Abstract
Valve-in-valve transcatheter aortic valve implantation (ViV-TAVI) has emerged as a feasible alternative to reoperative surgery in patients with degenerated surgical bio-prosthesis. However, data regarding the choice of valve type in ViV-TAVI remain inconclusive. This meta-analysis compares the procedural and clinical outcomes of self-expanding (SE) vs. balloon-expandable (BE) valves in ViV-TAVI. MEDLINE and Scopus were queried to identify studies reporting outcomes of ViV-TAVI by SE/BE valve type or comparing outcomes between SE or BE valves for ViV-TAVI. The primary outcome was incidence of all-cause mortality at 30 days. Data were presented as incidence of outcomes, analyzed via random effects model using inverse variance method with 95 % confidence intervals. Further incidence rates of primary and secondary outcomes were presented as subgroups of BE and SE, with comparison in incidence rates between the subgroups made using p-interaction of proportions. 27 studies with 13,182 patients (SE: 7346; BE: 5836) were included. There were no significant differences between the BE vs. SE valves in 30-day mortality (BE 4 % vs. SE 3 %, p = 0.44), 1-year mortality (BE 12 % vs. SE 10 %, p = 0.60), and moderate-to-severe AR at 1 year (BE 1 % vs. SE 3 %, p = 0.36). However, patients with SE valves had higher rates of new permanent pacemaker insertion (BE 4 % vs. SE 9 %, p = 0.0019). There were no significant differences in the incidence of 30-day safety outcomes, including stroke, AKI, coronary obstruction, major bleeding, and major vascular complications. Both BE and SE valve types showed comparable mortality and safety outcomes in ViV-TAVI, except pacemaker insertion, which was higher in SE compared with BE valves.