Breath-hold T2-weighted MRI of the liver at 3 T using the BLADE technique: impact upon image quality and lesion detection

AimTo compare image quality and lesion detection in the liver using magnetic resonance imaging (MRI) at 3 T between T2-weighted imaging using a standard rectilinear k-space trajectory (standard T2WI) and using the BLADE technique (BLADE-T2WI), a technique that employs periodically rotated overlapping parallel lines with enhanced reconstruction for motion correction.Materials and methodsTwenty-eight consecutive patients who underwent MRI examination of the liver at 3 T including standard T2WI and BLADE-T2WI, both performed using multiple breath-holds, comprised the study cohort. Images were reviewed in consensus by two radiologists during separate sessions for a number of measures regarding artefacts and image quality. These two readers also assessed the two image sets for the presence of liver lesions and measured liver-to-lesion contrast. Binary logistic regression for correlated data was used to compare the sequences in terms of sensitivity and positive predictive value (PPV) for lesion detection.ResultsBLADE-T2WI received significantly higher scores than did standard T2WI for in-plane respiratory motion (p = 0.0195), other ghosting artefacts (p < 0.0001), sharpness of the liver edge (p = 0.0004), sharpness of intra-hepatic vessels (p < 0.0001), flow signal suppression (p < 0.0001), and overall image quality (p < 0.0001). There was a non-significant trend toward improved B1-inhomogeneity artefact with BLADE-T2WI (p = 0.0571). There was no difference in through-plane respiratory motion (p = 0.6836). BLADE-T2WI demonstrated a significant improvement in PPV for lesion detection (p = 0.0129) as well as in liver-to-lesion contrast (p = 0.0054). There was no difference regarding lesion sensitivity (p = 1.0).ConclusionsUse of the BLADE technique for T2-weighted MRI of the liver at 3 T may lead to a significant improvement in image artefacts and improved PPV for lesion detection.