If fatty liver is identified on computed tomography (CT) scan, should an ultrasound (US) be obtained for confirmation and assessment?

Should Ultrasound Be Obtained After CT-Detected Fatty Liver?

No, ultrasound should not be routinely obtained after fatty liver is identified on CT, as ultrasound does not add diagnostic value and has significant limitations including high false-positive rates and inability to accurately grade steatosis severity. 1

Rationale Against Obtaining Ultrasound

CT Already Provides Adequate Detection

• CT has already established the diagnosis of hepatic steatosis, making additional imaging with ultrasound redundant for confirmation 1
• CT demonstrates strong correlation with quantitative measures of liver fat, with HU values inversely correlating with steatosis severity (65.2 ± 5.3 HU in normal liver to 35.6 ± 5.5 HU in severe steatosis) 2
• CT shows high diagnostic accuracy with 90% sensitivity and 90% specificity for moderate steatosis, and 95% sensitivity and 95% specificity for severe steatosis 2

Ultrasound Has Significant Limitations

• Ultrasound has a positive predictive value of only 47-62% for fatty liver in children, with similar limitations in adults, meaning nearly half of positive ultrasound findings may be false positives 1
• Ultrasound requires at least 20-30% hepatic fat content for reliable detection, missing mild steatosis below this threshold 3, 4
• Ultrasound cannot distinguish alcoholic from non-alcoholic causes of steatosis, cannot differentiate simple steatosis from steatohepatitis, and cannot identify inflammation 3
• Ultrasound correctly identifies steatosis severity in only 40-60% of cases when compared to quantitative measures 1

High False-Positive Rate of Ultrasound

• When ultrasound shows mild steatosis (Grade 1), 72% of cases may have negative findings on quantitative imaging (MRI) 1
• Of patients with no steatosis on MR spectroscopy, 25 out of 56 (45%) had ultrasound findings suggesting mild or moderate steatosis 1
• The false-positive rate is particularly problematic in mild disease, where 12 out of 18 children (67%) with mild steatosis by ultrasound had negative MRI 1

Appropriate Next Steps After CT-Detected Fatty Liver

Clinical and Laboratory Evaluation

• Obtain a complete hepatic panel including AST, ALT, alkaline phosphatase, bilirubin, albumin, and INR to assess liver function and differentiate causes (AST/ALT >2 suggests alcoholic origin; <1 suggests NAFLD) 5
• Screen for alternative causes: viral hepatitis serologies (HBsAg, anti-HBc, anti-HCV), hemochromatosis (ferritin, transferrin saturation), Wilson's disease (ceruloplasmin if age <40), and autoimmune hepatitis (ANA, anti-smooth muscle antibodies) 5
• Perform metabolic evaluation: fasting glucose, HbA1c, complete lipid profile, CBC, and creatinine 5
• Obtain detailed alcohol consumption history (≥21 drinks/week in men defines significant consumption) 5

Risk Stratification for Fibrosis

• Calculate NAFLD fibrosis score or FIB-4 score using age, transaminases, platelets, and albumin to stratify risk of advanced fibrosis 5, 6
• If scores indicate intermediate or high risk, consider transient elastography (FibroScan) with Controlled Attenuation Parameter (CAP) to quantify both hepatic stiffness and fat content 5, 3
• CAP provides quantitative assessment with cutoffs of 250 dB/m (mild), 299 dB/m (moderate), and 327 dB/m (severe steatosis) 3

When Advanced Imaging May Be Appropriate

• MRI with proton density fat fraction (PDFF) is the gold standard if quantification of steatosis severity is clinically necessary, showing 95% sensitivity and 92% specificity for detecting hepatic fat fraction >5% 1
• Liver biopsy remains necessary only when aggressive treatment is considered, cofactors are suspected, or histologic confirmation of steatohepatitis is required 3

Common Pitfalls to Avoid

• Do not assume normal transaminases exclude significant liver disease—NAFLD and NASH can occur with normal ALT/AST 5
• Do not interpret isolated elevated ferritin as hemochromatosis—it is common in NAFLD as an epiphenomenon 5
• Do not neglect fibrosis risk stratification—calculation of FIB-4 or NAFLD fibrosis scores identifies patients requiring hepatology referral 5, 7
• Recognize that 18% of patients with incidentally noted steatosis have significant fibrosis (≥F2), emphasizing the importance of proper evaluation rather than repeat imaging 7