ALTERNATIVE VERSUS PERCUTANEOUS TRANSFEMORAL ACCESS IN TAVI: PERIPROCEDURAL COMPLICATIONS AND FIVE-YEAR SURVIVAL IN A MULTICENTRE REGISTRY

Introduction Percutaneous transfemoral access is the reference route for transcatheter aortic valve implantation (TAVI), and the guidelines reserve the alternative routes for patients who are not suitable for it. The comparison between the two strategies is nevertheless burdened by confounding by indication, since patients who receive an alternative access are selected for an unfavourable vascular anatomy, and by the evolution of the technique over time. The present work assesses the periprocedural complications associated with the access route in an unselected regional cohort, the factors associated with the use of an alternative access and long-term survival. Methods We conducted a multicentre retrospective-prospective observational study including patients who underwent TAVI for severe aortic stenosis between 2010 and 2022 in six centres in the Emilia-Romagna region, Italy. Patients were classified according to percutaneous transfemoral or alternative access. The primary endpoint was a composite of periprocedural complications defined according to the Valve Academic Research Consortium-2 criteria, comprising acute kidney injury, major bleeding, major vascular complication, cerebrovascular event within 30 days and the need for transfusion. Secondary endpoints were the individual components, the length of stay and periprocedural, 30-day and one-year mortality. Associations were explored with multivariable logistic regression adjusted for clinical characteristics, valve type and procedural era, and survival was estimated with the Kaplan-Meier method. Results A total of 1308 patients were included, 1051 treated through percutaneous transfemoral access and 257 through an alternative access. The primary composite endpoint occurred in 29.6% of patients with alternative access against 21.7% of those treated transfemorally, and the association remained significant after adjustment (odds ratio 2.10; p = 0.003). The excess was driven mainly by acute kidney injury, which occurred in 23.7% against 8.9% (adjusted odds ratio 3.28; p < 0.001) and for which alternative access was the strongest independent predictor, despite comparable baseline renal function. The length of stay was also longer (median 8 against 7 days; p < 0.001) and periprocedural mortality higher (2.7% against 0.9%; p = 0.025), whereas the difference did not reach significance when in-hospital and 30-day deaths were considered together. Breaking the group down by individual route showed substantial heterogeneity, since the excess was concentrated in the thoracic routes, transapical and transaortic, whereas femoral access with surgical exposure did not differ from the percutaneous one. All-cause survival did not differ up to five years. The use of an alternative access was associated with peripheral artery disease, with previous coronary artery bypass grafting and, to a far greater extent, with the procedural era. Conclusions In this unselected multicentre cohort, alternative access was associated with an excess of periprocedural complications, driven mainly by acute kidney injury and persisting after adjustment, together with a longer stay and higher periprocedural mortality. The excess was not uniform within the group but concentrated in the approaches that require opening the chest. The observed cost was confined to the periprocedural window, with no detectable association with mortality up to five years.