Determinants of paravalvular leakage following transcatheter aortic valve replacement in patients with bicuspid and tricuspid aortic stenosis

Aims Paravalvular leakage (PVL) after transcatheter aortic valve replacement (TAVR) is a common complication in patients with bicuspid aortic valve (BAV). However, predictors and mechanisms of PVL are not well understood in this subset. The aim of this study was to analyse determinants and mechanisms of PVL in BAV and tricuspid aortic valve (TAV). Methods and results Of the 2394 consecutive patients undergoing transfemoral TAVR using new-generation valves at our centre, we identified 242 cases with BAV. To adjust for baseline differences, we performed 3 : 1 propensity score matching (TAVPS n = 726). We analysed the aortic root anatomy and calcification as well as the number, circumferential distribution, and predilection sites of PVL using pre-procedural multidetector computed tomography and post-TAVR echocardiography. In the matched cohort, the incidence of PVL ≥mild (BAV 51.9% vs. TAVPS 51.7%; P = 0.955) and PVL ≥moderate (BAV 5.0% vs. TAVPS 3.7%; P = 0.393), the circumferential distribution, and independent predictors were similar between BAV and TAVPS. Both the presence of peri-annular calcium chunks or LVOT calcification were highly associated with PVL in BAV and TAVPS patients, whereas in BAV patients neither the presence of a calcium bridge nor the volume of its calcification was related to PVL. Notably, the spatial localization of these lesions did not necessarily match the circumferential leak position. Conclusion The incidence, circumferential distribution, predilection sites, and predictors of PVL were similar in matched population of BAV and TAVPS patients undergoing transfemoral TAVR using new-generation devices. These novel findings suggest a common underlying mechanism of PVL in both entities.