Valve phenotype and likelihood of surgery in patients with bicuspid aortic valve

Abstract

Abstract Background It is well established, that patients with bicuspid aortic valve (BAV) are at increased risk of developing severe aortic valve and/or aortic disease early in life. Knowledge of factors placing patients at risk of early surgery is therefore essential. Several studies have found associations between aortic valve phenotype, aortopaty and type of valve dysfunction. Purpose To characterize the valve phenotype associated with increased likelihood of early surgery of the aortic valve and aorta in a large cohort of BAV patients. Methods A retrospective study of adult BAV patients seen in the outpatient clinics at two hospitals in Denmark from 2006 until May 2020. Clinical and anatomical data were obtained retrospectively from electronic health charts and hospital echocardiography databases. Bicuspid valve morphology was classified according to Sievers Classification; no raphe (Type 0), one raphe (Type 1 with fusion of the right-noncoronary cusps (R/N), Left-noncoronary cusp fusion (L/N) and Left-right coronary cusp fusion (L/R)) or 2 raphes (Type 2). Likelihood of surgery was calculated using odds ratio (OR). We performed multivariate regression models to adjust for potential confounding by sex, age, coarctatio aorta, aortic dilatation and cardiovascular risk factors. Results A total 983 BAV patients were identified of whom 877 had an available baseline echocardiography and were included. Clinical and echocardiographic characteristics are seen in Table 1. Noteworthy is that Type 2 patients had significantly higher occurrence of moderate-severe aortic regurgitation when compared to the whole population (38.9% vs 18.3%, p<0.01). During the study period 305 patients (34.8%) underwent surgery. Median age at time of surgery was 62 (IQR 55; 69) years. Using the most common phenotype (Type 1 L/R fusion) as a reference, patients with Type 0 had a lower likelihood of surgery (unadjusted OR 0.58, 95% CI: 0.39–0.85), while patients with Type 2 had a significantly higher likelihood of surgery (OR 2.76, 95% CI: 1.05–7.23). In a multiple regression analysis, adjusting for age, sex, coarctatio aorta and aortic dilatation did not change the primary finding of association between BAV phenotype and OR for surgery. Likelihood of surgery was lower for women (OR 0.66, 95% CI: 0.46–0.96) and increased with age. Further adjustments for cardiovascular risk factors (mentioned in Table 1) did not change the results. Median age at time of surgery was younger for patients with Type 2 (59 years, IQR 44; 65). Indications for valve surgery are shown in Table 2. Conclusion In this study we found significant association between valve phenotype and likelihood of surgery of the valve or aorta in patients with bicuspid aortic valve. Using BAV Type 1 L/R as reference, likelihood of surgery was lower in patients with BAV type 0, and higher in patients with BAV type 2. Results were consistent after adjustment for confounders in multivariate analyses. Funding Acknowledgement Type of funding sources: None.