Cardiac Computed Tomography Angiography to define variables associated with minor peri-device leak after watchman implantation for left atrial appendage percutaneous occlusion

Abstract

Percutaneous occlusion of left atrial appendage (LAA) with a Watchman device (WM) is indicated in patients with non-valvular atrial fibrillation, a CHA2DS2VASc score ≥ 4 and contraindication on long-term anticoagulation. Even though minor (m) peri-device leak (PL) has not been associated with recurrence of embolic event, obtaining a complete occlusion of LAA remains a primary objective when implanting a WM. Complex anatomy of LAA has led Debiase to define 4 main categories based on the morphology of the dominant lobe (DL) and localisation of secondary lobes (SL), Windsock, Cactus and Cauliflower and Chicken Wing type (CW). Beside, LAA ostium may be round or more ovale. Success of implantation is obtained when no major PL or protrusion is noted by trans oesophageal echocardiography (TEE) with good anchoring to prevent migration. We aimed to evaluate the impact of anatomy or LAA defined by its type and shape of the ostium (os) on the incidence of mPL. Cardiac Computed Tomography Angiography (CCTA) was done before implantation to assess LAA morphology and measure maximal (Dmax) and minimal (Dmin) diameter at the os. Os was round if Dmax/Dmin was ≤ 1.13. Perileak was defined by TEE as residual flow < 5 mm. From 70 out of 74 patients (95%) with successful implantation, 45 (64%) had a non-CW morphology (6 cactus) and os was not-round in 45 (64%). A 24 with mPL was detected in 24 patients (34%) with difference between CW and non-CW type [20 (44%) vs. 4 (26%), P = ns] or round or not-round os [15(33%) vs. 9 (36%), P = 1). Even if a tendency to more mPL is noted in patients with CW type, other parameters should be studied as proximal location of SL or diameter of DL distal to the ostium. A classification of LAA anatomy dedicated to the procedure should be established to optimize implantation.