Comparison of routine pelvic US and MR imaging in patients with pathologically confirmed endometriosis.

Abstract

To estimate the benefit of pelvic magnetic resonance (MR) imaging after routine pelvic ultrasound (US) in patients with pathologically or surgically proven endometriosis. Patients with surgically or pathologically proven endometriosis who had routine pelvic US followed by pelvic MR within 6months prior to surgery were included. Patients were excluded if they had previously confirmed endometriosis, pregnancy, or surgery > 6months after MR. The detection rate of endometriosis by pelvic US and MR was compared to the surgical/pathological reference standard. 83 female patients (mean age 40 ± 9) met inclusion criteria and had surgical/pathological confirmation of endometriosis. The mean time interval between pelvic US and MR was 33 ± 43days, with 64 ± 69days between MR examination and surgery. US detected endometriosis in 22% (18/83) of patients compared to 61% (51/83) for MR (p < 0.0001). 51% (33/65) of patients with a negative pelvic US exam had a positive MR. MR identified additional sites or sequela in the majority of patients with a positive US (14/18; 78%), including extraovarian locations [e.g., fallopian tubes 7/18 (39%), uterus 7/18 (39%), uterine ligaments 6/18 (33%), posterior cul de sac 5/18 (28%), pelvic side walls 5/18 (28%), abdominal wall 1/18 (6%)] and sequela [ovarian tethering 5/18 (28%), 6/18(33%) bowel adhesive disease, posterior cul de sac obliteration 2/18 (11%), hydrosalpinx 2/18 (11%), and hydronephrosis 1/18 (6%)]. 3 T MR detected endometriosis in 33/46 (72%) patients compared to 18/37 (49%) for 1.5 T MR (p = 0.03). Pelvic MR imaging had a higher detection rate of surgically/pathologically proven endometriosis and provides more information about disease location and sequela compared to routine pelvic US.