Continuous Hepatic Arterial Multiphase Magnetic Resonance Imaging During Free-Breathing

The aim of this study was to evaluate the feasibility of a prototype volume-interpolated breath-hold examination (VIBE) sequence using compressed sensing (VIBECS) for rapid multiphase arterial magnetic resonance imaging (MRI) at different temporal resolution during free-breathing in comparison with a conventional breath-hold approach (VIBESTD).

A total of 40 patients with liver malignancies were prospectively included in this study and underwent contrast-enhanced liver MRI at 1.5 T to evaluate the performance of VIBECS for rapid arterial multiphase imaging. An additional 40 patients examined with a VIBESTD were included serving as standard of reference. The VIBECS study cohort was subdivided into 2 groups (each n = 20). In both groups, VIBECS was continuously acquired for 60 seconds starting with the contrast agent administration (group A, temporal resolution 4 seconds; group B, temporal resolution 8 seconds). Subsequently, the time point with the subjectively best image quality was selected and defined as hepatic arterial dominant (HAD) phase. Overall image quality, lesion conspicuity, vessel contrast, and artifacts of HAD phase were assessed by 2 radiologists independently on a 5-point Likert scale (5 = excellent) and compared with arterial phase images of VIBESTD. In addition, signal attenuation/time curves of VIBECS were plotted for each patient to quantify the hepatic arterial enhancement.

No patients were excluded and all HAD phases were reliably recorded in the investigated VIBECS cohort. Most commonly, HAD was observed at the ninth time point (36 seconds after intravenous contrast injection) in group A and at the fifth time point (40 seconds after intravenous contrast injection) in group B. Timing with VIBESTD was only adequate in 65% (26/40). Image quality, lesion conspicuity, and vessel contrast were good to excellent without significant differences between both VIBECS groups (P ≥ 0.2) and with significantly higher reading scores as compared with VIBESTD with respect to lesion conspicuity (P ≤ 0.006) and image quality (group B; P < 0.001). VIBECS showed reconstruction artifacts, which were significantly higher in group A (P = 0.001). Mean peak arterial enhancement was observed at the ninth time point (36 seconds) in group A and at the sixth (48 seconds) in group B.

VIBECS allows for robust multiphase arterial imaging during free-breathing at high spatial and temporal resolution (preferably 8 seconds) with improved image quality and lesion conspicuity as compared with VIBESTD.