Introduction: Given the negative impact of atrial fi-brillation (AF) in patients (pts) with hypertrophic car-diomyopathy (HCM), finding new and better predic-tors of AF is clinically important, especially for patients considered at low or intermediate risk based on current recommendations (i.e. left atrial diameter, LAD <45 mm).

Objective: To assess the relationship between left atrial (LA) remodelling (size and function) and the presence of paroxysmal AF in HCM patients with and without increased LAD.

Methods: A comprehensive echocardiogram was per-formed in 110 consecutive patients (52 ± 17 years, 50 men) with HCM, in sinus rhythm. Indexed LA volu-me (LAVi), maximum left ventricular wall thickness (LVWT), LV ejection fraction, E/e’ ratio were determi-ned. Global longitudinal LV strain (GLS) and LA strain parameters (LAr, ASr) were assessed by speckle tracking echocardiography. Patients were divided into two groups according to the presence (30 pts) or ab-sence (80 pts) of documented paroxysmal AF (24/48 h ambulatory ECG recordings).

Results: Patients with AF were older than patients wi-thout AF (p<0.001). LAD, LAVi, E/e’ were significantly higher, while LAr, ASr were signifi cantly lower in pts with AF compared to pts without AF (p<0.05 for all). There were no significant differences between patients with and without AF regarding: gender, LVWT, GLS, the presence and severity of LV outflow tract obstruc-tion (p>0.05 for all). The correlates of AF in the whole HCM study population were: age (OR=1.05, p=0.001), ASr (OR=10.1, p<0.001), LAVi (OR=1.03, p=0.004), LAD (OR=1.2, p=0.001), E/e’(OR=1.05, p=0.02) and mitral regurgitation degree (OR=1.6, p=0.04). ASr had the best area under the curve (AUC: 0.74) with a cutoff of -0.88 s-1 for identifying HCM patients with AF (sen-sitivity: 80%, specificity: 65%). At multivariable logistic regression analysis, age, LAVi and ASr emerged as the only independent correlates of AF. 14 of the 71 patients with a LAD <45 mm had paroxysmal AF. In this selec-ted population, patients with AF were older (p=0.001), had higher values for E/e’ (p=0.04) and lower values for ASr (p=0.02) than patients without AF. Moreover, in this subgroup of patients, at multivariable logistic regression analysis, ASr correlated with AF indepen-dently of age, LA dimensions, LV hypertrophy or E/e’ (OR =10.008, 95% CI 1.297-77.219, p=0.02).

Conclusions: In patients with HCM, age, LAVi and ASr, were independently related to the presence of AF, while LAD was not. Moreover, in the subgroup of HCM pati-ents with LAD <45 mm, ASr (reflecting LA contractile function) was the only LA remodelling parameter cor-related with the presence of AF. The assessment of LA volume and function can provide further insights into the risk stratification of patients with HCM, especially in patients considered at lower risk for AF based on the assessment of classical risk parameters, such as LAD.