Deep infiltrating endometriosis (DIE) diagnosed by ultrasound in subfertile women - effect on Anti-Müllerian Hormone (AMH), antral follicle count (AFC) and fertility treatment outcome.

Project: Research


Endometriosis and adenomyosis
Endometriosis is a benign gynecological disease affecting up to 5-10% of all women of reproductive age and up to 30% of infertile women (Bulun 2009). The disease is characterized by the presence of endometrial tissue outside the uterine cavity, in various clinically distinct forms depending on it´s location; peritoneal endometriosis, ovarian endometriosis (endometrioma) and deep infiltrating endometriosis engaging bowel.
Adenomyosis is a related disease often coexisting with endometriosis, with endometrial glands and stroma invading the uterine myometrium and thus disrupting the endo-myometrial junction (the junctional zone, JZ). The condition affects approximately 20% of all women (Naftalin 2017), is more common among infertile women and in women with recurent miscarriages, and increases with age and parity. Symptoms of endometriosis is pelvic pain, dysmenorrhea, coital pain and blood during defecation or urination, while adenomyosis is frequently associated with dysmenorrhea, abnormal uterine bleeding and an enlarged and tender uterus. Approximately 30% of women with adenomyosis are believed to be asymptomatic.
The mechanism behind a possible association between adenomyosis, endometriosis and infertility is poorly understood. Studies have shown significantly lower pregnancy rates after ART in women with adenomyosis (23,6%) compared with no adenomyosis (44,6%) (Benaglia 2014, Costello 2011), and significantly higher pregnancy rates in women with endometriosis but without adenomyosis compared with those with concurrent endometriosis and adenomyosis after laparoscopic excision of DIE (Landi 2008). However, others have disputed a detrimental effect of adenomyosis on pregnancy rate in asymptomatic women (Benaglia). It has been suggested that patients with coexisting deep infiltrating endometriosis and adenomyosis may constitute a subgroup with a particularly poor reproductive prognosis (Eisenberg 2017, Vercellini 2014), but more studies on IVF/ICSI-outcome in relation to adenomyosis with and without concomitant endometriosis have been requested (Maubon 2010).
It has been observed that pelvic endometriosis, especially in severe stages, is strongly associated with thickening of the Junctional Zone, a feature also known to correlate with adenomyosis. Therefore, the JZ has been proposed to be involved in the process that determines pelvic endometriosis. Evaluation of the JZ and its alterations is increasingly important for the understanding of infertility in women with adenomyosis and endometriosis, but more studies are needed.
Anti-Müllerian Hormone and the ovarian reserve
Anti-Müllerian hormone (AMH), is a glycoprotein exclusively produced by the granulosa cells of pre-antral and small antral follicles in the ovaries, presumably independently of follicle-stimulating hormone (FSH) supporting the initial follicular development. AMH has emerged as a quantitative marker of ovarian reserve and response, and is thought to be equivalent to the Antral Follicle Count (AFC) in assessment of the ovarian reserve. However, it has been suggested that endometriosis negatively affects the ovarian reserve (Benaglia 2013, 2017), and that markers for the ovarian reserve might behave differently in women with and without endometriosis.
A possibly detrimental effect of endometrioma on the surrounding ovarian tissue and thus the ovarian reserve has been proposed, with free iron, reactive oxygen species and inflammatory molecules causing damage to primordial follicles, and present cysts causing stretch of the ovarian cortex with concomitant disruption of ovarian function. Studies have shown that serum-AMH levels are not significantly reduced in women with unoperated unilateral endometrioma (Benaglia 2017), and that a reduction in ovarian responsiveness could only be observed in women with larger (Ferrero 2017) or bilateral (Benaglia 2013) endometriomas. However, further analysis on the relation between AMH and the size of the endometrioma, in particular those > 5 cm in diameter, would be of interest.
Lower basal AMH-levels, as well as a lower oocyte retrieval, have been demonstrated among women with endometriosis compared to women of the same age without endometriosis (Yoo 2011). However, few studies have evaluated the use of AMH in combination with specific ultrasound criteria of endometriosis or adenomyosis for assessment of the ovarian reserve and prediction of outcome in women with endometriosis or adenomyosis undergoing IVF/ICSI-treatment.

Transvaginal Ultrasound
Traditionally, the diagnosis of endometriosis was made by laparoscopy with biopsies of suspected lesions, while adenomyosis was diagnosed by histopathology after hysterectomy. Over the last years, improved resolution of transvaginal ultrasonography (TVS), 3D-ultrasonography (3D-TVS) and magnetic resonance imaging (MRI) has made it possible to make a non-invasive but yet precise and reliable diagnosis of endometriosis and adenomyosis, without the need for histological examination. Several metaanalyses have found the diagnostic performance of TVS comparable to that of MRI, and TVS has been proposed as the first-line imaging technique for correct diagnosis because of it´s availability, low cost and dynamic nature. 3D ultrasonography allows a more detailed evaluation of the JZ, and seems to be more accurate than conventional 2D-TVS to detect adenomyosis (Exacoustos). Possibly, this new technique enables detection of features of adenomyoisis and endometriosis that have not been previously seen, and also allows detection of earlier steps of the disease.
However, different ultrasond criteria for the disease have been used in different studies, why comparison of results is somewhat difficult.
Studies report considerable differences in the prevalence of adenomyosis and endometriosis and its impact on IVF/ICSI outcome (Mavrelos 2017). Partly, this could be explained by differences in the diagnostic ultrasound criteria and methods used to detect adenomyosis and endometriosis, and that studies did not specifically examine the differential impact of each feature of the disorders.

Aims of the project
The overall objective of this study is to describe the ultrasound appearance of endometriosis and adenomyosis in women suffering from infertility, and to evaluate the association between specific ultrasound features of adenomyosis and endometriosis with the ovarian reserve and the outcome after IVF/ICSI treatment. We also wish to establish whether Anti-Müllerian Hormone (AMH) an antral follicular count (AFC) can predict of ART outcome in infertile women suffering from adenomyosis or endometriosis, to elucidate whether AMH levels in combination with specific ultrasound features of endometriosis and adenomyosis are predictive of live birth.
Preliminary, five papers are planned. The first manuscript, investigating the use of AMH for the prediction of live birth rate after ART, has been submitted.
In the other four studies, specifically we will focus on the use of ultrasound in combination with AMH for the prediction of outcome after IVF treatment in patients with
a) endometriosis with endometrioma only
b) deep infiltrating endometriosis (DIE) with or without endometrioma
c) deep infiltrating endometriosis + adenomyosis
d) adenomyosis with and without DIE or endometrioma
Primary outcome will be live birth rate, but we also wish to analyze oocyte quality, pregnancy and miscarriage rates and possible adverse pregnancy outcomes such as preeclampsia, IUGR, preterm birth, placental abruptio and obstetric bleeding. We also plan a fifth prospective study, were patients with a clinical signs of endometriosis, but with no signs of the disease on ultrasound examination, will be followed up after 5 years, to see if any ultrasonographic signs of the disease then appears.

Our hypothesis is that endometriosis affect the ovarian reserve negatively, and that this effect is related to the size and bilaterality of the endometrioma, while deep infiltrating endometriosis (DIE) on bowel or urinary bladder has a greater effect on the tubes; thus it might be important to differ ovarian only endometriosis from deep infiltrating endometriosis when counseling infertile couples. We also think that adenomyosis is associated with a lower implantation rate and increased pregnancy loss but does not affect the ovarian reserve per se. We believe that by using modern ultrasound technology, it will be possible to detect early features of adenomyosis and endometriosis and to establish which features are related to infertility as well as an adverse pregnancy outcome after ART. Hopefully, this could enable to give a relevant counselling about the prognosis of ART, identification and selection of appropriate treatment for patients with affected reproductive health.
An application will be made to the Ethics committee at Lund University for approval before study start.

Study subjects
Patients who are admitted to the Reproductive Medical Centre, Malmö, Skåne University Hopsital, for IVF or ICSI treatment, and with a history of dysmenorrhea, chronic pelvic pain, coital pain or blood in urine or during defecation will be asked to participate. Patients will be referred to the Ultrasound Department, Kvinnokliniken, Malmö, SUS, for ultrasound examination by either one of three operators (Sara Alson, Ligita Jokubkiene and Povilas Sladkevicius). Patients with any sign of endometriosis; endometrioma, adenomyosis or deep infiltrating endometriosis according to specific criteria, will be prospectively included in the study following written consent. Patients without any signs of endometriosis will be used as control group. All patients will have blood drawn for AMH-analysis at RMC as part of routine work-up.
Another group of the patients that will be included in the study are the patients referred for ultrasound diagnostics from the Endometriosis Centrum.
Additional information will be collected from patient journals (WinIVF, Melior and Obstetrix), and all data will be saved in an Excel file for statistical analysis.
Inclusion criteria:
- Subfertile women
- non-smoking women
- BMI <30 kg/m2,
- Age between 20-39 years
- both ovaries present.
- Known or suspected endometriosis or adenomyosis (history of dysmenorrea, coital pain, blood in urine or during defecation,verified by ultrasound or laparoscopy previously).

Exclusion criteria:
- infertility due to male factor
- only one ovary
- BMI ≥30kg/m2
- Uterine malformations and large fibroids (> 3 cm)

Blood test:
Serum AMH will be drawn at RMC before the start of IVF-treatment. Serum will be analyzed at the Department of Clinical Chemistry, Skåne University Hospital in Malmö, Sweden. Serum levels of AMH are measured using the Electro Chemi Luminiscence Immuno assay (ECLI) provided by Roche Elecsys AMH. The lowest detectable level is 0.07 pmol/l, and coefficients of variation are 2% at 6.86 pmol/l.
Ultrasound examination:
Patients with known or suspected endometriosis will be examined according to a structured protocol. Patients will undergo 2D, 3D and power Doppler TVS of the pelvic organs in a single examination before the start of IVF treatment. Each scan will be performed by the same operators (SA, LJ, PS), using an E8 or E10 (GE Healthcare, Zipf, Austria) ultrasound machine equipped with a multifrequency 3D volume endovaginal probe (2.8–10 MHz).
The 2D-TVS examination will include evaluation and measurement of the pelvic organs. The uterus, endometrium and adnexa will be measured in three planes and evaluated for any abnormalities. Antral Follicle Count (AFC, the sum of all follicles 2 to 10 mm in size) will be evaluated. Any lesions (myomas and signs of adenomyosis, ovarian cysts) will be described and measured.
Using 3D-TVS, a volume of the uterus will be acquired in order to obtain the coronal view. Two to four static gray-scale volumes of the uterus will be obtained from the sagittal plane and from the transverse plane. The volume acquisition technique will be standardized according to the following criteria: frequency, 6–9 MHz; magnification of the uterus up to half of the screen; sweep angle, 120◦; sweep velocity, adjusted from medium to maximum quality; 3D volume box exceeding the uterus by 1 cm on each side. The coronal view reconstruction technique involve placing a straight or curved line (rendering mode) along the endometrial stripe on the sagittal and transverse views (Panel A and B of the multiplanar view). The multiplanar view will then be manipulated until a satisfactory coronal image is obtained of the uterine external profile and the cavity, with visualization bilaterally of the interstitial portion of the Fallopian tube. Volume contrast imaging (VCI) will be applied (2–4 mm slice thickness) with volume rendering (mixed light surface and gradient light). On the coronal view the JZ appear as a hypoechoic zone around the endometrium. Using VCI modality with 2–4-mm slices the JZ can be viewed clearly in all planes of the multiplanar view.
Power Doppler:
Power Doppler will be performed using fixed preinstalled settings: frequency, 6–9 MHz (‘normal’); pulse repetition frequency, 0.6–0.3 kHz; gain, −4.0; wall motion filter, ‘low 1’ (40 Hz). If necessary, power Doppler gain will be reduced until all color artifacts disappear. This modality will be used to evaluate the vascularization of the myometrial tissue, to distinguish between a myometrial cyst and a vascular component, and between leiomyoma and focal adenomyosis. Localized adenomyosis and adenomyoma are characterized by the presence of rare, diffuse vessels, while fibroids have flow aligned along the external myoma capsule, appearing on imaging as a vascular ring.
Image processing and analysis:
Following acquisition, ultrasound digital and photographic images and volumes will be saved and stored in hospital PACS system for subsequent retrieval for offline analysis in ViewPoint.

Specific ultrasound features:
The presence of certain TVS features associated with adenomyosis, endometrioma and deep infiltrating endometriosis will be described:
1. Adenomyosis:
- Globally enlarged uterus
- Myometrial AP asymmetry unrelated to leiomyoma
- Ill-defined lesion or focal adenomyosis
- Fan-shaped shadowing (myometrial hypoechoic linear striations as a pattern of thin acoustic shadowing)
- Mixed echogenicity of myometrium (indistinctly defined myometrial area with decreased or increased echogenicity)
- Cysts (round, anechoic areas within the myometrium)
- Hyperechogenic islands, subendometrial lines and buds
- Vascularity: Translesional flow
- Junctional zone:
- Thickened: irregular or ill-defined, interrupted.
- JZ thickness: On the multiplanar coronal and longitudinal views obtained by 3D-TVS, the maximum and minimum JZ thickness will be measured from the basal endometrium to the internal layer of the outer myometrium (JZmax, JZmin), the difference between them (JZdif = JZmax − JZmin) and the ratio JZmax/total maximum myometrial thickness. The 3D-TVS markers JZdif ≥4 mm and JZ infiltration and distortion have high sensitivity (88%) and the best accuracy (85% and 82%, respectively). JZmax ≥ 8 mm and JZdif ≥4 mm are significantly more associated with adenomyosis than other two-dimensional features (Exacoustous)

2. Endometrioma:
- Unilocular cyst with ground glass echogenicity.
- Size
- Numbers,
- Unilateral/ bilateral

3. Deep Infiltrating Endometriosis, extent of disease:
- Site Specific Tenderness
- Kissing ovaries
- Pelvic adhesions
- Sactosalpinx
- Urinary bladder
- Ureters
- Rectovaginal septum
- Vaginal wall
- Uterosacral ligaments
- Rectum, rectosigmoid junction, sigmoid
- POD obliteration
- Number of lesions, size

Patient treatment
All patients will undergo IVF or ICSI treatment according to either an antagonist or agonist protocol. Stimulation protocol will be chosen by treating reproduction medicine specialist according to individual patient characteristics. Recombinant FSH, or human menopausal gonadotropins (hMG) in individual doses will be used for ovarian stimulation. Follicle development will be monitored by vaginal ultrasound, and when three or more follicles reach 17 mm, ovulation will be induced with human chorionic gonadotropin (hCG) and transvaginal follicle aspiration will be performed 35-36 hours later.
Aspirated oocytes will be assessed, and either inseminated or injected with sperm depending on semen quality parameters. Fertilization will be recorded and embryos assessed through time-lapse technique and defined as Good Quality Embryos (GQE) according to the Gardner blastocyst grading scale. Embryo transfer (ET) will be done two, three or five days after ovum pick up (OPU). Luteal phase support with progesterone vagitories will be given vaginally for two weeks after OPU. Surplus embryos of good quality will be cryopreserved on day 5-6, and frozen-thawed ETs will be carried out in either natural or hormone replacement cycles depending on the woman´s ovulatory status. All fresh and frozen embryos from the same index treatment cycle will be used until pregnancy is achieved or no embryos remain, resulting in some women contributing with more than one embryo transfer.
Pregnancy will be monitored by vaginal ultrasound in the age of 7-8 gestational weeks , and miscarriages registered (< 8 weeks, < 12 weeks, ≥ 12 weeks). Live births, defined as delivery of one living child in gestational week > 22, will be registered. Data regarding pregnancy length, birth weight, mode of delivery (vaginal delivery/caesarean section), obstetric bleeding and diagnosis of preeclampsia will be drawn from patients´ medical journals (Obstetrix, Melior).
Statistical analysis
Statistical methods will be applied using SPSS software. Power analysis, to establish the size of the patient groups, will be performed prior to starting the studies.
Time plan
We estimatet that approximately 1000 patients will be needed. 1400 Patients are treated at RMC yearly, and we believe that data collection will take 2,5 years. Data analysis will be made continuously. The first manuscript has been submitted, and we estimate that the whole project will take 4 years.

Layman's description

Endometrios är en sjukdom som drabbar 10 – 20% av kvinnor i fertil ålder. Sjukdomen har varierande svårighetsgrad och symptom, men yttrar sig ofta som svår dysmenorré, kronisk buksmärta, samlagssmärta, smärta vid defekation samt miktionsbesvär. Sjukdomen finns i kliniskt distinkta former beroende på lokalisation; på ovarier (endometriom), peritoneum eller djupt infiltrerande i tarm (DIE). Adenomyos är ett relaterat tillstånd, som ofta leder till smärtsamma och rikliga menstruationer. Hos många kvinnor orsakar endometrios och adenomyos sub – eller infertilitet. Relationen mellan olika former av endometrios/adenomyos och påverkan på ovarialreserv och fertilitet är dock ofullständigt kända.
Traditionellt har endometrios diagnosticerats via laparoskopi med biopsier, och adenomyos i samband med hysterektomi. Modern ultraljudsteknologi har dock gjort det möjligt att på ett precist sätt diagnosticera sjukdomen utan invasiv ingrepp dvs histologi. Förhoppningen är att man med modern ultraljudsteknologi ska kunna möjliggöra upptäckt av nya kännetecken hos sjukdomen samt en tidigare diagnos.
Syftet med projektet är dels att undersöka och beskriva vilka ultraljudsmässiga kriterier för diagnos av endometrios och adenomyos upptäcks i population av subfertila kvinnor, och dels att utvärdera hur Anti-Müllerian Hormone (AMH) och atrial follicullar count (AFC) ändras hos kvinnor med specifika ultraljudsmässiga kännetecken på endometrios och adenomyos, hur kan det prediktera utfallet efter assisterad befruktning hos dessa kvinnor.
Preliminärt planeras fem delarbeten, som kommer att utföras som prospektiva studier. I det första arbetet har vi undersökt hur AMH kan prediktera utfallet vid ART. I de övriga arbetena ska vi studera vilka ultraljudsmässiga kriterier av endometrios och adenomyos i kombination med AMH och AFC kan prediktera utfallet efter ART hos infertila patienter jämfört med kvinnor utan tecken på endometrios och adenomyos: 1) djupt infiltrativ endometrios (DIE) med respektive utan endometriom, 2) DIE + adenomyos, 3) adenomyos med respektive utan DIE eller endometriom. Ytterligare ett delarbete planeras, en prospektiv studie där kvinnor med kliniska symptom på endometrios men utan fynd vid ultraljudsundersökning följs upp efter 5 år, för att studera om synliga förändringar har utvecklats.

Primärt utfall kommer att vara födelsetal, men vi kommer också att analysera oocytkvalitet, graviditets – och missfallsfrekvens samt eventuellt negativt utfall såsom preeclampsi, intrauterin tillväxthämning, prematurbörd, ablatio samt obstetrisk blödning.
Short titleEndometriosis and subfertility
Effective start/end date2018/09/032022/09/05

Collaborative partners