Non-invasive methods of assessing liver fibrosis, such as transient elastography

The goal

Goal: identify ADVANCED FIBROSIS and PORTAL HYPERTENSION without a biopsy
- Fibrosis stage is the single strongest predictor of liver-related outcomes - the grade of inflammation is not
Two families of test:
- Serum panels - simple (FIB-4, APRI, NAFLD fibrosis score) or proprietary (ELF, FibroTest)
- Imaging elastography - vibration-controlled transient elastography (FibroScan), 2D shear-wave elastography, MR elastography
Two-step pathway is now standard: a cheap, sensitive rule-OUT test first (FIB-4), then elastography only in the indeterminate/high group

MASLD affects ~1 in 3 Australian adults; ~20-25% have MASH, ~5% advanced fibrosis
- Population prevalence is why biopsying everyone is impossible - hence these tests
FIB-4 <1.3 excludes advanced fibrosis in ~60-70% of a primary-care MASLD population at one stroke
FibroScan failure/unreliable measurement ~5-15%, almost all obesity-related

Elastography measures the speed of a shear wave through liver tissue; stiffness (kPa) rises with collagen content
*Stiffness is not fibrosis alone* - it rises with anything that raises tissue pressure:
- Acute hepatitis (high ALT), cholestasis/biliary obstruction, hepatic congestion (right heart failure), food in the stomach, amyloid, infiltration
- -> fast for at least 3 hours, and do not measure during an ALT flare
Serum panels use surrogates of fibrosis physiology: falling platelets (portal hypertension, dec thrombopoietin), rising AST:ALT ratio, and matrix turnover markers (ELF)

A. Serum scores

Score	Components	Interpretation
FIB-4	Age, AST, ALT, platelets	<1.3 rules out advanced fibrosis (NPV >90%); >2.67 rules in. Use <2.0 as the lower cut-off if age >65 - it over-calls in the elderly
APRI	AST:platelet ratio	Good negative predictive value for cirrhosis; <0.5 rules out, >1.0 rules in
NAFLD fibrosis score	Age, BMI, IFG/diabetes, AST/ALT, platelets, albumin	Useful where obesity limits FibroScan
ELF panel	Hyaluronic acid, PIIINP, TIMP-1	Direct matrix markers; >9.8 suggests advanced fibrosis. An alternative in obesity

All serum panels perform worse in the middle of their range - that is the point of the two-step pathway

B. Transient elastography (FibroScan)

Median of >=10 valid measurements; report requires IQR/median <=30%
Cirrhosis: <12.5 kPa makes it unlikely (very high NPV); >12.5 kPa makes it likely
Baveno VIII "Rule of Five" for compensated advanced chronic liver disease (cACLD)

LSM (kPa)	Meaning
<5	Normal - rules out cACLD
5-10	cACLD unlikely without other evidence
10-15	cACLD likely
15-20	cACLD highly likely
>=20-25	Clinically significant portal hypertension likely

LSM <15 kPa + platelets >150 = CSPH excluded -> varices unlikely -> endoscopy can be avoided (Baveno favourable criteria)
*In MASLD with obesity, the >=25 kPa rule-in threshold for CSPH is unreliable* - integrate BMI, platelets, spleen stiffness and other NITs (Baveno VIII)
Spleen stiffness measurement is now part of the backbone of non-invasive CSPH assessment
CAP (controlled attenuation parameter) on the same machine quantifies steatosis, not fibrosis

C. Where each fails

Problem	Effect	Alternative
Obesity, narrow intercostal spaces, ascites	Failed or unreliable FibroScan	XL probe, MR elastography, NAFLD fibrosis score or ELF
Acute hepatitis, cholestasis, congestion, recent meal, alcohol binge	Falsely high stiffness	Repeat when settled
Age >65	FIB-4 over-calls	Raise the cut-off to 2.0

D. Liver biopsy - still the reference

Indications now narrow: discordant non-invasive tests, suspected coexisting disease, unexplained abnormality, or a trial/drug eligibility requirement
Samples ~1/50,000 of the liver - sampling error, inter-observer variation
Risks: pain ~30%, significant bleeding ~1 in 500, death ~1 in 10,000
Transjugular route if coagulopathy or ascites (and allows HVPG measurement)

Low risk (FIB-4 <1.3 / LSM <8 kPa) -> manage metabolic risk in primary care; repeat in 2-3 years
Indeterminate -> elastography or ELF; if still indeterminate, specialist referral
High risk (LSM >=10 kPa, FIB-4 >2.67) -> hepatology referral
- HCC surveillance: 6-monthly ultrasound once cirrhotic
- Varices - endoscopy unless Baveno favourable criteria met; carvedilol if CSPH
- Vaccinate (hepatitis A/B), avoid alcohol, review hepatotoxic drugs
MASH with F2-F3 fibrosis -> now a treatment indication, not just observation
- Resmetirom (thyroid hormone receptor-beta agonist) - first agent approved for MASH with F2-F3; not TGA/PBS-funded in Australia
- GLP-1 RA (semaglutide) and tirzepatide - weight loss, MASH resolution
- Bariatric surgery where BMI criteria met
- Weight loss targets: 5% for steatosis, 7-10% for MASH resolution, >10% for fibrosis regression

Repeat non-invasive testing rather than repeat biopsy - annually in high risk, 2-3 yearly if low
*A falling LSM after treatment (or alcohol abstinence, or HCV cure) predicts improved outcomes* - but does not remove HCC surveillance if cirrhosis was established

MASLD/MASH - the dominant use case; renamed from NAFLD/NASH (2023 Delphi consensus)
- MetALD = metabolic dysfunction + significant alcohol intake
Chronic hepatitis B and C - staging, and post-cure follow-up
Alcohol-related liver disease - avoid measuring during a binge or acute hepatitis; stiffness falls substantially with abstinence
Haemochromatosis - FibroScan and MRI have largely replaced biopsy for staging
PBC/PSC - elastography validated, but cholestasis raises stiffness independently
Right heart failure / Fontan circulation - congestive hepatopathy raises stiffness with little fibrosis
Amyloid, lymphoma, extramedullary haematopoiesis - infiltration raises stiffness

*Fibrosis stage, not steatosis or inflammation, predicts liver-related and all-cause mortality*
- F0-F1: no excess liver mortality. F3-F4: markedly increased
LSM predicts outcomes continuously - each increment raises decompensation and HCC risk
Fibrosis regresses with the removal of the driver (alcohol, HCV cure, weight loss, venesection) - cirrhosis can even "recompensate" (Baveno)
Repeat testing beats a single value - a rising LSM over years identifies the progressor
The commonest cause of death in MASLD remains cardiovascular, not liver - manage the cardiometabolic risk regardless of the fibrosis result

10 more sections, plus exam facts

Premium unlocks every note across every specialty, and the full exam fact library behind it.