Association of Left Ventricular Remodeling Assessment by Cardiac Magnetic Resonance With Outcomes in Patients With Chronic Aortic Regurgitation

IMPORTANCE Chronic aortic regurgitation (AR) causes left ventricular (LV) volume overload, which results in progressive LV remodeling negatively affecting outcomes. Whether cardiac magnetic resonance (CMR) volumetric quantification can provide incremental risk stratification over standard clinical and echocardiographic evaluation in patients with chronic moderate or severe AR is unknown. OBJECTIVE To compare LV remodeling measurements by CMR and echocardiography between patients with and without heart failure symptoms and to verify the association of remodeling measurements of patients with chronic moderate or severe AR but no or minimal symptoms with clinical outcomes receiving medical management. DESIGN, SETTING, AND PARTICIPANTS This multicenter retrospective cohort study included consecutive patients with at least moderate chronic native AR evaluated by 2-dimensional transthoracic echocardiography and CMR examination within 90 days from each other between January 2012 and February 2020 at Allina Health System. Data were analyzed from June 2021to January 2022. EXPOSURES Clinical evaluation and risk stratification by CMR. MAIN OUTCOMES AND MEASURES The end point was a composite of death, heart failure hospitalization, or progression of New York Heart Association functional class while receiving medical management, censoring patients at the time of aortic valve replacement (when performed) or at the end of follow-up. RESULTS Of the 178 included patients, 119 (66.9%) were male, 158 (88.8%) presented with no or minimal symptoms (New York Heart Association class I or II), and the median (IQR) age was 58 (44-69) years. Compared with patients with no or minimal symptoms, symptomatic patients had greater LV end-systolic volume index (LVESVi) by CMR (median [IQR], 66 [46-85] mL/m(2 )vs 42 [30-58] mL/m(2); P < .001), while there were no significant differences by echocardiography (LVESVi: median [IQR], 38 [30-58] mL/m(2) vs 27 [20-42] mL/m(2); P = .07; LV end-systolic diameter index: median [IQR], 21 [17-25] mm/m(2) vs 18 [15-22] mm/m(2); P = .17). During the median (IQR) follow-up of 3.3 (1.6-5.8) years, 50 patients with no or minimal symptoms receiving medical management developed the composite end point, which, in multivariate analysis adjusted for age and EuroSCORE II, was independently associated with LVESVi of 45 mL/m(2) or greater and aortic regurgitant fraction of 32% or greater, the latter adding incremental prognostic value to CMR volumetric assessment. CONCLUSIONS AND RELEVANCE In patients with chronic moderate or severe AR, patients presenting with heart failure symptoms have greater LVESVi by CMR than those with no or minimal symptoms. In patients with no or minimal symptoms, CMR quantification of LVESVi and AR severity may identify those at risk of death or incident heart failure and therefore should be considered in the clinical evaluation and decision-making of these patients.