Do liver enzymes tell me anything useful about hepatic fat?

I have three data points across nine months, which I hope is enough to see a trend.

I suspect the usual explanation for this is wrong, or at least incomplete.

I am aware this may have a boring answer. I would still like the boring answer stated clearly.

Can someone derive this rather than assert it?

Worth being precise here: reference intervals for alanine aminotransferase differ between laboratories and have been revised downward over time, so the same value can be normal on one report and elevated on another.

Reference upper limits around forty units per litre for ALT are conventional rather than physiological; several groups have argued for limits closer to thirty for men and twenty-five for women.

Very rapid weight loss can transiently worsen liver biochemistry, which is one of several arguments against pursuing the steepest possible trajectory.

Get a baseline before you start anything, because it converts an uninterpretable result into an interpretable one for the cost of one blood draw.

Start with the pattern. A hepatocellular pattern with alanine aminotransferase above aspartate aminotransferase points one way; a cholestatic pattern with alkaline phosphatase and gamma-glutamyl transferase points another.

Weight loss of ten per cent or more typically reduces transaminases substantially in people whose elevation was driven by hepatic fat, which is the majority of mild elevations in this population.

Creatine kinase, alkaline phosphatase and bilirubin together tell you which compartment the abnormality is in, and ordering the transaminases alone throws that information away.

Hy's law and its variants are the standard framework for identifying drug-induced liver injury in trials and are why bilirubin is measured alongside transaminases rather than instead.

Read the pattern before the magnitude, and the magnitude in multiples of the upper limit.

edited 26 Jun 2026 by lipid_panel_q — updated for the 2026 guidance change

Answer first: mild transaminase elevation is common in this population before any drug is involved, and the usual cause is hepatic steatosis rather than anything acute.

Gamma-glutamyl transferase is sensitive and unspecific: it rises with alcohol, with several medications and with fatty liver, and an isolated elevation rarely changes anything on its own.

To be exact about it, an isolated result three weeks after starting anything is difficult to interpret. A baseline taken before starting makes the same result trivially interpretable, which is the argument for baselines.

Trials in this class have generally reported improvement rather than deterioration in liver biochemistry, consistent with the weight-mediated mechanism.

New training, alcohol and over-the-counter products first. Then the interesting hypotheses.

Before attributing anything to a compound, account for alcohol, muscle injury from new training, and any over-the-counter product being taken alongside.

Alanine aminotransferase is relatively liver-specific; aspartate aminotransferase is also present in muscle, heart and red cells. A raised AST with a normal ALT after heavy resistance training is usually muscle, and creatine kinase settles the question.

The caveat is direct: rising liver enzymes with jaundice, dark urine or right-upper-quadrant pain is a same-day clinical problem, not a forum question.

Transaminases plus bilirubin plus alkaline phosphatase, or you have not measured enough to conclude anything.

The short version: enzymes in this class usually fall as weight falls, and a rise is a reason to look for a cause rather than to assume a drug effect.

Hy's law describes the combination that matters: transaminases above three times the upper limit together with bilirubin above twice the upper limit and no cholestatic explanation. That combination is a signal; isolated mild transaminase elevation is not.

Population data show a substantial fraction of adults with mild transaminase elevation attributable to hepatic steatosis, which sets the base rate against which any new finding should be read.

Attributing an enzyme change to a compound requires a baseline, and most people asking have not got one.

Isolated mild elevation is common and usually improves with weight loss.

edited 8 May 2026 by fib4_reader — clarified the distinction between purity and content