Decoding Atrial Fibrillation Heterogeneity: A Nationwide Phenome-Wide Association Study in England.

BACKGROUND

Age and sex influence atrial fibrillation (AF) prevalence and outcomes, yet comprehensive data on sex- and age-specific health care utilization in patients newly diagnosed with AF remain limited.

OBJECTIVES

We aimed to address this gap using a phenome-wide association study (Phe-WAS) approach.

METHODS

We conducted a retrospective cohort study using nationwide Clinical Practice Research Datalink data linked to Hospital Episodes Statistics. Adults diagnosed with AF between 1998 and 2016 (n = 199,433) were matched 1:1 by age and sex to non-AF control subjects. Primary and secondary care interactions 5 years post diagnosis were analyzed using Phe-WAS with Bonferroni-adjusted significance thresholds. Logistic regression estimated odds ratios for conditions in AF vs control groups, stratified by sex and age. Associations were assessed for novelty through literature review and causality using published Mendelian randomization studies.

RESULTS

Patients with AF had higher rates of cardiovascular and noncardiovascular conditions than control subjects, with marked sex- and age-specific patterns. Women had more bleeding complications, pneumonia, Clostridium difficile enterocolitis, cholangitis, and electrolyte disturbances, whereas men had higher rates of alcohol-related disorders, ventricular arrhythmias, cardiac arrest, and cardiomyopathy. Younger patients with AF (<50 years) more often had alcohol-related mental and behavioral disorders, gout, and dilated cardiomyopathy, while older patients (>60 years) more often had heart failure, ischemic stroke, renal failure, and respiratory infections. Of 51 phenotypes more common in AF, 28 were previously reported, and 23 were potentially novel. Mendelian randomization analyses suggested AF was causal for 19 associations.

CONCLUSIONS

This Phe-WAS demonstrates the broad spectrum of health care interactions in AF beyond traditional cardiovascular complications.