Hemodynamics associated with atrial fibrillation directly alters thrombotic potential of endothelial cells

Highlights

• LAA atrial fibrillation (AFib) hemodynamics were applied to endothelial cells.

• AFib-exposed endothelial cells exhibited increased fibrin deposition.

• Markers of thrombosis are altered in AFib-exposed endothelial cells.

• Fibrin deposition thickness is reduced by apixaban in our experimental model.

Abstract

An experimental in vitro model of the hemodynamics that occur in atrial fibrillation (AFib) in the left atrial appendage (LAA) was developed to study changes in human endothelial cell thrombotic potential. We applied human-derived sinus rhythm and AFib hemodynamic shear stress patterns to primary human endothelial cells (ECs) in culture. We found that ECs exposed to AFib hemodynamics have increased thrombotic potential as measured by increased expression of pro-thrombotic gene markers and fibrin deposition on the endothelium. Treatment with the factor Xa inhibitor, apixaban, attenuated fibrin deposition thickness while increasing fibrin density at the endothelial cell surface. This study suggests that altered hemodynamics associated with AFib play a key role in driving the thrombotic potential of the LAA endothelium.

Introduction

Atrial fibrillation (AFib) is a sustained cardiac arrhythmia of the upper chambers of the heart that creates abnormal atrial blood flow patterns. A serious complication of AFib is blood stasis in the atrium that promotes clot, or thrombus formation. Thrombi can then dislodge and embolize in cerebral and other arteries thereby causing ischemic injury or stroke.

AFib is the most common serious cardiac arrhythmia affecting patients today, and disease prevalence continues to rise worldwide [1]. It is estimated that ~ 15–20% of all ischemic strokes are attributable to AFib, the majority caused by thrombi originating from the left atrial appendage (LAA; Fig. 1A) [2], [3]. AFib risk is increased with advancing age and is also frequently associated with co-morbidities such as hypertension, diabetes, and heart failure [4]. Stroke prevention in AFib is the primary objective for oral anticoagulant drugs, which are indicated for patients with non-valvular AFib. Other means to minimize stroke risk in AFib include cardioversion and ablation procedures to restore normal rhythm, devices and procedures that limit pooling of blood in the abnormal atrial appendage, and anti-arrhythmic medications [1].

The nature of thrombotic risk in AFib is clearly dependent on thrombin production, which may be triggered by: 1) the low-shear and blood-pooling in the dysfunctional LAA, 2) a generalized pro-thrombotic state of the blood characterized by hyperactive platelets, and/or 3) pro-coagulant factors expressed in the atrial endocardium partly attributable to endothelial dysfunction or damage. Moreover, pulsatility of blood flow and rheology are altered in the LAA during AFib and may influence vascular endothelial cell phenotype.

Preclinical model systems to study thromboembolic risk in AFib are conspicuously lacking. As a result, the proximal molecular triggers for clotting in AFib remain speculative. Pharmacologists seeking to characterize drug candidates measure blood coagulation in vitro triggered by thrombin, tissue factor, Russell's viper venom, and diatomaceous earth/celite, or crude in vivo models of venous stasis with thrombosis triggered by tissue factor or injury induced by ferric chloride in animals [5], [6], [7]. Anticoagulant drugs prescribed chronically to prevent stroke in AFib now include vitamin K antagonists like warfarin, direct thrombin inhibitors like dabigatran, and factor Xa inhibitors like apixaban and rivaroxaban, but all of these drugs carry significant risk of major bleeding [6]. Better understanding of endogenous triggers leading to stroke in AFib would enable discovery and development of potentially safer and more effective treatments.

We envisioned that the alteration in atrial hemodynamics caused by the AFib rhythm disturbance could promote a pro-thrombotic phenotype of the atrial endocardial endothelium, as it is well-appreciated that altered hemodynamics associated with disturbed flow promotes atherogenesis and an activated, thrombogenic endothelium [8]. To study this we modeled LAA blood flow patterns in normal sinus rhythm and in AFib using an in vitro human vascular surrogate system to investigate phenotypic contrasts and identify mediators of risk for atrial thrombosis and thromboembolism. This experimental approach enables us to mimic altered LAA hemodynamics during AFib, to reveal relevant mediators of endothelium-dependent thromboembolic risk, to elicit functional fibrin deposition with AFib simulation, and to explore the pharmacology of therapeutic agents in this in vitro human vascular surrogate system.