Atrial fibrillation is a heartbeat irregularity that slips in and out of notice, often escaping the moments when it is measured, and so it has long resisted early discovery. A new study drawing on more than 90,000 people suggests that the eye may carry faint evidence of this hidden rhythm, with thinner retinal layers and altered vessels appearing alongside the condition. Whether a few seconds of routine retinal imaging can become a meaningful early signal, and whether earlier knowledge will truly change lives, remains an open question that the researchers themselves have yet to answer.
Retinal Scans May Reveal Atrial Fibrillation, Large Study Suggests
A common heart condition leaves visible traces in the retina
So what did they actually find?
That people with atrial fibrillation had thinner layers in the macula, especially the ganglion cell-inner plexiform layer, and different vessel measurements, across more than 90,000 people in two cohorts.
Worth being clear: the cross-sectional part shows an association. It doesn't say the thinning comes from AF or the reverse.
Is there anything that points forward in time?
In UK Biobank, 1.3% of 39,013 people without AF at baseline developed it over about four years. Each standard deviation of thinner mGCIPL carried a 27% higher hazard after adjustment.
That's a hazard associated with a measurement, not a screening test. The source gives no accuracy figures, like sensitivity or how well it would pick out individuals.
Why does the new inner nuclear layer finding matter?
Nobody had described it in AF before, so it may open a question about mechanism.
Right, but it's one finding, and the authors themselves say the causal link needs more evidence.
And the people with AF weren't otherwise similar to the others?
They were older, more often male, with more hypertension, diabetes and prior cardiovascular events.
Which is why the adjustment matters, and why we can't say how much is AF itself versus broader vascular disease. Huemer says as much: whether earlier detection improves outcomes is still unknown.
Il Polso
- Atrial fibrillation hides between heartbeats, slipping past the brief windows when clinicians think to look for it.
- Researchers in two large cohorts found thinner macular retinal layers in people with the condition, a quiet signal inside a routine eye scan.
- Each standard deviation drop in one retinal layer's thickness was tied to a 27% higher hazard of developing the condition, a striking statistical pull.
- The team is now planning multimodal imaging to test whether the retina reflects cause or only company, while the question of whether earlier detection helps patients stays unanswered.
Atrial fibrillation is a heartbeat irregularity that slips in and out of notice, often escaping the moments when it is measured, and so it has long resisted early discovery. A new study drawing on more than 90,000 people suggests that the eye may carry faint evidence of this hidden rhythm, with thinner retinal layers and altered vessels appearing alongside the condition. Whether a few seconds of routine retinal imaging can become a meaningful early signal, and whether earlier knowledge will truly change lives, remains an open question that the researchers themselves have yet to answer.
A scan that takes only seconds in an optometrist's chair may hold clues about the heart. A team led by Josef Huemer and colleagues, publishing in PLOS Digital Health, set out to test whether the retina carries visible traces of atrial fibrillation, a common and dangerous rhythm disorder that is difficult to catch because it comes and goes.
The researchers drew on two large groups: a retrospective set of more than 50,000 patients and a prospective UK Biobank cohort of nearly 40,000 people. Using macular optical coherence tomography and color retinal photographs, they compared eyes of people with and without the condition. Those with atrial fibrillation showed thinner retinal layers and changes in vessel size and branching, and the thinning of one inner layer, the inner nuclear layer, had not been described in this setting before.
The study also followed participants over time. Among nearly 39,000 people free of the condition at the start, about 1.3 percent developed it over roughly four years, and those with thinner ganglion cell layers faced a notably higher risk. The authors see promise in noninvasive eye imaging, which is already widespread in primary care, as a tool for earlier detection and risk assessment.
Huemer and his colleagues are careful about what comes next. Multimodal imaging will be needed to test whether the retinal changes drive the disease or merely accompany it, and whether finding atrial fibrillation sooner truly improves outcomes is a question future studies must still answer.
A scan that takes a few seconds in an optometrist's office may carry information about the heart. That is the argument behind a new study from Josef Huemer, MD, and colleagues, published in PLOS Digital Health, which found that people with atrial fibrillation show measurable structural differences in the retina.
The work drew on more than 90,000 people in two independent cohorts. AlzEye was a retrospective set of 50,561 patients aged 40 and older, seen between 2008 and 2018. UK Biobank was a prospective cohort of 39,539 people aged 40 to 69, recruited between 2006 and 2010. Retinal layer thicknesses came from macular optical coherence tomography, and vascular measures came from color retinal photographs. Atrial fibrillation, or AF, was identified through patient reports and hospital admissions. Huemer is affiliated with the NIHR Moorfields Biomedical Research Centre and the Institute of Ophthalmology at University College London, and with the Department of Ophthalmology and Optometry at Kepler University Hospital in Linz, Austria.
The motivation is practical. AF is common, and its prevalence is expected to double again over the next two decades. It contributes to ischemic stroke, myocardial infarction, vascular dementia and death. Yet it is hard to catch, because it comes and goes and may not appear when an ECG is running. The authors note that AF episodes themselves do not map directly onto clotting-related outcomes. They describe AF as a thrombo-inflammatory condition closely tied to underlying vascular disease, and they argue that anticoagulants and treatments aimed at vascular disease could have substantial public health impact if diagnosis came earlier and reached more communities. Their reasoning for looking at the eye was that the retina shares structural and vascular features with the brain and heart.
In both cohorts, participants with AF were on average older than those without it. They were also more likely to be male, to have hypertension and diabetes, and, most strikingly, to have had a prior major adverse cardiovascular event.
In AlzEye, where AF prevalence was 11%, the condition was associated with a smaller retinal venular caliber and thinner layers across the macula: the retinal nerve fiber layer, the ganglion cell-inner plexiform layer, the inner nuclear layer and the photoreceptor layer. Fractal dimension, a measure of vessel branching complexity, was lower in the venules but higher in the arterioles of people with AF.
UK Biobank also showed a thinner macular ganglion cell-inner plexiform layer (mGCIPL) in people with AF. Of 39,013 participants free of AF at baseline, 1.3% developed it over a mean of 4.0 years. After adjusting for sociodemographic, clinical and lifestyle factors, each standard deviation drop in mGCIPL thickness was linked to a 27% higher hazard of AF.
The thinner inner nuclear layer was a new finding. The authors say mGCIPL thinning has a plausible connection to AF-related microvascular injury, but reduced inner nuclear layer thickness had not been described in this setting before.
The authors see potential for noninvasive retinal imaging to support early detection and risk stratification, and to offer an accessible way to study AF pathophysiology and assess end-organ damage. They point to the wide availability of retinal imaging in primary care optometry across developed countries as a reason to pursue it further.
Huemer said a notoriously hard-to-diagnose heart condition leaves visible traces in the retina that quick scans can capture. The next step is multimodal imaging to build stronger evidence for a causal link. Whether earlier detection actually improves patient outcomes, he said, is something future studies will have to establish.
Citazioni salienti
Whether earlier detection translates into better outcomes for patients will need to be established by future studies; next comes multimodal imaging to test causality.— Josef Huemer, MD, lead author