Ultrasonography is Useful to Detect Severity of Fatty Infiltration in Alcoholic Liver Diseases Independent of Body Mass Index, Liver Inflammation, Fibrosis and Cirrhosis
Objective: Fatty infiltration is the earliest and reversible change in alcoholic liver diseases (ALD). Whether body mass index (BMI), liver inflammation, fibrosis or cirrhosis influences ultrasonographic detection of the degree of fatty change in ALD is rarely reported.
Methods: From July 2005 to April 2008, 120 consecutive ALD patients had abdominal ultrasonography and fatty liver was graded on a scale of 0-3 for none, mild, moderate, and severe degree based on increased echogenecity, increased attenuation, and decreased visualization of veins. Liver biopsy was performed in 85 patients (M/F 78/7, age 29-65, BMI 15.6-35.1) and graded on a scale of 0-4 for fatty infiltration, 0-3 for sclerosing hyaline necrosis, 0-3 for peri-sinusoidal and peri-venular fibrosis, and presence of hepatitis and cirrhosis. We excluded 35 patients without biopsy due to bleeding tendency, thrombocytopenia, ascites or patients’ refusal. The correlation of ultrasonography and histology grading of fatty liver was assessed by Spearman’s correlation. Multiple linear regression was used to evaluate the effect of other variables.
Results: Ultrasonographic grading of fatty liver correlates with histology grading (p=0.001), in which histology scale (HS)=0.62+0.52 x ultrasonography scale (US). This correlation is maintained even after adjustment for either BMI (HS=0.96+0.52xUS; p=0.002), sclerosing hyaline necrosis (HS=0.62+0.50xUS; p=0.003), peri-sinusoidal and peri-venular fibrosis (HS=0.27+0.52xUS; p=0.002), or hepatitis and cirrhosis (HS=0.63+0.52xUS; p=0.003).
Conclusion: Ultrasonography may semiquantitatively predict severity of fatty infiltration on histology in ALD independent of BMI, liver inflammation, fibrosis, and cirrhosis.