Effect of steatosis on liver signal and enhancement on multiphasic contrast-enhanced magnetic resonance imaging

Abstract

To investigate the effect of steatosis on liver signal and enhancement in multiphasic contrast-enhanced (MCE) MRI. In this IRB-approved, HIPAA-compliant, retrospective, observational study, 1217 MCE abdominal MRIs performed during 2014 at a single institution were reviewed. Of these, 1085 were excluded, due to potential factors other than steatosis that may affect liver signal intensity and/or enhancement. In the remaining 132, liver fat fraction (FF) was calculated from the in- and opposed-phase 2D T1-weighted images. Liver signal intensity, absolute enhancement, and relative enhancement on fat-suppressed (Dixon method) 3D T1-weighted images before and after injection of gadobutrol (arterial, portal venous, and equilibrium phases) were plotted against co-localized FF values and the linear trend was evaluated by Pearson correlation coefficient (r). P values <0.05 were considered statistically significant. Liver signal intensity negatively correlated with FF for all phases (r=-0.388 to -0.544, p<0.001). Absolute enhancement negatively correlated with FF for the portal venous and equilibrium phases (r=-0.286 and -0.289, respectively, p<0.001), but not for the arterial phase (r=-0.042, p=0.632). Relative enhancement did not significantly correlate with FF for any phase (p≥0.125). Steatosis reduces liver signal intensity in MCE MRI. This effect of steatosis was reduced in calculated absolute enhancement and eliminated in calculated relative enhancement.