Non-invasive quantification of liver fat content by different Gradient Echo MRI sequences in patients with Non-Alcoholic Fatty Liver Disease (NAFLD)

Introduction: Non-invasive quantification of liver fat by Gradient echo (GRE) Technique is an interesting issue in quantitative MRI. Despite the numerous advantages of this technique, fat measurement maybe biased by confounding and effects. The aim of this study was to evaluate the GRE pulse sequences with different   and  weighting for liver fat quantification in patients with Non-Alcoholic Fatty Liver Disease (NAFLD). Materials and Methods: This prospective, cross-sectional study was performed on 30 NAFLD-patients. Sixteen GRE protocols with different   weighting contrast were performed with Siemens 1.5 Tesla with four Echo Times ( . For each sequence, TR or flip angle was changed and other scan parameters were fixed.To assessment of signal intensity changes, Regions of Interest (ROIs) which placed on  images were copied into other images. 48 fat indexes were calculated based on three equations (Equ 1-3). ( )   (1)   (2) (   (3);  =  ln ( Where  is the time difference between  and .Correlation between biopsy findings and MRI were performed using Pearson coefficient at the significant level of 1%. Results:Mean fat indexes which obtained from equation 3 have the maximum values in comparison to the other fat indexes. The maximum fat index was 23.58% which related to heavily weighted contrast pulse sequence (sequence 4 TR=50msec, α= defined by equation 3. The minimum fat index was -2.49% which correspond to the minimal weighted pulse sequence (sequence 13TR=200msec, α=  defined by equation 2.Relationship between calculated fat indexes and flip angle reveals that, these indexes increased with flip angle especially at low angles. Increasing TR parameter decline fat indexes gradually.Correlational analysis between measured fat indexes and biopsy findings indicate that fat indexes calculated by equation 1 and 3 have slightly stronger correlatecpmpared to resulted data from equation 2. Conclusion:In general, the relaxation effects more critical than in fat quantification. Diagnostic accuracy of First dual GRE and Triple GRE techniques are better than Second dual GRE. There was no significant difference between First dual GRE and Triple GRE technique. In fat quantification with GRE techniques maybe the third and fourth echoes are unnecessary. Findings suggest that flip angle could be a major factor, if not the only one, causing the overestimation of measured fat content. Consequently, GRE techniques for fat quantification are more desirable with lower flip angle.