Background: Atrial fibrillation (AF) is a potential source of thromboembolic stroke in patients with cryptogenic stroke (CS), with incidence of AF at 3 years of ~30%. Lack of anticoagulation (AC) prior to diagnosis of AF may pose a risk for recurrent stroke in these patients and warrants novel strategies for the prediction of AF. Structural and functional characteristics of the left atrium (LA) play an important role in the genesis of AF, yet their role in predicting AF in patients with CS has not been reported.
Objective: To assess the role of structural [LA volume index (LAVI)] and functional [strain (ε), strain rate (SR), LA appendage (LAA) velocity] parameters in predicting AF in patients with CS.
Methods: Transthoracic (TTE) and transesophageal (TEE) echos of all patients with CS implanted with ILR were reviewed. LAVI was determined from TTE and LAA velocity from TEE images. Global ε and SR analyses were averaged from TTE images in 2, 3 and 4-chamber views using a commercially available software AutoSTRAIN© (Tomtec Inc.). Incidence of AF with ILRs was determined, LA structural (LAVI) and functional (ε and SR and LAA velocity) parameters were compared in patients with and without AF. Multivariable analysis was performed to assess the predictors of AF in CS.
Results: A total of 579 patients comprised the study. During a median follow up of 20 months (IQR: 25), 85 (14.6%) patients were diagnosed with new-onset AF. When compared to those with no AF, those with AF had significantly higher CHADS2VASc score (4.4±1.4 vs 5±1.3; p=0.001) and LAVI (30±11 vs 35±14; p=0.001). LAA velocity was significantly lower in AF patients compared to non-AF patients (64±22 vs 70±32; p=0.001). Similarly, patients with AF had significantly lower ε (23±7 vs 28±7; p=0.01) and SR (0.92±0.27 vs 1.07±24; p=0.01) compared to non-AF patients. CHADS2VASc score, LAVI, global ε and SR predicted new-onset AF.
Conclusion: This large, single-center experience is the first evidence that LA structural and functional parameters predict the incidence of AF in CS patients. A predictive model incorporating LA structural and functional parameters may serve to identify patients at high risk of AF who may benefit from early initiation of anticoagulation. Further studies are needed to validate these findings.