin Med, 2021;10(2):344. DOI
- Busnelli A, Di Simone N, Somigliana E. Hum Reprod Update, 2024;30(6):751-788. DOI
- Geysenbergh B, Debrock S, Lie Fong S, et al. Hum Reprod, 2026;41(8):1420-1429. DOI
- Saraswat L, Ayansina D, Cooper K, et al. BJOG, 2017;124(3):444-452. DOI
- Practice Committee of the American Society for Reproductive Medicine. Fertil Steril, 2012;98(3):591-598. DOI
- Sanchez AM, Vanni VS, Bartiromo L, et al. J Ovarian Res, 2017;10(1):43. DOI
- Zondervan KT, Becker CM, Koga K, et al. Nat Rev Dis Primers, 2018;4:9. DOI
- US Centers for Disease Control and Prevention. Assisted Reproductive Technology (ART) success rates. cdc.gov/art
- Society for Assisted Reproductive Technology. National summary report, 2022. SART CORS
About this article: researched and written by the ProIVF Medical Editorial Team following Google E-E-A-T health content guidelines and reviewed by the ProIVF Medical Advisory Board. Odds ratios, risk ratios and mean differences are population-level references from published meta-analyses and cohorts, not predictions of individual outcome — confirm your personal prognosis with a reproductive endocrinologist.
This article is for informational purposes only and is not medical advice.
Editorial standards: ProIVF About page. Last updated: 2026-09-20.
