Quantification of Hepatic Steatosis with 3-T MR Imaging: Validation in ob/ob Mice

Purpose: To validate quantitative imaging techniques used to detect and measure steatosis with magnetic resonance (MR) imaging in an ob/ob mouse model of hepatic steatosis. Materials and Methods: The internal research animal and resource center approved this study. Twenty-eight male ob/ob mice in progressively increasing age groups underwent imaging and were subsequently sacrificed. Six ob/+ mice served as control animals. Fat fraction imaging was performed with a chemical shift-based water-fat separation method. The following three methods of conventional fat quantification were compared with imaging: lipid extraction and qualitative and quantitative histologic analysis. Fat fraction images were reconstructed with single-and multiple-peak spectral models of fat and with and without correction for T2* effects. Fat fraction measurements obtained with the different reconstruction methods were compared with the three methods of fat quantification, and linear regression analysis and two-sided and two-sample t tests were performed. Results: Lipid extraction and qualitative and quantitative histologic analysis were highly correlated with the results of fat fraction imaging (r(2) = 0.92, 0.87, 0.82, respectively). No significant differences were found between imaging measurements and lipid extraction (P=.06) or quantitative histologic (P=.07) measurements when multiple peaks of fat and T2* correction were included in image reconstruction. Reconstructions in which T2* correction, accurate spectral modeling, or both were excluded yielded lower agreement when compared with the results yielded by other techniques. Imaging measurements correlated particularly well with histologic grades in mice with low fat fractions (intercept, -1.0% +/-1.2 [standard deviation]). Conclusion: MR imaging can be used to accurately quantify fat in vivo in an animal model of hepatic steatosis and may serve as a quantitative biomarker of hepatic steatosis. (C) RSNA, 2010