Background: Although the prognostic implications of severe mitral regurgitation (MR) are well recognised, they are less clear in moderate MR. We therefore explored the prognostic impact of both moderate and severe MR within the large National Echocardiography Database Australia cohort.
Methods: Echocardiography reports from 608 570 individuals were examined using natural language processing to identify MR severity and leaflet pathology. Atrial (aFMR) or ventricular (vFMR) functional MR was assessed in those without reported leaflet pathology. Using individual data linkage over median 1541 (IQR 820 to 2629) days, we examined the association between MR severity and all-cause (153 612/25.2% events) and cardiovascular-related mortality (47 840/7.9% events).
Results: There were 319 808 men and 288 762 women aged 62.1±18.5 years, of whom 456 989 (75.1%), 102 950 (16.9%), 38 504 (6.3%) and 10 127 (1.7%) individuals had no/trivial, mild, moderate and severe MR, respectively, reported on their last echo. Compared with those with no/trivial MR (26.5% had leaflet pathology, 19.2% died), leaflet pathology (51.8% and 78.9%, respectively) and actual 5-year all-cause mortality (54.6% and 67.5%, respectively) increased with MR severity. On an adjusted basis (age, sex and leaflet pathology), long-term mortality was 1.67-fold (95% CI 1.65 to 1.70) and 2.36-fold (95% CI 2.30 to 2.42) higher in moderate and severe MR cases (p<0.001) compared with no/trivial MR. The prognostic pattern for moderate and severe MR persisted for cardiovascular-related mortality and within prespecified subgroups (leaflet pathology, vFMR or aFMR, and age<65 years).
Conclusions: Within a large real-world clinical cohort, we confirm that conservatively managed severe MR is associated with a poor prognosis. We further reveal that moderate MR is associated with increased mortality, irrespective of underlying aetiology.
Trial registration: Australian New Zealand Clinical Trials Registry (ACTRN12617001387314).
Keywords: Echocardiography; Mitral regurgitation.
© Author(s) (or their employer(s)) 2024. No commercial re-use. See rights and permissions. Published by BMJ.