Fusion imaging for evaluation of deep infiltrating endometriosis: feasibility and preliminary results

ABSTRACT Objective Magnetic resonance imaging ( MRI ) and ultrasound scanning complement each other in screening for and diagnosis of endometriosis. Fusion imaging, also known as real‐time virtual sonography, is a new technique that uses magnetic navigation and computer software for the synchronized display of real‐time ultrasound and multiplanar reconstructed MR images. Our aim was to evaluate the feasibility and ability of fusion imaging to assess the main anatomical sites of deep infiltrating endometriosis ( DIE ) in patients with suspected active endometriosis. Methods This prospective study was conducted over a 1‐month period in patients referred to a trained radiologist for an ultrasound‐based evaluation for endometriosis. Patients with a prior pelvic MRI examination within the past year were offered fusion imaging, in addition to the standard evaluation. All MRI examinations were performed on a 1.5‐T MRI machine equipped with a body phased‐array coil. The MRI protocol included acquisition of at least two fast spin‐echo T2 ‐weighted orthogonal planes. The Digital Imaging Communications in Medicine dataset acquired at the time of the MRI examination was loaded into the fusion system and displayed together with the ultrasound image on the same monitor. The sets of images were then synchronized manually using one plane and one anatomical reference point. The ability of this combined image to identify and assess the main anatomical sites of pelvic endometriosis (uterosacral ligaments, posterior vaginal fornix, rectum, ureters and bladder) was evaluated and compared with that of standard B‐mode ultrasound and MRI . Results Over the study period, 100 patients were referred for ultrasound examination because of endometriosis. Among them were 20 patients (median age, 35 (range, 27–49) years) who had undergone MRI examination within the past year, with a median (range) time interval between MRI and ultrasound examination of 171 (1–350) days. All 20 patients consented to undergo additional evaluation by fusion imaging. However, in three (15%) cases, fusion imaging was not technically possible because of changes since the initial MRI examination resulting from either interval surgery ( n = 2; 10%) or pregnancy ( n = 1; 5%). Data acquisition, matching and fusion imaging were performed in under 10 min in each of the other 17 cases. The overall ability of each technique to identify and assess the main anatomical landmarks of endometriosis was as follows: uterosacral ligaments: ultrasound, 88% (30/34); MRI , 100% (34/34); fusion imaging, 100% (34/34); posterior vaginal fornix: ultrasound, 88% (30/34); MRI , 100% (34/34); fusion imaging, 100% (34/34); rectum: ultrasound, 100% (17/17); MRI , 82.3% (14/17); fusion imaging, 100% (17/17); ureters: ultrasound, 0%; MRI , 100% (34/34); fusion imaging, 100% (34/34); and bladder: ultrasound, 100%; MRI , 100%; fusion imaging, 100%. Conclusion Fusion imaging is feasible for the assessment of endometriotic lesions. Because it combines information from both ultrasound and MRI techniques, fusion imaging allows better identification of the main anatomical sites of DIE and has the potential to improve the performance of ultrasound and MRI examination. Copyright © 2014 ISUOG. Published by John Wiley & Sons Ltd.