ABSTRACT NUMBER: ESOC2026A2472 EARLY P-WAVE PREDICTORS OF AF IN PATIENTS WITH ACUTE ISCHEMIC STROKE: EVIDENCE FROM INSERTABLE CARDIAC MONITORS

Abstract Background and aims Undetected asymptomatic atrial fibrillation (AF) is a major cause of cryptogenic ischemic stroke, and early identification is essential to guide secondary prevention. Prolonged monitoring with insertable cardiac monitors (ICM) improves AF detection, but early electrocardiographic markers are needed to stratify risk in the acute phase. Methods We conducted a single-center retrospective study of 43 patients admitted with acute ischemic stroke and discharged with cryptogenic stroke who underwent ICM implantation. Admission 12-lead ECGs were blindly reanalyzed for P-wave morphological abnormalities. Findings were tested in an independent replication cohort of 56 patients with comparable clinical characteristics. Results In the discovery cohort, ICM monitoring identified subclinical AF in 19 patients (44%). P-wave abnormalities were present in 19 patients, of whom 12 (63%) developed AF during follow-up, showing a significant association with AF detection (χ2=4.96, p=0.02). AF patients had longer hospitalization and greater delta-NIHSS improvement. In the replication cohort, AF was detected in 19 patients. P-wave abnormalities occurred in 22 patients, 11 of whom (50%) developed AF. Again, abnormal P-wave morphology was significantly associated with AF during prolonged monitoring (χ2=4.17, p=0.04). Conclusions Early P-wave morphological analysis on admission ECGs is associated with subsequent detection of asymptomatic AF in cryptogenic stroke across independent cohorts. This readily available, noninvasive marker may support early risk stratification, selection for prolonged cardiac monitoring, and timely initiation of anticoagulation, improving secondary stroke prevention and highlighting clinically relevant neurocardiac interactions. These findings reinforce the clinical value of standard ECG interpretation in acute stroke pathways and multidisciplinary neurocardiology decision making. Conflict of interest Nothing to disclose Figure 1 - belongs to Results