Deep and bowel disease, surgical outcomes, imaging and staging, recurrence, medical treatment after surgery, adenomyosis and fertility or miscarriage.
This page is written for clinicians and referrers and is not advice for individual patients. If you are a patient, the plain-language pages are Endometriosis and Adenomyosis .
Each week the literature is searched for new papers on endometriosis and adenomyosis, screened to the few that meet the evidence bar, and each one is verified on PubMed before it is summarised. I read every entry and give one verdict: whether the paper changes what I tell patients or do in theatre, supports current practice, or leaves it unchanged. Papers appear here newest first, grouped by the week they were found, with the study design, the main result and its effect size, and the reason for the verdict. Abstracts are not reproduced; the summaries are my own.
Digest of 5 November 2026
Search window 30 October 2026 to 5 November 2026
At hysterectomy for endometriosis, removing both ovaries went with fewer further operations (about 1 in 20 against 1 in 5), but no fewer doctor visits or pain prescriptions
Population-based retrospective cohort of 990 people aged 19 to 50 with pathologically confirmed endometriosis who had a hysterectomy between 2001 and 2016.
Finding
Reoperation: 19.2% with the ovaries conserved, 13.2% after one ovary was removed (adjusted HR 0.56, 95% CI 0.36 to 0.88) and 4.9% after both were removed (adjusted HR 0.20, 0.12 to 0.35). Adhesions, recurrent endometriosis and endometrioma were the commonest reasons. Physician visits and opioid, pill and endometriosis medicine prescriptions at 3 to 12 months did not differ between ovarian conservation and removal of both ovaries. Only 67.2% of those in premature surgical menopause filled a prescription for HRT. In natural frequencies, about 1 in 5 women who kept their ovaries had a further operation, against about 1 in 20 who had both removed.
Supports current practicemoderate evidence
It puts a number on a trade-off the hysterectomy page already describes, without showing better pain outcomes, so it informs the conversation rather than changing advice.
What it means
For a woman choosing what happens to her ovaries, this puts a figure on what the hysterectomy page says only in words: the chance of another operation is lower if both ovaries go.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
Quality of life improved after endometriosis surgery whatever the stage or site of disease, and stayed improved at a year
Prospective multicentre cohort (NCT05624567) of consecutive women aged 18 to 50 having surgery for suspected endometriosis at certified centres in Austria, Germany and Switzerland, with histologically confirmed disease and a baseline EHP-30.
Finding
Median EHP-30 fell from 50.8 (IQR 32.4 to 64.2) before surgery to 34.4 at 6 months and 31.1 at 12 months (both p<0.001; lower is better), with no significant change between 6 and 12 months. The improvement was seen across all #ENZIAN compartments, including deep and multifocal disease; neither anatomical distribution nor rASRM stage was associated with baseline quality of life or recovery. Baseline symptom burden and surgeon-reported complete resection (p=0.011 at 6 months, not significant at 12 months) were the strongest predictors of early improvement.
Supports current practicemoderate evidence
It supports offering surgery on symptoms rather than stage and aiming for complete excision, as he does, without a control group to prove cause.
What it means
A woman with superficial disease may be told she can expect as much improvement as a woman with extensive disease, and the amount of disease he sees should not by itself predict how she will feel.
Conflicts of interest: No external funding; the authors declare no competing interests.
Reviewed by Mr Jan, .
Digest of 29 October 2026
Search window 23 to 29 October 2026
For infertility with an endometrioma or deep endometriosis, going straight to IVF gave about twice the live birth rate of surgery followed by trying naturally
Pragmatic partially randomised patient preference trial (NCT04743167) at three Italian tertiary centres expert in both endometriosis surgery and IVF, March 2021 to September 2024.
Finding
Conception leading to live birth within 12 months: in the randomised cohort 16 of 28 (57%) with IVF against 7 of 27 (26%) with surgery, RR 2.20 (95% CI 1.08 to 4.50); in the preference cohort 32 of 50 (64%) against 7 of 25 (28%), RR 2.29 (1.18 to 4.43); combined, RR 2.29 (1.41 to 3.70). Per-protocol and survival analyses agreed. In natural frequencies, about 6 in 10 women had a baby within a year with IVF, compared with about 1 in 4 after surgery.
Changes practicemoderate evidence
The first randomised evidence that, for infertility alone with ultrasound-diagnosed endometrioma or deep endometriosis, IVF first roughly doubles the live birth rate at a year, which should shape how surgery and IVF are presented to women whose main aim is pregnancy.
What it means
For a woman under 40 whose main aim is a baby, with a small endometrioma or deep endometriosis seen on a scan and no other reason to operate, this trial suggests IVF first gives roughly twice the chance of a live birth within a year.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
Leuprolide injections relieved endometriosis pain more than oral GnRH antagonists, at the cost of more bone loss
Systematic review and meta-analysis (PROSPERO CRD420261289010) of five randomised trials comparing oral GnRH antagonists with leuprolide-based GnRH agonists for endometriosis, pooling pain, pelvic signs, bone mineral density, quality of life and hormone outcomes.
Finding
Leuprolide gave greater improvement in dysmenorrhoea (SMD 0.69, 95% CI 0.48 to 0.91), non-menstrual pelvic pain (SMD 0.33, 0.16 to 0.50), overall pelvic pain (SMD 0.25, 0.13 to 0.38) and dyspareunia (SMD 0.14, 0.02 to 0.26), with a larger benefit after 12 weeks; antagonists caused less bone loss (SMD 0.64, 0.39 to 0.89) at 24 weeks. Most quality-of-life domains did not differ.
Supports current practicemoderate evidence
It supports offering either route with add-back and choosing by the woman's priorities, without showing one is clearly better.
What it means
Injections that switch off the ovaries may control pain a little better than antagonist tablets, but tablets are gentler on bone and wear off quickly.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
Draining an endometrioma with alcohol before IVF: possibly more pregnancies, no proven effect on live births, very low certainty
Systematic review and meta-analysis of six studies (one randomised trial, one prospective comparative study, four retrospective cohorts) to November 2025 comparing transvaginal ethanol sclerotherapy of ovarian endometriomas before IVF or ICSI with expectant management.
Finding
Live birth did not differ significantly (2 studies, 124 women; OR 2.20, 95% CI 0.66 to 7.32). Clinical pregnancy was higher with sclerotherapy (4 studies, 221 women; OR 2.14, 1.21 to 3.79), with more oocytes (MD 2.20, 0.06 to 4.34) and embryos (MD 0.56, 0.07 to 1.05); mature oocytes did not differ significantly (MD 1.56, -0.14 to 3.26). GRADE certainty was very low for every outcome.
No changeweak evidence
Very low certainty evidence with no significant effect on live birth.
What it means
Sclerotherapy may be an option to discuss for a recurrent endometrioma before IVF where repeat surgery would cost ovarian reserve, but there is no reliable evidence that it gives more babies.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
Focused ultrasound treatment of fibroids is followed by a small fall in AMH, with no change in antral follicle count
Systematic review and single-arm meta-analysis of ten studies (two randomised trials and eight case series, 667 women) to January 2026 measuring ovarian reserve before and 3 to 24 months after high-intensity focused ultrasound (HIFU) for fibroids or adenomyosis.
Finding
AMH fell slightly after HIFU (MD -0.12 ng/mL, 95% CI -0.20 to -0.04, I2 0%), and similarly in fibroid-only studies (MD -0.13, -0.22 to -0.05); antral follicle count did not change (2 studies, 140 women; MD -0.05, -0.28 to 0.19). There were too few data for an adenomyosis-specific estimate.
No changeweak evidence
A small, uncontrolled AMH fall that does not change advice on fibroid treatment.
What it means
If a patient asks about HIFU instead of myomectomy, she can be told the effect on egg reserve appears small, but the evidence cannot show it is safe for fertility.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
MRI after an expert ultrasound finds more bowel and uterosacral disease in complex deep endometriosis, at the cost of more false positives
Prospective diagnostic cohort of 221 women with suspected deep endometriosis scheduled for surgery after expert transvaginal ultrasound (IDEA protocol); 189 had surgery and 112 also had MRI, selectively for suspected advanced disease, inconclusive ultrasound or severe symptoms with a negative scan.
Finding
Adding MRI raised sensitivity for rectosigmoid disease by 10.3% (6 more true positives, 3 more false positives) and for uterosacral ligaments by 29.2% (14 more true, 12 more false), but lowered specificity for endometriomas (by 19.0%), pouch of Douglas obliteration (13.3%) and uterosacral ligaments (20.3%); accuracy for bladder and ureters fell from 82.2% to 75.7%. Most lesions missed on ultrasound were in the mid or upper rectum, and the gain for rectosigmoid disease was greatest after previous surgery (exploratory).
Supports current practicemoderate evidence
It supports MRI as a targeted second test for complex deep disease after an expert scan.
What it means
For a woman with suspected bowel disease, especially after previous surgery, MRI adds most for the mid and upper rectum, which is what the colorectal team needs to plan; it adds little, and can mislead, for cysts and the bladder.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
Digest of 22 October 2026
Search window 20 to 22 October 2026
Supplements for endometriosis: some small trials report less pain with curcumin or selenium, no higher pregnancy rates and no less recurrence after surgery
Systematic review of 24 clinical trials in 2,019 women with endometriosis testing dietary supplements (vitamins, antioxidants, curcumin, selenium and others), alone or with standard hormonal treatment.
Finding
Some trials reported improvements in pain and quality of life, particularly with specially formulated curcumin or selenium, alone or with dienogest. Fertility-related measures improved in some trials but pregnancy rates did not rise. Some combinations reduced ovarian endometriosis, but supplements did not reduce recurrence after surgery. No pooled effect sizes are given; the authors conclude supplements may support standard treatment but should not replace it.
Supports current practiceweak evidence
It supports telling women that supplements may sit alongside treatment but do not replace it and do not stop endometriosis coming back.
What it means
Women often ask whether supplements help.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
Women with endometriosis have more autoimmune, allergic and chronic pain conditions, but the studies are of low quality
Systematic review and meta-analysis (PROSPERO CRD420251258716) of 55 observational studies (cohort, case-control, cross-sectional) to December 2025, 48 pooled, on the association between endometriosis and autoimmune, allergic, chronic pain and endocrine conditions.
Finding
In cohort studies, women with endometriosis had a higher risk of lupus (RR 1.81, 95% CI 1.50-2.18), Sjögren's syndrome (RR 1.54, 1.40-1.71) and rheumatoid arthritis (RR 1.71, 1.40-2.09); case-control and cross-sectional studies showed higher odds for these and for multiple sclerosis (OR 2.30, 1.57-3.37) and coeliac disease (OR 4.50, 1.94-10.42), but not significantly for inflammatory bowel disease (OR 2.82, 0.58-13.61). Asthma (RR 1.27, 1.14-1.40), eczema (OR 1.58, 1.34-1.87), fibromyalgia (RR 1.65, 1.42-1.92), chronic fatigue syndrome (RR 1.41, 1.31-1.51) and irritable bowel syndrome (RR 2.51, 1.55-4.07) were also more common; thyroid disorders were not.
Supports current practiceweak evidence
It supports a whole-patient, multidisciplinary approach to pain without changing any treatment.
What it means
It confirms what women with endometriosis often report, that other conditions, especially irritable bowel syndrome, fibromyalgia and fatigue, travel with it, which supports asking about them and involving pain and bowel colleagues.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
An ultrasound angle measuring how far the womb is pulled back may grade endometriosis adhesions before surgery
Mixed prospective-retrospective case-control study comparing women with surgically staged endometriosis with benign and normal controls (1:1:1).
Finding
The angle rose with rASRM stage and adhesion severity (both P<0.001) and was driven mainly by the adhesion score (beta 0.441). An angle of 220 degrees or more defined the 'question mark' sign with an area under the curve of 0.958. Controls had a median of about 178 to 179 degrees.
No changeweak evidence
An interesting objective measure that needs validation in unselected women before it changes scanning or surgical planning.
What it means
It puts a number on a sign his sonographers already look for, which may help describe a fixed, retroverted womb on a scan report.
Conflicts of interest: Funding is not stated in the abstract.
Reviewed by Mr Jan, .
Digest of 19 October 2026
Search window 13 to 19 October 2026
One in six women needs further surgery after an operation for ovarian endometriosis, one in three after ovarian-conserving surgery
Finnish national register cohort of 13,912 women whose first surgically verified diagnosis of ovarian endometriosis came with an ovarian procedure between 1987 and 2012, followed through the registers to 2019 (median 20.7 years).
Finding
The 30-year cumulative risk of any reoperation was 17.3% (about one in six) and of ovarian reoperation 16.5%, with a median 3.4 years to the first and under two years to each later one. Compared with oophorectomy, enucleation or resection carried a higher risk of any reoperation (aHR 1.43, 95% CI 1.28-1.60) and other ovarian procedures higher still (aHR 2.20, 95% CI 1.67-2.89); the authors put this at roughly one in three reoperated after ovary-conserving surgery against one in ten after oophorectomy. Young age, a family history of endometriosis, a primary diagnosis of endometriosis and prior infertility each raised the risk, and hysterectomy before or at the index operation lowered it. Reoperated women had fewer births (61.6% vs 70.7%) and more infertility (51.5% vs 17.1%); reoperation was more common after index surgery in 1996-2012 than in 1987-1995 (19.0% vs 12.4%), and by the fourth or later operation 13.4% of procedures were for deep infiltrating endometriosis.
Supports current practicestrong evidence
Gives long-term national figures for counselling before endometrioma surgery and backs an individualised, fertility-conscious choice between keeping and removing the ovary rather than any change in technique.
What it means
For a woman deciding about surgery for an endometrioma, this gives a figure I can quote: about one in six women needs another operation over the following decades, sooner in younger women, those with a family history and those with infertility, and keeping the ovary roughly trebles that chance compared with removing it.
Conflicts of interest: Funding not stated in the abstract.
Reviewed by Mr Jan, .
Vasopressin during endometrioma cystectomy speeds haemostasis but does not protect ovarian reserve: randomised trial
Single-centre randomised controlled trial at a Korean university hospital, September 2018 to July 2021: 75 women having laparoscopic cystectomy for a unilateral endometrioma, stratified by baseline AMH and randomised to vasopressin injection or no injection; the primary outcome was the change in serum AMH at 12 months, analysed by intention to treat and per protocol.
Finding
Vasopressin shortened bleeding control (2.2 vs 6.5 minutes, p = 0.002) and cyst stripping (8.9 vs 14.4 minutes, p = 0.054), but the fall in AMH at 12 months was the same in both arms (-0.96 vs -1.14 ng/mL, p = 0.488), with the same pattern in women with baseline AMH below 3 ng/mL and 3 to 7 ng/mL. AMH fell significantly at three months and partly recovered at six and twelve months in both groups (group-by-time interaction p = 0.816), and the specimens contained the same thickness and proportion of ovarian tissue in each group.
No changemoderate evidence
Vasopressin shortens the operation but should not be presented as protecting the ovary; the trial is small and single-centre, and it leaves the egg-sparing technique neither proven nor disproven.
What it means
For the woman having an endometrioma removed, the loss of ovarian reserve after cystectomy (AMH down by about 1 ng/mL at a year, with partial recovery from three months) happened whether or not vasopressin was used to reduce bleeding and diathermy.
Conflicts of interest: Funding not stated in the abstract.
Reviewed by Mr Jan, .
Structured MRI reports for deep endometriosis are more complete and more useful to surgeons, but take longer: randomised reader study
Single-centre randomised reader study in France: 20 radiology residents were assigned to free-text or structured reporting, 15 completed the same 22 pelvic MRI examinations (330 reports: 154 free-text, 176 structured), judged against a reference standard set by two pelvic radiologists; in an exploratory paired assessment two expert gynaecological surgeons rated 44 reports, one of each format per examination.
Finding
Structured reports were more complete (+8.1 percentage points, 95% CI 4.1-12.1; clinically weighted +5.6, 95% CI 1.3-10.0) and were rated higher by the surgeons for completeness, usefulness for decision-making and adequacy to clinical needs, with similar clarity; they took 4.52 minutes longer (95% CI 3.05-5.99) and carried nearly three times as many explicit negative statements that contradicted the expert reference (rate ratio 2.94, 95% CI 1.73-4.98), mainly compartment-based 'no disease' lines.
Supports current practicemoderate evidence
Backs asking for compartment-based structured MRI reports before deep endometriosis surgery, with the caveat that an explicit 'no disease' statement for a compartment is not always right and the readers were trainees.
What it means
For the woman with suspected deep endometriosis, this supports asking for an MRI report that works through each compartment of the pelvis, as the specialist endometriosis radiologist in my multidisciplinary team does (EXPERTISE, deep endometriosis; endometriosis page, 'How I assess endometriosis'): structured reports were more complete and surgeons found them more useful.
Conflicts of interest: Funding not stated in the abstract.
Reviewed by Mr Jan, .
Estetrol/drospirenone for endometriosis pain: better than placebo, no better than other hormonal treatment, on 198 women
Systematic review to Cochrane and PRISMA standards (PubMed, Embase, CENTRAL; two independent reviewers, risk-of-bias assessment) of estetrol 15 mg with drospirenone 3 mg for endometriosis symptoms: two randomised trials and two observational cohorts, 198 women in all; no meta-analysis was possible.
Finding
The pill reduced pelvic pain compared with placebo, with no significant difference against other active hormonal regimens; it improved dysmenorrhoea and quality of life (sleep, global improvement, overall health status); dyspareunia results were inconsistent; intermenstrual bleeding or spotting was the commonest side effect. No effect sizes are given in the abstract and nothing could be pooled.
No changeweak evidence
Too little evidence to prefer this pill over the combined pills and progestogens already used; it is an option for women who want it.
What it means
For a woman choosing a hormonal treatment for endometriosis pain, this says the newer estetrol pill reduces pain compared with placebo, as combined pills generally do, with no evidence that it beats the pills and progestogens already used, and spotting as its commonest side effect.
Conflicts of interest: Funding not stated in the abstract; proprietary product, so the included trials' sponsorship should be checked in the full text.
Reviewed by Mr Jan, .
About 1 in 100 women are first diagnosed with endometriosis after the menopause, mostly after a surgical menopause; HRT after a natural menopause did not raise the risk: Nurses' Health Study II
Prospective analysis of the population-based Nurses' Health Study II in the United States: 80,148 women with no report of endometriosis before the menopause, followed from the first questionnaire reporting menopause (baseline 1989) to 2019 for a laparoscopically confirmed first diagnosis of endometriosis after the menopause, analysed by age at menopause, surgical or natural menopause and hormone use with multivariable Cox models.
Finding
883 women (1.1%, about 1 in 90) reported a first diagnosis of endometriosis after the menopause, at a median age of 50.6 years (48.5 after a surgical menopause, 54.7 after a natural one), a median 4.0 years after it; 69.9% of the diagnoses followed a surgical menopause and 69.5% were in current hormone users. Compared with a natural menopause without hormones (110 cases), hormone use after a natural menopause carried no excess (137 cases; HR 1.16, 95% CI 0.89-1.49), whereas a surgical menopause carried a higher risk both without hormones (78 cases; HR 5.79, 95% CI 4.29-7.81) and, to a lesser degree, with them (477 cases; HR 3.86, 95% CI 3.08-4.82). The authors read the cases as a mixture of unrecognised persistent or recurrent disease and new lesions.
Supports current practicemoderate evidence
Confirms that endometriosis can present after a surgical menopause, which is why combined rather than oestrogen-only HRT is advised where disease is left behind, and that HRT after a natural menopause does not bring it on; it does not test the choice of HRT regimen.
What it means
For a woman considering HRT after a natural menopause, this is reassuring: hormone use did not increase the chance of endometriosis being found.
Conflicts of interest: Funding not stated in the abstract.
Reviewed by Mr Jan, .
Every paper on this site's evidence pages is also available as open data at /evidence/papers.json. The search and screening rules are on the evidence index.
Last reviewed .Reviewed by Mr Haider Jan, Consultant Gynaecologist.
For patients: Endometriosis and Adenomyosis
The plain-language pages on what the condition is, how I assess it and the choices, written for the women I see.
