Differences in Cardiac Remodeling in Left-Sided Valvular Regurgitation: Implications for Optimal Definition of Significant Aortic Regurgitation

Background: Grading of aortic regurgitation (AR) and mitral regurgitation (MR) is similar in the cardiology guidelines despite distinct differences in left ventricular (LV) adaptive pathophysiology. Objectives: This study compared differences in LV remodeling in patients with similar degrees of AR and MR severity and evaluated optimal cutoffs for significant AR in relation to the outcome of aortic valve replacement or repair (AVR) during follow-up. Methods: From 2008 to 2018, consecutive patients with isolated AR or MR who had cardiac magnetic resonance (CMR) were identified and CMR parameters were compared. Patients with left ventricular ejection fraction (LVEF) <50%, ischemic scar >5%, valve stenosis, or concomitant regurgitation were excluded. Patients were followed longitudinally for AVR. Results: Baseline characteristics of isolated AR (n = 418) and isolated MR (n = 1,073) were comparable except for higher male proportion and hypertension in AR, while heart failure was more prevalent in MR. Indexed LV end-diastolic and end-systolic volumes and mass were higher in AR compared with MR at the same level of regurgitant fraction. During follow-up (mean 2.1 years), 18.7% of AR patients underwent AVR based on symptoms or LV remodeling. Interestingly, 38.0% of patients that underwent AVR within 3 months after CMR did not meet severe AVR by current guidelines of AR severity. AR regurgitant fraction>35% had high sensitivity (86%) and specificity (88%) for identifying patients who underwent AVR. Conclusions: For similar regurgitation severity, LV remodeling is different in AR compared with MR. Cardiac symptoms and significant LV remodeling in AR requiring AVR occur frequently in patients with less severity than currently proposed. The study findings suggest that the optimal threshold for severe AR with CMR is different than MR and is lower than currently stated in the guidelines.