N-terminal pro-brain natriuretic peptide (NT-proBNP) measurements can be used to rule out heart failure in patients with sinus rhythm. Atrial fibrillation often coexists with heart failure but affects NT-proBNP levels. This study aims to identify the optimal NT-proBNP cut-off value for ruling out heart failure among atrial fibrillation patients.

This prospective study included 409 atrial fibrillation patients admitted to the emergency department. The inclusion criterion was documented atrial fibrillation on a 12‑lead electrocardiogram. All patients completed a NT-proBNP blood sample, a chest X-ray and an echocardiogram. Heart failure was defined as a left ventricular ejection fraction of <40%.

In total, 409 patients were included (mean age: 75.2 ± 11.6). The median NT-proBNP level was 2577 ng/L (quartiles: 1185-5438) and 21% had heart failure. We found a lower median NT proBNP level of 3187 ± 3973 ng/L in patients without heart failure compared to 9254 ± 8008 ng/L in patients with heart failure (absolute difference: 4131, 95% (CI): 3299-4986, p < 0.001). The area under the receiver operating characteristic curve for diagnosing heart failure was 0.82 (95% confidence interval: 0.77–0.87). The optimal cut-off value for ruling out heart failure was 739 ng/L with a sensitivity of 99%, a specificity of 18%, and a negative predictive value of 98%.

NT-proBNP can be used to rule out heart failure in atrial fibrillation patients with a high negative predictive value, but low specificity.

Trial registration number: NCT04125966. NCT04125966.