N-terminal pro-brain natriuretic peptide (NT-proBNP) is a strong marker of cardiovascular disease with recent evidence that inflammation may also influence its levels; discrimination of this confounding variable is of particular interest in rheumatic diseases. Therefore, we evaluated NT-proBNP in ankylosing spondylitis (AS) patients pre- and post-TNF blocker to determine the possible association between NT-proBNP levels and inflammatory parameters. Forty-five consecutive AS patients without previous/current cardiovascular disease or systolic myocardial dysfunction, who were eligible to anti-TNF therapy, were prospectively enrolled. All patients received TNF blockers and they were evaluated for circulating NT-proBNP levels, clinical and laboratory parameters of disease activity, traditional cardiovascular risk factors, and conventional and tissue Doppler imaging echocardiography at baseline (BL) and 6 months after (6M) treatment. At BL, all patients had active AS, NT-proBNP levels had a median of 36 (20-72) pg/mL and 11 % were high in spite of no systolic alteration. Multiple linear regression analysis revealed that this peptide, at BL, was independently correlated with erythrocyte sedimentation rate (ESR) (p < 0.001), age (p = 0.01), and pulse pressure (p = 0.01). After 6M, all disease parameters improved and NT-proBNP levels were significantly reduced [24 (16-47) pg/mL, p = 0.037] compared to BL. Changes in NT-proBNP were positively correlated with ESR changes (r = 0.41, p = 0.006). Cardiovascular risk factors remained stable during follow-up. In conclusion, our data suggest that elevations of NT-proBNP should be interpreted with caution in active AS patients with no other evidence of cardiovascular disease. The short-term reduction of NT-proBNP levels in these patients receiving anti-TNF therapy appears to reflect an improvement in inflammatory status.