Transvaginal ultrasound for diagnosis of deep endometriosis involving uterosacral ligaments, torus uterinus and posterior vaginal fornix: prospective study

Autor: Meritxell Gracia, M Valdés-Bango, Marta Tortajada, C. de Guirior, Isabel Matas, M.A. Martínez-Zamora, Francesc Carmona, Mariona Rius, Cristina Ros, E Mension
Rok vydání: 2021
Předmět:
Zdroj: Ultrasound in Obstetrics & Gynecology. 58:926-932
ISSN: 1469-0705
0960-7692
Popis: OBJECTIVE To evaluate the accuracy of transvaginal ultrasound (TVS) in diagnosing deep endometriosis (DE) involving the uterosacral ligaments (USLs), torus uterinus (TU) or posterior vaginal fornix (PVF) in women with suspected endometriosis scheduled for laparoscopic surgery. METHODS In this prospective study, consecutive women with clinically suspected pelvic endometriosis who were scheduled for laparoscopic surgery were invited to participate. TVS was performed before surgery. TVS findings were compared with those obtained at laparoscopy and confirmed histologically. The accuracy of TVS for diagnosing DE involving the USLs, TU or PVF was assessed. Additionally, the association of DE involving the USLs, TU or PVF on TVS with symptoms and with DE affecting other pelvic locations was assessed. RESULTS In total, 172 patients were included. The global sensitivity and specificity of TVS in diagnosing DE affecting the USLs, TU and/or PVF were 92% and 87%, respectively. For DE involving the USLs, the accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio and negative likelihood ratio of TVS were 89.5%, 96.6%, 82.1%, 85.0%, 95.8%, 5.41 and 0.04, respectively; the respective values for DE involving the TU were 86.6%, 83.9%, 89.4%, 89.0%, 84.4%, 7.92 and 0.18, and the respective values for DE involving the PVF were 93.6%, 87.0%, 94.6%, 71.4%, 97.9%, 16.20 and 0.14. Logistic regression analysis showed a significant association between DE affecting the USLs, TU and/or PVF and DE affecting the rectosigmoid (odds ratio, 5.43; P
Databáze: OpenAIRE