The DAFI75 scale predicted atrial thrombosis in patients with persistent nonvalvular atrial fibrillation with an AUC of 0.847 and a negative predictive value of 98.9% in the validation cohort.
Cohort (n=551)
Randomly split into training and validation cohorts
Can the DAFI75 scale predict left atrial appendage thrombosis in patients with persistent nonvalvular atrial fibrillation?
The DAFI75 scale, incorporating clinical and echocardiographic parameters, can reliably predict the absence of left atrial appendage thrombosis in patients with persistent nonvalvular AF, potentially reducing the need for routine TEE before cardioversion.
Effect estimate: AUC 0.847 (95% CI 0.761-0.934)
Introduction. The clinical significance of atrial fibrillation (AF) is associated with the development of thromboembolic complications that occur when thrombus from the left atrial appendage enter the systemic circulation. Transesophageal echocardiography can detect the left atrial appendage thrombus, but due to lack of availability, high cost and complexity of performing such a routine examination is unlikely. Therefore, the search for predictors of the left atrial appendage thrombosis is relevant, the presence of which may become the basis for a more in-depth instrumental examination of patients with AF. Aim. To identify predictors of atrial thrombosis in patients with persistent nonvalvular atrial fibrillation. Materials and methods . The 551 patients with persistent nonvalvular atrial fibrillation underwent transesophageal echocardiography before cardioversion, thrombus in the left atrial appendage was detected in 74 (13.4%) patients. All patients were selected into training (400 people) and validation (151 people) cohorts randomly. Multivariate logistic regression analysis was performed to identify predictors of atrial thrombosis in the derivation cohort. Results. Five factors influenced the atrial thrombosis independently. They are the ratio of the transmitral velocity to the mitral annulus early Diastolic velocity (E/e’) ≥ 12 (D), the absence or inadequate Anticoagulant therapy (A), atrial Fibrillation, not flutter (F), left atrial volume Index (I), and age ≥ 75 years (DAFI75 scale). The number of predictors corresponds the risk of detecting atrial thrombosis: the area under the characteristic curve was 0.818 (0.768–0.868) in the derivation cohort and 0.847 (0.761–0.934) in the validation cohort. The sensitivity of the DAFI75 criterion ≥ 3 in the derivation and validation cohorts is 91.7 and 92.9%, the specificity is 58.8 and 65.7%, the predictive value of a positive result is 28.2 and 21.7%, and the predictive value of a negative result is 97.6 and 98.9%. Conclusion. The presence of three or more predictors score DAFI75 allows predicting the absence of atrial thrombosis more than in 97% of case.
Мазур et al. (Sat,) conducted a cohort in Persistent nonvalvular atrial fibrillation (n=551). DAFI75 scale was evaluated on Left atrial appendage thrombosis (AUC 0.847, 95% CI 0.761-0.934). The DAFI75 scale predicted atrial thrombosis in patients with persistent nonvalvular atrial fibrillation with an AUC of 0.847 and a negative predictive value of 98.9% in the validation cohort.