Abstract
PURPOSE:
This research aims to present the findings of contrast-enhanced ultrasound (CEUS) in a series of patients with proven deep endometriosis (DE) and provide an updated literature review.
MATERIALS AND METHODS:
Between January 2018 and October 2022, seven patients with DE lesions had their imaging and medical records retrospectively reviewed. Clinical data, recorded images of a standardized conventional B-mode ultrasound, and Sonovue® CEUS were interpreted by two blinded, independent, experienced radiologists in consensus. The enhanced characteristics of the DE lesion on CEUS were also assessed using VueBox® software quantitatively.
RESULTS:
DE lesion appeared as irregular hypoechoic or heterogeneous on conventional ultrasound with dotted blood flow signal on color Doppler. Six of seven DE lesions showed heterogeneous hypo-enhancement in arterial phases. All the lesions showed a heterogeneous washout rapidly that began in the late arterial phase. In quantified analysis, the mean relative peak enhancement compared with adjacent tissue was 0.47±0.25.
CONCLUSION:
Our findings and literature review suggested that CEUS could be a feasible and promising non-invasive modality for diagnosing DE.
Introduction
Deep endometriosis (DE) presents as endometrium-like tissue lesions in the abdomen, extending on or under the peritoneal surface that is usually nodular, able to invade adjacent structures, and associated with fibrosis and disruption of normal anatomy [1]. 15% –30% of endometriosis patients suffer from DE, the most severe condition. DE affects the pelvic viscera and connective-fascial structures [2]. Imaging modalities play a significant part in diagnosing patients with suspected DE because it is impossible to properly diagnose DE based solely on symptoms, signs, and clinical examination [3].
Contrast-enhanced ultrasound (CEUS) is the most effective technique for visualizing the microvasculature of pelvic diseases. Sonovue, a contrast agent with SF6 microbubbles with an average diameter of 2.5 mm, is the most widely used contrast media. Ninety percent of the microbubbles have diameters less than 6 mm. These bubbles can easily pass through capillaries, making it possible to observe microvascular blood flow of the abdominal and pelvic lesion [4–6]. Over the past few years, CEUS has been widely used in diagnosing ovarian endometrial cysts, adenomyosis and abdominal wall endometriosis(AWE) [4, 8]. No study has systematically described the enhanced pattern of DE lesions except for some sporadic case reports [9, 10].
Our study aimed to introduce CEUS to diagnose DE and describe the imaging features and enhanced pattern of DE with a case series and literature review.
Materials and methods
Patients features
From January 2018 to October 2022, a unicentric retrospective analysis of 7 female patients with pathologically confirmed DE (mean age, 35.4±9.3 years; range 18–45 years) was performed (Table 1). The average BMI at diagnosis was 24.9±1.2. The average value of CA125 at diagnosis was 44.4±8.6U/ml. According to the Visual Analogue Scale (VAS), participants rated the cyclic pain between 3 and 5 points. Histological proof of all lesions was obtained through surgery.
Summary of clinical manifestations of seven patients
Summary of clinical manifestations of seven patients
PoD, pouch of Douglas; VAS, Visual Analogue Scale.
All patients underwent conventional B mode ultrasound (BMUS), color Doppler flow imaging (CDFI), and CEUS. The BMUS was used to measure three orthogonal diameters of the lesions, including the largest diameter and two perpendicular diameters. The dosage of SonoVue® was 2.4 mL, followed immediately by the injection of 5 mL sodium chloride solution. Images were recorded for 3 min after contrast agent injection.
Imaging analysis
Two independent radiologists (Z.J.Z and N.D) interpreted all examinations in consensus. A comparison was made between the CEUS features of DE lesions and the adjacent infiltrated structures. Criteria of imaging evaluation included: the diameter of each lesion, echogenicity, shape, and color Doppler features. The CEUS imaging review focused on: contrast enhancement patterns of lesions (hypo-, hyper-, iso-enhancing depending on whether their echogenicity was less than, greater than, or equal to the adjacent structure, respectively). The adjacent structures were defined as the normal tissue adjacent to the DE infiltrating area. In heterogeneous enhanced lesions, the active portion of the lesion was evaluated for enhancement. Two enhancement phases were analyzed, the arterial phase (10–30 s after contrast agent administration) and the venous phase (30–120 s after contrast agent administration) [11].
The CEUS evaluation software VueBox®(version 7.1, subprogram GI-Perfusion for generic perfusion quantification) was used to quantify the contrast response of the DE lesion and surrounding tissue [12, 13]. The following perfusion parameters were obtained: Wash-in rate(WiR), washout rate(WoR), Fall time(FT), Peak enhancement(PE), Time to peak(TTP).
Search strategy and study selection
The databases of Google Scholar, Embase, PubMed, and Web of Science were searched for previously published studies until February 20, 2023. Keywords and search strategies in each database were detailed in Electronic Supplementary Material. We included articles or case reports describing the CEUS features of DE. Review articles and conference abstracts were excluded.
Statistical analysis
Continuous variables were expressed as mean±standard deviation and categorical variables were presented as percentages or frequencies.
Results
Conventional ultrasound
6 of the examinations of the included patients were conducted with a transvaginal probe and 1 with the transabdominal probe. The location of the DE lesions was described using the definition of anatomical location according to the IDEA group consensus [14]. 3 of them were located in the bladder, and the others were located in the Small bowel, rectum, and pouch of Douglas(POD). Further examinations showed that all DE lesions had unclear margins. 4 DE lesions (4/7) were hypoechoic, and 3 (3/7) were heterogeneous. 6 patients presented irregular shapes, while the smallest lesion in the bladder was in regular shape. A small amount of dotted blood flow signal was detected on color or power Doppler in all 7 lesions (Table 2).
Qualitative results of CUS and CEUS
Qualitative results of CUS and CEUS
CUS, Conventional ultrasound; CEUS, Contrast-enhanced ultrasound; AP, arterial phase; VP, venous phase.
6 of the 7 DE lesions showed heterogeneous hypo-enhancement in arterial phases. However, the homogeneous enhanced pattern was shown in case 7, which is the smallest lesion. In addition, case 7 showed iso-enhancement in the venous phase, while other lesions exhibited hypo-enhancement. All the DE lesions showed heterogeneous washout rapidly that began in the late arterial phase. During the venous phase, the enhancement decreased to a level of hypoenhancement (Table 2).
Quantitative analysis of CEUS
The CEUS quantification of DE lesions in VueBox® yielded values for the mean time to peaks(TTP) of 12.7±6.1 s(range 7.9–25.4 s), fall time(FT) of 11.3±6.7 s(range 5.0–25.4 s), mean relative peak enhancement (PE) compared with adjacent tissue of 0.47±0.25 (Table 3). The relative washout rate was found to be faster in three cases, namely case 1 (204%), case 2 (260%), and case 4 (193%), compared to the adjacent structure. Conversely, the remaining four cases exhibited a slower washout rate. Similarly, the relative wash-in rate was faster in case 1 (175%), case 2 (175%), and case 4 (121%), while the other four cases exhibited a slower wash-in rate. In other words, Lesions with a rapid wash-in rate also exhibited a rapid washout rate, while lesions with a slow wash-in rate had a corresponding slow washout rate. Figures 1 and Figure 2 shows the representative cases.
Quantitative results of CEUS
Quantitative results of CEUS
TTP, Time to peak; PE, Peak enhancement; WiR, Wash-in rate; WoR, washout rate; FT, Fall time; POD, pouch of Douglas.

A 36-year-old woman(case 2) presented with unexplained blood and mucus in the stool and alternating diarrhea and constipation for three months. Her CA125 level was 56.0 U/ml. Standard TVS showed a 64×20×34 mm hypoechoic mass in the middle and upper rectum with unclear boundaries, possibly infiltrating the anterior wall of the rectum(A). A small amount of dotted blood flow signal could be detected by CDFI(B). The contrast agent was then injected intravenously after determining the optimum viewing plane. The lesion had mild enhancement (weaker than the adjacent rectal wall) and showed rapid washout. On delayed imaging, the lesion showed intense washout (E-H). Linearized time-intensity curve of the perfusion-echo-power signal over the time course with a weak heterogeneous enhancement of the mass and rapid washout starting 12s after contrast administration(C-D). After the patient signed the informed consent form, a pelvic tumor biopsy with ultrasonic guidance was carried out through the vagina with local anesthesia. Endometriosis was confirmed with the pathology of the puncture biopsy. After a thorough explanation, the patient finally received a gynecological-surgical laparotomy and enterostomy, and histopathology indicated endometriosis invading the mucous and muscular layers of the rectal wall.

A 45-year-old woman (case 5) was referred to our hospital for bladder neoplasm detected during a routine ultrasound in periodic health examination without specific symptoms such as bladder irritation sign or hematuresis. Except for a mild increase in CA125 (56.0 U/ml), the other laboratory data were all within the normal range. Cystitis glandularis was suspected after conducting a cystoscopy. TVS showed a 38×15×32 mm heterogeneous mass in the dome of the bladder with an unclear boundary(A). A small amount of dotted blood flow signal could be detected by CDFI(B). Next, an intravenous contrast agent was administered after determining the targeted imaging plane of the lesion. The lesion had a slightly enhanced and mild washout that began nine seconds after contrast administration(E-H). Linearized time-intensity curve of perfusion-echo-power signal over the time course with a weak homogenous enhancement of the mass and mild washout starting 9 s after injection of contrast agents(C-D). After fully informing the patient, she underwent transurethral resection of the bladder tumor (TURBT), and histopathology revealed endometriosis infiltrating the urinary bladder wall.
DE is relatively uncommon in endometriosis patients, and ectopic endometrium can occur almost anywhere in the abdomen [15]. However, It is unfortunate that DE does not receive sufficient attention and can easily be misdiagnosed in clinical practice. There are a small number of studies describing the image pattern of DE in CEUS. All were in the form of case report or cases reported in literatures [9, 16–18]. Table 4 summarizes the four published reports of five DE patients that were conducted CEUS in the literature review. Including our seven cases in Table 1, 12 cases have been reported (urinary bladder, 3; ureter,2; colon, 2; renal pelvis, 1; POD, 1; rectum, 1; small intestine,2).
Summarized clinical, conventional Transabdominal US and CEUS characteristics of patients with DE confirmed pathologically in literature review
Summarized clinical, conventional Transabdominal US and CEUS characteristics of patients with DE confirmed pathologically in literature review
NS, not specified; DC, descending colon.
Classical symptoms of endometriosis, such as pelvic or abdominal pain(5), dysmenorrhea(1), and Hematochezia(2), could be observed in these 8 of the 12 cases. However, three DE in the urinary bladder were discovered incidentally without symptoms like dysuria and cyclical hematuria. Laboratory tests showed no abnormalities in tumor marker except that the level of CA125 was elevated slightly in our 7 patients, measuring 44.4±8.6U/ml (upper limit of normal: 35 U/ml), similar to the clinical guideline [19].
TVS has been recommended as the first-line diagnostic tool for patients with suspected DE. It may be beneficial to perform an additional transabdominal ultrasound(TAB) examination in cases of extra pelvic disease, extensive findings, or limited transvaginal access [19]. Consequently, To describe the ultrasound imaging pattern of DE completely, We reviewed both TAB and TVS cases in our institute. In the present study and the included cases, BMUS characteristics of DE are typically demonstrated as a solid heterogeneous(predominantly hypoechoic) or hypoechoic mass with an unclear or clear margin dependent on the infiltrated organs. The BMUS results of the present study are similar to the previous consensus literature [20]. BMUS appearance of DE lesions is different depending on its anatomical location.
To achieve a uniform evaluation of patients with suspected DE, the IDEA group suggested reporting the findings by pelvic compartments. The anterior compartment is composed of the urinary bladder, uterovesical region, and ureters, while the posterior compartment contains the uterosacral ligaments(USLs), posterior vaginal fornix, anterior rectum, anterior rectosigmoid junction, and the sigmoid colon. BMUS features of anterior compartment DE lesions: lesions in the bladder can present as hypoechoic, linear, or spherical structures, which may affect the muscularis or (sub)mucosal layers. These lesions can display either regular or irregular contours. The ureters infiltrated by DE lesions appear as long tubular hypoechoic structures with a thick hyperechoic mantle extending from the lateral aspect of the bladder base to the common iliac vessels. Endometriosis causes dilatation of the ureter due to strictures. Posterior compartment DE lesions appear as hypoechoic thickening of the bowel or vagina wall or as hypoechoic solid nodules, which may vary in size and have smooth or irregular contours. The comet sign, moose antler sign, or pulling sleeve sign can be seen in bowel DE [21].
According to a Cochrane review, TVS has a high diagnostic accuracy for diagnosing DE (79% in mean sensitivity and 94% in mean specificity) [22]. However, In clinical practice, it is not easy for even a sophisticated examiner to differentiate DE from other diseases, like bladder and colon cancer. Therefore, more imaging features are needed for differential diagnosis, Although Color Doppler has been recommended for the differential diagnosis between DE and bowel or rectal cancer [20]. Compared to Doppler US, microbubble can visualize vessels with smaller diameters and blood flow at lower velocities. The use of Sonovue with a low MI enables the display of simple backscattering sound waves associated with microbubbles, with minimal bubble destruction, thereby permitting a prolonged evaluation of the hemodynamic distribution of the contrast agent in real time [23].
CEUS is a better modality to evaluate the vascularity of DE lesions compared to Color Doppler. As a result, we introduced CEUS to further evaluated the enhanced pattern and level. Most DE lesions show hypo- or iso-enhancement in the arterial phase and venous phase. This enhanced pattern in the arterial phase of CEUS may suggest a relatively weak blood supply in DE lesions. However, 2 cases in the sigmoid colon and proximal ureter in previous literature [9, 18] showed hyperenhancement in the arterial phase. This hyperenhancement may result from the muscularis layer infiltrated, which is related to reactive inflammation of the DE lesion and surrounding area, including inflammatory activation of blood vessels, the proliferation of smooth muscle cells, and fibrosis adhesions [24]. This discrepancy may suggest that the enhanced level could somewhat reflect the infiltration depth.
To describe the enhanced pattern in more detail, we applied vuebox® software to quantify the CEUS results. The Time-intensity curve(TIC) results suggested that the DE lesion’s peak enhancement was less than the adjacent structure. In addition, the lesions with a rapid wash-in rate showed a rapid washout rate, and lesions with a slow wash-in rate had a slow washout rate. CEUS characterized the lesion based on its microvessel density, which was highly correlated with its malignancy level. CEUS has a high sensitivity to detect low-speed blood flow and small vessels, which can be used for real-time dynamic observation of microcirculation perfusion [11]. Most of the abdomen and pelvis malignancies are present as hyperenhanced occupied lesions. For example, Bladder malignant tumors showed a fast-into and slow-out and hyperenhanced imaging mode compared to their bladder wall tissue, while rectal adenocarcinomas displayed noted and heterogeneous signal enhancement after contrast administration [25, 26]. Therefore, these enhanced patterns and quantified contrast-enhanced ultrasound features may provide a trend to differentiate DE from other gastrointestinal tract and urinary tract diseases.
Our study has several limitations. (1) The sample size of our study was only 7, and larger samples will be required for future studies. (2) our study lacked statistical analysis comparing CEUS with other enhanced imaging examinations like contrast-enhanced MRI. It will be necessary to conduct a prospective study with a larger sample size in the future.
Our results and the cases abstract from previous articles suggested that CEUS may be a promising and convenient non-invasive diagnostic tool for DE.
Footnotes
Acknowledgments
This work is supported by the National Natural Science Foundation of China(Grant No.82060320, Grant No.82260348) and the Natural Science Foundation of the Xinjiang Uygur Autonomous Region(Grant No.2021D01C009).
Conflict of interest disclosure
No conflict of interest exists in this article.
