Planimetry of Aortic Valve Area in Aortic Stenosis by Magnetic Resonance Imaging

Authors:
  • BDBehrus Djavidani
    Behrus Djavidani
  • JSJohannes Seitz
    Johannes Seitz
  • WNWolfgang Nitz
    Wolfgang Nitz
  • FSFranz-Xaver Schmid
    Franz-Xaver Schmid
  • FMFrank Muders
    Frank Muders
  • SBStefan Buchner
    Stefan Buchner
  • SFStefan Feuerbach
    Stefan Feuerbach
  • GRGünter Riegger
    Günter Riegger
  • ALAndreas Luchner
    Andreas Luchner
Publisher website
Google Scholar
Abstract
The aim of the study was to determine whether noninvasive planimetry of aortic valve area (AVA) by magnetic resonance imaging (MRI) is feasible and reliable in patients with valvular aortic stenosis in comparison to transesophageal echocardiography (TEE) and catheterization. Planimetry of AVA by MRI (MRI-AVA) was performed on a clinical magnetic resonance system (1.5-T Sonata, Siemens Medical Solutions) in 33 patients and compared with AVA calculated invasively by the Gorlin-formula at catheterization (CATH-AVA, n = 33) as well as to AVA planimetry by multiplane TEE (TEE-AVA, n = 27). Determination of MRI-AVA was possible with an adequate image quality in 82% (27/33), whereas image quality of TEE-AVA was adequate only in 56% (15/27) of patients because of calcification artifacts (P = 0.05). The correlation between MRI-AVA and CATH-AVA was 0.80 (P < 0.0001) and the correlation of MRI-AVA and TEE-AVA was 0.86 (P < 0.0001). MRI-AVA overestimated TEE-AVA by 15% (0.98 +/- 0.31 cm2 vs. 0.85 +/- 0.3 cm2, P < 0.001) and CATH-AVA by 27% (0.94 +/- 0.29 cm2 vs. 0.74 +/- 0.24 cm2, P < 0.0001). Nevertheless, a MRI-AVA below 1,3 cm2 indicated severe aortic stenosis (CATH-AVA < 1 cm2) with a sensitivity of 96% and a specificity of 100% (ROC area 0.98). Planimetry of aortic valve area by MRI can be performed with better image quality as compared with TEE. In the clinical management of patients with aortic stenosis, it has to be considered that MRI slightly overestimates aortic valve area as compared with catheterization despite an excellent correlation.

This publication has 16 references indexed in Scilit: