Magnetic resonance derived regurgitant fraction, 3D vena contracta area, and B-natriuretic peptide: the most accurate prognostic markers in patients with chronic asymptomatic aortic regurgitation

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Ministry of Health of the Czech Republic 17-28265A. Background The optimal timing of surgical treatment is challenging in asymptomatic patients with severe aortic regurgitation (AR). A significant number of patients undergoing aortic valve surgery as per current guidelines have irreversible left ventricular (LV) myocardial damage with a negative impact on their long-term postoperative outcome. Purpose We aimed to determine the value of magnetic resonance-, advanced echocardiography–derived indices and serum B-natriuretic peptide level (BNP) to identify patients with chronic asymptomatic AR who are at risk of early disease progression. Methods We prospectively enrolled a group of asymptomatic patients with pure chronic severe AR, without an indication for surgical treatment as per the current guidelines, in a multicenter study. Baseline examination consisted of serum BNP level analysis, comprehensive echocardiography including 3D vena contracta area (VCA) and 2D global longitudinal strain (GLS), and a comprehensive MRI including regurgitant fraction (RF) measurement; all imaging data were analyzed in Corelab. Patient follow-up was every 6 months. The primary endpoint was a new guideline-directed indication for surgical treatment. Results In total, 129 patients were enrolled between 2015 and 2019, the endpoint occurred in 43 patients during a median follow-up of 1375 (IQR 1041-1783) days. Baseline clinical data did not differ between patients with the endpoint (surgical group) and stable patients (medical group). Baseline BNP levels were significantly higher in the surgical group (55 vs. 21 ng/L, P = 0.008). MRI-derived LV ejection fraction was not different (61 vs 61 %, P = 0.815) at baseline while both CMR-derived end-systolic and end-diastolic LV volumes were significantly higher in the surgical group (all P < 0.01). Both baseline 3D VCA and 2D GLS were different between the surgical and medical groups (33 vs. 25 mm2, P = 0.006 and 18 vs 19 %, P = 0.07). Magnetic resonance derived RF was significantly higher in the surgical group (46 vs 32 %) with P < 0.01. A combined model including independent markers such as BNP (cut-off > 31 ng/L), MRI RF (cut-off value > 34 %), 3D VCA (cut-off value 36 mm2) or 2D GLS (cut-off value < 17 %) showed the highest predictive accuracy (P = 0.002) to predict the primary endpoint. (Figure 1). Conclusion A combination of baseline parameters of AR severity (3D VCA, MRI-derived RF) and myocardial impairment (2D GLS, serum BNP level) present an accurate tool for the identification of patients with chronic asymptomatic AR who are at risk of early disease progression. Abstract Figure 1