We assessed trends in the performance of transcatheter aortic valve implantation in the United Kingdom from the first case in 2007 to the end of 2012. We analyzed changes in case mix, complications, outcomes to 6 years, and predictors of mortality. Annual cohorts were examined. Mortality outcomes were analyzed in the 92% of patients from England and Wales for whom independent mortality tracking was available. A total of 3980 transcatheter aortic valve implantation procedures were performed. In successive years, there was an increase in frequency of impaired left ventricular function, but there was no change in Logistic EuroSCORE. Overall 30-day mortality was 6.3%; it was highest in the first cohort (2007-2008), after which there were no further significant changes. One-year survival was 81.7%, falling to 37.3% at 6 years. Discharge by day 5 rose from 16.7% in 2007 and 2008 to 28% in 2012. The only multivariate preprocedural predictor of 30-day mortality was Logistic EuroSCORE ≥40. During long-term follow-up, multivariate predictors of mortality were preprocedural atrial fibrillation, chronic obstructive pulmonary disease, creatinine >200 μmol/L, diabetes mellitus, and coronary artery disease. The strongest independent procedural predictor of long-term mortality was periprocedural stroke (hazard ratio=3.00; P<0.0001). Nonfemoral access and postprocedural aortic regurgitation were also significant predictors of adverse outcome. We analyzed transcatheter aortic valve implantation in an entire country, with follow-up over 6 years. Although clinical profiles of enrolled patients remained unchanged, longer-term outcomes improved, and patients were discharged earlier. Periprocedural stroke, nonfemoral access, and postprocedural aortic regurgitation are predictors of adverse outcome, along with intrinsic patient risk factors.