Computed tomography-based sizing recommendations for transcatheter aortic valve replacement with balloon-expandable valves: Comparison with transesophageal echocardiography and rationale for implementation in a prospective trial.

Background: Computed tomography (CT) annular measurements are predictive of paravalvular regurgitation after transcatheter aortic valve replacement (TAVR) which is a predictor of mortality and morbidity.

Objectives: To demonstrate the rationale and potential implications of new CT sizing recommendations for TAVR.

Methods: The CT sizing recommendations aim to ensure routine transcatheter heart valve (THV) oversizing of the aortic annular area [(THV external area/systolic annular area - 1) × 100; range, 1%-20%; target, 10%-15%]. Consecutive patients (n = 120) underwent CT before TAVR with balloon-expandable valves sized by transesophageal echocardiography (TEE). Retrospectively, the CT-recommended THV size was compared with the actual size implanted.

Results: Compared with TEE, application of the newly developed CT-based sizing recommendations would have led to implantation of a larger valve in 33.3% (40/120), no change in valve size in 55.8% (67/120), and a smaller valve in 10.8% (13/120). In patients when CT recommended a larger valve, the incidence of at least moderate paravalvular regurgitation was 25% (10/40) compared with 4.5% (3/67; P < 0.01) when both TEE and CT recommendations were in agreement. Using diastolic versus systolic CT measurements results in 20% of patients receiving smaller THVs. TEE sizing resulted in 33.3% (40/120) of valves being undersized (THV area < CT systolic annular area) with a mean annular oversizing of 9.4% ± 17.4% (range: -21.5% to 65.9%) without annular rupture. In contrast, the CT sizing recommendations results in mean annular oversizing of 13.9% ± 8.0% (range, 1.3%-29.8%).

Conclusion: These CT sizing recommendations enable standardized moderate overexpansion of the aortic annulus. Clinical outcomes from these recommendations are being prospectively assessed in a multicenter trial.