Design: Prospective cohort study with multivariable-adjusted Cox models and sex-stratified analyses
Population: 318,308 adults free of clinical liver and cardiovascular disease at baseline, median follow-up 14.0 years
Primary outcome: Major incident cardiovascular events: myocardial infarction, heart failure, atrial fibrillation, ischaemic stroke and cardiovascular death
Effect: 32,637 events (10.3%). Top vs bottom quartile: fatty liver index adjusted hazard ratio 1.46 (95% CI 1.40-1.52), FIB-4 1.66 (1.60-1.73), albumin-bilirubin 1.42 (1.37-1.47); top quartile of all three 2.53 (2.38-2.70)

Fatty liver has been linked to cardiovascular risk repeatedly, usually one marker at a time. This prospective cohort of 318,308 adults, all free of clinical liver and cardiovascular disease at baseline, asked what three separate non-invasive indices contribute independently and together: the fatty liver index for steatosis, FIB-4 for fibrosis, and the albumin-bilirubin score for functional reserve.

Over a median 14 years there were 32,637 cardiovascular events, 10.3% of the cohort. Each index predicted independently after multivariable adjustment: comparing top with bottom quartile, adjusted hazard ratios were 1.46 for the fatty liver index (95% CI 1.40-1.52), 1.66 for FIB-4 (1.60-1.73) and 1.42 for albumin-bilirubin (1.37-1.47). Participants in the top quartile of all three carried a hazard ratio of 2.53 (2.38-2.70). The pattern differed by sex — the steatosis index was more strongly associated in women, the fibrosis and function indices in men.

Every one of these is computed from tests already on a routine panel: transaminases, platelets, albumin, bilirubin, triglycerides, plus waist circumference and BMI. That makes this unusually easy to act on, and the fibrosis index doing the most work is the finding worth carrying — it is fibrosis rather than fat that marks the risk.

Calculate FIB-4 from tests you already have when assessing cardiovascular risk in metabolic disease
Note that the fibrosis index outperformed the steatosis index; fat alone is the weaker signal
Treat concordant elevation of all three as a substantially higher risk group
Do not use these to replace an established cardiovascular risk score; no comparison was made
Remember waist circumference is needed for the fatty liver index and is often not measured

Why it matters

It turns three scores already computable from routine bloods into cardiovascular risk information, and puts fibrosis ahead of fat.

The statistics, in plain English

Hazard ratios of 1.42 to 1.66 with intervals this narrow reflect an enormous sample rather than a large effect: these are real but modest associations, and the width tells you about precision, not importance. The combined figure of 2.53 is described as synergistic, but it is what you would expect from three correlated markers of the same metabolic process being present together, and no formal interaction test is reported. Nothing here shows these indices add to an existing risk score, which is the question that would decide clinical use.