LAUSR

Contrast-enhanced body magnetic resonance imaging (MRI) requires high spatial resolution for optimal diagnostic assessment of angiographic and arterial phases with robust fat suppression and minimal motion artifacts. Differential subsampling with Cartesian ordering (DISCO) is a new spatiotemporal resolution dynamic contrast-enhanced MRI technique, which has been recently introduced to capture the enhancement dynamics of tumors and vascular structures at high resolution. First introduced in 2012 by Saranathan et al using phantom experiments and clinical scans of abdominal organs from volunteers and patients, the feasibility of DISCO MRI has now been tested in the breast, pancreas, kidney, and liver. The DISCO pulse sequence combines variable density pseudo-random k-space segmentation, view sharing, a Dixon-based fat–water separation, and parallel imaging. From a technical point of view, the DISCO technique samples an elliptically ordered central k-space region every time and subsamples the outer regions with pseudo-random segmentation to obtain aliasing artifacts from subsample incoherence. A dual-echo spoiled gradient echo (SPGR) acquisition for Dixon-based water–fat separation is required for the use of a pseudo-random k-space segmentation scheme, because, in conventional fat suppression schemes, sequential or linearly segmented k-space is required for optimal performance but cannot be integrated into the DISCO sequence. In DISCO imaging, breath-hold mismatching is reduced by restricting the view sharing to a breath-hold, thereby reducing its temporal footprint. Compared with the T1-weighted 3D SPGR, referred to as a LAVA-Flex sequence, previous studies confirmed the superiority of DISCO toward the reduction of late arterial phase motion and motion artifacts. Above that, DISCO achieves multiple arterial phase acquisitions (MAP) in a single breath-hold period at high spatial resolution. In hepatic MRI, MAP acquisition is preferred to single arterial phase acquisition (SAP) techniques because of its potential to cover the whole arterial phase process including early and late arterial phases. Using MAP acquisition, the motion artifacts scoring of the late arterial phase (P < 0.001) and multiple arterial phases (P < 0.05) are significantly higher compared to the SAP acquisition technique. Whereas early arterial phases, might be important for the evaluation of presurgical hepatic vascular anatomies such as liver resection and liver transplantation, most malignant hepatic tumors are best visualized in the late arterial phase. In a recent study, Wei et al confirmed that for visualization of the hepatic artery and display of small hepatic artery branches, the use of MAP acquisition with DISCO is superior to SAP acquisition with DISCO (P < 0.01). Furthermore, MAP acquisition with DISCO reaches comparable image quality to computed tomography angiography (CTA). Optimal timing for the late arterial phase DISCO imaging is, above that, significantly more often achieved in MAP (95.4%) compared to SAP acquisition (73.1%; P < 0.001), underlining the clinical superiority of the first. Optimal timing of late-phase arterial imaging is important for the detection of hepatic malignant lesions. In a retrospective study of malignant liver nodules, ring-enhancement as a major imaging characteristic of hepatocellular carcinoma was diagnosed significantly more confidently using MAP than SAP acquisition DISCO imaging.