Prognostic value of the cardiac damage staging system and CPETecho across the entire spectrum of aortic stenosis

Abstract Background Cardiac damage correlates with adverse events in patients with severe aortic stenosis (AS). Purpose To evaluate the prognostic value of the cardiac damage staging system and the integrated use of cardiopulmonary exercise testing and echocardiography (CPET-echo) across the entire spectrum of aortic stenosis. Methods A total of 519 patients (65% men; 74±10 years) with ≥mild AS and no other significant primary valvular disease underwent CPET-echo. On the basis of a cardiac damage staging scheme, patients were categorized into five independent stages: Stage 0 – no other cardiac damage detected; Stage 1 – LV damage (LV hypertrophy, diastolic or systolic dysfunction); Stage 2 – LA or mitral valve damage (enlarged LA; presence of atrial fibrillation or ≥moderate functional mitral regurgitation); Stage 3 – pulmonary hypertension or ≥moderate tricuspid regurgitation (TR) and Stage 4 – RV damage. Patients were subsequently evaluated for the occurrence of the combined endpoint heart failure (HF) death and HF hospitalizations. Results Severe, moderate and mild AS was present in 55% (40% low flow, low gradient), 26% and 19% of patients, respectively. Cardiac damage of any grade was observed in 79% of all patients. Among them, 31% had Stage 1 involvement, and an equal proportion presented with Stage 2. Stage 3 was identified in 3% of cases, while right ventricular damage (Stage 4) was found in 17% of patients. Remarkably, 80% of patients with mild AS had cardiac damage, most of them (56%) Stage 2 or higher. Over a median follow-up of 39 months [23-49], 71 patients reached the primary endpoint: 51 with severe AS (18%), 10 with moderate AS (7%) and 10 with mild AS (10%). In the multivariable Cox regression, with aortic valve replacement included as time-dependent covariate, cardiac damage stage, NT-proBNP, elevated left ventricular filling pressures at rest or exercise (positive diastolic stress test) and exercise induced pulmonary hypertension (exPHT) were independently associated with the adverse outcome (Figure 1). Incorporating cardiac damage and CPETecho derived parameters (exPHT and elevated left ventricular filling pressures at rest or exercise) to a baseline model (age, AS severity and NT-proBNP) significantly improved event prediction: likelihood χ2=51, p <0.001 and AUC 0.702 versus 0.766, p=0.03 (Figure 1). Figure 2 depicts event-free survival at 48 months stratified according to the cardiac damage stage. Conclusions Cardiac damage is present in a significant proportion of patients with even mild AS. It is significantly associated with HF death and HF hospitalizations, independently of AS severity. CPET-echo may further refine risk stratification for HF across the entire spectrum of aortic stenosis.