Anatomic estimation of aortic stenosis severity vs "fusion" of data from computed tomography and Doppler...

Two-dimensional, transthoracic echocardiography does not account for the noncircular anatomy of the left ventricular outflow tract (LVOT) and may therefore underestimate LVOT area. Fusion of computed tomography (CT)-derived LVOT area and Doppler-derived flow data has been proposed to improve assessment of aortic valve area (AVA) and classification of aortic stenosis severity. For hemodynamic reasons, effective AVA has to be smaller than anatomic AVA. The aim of the study was to test the "fusion approach" by comparing effective CT-derived AVA with anatomic AVA from CT planimetry.

Data of 244 consecutive patients (mean age 81 ± 5 years, 61% female) with aortic stenosis were retrospectively analyzed comparing effective AVA (calculated from the continuity equation using CT-LVOT and transthoracic Doppler measurements) with anatomic AVA based on CT planimetry. Substituting the LVOT area from transthoracic echocardiography (TTE) by the CT-LVOT resulted in an increase in AVA from 0.74 ± 0.15 to 0.92 ± 0.18cm² (P < .01), which was larger than anatomic AVA (0.82 ± 0.15cm²). Similar results were obtained based on planimetry from transesophageal echocardiography (TEE; AVA 0.79 ± 0.14cm², P < .01 vs CT-LVOT) and in the subgroup presenting with low-gradient severe aortic stenosis and preserved ejection fraction (n = 67, AVA from TTE 0.76 ± 0.09; from CT-LVOT 0.97 ± 0.14; CT planimetry 0.86 ± 0.12; TEE planimetry 0.82 ± 0.13cm²).

Fusion of CT-derived LVOT area with Doppler echocardiography results in a calculated effective AVA that is larger than the corresponding anatomic AVA. Therefore, adjustment of partition values may be warranted when using this approach.