Background: Little is known about the prognostic importance of murmur in unselected patients. It is difficult to distinguish between innocent and significant murmurs. N-terminal pro-brain natriuretic peptide (NT-pro-BNP) and BNP have recently been shown to be useful in small series of patients with valvular heart disease. We wanted to test whether murmur predicts mortality in unselected patients admitted to the hospital and whether NT-pro-BNP is capable of distinguishing between innocent and significant murmurs.
Methods: Consecutive patients (n = 2977) older than 40 years admitted to a local hospital were studied. Auscultation, echocardiography were performed and levels of natriuretic peptides were measured.
Results: A total of 21.8% of the 2977 patients had a murmur. After adjusting for sex and age there was a significant difference in the one-year mortality of patients with and without murmur (OR = 1.57, 95% CI = 1.27-1.94). NT-pro-BNP gave additional prognostic information for both patients with and without murmurs. Presence of a murmur was an independent predictor of 1-year mortality (OR = 1.36, 95% CI = 1.03-1.80) in a multivariate analysis. In patients with a murmur but normal NT-pro-BNP, discovery of valvular heart disease by echocardiography yielded no additional prognostic information.
Conclusions: Detection of a cardiac murmur during routine medical examination of hospitalized patients is associated with increased risk of death within a year. A blood test for NT-pro-BNP gives significant additional prognostic information of a murmur and could obviate the need for echocardiography in selected patients with a murmur and normal NT-pro-BNP for whom surgery is not feasible.