Fatty liver: what actually makes it regress

In brief

In 2023, an international consensus renamed non-alcoholic fatty liver disease as MASLD, and this is no matter of vocabulary: the diagnosis is no longer made by excluding alcohol, but by finding at least one cardiometabolic risk factor. Fatty liver becomes a metabolic signal, not an isolated liver problem. Two further facts are worth knowing: being slim is no protection, and improvement follows a very clear gradient. In a biopsy study, every patient who lost at least 10 % of their weight saw their liver findings improve, and 90 % saw their steatohepatitis disappear.

Olive oil, green vegetables, nuts and berries on a wooden board, next to a water carafe
Diet remains the first documented lever on liver fat (illustration).

“Your liver is a bit fatty, it’s nothing serious.” Millions of people have left a consultation with that sentence and understood nothing more. The term itself did not help: it was called non-alcoholic fatty liver disease, in other words it was defined by what it was not.

In 2023, an international consensus bringing together 236 experts and patient representatives from 56 countries put an end to that wording[1]. The name change was widely covered in medical journals, and almost not at all with the public. Yet it changes the very way the diagnosis is made. This article explains what changed, and above all what really makes liver fat regress. It is informational and does not replace your doctor’s advice.

Fatty liver: what exactly are we talking about?

A liver that stores fat

Hepatic steatosis means the accumulation of fat in liver cells. A healthy liver contains very little. Beyond a certain threshold, this accumulation becomes visible on ultrasound or measurable by other examinations. It is by far the leading cause of liver abnormality in Western countries, and it now affects around one adult in three worldwide[2].

From plain fat to inflammation

The initial stage is simple steatosis: fat, with no notable inflammation. In some of those affected, it progresses to steatohepatitis, where inflammation and damage to liver cells are added. It is this second stage that can, in the long run, lead to fibrosis and then to more serious complications. The distinction matters: most people with simple steatosis will not progress that way.

Why it goes unnoticed

The liver does not hurt. It is almost always found by chance: an abdominal ultrasound done for something else, or slightly raised liver enzymes on a routine blood test. Many people live for years with steatosis without knowing it, which explains the scale of the phenomenon and the difficulty of making it a visible public health topic.

2023: the name change that changes the diagnosis

What was decided

The consensus process was run by three major liver societies, over four rounds of consultation. Nearly three quarters of participants felt the old terminology was flawed enough to change. The word “non-alcoholic” was judged stigmatising by 61 % of them, the word “fatty” by 66 %[1]. Non-alcoholic fatty liver disease thus became MASLD, metabolic dysfunction-associated steatotic liver disease.

The real change is not the word

It is the definition. Previously, the diagnosis was made by exclusion: fat in the liver, and no alcohol involved. Now a positive criterion is required: the presence of at least one of the five cardiometabolic risk factors, such as a raised waist circumference, disturbed blood sugar, high blood pressure or an abnormality of blood lipids[1]. The reasoning is reversed: you no longer look for what is missing, you record what is present.

What this changes for you

Fatty liver stops being an isolated liver problem and becomes a metabolic signal. In practice, this means that steatosis found on ultrasound warrants looking at blood sugar, blood pressure, waist circumference and lipids, not just rechecking liver enzymes in six months. It is probably the most useful information in this article, and it is almost absent from consumer-facing content.

Former approachSince 2023
NameNon-alcoholic fatty liver diseaseMASLD, metabolism-associated steatotic liver disease
Logic of the diagnosisBy excluding alcoholBy finding at least one cardiometabolic factor
Intermediate alcohol useBlurred categoryIts own category, called MetALD
Reading of the problemLiver abnormalityOverall metabolic signal

Being slim is no shield

An underestimated reality

The stock image links fatty liver to excess weight. Statistically that is common, but far from exclusive. A recent review estimates that the form affecting lean people concerns around 13 % of the world population, with a more marked frequency in Asia[2]. In other words, a normal waist circumference does not rule out the diagnosis.

The point that surprises most

The lean people affected do not have a better prognosis. That same review finds in them an overall mortality equal to or higher than that of affected non-lean people, and an increased risk of liver cancer[2]. People with excess weight, for their part, show more cardiovascular events and more type 2 diabetes. The risk profile differs, but severity does not ease off with slimness.

The practical consequence

If steatosis is found in you while your weight is normal, it is neither an ultrasound error nor a negligible detail. It warrants the same metabolic work-up and the same follow-up. The authors also stress that advanced age, liver fibrosis and type 2 diabetes remain the main risk modifiers, whatever the build.

What makes it regress: a very clear gradient

The study that put numbers on it

It is one of the rare pieces of work to have measured the effect on the liver itself, by biopsy, before and after. For one year, 293 people with steatohepatitis followed a lifestyle support programme; 261 had two comparable biopsies. The result follows a striking progression according to the extent of weight loss[3].

The figures

Among people who lost at least 5 % of their weight, 58 % saw their steatohepatitis disappear. Among those who lost at least 10 %, all of them saw their liver findings improve, 90 % had resolution of the steatohepatitis and 45 % a regression of fibrosis[3]. This last point deserves emphasis: fibrosis, long considered hard to reverse, receded in nearly one person in two in that group.

How to read these figures without losing heart

Losing 10 % of your weight is demanding, and only 30 % of participants reached the 5 % threshold within a year. So the message is not “all or nothing”. This gradient mainly indicates that every step counts, and that the goal is set in stages, with support. It also gives a concrete benchmark where the usual advice stays vague: it is not “eat better”, it is a measurable percentage, to be discussed with your doctor.

Moving works, even without losing weight

What physical activity shows

Exercise has been studied for its effect on so-called ectopic fat, the fat that settles where it should not, including liver fat. A review gathering 24 randomised trials and nearly 1,400 participants with type 2 diabetes found a significant reduction in visceral fat, and a reduction in hepatic steatosis on the edge of significance[4].

Which activity, precisely

In that analysis, it was endurance exercise that reduced visceral fat in overweight or obese people with type 2 diabetes; resistance training alone did not show that effect[4]. The authors recommend making it a central element of programmes. They also note that the effect of exercise in its own right, independently of weight loss, still needs clarifying.

The complementary dietary lever

Recommendations converge on simple principles: cut sugary drinks, limit ultra-processed products, increase the share of vegetables and fibre, favour vegetable oils. These measures work first because they bear on overall energy balance and on metabolism, not because a particular food would “cleanse” the liver. Detox cures, for their part, rest on no data at all.

Alcohol: the grey zone now has a name

Why an intermediate category was needed

The old classification set two worlds against each other: alcoholic liver disease on one side, “non-alcoholic” steatosis on the other. Between the two, a very common reality had no box: people with both metabolic factors and moderate to intermediate alcohol consumption.

What the consensus created

That situation now has a name, MetALD, and quantified reference figures: a weekly consumption between 140 and 350 g of alcohol in women, and between 210 and 420 g in men[1]. The benefit is concrete: acknowledging that drinking which is not dependence can nonetheless weigh on a liver already loaded by metabolism, without placing the person in the alcoholic category.

Let us sum up what matters. Fatty liver changed its name because it changed its status: it is no longer a diagnosis of exclusion but a metabolic marker, one that invites a look at blood sugar, blood pressure and waist circumference. Slimness is no protection. And improvement follows a gradient documented on biopsies, where every step of weight loss counts.

If these questions interest you from the digestive side, two reads extend this one: our article on SIBO, that bacterial overgrowth of the small intestine often confused with other disorders, and the one devoted to belly fat, which shares a good part of its mechanisms with steatosis.

Frequently asked questions

Is fatty liver reversible?

Biopsy data show that improvement is possible and that it depends on the extent of weight loss. In a one-year study in people with steatohepatitis, everyone who lost at least 10 % of their weight saw their liver findings improve, 90 % saw their steatohepatitis resolve and 45 % even saw their fibrosis regress. At 5 % loss, more than half already achieved resolution. None of this is decided alone: this path is organised with a doctor.

Why do we now talk about MASLD?

Because a 2023 international consensus bringing together 236 experts from 56 countries judged the old terminology flawed and stigmatising. The term non-alcoholic bothered 61 % of participants, the word fatty 66 %. Beyond vocabulary, the definition changed: diagnosis now requires at least one cardiometabolic risk factor, instead of being made by excluding alcohol. Fatty liver thereby becomes a marker of metabolism.

Can you have a fatty liver without being overweight?

Yes, and it is more common than people think. A recent review puts this form at around 13 % of the world population, with higher prevalence in Asia. The important point is that these people do not have a better prognosis: their overall mortality is equal to or higher than that of affected non-lean people, with an increased risk of liver cancer. A normal weight therefore exempts you from neither the work-up nor the follow-up.

What kind of exercise works on liver fat?

Endurance comes first. In a review of 24 randomised trials in people with type 2 diabetes, endurance exercise reduced visceral fat, unlike resistance training alone, and the effect on hepatic steatosis was close to the significance threshold. The authors make it a central element of programmes. Whether the effect is independent of weight loss remains an open question.

Is there a supplement for fatty liver?

No, and claims of that kind should be treated with suspicion. A food supplement does not treat, prevent or cure any disease, liver disease included, and both Swiss and European law forbid implying otherwise. The documented levers are the ones described here: gradual, supported weight loss, endurance activity, fewer sugary drinks and fewer ultra-processed products. Detox cures rest on no data at all.

What is MetALD?

It is the category created in 2023 for intermediate situations, where metabolic factors coexist with alcohol consumption that is neither negligible nor massive. The reference figures adopted are 140 to 350 g of alcohol a week in women and 210 to 420 g in men. That box was missing: it makes it possible to acknowledge that moderate drinking can weigh on an already strained liver, without placing the person in the alcoholic category.

Which examinations detect it?

It is most often found by chance, on abdominal ultrasound or through moderately raised liver enzymes. Other examinations then allow the amount of fat and, above all, the fibrosis to be assessed, fibrosis being the most important prognostic parameter. The choice of these examinations is the doctor’s. What has changed since 2023 is that a steatosis found must be accompanied by a look at cardiometabolic parameters.

Do you have to stop drinking altogether?

That is a medical decision, depending on the stage and on the rest of the picture. What the new classification brings is a framework for the discussion: it quantifies the thresholds that move a person from one category to another and acknowledges the existence of an intermediate zone. This makes it possible to address the question without labels, by looking at actual quantities rather than setting abstinence against alcoholism.

Sources and references (verified on PubMed)

4 sources
  1. Rinella M.E. et al. (2023). A multisociety Delphi consensus statement on new fatty liver disease nomenclature. | Journal of Hepatology | Delphi consensus, 236 experts and patient representatives, 56 countries: NAFLD becomes MASLD, diagnosis conditional on at least one cardiometabolic risk factor, creation of the MetALD category
  2. Danpanichkul P. et al. (2025). Long-term outcomes and risk modifiers of metabolic dysfunction-associated steatotic liver disease between lean and non-lean populations. | Clinical and Molecular Hepatology | review: the form affecting lean people concerns around 13 % of the world population, with overall mortality equal to or higher and an increased risk of hepatocellular carcinoma
  3. Vilar-Gomez E. et al. (2015). Weight Loss Through Lifestyle Modification Significantly Reduces Features of Nonalcoholic Steatohepatitis. | Gastroenterology | prospective study, 293 patients, 261 paired biopsies at 52 weeks: beyond 10 % weight loss, histological improvement in all, resolution in 90 % and regression of fibrosis in 45 %
  4. Sabag A. et al. (2017). Exercise and ectopic fat in type 2 diabetes: A systematic review and meta-analysis. | Diabetes & Metabolism | meta-analysis of 24 randomised trials, 1,383 participants: significant reduction in visceral fat with endurance exercise, effect on hepatic steatosis close to the significance threshold