Non-Invasive Fibrosis Assessment

Liver biopsy has long been the reference standard for fibrosis staging, but its routine use in chronic liver disease is neither practical nor safe at scale. Key limitations:

• Invasive and costly: Requires trained operator, monitoring, and frequently sedation or inpatient observation
• Sampling error: A core biopsy samples approximately 1/50,000 of total liver mass — the liver is not homogeneous in fibrosis distribution
• Complication risk: Approximately 1 in 1,000 procedures results in significant hemorrhage; rare but serious complications include bile peritonitis and hepatic artery laceration
• Patient acceptability: Many patients decline repeat biopsy for treatment monitoring, limiting longitudinal utility

Non-invasive tests (NITs) fall into two broad categories:

When a mechanical impulse is applied to tissue, two wave types propagate:

• Compression (longitudinal) waves — the basis of standard diagnostic ultrasound; travel at approximately 1,500 m/s through soft tissue
• Shear (transverse) waves — travel perpendicular to the compression wave, much slower (approximately 1–10 m/s in soft tissue); their speed is determined by tissue stiffness

The relationship between shear wave speed and tissue stiffness is described by Young's modulus:

E ≈ 3ρv²

Where E is stiffness (Young's modulus, in kPa), ρ is tissue density, and v is shear wave velocity. Because tissue density is relatively constant (ρ ≈ 1,000 kg/m³ across liver parenchyma), shear wave speed is a direct surrogate for tissue stiffness. Results are reported as liver stiffness measurement (LSM) in kilopascals (kPa).

The intuition: think of squeezing a peach. A ripe, soft peach transmits vibration slowly (normal liver). An unripe, hard peach transmits vibration quickly (fibrotic liver). The fibrotic liver behaves like the unripe peach — stiffer matrix conducts shear waves faster, generating a higher LSM.

Vibration-controlled transient elastography (VCTE), sold as FibroScan® (Echosens, Paris), is a dedicated probe system — not integrated into a standard ultrasound machine. It is the most widely validated elastography modality and the preferred second-line fibrosis test in MASLD per AASLD 2023.

Mechanism

1. A mechanical vibrator on the probe tip generates a low-frequency (50 Hz) elastic pulse that propagates into liver tissue as a shear wave
2. A 3.5 MHz ultrasound beam simultaneously tracks shear wave velocity as it propagates through the liver
3. The system calculates shear wave speed and converts it to LSM in kPa using the relationship E ≈ 3ρv²
4. Measurement volume: a cylinder approximately 1 cm wide × 4 cm long, positioned 25–65 mm below the skin surface — samples roughly 1/500 of liver mass, far larger than a biopsy core
5. Ten valid measurements are obtained and averaged; an IQR/median ratio <30% indicates a reproducible, reliable acquisition

Probe Selection

Controlled Attenuation Parameter (CAP)

CAP is measured simultaneously with LSM and quantifies hepatic steatosis by measuring ultrasound signal attenuation (dB/m). CAP thresholds for steatosis grading:

LSM Cutoffs by Disease Context

Cutoffs differ by etiology — use the appropriate column for clinical decision-making:

Confounders That Falsely Elevate LSM

• Active hepatitis with ALT elevation >5× ULN (inflammation stiffens hepatocytes independent of fibrosis)
• Right heart failure and hepatic venous congestion
• Cholestasis and biliary obstruction
• Recent food intake (within 3 hours)
• Amyloidosis and other infiltrative diseases of the liver

Technical Limitations

• BMI >40: higher probe failure rate even with XL probe (~15–20% failure in severe obesity)
• Narrow intercostal spaces: limits acoustic window access
• Ascites: shear waves do not propagate through fluid — ascites prevents valid measurement
• Cannot distinguish fibrosis from congestion or active inflammation as causes of elevated LSM

SWE uses the same physics as VCTE but is performed on standard ultrasound platforms (GE, Siemens, Philips, Canon) integrated into the conventional B-mode workflow — no dedicated machine is required. Two technical variants exist:

Advantages over VCTE: Integrated into standard ultrasound workflow; real-time B-mode targeting allows deliberate sampling; 2D SWE provides a spatial stiffness map rather than a single number.

Limitations: More operator-dependent than VCTE; cutoffs across etiologies are less uniformly validated; same confounders apply (obesity, ascites, active inflammation, congestion).

MRE is the most accurate non-invasive fibrosis test available and serves as the reference standard among NITs for advanced fibrosis and cirrhosis detection. It is particularly valuable when ultrasound-based elastography fails or is indeterminate.

Mechanism

1. A pneumatic driver (passive driver placed on the abdominal wall) generates continuous 60 Hz mechanical vibrations that are transmitted through the body wall into the liver
2. A modified MRI pulse sequence (gradient-echo or spin-echo EPI) uses motion-encoding gradients to detect shear wave propagation as phase shifts in the MR signal
3. An inversion algorithm computes tissue stiffness at each voxel, generating a quantitative stiffness map (elastogram) in kPa
4. A radiologist places a region of interest over liver parenchyma, excluding major vessels and bile ducts
5. MRE samples the entire liver in the imaging plane — a much larger sampling volume than biopsy, VCTE, or pSWE

MRE Stiffness Cutoffs

Advantages Over Ultrasound Elastography

• Not affected by obesity, ascites, or narrow intercostal spaces — major advantage in the MASLD population
• Samples the entire liver in the imaging plane, minimizing sampling variability
• Superior diagnostic accuracy for F2–F3 staging (AUROC ~0.90–0.94 vs. ~0.85–0.90 for VCTE)
• Can be combined with standard liver MRI to simultaneously quantify steatosis (PDFF), iron content (T1/T2* mapping), perfusion (DCE), and screen for HCC (LI-RADS sequences)

Limitations

• Requires MRI access: expensive, less universally available than ultrasound, longer acquisition times
• Contraindicated with MRI-incompatible implants (older pacemakers, cochlear implants, certain metallic implants)
• Severe iron overload reduces MR signal quality; gradient-echo MRE is preferred over spin-echo sequences in the presence of mild-to-moderate iron overload
• Breath-hold of approximately 15–20 seconds required per acquisition — challenging for patients with dyspnea or ascites
• Post-processing requires dedicated software and radiologist expertise

FIB-4 is the AASLD 2023 first-line recommended fibrosis assessment for all patients with suspected MASLD. It is derived entirely from routine laboratory values available on a standard CMP plus CBC — no additional testing is required.

FIB-4 = (Age [years] × AST [U/L]) / (Platelet count [10⁹/L] × √ALT [U/L])

Why It Works

Each component captures a physiologic consequence of fibrosis progression: AST rises as hepatocyte necrosis increases; platelet count falls as splenomegaly and thrombopoietin reduction occur (hypersplenism from portal hypertension); age is included because hepatic fibrosis accumulates over decades and age correlates with fibrosis burden at the population level.

The AST-to-Platelet Ratio Index (APRI) was one of the earliest validated serum-based fibrosis indices, developed initially for hepatitis C staging.

APRI = (AST / AST ULN) × (100 / Platelet count [10⁹/L])

Current role: APRI has been largely supplanted by FIB-4 in clinical practice due to FIB-4's superior AUROC. However, APRI remains relevant in two contexts: low-resource settings where expedient calculation is needed, and HCV fibrosis staging globally — the World Health Organization (WHO) endorses APRI for HCV fibrosis assessment in resource-limited countries where elastography is unavailable.

FibroTest (the European licensed name) and FibroSure (the US licensed name) are the same proprietary algorithm. The score is calculated from six components using a validated formula — it cannot be replicated from standard labs and requires a proprietary licensed assay sent to a reference laboratory.

Components

1. Alpha-2-macroglobulin (α2M) — elevated in hepatic fibrosis; also an acute-phase reactant
2. Haptoglobin — decreased in hemolysis and in hepatic dysfunction; low haptoglobin raises the FibroTest score
3. Apolipoprotein A1 (ApoA1) — decreased with hepatocyte synthetic dysfunction
4. GGT — elevated with fibrosis, cholestasis, and alcohol use
5. Total bilirubin — elevated with reduced hepatic conjugation and excretory function
6. Age and sex — incorporated as correction factors in the proprietary algorithm

Advantages: Well-validated across HCV, HBV, ALD, and MASLD; FDA-cleared in the US; standardized proprietary algorithm reduces inter-laboratory variability.

Limitations: Requires a specific licensed laboratory kit — not calculable from standard chemistry panels. Falsely elevated by hemolysis (decreases haptoglobin artifactually), Gilbert's syndrome (unconjugated bilirubin elevation), and acute inflammation (α2M as an acute-phase reactant rises independently of fibrosis).

Unlike FIB-4 and APRI, which measure indirect consequences of fibrosis (thrombocytopenia, transaminase elevation), the ELF score measures direct markers of extracellular matrix remodeling — components produced during active fibrogenesis. This mechanistic directness is its major advantage.

Components

1. TIMP-1 (Tissue Inhibitor of Metalloproteinase-1) — inhibits matrix metalloproteinases that normally degrade collagen; elevated when fibrosis is actively accumulating and collagen degradation is impaired
2. PIIINP (N-terminal propeptide of type III procollagen) — cleaved from procollagen III during collagen synthesis; directly reflects active fibrogenesis
3. Hyaluronic acid (HA) — a glycosaminoglycan deposited in perisinusoidal spaces during fibrosis; elevated HA also reflects impaired hepatic sinusoidal clearance

ELF = 2.278 + (0.851 × ln[HA]) + (0.751 × ln[PIIINP]) + (0.394 × ln[TIMP-1])

Advantages: Mechanistically direct — measures fibrosis components rather than indirect surrogate consequences; not affected by age (unlike FIB-4); AASLD-approved as a second-line test in MASLD when FIB-4 is indeterminate; accepted alongside VCTE as sufficient to establish F2–F3 fibrosis for resmetirom prescribing decisions without biopsy.

Limitations: Requires specialized immunoassay platform (Siemens ADVIA Centaur); not universally available in routine clinical labs; PIIINP is elevated during rapid growth phases (adolescents) and in active systemic inflammation independent of hepatic fibrosis.

The Baveno VII consensus workshop (2021) established non-invasive criteria to rule out clinically significant portal hypertension (CSPH), defined as a hepatic venous pressure gradient (HVPG) ≥ 10 mmHg. CSPH is the threshold above which varices form and the risk of hepatic decompensation (ascites, variceal hemorrhage, hepatic encephalopathy) rises substantially.

Traditionally, all patients with compensated advanced chronic liver disease (cACLD) underwent upper endoscopy (EGD) to screen for gastroesophageal varices. Baveno VII provides a validated alternative pathway.

The AASLD 2023 Practice Guidance establishes a stepwise non-invasive strategy for all patients with suspected MASLD. Serum-based testing is always first; elastography is reserved for indeterminate results.