Dual ultrasound combination improves the accuracy of preoperative assessment in rectal endometriosis: a prospective cohort study

Endometriosis is a chronic, inflammatory, estrogen-dependent, and complex gynecological disorder characterized by the presence of endometrial-like tissue outside the uterus[ 1 ]. Deep infiltrating endometriosis (DIE), as the most severe type, can infiltrate the bladder, bowel, and ureters, leading to occlusion and ureterohydronephrosis [ 1 , 2 ] . Bowel endometriosis is the most commonly affected type, with the rectum and recto-sigmoid junction being the predominant sites of involvement in 70%–93% of cases [ 3 , 4 ] . Rectal endometriosis is associated more frequently with dyschezia, painful defecation, rectal bleeding, and even intestinal obstruction. Notably, this condition often presents concomitantly with other subtypes of endometriosis or adenomyosis, which pose challenges for comprehensive disease evaluation[ 5 ]. Surgery is a key treatment for rectal endometriosis, including shaving, disc excision, and colorectal resection [ 6 – 9 ] . Surgery has helped patients alleviate symptoms and improve their quality of life [ 10 , 11 ] . However, the reported benefits are limited by major preoperative and postoperative complications, such as bowel leakage, anastomotic leakage, anastomotic stenosis, rectovaginal fistula, and voiding dysfunction. The complication rate according to shaving, disc excision, and colorectal resection was 2.2%, 9.7%, and 9.9%, respectively[ 12 ]. The variety of techniques highlights the considerable heterogeneity of management, especially concerning the most appropriate technique. This decision depends not only on different factors related to the patient’s desire and condition but also on the surgeon’s experience[ 12 ]. In addition, due to deep nodules frequently affecting the retrocervical region, uterosacral ligaments, vagina, and rectosigmoid, meticulous preoperative assessment of lesions is necessary to optimize surgical strategy[ 3 ]. Transvaginal ultrasound (TVS) is highly accurate in detecting ovarian endometriosis and adenomyosis, which is also often used for diagnosing DIE[ 13 ]. However, diagnosing DIE using TVS does not allow for a detailed assessment of bowel wall involvement. Endorectal ultrasound (ERUS), on the other hand, provides a valuable tool for imaging the rectal wall, in resolving the layers, and facilitating the assessment of the depth of invasion, but its focus is limited to the rectum. The 2022 European Society of Human Reproduction and Embryology guidelines recommend multimodal imaging for endometriosis assessment, advocating the combined use of TVS and Magnetic Resonance Imaging (MRI)[ 14 ]. However, given MRI’s limited accessibility in clinical practice, the integration of TVS with ERUS may offer greater practical advantages due to its convenience and real-time imaging capabilities. Despite its importance, preoperative assessment remains inconsistent across institutions due to a lack of standardized protocols. This study aims to evaluate the efficacy of combining TVS and ERUS in the preoperative assessment of rectal endometriosis. By integrating these two imaging techniques, we seek to improve lesion mapping accuracy, particularly for rectal involvement, which is challenging with TVS alone. Accurate preoperative mapping is crucial for surgical planning, as it allows for better anticipation of disease extent, facilitates multidisciplinary surgical strategies, and minimizes the risk of incomplete resection. This cohort/cross-sectional/case-control study has been reported in line with the Strengthening the reporting of cohort, cross-sectional and case-control studies in surgery (STROCSS) Guidelines.

This was a single-center, prospective cohort study conducted at the Department of Obstetrics and Gynecology, Peking Union Medical College Hospital (PUMCH), involving women suspected of rectal endometriosis. The study was conducted between August 2023 and September 2024. The study was approved by the ethics committee of PUMCH (I-23PJ653) and followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines and Revised STROCSS Guideline [ 15 , 16 ] . The date of Institutional Review Board approval was April 17, 2023. Written informed consent was obtained from all participants before data collection. The clinical trial was registered on the Chinese Clinical Trial Registry ( https://www.chictr.org.cn/ ). The registration number was ChiCTR2300074842. Details of the Method section are described in Supplemental Digital Content Material, available at: http://links.lww.com/MS9/B13 . Inclusion criteria were women between 20 and 45 years old and a clinical diagnosis of rectal endometriosis. The diagnosis of rectal endometriosis, other endometriosis, and adenomyosis adhered to the criteria outlined in the Chinese Guidelines for the Diagnosis and Treatment of Endometriosis (third edition)[ 17 ] (see in the Supplemental Digital Content Material, available at: http://links.lww.com/MS9/B13 ). Exclusion criteria were in Supplemental Digital Content Material, available at: http://links.lww.com/MS9/B13 . HIGHLIGHTS Combined endorectal ultrasound (ERUS) and transvaginal ultrasound (TVS) improves diagnostic accuracy in rectal endometriosis. Dual ultrasound combination enables more precise lesion mapping before surgery. Preoperative ERUS + TVS facilitates personalized surgical planning in rectal endometriosis. Enhanced imaging strategy optimizes management of rectal endometriosis. Combined endorectal ultrasound (ERUS) and transvaginal ultrasound (TVS) improves diagnostic accuracy in rectal endometriosis. Dual ultrasound combination enables more precise lesion mapping before surgery. Preoperative ERUS + TVS facilitates personalized surgical planning in rectal endometriosis. Enhanced imaging strategy optimizes management of rectal endometriosis. Demographic characteristics of the patients were recorded at the time of inclusion in the study. All patients underwent TVS and ERUS within 1 month of enrollment at our institution. Surgery was performed within 3 months after the TVS and ERUS examination. Before ERUS examination, an oral laxative should be administered a few hours before the examination. The ERUS and TVS examinations were performed independently by two separate ultrasound physicians (G. Zhong and Q.D.). Only two-dimensional images were obtained during the study. The procedure was performed in a standard fashion based on the guidelines of the International Deep Endometriosis Analysis group[ 18 ]. The presence of rectal endometriosis was defined as endometriotic lesions invading at least the intestinal muscular layer[ 18 ]. Standardized patient symptom data and questionnaires were collected at baseline. The questionnaires were also assessed at the postoperative 3-month follow-up. Pain symptom was assessed through a numeric rating scale. Digestive symptoms were assessed through the Gastrointestinal Quality of Life Index (GIQLI)[ 19 ]. GIQLI was assessed at baseline, postoperative 3-month follow-up. Endometriosis Health Profile-30 (EHP-30, where 0 is the best score and 100 is the worst score) is a validated condition-specific tool to assess the impact on quality of life of endometriosis ( https://innovation.ox.ac.uk/outcome-measures/endometriosis-health-profile-ehp/ )[ 18 ]. EHP-30 was assessed at baseline and at the postoperative 3-month follow-up. The surgical procedures were performed by two experienced gynecologists (J.L., Y.D.), with two experienced colorectal surgeons (Y.X., G. Zhang) assisting when bowel surgery was required. The surgeons were informed of the findings from both TVS and ERUS examinations. All patients underwent multidisciplinary team consultation, where surgical approaches were decided according to TVS and ERUS. Rectosigmoid endometriotic nodules were removed using one of three techniques: shaving, disc resection, or colorectal resection. All the perioperative data were documented. Postoperative complications were documented until 42 days post-procedure according to the Clavien-Dindo classification[ 20 ]. Postoperative complications were defined as any event that represents a deviation in the expected postoperative course[ 20 ]. Histological confirmation was evaluated in a standard way in the Department of Pathology, PUMCH. Clinicopathological details were depicted using mean (standard deviation) and median (interquartile range [IQR]) for continuous variables, whereas totals and frequencies were utilized for categorical variables. Parametric and nonparametric data, with the t -test and Mann–Whitney U test. Statistically significant differences were defined as those with P value of 0.05. Statistical analyses were performed using SPSS (version 26.0; IBM Corp., Armonk, NY, USA).

The study recruited 33 patients between August 2023 and September 2024. Of these, two declined participation, six did not undergo surgery, and two were lost to follow-up. Therefore, 23 patients were included (Fig. 1 ). A total of 87.0% of patients suffered from dysmenorrhea, while 13.0% reported chronic pelvic pain (Table 1 ). Regarding digestive symptoms, 43.5% of patients experienced increased bowel movement frequency during menstruation, while 17.4% had hematochezia. Additionally, 8.7% reported dyschezia, and 13% experienced painful defecation, pencil-thin stools, and constipation. One patient developed intestinal obstruction. Over half of the patients ( n = 12, 52.2%) were diagnosed with adenomyosis. Furthermore, ovarian endometrioma was present in 15 (65.2%) patients, ureteral endometriosis in three patients (13.0%), and bladder endometriosis in one patient (4.3%) (Table 1 ). Serum CA-125 and CA-199 levels had median values of 36.7 U/L (IQR: 21.6–73.0) and 16.0 U/L (IQR: 9.9–23.7), respectively. Figure 1. Flowchart of study design. TVS, transvaginal ultrasound; ERUS, endorectal ultrasound. Table 1 Clinical and demographic characteristics Characteristics Total population n = 23 (%) Age, years (IQR) 40.0 (37.0–43.0) BMI, kg/m 2 (IQR) 20.4 (19.3–20.0) Previous cesarean section 9/23 (39.1) Previous surgery for endometriosis 3/23 (13.0) Gravidity (mean ± SD) 1.0 ± 1.0 Parity (mean ± SD) 0.7 ± 0.3 Symptoms  Dysmenorrhea 20/23 (87.0)  NRS for dysmenorrhea 7.2 ± 2.5  Chronic pelvic pain 3/23 (13.0)  NRS for chronic pelvic pain 4.5 ± 1.3  Dyschezia 2/23 (8.7)  Painful defecation 3/23 (13.0)  Increased bowel movement frequency during menstruation 10/23 (43.5)  Pencil-thin stools 3/23 (13.0)  Constipation 3/23 (13.0)  Hematochezia 4/23 (17.4)  Intestinal obstruction 1/23 (4.3) Coexistence with adenomyosis or endometriosis (pathologically confirmed)  Adenomyosis 12/23 (52.2)  Ovarian endometriosis 15/23 (65.2)  Unilateral 7/23 (30.4)  Bilateral 8/23 (34.8)  Ureteral endometriosis 3/23 (13.0)  Unilateral 3/23 (13.0)  Bilateral 0/23 (0.0)  Bladder endometriosis 1/23 (4.3)  Abdominal wall endometriosis 0/23 (0.0) Preoperative characteristics  Preoperative anemia 2/23 (8.7)  Preoperative HGB, g/L (IQR) 126.0 (120.0–140.0)  Preoperative CA-125, U/L (IQR) 36.7 (21.6–73.0)  Preoperative CA-199, U/L (IQR) 16.0 (9.9–23.7)  Preoperative treatment 23/23 (100.0)  GnRHa agonists 23/23 (100.0) Flowchart of study design. TVS, transvaginal ultrasound; ERUS, endorectal ultrasound. Clinical and demographic characteristics All patients underwent TVS and ERUS examinations at baseline before surgery. ERUS identified luminal stenosis in two patients. Additionally, 82.6% of patients had a single rectal nodule, while 17.4% had two rectal nodules. All rectal nodules were located on the anterior rectal wall, with a mean proximal edge distance of 10.4 ± 1.9 cm and a distal edge of 8.2 ± 2.2 cm from the anal verge. All nodules were hypoechoic, with 56.5% presenting as exophytic and 65.2% as triangular-shaped. The mean nodule length and width were 2.1 ± 0.7 and 1.1 ± 0.6 cm, respectively, involving an average of 20.5 ± 12.4% of the rectal circumference and a mean cross-sectional area of 1.3 ± 1.1 cm 2 . Regarding the depth of invasion, 95.7% of patients extended into the perirectal fat, 100.0% involved the muscular layer, while 30.4% and 17.4% infiltrated the submucosa and mucosa, respectively (Table 2 ). A representative ERUS image is shown in Fig. 2 A, depicting an exophytic, triangular-shaped lesion infiltrating the perirectal fat, muscularis, and submucosa. Compared to ERUS, TVS identified 60.9% of patients with single rectal nodules. Rectal nodules had a mean length of 2.7 cm (range: 2.1–3.2 cm), width of 1.7 cm (range: 1.1–2.0 cm), and thickness of 1.1 cm (range: 0.8–1.8 cm). Ovarian endometriosis was detected in 47.8% (11/23) of patients, with six patients having unilateral involvement and five patients having bilateral involvement. Additionally, adenomyosis was present in 47.8% (11/23) of patients (Table 2 ). Typical images were presented in Fig. 2 B and C, showing a hypoechoic lesion posterior to the uterus and a uterine condition. Figure 2. Endorectal ultrasound, intraoperative image, gross specimens, and microscopic findings. (A) Endorectal ultrasound revealed a triangular-shaped, hypoechoic rectal nodule with exophytic infiltration into the perirectal fat, muscular layer, submucosa, and mucosa (white arrow and yellow dotted line, area: 1.81 cm 2 ). (B) Transvaginal ultrasound identified a hypoechoic lesion posterior to the uterus (white arrow and yellow dotted line, dimensions: 3.92 cm × 2.07 cm). (C) Transvaginal ultrasound measured the uterus at 4.7 × 3.9 × 3.0 cm, with an endometrial thickness of 0.7 cm. The myometrium exhibited homogeneous echogenicity. (D) Rectal lesion and total Douglas pouch obliteration when laparoscopic exploration (Yellow star: rectal lesion). (E) Rectal disc excision and bowel adhesion lysis. (F) Postoperative uterus, bilateral adnexa, and rectum. (G) Lesion of rectal endometriosis during the surgical procedure. (H) Gross specimen of rectal endometriosis. (I) Ectopic endometrial glands and stroma (black arrow, HE staining, original magnification ×40). Table 2 Preoperative endorectal ultrasound (ERUS) and transvaginal ultrasound (TVS) Endorectal ultrasound n (%) Luminal stenosis 2/23 (8.7) Rectal nodules (up to 15 cm from the anal verge)  1 19/23 (82.6)  2 4/23 (17.4) Anterior wall of rectal 23/23 (100.0) Posterior wall of rectal 0/23 (0.0) Distance from the anal verge  Proximal edge (mean ± SD) 10.4 ± 1.9  Distal edge (mean ± SD) 8.2 ± 2.2 Echogenicity  Hypoechoic 23/23 (100.0)  Heterogeneous 0/23 (0.0)  Hyperechoic 0/23 (0.0)  Exophytic lesion 13/23 (56.5)  Triangular-shaped lesion 15/23 (65.2)  Length of rectal nodule, cm (mean ± SD) 2.1 ± 0.7  Width of rectal nodule, cm (mean ± SD) 1.1 ± 0.6  Percentage of circumference involved, % (mean ± SD) 20.5 ± 12.4  Cross-sectional area, cm 2 (mean ± SD) 1.3 ± 1.1 Depth of the rectal nodules Perirectal fat 22/23 (95.7) Muscular layer 23/23 (100.0) Submucosa 7/23 (30.4) Mucosa 4/23 (17.4) Transvaginal ultrasound  Rectal nodule 14/23 (60.9)  1 14/14 (100.0)  2 0/14 (0.0)  Length of rectal nodule, cm (mean ± SD) 2.7 (2.1–3.2)  Width of rectal nodule, cm (mean ± SD) 1.7 (1.1–2.0)  Thickness of rectal nodule (mean ± SD) 1.1 (0.8–1.8)  Ovarian endometriosis 11/23 (47.8)  Unilateral 6/23 (26.1)  Bilateral 5/23 (21.7)  Adenomyosis 11/23 (47.8) Endorectal ultrasound, intraoperative image, gross specimens, and microscopic findings. (A) Endorectal ultrasound revealed a triangular-shaped, hypoechoic rectal nodule with exophytic infiltration into the perirectal fat, muscular layer, submucosa, and mucosa (white arrow and yellow dotted line, area: 1.81 cm 2 ). (B) Transvaginal ultrasound identified a hypoechoic lesion posterior to the uterus (white arrow and yellow dotted line, dimensions: 3.92 cm × 2.07 cm). (C) Transvaginal ultrasound measured the uterus at 4.7 × 3.9 × 3.0 cm, with an endometrial thickness of 0.7 cm. The myometrium exhibited homogeneous echogenicity. (D) Rectal lesion and total Douglas pouch obliteration when laparoscopic exploration (Yellow star: rectal lesion). (E) Rectal disc excision and bowel adhesion lysis. (F) Postoperative uterus, bilateral adnexa, and rectum. (G) Lesion of rectal endometriosis during the surgical procedure. (H) Gross specimen of rectal endometriosis. (I) Ectopic endometrial glands and stroma (black arrow, HE staining, original magnification ×40). Preoperative endorectal ultrasound (ERUS) and transvaginal ultrasound (TVS) All 23 patients underwent laparoscopic surgery. Additional procedures included hysterectomy (21.7%), myomectomy (47.8%), adenomyomectomy (8.7%), ovarian cystectomy (60.9%), unilateral and bilateral ovarian involvement in both 30.4% and unilateral salpingo-oophorectomy (8.7%). A total of 82.6% of patients underwent uterosacral ligament resection, with 43.5% of cases having the procedure performed unilaterally and 39.1% bilaterally. Total Douglas pouch obliteration was found in 91.3% of cases. All rectal nodules were located on the anterior rectal wall (100.0%), with 91.3% of patients having a single rectal nodule. Compared to ERUS, which detected multifocal lesions in four patients, two patients were confirmed to have multifocal lesions during surgical procedures. The mean rectal nodule length was 2.7 ± 0.9 cm, and the width was 1.9 ± 0.9 cm, similar to TVS results. The distance from the anal verge was 8.2 ± 1.6 cm, similar to the ERUS results. Rectal wall invasion was observed in 73.9% of cases involving the mucosa and 26.1% involving the muscular layer. Sigmoid nodules were present in 17.4% of cases. Vaginal infiltration was observed in 13.0%, and rectovaginal septum nodule excision was performed in 47.8% of cases (Table 3 ). Table 3 Intraoperative findings, postoperative surgical complications, and pathological findings Findings Total population n = 23 (%) Surgical procedure Operative route  Laparoscopy 23/23 (100.0)  Laparotomy 0/23 (100.0)  Hysterectomy 5/23 (21.7)  Myomectomy 11/23 (47.8)  Adenomyomectomy 2/23 (8.7) Ovarian cystectomy  No 9/23 (39.1)  Unilateral 7/23 (30.4)  Bilateral 7/23 (30.4) Salpingo-oophorectomy  No 21/23 (91.3)  Unilateral 2/23 (8.7)  Bilateral 0/23 (0.0) Uterosacral ligament resection  No 4/23 (17.4)  Unilateral 10/23 (43.5)  Bilateral 9/23 (39.1) Douglas pouch obliteration  No 1/23 (4.3)  Partial 1/23 (4.3)  Total 21/23 (91.3) Location of rectal nodule  Anterior wall 23/23 (100.0)  Posterior wall 0/23 (0.0) Number of rectal nodule  1 21/23 (91.3)  2 2/23 (8.7) Number of sigmoid nodule  0 19/23 (82.6)  1 4/23 (17.4)  Length of rectal nodule, cm (mean ± SD) 2.7 ± 0.9  Width of rectal nodule, cm (mean ± SD) 1.9 ± 0.9  Distance from the anal verge, cm (mean ± SD) 8.2 ± 1.6 Depth of rectal wall involvement  Muscular layer 6/23 (26.1)  Mucosa 17/23 (73.9)  Vaginal infiltration requiring concomitant vaginal excision 3/13 (13.0)  Rectovaginal septum nodule excision 11/23 (47.8) Surgical procedures on digestive tract  Rectal shaving 12/23 (52.2)  Rectal disc excision 8/23 (34.8)  Rectal segmental resection 3/23 (13.0)  Sigmoid colon shaving 4/23 (17.4)  Sigmoid colon disc excision 0/23 (0.0)  Sigmoid colon resection 0/23 (0.0)  Resection of right/transverse colon 0/23 (0.0)  Appendectomy 0/23 (0.0)  Resection of small bowel 0/23 (0.0) Temporary diverting stoma  Colostomy 1/23 (4.3)  Ileostomy 0/23 (0.0) Surgical procedure on urinary tract  Preoperative D-J stent placement 3/23 (13.0)  Resection of the bladder 1/23 (4.3)  Advanced ureterolysis 20/23 (87.0)  Ureteral resection and ureterocystostomy 2/23 (8.7)  Operative time, min (mean ± SD) 189.5 ± 66.9  Blood loss, ml (mean ± SD) 43.9 ± 22.7  Hospital stay 13.6 ± 2.2  Tmax after surgery 37.6 ± 0.6  Catheter removed on postoperative day (IQR) 1.0(1.0–2.0) Total score rASRM (mean ± SD) 58.2 ± 23.8 rASRM stage  I 1/23 (4.3)  II 1/23 (4.3)  III 1/23 (4.3)  IV 20/23 (87.0) Postoperative complications Clavieno-Dindo complications Grade I  Bladder atony (max 7 days) 2/23 (8.7)  Vaginal bleeding 1/23 (4.3)  Fever 12/23 (52.2)   Clostridium difficile infection 1/23 (4.3) Grade II  Rectal bleeding 1/23 (4.3)  Ileus 0/23 (0.0) Grade III  Anastomotic leakage 0/23 (0.0)  Rectovaginal fistula 0/23 (0.0) Grade IV  Rectovaginal fistula 0/23 (0.0)  Ureteral fistula 0/23 (0.0) Specimen characteristics  Length of specimen 2.4 ± 1.1  Height of rectal nodule 1.7 ± 0.7  Depth of rectal wall involvement 1.0 ± 0.5  Perirectal fat 1/23 (4.3)  Muscular layer 17/23 (73.9)  Submucosa 4/23 (17.4)  Mucosa 1/23 (4.3) Intraoperative findings, postoperative surgical complications, and pathological findings For digestive tract surgery, rectal shaving (52.2%), rectal disc excision (34.8%), rectal segmental resection (13.0%), and sigmoid colon shaving (17.4%) were performed. No sigmoid disc excision or resection was required. A temporary colostomy was performed in one patient (4.3%). For urinary tract procedures, three patients received D-J stent placement. Advanced ureterolysis was performed in 87.0%, while ureteral resection and ureterocystostomy were required in 8.7%. One patient (4.3%) underwent partial bladder resection, consistent with preoperative TVS assessment (Table 3 ). Typical intraoperative images are present in Fig. 2 D – G. The total rASRM score was 58.2 ± 23.8. In terms of rASRM staging, 4.3% of patients were in stage I, 4.3% in stage II, 4.3% in stage III, and 87.0% in stage IV. Regarding complications, no Grade III or IV complications were reported according to the Clavien-Dindo classification system. Only one patient experienced rectal bleeding (Grade II). Most postoperative complications were Grade I, including fever (52.2%), bladder atony (8.7%), and rectal bleeding (4.3%). There were no cases of anastomotic leaks or rectovaginal fistulas (Table 3 ). Twenty-three patients were pathologically confirmed with rectal endometriosis. The rectal specimen characteristics were as follows: the average length of the specimen was 2.4 ± 1.1 cm, with a rectal nodule height of 1.7 ± 0.7 cm and a rectal wall involvement depth of 1.0 ± 0.5 cm. Histological involvement was observed in the following proportions: muscular layer (73.9%, 17/23), submucosa (17.4%, 4/23), mucosa (4.3%, 1/23), and perirectal fat (4.3%, 1/23). The gross specimen and microscopic images are present in Fig. 2 H and I. The sensitivity of ERUS for detecting mucosa and submucosa was found to be 100% (5/5), whereas the specificity was 73.9% (17/23). Table 4 presents the ERUS findings for three different rectal surgical approaches. No cases of luminal stenosis were observed in the rectal shaving ( n = 12) and rectal disc excision ( n = 8) groups (0/3, 0.0%), while rectal segmental resection ( n = 3) showed luminal stenosis in two out of three cases (2/3, 66.7%). Regarding the size of the rectal nodules, the rectal segmental resection group exhibited significantly larger nodule lengths, widths, and cross-sectional areas compared to the other two groups, with means of 3.4 ± 0.5 cm, 2.1 ± 0.8 cm, and 56.8 ± 0.4 cm 2 , respectively. In contrast, the rectal shaving and rectal disc excision groups had smaller nodules, with mean lengths of 2.0 ± 0.3 and 1.7 ± 0.6 cm, widths of 0.9 ± 0.3 and 0.9 ± 0.4 cm, and cross-sectional areas of 1.0 ± 0.4 and 1.2 ± 0.6 cm 2 , respectively. Table 4 ERUS findings in three rectal surgical approaches Rectal shaving ( n = 12) Rectal disc excision ( n = 8) Rectal segmental resection ( n = 3) Luminal stenosis ( n = 3) 0/3 (0.0) 0/3 (0.0) 2/3 (66.7) Length of rectal nodule, cm (mean ± SD) 2.0 ± 0.3 1.7 ± 0.6 3.4 ± 0.5 Width of rectal nodule, cm (mean ± SD) 0.9 ± 0.3 0.9 ± 0.4 2.1 ± 0.8 Percentage of circumference involved, % (mean ± SD) 17.7 ± 6.4 16.6 ± 3.3 5.7 Cross-sectional area, cm 2 (mean ± SD) 1.0 ± 0.4 1.2 ± 0.6 56.8 ± 0.4 Depth of rectal wall involvement  Perirectal fat 12/12 (100.0) 7/8 (87.5) 3/3 (100.0)  Muscular layer 12/12 (100.0) 8/8 (100.0) 3/3 (100.0)  Submucosa 3/12 (25.0) 1/8 (12.5) 3/3 (100.0)  Mucosa 1/12 (8.3) 1/8 (12.5) 2/3 (66.7) ERUS findings in three rectal surgical approaches In terms of nodule depth, all patients had involvement of the perirectal fat and muscular layer, with 100% involvement in both layers for the rectal shaving and rectal disc excision groups. However, there was greater variability in the submucosa and mucosa layers, with the rectal segmental resection group showing higher rates of involvement, with 100% of cases affecting the submucosa and 66.7% affecting the mucosa. Compared to the preoperative assessment (T0), the postoperative 3-month follow-up (T3) showed significant improvements in several areas of the EHP-30. Specifically, pain (49.1 ± 34.0 vs. 16.6 ± 22.4), control and powerlessness (56.4 ± 31.5 vs. 25.3 ± 23.4), and emotional well-being scores (40.6 ± 27.3 vs. 25.8 ± 26.0) all decreased significantly ( P < 0.05). However, the changes in social support and self-image between T0 and T3 were not statistically significant (Fig. 3 A and Supplemental Digital Content Table 1, available at: http://links.lww.com/MS9/B13 ). Figure 3. Change between preoperative and postoperative digestive function and quality of life. (A) Scores of EHP-30 core instrument at T0 and T3. (B) Scores of EHP-30 include six supplementary modules at T0 and T3. (C) GIQLI score at T0 and T3. T0, preoperative assessment; T3, postoperative 3-month follow-up; EHP, endometriosis health profile; GIQLI, Gastrointestinal Quality of Life Index. Change between preoperative and postoperative digestive function and quality of life. (A) Scores of EHP-30 core instrument at T0 and T3. (B) Scores of EHP-30 include six supplementary modules at T0 and T3. (C) GIQLI score at T0 and T3. T0, preoperative assessment; T3, postoperative 3-month follow-up; EHP, endometriosis health profile; GIQLI, Gastrointestinal Quality of Life Index. Regarding the six supplementary modules in EHP-30, significant improvements were observed in work-related issues 45.6 ± 29.3 vs. 16.7 ± 24.2) and sexual relationship quality (52.4 ± 23.6 vs. 33.9 ± 21.1) ( P < 0.05). However, no significant changes were noted in the relationship with child/children, feelings about the medical profession, and treatment. Additionally, data on feelings about infertility were not fully available, as no changes were observed after treatment (Fig. 3 B and Supplemental Digital Content Table 1, available at: http://links.lww.com/MS9/B13 ). The GIQLI score increased from 102.2 ± 20.4 at T0 to 112.1 ± 12.0 at T3, representing a change of 9.6 ± 21.1. However, this change was not statistically significant ( t = −1.731, P = 0.110) (Fig. 3 C and Supplemental Digital Content Table 1, available at: http://links.lww.com/MS9/B13 ).

Rectal endometriosis is a challenging and complex condition. Colorectal surgery for endometriosis has been a topic of debate for years, particularly regarding the risk of major complications such as rectovaginal fistulas, anastomotic leakage, and the possibility of requiring a permanent ostomy. Therefore, the decision to proceed with surgery must be made with great caution. An accurate preoperative evaluation is crucial to assess the full extent of the disease, to determine the best surgical approach, and to weigh the risks and benefits. Ultrasonography is a valuable tool for preoperatively assessing the size and lateral extension of lesions[ 21 ]. Larger lesions are associated with an increased risk of complications following surgery, such as urinary retention. The study first proposes that a dual ultrasound combination (TVS + ERUS) should be incorporated as a routine component of preoperative evaluation. Our study highlights that by utilizing both TVS and ERUS, clinicians can achieve a more comprehensive view of the lesion’s size, location, rectal wall involvement, and other pelvic conditions. TVS demonstrates 75%–98% sensitivity as a first-line tool for detecting DIE, effectively identifying lesions in uterosacral ligaments, rectovaginal space, and rectosigmoid junction[ 3 ]. However, its sensitivity varies by location: ovarian (92%/87%), pouch of Douglas (64%/95%), and rectosigmoid (86%/84%)[ 22 ]. TVS cannot assess mucosal infiltration depth, limiting surgical approach selection (shaving/disc/colorectal resection)[ 23 ]. ERUS provides precise rectal measurements (size, distance from anal verge) and demonstrates 100% sensitivity for mucosa/submucosa detection (specificity 73.9%) in DIE, with strong pathological correlation. While ERUS achieves 76% sensitivity for muscularis propria infiltration[ 24 ], its 2003 data show submucosal detection at 66%. The complementary use of TVS (pelvic-wide assessment) and ERUS (rectal depth evaluation) enhances preoperative mapping. ERUS’s 73.9% specificity suggests occasional overestimation of deeper invasion, necessitating combined interpretation. This multimodal approach optimizes surgical planning by delineating disease extent while avoiding unnecessary aggressive procedures. Our study noted that TVS-detected lesions matched specimen size, while ERUS measurements appeared smaller, likely due to posterior uterine-rectal adhesions causing ERUS assessment limitations. TVS provided a comprehensive pelvic evaluation, including uterine assessment, whereas ERUS was confined to rectal imaging. Surgically confirmed rectal nodules were exclusively anterior wall lesions. Significantly, nearly 50% of rectal endometriosis patients had coexisting adenomyosis, suggesting shared pathophysiology[ 25 ]. This overlap highlights the potential for shared therapeutic strategies and emphasizes the need for further research into the underlying mechanisms linking these two disorders. Colorectal resection for bowel endometriosis risks permanent ostomy [ 26 , 27 ] . Conservative techniques (disc excision/shaving) reduce resection extent and preserve function. Meta-analysis shows rectal shaving lowers rectovaginal fistula and bowel leakage risk versus disc excision/resection, while disc excision reduces stenosis risk[ 12 ]. Our cohort included segmental resection ( n = 12), disc excision ( n = 8), and colorectal resection ( n = 3), with luminal stenosis exclusively in segmental cases. Preoperative luminal evaluation and larger nodules with submucosal invasion in this group indicated the necessity for aggressive resection to avoid obstruction. ERUS precisely identified invasion depth for selecting segmental resection candidates, while limited disease benefited from shaving/disc excision. However, ERUS detection range (≤15 cm from anus) limits the evaluation of upper sigmoid lesions, necessitating MRI and colonoscopy. The preoperative and postoperative EHP-30 and GIQLI assessments revealed significant pain reduction, aligning with treatment goals and improving physical and psychological well-being. Work-related difficulties also decreased, suggesting enhanced social reintegration and reduced socioeconomic burden. GIQLI showed modest digestive function improvement (statistically non-significant), indicating gradual symptom alleviation, possibly requiring extended follow-up or complementary therapies. Alongside improved quality of life and digestive function, postoperative outcomes demonstrated favorable safety: no severe complications (Clavien-Dindo ≥III) occurred. Observed complications included one Grade II rectal bleeding case and five Grade I events (bladder atony, vaginal bleeding, fever, Clostridium difficile infection), all manageable conservatively. These findings support surgical safety for rectal endometriosis. However, limitations include a relatively small sample size limiting generalizability, and lack of long-term follow-up data despite preoperative imaging (TVS/ERUS) guiding surgical decisions effectively. Direct comparisons between multimodal ultrasound and MRI regarding clinical utility and economic outcomes were absent. Future multicenter studies with larger cohorts and extended follow-up are needed for validation.

In summary, our study demonstrates that preoperative ERUS, in conjunction with TVS, enhances the accuracy of preoperative assessments and, by extension, helps tailor the surgical approach to the individual patient’s disease characteristics. This can significantly improve surgical outcomes by ensuring that the most appropriate technique is chosen based on the extent of the lesion, ultimately leading to better management of rectal endometriosis.