Transvaginal sonography for the assessment of ovarian and pelvic endometriosis: how deep is our understanding?

editorial OA: bronze CC0 ⤵ 80 in-corpus citations
AI-generated summary by claude@2026-06, 2026-06-08

Transvaginal sonography effectively assesses ovarian and pelvic endometriosis by identifying typical and atypical endometriomas, deep infiltrating endometriosis in the uterosacral ligaments, rectum, vagina, and bladder, and adenomyosis.

One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works

Abstract

Endometriosis is a chronic disease affecting 10–15% of women in their reproductive years1. It is characterized by the presence of functional endometrial glands and stroma in sites outside the uterine cavity. Its treatment is both complex and challenging, involving medical or surgical approaches, or a combination of both. Laparoscopic excision of endometriotic lesions is the only treatment which aims at eliminating the disease2, 3, but surgical removal of deep infiltrating endometriosis (DIE) is risky, requiring long operating times and excellent technical skills3, 4. Thus, accurate preoperative mapping of all endometriotic lesions is always required in order to evaluate the various treatment options. Ultrasound is the imaging modality of choice in the assessment of an adnexal mass, given its high accuracy in evaluating the likelihood of malignancy. Moreover, for some histological types of ovarian cyst, such as endometriomas, it is possible not only to predict if the cyst is benign or malignant by means of TVS, but even to predict the likely histological nature of the mass (so-called 'sonohistology'). Several studies have described the ultrasound characteristics of endometriomas and defined their typical ultrasound features11-13. A 'typical' endometrioma is a unilocular cyst with a regular wall and homogeneously low-level echogenicity (so-called 'ground glass' appearance) of the cyst's content (Figure 1). Hyperechogenic wall foci can be seen in up to a third of endometriomas and are quite distinctive, as they are rarely found in other benign non-resolving ovarian cysts13. The pathological basis of these foci has not yet been established; although it can be hypothesized that they represent cholesterol or salt crystal deposits following the breakdown of cell membranes. These foci accumulate over time along the inner surface of the cyst wall and are rarely seen in newly formed endometriomas; thus, from personal experience, this feature may help in estimating the age of the mass. Almost half of endometriomas manifest ultrasound characteristics other than 'unilocular cyst with ground glass echogenicity'; moreover, endometriomas may look different in pre- and postmenopausal patients (Van Holsbeke and the IOTA study group, unpubl. data). Septations are a frequent finding in endometriomas, giving the cyst a multilocular appearance. The wall of an endometrioma is usually smooth and clearly visible, but irregularity of the profile or even frank papillary projections can be present as a result of several histological processes involving the wall of the cyst, including inflammation, necrosis, hemorrhage and decidualization, to name but a few (Figure 2). These so-called 'atypical' endometriomas can lead to diagnostic problems and, in rare cases, the differential diagnosis with an ovarian malignancy can be difficult. The presence of ovarian cancers among the masses included in the differential diagnosis of endometriomas is of concern; it seems that the highest risk of misclassifying an ovarian cancer as an endometrioma is in postmenopausal women with an ovarian cyst of low-level echogenicity (Van Holsbeke and the IOTA study group, unpubl. data). Transvaginal ultrasound image of a typical endometrioma. The content is homogeneous and composed of low-level echoes and the cyst wall is regular and smooth. Transvaginal ultrasound image of an atypical endometriotic cyst. Note the presence of focal wall nodularity with absence of blood flow. Power Doppler depicts sparse vascularization, the few blood vessels being confined to the cyst wall. The differential diagnosis of endometriomas includes luteal cysts, cystadenomas, pyosalpinges, dermoids and ovarian cancers, because, in these masses, the cyst content (blood, mucus or pus) may display low-level echoes on ultrasound. On TVS, an experienced sonologist should be able, in most cases, to distinguish between endometriomas and other types of adnexal mass by means of pattern recognition14. Moreover, power Doppler can be of help in showing the absence of flow within the cyst content when heterogeneous inner portions are found due to intracystic hemorrhage or accumulation of the dense parts of the content15. Endometriotic cysts are associated with scanty vascularization of the cyst wall, while non-endometriotic cysts, particularly luteal cysts and ovarian cancers, are characterized by rich vascularization of the wall and the presence of arterial flow within papillary projections and echogenic areas of the cyst. Another distinct feature of endometriomas is that they usually form adhesions. Thus, they can be found far from the true adnexal site (eg. attached to the pouch of Douglas), being fixed and painful when pressure is applied with the vaginal probe. These features may be revealed only if gentle pressure is exerted while visualizing the cyst on the screen (so-called 'sliding sign'). When endometriomas are bilateral, they can even be found adhering to each other behind or above the uterus (so-called 'kissing ovaries'). One of the most frequent locations of extraovarian endometriosis is the uterosacral ligaments (USLs) and the torus uterinus8, defined anatomically as a small transverse thickening joining the insertion of the USLs to the posterior wall of the uterus6, 16. Involvement of these structures by endometriotic implants produces a discrete sheet-like or stellate hypoechoic nodule with irregular outer margins, usually located in close proximity to the uterine cervix (Figure 3)8. A normal USL is barely detectable on TVS, but, when harboring DIE, it thickens and becomes visible17. I find it helpful to evaluate both USLs by means of the split-screen technique: simultaneous comparison of the left and right USL will easily depict a nodule if it is unilateral; asymmetry between the two ligaments and irregularity of their profiles are more specific for the presence of endometriosis than is a simple thickness measurement17. Moreover, deep endometriotic implants are fixed and painful under pressure from the transvaginal probe; the mobility of a given structure and the pain evoked during examination will help in diagnosing a possible USL endometriotic nodule. Transverse transvaginal ultrasound image at the level of the uterine cervix, showing the presence of a hypoechoic nodule of irregular morphology and blurred margins located in the median third of the left uterosacral ligament. I recommend routinely assessing the sliding of various anatomical structures and the mapping of painful sites in the pelvis while asking the patient about her symptoms. Does the pain evoked by the pressure of the probe resemble the deep dyspareunia experienced? This will help in making a correct diagnosis. Finally, with experience, it is possible to perceive the resistance offered by the nodule to the slide of the transvaginal probe, similar to the findings on bimanual pelvic examination. Another two common locations of extraovarian endometriosis are the antimesenteric portion of the rectosigmoid junction and the rectum. It is crucial to diagnose correctly the presence of DIE in these locations in order to accurately predict the risks and difficulties of surgery if this has to be performed. Moreover, the duration of the operation, the need for a general surgeon to be on call, the length of hospital stay and extent of surgical complications are strictly dependent on whether or not intestinal DIE is present. Clinical examination is of limited use in establishing both the presence and the extent of DIE lesions6, 7, so preoperative imaging modalities are often required when DIE is suspected. TVS has been demonstrated to be less invasive, better accepted and more accurate than are transrectal ultrasound9 and rectal endoscopic sonography8, and to be more accurate than are MRI and clinical examination7. On TVS, a normal rectal wall exhibits a thin (< 3 mm) hypoechoic smooth muscle layer and a hyperechoic internal layer corresponding to the rectal submucosa and mucosa. Histologically, the endometriotic nodule progressively infiltrates the serosal surface of the bowel and then reaches the muscolaris propria, forming a bulky nodule composed of smooth muscle cells, collagen, fibroblasts, islands of scattered endometrial stroma and glands18. The submucosal and mucosal layers are rarely affected, which explains the high proportion of false negatives on colonoscopy19. Diagnosis of bowel endometriosis on TVS is straightforward when a hypoechoic fixed nodule is seen behind the cervix, attached to the bowel wall (Figure 4). The external margins of the nodule are hyperechoic due to the presence of congested adipose tissue, submucosa and mucosa. Some nodules manifest internal hyperechoic spots, probably due to calcified portions; power Doppler invariably shows few blood vessels within and around the nodule. Moreover, as most of the nodules obliterate the pouch of Douglas, I again recommend considering TVS a dynamic examination and evaluating the sliding of the cervix along the rectum by gently pushing the vaginal probe while looking for the presence of DIE. When intestinal DIE is found, the examiner should be aware that, in up to 93% of cases, there is a second DIE location (USL, vagina, bladder, ureter)6, 9. TVS is of limited value in locating DIE in the sigmoid colon, above the level of the uterine fundus, far from the tip of the probe. These lesions may be difficult to visualize, especially if air and stool are interposed. Moreover, on TVS, it is impossible to measure the distance between the lower limit of the lesion and the anal canal. In conclusion, after a thorough anamnesis, clinical bimanual pelvic examination should always be coupled with TVS, possibly performed by the same trained physician, as the combination of both will give a sensitivity of approximately 95–98% for lesions affecting the rectosigmoid colon and rectum7, 10. The superiority of TVS over other imaging modalities, especially MRI, is in part because of the ease of detecting small implants of endometriosis on TVS, and because bowel movements do not produce artifacts on TVS, while they hinder diagnosis on MRI7. Sagittal transvaginal ultrasound image of the posterior compartment of the pelvis, showing the presence of an endometriotic nodule (deep infiltrating endometriosis) located on the anterior wall of the rectum, extending caudally towards the rectovaginal septum. The implant appears as a solid hypoechoic nodule with blurred margins and a hyperechoic rim, disrupting the regular aspect of the muscolaris propria. Superficial peritoneal endometriotic implants can be found in up to 15% of normal healthy women but are barely visible with any imaging modality. Moreover, one should consider that almost 100% of patients with ovarian endometriomas have superficial disease elsewhere along the pelvic peritoneum or intestinal tract20. As described previously, TVS can be used to evaluate the mobility of a given structure (eg. an endometrioma), as well as site-specific tenderness. Okaro et al.21 have used TVS to evaluate ovarian mobility, site-specific tenderness and loculated peritoneal fluid ('peritoneal pseudocysts') in women with chronic pelvic pain. They reported that such features can be considered useful soft markers for the presence of adhesions and superficial endometriosis. They found that the addition of soft marker analysis can improve the sensitivity of TVS in diagnosing peritoneal endometriosis from 34 to 87%, with a high negative predictive value of 84%. The vagina is considered to be involved when its wall is thickened or when a nodule with an irregular outer contour and spiculations is seen attached to it. The most frequently affected portion of the vaginal canal is the posterior fornix. Involvement of the true rectovaginal septum alone is rare, and this is usually secondary to the presence of a huge bulky nodule, originating in the pouch of Douglas or the rectum, which has extended caudally. The sensitivity of TVS for the diagnosis of isolated endometriosis located on the vaginal wall and rectovaginal septum is reported to be as low as 29%17. This seems to be due to the configuration of the vaginal probe, with the receiver on its tip, and the fact that the symphysis pubis limits the inclination of the probe toward the posterior vaginal wall. Thus, implants located on the posterior vaginal fornix close to the uterine cervix can be visualized, while those located in the posterior vaginal wall and the rectovaginal septum may be missed. A technique has been described involving the creation of a sonographic stand-off, obtained by increasing the amount of ultrasound gel inside the probe cover in order to visualize the area near the posterior vaginal wall and rectovaginal septum22. This promising approach showed good specificity and sensitivity for these specific locations of DIE. Involvement of the urinary tract occurs in approximately 1–2% of patients with endometriosis and in 90% of these cases the bladder is involved23. Once considered a rare pathological condition, bladder endometriosis is probably underdiagnosed because of its non-specific symptoms, often mimicking recurrent cystitis with dysuria, urgency, frequency, suprapubic pain, vesical tenesmus, incontinence and hematuria6. In order to identify the presence of bladder endometriosis, the transducer should be positioned in the anterior vaginal fornix and tilted upward to visualize the vesicouterine space and the bladder, using both longitudinal and transverse sections. In these planes, the bladder wall is visualized easily if a moderate amount of urine is present. Diagnostic criteria suggestive of a bladder endometriotic nodule include the presence of a hypo- or isoechogenic nodule within the bladder wall17 and the presence of a nodule with heterogeneous echostructure containing numerous anechoic ('bubble-like') areas24. The lesion can be located in the bladder base (close to the ureteral ostia) or in the bladder dome and can have a 'nodular' or a 'comma' shape25. Small internal anechoic cystic areas are seen in approximately 30% of the nodules. As with other locations, patients with bladder endometriosis report pain under pressure from the transvaginal probe. Adenomyosis is a common disorder which affects women in their late reproductive years. It is characterized by the presence of endometrial glands and stroma intermingled with the uterine smooth muscle, showing signs of hyperplasia. It can be confined to a specific area of the uterus or diffused throughout the whole organ, but the cervix is rarely involved. On TVS, adenomyosis can manifest as the presence of uterine enlargement without signs of myomas or as asymmetrical thickening of the anterior or posterior uterine walls26. Other signs of adenomyosis include: hypoechoic areas scattered through the myometrium, heterogeneous myometrial echotexture27, 28 and linear striations radiating out from the endometrium (Figure 5)29, 30. The endometrial–myometrial interface can be fuzzy and ill-defined, with a pseudo-thickening of the endometrium29. These sonographic findings are usually subtle and easier to identify during the course of real-time (dynamic) TVS examination rather than from still images28. The overall diagnostic accuracy of TVS is good in clinically suspected cases of adenomyosis, in line with that of MRI in unselected patients without myomas. In the presence of myomas, especially if they are multiple or huge, the diagnostic accuracy of TVS for adenomyosis is reduced31. Transvaginal ultrasound image of an adenomyotic retroflexed uterus showing a heterogeneous myometrial echotexture, linear myometrial striations and ill-defined endometrial–myometrial borders. In conclusion, TVS should be regarded as the first-line imaging modality in the evaluation of patients with suspected endometriosis. The accuracy of TVS has greatly improved over recent years as knowledge has increased regarding the various sonographic aspects of endometriosis. The performance of ultrasound is heavily operator-dependent, which is both a limitation and a strength of this technique. It is important to 'think endometriotic' and to familiarize oneself not only with endometriosis of the ovaries but also with the characteristics which the bowel, USLs, vagina and bladder show when affected by DIE. Good training, skills and passion are prerequisites for evaluation of patients in order to make the diagnosis of endometriosis less elusive.

My notes (saved in your browser only)

Condition tags

mesh:D004715mesh:D017699endometriosisadenomyosisdie_deep_infiltratingendometriomabowel_endometriosisbladder_endometriosischronic_pelvic_paindyspareunia

MeSH descriptors

Endometriosis Endosonography Ovarian Diseases Pelvic Pain Adult Clinical Competence Endometriosis Endosonography Endosonography Female Humans Ovarian Diseases Pelvic Pain

Citation neighborhood

Papers in the corpus that this work cites (lower rings, blue) and that cite this one (upper rings, green). Dot size scales with the paper's in-corpus citation count — bigger dot = more influential within the endo/adeno field. Click a dot to open that paper. [ expand to 2 hops ] — adds papers reached through this work's immediate citers/citees. Heavier; up to 60 extra dots.

References (32)

Cited by (50)

Source provenance

europepmc
last seen: 2026-06-04T01:30:01.192114+00:00
openalex
last seen: 2026-06-10T17:14:06.276822+00:00
pubmed
last seen: 2026-05-13T22:14:11.755070+00:00
License: CC0 · commercial use OK