Background: A B-type natriuretic peptide (BNP) threshold of 100 pg/ml is used in practice for identification of heart failure, but data about the "normal" distribution of BNP in a large population in primary prevention are rare. We aimed to characterize the BNP distribution in a healthy subset of a population-based cohort and to evaluate the association of elevated BNP levels with major events.
Methods: In a first step, we determined gender-specific distribution and 90th percentiles of BNP in participants who were at baseline free from known determinants of increased BNP, i.e. cardiovascular disease, hypertension or chronic kidney disease. Consecutively, the association of BNP levels above these 90th percentiles with subsequent cardiovascular and coronary events was assessed in the entire cohort.
Results: In the BNP-normal sub-sample (n = 1,639), we defined gender-specific 90th percentile of BNP (31.3 pg/ml for men, 45.5 pg/ml for women). From overall 3,697 subjects (mean age 59.4, 52.4 % female), 194 subjects developed a major cardiovascular event and 122 myocardial infarction during a mean follow-up period of 8.0 ± 1.5 years. The 90th percentiles derived from the normal subset as threshold showed strong associations with major events in the entire cohort even after adjusting for traditional risk factors: hazard ratio (95% CI): 1.86 (1.37; 2.53), p < 0.0001 for cardiovascular, and 1.77 (1.19; 2.62), p = 0.005 for coronary events.
Conclusion: The gender-specific 90th percentile of BNP (31 pg/ml for males and 45 pg/ml for females) obtained from a BNP-normal sub-sample is associated with incident major cardiovascular and coronary events, suggesting that even BNP values lower than 100 pg/ml could imply prognostic information in the general population.