Commentary

Results of the study by Meuller et al are in agreement with the many other studies (now >40) that show that natriuretic peptides assist in the diagnosis of CHF. The study also compared the diagnostic accuracy of 2 forms of BNP—the biologically active hormone and NT-proBNP (the inactive peptide cleaved from the prohormone after release by cardiac myocytes). This study shows that they are equivalent for diagnosing CHF, although another paper by the same authors found that BNP was more accurate in patients with left ventricular ejection fraction (LVEF) <35% and that NT-proBNP was more accurate in patients with LVEF <60%.¹

The diagnostic accuracy of BNP was measured against a reference standard that combined Framingham criteria with echocardiographic evidence of systolic or diastolic dysfunction. It is questionable whether this is the best reference standard for CHF. It may be that BNP is a better method than current measures for identifying patients with cardiac dysfunction and for predicting prognosis.²

Clinicians would have most use for a test that can differentiate respiratory from cardiac causes in patients with dyspnoea. As this and other studies have shown, BNP can assist in diagnosing CHF, but clinicians should remember that respiratory diseases, such as pulmonary embolism and chronic obstructive pulmonary disease, also cause BNP levels to rise because of right ventricular strain. Current treatment also needs to be considered, because treatment with agents such as angiotensin converting enzyme inhibitors can normalise BNP levels.

Finally, no clear agreement exists on the cutpoints for BNP levels that should be used to define CHF. BNP values are higher in healthy women and in the elderly, although this may be caused in part by undetected diastolic heart failure in these populations. It is not yet clear at what values functional status declines and the risk of cardiovascular events begin to rise.