Learning Curve for the Detection of Deep Infiltrating Endometriosis and Adenomyosis with 3-D Transvaginal Water Contrast Sonography.

Three-dimensional rectosonography (RSG) is a transvaginal sonography technique using rectal water contrast and 3-D acquisitions. The main points of interest of 3-D RSG could be its easy accessibility and its quick learning curve, especially with respect to rectosigmoid lesions. The objective of this prospective observational study was to assess the learning curve of 3-D RSG for the diagnosis of rectosigmoid lesions and for various other locations of deep endometriosis (DE), endometriomas and adenomyosis. From April 2017 to November 2017, 116 patients with suspected pelvic endometriosis were referred to the Croix-Rousse University Hospital, and those who underwent 3-D RSG were included in our study. After a short training period, four residents were asked to perform 3-D RSG by themselves. Each procedure was systematically controlled immediately afterward by a single expert sonographer. The success of the procedure involved the correct identification of various locations of endometriosis (rectosigmoid, uterosacral ligament, retrocervical space, vagina, bladder and ovaries) and adenomyosis, using the expert sonographer's examination as the reference technique. The learning curve was generated using these data and assessed using the Learning Curve Cumulative Summation Test (LC-CUSUM) method. The pooled LC-CUSUM revealed that the required level of achievement was reached after 24 3-D RSGs were performed for the diagnosis of rectosigmoid lesions. All four residents were significantly competent in diagnosing rectosigmoid lesions at the end of their training period, with an α risk <0.05 (T1, p = 0.03; T2, p = 0.0002; T3, p = 0.05; T4, p = 0.02). The LC-CUSUM analysis confirmed that competency was achieved for vaginal DE, torus uterinum DE, US DE, bladder DE, endometriomas and adenomyosis within 17, 27, 38, 19, 17 and 33 scans, respectively. This study provides evidence that the skills required to diagnose endometriosis lesions and adenomyosis with 3-D RSG can be acquired after a brief learning period in an expert center.

Comparison of three-dimensional rectosonography, rectal endoscopic sonography and magnetic resonance imaging performances in the diagnosis of rectosigmoid endometriosis.

STUDY OBJECTIVE: Comparison of 3D-rectosonography (3D-RSG), rectal endoscopic sonography (RES), and MRI performances in the diagnosis of rectosigmoid endometriosis using surgery as the Gold Standard. DESIGN: Monocentric retrospective longitudinal study on diagnostic procedures. DESIGN CLASIFICATION: Canadian Task Force II-2. SETTING: University Hospital of Lyon Croix-Rousse. PATIENTS: A total of 37 patients treated surgically for pelvic endometriosis. INTERVENTIONS: Expert 3D-RSG (3D Transvaginal sonography with water contrast in the rectum), MRI and RES performed by expert examiners. Sensitivity, specificity, accuracy, positive and negative predictive value, positive and negative likelihood ratios were calculated. Depth, size, and volume of intestinal lesions were also compared to the type of surgery performed (shaving versus segmental resection). MEASUREMENTS AND MAIN RESULTS: Rectosigmoid endometriosis lesion was confirmed by surgery in 31 patients on 37 (84%). Sensitivity, specificity, accuracy, positive and negative predictive value, positive and negative likelihood ratios for 3D-RSG were 94%, 100%, 95%, 100%, 75%, +∞ and 0.06 respectively; for RES 81%, 100%, 84%, 100%, 50%, +∞ and 0.19 respectively; while for MRI 90%, 100%, 92%, 100%, 67%, +∞ and 010 respectively. There was no significant difference between the 3 procedures (p > 0.05). CONCLUSION: 3D-RSG, RES and MRI seem to be 3 effective procedures in the diagnosis of rectosigmoid endometriosis. Their performances seem equivalent.