Fatty liver signs are mostly silent until fat-driven inflammation or scarring begins to impair liver function, so early clues often appear in blood work or imaging rather than in how you feel. When symptoms do surface—persistent fatigue, right-upper-abdominal discomfort, or unexplained metabolic changes—they typically mean the liver has already been accumulating fat for years. This explainer walks through the mechanisms behind each sign, the numbers that define them, and which findings deserve urgent attention.

The Mechanism Behind a Quiet Liver

Metabolic dysfunction-associated steatotic liver disease (MASLD, the current umbrella term replacing "NAFLD") begins when hepatocytes accumulate triglyceride droplets faster than they can export or burn them. Insulin resistance is the usual trigger: it drives lipolysis in adipose tissue, floods the liver with free fatty acids, and simultaneously increases de novo lipogenesis—the liver's own fat-synthesis pathway. The result is steatosis, defined histologically as fat in more than 5% of hepatocytes.

Simple steatosis is often reversible. The danger emerges when fat triggers lipotoxic stress, mitochondrial dysfunction, and inflammatory signaling. This is the steatohepatitis stage, where ballooned hepatocytes and lobular inflammation appear. Over years, repeated injury activates stellate cells, which deposit collagen and drive fibrosis. Fibrosis is the variable that predicts outcomes: cirrhosis, portal hypertension, and hepatocellular carcinoma.

A key point: the liver has no pain fibers in its parenchyma. Discomfort comes from stretching of the capsule or from inflammation reaching it. This is why early steatosis produces no reliable symptoms at all.

Quantified Context: What the Numbers Actually Mean

Radiology and pathology use different thresholds, and reconciling them matters.

On ultrasound, steatosis becomes visible when hepatic fat exceeds roughly 20–30% of liver volume—well above the 5% histologic threshold. That gap explains why a "normal" ultrasound does not exclude early disease. Controlled attenuation parameter (CAP) on transient elastography estimates fat with a cutoff near 248 dB/m for mild steatosis and above 280 dB/m for moderate-to-severe. MRI-PDFF is the most sensitive noninvasive measure, detecting fat fractions as low as 5%.

Liver enzymes are a poor screening tool. Aminotransferases can be normal in up to 80% of people with steatosis, and ALT thresholds historically used (around 40 U/L) miss disease in many patients. Major gastroenterology societies have proposed lower upper limits—roughly 30 U/L for men and 19 U/L for women—precisely because the old cutoffs under-detect.

Fibrosis staging uses the F0–F4 scale. F2 (significant fibrosis) is the usual threshold for specialist referral, and F4 means cirrhosis. Vibration-controlled transient elastography gives a liver stiffness reading in kilopascals; values above roughly 8 kPa suggest significant fibrosis, and above 12–15 kPa suggest advanced disease, though inflammation and congestion can inflate readings.

Symptoms, Stratified by Disease Stage

Most people with simple steatosis report nothing. When symptoms appear, they tend to cluster at the steatohepatitis and fibrosis stages.

Fatigue is the most common complaint, but it is nonspecific and overlaps with sleep apnea, which is itself strongly associated with fatty liver. Right-upper-quadrant discomfort is dull and poorly localized, reflecting capsular stretch rather than liver pain. Hepatomegaly (enlarged liver) may be palpable on exam in advanced steatosis. Pruritus and jaundice suggest cholestatic or decompensated disease and warrant prompt evaluation. Ascites, confusion, or gastrointestinal bleeding indicate cirrhosis complications and are emergencies.

Plain language: by the time you feel something, the liver has usually been sick for a long while.

The counterintuitive detail: symptom severity correlates poorly with fat volume. A person with 40% hepatic fat and no inflammation may feel fine, while someone with modest fat and active steatohepatitis feels unwell. The biology that matters is inflammation and fibrosis, not the fat percentage alone.

Reading the Signs: Comparison of Detection Methods

Method What it detects Key threshold Limitation
Ultrasound Moderate-to-severe steatosis ~20–30% fat Misses mild steatosis
CAP (elastography) Steatosis grade ~248 dB/m mild Confounded by obesity
MRI-PDFF Precise fat fraction ≥5% abnormal Cost, availability
ALT/AST Liver injury Lower cutoffs: 30/19 U/L Normal in most cases
FIB-4 index Fibrosis risk >1.3 indeterminate, >2.67 high Age-dependent accuracy
Transient elastography Fibrosis (stiffness) >8 kPa significant Inflammation inflates values
Liver biopsy Reference standard Histology Invasive, sampling error

Reconciling thresholds explicitly: a normal ultrasound plus a normal ALT does not rule out MASLD. If metabolic risk factors are present—central obesity, type 2 diabetes, dyslipidemia—a FIB-4 calculation and elastography are more informative than either test alone.

Edge Cases That Break the Pattern

Lean MASLD occurs in people with normal body mass index, driven by visceral adiposity, genetic variants (notably PNPLA3), or rapid weight loss. These patients are often missed because clinicians do not suspect fatty liver in a lean person.

Alcohol-related liver disease can look identical on imaging, and the distinction is clinical, not radiological. Current frameworks separate MASLD from alcohol-associated disease and from a combined category (MetALD) based on alcohol intake thresholds—generally more than 20–30 g per day for women and 30–40 g per day for men, though definitions vary by guideline.

Medications are an underappreciated cause: amiodarone, methotrexate, tamoxifen, and corticosteroids can produce steatosis or steatohepatitis independent of metabolic risk. Rapid weight loss—more than about 1.5 kg per week—can transiently worsen inflammation even as fat falls.

Pregnancy-associated acute fatty liver is a rare obstetric emergency with a different mechanism entirely and does not belong in the routine MASLD conversation.

Professional caveat: an isolated abnormal ALT in a patient with obesity is more likely steatohepatitis than steatosis alone, but it is not diagnostic. Normal enzymes do not exonerate the liver, and abnormal enzymes do not confirm the cause.

Practical Actions and Thresholds

If you have type 2 diabetes, obesity, or metabolic syndrome, ask for a FIB-4 calculation—it requires only age, AST, ALT, and platelets, and costs effectively nothing beyond a standard panel. A FIB-4 above 1.3 warrants elastography or specialist referral; above 2.67 suggests advanced fibrosis risk. Do not rely on a single normal ALT. Request the lower-threshold interpretation (30 U/L men, 19 U/L women) when metabolic risk factors exist. A 5–10% body-weight loss reduces hepatic fat substantially; 7–10% loss improves steatohepatitis histology in trial data. Aim for gradual loss, not crash dieting. Limit added sugars and fructose-sweetened beverages; they feed de novo lipogenesis directly. Avoid alcohol entirely if steatohepatitis or fibrosis is suspected—there is no safe threshold once injury is present.

Time Course: What Happens and When

Timeframe What changes
Weeks Hepatic fat rises with sustained caloric surplus and insulin resistance
3–6 months Steatosis becomes detectable on imaging as fat exceeds ~20%
6–12 months Inflammation may develop; ALT can rise or stay normal
Years (typically 5–10+) Fibrosis progresses in a subset; rate varies widely
10–20 years A minority progress to cirrhosis; HCC risk emerges
Post weight loss Fat falls within weeks; fibrosis regresses slowly over years

The progression rate is highly individual—some people remain at simple steatosis for decades, while others fibrose quickly. Genetic background, diabetes, and alcohol use shift the trajectory.

Limitations and What the Evidence Cannot Yet Resolve

No blood test yet diagnoses steatohepatitis without imaging or biopsy; the field relies on composite scores with imperfect sensitivity. FIB-4 performs less well in people under 35 and over 65. Elastography readings vary with operator technique, probe choice, and recent food intake, so fasting before the exam matters.

We also lack long-term outcome data on how aggressively to treat lean MASLD or how to monitor patients whose enzymes normalize on therapy. The mechanisms are well understood; the personalized thresholds are not. Treat any single number as a clue, not a verdict, and interpret it alongside metabolic context and, where indicated, specialist assessment.