Comparison of three noninvasive quantitative MRI-based methods for estimating liver steatosis

Introduction: Magnetic resonance imaging-proton density fat fraction (MRI-PDFF) has been validated for diagnosing liver steatosis.However, comparisons of MRI-based methods have not been sufficiently addressed.This study compared three noninvasive quantitative MRI methods for evaluating liver steatosis and assessed the correlations between body mass index (BMI) and abdominal circumference with liver fat in Mexican patients.Material and methods: This cross-sectional study included adults without a prior clinical or laboratory diagnosis of liver disease.Liver fat was quantified using LiverLab software with three MRI-based methods: automatic MRI-PDFF segmentation of the entire liver, manual MRI-PDFF ROI, and magnetic resonance spectroscopy of fat fraction (MRS-FF).Results: Forty-one participants with a mean age of 40.6 13.7 years were included; 21 (51.2%) were men, and 20 (48.8%) were women.The mean BMI was 27.1 4.6 kg/m.The prevalence of hepatic steatosis was 43.9% (n = 18).The median liver steatosis for automatic MRI-PDFF segmentation was 5.0% (IQR 6.4%); for MRI-PDFF ROI, 2.9% (IQR 6.9%), and for MRS-FF, 5.7% (IQR 8.4%), with a significant difference (p = 0.006).Liver volume was positively correlated with BMI (r = 0.50; p < 0.001); a higher BMI was associated with greater liver volume.Abdominal circumference was positively correlated with the liver fat fraction (r = 0.46, p = 0.001).Conclusion: Automated MRI-PDFF segmentation and MRS-FF showed comparable results for quantifying liver fat, while MRI-PDFF ROI yielded significantly lower values.Liver steatosis correlated directly with higher BMI and larger abdominal circumference.This is the first study in Mexican patients to report liver fat estimation using three MRI-based methods.