Long-term Evolution of Mitral Regurgitation and Cardiac Remodeling After Aortic Valve Replacement: A Single-Center Retrospective Cohort Study

Objective Mitral regurgitation (MR) frequently accompanies aortic valve disease and may improve after aortic valve replacement (AVR) through altered loading conditions and reverse cardiac remodeling. Evidence describing paired preoperative and long-term echocardiographic changes in routine clinical practice remains limited. We evaluated long-term MR burden and changes in MR severity, cardiac structure, systolic function, and diastolic indices after AVR. Methods This single-center retrospective cohort included adults who underwent AVR from January 2015 through December 2024, had at least one postoperative TEE within 10 years after AVR, and did not undergo concomitant mitral valve intervention. All echocardiographic examinations included in the analysis were TEE examinations; TTE and TEE examinations were not mixed. The preoperative TEE closest to AVR within the preceding 365 days and the latest TEE with a reportable MR grade more than 365 days after AVR were selected. Long-term TEE was obtained through routine clinical care rather than a protocol-mandated surveillance schedule. MR was graded ordinally from none/not detected to severe. Because paired differences were frequently skewed and MR severity was ordinal, paired changes were assessed using the Wilcoxon signed-rank test or exact McNemar test. Between-group comparisons used the Mann-Whitney U test or Fisher exact test. Results The cohort included 735 patients, of whom 263 (35.8%) had a long-term TEE with a reportable MR grade at a median of 1093.0 [644.5, 1791.0] days after AVR. Long-term moderate-or-worse MR was present in 8/263 patients (3.0%; 95% CI, 1.5%-5.9%). Among 189 patients with paired TEE MR grades, MR improved in 33 (17.5%; 95% CI, 12.7%-23.5%), remained unchanged in 131 (69.3%; 95% CI, 62.4%-75.4%), and worsened in 25 (13.2%; 95% CI, 9.1%-18.8%). Moderate-or-worse MR decreased from 10/189 (5.3%) preoperatively to 6/189 (3.2%) at long-term follow-up (P=0.344), whereas MR ordinal score was unchanged overall (P=0.865). Paired analyses showed reductions in LVEDD, LVESD, IVSd, LVPWd, and pulmonary artery systolic pressure (all P<0.001), with a modest increase in LVEF (P=0.021). E/A and lateral e′ increased (P=0.003 and P<0.001), whereas septal e′, average E/e′, and LAD did not change significantly. Atrial fibrillation was more frequent among patients with long-term moderate-or-worse MR (50.0% vs 2.7%, P<0.001), although only 8 outcome events were available. Conclusions Among patients with a reportable long-term TEE after AVR, clinically important long-term MR was uncommon and MR severity was most often stable, while ventricular dimensions, wall thickness, and pulmonary pressure decreased over time. These findings suggest that structural reverse remodeling after AVR may occur even when ordinal MR grade remains unchanged. Confirmation in cohorts with more complete and systematic follow-up, quantitative MR assessment, and analyses that address differential follow-up is warranted.