Related Experiment Video
Updated: Nov 30, 2025
Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
Predicting Atrial Fibrillation with High Risk of Embolization with Atrial Strain and NT-proBNP
Jorge Pagola1, Jesus Juega2, Jaume Francisco-Pascual3
1Stroke Unit, Medicine Department, Vall d'Hebrón Hospital and Autonomous University of Barcelona, Passeig Vall d'Hebrón, 119-129, 08035, Barcelona, Spain. jpagola@vhebron.net.
Insights
Predicting high-risk paroxysmal atrial fibrillation (HpAF) in cryptogenic stroke patients is crucial. Peak atrial longitudinal strain (PALS) and NT-proBNP levels effectively identify individuals at higher risk for HpAF episodes.
Area of Science:
- Cardiology
- Neurology
- Biomarkers
Background:
- Cryptogenic stroke often results from undetected paroxysmal atrial fibrillation (pAF).
- Identifying patients with high-risk pAF (HpAF) is essential for preventing recurrent embolic events.
Purpose of the Study:
- To identify markers of atrial dysfunction predicting HpAF in cryptogenic stroke patients.
- To evaluate the predictive value of echocardiographic parameters and NT-proBNP for HpAF.
Main Methods:
- Patients from the Crypto-AF study were classified based on pAF duration detected by Holter monitoring.
- Echocardiography (LAVI, PALS, PACS) and NT-proBNP levels were assessed.
- Follow-up for 2 years monitored pAF recurrence and stroke recurrence.
Main Results:
- Age and the combination of PALS and NT-proBNP independently predicted HpAF.
- PALS and NT-proBNP demonstrated higher predictive validity (AUC 0.82) than age.
- Patients with PALS < 25% and NT-proBNP > 283 pg/ml had significantly higher pAF detection rates during follow-up.
Conclusions:
- Multimodal assessment of atrial dysfunction using PALS and NT-proBNP improves prediction of HpAF in cryptogenic stroke.
- These markers can aid in identifying patients requiring intensified monitoring or treatment for stroke prevention.
Abstract:
The aim of the study was to determine markers of atrial dysfunction in patients with cryptogenic stroke to predict episodes of paroxysmal atrial fibrillation with high risk of embolization (HpAF). We classified patients included in the Crypto-AF study, Cryptogenic Stroke registry, to detect paroxysmal atrial fibrillation (pAF) with wearable Holter, according to the longest episode of pAF in three groups: without pAF detection, episodes of pAF shorter than 5 h, and episodes of pAF longer than 5 h (HpAF). Atrial dysfunction surrogates were evaluated: EKG pattern, Holter record and echocardiography parameters (left atria volume (LAVI), and peak atrial longitudinal and contraction strain (PALS and PACS). The level of N-terminal pro b-type natriuretic peptide (NT-proBNP) was determined. All patients were followed for 2 years to detect pAF and stroke recurrence. From 308 patients, 253 patients with high quality Holter analysis were selected. The distribution was No pAF 78.6% (n = 199), pAF < 5 h 7.9% (n = 20), and HpAF > 5 h 13.4% (n = 34). Age of the patients and combination of PALS and NT-proBNP independently predicted HpAF OR 1.07 (1.00; 1.15) and OR 3.05 (1.08; 8.60) respectively. The validity of PALS and NT-proBNP to detect patients at risk of HpAF was higher than the validity of age (AUC 0.82, sensitivity 78.95%, specificity 63%). Patients with PALS < 25% and NT-proBNP > 283 pg/ml had more detection of pAF during follow-up 35% vs. 5.1% OR 2.33 (1.05-5.13) (p < 0.001). Multimodal assessment of atrial dysfunction with PALS and NT-proBNP improved the prediction of pAF episodes with high embolic risk in patients with cryptogenic stroke.
