What is the diagnostic workup and treatment approach for a patient suspected of having fatty liver disease (steatosis)?

Diagnostic Workup for Fatty Liver Disease

Who Should Be Evaluated

Screen patients with type 2 diabetes, metabolic syndrome (≥2 metabolic risk factors), obesity (BMI >25 kg/m²), or persistently elevated liver enzymes, as these populations have the highest risk of clinically significant disease. 1, 2

• Patients with incidental hepatic steatosis on imaging should undergo further evaluation, particularly if aminotransferases are elevated 3, 1
• Screen patients with ≥2 metabolic conditions including central obesity, hypertriglyceridemia (≥150 mg/dL), low HDL (<40 mg/dL men, <50 mg/dL women), hypertension (≥130/85 mmHg), or elevated fasting glucose (100-125 mg/dL) 3
• Do not exclude NAFLD based on normal ALT alone—half of NAFLD patients have normal transaminases 1

Initial Laboratory and Clinical Assessment

Obtain a comprehensive metabolic panel, complete blood count with platelets, fasting lipid profile, fasting glucose or HbA1c, and INR as the initial laboratory workup. 1, 2

• Exclude significant alcohol consumption: >210/140 g weekly (21/14 drinks) for men/women per American guidelines, or >30/20 g daily per European guidelines 3
• Screen for alcohol use disorders using validated tools (AUDIT, AUDIT-C, or single-question screening) 3
• Rule out competing causes of steatosis: hepatitis C (especially genotype 3), hepatitis B, medications, Wilson's disease, hemochromatosis, and autoimmune liver disease 3, 2
• Assess for metabolic syndrome components including waist circumference, blood pressure, and evaluate for diabetes and dyslipidemia 3

Handling Elevated Ferritin and Autoantibodies

• Elevated ferritin is common in NAFLD and does not necessarily indicate iron overload 3
• Test for HFE gene mutations if both ferritin and transferrin saturation are elevated; consider liver biopsy if C282Y homozygote or compound heterozygote mutations are present 3
• Positive autoantibodies (ANA >1:160 or ASMA >1:40) occur in 21% of NAFLD patients and are generally epiphenomena unless accompanied by very high aminotransferases and elevated globulin 3

Imaging for Steatosis Detection

Use abdominal ultrasonography as the first-line imaging modality—it is widely available, cost-effective, and has high accuracy for moderate-to-severe steatosis. 3, 1, 2

• Ultrasound has limited sensitivity for mild steatosis (<30% hepatic involvement) and is operator-dependent 3, 4
• In high-risk patients (diabetes, metabolic syndrome, obesity), proceed directly to fibrosis risk stratification without requiring ultrasound confirmation of steatosis 3
• MRI with proton density fat fraction (MRI-PDFF) is the most accurate method for quantifying steatosis but is expensive and primarily used in research 3, 5
• Controlled attenuation parameter (CAP) via transient elastography can simultaneously assess steatosis and fibrosis 3, 2
• CT can detect steatosis but only at thresholds ≥30% and involves radiation exposure 6

Non-Invasive Fibrosis Assessment

Calculate the FIB-4 score as the preferred initial test for fibrosis assessment using age, AST, ALT, and platelet count. 1, 2

FIB-4 Interpretation Algorithm

• FIB-4 <1.3 (or <2.0 if age >65): Advanced fibrosis excluded; repeat testing in 2-3 years 1, 2
• FIB-4 1.3-2.67 (indeterminate zone): Proceed to second-tier testing with transient elastography, magnetic resonance elastography (MRE), or Enhanced Liver Fibrosis (ELF) panel 1
• FIB-4 >2.67: High risk for advanced fibrosis; refer to hepatology for further evaluation 1

Alternative non-invasive scores include:

• NAFLD Fibrosis Score (uses age, BMI, hyperglycemia, platelet count, albumin, AST/ALT ratio) 3, 1
• Hepatic Steatosis Index (HSI) for steatosis assessment (uses ALT/AST ratio, BMI, diabetes, sex) 3
• Fatty Liver Index (FLI) for steatosis assessment (uses triglycerides, GGT, BMI, waist circumference) 3, 4

Advanced Imaging for Fibrosis

• Magnetic resonance elastography (MRE) is the most accurate non-invasive test for fibrosis assessment in NAFLD 3
• Transient elastography (FibroScan) has moderate accuracy but is limited by obesity and severe steatosis 3
• MRE combined with MRI-PDFF can differentiate NASH from simple steatosis with AUC 0.82-0.93 3

When to Perform Liver Biopsy

Consider liver biopsy in patients with suspected NASH and advanced fibrosis, indeterminate non-invasive test results, or when concurrent chronic liver disease cannot be excluded. 3, 1

• Biopsy is the gold standard for diagnosing NASH and staging fibrosis but is invasive, expensive, and carries risk 3
• Biopsy provides prognostic information—presence of NASH and/or fibrosis predicts risk of progression to cirrhosis 3
• Biopsy is essential when diagnosis remains uncertain despite workup, particularly with persistently elevated aminotransferases 3
• Do not perform routine biopsy in asymptomatic patients with incidental steatosis on imaging and normal liver biochemistries 3

Diagnosing NASH Non-Invasively

Non-invasive diagnosis of NASH remains limited; definitive diagnosis requires liver biopsy. 3

• Cytokeratin-18 fragments have modest accuracy (sensitivity 66%, specificity 82%) for NASH diagnosis 3
• Multiparametric MRI combining MRE, MRS, and T1 mapping shows promise (sensitivity 80%, specificity 85.2%) but is not widely available 3
• Shear wave dispersion imaging may help diagnose NASH based on liver parenchymal viscosity 3

Screening for Complications in Cirrhotic Patients

• Screen patients with established cirrhosis for hepatocellular carcinoma and esophageal varices per AASLD guidelines 3
• No current HCC screening guidelines exist for NASH patients without cirrhosis 3
• Assess cardiovascular risk, as higher hepatic fat content is associated with increased cardiovascular mortality 3, 2

Key Pitfalls to Avoid

• Do not assume AST:ALT ratio >1 excludes NAFLD—this ratio reverses in advanced fibrosis and cirrhosis 1
• Do not screen the general population or all primary care patients—screening is not cost-effective in unselected populations 3
• Do not rely solely on ultrasound sensitivity for mild steatosis—it misses disease when <30% of liver is involved 3
• Recognize that screening family members may be warranted given heritability estimates of 0.386 for liver fat fraction, though conclusive studies are lacking 3