Coronary artery disease (CAD) is the leading cause of death in adults, accounting for more than 480,000 deaths in the US and 1.95 million deaths in Europe each year. There is considerable interest in early diagnosis of CAD, because obstructive coronary artery disease can exist with minimal or no symptoms and can progress rapidly. In patients with symptoms suggestive of CAD, a variety of tests and biomarkers are available to assess the risk of future cardiovascular events and to identify patients for further diagnostic work-up with coronary angiography. B-type natriuretic peptide (BNP) and the inactive N-terminal fragment of its prohormone (NT-proBNP) provide prognostic information in the general population and in a variety of cardiac diseases, including heart failure and acute coronary syndromes (ACS).
The prognostic value of NT-proBNP in patients with stable coronary artery disease has been clearly established. Elevated levels of natriuretic peptides are associated with an increased risk of death and myocardial infarction (MI). The clinical utility of natriuretic peptide testing for the diagnosis of coronary disease is a matter of on-going research. A relationship of NT-proBNP with the severity of coronary stenosis and with myocardial ischaemia has been demonstrated in several reports. Establishing a causal link between natriuretic peptides and specific pathophysiologic conditions is a challenging task, as natriuretic peptide levels may be elevated in various clinical conditions, in particular in those associated with acute or chronic impairment of left ventricular function. Accordingly, BNP and NT-proBNP are established markers of prognosis and disease progression in patients with heart failure. Interpretation of elevated natriuretic peptide levels for diagnosis of coronary disease must always take into account left ventricular ejection fraction (LVEF).
In a recent study the author demonstrated that NT-proBNP measured at rest provides diagnostic information for non-invasive prediction of CAD in 781 patients with normal systolic left ventricular function referred for coronary angiography. Elevations of NT-proBNP levels were correlated with the extent of CAD. NT-proBNP levels were increased not only among patients with significant coronary stenosis, but also among those with non-obstructive CAD compared with patients with angiographically normal coronary arteries. This finding might partly be explained by transient myocardial ischaemia. Ischaemia promotes the expression and release of NT-proBNP in patients with CAD and no evidence of heart failure. Additional mechanisms may also be important, such as upregulation of natriuretic peptides in the setting of subclinical cardiovascular disease. Natriuretic peptides have a fundamental role in vascular function and remodelling, by potentiating and complementing the effects of nitric oxide (NO), inhibiting oxidised low-density lipoprotein (LDL)-induced migration of smooth muscle cells in the vascular wall, and increasing parasympathetic tone.
Community-based studies have demonstrated that normal values for BNP and NT-proBNP are age-and sex-specific. The author has have extended these results to patients with suspected CAD. NT-proBNP levels are higher in women irrespective of the presence or absence of CAD, compared with men with similar angiographic diagnosis. The correlation of NT-proBNP and age may be caused by multiple factors. Alteration in the degradation, clearance or endogenous production of BNP may occur with ageing. Furthermore, natriuretic peptide levels may be influenced by renal function, although a pathophysiologic mechanism has not been established. Cardiac arrhythmias may cause increase in natriuretic peptide levels. In individuals with atrial fibrillation (AF) and with normal ventricular function, BNP and atrial natriuretic peptide are increased compared with controls. NT-proBNP secretion in AF may be due to atrial enlargement, or the fibrillatory activity per se. However, increased gene expression of NT-proBNP has only been found in persistent, but not in paroxysmal, AF.
Screening for pre-clinical, asymptomatic heart disease by using moderately elevated BNP levels is an interesting possibility, because treatment of heart disease in advanced stages is costly and less effective. NT-proBNP is a marker of long-term mortality in patients with stable angina pectoris providing prognostic information above and beyond that provided by conventional risk markers and left ventricular systolic dysfunction. In a large community-based study, during a follow-up of five years, elevated levels of BNP were associated with an increased risk of death, heart failure, AF and stroke, but not with the risk of CADevents. Therefore, BNP and NT-proBNP might be general markers of cardiovascular disease, which could be useful to identify individuals with asymptomatic disease. Recently, normal values of natriuretic peptides have been reported, which could be used for screening purposes. However, cut-off values that are appropriate for screening for left ventricular systolic dysfunction are not sensitive enough to reliably identify patients with CAD. Therefore, lower cut-off values will have to be defined to screen for cardiovascular disease including asymptomatic CAD. Although NT-proBNP is a strong independent predictor of CAD, its value as a stand-alone screening test for diagnosis of angiographically significant CAD is limited by low specificity for CAD. A recent study using NT-proBNP at a cut-point of 125pg/ml for both men and women found poor test characteristics for NT-proBNP as a screening test for CAD. Using a cut-point of 85pg/ml for men and 165pg/ml for women, the accuracy to predict significant coronary stenosis could be improved to 65%, which is similar to the diagnostic accuracy of a standard exercise stress test.
However, the diagnostic information provided by natriuretic peptide levels is incremental to that provided by clinical information and stress testing. Using a multivariate model integrating clinical data, stress testing and sex-specific NT-proBNP measurements, patients could be correctly classified with an accuracy of 87%.
For optimised non-invasive assessment of CAD risk and estimation of prognosis, an integrated approach using risk factor assessment, patient history and stress testing should be applied in combination with natriuretic peptide testing. The author has demonstrated that a diagnostic score using BNP, exercise test, and cardiovascular risk factors can improve non-invasive prediction or exclusion of CAD in patients routinely referred for coronary angiography in daily practice (see Table 1). Especially, the addition of resting BNP levels can significantly improve the discriminatory ability of non-invasive evaluation of patients with suspected CAD and normal LVEF. In more than half of patients relevant CAD could be diagnosed or ruled out with high accuracy.
Natriuretic peptide levels are elevated in patients with coronary disease. Plasma levels of NT-proBNP are correlated with the severity of coronary stenosis. Elevated levels may even be found in asymptomatic patients with coronary plaques without significant stenosis. Age- and sex-specific normal ranges have to be applied for screening and diagnostic purposes. Due to low specificity, the utility of natriuretic peptides alone for non-invasive assessment of CAD probability is limited, and integrative approaches have to be applied. Natriuretic peptide measurements should be used in combination with patient history and risk factor assessment, echocardiographic evaluation of left ventricular function, and stress testing.