BNP and Diastolic Dysfunction

Abstract & Commentary

Synopsis: In ambulatory hypertensive patients with heart failure symptoms but normal cystolic LV function, BNP is higher in those with diastolic dysfunction. However, it is usually in the normal range and thus, it is of little diagnostic value.

Source: Mottram PM, et al. Am J Cardiol. 2003;92:1424-1438.

B-type natriuretic peptide (BNP) levels are gaining in popularity as a screening test for heart failure in patients with suggestive symptoms. Although there are reports that BNP is elevated in systolic as well as diastolic dysfunction, its diagnostic use for the latter is unclear. Thus, Mottram and associates prospectively studied 72 ambulatory patients with hypertension under treatment and exertional dyspnea. Patients were excluded if their symptoms warranted hospitalization; if ischemic heart disease was present by history or resting echo; if respiratory disease was present; if significant valvular disease was detected; or if their left ventricular ejection fraction was < 50%. BNP values were classified as normal or abnormal based upon the upper limit of normal adjusted for age and sex. Left ventricular (LV) diastolic function was classified as either normal, impaired relaxation, pseudonormal, or restrictive using standard Echo-Doppler criteria. In addition, tissue Doppler was used to measure LV strain and strain rate. Also, Echo and blood pressure measurements were used to calculate LV mass and wall stress. By the standard Doppler criteria, about half of the patients had isolated diastolic dysfunction, most had impaired relaxation, and a few had pseudonormal filling. Those with normal diastolic function were younger, but there was no difference in LV mass or blood pressure between the 2 groups. BNP levels were higher in those with diastolic dysfunction (46 vs. 20 pg/mL; P = .004) and were independently positively associated with systolic blood pressure, strain rate, left atrial area, systolic wall stress and age, and negatively with diastolic blood pressure and mitral annular late diastolic velocity.

BNP levels were in the normal range in 97% of the patients with normal diastolic function but also in 79% of the patients with abnormal diastolic function. Mottram et al concluded that in ambulatory hypertensive patients under treatment with symptoms of heart failure and normal systolic LV function, BNP is elevated with diastolic dysfunction but is usually within the normal range and, therefore, is of little diagnostic value.

Comment by Michael H. Crawford, MD

BNP is the new troponin. Everyone in the emergency department and most hospitalized patients now get them done in the hopes of a positive value, which ensures a transfer to the cardiology service. Thus, it is welcome news that most ambulatory patients with hypertension, dyspnea, and a mild LV relaxation abnormality have a normal value. However, more than 60% of the few patients with more advanced diastolic dysfunction did have elevated values. These results are consistent with other studies done in hospitalized patients, which showed a high incidence of elevated BNP in diastolic heart failure patients. Somewhat disturbing was the observation confirmed in this study that BNP is related to the blood pressure level and perhaps the pulse pressure. Thus, uncontrolled hypertension may cause elevations and provoke inappropriate hospital admissions.

The major weakness of this study is that it included few patients with advanced diastolic dysfunction. It is not surprising that in a group of ambulatory patients, not considered sick enough for admission, with symptoms only on exertion, that Echo-Doppler estimated filling pressures were normal, as was BNP. A major strength of this study was the sophisticated Echo-Doppler parameters measured; this was a state of the art study. Unfortunately, there were few associations between any of these sophisticated measures and BNP. By multivariate analysis, the only sophisticated parameter of diastolic function related to BNP was mitral annular late diastolic velocity, which is a measure of the vigor of atrial function and LV stiffness. BNP was independently related to more mundane measures such as blood pressure, age, left atrial size, and systolic wall stress. In summary, it would appear that the measure of BNP levels should be restricted to patients with marked symptoms, requiring hospitalization, where an elevated value will point to a cardiovascular cause. Thus, there may be little reason for an emergency department physician to order a BNP level. It can be done by the admitting physician in those requiring hospitalization.

Dr. Crawford, Professor of Medicine, Associate Chief of Cardiology for Clinical Programs University of California San Francisco, is Editor of Clinical Cardiology Alert.