Abstract In subjects with aortic regurgitation (AR) or mitral regurgitation (MR), transthoracic echocardiography (TTE) is recommended for surveillance. Few prospective studies have directly compared the ability of TTE and cardiac magnetic resonance (CMR) to predict clinical outcomes in AR and MR. We hypothesized that, given its higher reproducibility, CMR would predict the need for valve surgery or heart failure (HF) hospitalization better than TTE. Quantitative TTE and CMR were performed on the same day for 51 subjects: 29 with chronic AR and 22 with chronic, primary MR for quantification of valve regurgitation. Baseline measures of valve regurgitation were compared to the combined primary endpoint of new HF and valve surgery using receiver operating characteristics, simple logistic regression and Kaplan-Meier survival analyses. The primary endpoint occurred in 5 AR subjects (all surgery) and 8 MR subjects (7 surgery, 1 HF) after a mean follow-up of 4.4 ± 1.5 years. For AR, CMR-derived regurgitant volume >50mL identified those at high risk with 50% undergoing valve surgery versus 0% for those with regurgitant volume ≤50 ml, and was more strongly associated with outcomes than regurgitant volume by TTE (p<0.05). For MR, 6.8% of those with regurgitant volume by TTE ≤30 mL developed the primary endpoint versus 70% in those with regurgitant volume >30 mL. Regurgitant volume by CMR showed no significant separation of survival curves for MR. In conclusion, regurgitant volume by CMR was more predictive of outcomes than by TTE in subjects with AR. In MR, the two modalities performed similarly.