Older adults and working-age people whose atrial fibrillation is caught through screening, rather than through symptoms or clinical visits, face roughly triple the risk of developing heart failure compared to those without the arrhythmia. That finding, drawn from multiple peer-reviewed cohorts spanning different age groups and countries, challenges the assumption that screen-detected atrial fibrillation is a harmless incidental finding. With population-level screening programs expanding across Europe and under active discussion elsewhere, the question is no longer just whether to screen but what to do once the irregular rhythm is found.
Screen-detected atrial fibrillation and the heart-failure gap
The risk signal is sharpest in a recent analysis of working-age adults published in JAMA Network Open, which tracked cardiovascular outcomes in people whose atrial fibrillation was identified through screening rather than through standard clinical care. In that cohort, screen-detected atrial fibrillation was associated with a markedly higher incidence of subsequent heart failure. The finding matters because working-age populations are not the traditional target of atrial fibrillation screening. If the arrhythmia already carries elevated heart-failure risk in younger, presumably healthier adults, the implications for older screening populations are even more pressing.
Parallel evidence from Sweden’s STROKESTOP program reinforces that concern. The original trial was a multicentre randomised study that invited 75- and 76-year-old men and women for intermittent ECG screening, as reported in a Lancet analysis of the program. That work established the infrastructure and cohort that later investigations have used to examine downstream outcomes, including heart failure. Its successor, STROKESTOP II, added a blood biomarker step: participants first had their N-terminal B-type natriuretic peptide (NT-proBNP) levels measured, and those with elevated results were then directed to handheld ECG monitoring. The stepwise design was intended to improve screening efficiency, but it also generated a dataset linking a cardiac stress biomarker to the presence or absence of silent atrial fibrillation at the point of detection.
A separate line of evidence from the PREVEND study, published in Open Heart, directly compared outcomes between screen-detected and clinically detected atrial fibrillation. The researchers found that screen-detected cases carried adverse outcome profiles similar to those found through routine clinical care. In other words, the absence of symptoms at the time of detection did not translate into a lower risk of complications. That result dismantles the reassuring notion that if a patient feels fine, the arrhythmia is unlikely to cause trouble.
STROKESTOP, PREVEND, and the converging data on silent arrhythmia
The STROKESTOP program’s design is central to understanding why the heart-failure signal is credible. The original Swedish trial, described in detail in a Circulation report, screened 75- and 76-year-old men and women using intermittent ECG recordings and an invitation-based strategy. Participants who were found to have previously undetected atrial fibrillation could then be offered anticoagulation therapy to reduce stroke risk. The trial’s primary endpoint was a composite clinical outcome, but the cohort’s size and follow-up duration have made it a rich resource for secondary analyses, including those focused on heart failure and other downstream events.
STROKESTOP II refined the approach. Its protocol used NT-proBNP as a first-pass risk filter before ECG screening. NT-proBNP is a peptide released by the heart under mechanical stress, and elevated levels can signal early cardiac dysfunction even before symptoms appear. By measuring NT-proBNP at the screening stage, STROKESTOP II created a natural experiment: researchers could later examine whether participants with both elevated NT-proBNP and newly detected atrial fibrillation went on to develop heart failure at rates higher than those with atrial fibrillation alone.
That layered dataset raises a hypothesis worth testing. If serial NT-proBNP measurements, taken not just at the initial screen but at regular intervals afterward, could identify a subgroup whose biomarker levels climb faster than average, clinicians might be able to flag patients whose heart-failure risk exceeds even the tripled baseline rate. Such a subgroup could be candidates for earlier intervention, whether through rate or rhythm control, blood pressure management, or closer surveillance for early signs of fluid retention and reduced cardiac output. No published trial has yet reported individual-level follow-up data linking specific NT-proBNP thresholds at screening to subsequent heart-failure events, so the hypothesis remains untested in a prospective design.
The PREVEND study adds geographic and methodological diversity to the evidence base. Conducted in the Netherlands, PREVEND used a community-based cohort and compared cardiovascular outcomes between participants whose atrial fibrillation was found through systematic screening and those diagnosed through conventional clinical pathways. The study’s conclusion, that screen-detected atrial fibrillation is not a benign variant but instead tracks with similar rates of heart failure, stroke, and mortality as clinically detected arrhythmia, suggests that the mechanism of discovery does not dilute the underlying disease risk. In practice, a rhythm strip obtained at a public screening event may carry the same prognostic weight as one recorded during an emergency visit for palpitations.
From detection to management: closing the heart-failure loop
Together, these data streams point to a gap between detection and management. Screening programs were initially conceived mainly as stroke-prevention tools, focused on identifying candidates for anticoagulation. Yet if screen-detected atrial fibrillation carries a substantial and early risk of heart failure, then stroke prevention alone is too narrow a goal. Clinicians and health systems need structured pathways that address rhythm control, comorbidity optimisation, and heart-failure surveillance from the moment the arrhythmia is found.
One implication is that screening initiatives should incorporate baseline assessment of cardiac structure and function. For some patients, that may mean echocardiography to look for left atrial enlargement, reduced ejection fraction, or valvular disease. For others, particularly in resource-limited settings, a combination of NT-proBNP testing, blood pressure measurement, and careful clinical history may offer a pragmatic risk stratification approach. The key is to move beyond a binary “atrial fibrillation: yes or no” framework toward a graded understanding of heart-failure vulnerability.
Another implication is the need for follow-up that is both timely and sustained. Screen-detected atrial fibrillation often emerges in people who have not previously engaged intensively with cardiac care. Without clear arrangements for repeat visits, medication titration, and monitoring of symptoms such as exertional breathlessness or ankle swelling, the opportunity to blunt heart-failure progression may be lost. Embedding screening programs within primary care networks, rather than running them as stand-alone campaigns, could help ensure that newly diagnosed patients are not left navigating complex treatment decisions alone.
For policymakers, the emerging evidence base poses a challenge and an opportunity. On one hand, the prospect of tripled heart-failure risk raises questions about the cost-effectiveness of large-scale screening if downstream care remains fragmented. On the other, it suggests that well-designed programs-those that combine targeted biomarker use, structured pathways to cardiology assessment, and ongoing primary care follow-up-might prevent not just strokes but also a significant burden of heart-failure hospitalisations.
Future research will need to clarify which elements of such pathways deliver the greatest benefit. Randomised comparisons of different follow-up strategies for screen-detected atrial fibrillation, including varying intensities of NT-proBNP monitoring and different thresholds for initiating rhythm-control therapies, could help refine practice. Until those data arrive, the converging message from STROKESTOP, PREVEND, and working-age cohorts is that silent arrhythmia should not be treated as a minor incidental finding. Once atrial fibrillation is uncovered-whether by opportunistic pulse checks, handheld devices, or formal population screening-the work of preventing heart failure has already begun.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
