P1797 Left atrial mechanics in patients with stoke: Implications for emprical anticoagulation

Abstract

Abstract Purpose Detection of atrial fibrillation (AF) after stroke impacts patient management. However, detection of AF is difficult due to its paroxsysmal nature. We sought to test the hypothesis that echocardiographic quantification left atrial (LA) mechanics in patients with stoke can be an imaging biomarker to predict AF. Methods We enrolled prospectively and consecutively 131 patients in normal sinus rhythm with a new diagnosis of ischemic stroke. All patients underwent 48-hour Holter monitorization, transthoracic echocardiography with saline injection in addition to routine work-up of stroke evaluation. Two-dimensional and Doppler studies, together with most up-to-date LA quantification tools were performed including 3-dimensional LA phasic volumes (maxiumum and minimum LA Volume index - LAVImax, LAVImin) and LA strain quantification (average strain from apical 4- and 2-chamber during reservoir (Ss) and contraction (Sa) phases). Patients were followed-up and underwent second 48-hour Holter monitorization 6-12 months later if no etiologic cause was found. Any AF episode lasting >30 seconds was considered significant parosysmal AF. Results In 49 patients, causes other than AF were identified (Noncryptogenic). In 43 patients no discernable cause was found (NoAF Cryptogenic). Paroxysmal AF episodes lasting longer than 30 sec were documented in 27 patients at first 48-hour Holter and in 12 patients during follow-up (9.2 ± 3.1 months) either by second 48-hour Holter monitorization or clinically (AF cryptogenic). LAVImax and LAVImin were significantly increased, Ss and Sa were significantly reduced in patients with AF as compared to No AF and Noncryptogenic groups (Table). LA volumes and strain measures predicted AF development independently of CHA2DS2-VASc score. Better discrimination between No AF and AF groups was obtained after second Holter monitorization that enabled detection of more cases with paroxysmal AF (Figure). Conclusions Our findings underscore the value of echocardiographic assessment of LA function as a marker of AF development and for selection of patients who could benefit from empiric anticoagulation. Noncryptogenic AF cryptogenic No AF cryptogenic 3D LAVI max (ml/m²) 30.1 ± 9,4 38.1 ± 12.7 28.9 ± 9.2 3D LAVI min (ml/m²) 13.9 ± 7.5 20.7 ± 10.7 13.4 ± 5.3 3D LA EF (%) 55.5 ± 9.4 50.0 ± 10.5 53.8 ± 10.0 Ss (%) 17.4 ± 8.5 12.7 ± 5.3 17.5 ± 8.5 Sa (%) 12.7 ± 6.3 9.0 ± 5.0 12.8 ± 6.3 CHA2DS2-VASc 4.0 ± 1.6 5.0 ± 1.6 3.9 ± 1.5 Abstract P1797 Figure.