Objective: T he aim of the study is to investigate whether peak left atrial longitudinal strain in reservoir phase (PALS) is a predictor of cardiac events (CE) in patients with heart failure (HF) with preserved or bor-derline ejection fraction.

Methods: We prospectively enrolled 182 patients (94 women, mean age 65 ± 11 years) with non-acute HF (ejection fraction ≥45%). Conventional and speckle tracking 2D rest echocardiography were performed and clinical variables were recorded. We excluded pa-tients with acute coronary syndromes, severe valvular disease, cardiomyopathies, arrhythmia and class IV NYHA. Cardiac events included: cardiac death, acute coronary syndromes, worsening heart failure, atrial fi-brillation, stroke.

Results: During a median follow up of 20 months (in-terquartile range 18-26 months) there were 49 cardi-ac events recorded including 2 cardiac deaths, 9 acute coronary syndromes, 22 hospitalisations for worsening heart failure, 11 episodes of atrial fibrillation, 5 strokes. In univariate analysis lower PALS was associated with a higher risk of cardiac events [Hazard Ratio (HR ) 1,38 95% CI (confidence interval) 1,28 -1,44; p<0,001 per 1% decrease)]. On multivariate analysis this associati-on was independent after adjustment for age, hyper-tension, diabetes, left atrial indexed volume (LAVI), left ventricular global longitudinal strain (LVGLS), (HR 1,08 per 1% decrease 95% CI 1,03 -1,16; p=0,023). Worse outcomes were observed between lower PALS and a composite of worsening heart failure, atrial fi-brillation and stroke (HR 1,49 per 1% decrease 95% CI 1,39 -1,58; p<0,001). In the subgroup defined by de-pressed LVGLS (more positive than – 20%) and LAVI higher than 34ml/m² adding lower PALS, with a cut off point of 25%, significantly improved prediction of CE (C-statistic increased from 0,72 to 0,83, p<0,001).

Conclusions: PALS is an independent predictor of cardiac events in patients with heart failure (HF) with preserved or borderline ejection fraction. When added to reduced global longitudinal strain and high left atri-al indexed volume, atrial strain significantly improves prediction of cardiac events.