FibroScan Accuracy for Liver Fibrosis Assessment
FibroScan (transient elastography) demonstrates good to excellent accuracy for detecting liver fibrosis, with performance improving as fibrosis severity increases—achieving sensitivities of 81-90% and specificities of 83-89% for cirrhosis detection across major liver diseases. 1
Overall Diagnostic Performance
Based on the most comprehensive 2025 AASLD systematic review analyzing 240 studies with over 61,000 patients 1:
Sensitivity ranges:
- Significant fibrosis (F2-4): 51-95%
- Advanced fibrosis (F3-4): 70-100%
- Cirrhosis (F4): 60-100%
Specificity ranges:
- Significant fibrosis (F2-4): 36-100%
- Advanced fibrosis (F3-4): 74-100%
- Cirrhosis (F4): 67-99%
The accuracy increases progressively from detecting early fibrosis to cirrhosis, with FibroScan performing best at identifying advanced disease stages. 1
Disease-Specific Performance
NAFLD/MASLD
Using standard cutoffs 1:
- F2-4 (7 kPa cutoff): 76% sensitivity, 73% specificity
- F3-4 (10 kPa cutoff): 82% sensitivity, 79% specificity
- F4 (13 kPa cutoff): 90% sensitivity, 89% specificity
A 2019 UK multicenter study with 404 NAFLD patients confirmed these findings with AUROCs of 0.77 for F≥2,0.80 for F≥3, and 0.89 for F4 2. The optimal cutoffs identified were 8.2 kPa, 9.7 kPa, and 13.6 kPa respectively.
Hepatitis B
Using standard cutoffs 1:
- F2-4 (7 kPa cutoff): 78% sensitivity, 79% specificity
- F3-4 (8 kPa cutoff): 87% sensitivity, 83% specificity
- F4 (11 kPa cutoff): 81% sensitivity, 87% specificity
A 2016 meta-analysis of 27 studies with 4,386 HBV patients demonstrated summary AUROCs of 0.88,0.91, and 0.93 for F≥2, F≥3, and F4 respectively 3. For HBV specifically, the 2017 AGA guideline recommends an 11.0 kPa cutoff for cirrhosis detection, deliberately favoring sensitivity over specificity to minimize false negatives 4.
Hepatitis C
Using standard cutoffs 1:
- F2-4 (7 kPa cutoff): 83% sensitivity, 74% specificity
- F3-4 (11.5 kPa cutoff): 80% sensitivity, 91% specificity
- F4 (14 kPa cutoff): 83-100% sensitivity, 84-93% specificity
Alcoholic Liver Disease
Using standard cutoffs 1:
- F2-4 (7.8 kPa cutoff): 80% sensitivity, 91% specificity
- F3-4 (11.6 kPa cutoff): 76-87% sensitivity, 81-92% specificity
- F4 (≥15 kPa cutoff): 90-93% sensitivity, 85-86% specificity
A 2018 Danish study found FibroScan achieved an AUROC of 0.90 for advanced fibrosis in alcoholic liver disease, comparable to FibroTest but superior to indirect serum markers 5.
Important Clinical Caveats
Factors That Do NOT Affect Accuracy
Contrary to common concerns, steatosis severity and probe type (M vs XL) do not significantly impact liver stiffness measurements—only fibrosis stage matters 2. This finding from multivariable analysis is clinically reassuring for NAFLD patients.
Factors That DO Limit Accuracy
Morbid obesity (BMI ≥40 kg/m²) significantly reduces FibroScan accuracy. A 2022 prospective study of 167 bariatric surgery patients found that at the optimal 12.8 kPa cutoff, accuracy for significant fibrosis was only 71.3%, with a positive predictive value of just 24.6% 6. In morbidly obese patients, combining FibroScan with hemoglobin A1c and alkaline phosphatase improves accuracy to 79.6% 6.
Prevalence Effects on Predictive Values
In low-prevalence populations (e.g., primary care), negative predictive values increase while positive predictive values decrease. This means FibroScan is excellent for ruling out advanced disease in screening settings but less reliable for ruling it in 4. For example, in primary care alcoholic liver disease patients, an ELF <10.5 had a 98% negative predictive value for advanced fibrosis 5.
Comparative Performance
MRE (magnetic resonance elastography) demonstrates superior accuracy compared to FibroScan across all fibrosis stages, though FibroScan has the largest body of supporting evidence 1. For NAFLD cirrhosis detection, MRE at 6.7 kPa achieves 91% sensitivity and 95% specificity versus FibroScan's 90% sensitivity and 89% specificity at 13 kPa 1.
FibroScan outperforms blood-based tests (APRI, FIB-4) for detecting advanced fibrosis and cirrhosis 1, 4. However, for HBV cirrhosis detection, FibroScan and FIB-4 show comparable sensitivity (86% vs 87%), with FibroScan's advantage being superior specificity (85% vs 65%) 4.
Practical Application Algorithm
- Use disease-specific cutoffs as outlined above rather than universal thresholds
- For ruling out advanced disease: Lower cutoffs maximize sensitivity (e.g., <8 kPa for F3-4 in most diseases)
- For ruling in cirrhosis: Higher cutoffs maximize specificity (e.g., ≥13-15 kPa depending on etiology)
- In morbid obesity: Supplement FibroScan with biochemical markers or consider MRE
- Serial measurements: Repeat testing after viral suppression or metabolic improvement, as liver stiffness decreases with disease control 4