Evaluation of a multimodality MR/US coregistration system for investigation of hepatic metastases from colorectal cancer.

Purpose: To evaluate a magnetic resonance (MR)/ultrasound (US) coregistration system with US used in follow-up diagnostic studies of lesions originally identified by MR imaging.

Materials And Methods: A single-center prospective study enrolled 21 consecutive patients (age, 64.0 y±7.5; eight men [age, 63.0 y±7.1] and 13 women [age, 65.0 y±7.3]) evaluated for potential surgical resection of liver metastases. Each patient underwent same-day MR examination and two US examinations: one regular and one with the MR/US coregistration system. Target lesions were identified on MR imaging, and US was used in follow-up diagnostic studies of lesions originally identified by MR imaging. During US, two outcome measures (target localization success and target localization time) were collected. Ratios of lesions found per patient were compared with a paired Wilcoxon test, and a Student t test was used to compare target localization time.

Results: Ratios of lesions found per patient with the coregistered system (93.7%) and conventional US (73.3%) were statistically different (P =.003). Lesions found by the coregistration system but not conventional US were statistically smaller than those found by both systems (6.1 mm±4.0 vs 15.6 mm±9.8; P<.0006). There was no statistical difference in target localization time for detection of lesions found by coregistered (100 s±89) and conventional US (108 s±86; P = .78).

Conclusions: The MR/US coregistration system increases the number of lesions detected with US versus standard US alone. This may prove beneficial in surgical and nonsurgical management of patients with focal liver lesions.