Diagnostic and Prognostic Utility of Cardiac Magnetic Resonance Imaging in Aortic Regurgitation.

Objectives: This study investigated the diagnostic and prognostic value of cardiac magnetic resonance (CMR) imaging in chronic aortic regurgitation (AR).
Background: Accurate quantification of AR severity by echocardiography frequently remains difficult. CMR is recommended as the complementary method; however, its accuracy and prognostic utility remain unknown.
Methods: A total of 232 consecutive patients (34.5% were females 55.5 ± 19.8 years of age) with chronic AR (including 40 with moderate to severe and 44 with severe AR on echocardiography) underwent CMR within 4 weeks of echocardiography. CMR included phase-contrast velocity-encoded imaging for the measurement of regurgitant volume and fraction at the sinotubular junction and assessment of holodiastolic retrograde flow (HRF) in the descending aorta. Significant AR was defined as the presence of HRF on CMR. Patients were followed prospectively, and multivariate Cox regression was applied for outcome analysis using a combination of heart failure, hospitalization, and cardiovascular death as primary endpoint.
Results: AR severity on the basis of echo was reclassified in a significant number of patients according to CMR: 6.8% with mild AR on echo had HRF on CMR, whereas 34.1% with severe AR on echo did not have HRF on CMR and were reclassified as having nonsignificant AR. In 40 patients with uncertain AR severity (moderate to severe) on echo, 45.0% had HRF on CMR, indicating severe AR. Patients were followed for 35.3 ± 26.6 months. During that period, 63 patients (27.2%) reached the combined endpoint, including 43 (18.5%) with heart failure hospitalizations and 20 (8.6%) with cardiovascular deaths. By multivariate regression analysis, including clinical as well as imaging parameters, only N-terminal pro-B-type natriuretic peptide concentration (hazard ratio: 2.184 [95% confidence interval: 1.468 to 3.248]; p < 0.001) and HRF on CMR (hazard ratio: 2.774 [95% confidence interval: 1.131 to 6.802]; p = 0.026) remained significantly associated with outcome.
Conclusions: In chronic AR, CMR has the potential to add important diagnostic and prognostic information.
(Copyright © 2019 American College of Cardiology Foundation. Published by Elsevier Inc. All rights reserved.)