Get access

Standardization of Plasma Brain Natriuretic Peptide Concentrations in Older Japanese—Relationship to Latent Renal Dysfunction and Ischemic Heart Disease


Address correspondence to Hideki Nomura, MD, PhD, Department of Internal Medicine, Chubu Hospital, 36–3 Gengo Morioka-cho, Obu, Aichi 474–8511, Japan. E-mail:


OBJECTIVES: To determine the contributors to elevating plasma brain natriuretic peptide (BNP) concentrations in older people with normal systolic function. To investigate the relationship between cyclic guanosine monophosphate (cGMP) and BNP in older people with and without ischemic heart disease (IHD).

DESIGN: Observational study.

SETTING: Hospitalized patients in Nagoya University Hospital from November 1997 to May 2000.

PARTICIPANTS: Younger patients (<65) without IHD (n = 31), older patients (≥65) without IHD (n = 37), and older patients with stable IHD (n = 32). All participants showed 45% or greater of their left ventricular ejection fraction (LVEF).

MEASUREMENTS: LVEF, peak atrial velocity/peak early velocity (A/E) ratio at the mitral valve, and left ventricular mass volume were measured using transthoracic echocardiogram. Plasma BNP level, cGMP, and serum creatinine (Scr) were measured. Creatinine clearance (CLcr) was calculated based on 24-hour urine collection.

RESULTS: Plasma BNP levels in older people with and without IHD were significantly greater than in younger patients (mean ± standard deviation = 76.4 ± 96.0 (P < .001), 165.2 ± 200.6 (P < .001), and 8.1 ± 7.0, respectively). By simple regression analysis, in the groups without IHD, the logarithm of plasma BNP (Log BNP) concentrations had a significant positive relationship with age (R = 0.657, P < .001), Scr (R = 0.449, P < .001), and A/E ratio (R = 0.326, P = .003) and a significant negative relationship with CLcr (R = −0.663, P < .001). A stepwise multiple regression analysis with Log BNP level as the dependent variable and age, Scr, CLcr, and A/E ratio as independent variables showed that CLcr was a significant independent contributor in groups without IHD (R = −0.766, P < .001). In this analysis, the regression coefficient of the intercept was 2.006, and that of CLcr was −0.010. The cGMP/BNP ratio in older subjects with stable IHD tended to be lower than in those without IHD (P = .063).

CONCLUSIONS: Elevated BNP levels in older patients with normal systolic function may be in part due to latent renal dysfunction, despite normal Scr levels. In healthy older people, it is important to exclude the effects of latent renal function in assessing cardiac function according to BNP level. In older subjects with stable IHD, the cGMP/BNP ratio tended to be lower than in those without IHD. This may be a reflection of a poor response of cGMP to BNP.