Since the first implantation by Cribier, transcatheter aortic valve implantation (TAVI) revolutionized the management of severe symptomatic aortic stenosis. Indeed, TAVI is indicated as an alternative to conventional surgery in patients at high surgical risk or contra-indicated to conventional surgery.

However, TAVI remains associated with specific complications related to the technics itself dominated by vascular complications and conductive disorders. Major vascular complications remain frequent after TAVI despite improvements in operators' experience, patient's selection and lower profile devices. Indeed, according to the Valve Academic Research Consortium 2 (VARC-2) criteria , major VC are still reported with an incidence of 1.5% to 15% of the procedures in registries and may be associated with unfavorable clinical outcomes. Currently, percutaneous approach (PC) in transfemoral TAVI is performed in routine, considered as a less invasive strategy than the traditional surgical cutdown (SC) performed in the first TAVI experience. Indeed, percutaneous approach may facilitate the local anesthesia and does not require the presence of the surgeon in the catheterization laboratory . However, surgical approach is still performed in many centers, allowing a better control of the puncture site with a low rate of vascular complications . Several non-randomized studies compared the two approaches with contradictory results]. No data are available comparing both approaches performed by the same team during the same period.

The aim of this study was to compare percutaneous and surgical approaches in terms of vascular complications and bleeding in patients undergoing transfemoral TAVI.

This prospective monocentric observational study aims to include all patients undergoing TAVI via transfemoral access from November 2017 at Montpellier University Hospital, France.

All procedures are performed in the same site by 7 medical teams including 4 different surgeons and 9 interventional cardiologists.

All patients have severe symptomatic aortic stenosis secondary to degenerative disease confirmed by transthoracic echocardiography (mean gradient >40 mmHg and/or valve area <1 cm2) and are deemed for TAVI after heart team decision.

Before the procedure, a multislice computerized tomography (MSCT) is systematically performed to evaluate the aortic annulus, the aorta and to determine the access site.

The choice of percutaneous or surgical is at operators' discretion, according to predictable access site difficulties (calcifications, tortuosity femoral bifurcation height) and global considerations as obesity.

Both procedures are performed during the same period, our center using both techniques in a common way.

TAVI procedure are performed under general anesthesia in the majority of cases.

Percutaneous approach The anterior wall of the common femoral artery are punctured under echographic guidance after selection of a site free of calcifications and above the femoral bifurcation and a 6-French sheath was inserted.

Two Perclose ProGlide® devices (Abbott Vascular, Santa Clara, CA) are used to perform the perpendicular preclosing. After the deployment of the prosthesis, the TAVI sheath is removed and the nodes from the two previously placed ProGlide® devices were tightened. Final femoral angiographic control is not systematic.

Surgical approach A surgical equivalent of the " preclose technique ", untitled the " zip technique " is performed routinely in our center. This technique avoids arterial crossclamping and purse string effect. Briefly, the femoral artery is exposed through a transversal groin incision, the anterior wall of the artery is cleaned from surrounding tissues up to a few centi- meters. The most suitable site is chosen free of calcifications allowing a puncture and insertion of a 6-French sheath. At the end of the procedure, both ends of the sutures are gently pulled to obtain coaptation of the arterial wall.

Study endpoints The primary endpoint of our study is a composite of vascular complications and bleeding defined by the modified VARC-2 classification.

Secondary end points included complications after TAVI, amount of contrast, amount of radiation, procedure duration, hospitalization length.

Data assessment Data on patients' baseline characteristics, procedural details and in-hospital outcomes is collected from a prospective TAVI database.

No additional testing or biological samples were specifically required for this study.

Inclusion criteria:

• Patients undergoing a TAVI at Montpellier University Hospital
• Since November 2017
• With implantation of a COREVALVE or EDWARDS
• By femoral approach

Exclusion criteria:

• Minor patient
• Patient unable to give consent for reasons of understanding or language barrier

November 2017 at Montpellier University Hospital, France, for severe symptomatic aortic stenosis

NC