Predictors of atrial fibrillation detection by implantable loop recorder after cryptogenic stroke

Background Atrial fibrillation (AF) is a major yet often silent cause of cryptogenic stroke. Implantable loop recorders (ILRs) enable continuous rhythm monitoring and improve late AF detection. However, clinical and echocardiographic predictors of AF detection in this population remain unclear. Identifying high-risk patients could guide early anticoagulation and optimize resource allocation. Methods We retrospectively analyzed consecutive patients who underwent ILR implantation for cryptogenic stroke between 2024 and 2025. Baseline clinical characteristics, comorbidities, echocardiographic parameters, and ongoing cardiovascular therapies were collected at implantation. AF occurrence was defined as any device-detected and visually confirmed episode during follow-up. Univariate analyses compared patients with and without AF detection. Variables with p < 0.20 or strong clinical relevance were entered into a multivariable logistic regression to identify independent predictors. Results A total of ≈120 patients were included (mean age ≈78 ± 8 years). AF was detected in 18% of patients during a median follow-up of 6 months. At univariate analysis, older age, higher left atrial volume index (LAVI), and presence of diabetes and sleep apneawere associated with AF detection (p < 0.05). In multivariable logistic regression, age (adjusted OR 1.07 [95 % CI 1.01–1.15], p = 0.03) and LAVI (adjusted OR 1.05 [95 % CI 1.00–1.10], p = 0.04) were independent predictors of AF detection. The model showed good discrimination (AUC = 0.81). The AUC = 0.81 means that this combination of predictors (age + LAVI) correctly differentiates patients who later developed AF from those who did not about 81% of the time — a good level of discriminative accuracy. Conclusions Among patients with cryptogenic stroke monitored with an ILR, older age and increased left atrial volumeindependently predicted subsequent AF detection. These results suggest that echocardiographic assessment of atrial size, in combination with age, may help refine post-stroke risk stratification and identify patients most likely to benefit from prolonged ILR monitoring or early anticoagulation.