Endometriosis is a chronic gynecological disease, defined as the growth of functional ectopic endometrial glands and stroma outside the uterus.
The different manifestations of endometriosis include ovarian endometriomas, endometrial adenomyosis, superficial peritoneal implants, deep pelvic endometriosis (DPE), and extra-pelvic endometriosis.
This condition affects 10–15% of women of reproductive age, and develops with a wide range of symptoms and signs, including chronic pelvic pain, dysmenorrhea, dyspareunia, and hematuria.
A poor correlation between symptoms and severity of lesions exists.
The non-specificity of clinical manifestations leads to a significant diagnostic delay from the onset of symptoms.
Multicentric studies conducted in some European countries quantify this delay in 7.4 years in the Netherlands (median), 11.4 years in Italy (mean), and 10.4 years in Austria and Germany (median) respectively.
Imaging plays a fundamental role in the attempt to contrast this diagnostic issue, which causes further pain and discouragement in patients as well as a therapeutic delay contributing to the disease’s worsening, which can culminate in infertility. 1.
MRI indications The 2022 updated European Society for Human Reproduction and Embryology (ESHRE) guidelines for endometriosis undermine the concept of laparoscopic surgery as the gold standard for diagnosis and conclude that both diagnostic laparoscopy and imaging associated with empirical treatment can be considered in women suspected of endometriosis and that no evidence of superiority of either approach exists; the diagnostic workup should be tailored to the patient, weighing up the benefit of laparoscopy against its risks.
Furthermore, explorative laparoscopy is limited in the evaluation of retroperitoneal structures.
New hormonal therapy has emerged as a feasible approach to delay or prevent surgical treatments.
As for imaging, the 2017 European Society of Urogenital Radiology (ESUR) guidelines on MR imaging of pelvic endometriosis suggest that transvaginal ultrasound (TVUS) should be regarded as a first-line imaging modality, whereas MRI is indicated as a second-line examination in symptomatic patients with clinical suspicion of endometriosis and negative US findings, as well as in the presurgical evaluation of DPE, owing to its capability of providing anatomical detail and a complete assessment of all pelvic compartments. 2.
MRI acquisition protocol According to the aforementioned 2017 ESUR guidelines, certain recommendations should be followed in the acquisition protocol and in patient preparation.
MR examination is to be performed on either 1.5T or 3T scanners and the use of pelvic phased array coils is recommended.
The guidelines recommend the acquisition of three 2D-T2 weighted (T2w) MRI sequences (sagittal, axial, oblique); the addition of a 3D-T2w MRI sequence is considered optional.
T1 weighted (T1w) MRI sequences with and without fat suppression are recommended in the assessment of adnexal endometriosis, and the “Dixon technique” can be used instead of a standard T1w sequence.
Some authors suggest that Short-tau inversion recovery (STIR) sequences should be avoided, since the signal from any tissue with a short T1 may be suppressed, including the blood contained in endometriomas, mimicking an adnexal mature cystic teratoma.
In addition, T2w sequences with fat saturation should be avoided since they would pose an obstacle to the contrast between the hyperintensity of fatty tissue and the low intensity of endometriomas.
No recommendation was reached about the use of diffusion-weighted (DW), susceptibility-weighted (SW), and post-contrast imaging in the assessment of DPE.
The employment of gadolinium is considered a viable option in the evaluation of indeterminate adnexal endometriosis.
At our institution, we acquire axial, sagittal and coronal 2D-T2w Turbo Spin Echo (TSE) sequences, an iso-volumetric T1 Dixon sequence with 1 mm thickness, and a DWI sequence with three values of b (0- 500- 800).
Contrast is used only at the discretion of the radiologist if findings are considered suspicious for malignant degeneration of endometriosis. 3.
Patient preparation The 2017 ESUR guidelines state that the protocol should be tailored to the main indication for pelvic MRI, including diagnosis or staging of DPE and characterization of an indeterminate adnexal mass.
Abdominal strapping is recommended in the assessment of pelvic endometriosis and the exam should be performed in the supine position, except for cases of claustrophobia, in which the prone position may be an option.
Fasting is recommended in the evaluation of DPE, but a variable duration of fasting is cited (3, 4 or 6 hours) in literature.
The administration of a spasmolytic agent is recommended.
Bowel preparation is defined as ‘best practice’ for the detection of DPE, and a moderately full bladder is recommended.
No recommendation is proposed for timing of MRI in relation to the menstrual cycle.
Vaginal opacification and rectal opacification are suggested as an option in the evaluation of pelvic endometriosis.
At our institution, neither vaginal nor rectal opacification is performed.
Butylscopolamine is administered intravenously at the beginning of the examination.
With respect to the timing of the menstrual cycle, no selection is applied. 4.
MRI features The different manifestations of endometriosis include ovarian endometriomas, endometrial adenomyosis, superficial peritoneal endometriosis, deep pelvic endometriosis and extra-pelvic endometriosis.
The appearance of endometriotic lesions is variable and depends both on the location and the amount of ectopic tissue, but also on the duration of the untreated disease and the consequent fibrosis. 4.1| Endometriomas Ovaries are the most common site of endometriosis implants (up to 40% of cases), which may present as superficial implants, micro intra-ovarian endometriomas and endometriotic cyst (endometriomas).
Micro intra-ovarian endometriomas are sub-centimetric implants within the ovaries, often multiple and bilateral.
At MRI, they present with high signal intensity at T1w images and variable signal intensity on T2w images.
Endometriomas (Fig. 1) are pseudocysts formed by the invagination of endometriosis foci within the ovarian cortex, resulting from cyclic hemorrhage into the endometrial implant, which causes an accumulation of blood products.
They often are multilocular and bilateral.
Endometriomas present with a marked hyperintense signal on T1w images (homogeneous or with fluid-fluid levels) and a corresponding low signal on T2w images.
T2 shading sign is a sensible yet non-specific sign for the diagnosis of endometriomas and may appear as fluid–fluid levels, a gradient in signal loss or as a homogeneous hypointensity.
It results from blood products from cyclic bleeding occurring over time within the cyst, thus decreasing the T2 relaxation time.
The so-called T2 dark spot sign is a more specific sign and the markedly hypointense foci within the cyst result from chronic hemorrhage.
It is useful in the differential diagnosis between a solitary endometrioma and a functional hemorrhagic ovarian cyst.
The presence of ovarian endometriomas is related to an increased number of deep pelvic endometriosis lesions.
In certain cases, adhesion formation between the peritoneal surfaces of ovaries may lead to their retraction behind the uterus in the pouch of Douglas, a sign known as kissing ovaries (Fig. 5).
Endometriomas often show low ADC values, therefore DWI does not play a significant role in differential diagnosis with cystic ovarian neoplasms.
A role for DWI has been suggested in differentiating endometriomas from hemorrhagic cysts, with significantly lower ADC values in endometriomas when compared to hemorrhagic ovarian cysts.
The differential diagnosis includes lesions with T1-hyperintensity, namely hemorrhagic cysts, mature cystic teratoma (which demonstrates loss of T1-hyperintensity on fat-suppressed images) and mucinous cystic neoplasms (which show proteinaceous hyperintensity on T1w images, lower than that of blood). 4.2 | Adenomyosis Adenomyosis (Fig. 2) is defined as the presence of ectopic endometrial glands and stroma within the myometrium, owing to the infiltration of the basal endometrium into the underlying myometrium.
This leads to hypertrophy and hyperplasia of smooth muscle.
TVUS and MRI are the main imaging methods for the diagnosis.
Adenomyosis usually develops from the junctional zone; the junctional zone represents the inner myometrium, usually visualized on MRI as a low T2 signal layer between the endometrium and the muscular stroma of the myometrium.
Adenomyosis appears as increased thickness of the junctional zone; when the thickness exceeds 12 mm, there is a high positive predictive value (PPV) for the diagnosis.
A junctional zone thickness lower than 8 mm generally rules out adenomyosis.
Some authors suggest that a thickness between 8 and 12 mm is not specific for adenomyosis, so the diagnosis requires additional findings, such as a difference between the maximum and the minimum thickness of the junctional zone in both anterior and posterior aspects of the uterus > 5 mm, and a ratio of maximal junctional zone thickness to myometrium thickness >40%.
Associated T2-hyperintense foci suggest heterotopic endometrial tissue, cystic dilatation of endometrial glands or hemorrhagic foci.
If hemorragic foci are present, they appear hyperintense on fat saturated T1w images, with a high PPV (95%).
Adenomyosis is generally located in the fundus, mainly along the posterior wall, of a large and asymmetric uterus.
Two types of adenomyosis have been described: diffuse, in which foci of adenomyosis are spread throughout the uterus, and focal (also called adenomyoma), which is represented by a discrete adenomyotic nodule within the myometrium. 4.3 | Superficial peritoneal implants Superficial peritoneal implants show a depth of peritoneal infiltration < 5 mm, involving the surface of pelvic organs or pelvic peritoneum.
At MRI, these small implants manifest as cystic or nodular lesions, characterized by T1-hyperintensity on fat-suppressed images and variable T2 signal, due to their hemorrhagic content (Fig. 3).
The inflammatory response induced by endometrial implants may result in post-contrast diffuse enhancement of peritoneal reflections over the rectouterine pouch and the uterus.
Adhesion formation may complicate endometriosis as a result of fibrosis.
Synechiae are recognizable on T2w images as hypointense peritoneal strands retracting pelvic organs, and their presence may also be suggested by the loss of the fatty cleavage planes separating pelvic structures, even in the absence of evident nodular lesions.
Indirect signs of adhesions include an angulated shape of the rectosigmoid colon (with a drop-like appearance of colonic walls) and bowel loops, posterior displacement of the uterus and adnexa (kissing ovaries sign), elevation of the posterior vaginal fornix, the presence of hydrosalpinx and loculated fluid collections or free fluid in a non-dependent distribution. 4.4 | Deep pelvic endometriosis In deep pelvic endometriosis, endometrial glands and stroma infiltrate more than 5 mm below the serosal or peritoneal surfaces.
The induced smooth muscle proliferation and fibrous reaction lead to the formation of solid nodules.
The most frequent sites of DPE implants include the pouch of Douglas, the uterosacral ligaments, the posterior vaginal fornix and the torus uterinus (i.e., a small transverse thickening at the insertion of uterosacral ligaments behind the posterior cervix), but any pelvic structure can be involved.
At MRI, solid endometriotic lesions show a hypointense nodular appearance with irregular or stellate margins on T2w sequences (Fig. 4).
T2-hyperintense foci within the mass indicate dilated ectopic endometrial glands.
On fat-suppressed T1w images, the signal intensity is generally intermediate, even though some hyperintense spots can be seen when red cell extravasation into the stroma occurs.
On post-contrast images, the nodules may show enhancement due to fibrosis and inflammation. 4.5 | Extra-pelvic endometriosis Endometriosis may involve extra-pelvic anatomical structures (Fig. 5).
Abdominal wall and thoracic endometriosis are relatively frequent.
Abdominal wall endometriosis is due to the presence of ectopic endometrial tissue superficial to the parietal peritoneum of the abdominal wall.
It can be primary or related to prior surgery (e.g., caesarean section and hysterectomy) and include scar, umbilical and inguinal endometriosis.
These patients generally present with a painful mass in the abdominal wall, which can be more symptomatic during menstruation.
Scar endometriosis occurs at the caesarean scar, at uterine scars and at the laparoscopic trocar sites.
Surgical resection is recommended, and radiological examinations are useful to determine the extent of the lesion before surgery, fascial involvement, and the depth of infiltration.
On MRI, the presence of a signal compatible with blood components within the mass is suggestive of endometriosis.
Umbilical endometriosis includes primary and secondary endometriosis, the first being more frequent.
Secondary endometriosis occurs after laparoscopic uterine surgery or endometriosis surgery.
These patients present with such symptoms as swelling, pain, or bleeding at the lesion site.
Physical examination and ultrasonography can lead to the diagnosis, but Computed Tomography (CT) and MRI are more accurate in assessing the size and extent of lesions.
Inguinal endometriosis is rare and is predominantly located in the right groin.
At MRI, T1-hyperintense hemorrhagic cysts can be detected, with or without the presence of hypointense solid masses.
Fat-saturated T1w images are particularly useful in the diagnosis of inguinal endometriosis.
Thoracic endometriosis is characterized by the presence of endometrial ectopic tissue in the thoracic cavity.
Catamenial pneumothorax is the most common clinical presentation of thoracic endometriosis, followed by catamenial haemothorax, catamenial hemoptysis and lung nodules.
In 90% of cases, catamenial pneumothorax occurs at the right side.
Catamenial pneumothorax (CP) is defined as recurrent pneumothorax occurring within 72 hours from the onset of menstruation.
However, since endometriosis-related pneumothorax (ERP) can occur during non-menstrual periods, CP does not equal ERP.
Imaging can be useful in the evaluation of thoracic endometriosis, CT being the first-line method.
MRI shows a greater sensitivity compared to CT in diagnosing diaphragmatic, pleural and hemorrhagic lesions.
VATS (video-assisted thoracoscopic surgery) is the gold standard for diagnosing and treating thoracic endometriosis.
Visceral extra-pelvic endometriosis is rare and includes hepatic, renal, pancreatic, and biliary endometriosis.
The symptoms are related to the organ involved.
CT is the generally used imaging method, even though MRI can be useful in the characterization of haemorragic cystic lesions.
Although exceedingly rare, non-abdominal and non-thoracic sites of endometriosis have been described in literature, involving the central nervous system (conus medullaris, brain), lumbar vertebrae, extra-pelvic muscles, peripheral nerves, and nasal cavity.