FibroScan vs Liver Biopsy: Which Test Do You Actually Need?
For most people with fatty liver or chronic hepatitis, FibroScan is enough to stage liver scarring: in a 450-patient UK study it detected cirrhosis with an accuracy (AUROC) of 0.89 against biopsy. A liver biopsy is still needed when the diagnosis is unclear, when two liver diseases may coexist, or when non-invasive results conflict. Biopsy is not perfect either — a second sample from the same liver disagreed by at least one fibrosis stage in 41% of patients.
— Dr Chetan Kalal, DM Hepatology, Gleneagles Hospital Mumbai
Key takeaways
- FibroScan measures liver stiffness (in kPa) as a stand-in for scarring, and a CAP score (in dB/m) for fat. It takes about ten minutes and needs no needle.
- Against biopsy in 450 UK patients with fatty liver, FibroScan stiffness identified significant fibrosis (F2+) with an AUROC of 0.77, advanced fibrosis (F3+) with 0.80, and cirrhosis with 0.89.
- Liver biopsy samples only a tiny fraction of the liver: in a study of paired biopsies, results differed by one fibrosis stage or more in 41% of patients.
- Stiffness readings can be falsely high with active hepatitis, liver congestion, blocked bile ducts or a recent meal, so results must be read in context.
- Biopsy remains essential when the cause of liver disease is uncertain, when autoimmune hepatitis, Wilson disease or drug injury is suspected, or when tests disagree.
“Do I really need a needle in my liver?” Patients ask it with obvious relief when I say, most of the time, no. Twenty years ago, staging liver scarring meant a biopsy for almost everyone. Today, for the large majority of people with fatty liver disease, hepatitis B or hepatitis C, a FibroScan and a few blood tests give an answer that is good enough to act on. But “most of the time” is not “always”, and choosing wrongly in either direction has consequences: an unnecessary biopsy carries risk, and a missed diagnosis carries more.
This guide explains what each test measures, how accurate each is, and how I decide between them at Gleneagles Hospital Mumbai.
What FibroScan measures — and how accurate it is
FibroScan (vibration-controlled transient elastography) sends a gentle mechanical pulse through the liver and measures how fast it travels. Stiff, scarred liver transmits it faster. The result is liver stiffness in kilopascals (kPa). The same probe also gives a controlled attenuation parameter (CAP), in decibels per metre, which estimates fat.
The best-quality evidence in fatty liver comes from a prospective UK study by Eddowes and colleagues, published in Gastroenterology in 2019. They performed FibroScan within two weeks of liver biopsy in 450 adults at seven centres, with biopsies read by two blinded expert pathologists. Stiffness detected significant fibrosis (F2 or worse) with an area under the curve of 0.77, advanced fibrosis (F3 or worse) with 0.80, and cirrhosis with 0.89. CAP detected any fat with 0.87. In that cohort the best-balanced stiffness cut-offs were 8.2 kPa for F2, 9.7 kPa for F3 and 13.6 kPa for cirrhosis. Importantly, neither the amount of fat nor the probe size significantly affected the stiffness reading.
What those numbers mean in practice is that FibroScan is very good at ruling advanced disease in or out, and less precise at distinguishing neighbouring stages such as F1 from F2. That is usually enough, because the clinical decisions turn on whether someone has advanced fibrosis or cirrhosis.
Cut-offs differ by disease — a stiffness value means something different in hepatitis B, hepatitis C, alcohol-related disease and fatty liver — and by the population tested. They should be interpreted by someone who knows which applies to you, not read off a generic chart. MR elastography measures stiffness across the whole liver and is more accurate still, but it is costlier and less widely available; I use it selectively.
What a liver biopsy shows — and its limits
A liver biopsy takes a thin core of tissue, usually through the skin under local anaesthetic and ultrasound guidance, or through a neck vein (transjugular biopsy) when blood clotting is poor or there is fluid in the abdomen. A pathologist then sees things no scan can: the pattern of inflammation, iron or copper deposits, features of autoimmune disease, drug injury, or a second unexpected condition.
Biopsy is still called the “reference standard”, but it is not flawless. In a classic French study, Ratziu and colleagues took two biopsy samples at the same sitting from 51 patients with fatty liver disease. The two samples disagreed by at least one fibrosis stage in 41% of patients. Of 17 patients with bridging fibrosis on one sample, six (35%) had mild or no fibrosis on the other, and ballooning — a key feature of steatohepatitis — would have been missed in 24% with a single sample. Scarring is unevenly spread through the liver, and a biopsy samples a tiny fraction of it.
Biopsy also carries small but real risks, principally pain and bleeding, which is why it is reserved for situations where the information will change management.
| FibroScan | Liver biopsy | |
|---|---|---|
| What it measures | Liver stiffness (kPa) and fat (CAP, dB/m) | Actual tissue: fibrosis stage, inflammation, cause |
| Invasive? | No — ten minutes, no needle | Yes — needle, local anaesthetic, observation afterwards |
| Accuracy for cirrhosis | AUROC 0.89 against biopsy in fatty liver | Reference standard, but 41% stage discordance between paired samples |
| Can identify the cause? | No | Often — autoimmune, Wilson, drug injury, overlap |
| Repeatable for monitoring? | Yes, easily | Rarely repeated |
| Falsely affected by | Active hepatitis, congestion, blocked bile ducts, recent meal | Small or fragmented sample |
When FibroScan is enough
For fatty liver disease, the 2024 European EASL-EASD-EASO guidelines place non-invasive tests at the centre: a FIB-4 blood score first, then elastography if needed. For most patients that pathway answers the key question — is there advanced fibrosis or cirrhosis? — without a biopsy. The same applies to most people with chronic hepatitis B or C, where treatment decisions rarely depend on biopsy today.
FibroScan has another advantage biopsy cannot match: it can be repeated. A falling stiffness after weight loss, antiviral treatment or alcohol abstinence is one of the most motivating results I can show a patient.
For a reliable reading, come having fasted for a few hours, and tell the operator if you have had a recent flare of hepatitis, heart failure or jaundice, because these can raise stiffness independently of scarring.
When a biopsy is still needed
I recommend a biopsy when the cause of liver disease is unclear after blood tests and imaging; when a second condition is suspected alongside the obvious one — autoimmune hepatitis with fatty liver, for example, or drug injury in someone with hepatitis B; when Wilson disease or another metabolic disease needs tissue confirmation; when FibroScan, blood scores and the clinical picture disagree; when liver enzymes stay unexplained and raised; and when a liver lump cannot be characterised by imaging. Some clinical trials, including those that led to the new fatty liver drugs, also required biopsy.
The right test is the one whose result will change what we do next. That judgement is best made with a hepatologist who can see your whole picture, not with a test chosen from a price list.
Frequently asked questions
Is FibroScan as good as a liver biopsy?
For detecting advanced fibrosis and cirrhosis, FibroScan performs well — an AUROC of 0.89 for cirrhosis against biopsy in a 450-patient UK study — and it avoids a needle. It cannot identify the cause of liver disease, which is when biopsy is still needed.
What is a normal FibroScan reading?
There is no single cut-off for everyone; it depends on the underlying liver disease. In fatty liver, one large study found the best cut-offs to be around 8.2 kPa for significant fibrosis, 9.7 kPa for advanced fibrosis and 13.6 kPa for cirrhosis. Your hepatologist will interpret your result for your condition.
Can a FibroScan result be wrong?
Yes. Stiffness can be falsely raised by active hepatitis, liver congestion from heart problems, blocked bile ducts or eating shortly before the test. Results are interpreted alongside blood tests and the clinical picture.
Is liver biopsy painful or risky?
It is done under local anaesthetic, and most patients feel only pressure and mild discomfort. The main risks are pain and, rarely, bleeding, which is why biopsy is reserved for when its result will change treatment.
Can a biopsy miss scarring?
Yes. Scarring is unevenly spread. When two biopsy samples were taken from the same liver, they differed by at least one fibrosis stage in 41% of patients in one study.
Book a consultation with Dr Chetan Kalal at Gleneagles Hospital, Mumbai — call +91 83294 76669 or use the appointment form on this site. International and NRI patients may request a teleconsultation.
References
- Eddowes PJ, Sasso M, Allison M, et al. Accuracy of FibroScan controlled attenuation parameter and liver stiffness measurement in assessing steatosis and fibrosis in patients with nonalcoholic fatty liver disease. Gastroenterology. 2019;156(6):1717–1730. PMID 30689971. doi:10.1053/j.gastro.2019.01.042
- Ratziu V, Charlotte F, Heurtier A, et al. Sampling variability of liver biopsy in nonalcoholic fatty liver disease. Gastroenterology. 2005;128(7):1898–1906. PMID 15940625. doi:10.1053/j.gastro.2005.03.084
- European Association for the Study of the Liver, European Association for the Study of Diabetes, European Association for the Study of Obesity. EASL-EASD-EASO Clinical Practice Guidelines on the management of MASLD: executive summary. Diabetologia. 2024;67(11):2375–2392. PMID 38869512. doi:10.1007/s00125-024-06196-3
