Deep infiltrating endometriosis: CT imaging evaluation

Sung Il Jung, Young Jun Kim, Hae Jeong Jeon, Kyung Ah Jeong

Research output: Contribution to journalArticlepeer-review

15 Scopus citations


Objective: To retrospectively evaluate the feasibility of computed tomography (CT) in depicting deep-infiltrating endometriosis. Materials: The study population included 54 patients (age: mean, 35.5 years; range, 23-48 years) with histologically confirmed ovarian endometriomas between January 2007 and July 2009. All the patients underwent preoperative CT imaging before laparotomy or laparoscopy. The CT images were evaluated for the presence of a tethered appearance of the rectum in the direction of the uterus, stranding of periuterine pelvic fat, thickening of the uterosacral ligament, and retroflexed uterus. Two radiologists performed a blinded and independent review for each CT finding. The sensitivity, the specificity, the positive predictive value, the negative predictive value, and the accuracy of each CT finding and κ statistics were determined. Results: Deep-infiltrating endometriosis was confirmed after surgery and pathologic examination in 34 patients (63.0%). The most specific finding for the diagnosis of deep-infiltrating endometriosis was tethered appearance of rectum in the direction of the uterus (90.0%). The mean sensitivity, specificity, positive predictive value, negative predictive value, and accuracy values of all the CT findings except that of retroflexed uterus were 56.9%, 70.0%, 78.1%, 60.4%, and 61.7%, respectively. The mean κ value was 0.82 (range, 0.67-0.96). Conclusions: Computed tomographic imaging may constitute another potential option as a complementary imaging modality for the evaluation of deep-infiltrating endometriosis.

Original languageEnglish
Pages (from-to)338-342
Number of pages5
JournalJournal of Computer Assisted Tomography
Issue number3
StatePublished - 2010


  • CT
  • Endometriosis


Dive into the research topics of 'Deep infiltrating endometriosis: CT imaging evaluation'. Together they form a unique fingerprint.

Cite this