Chua, Su-KiatSu-KiatChuaHuang, Pang-ShuoPang-ShuoHuangChen, Jien-JiunJien-JiunChenChiu, Fu-ChunFu-ChunChiuJUEY-JEN HWANGWang, Chih-HsienChih-HsienWangYI-CHIH WANGCHIA-TI TSAI2026-03-202026-03-202026-01-30https://www.scopus.com/pages/publications/105029047303https://scholars.lib.ntu.edu.tw/handle/123456789/736499Atrial fibrillation (AF) significantly increases the risk of ischemic stroke. This study evaluates whether transmitral atrial flow velocity (MVA), measured non-invasively via transthoracic echocardiography (TTE), predicts stroke risk in AF patients.To assess the independent association between TTE-derived MVA and stroke incidence in paroxysmal AF patients and its value in refining risk stratification, especially in low-risk groups.This cohort study included 10,150 paroxysmal AF patients from 2010 to 2021. The primary outcome was hospitalization for ischemic stroke. Multivariable Cox regression analyses adjusted for CHADS-VASc scores evaluated the relationship between MVA and stroke risk.Over a mean follow-up of 4.26 ± 3.52 years, 2,419 (23.8%) patients developed ischemic strokes (5.59% per 100 person-years). In multivariable analysis, adjusting for CHADS-VASc score, MVA was independently associated with stroke incidence. Every 10 cm/s reduction in MVA velocity conferred 4% higher stroke risk (adjusted hazard ratio [HR] 0.96 [0.94-0.97],  < 0.001). AF patients with MVA < 50 cm/s had a 39% increase in stroke risk compared to those with MVA ≥ 50 cm/s (adjusted HR 1.39 [1.22-1.58],  < 0.001). In patients with a CHADS-VASc score of 0 or 1, the stroke incidence increased from 1.33 to 2.28% when they had MVA < 50 cm/s, which was similar to that of patients with a CHADS-VASc score of 2 points (2.51%).TTE-derived MVA independently predicts stroke risk in paroxysmal AF patients. Incorporating MVA enhances risk stratification and guides targeted stroke prevention, particularly in low-risk populations.enjournal article10.1055/a-2760-813441544945