Outcomes of repeat transcatheter aortic valve replacement with balloon-expandable valves: a registry study

Background With increasing numbers of patients undergoing transcatheter aortic valve replacement (TAVR), data on management of failed TAVR, including repeat TAVR procedure, are needed. The aim of this study was to assess the safety and efficacy of redo-TAVR in a national registry.Methods This study included all consecutive patients in the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry from Nov 9, 2011, to Dec 30, 2022 who underwent TAVR with balloon-expandable valves in failed transcatheter heart valves (redo-TAVR) or native aortic valves (native-TAVR). Procedural, echocardiographic, and clinical outcomes were compared between redo-TAVR and native-TAVR cohorts using propensity score matching.Findings Among 350 591 patients (1320 redo-TAVR; 349 271 native-TAVR), 1320 propensity-matched pairs of patients undergoing redo-TAVR and native-TAVR were analysed (redo-TAVR cohort: mean age 78 years [SD 9]; 559 [42<middle dot>3%] of 1320 female, 761 [57<middle dot>7%] male; mean predicted surgical risk of 30-day mortality 8<middle dot>1%). The rates of procedural complications of redo-TAVR were low (coronary compression or obstruction: four [0<middle dot>3%] of 1320; intraprocedural death: eight [0<middle dot>6%] of 1320; conversion to open heart surgery: six [0<middle dot>5%] of 1319) and similar to native-TAVR. There was no significant difference between redo-TAVR and native-TAVR populations in death at 30 days (4<middle dot>7% vs 4<middle dot>0%, p=0<middle dot>36) or 1 year (17<middle dot>5% vs 19<middle dot>0%, p=0<middle dot>57), and stroke at 30 days (2<middle dot>0% vs 1<middle dot>9%, p=0<middle dot>84) or 1 year (3<middle dot>2% vs 3<middle dot>5%, p=0<middle dot>80). Redo-TAVR reduced aortic valve gradients at 1 year, although they were higher in the redo-TAVR group compared with the native-TAVR group (15 mm Hg vs 12 mm Hg; p<0<middle dot>0001). Moderate or severe aortic regurgitation rates were similar between redo-TAVR and native-TAVR groups at 1 year (1<middle dot>8% vs 3<middle dot>3%, p=0<middle dot>18). Death or stroke after redo-TAVR were not significantly affected by the timing of redo-TAVR (before or after 1 year of index TAVR), or by index transcatheter valve type (balloon-expandable or non-balloon-expandable).Interpretation Redo-TAVR with balloon-expandable valves effectively treated dysfunction of the index TAVR procedure with low procedural complication rates, and death and stroke rates similar to those in patients with a similar clinical profile and predicted risk undergoing TAVR for native aortic valve stenosis. Redo-TAVR with balloon-expandable valves might be a reasonable treatment for failed TAVR in selected patients.