Markers of atrial myopathy in the UK Biobank: prevalence, consequences, and risk stratification

Atrial myopathy is increasingly recognized as a distinct clinical entity, and an important underlying component in atrial fibrillation (AF). While histological examination of atrial biopsies is the gold-standard for diagnosis of atrial myopathy, it is often impractical in a clinical setting.

This study aimed to examine the prevalence and inter-relationships of indirect atrial myopathy markers, identify associations with incident AF, and explore clinical or genetic risk scores for risk-stratification in individuals with markers of atrial myopathy.

The study cohort consisted of European ancestry individuals from the UK Biobank with cardiac magnetic resonance imaging and electrocardiographic information, and no prior history of AF, heart failure, or cardiac conduction disorders. Three markers of atrial myopathy were examined: Left atrial (LA) dilation (indexed LA volume >60 ml/m2), mechanical dysfunction (LA emptying fraction <45%), and electrical dysfunction (P-wave duration >120 ms). Hazard ratios (HR) for AF were examined in a multivariate Cox regression adjusted for sex, age, body-mass index, hypertension, coronary artery disease and diabetes. The cohort was stratified by LA myopathy markers, genetic risk of AF (using a previously validated polygenic risk score [PRS]), and clinical risk of AF (using the HARMS2-AF score). Genetic risk was stratified by PRS tertiles, and clinical risk was stratified by HARMS2-AF score ≤2 (low risk), HARMS2-AF score between 3-6 (intermediate risk), and HARMS2-AF score ≥7 (high risk). Five-year incidences of AF were calculated using the Aalen-Johansen estimator, with all-cause mortality as a competing risk.

Among 26,026 individuals, 2,447 (9.7%) had at least one marker of atrial myopathy, while 244 (0.9%) had two or more markers. The largest overlap was seen between LA dilation and mechanical dysfunction. During a median follow-up of 5 years, 583 (2.2%) were diagnosed with incident AF. Having one LA myopathy marker conferred a HR of 2.53 (95% confidence interval [CI]: 2.06-3.11; P<0.001) for AF. Higher rates were observed in individuals with two or more markers (HR for AF: 6.34; 95% CI: 4.56-8.81; P<0.001). Individuals with one or more LA myopathy marker, and high clinical and genetic risk, had a 11.7% (95% CI: 7.5-16.0%) 5-year risk of incident AF, compared with a 0.7% (95% CI: 0.4-0.9%) risk in those with no LA myopathy markers, and low clinical and genetic risk.

In a cohort without prior history of AF, almost one in ten individuals had at least one marker of atrial myopathy. Although overlap between different markers was modest, a dose-response-like relationship was observed between amount of atrial myopathy markers and rates of AF. Integration of clinical and genetic risk scores showed a several-fold risk gradient, highlighting that PRS and clinical risk scores may be useful aids in risk stratification of individuals with markers of LA myopathy.

Co-occurence of atrial myopathy markers