Accurate simultaneous quantification of liver steatosis and iron overload in diffuse liver diseases with MRI

作者：Franca Manuela^*; Alberich Bayarri Angel; Marti Bonmati Luis; Oliveira Pedro; Costa Francisca Emanuel; Porto Graca; Vizcaino Jose Ramon; Sanchez Gonzalez Javier; Ribeiro Eduardo; Oliveira Joao; Miranda Helena Pessegueiro

来源：Abdominal Radiology, 2017, 42(5): 1434-1443.

DOI：10.1007/s00261-017-1048-0

• Fat quantification
• Iron overload
• R2*measurement
• Chronicliver diseases
• Quantitative imaging biomarkers

Purpose: To evaluate the diagnostic performances of 3 Tesla multi-echo chemical shift-encoded gradient echo magnetic resonance (MECSE-MR) imaging to simultaneously quantify liver steatosis and iron overload in a wide spectrum of diffuse liver diseases having biopsy as reference standard.
Methods: MECSE-MR-acquired images were used to calculate fat fraction and iron content in a single breath-hold in 109 adult patients. Proton density fat fraction (PDFF) was prospectively estimated using complex-based data reconstruction with multipeak fat modeling. Water R2* was used to estimate iron content. Biopsy was obtained in all cases, grading liver steatosis, siderosis, inflammation, and fibrosis. Differences in PDFF and R2* values across histopathological grades were analyzed, and ROC curves analyses evaluated the MR diagnostic performance.
Results: Calculated fat fraction measurements showed significant differences (p < 0.001) among steatosis grades, being unaffected by the presence of inflammation or fibrosis (p >= 0.05). A strong correlation was found between fat fraction and steatosis grade (R-S = 0.718, p < 0.001). Iron deposits did not affect fat fraction quantitation (p >= 0.05), except in cases with severe iron overload (grade 4). A strong positive correlation was also observed between R2* measurements and iron grades (R-S = 0.704, p < 0.001). Calculated R2* values were not different across grades of steatosis, inflammation, and fibrosis (p >= 0.05).
Conclusions: A MECSE-MR sequence simultaneously quantifies liver steatosis and siderosis, regardless coexisting liver inflammation or fibrosis, with high accuracy in a wide spectrum of diffuse liver disorders. This sequence can be acquired within a single breath-hold and can be implemented in the routine MR evaluation of the liver.

• 出版日期2017-5