Key takeaways
- MASLD is excess liver fat alongside metabolic dysfunction; MASH is the inflamed subset that drives scarring.
- Fat accumulates by three routes that insulin resistance opens at once: more delivery, more manufacture, and a fixed ceiling on disposal.
- There are few GLP-1 receptors on the hepatocyte, so the effect on the liver arrives upstream rather than directly.
- Liver fat and ALT improve early; fibrosis, the endpoint that predicts outcomes, moves slowest.
- Order ALT and fasting insulin, and get a fibrosis estimate rather than assuming a bright liver on ultrasound is benign.
Most fatty liver is found by accident. A routine panel comes back with a mildly raised ALT, or an abdominal ultrasound ordered for something else mentions a "bright" or "echogenic" liver, and the report closes with the words "clinically insignificant". For a lot of people that is the last time anyone looks. It is worth understanding what has actually been found, because fat in the liver is not a liver problem that happened to a metabolic patient. It is a metabolic problem that happens to be visible in the liver.
MASLD and MASH, defined
MASLD — metabolic-dysfunction-associated steatotic liver disease — is excess fat stored inside liver cells in someone who also has features of metabolic dysfunction: excess weight, insulin resistance, type 2 diabetes, high triglycerides, high blood pressure. It is common. Global estimates put it at roughly 25-30% of adults (Younossi et al., Hepatology 2016).
MASH is the subset where fat is accompanied by inflammation and visible injury to liver cells. Around a fifth of MASLD is thought to progress that far. MASH is the version that matters clinically, because inflammation drives scarring, and scarring is what eventually costs people liver function.
The renaming — from NAFLD and NASH — was not cosmetic. The old names defined the condition by what it was not (not alcohol). The new ones define it by what it is: a liver manifestation of metabolic disease. That tells you where to aim.
Why fat ends up in the liver
The liver is not a passive storage depot. Fat accumulates there through three routes running at once, and insulin resistance opens all three.
The first is delivery. Insulin normally restrains the release of free fatty acids from fat tissue. When fat tissue stops listening to insulin, fatty acids leak into the circulation continuously rather than only between meals, and the liver takes up whatever arrives.
The second is manufacture. The liver can build fat from carbohydrate, a process called de novo lipogenesis. High insulin is the signal that switches it on. In insulin resistance, insulin is chronically high — so the liver keeps building fat even while it is already drowning in it. This is the cruel part of the mechanism: the liver's fat-making machinery stays insulin-sensitive while its glucose-handling machinery goes deaf.
The third is disposal. Fat leaves the liver by being burned or packaged into VLDL and exported, and both routes have a ceiling. When delivery plus manufacture exceeds it for long enough, the surplus is stored — and steatosis is what that surplus looks like on a scan.
Inflammation enters when stored fat is handled badly, with fatty acids shunted into toxic intermediates rather than inert triglyceride droplets. That step separates a quiet fatty liver from MASH, and it is why two people with the same amount of liver fat can have very different disease.
Where GLP-1 therapy acts
There is no meaningful population of GLP-1 receptors on the human hepatocyte. Whatever these medications do to the liver, they do from upstream. That is not a weakness in the argument — it is the argument.
- Reduced energy intake lowers the substrate arriving at the liver and reduces visceral fat, which drains directly into the portal vein.
- Improved insulin sensitivity turns down de novo lipogenesis and restores insulin's brake on fatty acid release — closing two of the three taps at once (insulin sensitivity beyond weight loss).
- Lower systemic inflammation follows the fall in visceral fat, and bears on the inflammatory step rather than the storage step.
Note what is absent: any direct hepatic drug action. Understand it as "less fat delivered, less fat made, less inflammation to injure cells" and you will predict the clinical course correctly, including the parts that are slower than people expect.
What the evidence supports
Two lines of evidence matter here, and they are not equally strong.
The first is liver fat itself, measured by MRI-PDFF, which quantifies the proton density fat fraction and is far more informative than an ultrasound impression (Idilman et al., Semin Ultrasound CT MR 2016). Liver fat falls reliably with GLP-1 therapy, and it falls in proportion to how much metabolic improvement occurs.
The second is histology — biopsy-confirmed resolution of steatohepatitis — examined in dedicated randomised work (Sanyal et al., N Engl J Med 2025). A meaningful share of treated patients improve. Results vary by agent, dose and population, so the honest framing is directional rather than precise.
It is also worth saying plainly that weight loss achieved any way improves this condition — sustained lifestyle-driven loss produces dose-dependent improvement in steatosis, inflammation and even fibrosis on biopsy (Vilar-Gomez et al., Gastroenterology 2015). Medication makes that loss achievable for people who could not otherwise reach it. It is not a separate mechanism.
Fat and fibrosis run on different clocks
This is the distinction that determines what a reasonable person should expect. Liver fat is a storage state, and storage states unwind quickly — weeks to months. Fibrosis is deposited structural protein, and structure remodels on a scale of years.
Fibrosis is also the part that predicts outcomes. Long-term follow-up of biopsied patients shows fibrosis stage — not steatosis or inflammation — tracks with mortality and liver-related complications (Angulo et al., Gastroenterology 2015). The endpoint that matters most is the one that moves slowest.
Practically: F0-F1 fibrosis with a fatty liver is largely reversible over months. At F2-F3, expect enzymes and fat to improve early while the structural picture takes far longer — and expect hepatology co-management rather than watchful waiting.
The clinical pearl: liver enzyme normalisation is usually the first hard, lab-visible sign that metabolic therapy is working — often before body weight has moved much. For someone who did not know they had fatty liver, a falling ALT is the cleanest available proof that the underlying problem is being addressed rather than just the number on the scale.
What to measure, and in what order
The sequence matters more than the completeness of the panel.
- ALT, AST and GGT at baseline. ALT is the most liver-specific and usually the first to move. GGT tracks both fatty liver and alcohol intake, which is worth knowing when reading it — see GGT as a metabolic marker.
- Fasting insulin, not just glucose. Fatty liver and insulin resistance are the same process seen from different organs, and insulin rises years before glucose (fasting insulin).
- A fibrosis estimate. Non-invasive scores from routine labs, or elastography, separate "fatty liver that needs metabolic work" from "liver disease that needs a hepatologist" — the triage step society guidance is built around (Chalasani et al., Hepatology 2018).
- Repeat enzymes at three and six months, then annually. Elastography is an annual measure; repeating it quarterly measures noise.
What to expect, and in what order
Liver fat falls before weight loss looks impressive, because visceral and hepatic fat mobilise early. ALT typically follows within the first few months; AST moves more slowly. Fibrosis markers, if they move at all, move last and only with sustained treatment.
Two things do not change on their own. Alcohol still injures a liver that already holds metabolic fat, and the combination is worse than either alone. And muscle matters here, because skeletal muscle is the largest site of insulin-stimulated glucose disposal — losing it during weight loss works against the mechanism you are relying on. Resistance training is what protects the result; the wider stack is in the insulin resistance reversal protocol and the muscle-preservation programme.
Bottom line
Fatty liver is metabolic disease that has become visible. Fat accumulates because insulin resistance increases delivery to the liver and switches on the liver's own fat production, and it clears when that upstream picture improves. GLP-1 therapy acts upstream rather than on the liver directly, which is why fat and enzymes improve early while fibrosis — the endpoint that predicts outcomes — moves slowly if at all in the short term. Measure ALT and fasting insulin, get a fibrosis estimate rather than assuming a bright liver is benign, and keep the muscle that does most of the glucose disposal. Whether medication is appropriate is a physician's decision after evaluation; the 60-second assessment is where that starts.
Educational content, not medical advice. Laboratory interpretation and any treatment decision are made by a licensed physician after individual evaluation. Individual results vary.
