Prediction of early response to uterine artery embolization in fibroids: Value of MR signal intensity ratio

Abstract

Objective: To quantitatively assess magnetic resonance (MR) imaging findings that help predict early post-therapeutic response in fibroids following uterine artery embolization (UAE). Methods: Fifteen patients with a total of 52 fibroids underwent UAE. The signal intensity ratio (SIR) on T1-, T2-, diffusion weighted and gadolinium-enhanced images was calculated by dividing the mean signal intensity of fibroids by that of the abdominal rectus muscle. Fibroids were divided into the two groups: affected (post-UAE volume reduction rate > median of all fibroids) and unaffected (< median rate). The SIRs were compared between the two groups. ROC analysis was used to evaluate the predictive performance for differentiating the affected from unaffected lesions. Results: The SIRs of the affected group were significantly lower on T1-weighted images (0.85 ± 0.1 vs 0.95 ± 0.2) (P = 0.0001), but higher on T2-weighted (1.30 ± 0.6 vs 1.12 ± 0.9) (P = 0.026) and gadolinium-enhanced images (1.51 ± 0.2 vs 1.20 ± 0.4) (P = 0.0002) than those of the unaffected group. There was no significant difference in ADC values between the two groups (P = 0.510). The sensitivity, specificity, and area under the ROC curve (AUC) in the prediction of the affected lesions were 92%, 50%, and 0.712 with SIR on T1-weighted images, and 85%, 62%, and 0.731 with SIR on gadolinium-enhanced images, respectively. Conclusions: The SIRs on T1-weighted images and gadolinium-enhanced images were useful for the prediction of the changes in size of fibroids responding to UAE.