Introduction
Atrial fibrillation (AF) is the most common clinical arrhythmia. While
some rare monogenic or Mendelian forms of AF have identified non-AF
manifestations1, the common forms of AF do not. Recent
insights into the genetics of common AF suggest that the disease, while
manifesting in the atrial myocardium, may also have a somatic “non-AF”
phenotype2. Patients with AF often manifest with sinus
node dysfunction, heart block, atrial flutter, cardiac thrombi, coronary
disease and other atrial arrhythmias. The relationship between these
discrete phenotypes is thought to be secondary to overlapping upstream
biological risk factors. We sought to investigate the potential presence
of a non-AF phenotype of AF through associations with polygenic AF risk.
Since the first identified single nucleotide variant (SNV) predisposing
to AF was identified3, a cascade of other variants
have been implicated in the heritable contribution to AF. This has
allowed the creation of polygenic risk scores which aggregate risk
across multiple loci to approximate the risk to an individual. While the
clinical utility of such scores remains uncertain, they do offer
opportunity to define whether discrete clinical features are associated
with specific genetic risk. We sought to determine if more granular
clinical features might be associated with a defined polygenic risk to
AF.