A prospective study of image-based quantification of liver fat

Background: Fatty liver disease involves triglyceride accumulation in hepatocytes, especially among obese and diabetic individuals. Diagnosis relies on a combination of history, examination, and imaging, with biopsy being a definitive but invasive procedure. Aims and Objectives: This study aimed to quantify hepatic steatosis using chemical shift magnetic resonance imaging (MRI) and MR spectroscopy and correlate these findings with the hepatic attenuation index on unenhanced computed tomography (CT). Materials and Methods: This prospective study was conducted on 36 patients at the Barnard Institute of Radiology, Madras Medical College, Chennai, from March 2019 to September 2020. Patient data were collected from case sheets and records. Hepatic fat was quantified using chemical shift MRI with the two-point Dixon technique and single-voxel MR spectroscopy on a 3T system, while correlation with hepatic attenuation index from unenhanced CT was also assessed. Results: The study included 36 patients with a mean age of 45±14 years, predominantly males (69.4%). Common comorbidities were diabetes and fatty liver (16.7% each) and obesity (11.1%). MRI fat percentage negatively correlated with water lipid (WL) ratio (r=−0.385, P=0.02) and CT hepatic index (r=−0.903, P=0.0005), while WL ratio positively correlated with CT hepatic index (r=0.555, P=0.0005). Age and gender showed no significant correlations with these parameters. Normal MRI Fat percentage and CT Hepatic attenuation index were mostly observed in patients without comorbidities, whereas mild-to-severe hepatic fat accumulation was associated with diabetes, fatty liver, alcohol use, and obesity. Obesity cases exclusively showed severe hepatic involvement. Conclusion: Image-based techniques, particularly MRI fat percentage and MR spectroscopy, provide accurate, non-invasive quantification of liver fat, aiding disease assessment, treatment monitoring, and pre-surgical evaluation.