Abstract 16439: Is Pulmonary Venous Flow Pulsatility a Critical Determinant of Left Atrial Appendage Blood Stasis Risk?

Abstract

Introduction: Atrial fibrillation (AF) is the most common arrhythmia that leads to thrombus formation, most commonly in the left atrial appendage (LAA), and a resultant five-fold increase in the risk of ischemic stroke. The current standard of stratifying stroke risk, based on the CHA 2 DS 2 -VASc score, does not consider subject-specific LAA hemodynamics and blood stasis risk. We have developed a methodology for computing mean residence time of blood-borne particles in the LAA ( t m ), an index of LAA blood stasis, using 3D left atrial (LA) and LAA geometries and pulmonary venous (PV) inlet blood flow data. While subject-specific 3D geometry can be obtained readily, it is not easy to measure all LA inlet blood flow waveforms in vivo . Therefore, a sensitivity study of LAA t m to the inlet blood flow waveform characteristics (shape and magnitude) is needed. Hypothesis: PV flow pulsatility does not affect LAA t m . Methods: Cardiac computed tomography images of 10 AF subjects were obtained and LA and LAA geometries were reconstructed in 3D and subsequently meshed for computational fluid dynamics (CFD) analysis. We employed CFD analysis, modeling blood cells as neutrally buoyant tracers transported in the blood and used a systems-based approach, to quantify LAA t m . Each subject was simulated using four PV blood flow characteristics: (1) Normal-S , steady PV flow with cardiac output (CO) of 5.5 L/min; (2) AF-S , steady PV flow with reduced CO of 4.4 L/min; (3) Normal-P , normal pulsatile PV flow with normal CO = 5.5 L/min; and (4) AF-P , a typical AF pulsatile waveform where systolic phase and the reversal phase are diminished with reduced CO of 4.4 L/min. Results: Comparing the two steady flow groups, LAA t m significantly increased as CO was decreased ( Normal-S: 2.6±0.3 s, AF-S: 3.1±0.3 s, P =0.02). However, the presence of pulsatility did not affect LAA t m at both levels of CO: Normal-P: 2.4±0.2 s; Normal-S: 2.6±0.3 s, P =0.2 and AF-P: 2.9±0.3 s; AF-S: 3.1±0.3 s, P =0.3. Conclusions: LAA blood stasis risk, as quantified by LAA t m , is significantly affected by the steady PV flow magnitude (i.e., cardiac output) and not by PV flow pulsatility. Therefore, the subject-specific LAA blood stasis risk can be reliably estimated using subject-specific LA and LAA 3D geometries and subject-specific CO.