Assessment of rectovaginal endometriosis using three‐dimensional gel‐infusion sonovaginography

Deep infiltrating endometriosis (DIE) is characterized by the presence of foci of ectopic endometrial tissue, fibrosis and muscular hyperplasia that have penetrated more than 5 mm into the affected pelvic organs. Rectovaginal endometriosis is considered the most severe form of DIE, defined as infiltration by endometriotic lesions of both the rectum and the posterior vaginal fornix with possible extension to the rectovaginal septum1. The gold standard method for diagnosis of DIE is laparoscopy with biopsy2. However, this approach is inadequate for lesions located in the rectovaginal septum and posterior vaginal fornix because these structures are in the posterior cul-de-sac, below the peritoneal reflection, and sometimes this area is obliterated by the disease and is not visible by laparoscopy. Similarly, conventional transvaginal sonography (TVS) has lower sensitivity for detection of DIE in this area than in other sites3, 4. Therefore, the preoperative diagnosis of rectovaginal DIE can be challenging. Here, we describe an approach to visualize the rectovaginal septum and posterior vaginal fornix using three-dimensional (3D) TVS, which can complement the standard ultrasonographic protocol for evaluation of DIE5. The technique involves infusion of a high-viscosity ultrasonographic gel, free of air bubbles, into the vagina prior to performing 3D-TVS. This creates an anechoic and stable acoustic interface that provides an effective window through which to examine the vaginal fornix, rectovaginal septum and surface of the uterine cervix. After gel infusion, the transvaginal probe is introduced 2–3 cm into the vaginal introitus and 3D volume datasets are obtained. On image post-processing, the whole extent of the rectovaginal septum and vaginal fornix can be visualized in the coronal, sagittal and parasagittal planes. Additionally, using the rendering and multiplanar modes, the posterior vaginal fornix, as well as the parametrium and the cervicovaginal fascia, can be visualized in a coronal plane. The steps involved in this technique are described below. Ultrasound scans were performed using the Voluson E6 equipment (GE Medical Systems, Zipf, Austria). Immediately before the examination, the patient is asked to empty completely her bladder and then drink approximately 800 mL water. A complete routine TVS pelvic examination is performed. By the end of this initial examination, most patients have a partially full bladder. The vagina is infused slowly and carefully with 60–80 mL of a homogeneous and completely translucent anechoic gel (Clinical Gel RMC, Amparo, Brazil) using a 26-g rectal catheter (26-FG) connected directly to the bag. The catheter is introduced deeply into the vagina and then withdrawn slowly while infusing the gel. It is important to continue to inject gel into the vagina while carefully withdrawing the catheter in order to avoid the formation and retention of air and ensure an entirely anechoic acoustic window. Using the cervix as the central reference image, the transvaginal transducer is introduced 2–3 cm into the vagina and the focal zone is adjusted so that the volume box includes part of the bladder (cranial edge), the entire rectovaginal septum and rectum (caudal edge) and the cervix (posterior edge) in the sagittal plane (Figure S1). The following settings are recommended for acquisition of the volume datasets: frequency 6–9 MHz, scanning angle as wide as possible (120° maximum angle), average scanning speed and maximum quality. Using two-dimensional (2D)- and 3D-TVS, the rectovaginal septum and fornix can be visualized in the sagittal and parasagittal views (Figures 1 and S2). Post-processing of the coronal plane, either on the ultrasound machine itself or using the 4D-View program, allows visualization of structures that are difficult to access by conventional TVS, such as the parametria, cervicovaginal fascia, posterior fornix and rectovaginal septum. The presence of DIE in these areas can subsequently be assessed (Figures 2 and 3). The simultaneous partial filling of the urinary bladder and presence of anechoic gel in the vagina creates two acoustic windows that facilitate visualization of the vesicouterine pouch by 3D-TVS. This enables physicians to identify exophytic lesions of the fornix or nodular lesions that distort the vesicouterine pouch, especially using the rendering and multiplanar modes (sectional planes). The technique described here can improve substantially visualization of the posterior (rectovaginal septum, posterior fornix, cervix, retrocervical region, parametrium and cervicovaginal fascia) and the anterior (vesicouterine pouch and bladder) compartments achieved by TVS and consequently the detection of DIE in these areas. It differs from other TVS methods because it involves the use of a high-viscosity gel (free of air bubbles) to create an entirely anechoic and stable acoustic window and uses 3D-TVS to obtain images that provide additional information to those obtained by 2D gel sonovaginography. This technique has the potential to increase the sensitivity and overall accuracy of TVS in the diagnosis of DIE5. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.