Comparative prognostic impact of paravalvular leak versus permanent pacemaker implantation after transcatheter aortic valve implantation without reduced ejection fraction patients

Background

Paravalvular leak (PVL) and permanent pacemaker implantation (PPI) remain clinically important complications after transcatheter aortic valve implantation (TAVI). Procedural strategies aimed at reducing PVL, including valve oversizing and post-dilatation, may increase the risk of conduction disturbances and PPI. Conversely, strategies intended to minimize conduction disturbances and avoid PPI, such as selecting a smaller valve size and avoiding post-dilatation, may increase the risk of residual PVL. However, the relative prognostic impact of PVL versus PPI after TAVI remains unclear.

Purpose

The purpose of the present study was to compare the prognostic impact of mild-or-greater PVL and new-onset PPI after TAVI in a multicenter registry.

Methods

We analyzed data from a prospective, multicenter registry in Japan that enrolled 3,129 consecutive patients undergoing TAVI at seven centers between April 2010 and June 2022. After excluding patients with prior PPI, missing post-procedural PVL data, preprocedural left ventricular ejection fraction (LVEF) <40%, a history of surgical or transcatheter aortic valve replacement, or unsuccessful transcatheter valve implantation, 2,673 patients were included in the present analysis. PVL(+) was defined as mild or greater PVL on post-procedural transthoracic echocardiography, and PPI(+) as new-onset PPI during the index hospitalization. Patients were categorized into four groups: PVL(–)/PPI(–) (n=1,864), PVL(+)/PPI(–) (n=614), PVL(–)/PPI(+) (n=135), and PVL(+)/PPI(+) (n=60). The primary endpoint was three-year all-cause mortality and the secondary endpoint was three-year heart failure hospitalization (HFH).

Results

For all-cause mortality, Kaplan–Meier analysis across the four groups showed the highest mortality in the PVL(+)/PPI(+) group, and PVL(+)/PPI(–) patients had significantly higher three-year all-cause mortality than PVL(–)/PPI(+) patients (21.9% vs. 12.9%, P=0.02) (Figure 1A). However, in multivariable analysis, neither PVL(+) nor PPI(+) was an independent prognostic factor for all-cause mortality. In a subgroup analysis restricted to PVL(+)/PPI(–) and PVL(–)/PPI(+) patients, PVL(+) showed a consistent trend toward worse all-cause mortality than PPI(+) across most subgroups (Figure 2).

For HFH, Kaplan–Meier analysis demonstrated that HFH rates were comparable between PVL(+)/PPI(–) and PVL(–)/PPI(+) patients (10.1% vs. 10.6%, P=0.95) (Figure 1B). However, in multivariable analysis, PVL(+) was an independent predictor of HFH (HR 1.38, P=0.03), whereas PPI(+) was not (HR 1.52, P=0.07).

Conclusion

In patients undergoing TAVI with LVEF ≥40%, mild-or-greater PVL was associated with a stronger adverse prognostic factor than new-onset PPI. When a trade-off between PVL reduction and conduction safety is unavoidable, prioritizing PVL reduction may be reasonable, while individualized decision-making remains essential.Figure 1For image description, please refer to the figure legend and surrounding text.Figure 2For image description, please refer to the figure legend and surrounding text.