Extended Transvaginal Sonography in Deep Infiltrating Endometriosis

Abstract

The purpose of this study was to assess the diagnostic performance of extended transvaginal sonography for diagnosing deep infiltrating endometriosis. A prospective study was conducted comprising 51 women (mean age, 32.9 years; range, 23-43 years) with suspected deep infiltrating endometriosis based on clinical symptoms. All women underwent extended transvaginal sonography, which included assessment of 2 pelvic compartments (anterior compartment: bladder and distal ureters; and posterior compartment: posterior vaginal fornix, retrocervical area, pouch of Douglas, and rectosigmoid). The sliding sign for detecting pouch of Douglas obliteration was also assessed. All patients received bowel preparation before sonographic examinations. A single examiner performed all examinations. All women underwent laparoscopic surgery, and histologic confirmation of endometriosis was done. The sensitivity, specificity, positive likelihood ratio (LR+) and negative likelihood ratio (LR-) were calculated. Some women had more than 1 lesion, giving a total of 55 histologically confirmed lesions (rectosigmoid, n = 13; vagina, n = 5; retrocervical, n = 32; bladder, n = 5). The sensitivity, specificity, and LR+ for rectosigmoid involvement were 100%, 93%, and 14.0, respectively. The sensitivity, specificity, LR+, and LR- for vaginal involvement were 60%, 98%, 30.0, and 0.41. The sensitivity, specificity, LR+, and LR- for retrocervical involvement were 84%, 96%, 19.4, and 0.16. The sensitivity, specificity, and LR- for bladder involvement were 20%, 100%, and 0.80. The sensitivity, specificity, LR+, and LR- of the sliding sign for diagnosing pouch of Douglas obliteration were 89%, 92%, 10.7, and 0.12. Except for bladder involvement, extended transvaginal sonography has good diagnostic performance for deep infiltrating endometriosis.