Echocardiography-derived left atrial volumetric–mechanical coupling index predicts early new-onset atrial fibrillation after embolic stroke/transient ischemic attack of undetermined source

Aim In patients with embolic stroke or transient ischemic attack of undetermined source (ESUS/TIA), occult atrial fibrillation (AF) is a frequent underlying mechanism with critical therapeutic implications, yet selecting patients who warrant intensified rhythm monitoring remains challenging. The Left Atrioventricular Coupling Index (LACI = LA end-diastolic volume/LV end-diastolic volume) captures atrioventricular uncoupling, whereas the Left Atrial Volumetric–Mechanical Coupling Index (LACI-VM = LAVi/a′) integrates atrial size with mechanical performance; both may capture complementary domains of atrial remodelling relevant to subclinical AF. We aimed to investigate the association of LACI-VM and LACI with incident AF in ESUS/TIA. Methods and results In this retrospective single-center cohort, 285 consecutive patients with ESUS/TIA of undetermined source underwent transthoracic echocardiography at index hospitalization, including measurement of LACI-VM and LACI. The primary endpoint was new-onset AF during follow-up. Median age was 71 years and 43% were women. Over a median follow-up of 10.9 months, AF occurred in 24 patients (8.4%), with 62% of events occurring within the first year. LACI-VM was higher in patients who developed AF and remained independently associated with incident AF after multivariable adjustment (HR 6.64, CI 1.34–59.88; p = 0.041). LACI showed a non-significant association overall (HR 2.28, CI 0.74–7.01; p = 0.09), with predictive value confined to events within 12 months. A combined assessment of LACI-VM and LV filling pressures further refined risk stratification: patients with preserved LACI-VM (< 2.5) and normal filling pressures (E/e′ ≤ 10) had a very low AF incidence (~ 2%), identifying a truly low-risk subgroup; conversely, when both LACI-VM (≥ 2.5) and filling pressures (E/e′ > 10) were impaired, AF risk was ~ tenfold higher (~ 20%). Conclusion In ESUS/TIA, echocardiography-derived LACI-VM independently predicts incident AF and identifies patients at genuinely low risk. As a simple echocardiographic metric, its integration with LV filling pressures may support a more personalized use of rhythm monitoring.