The ALT Problem

ALT is the enzyme a liver cell spills when it is in distress, and a blood panel reads it as a yes-or-no question: inside the reference range or out. That framing fails quietly in fatty liver disease. A substantial share of people with MASLD — including some with inflammation and even cirrhosis — carry values their lab report stamps "normal," and the ranges themselves were built on populations that included undiagnosed liver disease. Three failure modes, the lower thresholds that catch more risk, and what a rational workup looks like.

What ALT Actually Measures

Alanine aminotransferase is an enzyme concentrated in liver cells; when hepatocytes are injured, ALT leaks into serum. It is a satellite photo of active damage, not of fat. A fatty liver storing triglycerides quietly — the simple steatosis that makes up most of MASLD — has few dying cells and little to leak, and ALT correlates only loosely with the amount of fat in the liver, more loosely still with inflammation. The enzyme's half-life is measured in days. And the outcome that matters most — fibrosis, the slow accumulation of scar tissue — is precisely what a leak-marker was never built to see.

Failure Mode One: Normal ALT, Sick Liver

The clearest demonstration comes from biopsies. Mofrad and colleagues (2003) studied 51 people with NAFLD and normal ALT values — most discovered through hepatomegaly or living-donor evaluation — and compared them with 50 consecutive patients whose ALT was elevated. Under the microscope the groups were statistically indistinguishable: similar fat, similar inflammation, comparable scarring. Within the normal-ALT group, twelve had bridging fibrosis and six had cirrhosis — the full histologic spectrum in people whose labs said nothing was wrong. Pooled analyses since (Ma et al., 2020) put the normal-ALT share of MASLD near 25%; even among those with steatohepatitis, about 19% stay in range. Normal is a reading of the enzyme, not of the liver.

Failure Mode Two: The Reference Range Is Contaminated

Where did the range on your lab printout come from? Mostly "healthy" reference populations — frequently blood donors — screened by questionnaire. Prati and colleagues (2002) re-derived the limits after excluding people with hepatitis C antibodies, risky exposures, and metabolic risk factors: roughly 30 U/L for men and 19 U/L for women, versus the 40 and up that many lab printouts say. Ranges are also lab-specific: across 11 NASH Clinical Research Network laboratories, upper limits ran 35–79 U/L in men and 31–55 U/L in women (Neuschwander-Tetri et al., 2008). The old ranges had absorbed sick livers and called them normal. US-population work by Ruhl and Everhart (2012), using NHANES data, found the cutoff best separating hepatitis C from low-risk adults near 29 U/L for men and 22 U/L for women; tuning for sensitivity pushed thresholds lower still, to roughly 24 and 18. Exact numbers differ by method and by lab; the direction is consistent. A value in the 30s that your lab calls normal is, on healthy-population arithmetic, high.

Failure Mode Three: Risk Is Graded Inside "Normal"

Even within the green zone, higher is not neutral. Kim and colleagues (BMJ, 2004) followed a large Korean cohort — 94,000 men and 47,000 women — for eight years and watched liver-disease mortality climb across aminotransferase bands that were all officially normal: compared with people under 20 IU/L, an ALT of 30–39 IU/L — still inside the normal range — carried roughly 9-fold higher liver-disease mortality in men (95% CI 7.9 to 11.5). The association runs through the range; it does not switch on at the printed line. Two honest caveats: absolute risks from a hepatitis-B-endemic country do not transfer cleanly, and association is not causation — a high-normal ALT is a flag for attention, not a diagnosis.

The Thresholds That Flag Earlier

Because the conventional ranges absorbed sick people, the evidence converges on lower, sex-specific ceilings: Prati's re-derivation (30 men / 19 women) and Ruhl and Everhart's cutoffs at 29 and 22, with sensitivity-tuned variants near 24 and 18. The shared direction: values in the 30s, especially in women, deserve a second thought even when stamped normal. The trade-off is real — the tighter cutoffs would label roughly a third of US men abnormal. A lower threshold is a trip-wire for conversation, not a verdict. Guidelines have partly converged: the American College of Gastroenterology puts a truly healthy ALT at 29–33 U/L in men and 19–25 U/L in women, and the 2023 AASLD practice guidance advises treating ALT above 30 U/L as abnormal (Kwo et al., 2017; Rinella et al., 2023).

When ALT Is High: A Sensible Workup Sequence

An isolated ALT elevation triggers a standard diagnostic cascade — clinician territory, described, not prescribed. The logic: exclude the dangerous and the common first.

• 🔁 First, repeat it: confirm persistence on a second draw weeks later, fasted, alongside AST, GGT, and bilirubin.
• 🦠 Exclude viral hepatitis: hepatitis B and C serology — cheap, decisive, first thing a workup owes you.
• 🍺 Level about alcohol and meds: an honest intake history plus a review of all drugs — both causes are common and only treatable when named.
• 🩸 Metabolic panel: fasting glucose or HbA1c, lipids, ferritin, TSH — MASLD announces itself through its traveling companions.
• 🔊 Image and score: ultrasound for structure; elastography or fibrosis scores (FIB-4) where risk persists — which livers need watching is the question that matters most.

If ALT Is Normal but Suspicion Is Not

The mirror problem is the person with every metabolic risk factor — central weight, high triglycerides, low HDL, rising glucose — whose ALT reads 24 and is told all is well. Normal ALT lowers the probability of inflammatory MASLD; it does not remove it, and Mofrad's cirrhosis-in-range patients are the caution. The decision point in fatty liver care is not the enzyme but the fibrosis estimate behind it, where scores and elastography do the sorting. And the response to suspected MASLD is the same lifestyle response regardless of enzyme — the dose-response what actually reverses it maps does not wait for a lab to turn red. For the wider principle, biomarkers that mislead is this site's tour.

What Moves ALT Day to Day

• 📉 Weight loss lowers it: in a 15-month diet-and-exercise program in 31 patients, ALT improvements tracked the weight lost and regressed when weight returned (Hickman et al., Gut, 2004).
• 🏋️ Muscle moves it up: strenuous exercise can transiently raise ALT through muscle, not liver — one reason single post-workout draws mislead.
• 🍔 Surplus calories raise it: sustained energy surplus and fast carbohydrate delivery — the assembly line why the liver gets fat explains — raise liver fat first, enzymes later, if ever.
• 🍷 Alcohol raises it honestly: the most common reversible cause of mild elevation; weeks off move the number.

Questions, Answered Briefly

• 🤔 "My ALT is 34 and the lab says normal." The printed range came from a mixed population; healthier-population studies put the men's limit near 30 (Prati et al., 2002). With metabolic risk factors, 34 is a conversation, not a clearance.
• 🩺 "Can fatty liver exist with normal labs?" Yes — about a quarter of MASLD in pooled analyses, and Mofrad's series included fibrosis and cirrhosis in range. ALT measures distress, not fat or scar.
• 🔁 "One high reading — now what?" Repeat in a few weeks, fasted, no hard training beforehand. Persistence, magnitude, and trend drive the workup.
• 🎯 "Which number should I watch?" The trend across panels — and once suspicion exists, fibrosis estimation (FIB-4, elastography), the number that predicts the future. The progression odds page explains why.

Related Topics