Predictors of long-term survival after percutaneous aortic valvuloplasty: Report of the mansfield scientific balloon aortic valvuloplasty registry

Abstract

Percutaneous balloon aortic valvuloplasty was used to prospectively treat 492 elderly, symptomatic, nonsurgical patients suffering from severe aortic stenosis in 27 centers in North America and Europe. At 1 year the overall survival rate was 64% and the event-free survival rate (survival free of valve replacement or repeat valvuloplasty) was 43%. Clinical, catheterization and procedural variables were assessed to define prognostic variables.Univariate analysis revealed that patients who survived had a lesser frequency of previous myocardial infarction (2% versus 6%, p < 0.005), lower incidence of severe ventricular dysfunction (22% versus 48%, p < 0.001) and lower incidence of symptoms of heart failure (60% versus 75%, p < 0.02). History of angina (56% versus 45%, p = NS) and syncope (23% versus 16%, p = NS) were similar for both groups. Values obtained at cardiac catheterization that differed in survivors and nonsurvivors included lower pulmonary artery systolic pressure (43 ± 1 versus 54 ± 2 mm Hg, p < 0.001), lower mean pulmonary artery pressure (28 ± 1.0 versus 36 ± 1.0 mm Hg, p < 0.001) and larger initial valve area (0.52 ± 0.01 versus 0.47 ± 0.02 cm2, p = 0.006).Discriminate function analysis was performed to identify variables that independently predicted improved probability of survival. Eight variables were significantly and independently predictive. These included age, initial cardiac output, initial left ventricular systolic pressures, initial left ventricular end-diastolic pressures, presence of coronary artery disease, New York Heart Association dyspnea classification, number of balloon inflations and final valve area. From this analysis, patient survival probabilities at 1 year varying between 80% and 20% could be calculated. Although both initial and final severity of aortic stenosis were prognostically important on univariate analysis, discriminant function analysis revealed that only final severity of aortic stenosis was prognostic.These observations suggest that subgroups of patients with lower or higher mortality rates can be identified after balloon aortic valvuloplasty. The presence of left ventricular dysfunction, the presence of coronary artery disease and the use of multiple balloon inflations adversely affect long-term prognosis. Conversely, effective valvuloplasty provides modest improvement in survival probability.