OC023: Sonographic evaluation of deep pelvic endometriosis: Endovaginal‐, transrectal‐ and vaginosonography to assess the extension of the disease

Magnetic Resonance Imaging (MRI) has emerged to be the best diagnostic imaging to evaluate the extent pelvic deep infiltrating endometriosis (DIE). Transvaginal sonography (TVS) is very accurate in diagnosing ovarian endometriomas but shows some diagnostic limits in the evaluation of the detection and extension of DIE. The aim of this study was to evaluate the accuracy of different sonographic approaches in the assessment of posterior DIE compared to laparoscopical and histological findings. Fifty patients with deep pelvic endometriosis underwent sonographic evaluation prior to laparoscopic surgery, using a transvaginal sectorial probe, first transvaginally (TVS) than transrectally (TRS). Furthermore vaginosonography was associated to the transrectal examination. An accurate mapping of the extension of the disease was performed at sonographic examination and during laparoscopy. The mapping was based on a detailed list of locations which described the endometriotic/fibrotic infiltrating tissue in terms of presence or absence on different sides of the posterior and anterior pelvis: pouch of Douglas, rectovaginal septum (RVS), uterosacral ligaments (USL), vaginal walls, rectal-sigmoid wall and bladder. The sensitivity in the diagnosis of endometriotic nodules of the USL and of the distal rectal and sigmoid wall was similar for TVS alone (88%, 90%) and after TRS and vaginography (89%, 90%). Whereas the accuracy in the diagnosis of rectovaginal nodules, in particular for those infiltrating the RVS and posterior vaginal fornix, improves significantly with TRS and vaginosonography (TVS alone 67% vs. 91%). TVS/TRS and vaginosonography performed by skilled sonographer provide accurate informations about the presence and the extension of pelvic DIE, similar to those reported for MRI. In particular it improves the diagnosis of the infiltration of the rectal wall, RVS and vaginal fornix.