The effect of transcatheter aortic valve implantation approaches on mortality.

OBJECTIVES We aimed to evaluate the effect of transcatheter aortic valve implantation (TAVI) approaches on mortality and identify effect modifiers and predictors for mortality. BACKGROUND Alternative access routes to transfemoral (TF) TAVI include the surgical intra-thoracic direct-aortic (DA) and transapical (TA) approach. TA TAVI has been associated with a higher mortality rate. We hypothesized that this is related to effect modifiers, in particular the left ventricular ejection fraction (LVEF). METHODS This multicentre study derived its data from prospective registries. To adjust for confounders, we used propensity-score based, stabilized inverse probability weighted Cox regression models. RESULTS In total, 5,910 patients underwent TAVI via TF (N = 4,072), DA (N = 524), and TA (N = 1,314) access. Compared to TF, 30-day mortality was increased among DA (HR 1.87, 95%CI 1.26-2.78, p = .002) and TA (HR 3.34, 95%CI 2.28-4.89, p < .001) cases. Compared to TF, 5-year mortality was increased among TA cases (HR 1.50, 95%CI 1.24-1.83, p < .001). None of the variables showed a significant interaction between the approaches and mortality. An impaired LVEF (≤35%) increased mortality in all approaches. CONCLUSIONS The surgical intra-thoracic TA and DA TAVI are both associated with a higher 30-day mortality than TF TAVI. TA TAVI is associated with a higher 5-year mortality than TF TAVI. The DA approach may therefore have some advantages over the TA approach when TF access is not feasible.