Wendaline M. VanBuren

Purpose To determine whether presence of endometriomas on routine pelvic ultrasound (US) can predict presence of deep infiltrating endometriosis (DE) on magnetic resonance imaging (MRI) and correlate endometrioma size with the presence of DE on MRI. Materials and methods In this IRB approved, HIPAA compliant study, radiology data base was queried for MRI exams dedicated for endometriosis evaluation in patients with surgically proven endometriosis over a three-year period (2016–2019). Imaging reports were reviewed for the presence of endometriomas and DE on MRI. For the patients with endometriomas, records were reviewed for concurrent routine pelvic US imaging, which were not tailored to DE protocols. US images were reviewed for the presence of endometriomas, their total number, laterality and largest dimension. Descriptive statistics and receiver operating curve (ROC) analysis were performed. Pathology and surgical notes were used as reference standard. Results 253 patients with surgically confirmed endometriosis underwent MRI for DIE over a 3-year duration. 47 patients had a concurrent US exam. 33/47 patients (70.2%) had endometriomas seen on US (size range 0.9–7.0 cm). 12 had bilateral endometriomas and 13 had multiple (more than 1) endometriomas. 27/33 (82%) of these patients had DE on MRI. 6/33 (18%) patients had additional sites of DE on surgery, which was not reported preoperatively on MRI. When endometriomas more than 3 cm as were evaluated, 19/21 (90%) of patients had DE on MRI. AUC was 0.7106. Using ROC analysis, a threshold of 3.8 cm, provided a sensitivity of 71% and specificity of 78% for detection of DE on MRI. Conclusion Presence of endometriomas on routine pelvic US is associated with a high frequency of DE detected on MRI performed for endometriosis. The frequency of DE on MRI was higher in patients with endometrioma size > 3 cm compared to endometriomas of all size range.

Background Rectosigmoid endometriosis (RSE) presents with a diverse array of MRI findings that impact surgical planning. No standardized reporting and data system has been established for RSE. Purpose We propose a novel MRI scoring system designed to predict the likelihood of muscularis propria (MP) involvement in RSE, which would, in turn, influence surgical planning. Materials and methods The records of patients with bowel endometriosis treated surgically from May 2018 to June 2022 were retrieved. Surgery was classified as partial thickness discoid, full thickness discoid, or segmental resection. Each pre-treatment MRI was scored based on the mutual agreement of two abdominal radiologists (reference score). The MRI score was defined as (1) score 0: no evidence of RSE, (2) score 1: minimal tethering involving the serosal surface without MP involvement, (3) score 2: intermediate soft tissue thickening involving the rectosigmoid colon with indeterminate MP involvement, or (4) score 3: definite mushroom cap sign or definite MP involvement. In the reader study, two radiologists independently scored each exam. The area under the curve (AUC) was evaluated for predicting the need for segmental or full thickness discoid resection. Results The cohort consisted of 95 patients (median age: 36 years); 16, 14, 30, and 35 patients had MRI score 0, 1, 2, and 3, respectively. Patients with MRI scores 3 and 2 underwent partial thickness discoid (6% vs. 50%), full thickness discoid (6% vs. 17%), and segmental resection (89% vs. 33%), respectively. All patients with MRI scores 1 or 0 either underwent partial thickness discoid resection or did not undergo rectosigmoid surgery. The AUCs were 92.2%, 84.5% and 93.9% for MRI scores of the reference, reader 1, and 2, respectively. Conclusion Our MRI scoring system based on suspected depth of bowel invasion showed good diagnostic performance to predict the type of surgical intervention needed.

Endometriosis is a multifocal disease of pelvis affecting mostly reproductive age women. It can involve genital and extra genital organs. The gastrointestinal tract including rectum and sigmoid colon is the most affected extragenital site. Bowel involvement can change the surgery from simple shaving to resection based on depth of invasion and disease distribution. Benefit of ultrasound for bowel endometriosis, especially for rectum and rectosigmoid colon have been reported, however MRI directed ultrasound, particularly for the lesions which demonstrate indeterminate depth of wall invasion and multifocal bowel involvement hasn’t been reported. Here we present a case of rectosigmoid endometriosis evaluated with preoperative MRI directed ultrasound, which provided better visualization of muscularis propria invasion of the bowel and multifocal lesions.

Background and purpose Endometriosis affects 5–10% of women of reproductive age. Despite its prevalence, diagnosing endometriosis through imaging remains challenging. Advances in deep learning (DL) are revolutionizing the diagnosis and management of complex medical conditions. This study aims to evaluate DL tools in enhancing the accuracy of multi-sequence MRI-based detection of endometriosis. Method We gathered a patient cohort from our institutional database, composed of patients with pathologically confirmed endometriosis from 2015 to 2024. We created an age-matched control group that underwent a similar MR protocol without an endometriosis diagnosis. We used sagittal fat-saturated T1-weighted (T1W FS) pre- and post-contrast and T2-weighted (T2W) MRIs. Our dataset was split at the patient level, allocating 12.5% for testing and conducting seven-fold cross-validation on the remainder. Seven abdominal radiologists with experience in endometriosis MRI and complex surgical planning and one women’s imaging fellow with specific training in endometriosis MRI reviewed a random selection of images and documented their endometriosis detection. Results 395 and 356 patients were included in the case and control groups respectively. The final 3D-DenseNet-121 classifier model demonstrated robust performance. Our findings indicated the most accurate predictions were obtained using T2W, T1W FS pre-, and post-contrast images. Using an ensemble technique on the test set resulted in an F1 Score of 0.881, AUROCC of 0.911, sensitivity of 0.976, and specificity of 0.720. Radiologists achieved 84.48% and 87.93% sensitivity without and with AI assistance in detecting endometriosis. The agreement among radiologists in predicting labels for endometriosis was measured as a Fleiss’ kappa of 0.5718 without AI assistance and 0.6839 with AI assistance. Conclusion This study introduced the first DL model to use multi-sequence MRI on a large cohort, showing results equivalent to human detection by trained readers in identifying endometriosis.

Endometriosis, a chronic and multifocal inflammatory condition with a substantial estrogen-dependent component, is often linked to infertility. Some patients with endometriosis may require surgical intervention or assisted reproductive technologies to conceive. Although many patients who achieve pregnancy have relatively uncomplicated outcomes because of the progesterone-induced regression of endometriotic lesions, complications can still arise during pregnancy and the peripartum period. Complications include the decidualization of endometriosis implants, with site-specific implications (e.g., decidualized endometrioma mimicking ovarian cancer, decidualized deep endometriosis infiltrating the myometrium leading to uterine rupture, spontaneous hemoperitoneum), placenta previa, preterm labor and premature rupture of membranes, postpartum hemorrhage, or systemic conditions such as hypertensive or coagulation disorders. Herein, we review the background of these conditions and the expected radiologic findings. Additionally, we review essential clinical concepts about the treatment available and the information needed to make health care decisions. This review aims to equip radiologists with essential insights into the challenges faced by patients with endometriosis, from infertility diagnosis through postpartum care. By enhancing radiologists’ understanding of these aspects and relevant imaging findings, we aspire to improve maternal and fetal outcomes affected by this complex condition.

Bowel endometriosis, most commonly involving the rectosigmoid colon, is best evaluated with transvaginal US with bowel preparation or with MRI, and knowledge of normal bowel anatomy and disease manifestations in different bowel segments and identification of coexisting sites of endometriosis are important for accurate diagnosis.

Endometriosis is the presence of hormonally sensitive endometrium-like tissue outside the uterus. It is a common condition, affecting 10% of reproductive-age people assigned as female at birth. Although usually occurring in the pelvis, it can rarely involve the thoracic cavity and diaphragm which is termed thoracic endometriosis. Thoracic endometriosis syndrome (TES) refers to four well-recognized clinical entities: catamenial pneumothorax, catamenial hemothorax, catamenial hemoptysis, and lung nodules. However, TES presentation can also be nonspecific, even absent. Optimal management is multidisciplinary, as it depends on patient presentation and lesion characteristics. While initial imaging of thoracic symptoms is with chest radiography and computerized tomography, these modalities have inferior soft tissue contrast resolution and tissue characterization compared to magnetic resonance imaging (MRI), especially for the detection of both hematic and cystic endometriotic implants. Therefore, a dedicated MRI protocol is essential for diagnosing thoracic endometriosis and aiding surgical planning. Considering the dome-shaped morphology of the diaphragm, sagittal and coronal projections can improve visualization of tiny endometriotic plaques or deposits that are inconspicuous on the axial plane. Breath-hold and respiratory-triggered or navigated techniques are critical for mitigating motion artifacts. T1-weighted fat-suppressed sequences are important for identifying intrinsic T1 hyperintensity and blood products associated with endometriotic lesions. T2-weighted fat-suppressed sequences increase sensitivity for cystic or vesicular tissue. Diffusion-weighted and postcontrast imaging can help diagnose alternative causes of symptomology, including malignancy.