BACKGROUND: There is debate whether the J-curve relationship between cardiac event risk and DBP is because of inherent cardiac risk or is a consequence of blood pressure (BP) lowering therapy. METHODS: We examined the association between the cardiovascular risk marker amino-terminal-pro-B-type natriuretic peptide (NT-proBNP) and DBP in 1781 women and 2211 men aged at least 60 years with one or more cardiovascular risk factors; exclusion criteria were known heart failure or cardiac abnormality on a cardiac imaging study. RESULTS: The lowest median serum NT-proBNP levels were for DBP 85-89 mmHg for both women and men. DBP less than 70 mmHg in women and less than 80 mmHg in men was associated with higher NT-proBNP levels than the levels at DBP 85-89 mmHg, and this relationship was present for those with SBP equal to or less than 140 and SBP greater than 140 mmHg. In conditional logistic regression models, the association of elevated NT-proBNP levels with low DBP in women was no longer statistically significant after adjustment for age, ischaemic heart disease (IHD), pulse rate, atrial fibrillation, haemoglobin and glomerular filtration rate, whereas the association in men was no longer statistically significant after adjustment for age and IHD. By contrast, the association between elevated NT-proBNP levels and low DBP remained statistically significant after adjustment for the number of antihypertensive drug classes alone or together with all antihypertensive drugs, including β-blocker therapy. CONCLUSION: There was a J-curve relationship between the cardiovascular risk marker NT-proBNP and DBP that was explained by the clinical variables and not by the BP-lowering therapy. ; Duncan J. Campbell, Michele McGrady, David L. Prior, Jennifer M. Coller, Umberto Boffa, Louise Shiel, Danny Liew, Rory Wolfe, Simon Stewart, Christopher M. Reid, Henry Krum