P1288 Male gender and left atrial volume index predict non valvular atrial fibrillation recurrence. The NeAfi echo registry

Abstract

Abstract Background In non-valvular atrial fibrillation (AF) patients, risk stratification scores such as CHA2DS2VASc and HASBLED allow the prediction of thromboembolic and bleeding risks, respectively. However, no risk score for the prediction of AF recurrence has been yet validated. Purpose To evaluate the mayor anthropometric and echocardiographic determinants of AF recurrence in non-valvular paroxysmal AF patients at 1 year follow-up. Methods Among 395 consecutive adult patients with non valvular AF enrolled in the Neapolitan Atrial Fibrillation (NeAfi) Echo registry, 177 (F/M = 87/90, age = 66.5± 11.9 years) had paroxismal AF and underwent 1-year follow-up. Fifteen patients had AF recurrence. Anthropometric parameters and blood pressure (BP) were recorded and CHA2DS2VASc and HASBLED scores were calculated. At baseline, patients underwent a comprehensive echo-Doppler exam, including quantification of left atrial (LA) size measurements, peak atrial longitudinal strain (PALS) and left ventricular (LV) global longitudinal strain (GLS). Binary logistic regression analysis was used to establish a mathematical model of the relationship between the variables and AF recurrence. The covariates for regression analysis were chosen as potential confounding factors based on their significance in independent T test analyses for continuous variables of chi-square for dichotomous variables, or on their biological plausibility. Results AF recurrence was higher in male than in female patients (14.4 vs 2.3%, p = 0.008). Patients with AF recurrence had similar body mass index, systolic and diastolic BP and heart rate compared to those without. The two groups were similar for LV end-systolic and end-diastolic volumes, ejection fraction, LV mass index, diastolic indexes, pulmonary artery systolic pressure and GLS. LA diameter (p = 0.235) and PALS (p = 0.375) were also similar between the two groups, whereas LA volume index (LAVi) was greater in patients experiencing AF recurrence (45.5 ± 15.7 vs. 36.7 ± 10.4 ml/m², p = 0.003). Binomial multiple regression analysis model explained 25% (Nagelkerke R²) of the variance in AF recurrence and correctly classified 95.0% of cases. Males were 8.9 times more likely to exhibit AF recurrence than females (p = 0.04). Greater LAVi was associated with an increased likelihood of exhibiting AF recurrence (OR = 1.07, p = 0.03), whereas CHA2DS2VASc >1 in men and >2 in women, HASBLED >3 and greater LA diameter or lower PALS did not add significant information to the model. Conclusions Male gender and, with a lower extent, LAVi appear to be major determinants of AF recurrence in non-valvular paroxysmal AF. The quantification of PALS does not seem to add valuable information in the prediction of recurrent AF.