Contrast-enhanced double inversion recovery sequence for patients with multiple sclerosis: feasibility of subtraction images between pre- and post-contrast images

Background: Few reports have examined the feasibility of a post-contrast double inversion recovery (DIR) magnetic resonance (MR) sequence in patients with multiple sclerosis (MS) because of partial or complete signal loss of enhancing MS lesions. Purpose: To compare subtracted images of DIR (pre-contrast - post-contrast DIR images) with contrast enhanced T1-weighted (CE-T1W) images in the depiction of contrast enhancement of MS lesions. Material and Methods: In total, 27 patients were included. Two neuroradiologists interpreted both images of CE-T1W imaging and subtracted DIR, and interpretation of the images was classified into a score of 1-5 (from 5, definitely superior contrast of lesions on DIR subtraction compared to conventional CE-T1W imaging, to 1, definitely superior contrast of lesions on CE-T1W imaging. The interrater agreement (kappa coefficient) was measured. The signal-to-noise ratio (SNR) and contrast-noise-ratio (CNR) of the lesion were compared. Results: A significant difference (P < 0.001) in scoring was seen between conventional CE-T1W imaging (2.1 +/- 1.5 with one reviewer and 2.4 +/- 1.5 with the other) and DIR subtraction (4.4 +/- 1.0 with one reviewer and 4.7 +/- 0.8 with the other). SNR from conventional CE-T1W imaging (24.8 +/- 14.7) was significantly superior to that from DIR subtraction (4.0 +/- 1.0; P < 0.001). CNR in DIR subtraction (326.4 +/- 250.0) was significantly superior to that in conventional CE-T1W imaging (0.8 +/- 5.5; P < 0.001). For interrater agreement in the evaluation of contrast enhancement of the lesions, kappa coefficients were 0.84 for conventional CE-T1W imaging and 0.72 for DIR subtraction. Conclusion: Subtracted DIR image enables more obvious contrast enhancement of the MS lesions compared with conventional CE-T1W imaging.