Usefulness of the B-Type Natriuretic Peptides in Low Ejection Fraction, Low-Flow, Low-Gradient Aortic Stenosis Results from the TOPAS Multicenter Prospective Cohort Study

Abstract

Background: Patients with low left ventricular ejection fraction (LVEF), low-flow, low-gradient (i.e. classical low flow [CLF]) aortic stenosis (AS) have a dismal short-term outcome without aortic valve replacement (AVR) but high operative mortality. We hypothesized that brain natriuretic peptides (BNP/NT-proBNP) can risk stratify patients with CLF AS and may assist in clinical decision-making. Methods: Patients with aortic valve area ≤1.2 cm2, mean transvalvular gradient <40 mmHg, and left ventricular ejection fraction <50%, were prospectively recruited. BNP and/or NT-proBNP were measured at baseline. Results: Among 234 patients (77 [68–83] years, 76% male), BNP > 550 pg/ml or NT-proBNP > 1,600 pg/ml (85% and 93% sensitivity, respectively, to correctly classify 1-year death) strongly predicted all-cause mortality (adjusted HR = 2.53 [1.68–3.81], p < 0.001) outperforming flow reserve and baseline LVEF (all likelihood ratio p ≤ 0.02). For both natriuretic peptides, spline curve analysis showed gradual increase in mortality with higher biomarkers levels, which was blunted by AVR. In a head-to-head comparison (n = 104), NT-proBNP appeared to have superior incremental prognostic value than BNP (likelihood-ratio p < 0.001 vs. p = 0.07). Baseline NT-proBNP ≥ 1,600 pg/ml or BNP ≥ 550 pg/ml, identified: i) a high-risk cohort with a dismal outcome under conservative management, but a markedly better survival associated with early AVR (adjusted HR = 0.41 [0.25–0.65], p < 0.001); and ii) a low-risk cohort with an excellent 1-year survival (94 ± 4%) with conservative management or deferred AVR. Conclusion: In patients with CLF AS, BNP/NT-proBNP have the potential to identify high-risk patients who may benefit from early AVR.