Does Genetic Risk Determine Who Benefits From AFib Screening?

This article covers what a genetic secondary analysis of the LOOP implantable loop recorder trial found about polygenic risk and screening benefit, and how that sits against the original LOOP results and broader AFib genetics research. It does not cover consumer genetic testing products or non-AFib polygenic scores.

A prespecified analysis of the LOOP trial tested whether a polygenic risk score for AFib changes who benefits from loop recorder screening for stroke prevention, and it did not find that genetic risk level changed the screening effect on stroke and systemic embolism. That means the original LOOP result, screening did not reduce stroke across the trial as a whole, held regardless of where someone fell on the genetic risk spectrum.

The assumption: high genetic risk should mean screening pays off more

There's a logical appeal to the idea that AFib risk isn't evenly distributed, and that screening should work best on the people whose biology makes them most likely to develop it. If genetic risk scores can identify who is quietly primed for AFib, it seems like screening ought to be aimed at that group, and skipped or deprioritized for everyone else. So that's the question this article is built around: does knowing your genetic risk tell you whether AFib screening is worth doing.

The LOOP study itself already complicated the simpler version of this question. It tested implantable loop recorder monitoring against usual care in older adults with stroke risk factors, tracking what happened with stroke and systemic embolism across the full trial population. A secondary analysis of that same trial then asked the sharper question: among people already enrolled, did genetic risk change who the screening helped.

What the genetic analysis actually tested

The polygenic risk score used in this analysis is built to predict susceptibility to developing AFib itself, not to predict who will suffer a stroke, and not to predict who will respond better to being monitored. Those are three different questions, and it's easy to blur them. The LOOP genetic analysis specifically modeled the interaction between someone's randomization arm (screened vs usual care) and their polygenic risk score, looking at whether that interaction changed the composite outcome of stroke and systemic embolism.

That framing matters because a genetic score could plausibly do several different jobs. It could identify who is likely to develop AFib at all. It could identify who, once AFib is found, is at higher stroke risk. Or it could identify who benefits more from the act of screening and treating with anticoagulation once AFib is detected. The analysis was built to test the last of those, screening benefit specifically, using the full trial's randomized comparison rather than just looking at who ended up with an AFib diagnosis.

A review of AFib genetics research has laid out this distinction explicitly: genetic variants are increasingly well characterized as risk factors for developing AFib, but whether that genetic knowledge translates into identifying who benefits from specific screening or treatment approaches is a separate, still-open question in the field. The secondary analysis of LOOP is one of the few pieces of evidence that has directly tested that translation for AFib screening rather than just AFib prediction.

Why this sits differently than other AFib screening results

Other AFib screening trials have shown that different testing methods find different numbers of AFib cases. A trial testing a 14-day continuous monitor against usual care in adults aged 70 and older was designed specifically to see whether screening plus anticoagulation lowers stroke and bleeding hospitalization rates, and that trial's stroke outcome results sit alongside LOOP as one of the direct tests of whether finding more AFib actually changes stroke rates, separate from the genetic question entirely.

The genetic analysis doesn't undo or replace those detection questions. It answers something narrower: given that LOOP already ran its randomized comparison of screening versus usual care, did the size of the screening effect on stroke depend on where someone landed on a genetic risk gradient. It didn't. That's a specific negative finding about effect modification, not a claim that genetic scores are useless for anything else in AFib. Identifying who is likely to develop the arrhythmia in the first place is a separate and better-supported use of these scores.

I'd also separate this from the broader conversation about whether finding more AFib with any device actually prevents strokes, since the genetic analysis is nested inside a trial that was already testing that exact question for loop recorders specifically.

What this doesn't settle

A single prespecified analysis within one trial population is not the same as a settled answer across all AFib screening contexts. The polygenic risk score tested here is specific to AFib susceptibility, and the population was already narrowed to older adults with stroke risk factors before genetic risk was even considered. Whether a different genetic score, a different age group, or a different screening technology would show a different pattern isn't addressed by this analysis.

The broader question of whether AFib screening reduces stroke at all remains its own live debate, separate from genetics. Some trials testing intensive monitoring approaches have looked at this directly, and the field is still working out where the benefit shows up, if anywhere, and for whom, independent of any genetic layer.

Common questions