Innovative multi-modality imaging to assess paravalvular leak

摘要

Imaging is a key point in the guidance of transcatheter treatment of paravalvular leak (PVL). Mostly explored by echocardiography, this imaging modality is limited by the acoustic shadowing from the valve prosthesis [1]. Cardiac computed tomography may be useful to assess anatomical characteristics (size, location, path) but is exposed to beam hardening artifacts and to cardiac movements [2]. Conventional cardiac magnetic resonance (CMR) seems to be relevant for functional evaluation of PVL after transcatheter aortic valve replacement, but is highly sensitive to metallic artefacts [3]. The question of a complete evaluation of paraprosthetic leakage therefore remains open and no conventional modality seems satisfactory. Because this question is also frequent in clinical practice, we have been led to consider an innovative imaging modality to complete the multimodal analysis; 4D flow imaging. We considered exploring new imaging techniques after percutaneous closure failure for a paravalvular mitral valve leak. The hypothesis put forward was a misguided selection of the prosthesis and therefore an unsatisfactory assessment of the leak. The initial evaluation was performed according to our usual practice, based on echocardiography (transthoracic and transesophageal) and cardiac computed tomography (CT) (Figures 1, 2). Considering that this initial evaluation was unsatisfactory, we favored an additional imaging modality rather than renewing potentially invasive or deleterious examinations. Cardiac magnetic resonance was then performed at 1.5 Tesla (Discovery MR 450 W, GEHC) in order to assess 4D flow imaging (TR 4.3, TE 2.4, VPS 3, 1.4 × 1.4 × 1.2 mm) (Figure 3). 4D flow imaging is an innovative modality of blood flow imaging with 3-dimensional time-resolved, phase-contrast cardiac magnetic resonance. This technique has already shown its feasibility and interest in many cardiovascular areas, including the study of prosthetic valves [4, 5]. Full examination was done in a free-breathing, single, acquisition (acquisition time 8 min, post-processing time 20 min) [6]. Retrospective valve tracking was used to quantify PVL after post-processing (Arterys, California, USA). Inlet and outlet dimensions were determined using planimetry on dynamic and almost isotropic acquisition after adjustment for every timeframe and direction of the regurgitant flow. Hemodynamic characteristics could be determined by direct measurement giving celerity and regurgitant volume....