The highest inflammation marker came with 43% more heart attacks and strokes

Nearly 480,000 British adults gave blood decades ago, and the pattern in those samples now looks uncomfortable. Those whose blood carried the most glycoprotein acetyls, an NMR-measured signal of chronic inflammation, went on to suffer far more cardiovascular events than people at the low end of the same scale.

Declan O’Regan, a professor at Imperial College London and the MRC Laboratory of Medical Sciences, led the work. His group reports that people in the top fifth for the marker had a 43% higher risk of heart attack and stroke than people in the bottom fifth, and that the signal showed up even in people with no prior heart disease.

The 43% figure and what it counts

Imperial College London’s news release quotes the finding directly: people with the highest levels of inflammation, the top 20%, had a 43% higher risk of heart attack and stroke than those with the lowest, the bottom 20%. The paper appeared in the European Journal of Preventive Cardiology in September 2026, with Mattia Corianò as lead author.

The wording of the endpoint matters, because the team’s own preprint draws it wider than the two events in the public summary. In that medRxiv full text, the authors define the primary outcome as a composite of cardiac arrest, heart failure, stroke or other cerebrovascular events, myocardial infarction, or cardiovascular death, and report an adjusted hazard ratio of 1.43 (95% confidence interval 1.38 to 1.49) for the highest GlycA quintile against the lowest. That is the 43%: a composite of several cardiovascular outcomes, not a count of heart attacks and strokes alone.

The journal page could not be opened for this report, so the published wording was not compared line by line with the preprint; the group, the quintile comparison and the 43% match in both.

GlycA as the inflammation signal

Glycoprotein acetyls are a cluster of sugar-decorated proteins whose combined signal can be read from a single blood sample by nuclear magnetic resonance spectroscopy, the same technique that profiled the UK Biobank blood. The preprint describes chronic inflammation as quantified this way. Unlike a standard C-reactive protein test, which can spike and fall with a passing infection, GlycA is thought to track a steadier, longer-running inflammatory state.

A separate 2023 preprint led by Daisy Crick, with Abigail Fraser among the authors, compared the two markers across the lifespan using ALSPAC and UK Biobank data. GlycA readings taken years apart agreed with each other more closely than high-sensitivity CRP readings did; between ages 15 and 18, the correlation was 0.52 for GlycA against 0.40 for CRP. That stability is the reason a single baseline measurement can say something about risk over many years.

Cohort, follow-up and the adjusted numbers

Figures in the O’Regan group’s preprint PDF show how much the estimate shifts as models add controls. The covariate-adjusted hazard ratio for the top quintile was 1.64; adding lipid profiles brought it to 1.46; and a model that treated death from non-cardiovascular causes as a competing risk gave 1.42. The 43% sits inside that range of 42% to 64%, and the lower, more conservative end is the one carried into the public summaries.

Follow-up was long. The preprint reports a median of 15.3 years, and 74,336 participants experienced a major cardiovascular event while 55,076 died from any cause. Analyses accounted for age, sex, body surface area, statin use and cholesterol measures including LDL, HDL, triglycerides and lipoprotein(a), which is why the result cannot be waved away as a restatement of high cholesterol. Imaging-adjusted models were smaller, drawing on 72,541 participants with complete heart scan data, while the lipid-adjusted analyses covered 454,821 people. Each model answers a slightly different question, and the single 43% carried into public summaries is one of several the authors computed from the same cohort.

Heart structure and what the researchers claim

The Imperial team also paired the blood data with heart imaging and genetic information. According to News-Medical’s account of the release, inflammation was linked with thicker heart walls, smaller chambers and impaired filling, changes that may develop silently for years. O’Regan said millions of people could be living with hidden inflammation that is slowly changing their heart and causing long-term damage.

Funding came from the Medical Research Council, the British Heart Foundation and the NIHR Imperial Biomedical Research Centre, as the ScienceDaily posting of the release lists.

What the data cannot do is prove that lowering GlycA would lower the risk. UK Biobank is an observational resource, and a hazard ratio of 1.43 describes association between a marker and later events, not a tested treatment effect. Whether acting on the marker would prevent a single heart attack is a question the Imperial group’s design leaves open.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview