Transthoracic echocardiography (TTE) and cardiac magnetic resonance (CMR) are current standard for assessing aortic regurgitation (AR). Regurgitant fraction (RF) can also be estimated by Doppler examination of the left subclavian artery (LSA-Doppler). However, a comparison of AR grading scales using these methods and a TTE multiparametric approach as reference is lacking. We evaluated the severity of AR in 73 patients (58 ± 15 years; 57 men), with a wide spectrum of AR of the native valve. Using a recommended TTE multiparametric approach the AR was divided in none/trace (n = 12), mild (n = 23), moderate (n = 12), and severe (n = 26). RF was evaluated by LSA-Doppler (ratio between diastolic and systolic velocity-time integrals) and by CMR phase-contrast imaging (performed in the aorta 1 cm above the aortic valve); the grading scales were then calculated. There were a good correlation between all methods, but mean RF values were greater with TTE compared with LSA-Doppler and CMR (39 ± 16% vs. 35 ± 18% vs. 32 ± 20%, respectively; p < 0.037). Mean differences in RF values between methods were significant in the groups with mild and moderate AR. Grading scales that best defined the TTE derived AR severity using CMR were: mild, < 21%; moderate, 22 to 41%; and severe, > 42%; and using LSA-Doppler: mild, < 29%; moderate, 30 to 44%; and severe, > 45%. RF values for AR grading using TTE, LSA-Doppler and CMR correlate well but differ in groups with mild and moderate AR when using a recognized multiparametric echocardiographic approach. Clinical prospective studies should validate these proposed modality adjusted grading scales.