Background: Valve reintervention after transcatheter aortic valve replacement (TAVR) failure has not been studied in detail. Objectives: The authors sought to determine outcomes of TAVR surgical explantation (TAVR-explant) vs redo-TAVR because they are largely unknown. Methods: From May 2009 to February 2022, 396 patients in the international EXPLANTORREDO-TAVR registry underwent TAVR-explant (181, 46.4%) or redo-TAVR (215, 54.3%) for transcatheter heart valve (THV) failure during a separate admission from the initial TAVR. Outcomes were reported at 30 days and 1 year. Results: The incidence of reintervention after THV failure was 0.59% with increasing volume during the study period. Median time from index-TAVR to reintervention was shorter in TAVR-explant vs redo-TAVR (17.6 months [IQR: 5.0-40.7 months] vs 45.7 months [IQR: 10.6-75.6 months]; P < 0.001], respectively. TAVR-explant had more prosthesis–patient mismatch (17.1% vs 0.5%; P < 0.001) as the indication for reintervention, whereas redo-TAVR had more structural valve degeneration (63.7% vs 51.9%; P = 0.023), with a similar incidence of ≥moderate paravalvular leak between groups (28.7% vs 32.8% in redo-TAVR; P = 0.44). There was a similar proportion of balloon-expandable THV failures (39.8% TAVR-explant vs 40.5% redo-TAVR; P = 0.92). Median follow-up was 11.3 (IQR: 1.6-27.1 months) after reintervention. Compared with redo-TAVR, TAVR-explant had higher mortality at 30 days (13.6% vs 3.4%; P < 0.001) and 1 year (32.4% vs 15.4%; P = 0.001), with similar stroke rates between groups. On landmark analysis, mortality was similar between groups after 30 days (P = 0.91). Conclusions: In this first report of the EXPLANTORREDO-TAVR global registry, TAVR-explant had a shorter median time to reintervention, with less structural valve degeneration, more prosthesis–patient mismatch, and similar paravalvular leak rates compared with redo-TAVR. TAVR-explant had higher mortality at 30 days and 1 year, but similar rates on landmark analysis after 30 days.