How obesity affects the cut-points for B-type natriuretic peptide in the diagnosis of acute heart failure. Results from the Breathing Not Properly Multinational Study.

BACKGROUND: B-type natriuretic peptide (BNP) is valuable in diagnosing heart failure (HF), but its utility in obese patients is unknown. Studies have suggested a cut-point of BNP > or = 100 pg/mL for the diagnosis of HF; however, there is an inverse relation between BNP levels and body mass index. We evaluated differential cut-points for BNP in diagnosing acute HF across body mass index levels to determine whether alternative cut-points can improve diagnosis. METHODS: The Breathing Not Properly Multinational Study was a 7-center, prospective study of 1586 patients who presented to the Emergency Department with acute dyspnea. B-type natriuretic peptide was measured on arrival. Height and weight data were available for 1368 participants. The clinical diagnosis of HF was adjudicated by 2 independent cardiologists who were blinded to BNP results. RESULTS: Heart failure was the final diagnosis in 46.1%. Mean BNP levels (pg/mL) in lean, overweight/obese, and severely/morbidly obese patients were 643, 462, and 247 for patients with acute HF, and 52, 35, and 25 in those without HF, respectively (P < .05 for all comparisons except 35 vs 25). B-type natriuretic peptide cut-points to maintain 90% sensitivity for a HF diagnosis were 170 pg/mL for lean subjects, 110 pg/mL for overweight/obese subjects, and 54 pg/mL in severely/morbidly obese patients. CONCLUSIONS: Body mass index influences the selection of cut-points for BNP in diagnosing acute HF. A lower cut-point (BNP > or = 54 pg/mL) should be used in severely obese patients to preserve sensitivity. A higher cut-point in lean patients (BNP > or = 170 pg/mL) could be used to increase specificity.

Duke Scholars

Author Nancy Marie Allen LaPointe Medicine, Cardiology

Author Elizabeth Ray DeLong Biostatistics & Bioinformatics, Division of Biostatistics

Author Bradley Gordon Hammill Population Health Sciences

Author Lawrence H. Muhlbaier Biostatistics & Bioinformatics, Division of Biostatistics

Author Judith Mae Kramer Medicine, General Internal Medicine