Roughly a third of adults have fat in the liver. Most will never come to harm from it. A minority will develop scarring that eventually matters a great deal, and the entire clinical problem is telling those two groups apart — which can largely be done with blood tests you have probably already had.
This chapter is in a cardiovascular guide for a reason that surprises people: the commonest cause of death in fatty liver disease is not liver disease. It is cardiovascular disease. The liver is where the metabolic problem shows up first and most measurably. The heart is where it kills you.
1. The name changed, and the change was not cosmetic
What was called non-alcoholic fatty liver disease (NAFLD) is now MASLD — metabolic dysfunction-associated steatotic liver disease. Three things changed with it.
- It became a positive diagnosis rather than an exclusion. NAFLD meant "fatty liver, and we have ruled out alcohol". MASLD requires fat in the liver plus at least one cardiometabolic risk factor — excess weight or waist, raised glucose or diabetes, raised blood pressure, raised triglycerides, or low HDL. That is a diagnosis of something, rather than a diagnosis of not-something.
- The word "alcoholic" went, and the stigma with it. Being told you have a liver condition defined by not drinking was neither accurate nor useful.
- It stopped pretending the two are separate. There is now a middle category, MetALD, for people with metabolic liver disease who also drink moderately — which is a very large number of people, and whose two problems compound rather than queue politely. See the alcohol chapter.
You will still see NAFLD and NASH on older results and in most existing literature. NASH — the inflamed, actively damaging form — is now MASH. The conditions are the same; the labels moved.
2. Fat is not the problem. Scarring is.
This is the single most useful thing on the page, and the thing an ultrasound report saying "fatty liver" almost never explains.
Steatosis — fat in liver cells — is extremely common and on its own carries little liver risk. In perhaps 10–30% of people it progresses to MASH, where the fat is accompanied by inflammation and cell injury, and inflammation over years produces fibrosis: scar tissue replacing working liver. Fibrosis is staged F0 to F4, where F4 is cirrhosis.
Fibrosis stage is what predicts outcomes. Not how much fat is visible, not how abnormal the liver enzymes look. Someone with a strikingly fatty liver on ultrasound and no fibrosis is in a much better position than someone with modest fat and F3 scarring.
Which is why normal liver enzymes are not reassurance. ALT and AST can be entirely normal in the presence of significant fibrosis, and a raised ALT is a poor guide to how much scarring there is. If liver enzymes were sufficient, none of the scoring below would need to exist.
3. The two-step pathway, and what each step actually does
Step one: FIB-4
An index combining age, AST, ALT and platelet count. It is free, it uses tests already performed, and it is designed as a rule-out.
| FIB-4 | Interpretation | What follows |
|---|---|---|
| Under 1.3 | Advanced fibrosis unlikely | Managed in primary care. Address the cardiometabolic risk factors and repeat in a few years. |
| 1.3 to 2.67 | Indeterminate — the largest group | Second-line test: elastography or an ELF blood test. European guidance also allows a year of intensive lifestyle and risk-factor management, then retesting. |
| 2.67 or above | Advanced fibrosis likely | Specialist referral. |
Two age caveats that materially change the answer, because age is a term in the formula and drags the score with it:
- Over 65, the lower threshold rises to about 2.0. Using 1.3 in an older person generates a large number of false alarms.
- Under about 35, FIB-4 performs poorly in either direction and elastography is the better first test.
Step two: elastography, or ELF
Vibration-controlled transient elastography — a FibroScan — measures how stiff the liver is by timing a pulse through it. Painless, about ten minutes, no needle. Below roughly 8 kPa suggests low risk; higher readings prompt specialist assessment. The ELF test is a blood alternative measuring markers of collagen turnover, used where elastography is not available.
Where this pathway is weaker than it looks. FIB-4's strength is its negative predictive value — around 90% — so a low score is genuinely reassuring. Its positive predictive value is poor, which is exactly why a raised score triggers a second test rather than a diagnosis. And the two steps disagree more often than the flowchart suggests: up to about 30% of patients have discordant FIB-4 and liver stiffness results, with the stiffness measurement generally the better reflection of true stage. A score is a triage tool, not a verdict. Moderate
4. Why this belongs in a heart guide
MASLD is not a liver disease that happens to occur in people with metabolic problems. It is the hepatic expression of the same insulin resistance that drives the lipid pattern in the glucose chapter — raised triglycerides, low HDL, small dense LDL, apoB high relative to LDL-C. The liver is not a bystander in that picture; it is where much of it is manufactured, through de novo lipogenesis.
So a MASLD diagnosis should do two things at once. It should trigger the fibrosis pathway above. And it should raise your estimate of cardiovascular risk, because that is what is statistically most likely to kill the person in front of you. A fatty liver found incidentally on a scan is a reason to check blood pressure, apoB and HbA1c — not only a reason to think about the liver. Strong
5. Treatment: weight, and now two drugs
Weight loss remains the core of it, and the dose-response is well described: modest loss improves the fat, more substantial loss improves the inflammation, and sustained loss of around 10% of body weight can improve fibrosis itself. Nothing else on this page matches it. Strong The specifics belong in eating for the heart and exercise, and exercise helps the liver even without much weight change.
Beyond that, coffee has a surprisingly consistent inverse association with fibrosis; alcohol should be reduced or stopped, since it compounds directly; and the cardiometabolic risk factors should be treated on their own merits rather than deferred.
What changed in 2026
For the first time there are licensed drugs. The MHRA authorised resmetirom (Rezdiffra), a liver-selective thyroid hormone receptor-β agonist, on 3 June 2026 through the International Recognition Procedure, for MASH with stage 2–3 fibrosis. In early July it granted semaglutide (Wegovy) a conditional marketing authorisation for MASH with moderate-to-advanced fibrosis — conditional, explicitly, on further results from the ongoing trial being submitted.
| Drug | Trial result |
|---|---|
| Resmetirom | MAESTRO-NASH, over 900 adults. Roughly 26–30% achieved MASH resolution without worsening fibrosis at 12 months, against about 10% on placebo, with fibrosis improvement also more common. |
| Semaglutide | ESSENCE, at 72 weeks. About 63% achieved resolution of steatohepatitis without worsening fibrosis against roughly 34% on placebo; fibrosis improvement in about 37% against 22.5%. |
MHRA authorisation is also not the same as NHS availability: NICE has to complete a technology appraisal before routine funding follows. In practice, for most people reading this, the treatment is still weight, alcohol, exercise and treating the cardiometabolic risk factors properly.
What to do with this
If you have been told you have a fatty liver, ask for a FIB-4
It may be calculable from bloods you have already had. The question that matters is fibrosis, and "fatty liver" on an ultrasound report does not answer it either way.
Treat it as a cardiovascular finding as well as a liver one
Blood pressure, apoB and HbA1c. The liver told you something about the rest of you.
Do not be reassured by normal liver enzymes
They can be normal with significant fibrosis, and abnormal with none.
If FIB-4 is raised, the next step is a second test, not a diagnosis
Elastography or ELF. Most raised FIB-4 scores do not turn out to be advanced fibrosis.