NT-proBNP for Risk Prediction in Heart Failure

• DOI: 10.1016/j.jchf.2021.05.014
• Publication Date: 2021-07-07T21:36:29Z
• Authors: Giuseppe Vergaro; Francesco Gentile; Laura M.G. Meems; Alberto Aimo; James L. Januzzi; A. Mark Richards; Carolyn S.P. Lam; Roberto Latini; Lidia Staszewsky; Inder S. Anand; Jay N. Cohn; Thor Ueland; Lars Gullestad; Pål Aukrust; Hans-Peter Brunne...; Antoni Bayes-Genis; Josep Lupón; Akiomi Yoshihisa; Yasuchika Takeishi; Michael Egstrup; Ida Gustafsson; Hanna K. Gaggin; Kai M. Eggers; Kurt Huber; Greg D. Gamble; Lieng H. Ling; Kui Tong Gerard L...; Poh Shuah Daniel Yeo; Hean Yee Ong; Fazlur Jaufeerally; Tze P. Ng; Richard Troughton; Robert N. Doughty; Gerry Devlin; Mayanna Lund; Alberto Giannoni; Claudio Passino; Rudolf A. de Boer; Michele Emdin
• Keywords: Underweight; Obesity paradox

The goal of this study was to assess the predictive power of N-terminal pro-B-type natriuretic peptide (NT-proBNP) and the decision cutoffs in heart failure (HF) across body mass index (BMI) categories.Concentrations of NT-proBNP predict outcome in HF. Although the influence of BMI to reduce levels of NT-proBNP is known, the impact of obesity on prognostic value remains uncertain.Individual data from the BIOS (Biomarkers In Heart Failure Outpatient Study) consortium were analyzed. Patients with stable HF were classified as underweight (BMI <18.5 kg/m2), normal weight (BMI 18.5-24.9 kg/m2), overweight (BMI 25-29.9 kg/m2), and mildly (BMI 30-34.9 kg/m2), moderately (BMI 35-39.9 kg/m2), or severely (BMI ≥40 kg/m2) obese. The prognostic role of NT-proBNP was tested for the endpoints of all-cause and cardiac death.The study population included 12,763 patients (mean age 66 ± 12 years; 25% women; mean left ventricular ejection fraction 33% ± 13%). Most patients were overweight (n = 5,176), followed by normal weight (n = 4,299), mildly obese (n = 2,157), moderately obese (n = 612), severely obese (n = 314), and underweight (n = 205). NT-proBNP inversely correlated with BMI (β = -0.174 for 1 kg/m2; P < 0.001). Adding NT-proBNP to clinical models improved risk prediction across BMI categories, with the exception of severely obese patients. The best cutoffs of NT-proBNP for 5-year all-cause death prediction were lower as BMI increased (3,785 ng/L, 2,193 ng/L, 1,554 ng/L, 1,045 ng/L, 755 ng/L, and 879 ng/L, for underweight, normal weight, overweight, and mildly, moderately, and severely obese patients, respectively) and were higher in women than in men.NT-proBNP maintains its independent prognostic value up to 40 kg/m2 BMI, and lower optimal risk-prediction cutoffs are observed in overweight and obese patients.