医学
子宫内膜异位症
阶段(地层学)
一致性
盆腔疼痛
腹腔镜检查
回顾性队列研究
恶性肿瘤
妇科
外科
内科学
生物
古生物学
作者
Maurício Simões Abrão,Marina Paula Andres,Julian A. Gingold,Mariona Rius,João Siufi Neto,Manoel Orlando Gonçalves,A. Di Giovanni,Mario Malzoni,Francisco Carmona
标识
DOI:10.1016/j.jmig.2022.11.003
摘要
To compare the accuracy of preoperative ultrasound (US) in predicting the laparoscopically defined 2021 American Association of Gynecologic Laparoscopists (AAGL) Endometriosis Staging.Retrospective multicenter study of patients treated at 3 specialized endometriosis centers.Three specialized endometriosis surgical centers in São Paulo (Brazil), Barcelona (Spain), and Avellino (Italy) participated.A total of 878 patients aged 15 to 45 years with no history of pelvic malignancy underwent laparoscopic (LPS) treatment for suspected endometriosis.Retrospective review of preoperative transvaginal and transabdominal US (index test) assessed for endometriosis at all sites used in the 2021 AAGL Endometriosis Classification and classified patients into AAGL-US stages 1 to 4. Results were compared with reference-standard LPS (AAGL-LPS) staging.The AAGL-US and AAGL-LPS stage were concordant in 586 cases (66.7%) (weighted kappa [WK] 0.759; intraclass correlation = 0.906), with the highest agreement observed in patients with no endometriosis (n = 70, 75.3% concordance), AAGL-LPS stage 1 (104, 50.7%) and stage 4 disease (358, 88.2%). Endometriosis was most accurately diagnosed in the rectum/sigmoid colon (WK 0.862), bladder (WK 0.911), and ovaries (WK 0.835/0.795 for right/left, respectively) and least accurately diagnosed at superficial peritoneal (WK 0.442), tubal (WK 0.391/0.363 for right/left, respectively), and retrocervical/uterosacral ligament (WK 0.656) sites.Sonographic estimation of the 2021 AAGL Endometriosis Staging is greatest in AAGL-LPS stages 1 and 4 and among patients with no endometriosis. US best identifies endometriosis of the ovaries, bladder, and bowel but is more limited for the tubes and superficial peritoneum.
科研通智能强力驱动
Strongly Powered by AbleSci AI