Long-term implications of left atrial appendage thrombus identified incidentally by pre-procedural cardiac computed tomography angiography in patients undergoing transcatheter aortic valve replacement

Aims: The prevalence and prognostic implications of left atrial appendage (LAA) thrombus (LAAT) in patients considered for transcatheter aortic valve replacement (TAVR) are incompletely defined. We, therefore, studied pre-procedural cardiac computed tomography angiography (CCTA) scans of TAVR candidates to determine the prevalence of LAAT and its association with late outcomes. Methods and results: Baseline clinical variables and CCTA findings from a prospective TAVR registry were analysed for the prevalence of pre-procedural LAAT and its impact on in-hospital outcomes and late mortality. LAAT was differentiated from LAA filling defects (LAAFD) reflecting stasis without clot. Patients (n = 561) with complete in-hospital and late mortality data were included in the study (median follow-up 31.6 months). LAAT and LAAFD were evidenced on pre-procedural CCTA in 24 (4.3%) and 26 (4.6%) patients, respectively. One hundred fourteen (20.3%) patients died during the study period. Though in-hospital adverse event rates (including stroke) did not differ among groups, mortality at long-term follow-up was higher among LAAT patients compared with those with or without LAAFD (58.3% vs. 11.5% vs. 19.0%, respectively; P < 0.003). By multivariable analysis, LAAT (but not LAAFD) was independently associated with all-cause mortality [hazard ratio (HR) = 3.33 (1.83-6.00), P < 0.001]. In patients with LAAT, oral anticoagulation at discharge was associated with lower mortality risk, independently of atrial fibrillation status. Conclusions: LAAT visualized by pre-procedural CCTA is an independent predictor of late mortality following TAVR, but not peri-procedural stroke. When reporting TAVR-CCTA, particular note should be made of LAA features and presence of LAAT which may have prognostic and management implications.