Plasma Brain Natriuretic Peptide Levels Increase in Proportion to Changes in Pulmonary Artery Pressure in Old‐old Persons

To the Editor: Brain natriuretic peptide (BNP) is well known to be an important biochemical marker of heart disease.1–4 Clinically, we often encounter elderly persons who are apparently healthy despite the presence of high plasma BNP concentrations,4 but, because most of the elderly have heart disease and are treated with drugs, it is uncertain whether the plasma BNP concentration increases because of aging. In the present study, we investigated the potential for an increase of BNP in healthy elderly persons who were receiving no drugs by repeating the measurements of plasma BNP and parameters on echocardiography. Of 211 individuals aged 80 and older who were measured for BNP in community health-screening programs at our hospital from August through December 1998, 25 were evaluated to be healthy. Of these, 20 had a normal echocardiographic study; 14 of these individuals were evaluated for maximum pressure gradient in tricuspid regurgitation (Tr max PG) and enrolled in the study. Healthy individuals were defined as those with no abnormalities on routine screening examinations diagnosed by a general physician, not taking medications, and having no history of serious disease. Measurements of plasma BNP and echocardiography were repeated after 36 months. All 14 subjects (3 men and 11 women; aged 80–88, average age 82.2) gave informed consent to participate in the study. Echocardiography was conducted within 2 weeks of plasma BNP measurements. Normality of the echocardiographic study was defined by normal left ventricular (LV) and right ventricular function and dimensions, normal left and right atrial dimensions, absence of valvular stenosis, absence of significant valvular insufficiency, absence of aortic root dilatation, and absence of pericardial disease. This definition was similar to that of McQuillan et al.5 In addition, four items were evaluated: interventricular septal thickness as an index of LV hypertrophy, LV ejection fraction as an index of capacity for LV contraction, atrial-to–early peak transmitral velocity ratio as an index of capacity for LV dilation, and Tr max PG as an index of the right heart. All 14 subjects had survived for 36 months and none had developed a cardiovascular disease. They had not received any drugs for a chronic disease such as hypertension. Characteristics at baseline and at the 36-month follow-up are shown in Table 1. Plasma BNP concentration and Tr max PG increased significantly during the follow-up period. Other variables did not increase significantly. Results of linear regression analyses of four echocardiographic variables showed a strong correlation of baseline plasma BNP levels with baseline Tr max PG (r = 0.83, P < .001). Similarly, follow-up plasma BNP levels were closely associated with follow-up Tr max PG (r = 0.85, P < .001). In addition, changes in plasma BNP correlated significantly with changes in Tr max PG (r = 0.57, P = .0348). Other variables were not significantly associated with plasma BNP levels. Of the four echocardiographic variables, only Tr max PG was closely associated with plasma BNP levels. Pulmonary artery systolic pressure (PASP) is inferred from the simplified Bernoulli formula:6,7 Pulmonary artery pressure and PASP are known to be associated with age.5–7 These findings suggest that, in the healthy elderly, plasma BNP levels and PASP may increase for the same reason. It is well known that PASP increases not only in the presence of heart disease, but also with pulmonary disease.3,8 Nagaya et al. observed increased plasma BNP concentrations in patients with pulmonary hypertension whose LV ejection fraction was normal and reported that plasma BNP concentration correlated with the ejection fraction and end-diastolic pressure of the right ventricle.8 Lang et al. reported increased plasma BNP concentrations in patients with chronic obstructive pulmonary disease.9 These reports suggest that plasma BNP concentration is associated with pulmonary diseases as well as PASP when LV function is normal. A homogeneous enlargement of the alveolar air spaces without fibrosis or destruction of their walls characterizes senile lungs.10 It was inferred that the increase of PASP in our subjects was indicative of asymptomatic pulmonary disease or a change in the respiratory system due to aging. Plasma BNP concentrations might reflect such a change. In conclusion, our data suggest that plasma BNP levels increase in proportion to changes of PASP that are extrapolated by measurement of tricuspid regurgitation in the elderly.