Proximal isovelocity surface area (PISA) has been proposed as a quantitative method to assess the severity of aortic regurgitation (AR). Yet the accuracy of this method in patients with eccentric AR jets is unknown. The aims of this study were to compare the accuracy of the PISA method for the quantification of AR severity in patients with central versus eccentric AR jets and to verify whether imaging from the left parasternal instead of the apical window improves the accuracy of the PISA method in patients with eccentric jets. Fifty patients with AR (21 with central jets and 29 with eccentric jets) underwent PISA and phase-contrast cardiac magnetic resonance (CMR) measurements of AR volume. In patients with eccentric AR jets, PISA measurements obtained from the left parasternal and apical windows were compared. In patients with central AR jets, CMR- and PISA-derived AR volumes were similar (28 +/- 19 vs 30 +/- 20 ml, p = 0.34), were strongly correlated (r = 0.92, p <0.0001), and differed minimally from each other (by 2 +/- 8 ml). In patients with eccentric AR jets, PISA-derived AR volumes underestimated those measured by CMR (38 +/- 22 vs 51 +/- 27 ml, bias -13 +/- 20 ml) and were correlated only fairly (r = 0.69, p <0.001). Imaging from the left parasternal window eliminated the differences between CMR- and PISA-derived AR volumes (51 +/- 27 vs 53 +/- 26 ml, p = 0.24) and improved the correlation between the 2 measures (r = 0.95). In conclusion, in patients with eccentric AR jets imaged from the apical window, the PISA method significantly underestimated AR severity. This was no longer the case when imaging was performed from the left parasternal instead of the apical window.