Can MRI (Magnetic Resonance Imaging) detect Non-Alcoholic Steatohepatitis (NASH)?

Can MRI Detect NASH?

MRI alone cannot reliably diagnose NASH, and liver biopsy remains the gold standard for definitive diagnosis. 1 However, multiparametric MRI techniques combining multiple sequences show promising diagnostic performance and may aid in identifying NASH patients when biopsy is not feasible.

Current Guideline Recommendations

The most recent guidelines (2021) explicitly state that non-invasive diagnosis of NASH remains limited and should be diagnosed by liver biopsy. 1 This represents the consensus position across major hepatology societies.

Why Standard MRI Falls Short

• Standard MRI sequences (including MRI-PDFF and MRS) can quantify hepatic steatosis but cannot distinguish NASH from simple steatosis (NAFL). 1
• Clinical, biochemical, and conventional imaging measures cannot differentiate NASH from steatosis alone. 1
• No clear diagnostic criterion has been established for using CT or MRI to differentiate NASH from NAFLD. 1

Emerging Multiparametric MRI Approaches

While not yet standard practice, advanced MRI techniques show potential:

Magnetic Resonance Elastography (MRE) Combined with MRI-PDFF

• The combination of MRE and MRI-PDFF achieved an AUC of 0.82-0.93 for differentiating NASH from NAFL. 1
• MRE alone can detect inflammation even before fibrosis develops, with patients having inflammation but no fibrosis showing higher liver stiffness than those with simple steatosis. 2

Multiparametric MRI Indices

• A Korean study using MRE, MRS, and T1 mapping demonstrated sensitivity of 80%, specificity of 85.2%, and AUC of 0.883 for NASH diagnosis. 1, 3
• A meta-analysis of all MRI studies for NASH differentiation showed pooled sensitivity of 87.4% and specificity of 74.3%. 1
• Quantitative susceptibility imaging combined with fatty acid composition analysis showed an AUC of 0.91 for NASH detection. 4

Clinical Algorithm for NASH Diagnosis

When NASH is suspected based on metabolic risk factors and elevated fibrosis markers:

1. First-line assessment: Use FIB-4 index to stratify fibrosis risk (cutoff <1.3 excludes advanced fibrosis). 5, 6

2. Second-line imaging: If FIB-4 is indeterminate (1.3-2.67), proceed with transient elastography or MRE to assess fibrosis. 6

3. Consider liver biopsy when:

   • NASH with advanced fibrosis is suspected based on clinical and imaging findings 1
   • Noninvasive tests are indeterminate 6
   • Concurrent chronic liver disease cannot be excluded 1
   • Definitive diagnosis is needed for treatment decisions 1

Key Limitations and Pitfalls

• Multiparametric MRI techniques remain primarily research tools and are not recommended for routine clinical practice. 1
• These advanced MRI methods have high cost and limited availability, restricting their use to specialized centers. 1, 5
• Cytokeratin-18 fragments show modest accuracy (66% sensitivity, 82% specificity) but do not perform better than ALT for identifying NASH. 1
• Approximately 20-58% of patients fall into indeterminate ranges with noninvasive tests, necessitating biopsy for definitive diagnosis. 5

Bottom Line for Clinical Practice

Use standard MRI (MRI-PDFF) to quantify steatosis and MRE to assess fibrosis, but proceed to liver biopsy when NASH diagnosis is clinically important for management decisions. 1 Multiparametric MRI may be considered in research settings or tertiary centers with expertise, but should not replace biopsy when definitive NASH diagnosis is needed. 3, 7