The Clinical Impact of Paravalvular Leaks With Transcutaneous Aortic Valve Implantation (TAVI) Versus Surgical Aortic Valve Replacement (SAVR): A Systematic Review and Meta-Analysis

Background

Aortic valve stenosis is a common cardiac condition that requires intervention for symptomatic and/or prognostic reasons. The two most common interventions are surgical aortic valve replacement (SAVR) and transcatheter aortic valve implantation (TAVI). The ratio of TAVI:SAVR has increased twofold over the past few years and is now being considered in intermediate-risk patients as well. One of the significant benefits of TAVI is that it is less invasive; however, one of the drawbacks is a high paravalvular leaks (PVLs) rate compared to SAVR. To assess the impact of PVLs on survival, progression of heart failure, and the need for re-intervention.

Method

We conducted a comprehensive systematic literature search from the conception of TAVI 2002 until December 2022 through Embase (Ovid), MEDLINE (Ovid), Science Direct, and CENTRAL (Wiley). We followed PRISMA guidelines and checklists. Review protocol registration ID in PROSPERO: CRD42023393742.

Results

We identified 28 studies that met our eligibility criteria, and only 24 studies were suitable for pooling in a meta-analysis (including their hazard ratio with a confidence interval of 95%) assessing our primary outcome (all-cause mortality). The remaining four studies were narratively synthesised.

RevMan V5.4 (Version 5.4. Cochrane Collaboration, 2020) was utilised to pool meta-analysis data to assess effect estimates of PVLs in both intervention arms, using a random effect model for calculation (hazard ratio 1.14 confidence interval 95% 1.08–1.21 [p<0.0001]), with a follow-up duration between 30 days to 5 years.

Conclusion

Patients with mild or higher degrees of PVLs in both intervention arms incurred unfavourable outcomes. The incidence of PVLs was significantly higher with TAVI; even a mild degree led to poor quality of life and increased all-cause mortality on long-term follow-up.