New onset left bundle branch block after transcatheter aortic valve implantation and the effect on long-term survival – a UK wide experience

Introduction New onset left bundle branch block (LBBB) is the most common conduction disturbance associated with transcatheter aortic valve implantation (TAVI). It has been shown to adversely affect cardiac function and increase re-hospitalisation, although its impact on mortality remains contentious. Methods We conducted an observational cohort analysis of all TAVI procedures performed by 13 heart teams in the United Kingdom from inception of their structural programmes until 31st July 2013. The primary outcome was 1-year all-cause mortality. Secondary outcomes included left ventricular ejection fraction (LVEF) at 30 days and need for a post-TAVI permanent pacemaker (PPM). Results 1785 patients were eligible for inclusion to the study. The primary analysis cohort was composed of 1409 patients with complete electrocardiographic (ECG) data pre- and post-TAVI. Pre-existing LBBB was present in 200 (14.2%) patients. New LBBB occurred in 323 (22.9%) patients post TAVI, which resolved in 99 (7%) patients prior to discharge. A balloon-expandable device was implanted in 968 (69%) patients, whilst 421 (30%) patients received a self-expandable valve. New LBBB was observed in 120 (12.4%) and 192 (45.6%) patients receiving a balloon- or self-expandable prosthesis respectively. Overall 1-year all-cause mortality post TAVI was 18.7%. New onset LBBB was not associated with an increase in 1-year all-cause mortality (p=0.416). Factors that were associated with mortality included an increasing logistic EuroScore (p=0.05), history of previous balloon aortic valvuloplasty (p=0.001), renal impairment (p=0.003), previous myocardial infarction with pre-existing LBBB (p=0.028) and atrial fibrillation (p=0.039). Lower baseline peak and mean AV gradients were also associated with greater mortality at 1 year (p=0.001), likely reflecting underlying left ventricular dysfunction. In the majority of patients, LVEF remained unchanged following TAVI. Interestingly, the presence or absence of new onset LBBB did not affect LVEF improvement at 30 days. 10% of patients required a PPM post TAVI. Predictors of PPM included new LBBB (OR 2.6, p<0.001), pre-TAVI left ventricular systolic impairment (OR 1.2, p=0.037), a self-expandable device (p<0.001), and pre-existing RBBB (OR 4.0, p<0.001). Conclusions These findings suggest that new onset LBBB post TAVI does not increase mortality at 1 year or adversely affect LVEF at 30 days. Funding Acknowledgement Type of funding source: None