Use of cardiovascular magnetic resonance imaging for TAVR assessment in patients with bioprosthetic aortic valves: comparison with computed tomography

Abstract

Purpose: Transcatheter aortic valve replacement (TAVR) has been successfully used to treat patients with failing aortic bioprostheses. Computed tomography (CT) is the usual method of pre-procedural imaging for TAVR in the native position; however, the optimal modality for valve-in-valve procedures has not been established. CT can assess intracardiac anatomy and is superior to cardiovascular magnetic resonance (CMR) in the assessment of coronary artery disease. However, CMR can provide superior haemodynamic information, does not carry the risk of ionising radiation, and may be performed without contrast in patients with renal insufficiency. In this study, we compared CT and CMR for the evaluation of TAVR in a small cohort of patients with existing aortic bioprostheses.

Materials and methods: 21 patients with aortic bioprostheses were prospectively evaluated by CT and CMR, as pre-assessment for TAVR; agreement between measurements of aortic geometries was assessed.

Results: 16/21 patients had aortic bioprostheses constructed with a metal ring, and 5/21 patients had a metal strut construction. Patients with metal struts had significant metal-artefact on CMR, which compromised image quality in this region. There was good agreement between CT and CMR measurements of aortic geometry. The mean difference (d) in annulus area-derived diameter was 0.5mm (95% limits of agreement [L.A] 4.2mm). There was good agreement between modalities for the cross-sectional area of the sinuses of valsalva (d 0.5 cm(2), L.A 1.4 cm(2)), sinotubular junction (d 0.9 cm(2), L.A 1.5 cm(2)), and ascending aorta (d 0.6 cm(2), L.A 1.4 cm(2)). In patients without metal struts, the left coronary artery height d was 0.7 mm and L.A 2.8mm.

Conclusions: Our analysis shows that CMR and CT measurements of aortic geometry show good agreement, including measurement of annulus size and coronary artery location, and thus provide the necessary anatomical information for valve-in-valve TAVR planning. However, in patients with metal strut aortic valve constructions, CT should be performed due to the presence of limiting metal artefacts on CMR. CMR may be considered as an appropriate alternative to CT in patients in whom iodinated contrast agents are contraindicated or where additional haemodynamic assessment with phase-contrast CMR is required.