- Email: [email protected]
Natriuretic peptides in patients with aortic stenosis
Valvular and Congenital Heart Disease
Natriuretic peptides in patients with aortic stenosis Wei Qi, MD,a Per Mathisen, MD,b John Kjekshus, MD, PhD,b Svein Simonsen, MD, PhD,b Reidar Bjørnerheim, MD, PhD,b Knut Endresen, MD, PhD,b and Christian Hall, MD, PhDa Oslo, Norway
Background Whereas atrial natriuretic peptide (ANP) is secreted mainly from cardiac atria, brain natriuretic peptide (BNP) is produced to a larger extent in ventricles. Their relative importance as markers of cardiac function and myocardial hypertrophy is not yet clarified. This study evaluated circulating BNP and ANP and the N-terminal part of their propeptides (NT-proBNP and NT-proANP) as markers of left ventricular hypertrophy and atrial pressure increase in patients with aortic stenosis.
Methods The plasma concentrations of BNP, NT-proBNP, ANP, and NT-proANP were measured by radioimmunoassay in 67 patients with aortic stenosis. Peptide plasma concentrations were related to measurements obtained by cardiac catheterization and echocardiography.
Results Receiver operating characteristic curves indicated that BNP and NT-proBNP performed best in the detection of increased left ventricular mass and NT-proANP in the detection of increased left atrial pressure. NT-proBNP was significantly increased in mild left ventricular hypertrophy (left ventricular mass index, 78 to 139 g/m2), whereas NT-proANP was not increased until left ventricular mass index was 141 to 180 g/m2. Conclusions Plasma BNP and NT-proBNP may serve as early markers of left ventricular hypertrophy, whereas ANP and NT-proANP reflect left atrial pressure increase. The repeated and combined measurements of natriuretic peptides might provide diagnostic information relevant to the evaluation of the stage of aortic stenosis. (Am Heart J 2001; 142:725-32.)
Brain natriuretic peptide (BNP) and atrial natriuretic peptide (ANP) constitute a cardiac hormone system mediating natriuresis, diuresis, and vasodilation.1-4 The circulating levels of these peptides are elevated in states of increased cardiac wall stress. In this respect, they may serve as markers of cardiac function and hypertrophy.5,6 Although BNP has a remarkable sequence homology to ANP, the intracardiac distribution of BNP is different from ANP, its production being capable of upregulation not only in atrial but to a larger extent also in ventricular tissue. The relative importance of BNP and ANP with regard to their utility as cardiac markers is currently under discussion. On secretion, ANP prohormone is cleaved into ANP and N-terminal proANP (NT-proANP).7 Because of its longer half-life and in vitro stability, NT-proANP has
From aResearch Institute for Internal Medicine and bMedical Department B, the National Hospital, the University of Oslo, Norway. Supported by Norwegian State Educational Fund and Medinnova SF, Norway. Submitted November 7, 2000; accepted April 16, 2001. Reprint requests: Christian Hall, Research Institute for Internal Medicine, The National Hospital, The University of Oslo, N-0027 Oslo, Norway. E-mail: [email protected] Copyright © 2001 by Mosby, Inc. 0002-8703/2001/$35.00 + 0 4/1/117131 doi:10.1067/mhj.2001.117131
been shown to be a more reliable marker than ANP.8,9 In 1995, Hunt et al10 first reported the measurement of a large molecular N-terminal form of BNP prohormone (1–108) thought to represent N-terminal proBNP (1–76). In comparison to ANP and NT-proANP, the relation between BNP and N-terminal proBNP (NT-proBNP) remains less studied. In aortic stenosis (AS), the left ventricle adapts to increased afterload by myocardial hypertrophy. In the early course of the disease, the enhanced “booster pump” function of the left atrium raises left ventricular end-diastolic pressure without producing a concomitant elevation of mean atrial pressure.11 Later, however, the left ventricle may decompensate with a reduction in the ratio between wall thickness and left ventricular cavity size, atrial and pulmonary wedge capillary pressures rise, and cardiac output decreases.11 The current study was designed to evaluate the relation of A-type (ANP and NT-proANP) and B-type natriuretic peptides (BNP and NT-proBNP) to hemodynamic changes and left ventricular hypertrophy in patients with AS. Because of the differences in intracardiac distribution of the peptides, we hypothesized that they might differ in their abilities to detect left ventricular hypertrophy and to discriminate between hypertrophy and left atrial distention.
American Heart Journal October 2001
726 Qi et al
Table I. Characteristics of patients with aortic stenosis Values Age (y) 70 ± 1 Sex (M/F) 28/39 New York Heart Association classification I 4 II 32 III 28 IV 3 Systolic blood pressure (mm Hg) 148 ± 23 Diastolic blood pressure (mm Hg) 86 ± 2 Serum creatinine (µmol/L) 91 ± 2 LVSF (%) 35 ± 2 LAD (cm) 4.0 ± 0.1 LVMI (g/m2) 189 ± 8 AVAI (cm2/m2) 0.33 ± 0.01 Mean aortic valve gradient (mm Hg) 56 ± 2 PCWP (mm Hg) 12 ± 1 RAP (mm Hg) 4 ± 0.3 LVEDP (mm Hg) 17 ± 1 CI (L/min per m2) 2.7 ± 0.1 BNP (pmol/L) 80 ± 14 NT-proBNP (pmol/L) 237 ± 34 ANP (pmol/L) 48 ± 6 NT-proANP (pmol/L) 1519 ± 122
(Range)
Table II. Comparison between patients with PCWP ≤12 mm Hg and >12 mm Hg
(36–82)
(89–220) (50–120) (67–136) (5–58) (2.2–6.1) (78–336) (0.12–0.66) (19–111) (3–38) (–1 to 14) (4–34) (1–5.7) (3–547) (24–1664) (9–245) (479–5020)
Values are mean ± SE.
Methods Patients
Age (y) LVSF (%) LVMI (g/m2) AVAI (cm2/m2) Mean aortic valve gradient (mm Hg) PCWP (mm Hg) RAP (mm Hg) LVEDP (mm Hg) CI (L/min per m2) BNP (pmol/L) NT-proBNP (pmol/L) ANP (pmol/L) NT-proANP (pmol/L)
PCWP ≤12 mm Hg (n = 41)
PCWP >12 mm Hg (n = 26)
69 ± 1 37 ± 2 173 ± 10 0.34 ± 0.02 56 ± 3
72 ± 2 29 ± 2† 213 ± 13* 0.31 ± 0.02 55 ± 3
8 ± 0.4 3 ± 0.3 15 ± 1 2.8 ± 0.1 49 ± 12 152 ± 24 31 ± 5 952 ± 80
19 ± 1.1‡ 5 ± 0.6‡ 20 ± 2† 2.4 ± 0.1* 130 ± 29* 371 ± 73† 74 ± 11‡ 2156 ± 221‡
Values are mean ± SE. *P < .05. †P ≤ .01. ‡P ≤ .001.
Left ventricular and atrial measurements were obtained from M-mode recordings, which included interventricular septal thickness at end diastole (IVSTd), posterior wall thickness at end diastole (PWTd), left ventricular internal dimension at end systole (LVDs) and at end diastole (LVDd), and left atrial diameter (LAD). Left ventricular mass (LVM) was calculated by the equation of Devereux and Reichek12: LVM (g) = 1.04 ([LVDd + IVSTd + PWTd]3 – LVDd3) – 14 and normalized to body surface area ( LVM index, LVMI). LVSF was calculated as (LVDd – LVDs)/LVDd. The pressure gradient across the aortic valve was estimated by the simplified Bernoulli equation from flow velocity detected by continuous wave Doppler integrated throughout systole. Aortic valve area was calculated by the continuity equation according to Ihlen and Molstad13 and divided by body surface area to give the aortic valve area index (AVAI). Mitral and aortic regurgitation was evaluated by Doppler color-flow mapping.
From June 1995 to September 1996, patients evaluated for aortic valve surgery for AS were consecutively enrolled into the study as they were undergoing diagnostic cardiac catheterization. Only patients with a confirmed diagnosis of AS were included. The exclusion criteria were severe (grade III) aortic (disturbing aortic valve gradient calculation) or mitral (atrial enlargement influencing secretion of natriuretic peptides) regurgitation or serum creatinine >150 µmol/L. Sixty-seven patients (28 men, 39 women) with a mean age of 70 years (range 36 to 82 years) constituted the study population (Table I). For subgroup analysis, the patients were divided into 2 groups: those with pulmonary capillary wedge pressure ≤12 mm Hg and those with pulmonary capillary wedge pressure >12 mm Hg (Table II). Among all patients, 53 had mild to moderate aortic regurgitation, 46 had mild to moderate mitral regurgitation (grade I to grade II), and 35 had significant (≥50%) coronary artery stenosis. Thirty-two percent (18 of 56) of the patients had evidence of left ventricular systolic dysfunction (left ventricular shortening fraction [LVSF] <29%). The patients were treated with the following cardiovascular drugs: digitalis (n = 13), diuretics (n = 28), angiotensin-converting enzyme inhibitors (n = 7), and β-blockers (n = 10). Informed consent was obtained from each patient. The study protocol was approved by the regional ethics committee.
Cardiac catheterization was performed to measure pulmonary capillary wedge pressure (PCWP), right atrial pressure (RAP), systemic arterial pressure, and left ventricular end-diastolic pressure (LVEDP), by using the fourth intercostal space in the anterior axillary line as zero reference level. Cardiac output, normalized to body surface area (cardiac index, CI), was determined according to the direct Fick principle, calculating O2 consumption from expired gas volume and O2 concentration. Coronary angiography was performed with the use of a standard technique.
Echocardiography
Measurement of natriuretic peptides
Echocardiography was performed with a Vingmed CFM 800, Vingmed Sound (Vingmed Sound AS, Horten, Norway).
During catheterization, arterial blood (10 mL) was sampled and transferred into prechilled EDTA Vacutainers for
Cardiac catheterization
American Heart Journal Volume 142, Number 4
analysis of BNP, NT-proBNP, ANP, and NT-proANP. Immediately after sampling, the tubes were placed on ice and centrifuged at 4°C before plasma aliquots were frozen at –70°C for later analysis. The plasma BNP concentration was measured by an IRMA kit for human BNP (Shionoria BNP, Shionogi & Co, Ltd, Osaka, Japan). The detection limit was 1.2 pmol/L and the between- and within-assay coefficients of variation were 9.3% and 3.3%, respectively. The plasma NT-proBNP concentration was measured by an in-house–developed radioimmunoassay directly in plasma with polyclonal antiserum raised in a rabbit immunized with N-terminal proBNP (1–21) (Peninsula Laboratories Inc, Belmont, Calif) as antigen. The detection limit was 9.7 pmol/L, and the between- and within-assay coefficients of variation were 9.0% and 7.3%, respectively. The plasma ANP (irANP 99–126) concentration was measured in plasma by an IRMA kit for α-human ANP (Shionoria ANP, Shionogi & Co, Ltd). The detection limit for ANP was 1.40 pmol/L, and the between- and within-assay coefficients of variation were 6% and 4.8%, respectively. The plasma NT-proANP (irANP 1–98) concentration was measured in unextracted plasma according to Sundsfjord et al.7 The detection limit for NT-proANP was 185 pmol/L, and the between- and within-assay coefficients of variation were 4.1% and 6.3%, respectively. The recovery of synthetic peptide added to plasma from healthy subjects was as follows: BNP: 107% (n =7), NTproBNP: 82% (n = 10), ANP: 75% (n = 9), and NT-ANP: 85% (n = 10). For control, the plasma concentrations of BNP, NTproBNP, ANP, and NT-proANP were measured in 19 volunteers with a mean age of 66 years (range 54 to 83 years; 9 men and 10 women). These control subjects were recruited from an outpatient practice and judged to be free from cardiopulmonary, renal, and infectious disease according to questionnaire and clinical examination. In these subjects, peptide levels were BNP, 7.0 ± 1.2 pmol/L; NT-proBNP, 41 ± 3 pmol/L; ANP, 7.6 ± 0.9 pmol/L; and NT-proANP, 701 ± 105 pmol/L.
Statistical analysis Continuous variables were presented by mean ± standard error (SE). The natriuretic peptide data were transformed by natural logarithms to fit the normal distribution (except for Figure 1). A 2-sample t test was used to examine the differences between groups and 1-sample t test to examine the differences from 0. Pearson correlation and univariate linear regression analyses were performed to examine the relation between continuous variables. Correlation between natriuretic peptides was analyzed by the Spearman rank method. To compare ability of peptide levels to predict the severity of aortic stenosis, natriuretic peptides were incorporated in a stepwise multivariate regression analysis. The area under the receiver operating characteristic (ROC) curve, which represents an index of overall test accuracy, was assessed by the program GraphROC (version 2.0, Turku, Finland). A probability value of <.05 was considered statistically significant. All analysis unless otherwise noted was performed with SPSS statistical analysis package (version 8.0, SPSS, Chicago, Ill).
Qi et al 727
Figure 1
Scatterplots showing relation between plasma concentrations of BNP, ANP, and their respective N-terminal propeptides, NTproBNP and NT-proANP, in 67 patients with AS.
Results Characteristics of patients As shown in Table I, the patients had obstruction to the left ventricular outflow with an AVAI of 0.33 ± 0.01 cm2/m2 (0.12 to 0.66 cm2/m2) and a mean aortic valve gradient of 56 ± 2 mm Hg (19 to 111 mm Hg). The two subgroups of patients with PCWP below and above 12 mm Hg did not differ with regard to age, AVAI, and aortic valve gradient. Patients with elevated PCWP had significantly higher LVMI and LVEDP and lower LVSF and CI (Table II).
Plasma levels of natriuretic peptides The plasma levels of BNP, NT-proBNP, ANP, and NTproANP in the total study population are presented in Table I. A close correlation was found between BNP and NT-proBNP (r = 0.93, P < .001) and ANP and NTproANP (r = 0.81, P < .001) (Figure 1). All peptides were elevated compared with control subjects in the
American Heart Journal October 2001
728 Qi et al
Figure 2
Plasma concentrations of NT-proBNP and NT-proANP in control subjects and in patients with AS, without and with elevated left atrial pressure, represented by PCWP ≤12 and >12 mm Hg. *P < .05 vs control subjects, ‡P < .001 vs control subjects.
group of patients with PCWP ≤12 mm Hg and more so in the group of patients with PCWP >12 mm Hg (Table II). In the control group, the molar concentration of BNP was comparable to that of ANP, whereas BNP was higher than ANP in patients with AS regardless of PCWP elevation. The relation between B-type and Atype natriuretic peptides is further illustrated by the Nterminal propeptides, NT-proBNP and NT-proANP (Figure 2). In the group of patients with PCWP ≤12 mm Hg, the molar concentration of NT-proBNP was increased 3.7-fold (P < .001) compared with control subjects, whereas NT-proANP was increased only 1.4-fold (P < .05). In the group of patients with PCWP >12 mm Hg, NT-proBNP was increased 9.1-fold (P < .001), and NTproANP was increased 3.1-fold (P < .001) compared with control subjects. A similar pattern was evident when patients were divided according to ventricular function (LVSF > 29%: NT-proBNP increase, 3.3-fold [P < .001]; NT-proANP increase, 1.7-fold [P < .005]; LVSF <29%: NTproBNP increase, 11.2-fold [P < .001]; NT-proANP increase, 3.3-fold [P < .001]). Figure 3 illustrates the relative increase of both N-terminal peptides with increasing LVMI. When the patients were grouped according to LVMI quartiles, NT-proBNP was significantly increased relative to control subjects already in the lowest LVMI quartile (78 to 139 g/m2), and it increased further as LVMI increased. In comparison, NT-proANP was increased over control values only from the second quartile (141 to 180 g/m2) and above. When the patients were subgrouped according to the presence of coronary
artery disease, none of the peptides were significantly different across groups.
Correlation of plasma levels of natriuretic peptides with variables of cardiac function and structure When all patients were considered together, all 4 peptides correlated with AVAI, LVMI, LVEDP, LVSF, PCWP, RAP, and LAD (Table III). Of the peptides, only BNP and NT-proBNP were correlated, although weakly, to mean aortic valve gradient. In the group of patients without elevated PCWP, BNP and NT-proBNP (each r = 0.47, P < .01) but not ANP or NT-proANP correlated with LVMI (Figure 4, BNP and ANP data not shown). Furthermore, BNP, NT-proBNP, and NT-proANP (r = –0.62, –0.60, and –0.54, respectively, each P ≤ .001) but not ANP correlated to AVAI (Figure 4). All 4 peptides correlated to mean aortic valve gradient (BNP and NT-proBNP: r = 0.58 and 0.52, P < .001; ANP and NT-proANP: r = 0.42 and 0.39, P < .01). In contrast, in the group of patients with elevated PCWP, none of the peptides correlated to either AVAI or mean aortic valve gradient.
Comparison of ability of peptide levels to predict severity of AS In multivariate analysis, the ability of peptides to predict LVMI, AVAI, and mean aortic valve gradient was examined. When all peptides except NT-proBNP were entered into the model, only BNP remained independently predictive of these parameters (P < .001). The
American Heart Journal Volume 142, Number 4
Qi et al 729
Figure 3
Plot showing relation of NT-proBNP and NT-proANP to LVMI. *P < .05 vs control subjects.
same analysis that included NT-proBNP but not BNP indicated the same finding for NT-proBNP (P < .001). In neither case was there any significant contribution from ANP and NT-proANP to the prediction. The areas under ROC curves for these peptides in the detection of an increased LVMI (>170 g/m2) and elevated PCWP (>12 mm Hg) are presented in Figure 5. BNP and NT-proBNP were the strongest markers of LVMI, whereas NT-proANP performed best in the diagnosis of PCWP elevation.
Discussion In the current study of patients with AS, we found that the concentrations of natriuretic peptides were increased compared with control subjects of the same age range. The increases were related to reduction in AVAI and increases in LVM, LVEDP, and left atrial pressure (represented by PCWP). A stronger association was found for BNP and NT-proBNP with AVAI and LVMI, whereas NTproANP was most closely associated with left atrial pressure. In the subgroup of patients without elevated left atrial pressure, a preferential increase of BNP and NTproBNP in the circulation reflected the presence of left ventricular hypertrophy. Our data showed that NT-proBNP was closely correlated with BNP in plasma, suggesting that the two peptides might be cosecreted into circulation. Probably because of a lower clearance from circulation, NT-
proBNP level in plasma was 3-fold higher than BNP, a relation resembling that of NT-proANP and ANP. In the current study, the association of NT-proBNP with left ventricular parameters was comparable with that of BNP. In AS, the elevated afterload induces left ventricular systolic wall stress, acting as a stimulus for myocardial cell hypertrophy characterized by the reexpression of fetal isogenes, including increased expression of natriuretic peptides in the ventricle.14,15 During the development of cardiac hypertrophy, a gradual disappearance of natriuretic peptide clearance receptor (NPR-C) mRNA was found in the rat heart.16 Thus downregulation of NPR-C might be an additional mechanism contributing to the observed increase of ANP and BNP plasma levels in patients with AS. The relation observed between the concentrations of natriuretic peptides and LVMI was closer for BNP and NT-proBNP than for ANP and NTproANP. This result corroborates the studies by Kohno et al17 and Yamamoto et al,18 who found that BNP and ANP plasma concentration was higher in hypertensive patients with left ventricular hypertrophy than in those without and that BNP was superior to ANP and NTproANP as a marker of ventricular hypertrophy. The natriuretic peptide increase in AS may result not only from left ventricular hypertrophy but also from increased left atrial pressure caused by left ventricular systolic dysfunction. In previous studies, no attempts were made to dissociate between these mechanisms.17-19 The observation that fewer than
American Heart Journal October 2001
730 Qi et al
Figure 4
Relations of NT-proBNP and NT-proANP (log scale) to AVAI and LVMI in patients with normal PCWP.
Table III. Correlation coefficient between plasma levels of natriuretic peptides and variables of left ventricle and atria
PCWP RAP LAD LVEDP LVMI Mean aortic valve gradient AVAI LVSF
BNP
NTproBNP
ANP
NTproANP
0.58‡ 0.31* 0.34† 0.47† 0.60‡ 0.32†
0.62‡ 0.42‡ 0.40† 0.45† 0.57‡ 0.25*
0.65‡ 0.36† 0.42‡ 0.44† 0.44‡ 0.15
0.76‡ 0.40‡ 0.48‡ 0.46† 0.52‡ 0.09
–0.56‡ –0.51‡
–0.52‡ –0.57‡
–0.33* –0.41†
–0.47‡ –0.49‡
PCWP (n = 67); RAP (n = 67); LAD (n = 56); LVEDP (n = 42); LVMI (n = 56); mean aortic valve gradient (n = 66); AVAI (n = 53); LVSF (n = 56). *P < .05. †P ≤ .01. ‡P ≤ .001.
50% of our patients were receiving diuretic therapy illustrates that hemodynamic derangement as well as clinical symptoms were variable and not always severe. The association of BNP and NT-proBNP to LVMI in patients with normal left ventricular systolic function and left atrial pressure suggests that in this setting, BNP secretion is primarily related to the process of left ventricular hypertrophy. ANP concentration was increased only when LVMI was elevated to a larger degree, possibly as a result of concomitant increase in atrial wall stretch. Corroborating this, the ROC analyses showed that BNP and NT-proBNP were most accurate in the detection of increased LVMI, whereas NT-proANP performed best in the detection of increased left atrial pressure. In the natural history of aortic stenosis, a latent period may exist in which left ventricular hypertrophy compensates for the increase in afterload with-
American Heart Journal Volume 142, Number 4
Qi et al 731
Figure 5
ROC curves for plasma concentrations of natriuretic peptides in predicting LVMI >170 g/m2 and PCWP >12 mm Hg. *Significantly different compared with BNP; ‡significantly different compared with NT-proANP.
out preload increase. During progression, decompensation may occur and atrial pressures rise. Thus an elevation of B-type natriuretic peptide levels in the circulation, with little or no increase in A-type natriuretic peptides, may indicate a relatively early stage of AS, whereas the combined elevation of both natriuretic peptides may indicate a more advanced stage. Prasad et al20 found a correlation between BNP, ANP, and mean aortic valve gradient in 30 patients with AS. However, no data on cardiac preload were reported. In our total study group of patients, BNP and NT-proBNP were only weakly related to aortic valve gradient, whereas it was considerably stronger in the subgroup without elevated left atrial pressure. The aortic valve gradient, being dependent on cardiac output, is not linearly related to the aortic valve area. Late in the course of AS, patients may have left ventricular pump failure with increased left atrial pressure, whereas the aortic valve gradient does not further increase or even decreases. Thus markedly increased natriuretic peptides with no or moderate aortic valve gradient elevation might indicate advanced disease with a decompensated cardiac condition. The clinical prognosis of patients with untreated AS, with symptoms of heart failure and angina, is poor.21 Irrespective of symptoms, progressive left ventricular dysfunction and cardiomegaly are considered indications for valve replacement. Conceivably, monitoring AS patients by repeated measurements of natriuretic
peptides may help in determining the stage of disease and thus provide information useful in deciding the timing of surgical intervention. In summary, ventricular hypertrophy and atrial pressure increase contribute to the specific secretion of natriuretic peptides. We suggest that the elevation of plasma levels of BNP and NT-proBNP, in isolation, may reflect early hypertrophic changes of the left ventricle, whereas further elevation of these two peptides combined with increase in NT-proANP indicates decompensation of cardiac function at a later stage of the disease. Thus repeated and combined measurements of natriuretic peptides may provide diagnostic information with regard to evaluation of the stage of aortic stenosis. We thank Ellen Lund Sagen and Hanne Schulz Jensen for technical assistance and the doctors and nurses at the catheterization laboratory for cooperation.
References 1. Ballermann BJ, Brenner BM. George E. Brown memorial lecture: role of atrial peptides in body fluid homeostasis. Circ Res 1986; 58:619-30. 2. Sudoh T, Kangawa K, Minamino N. A new natriuretic peptide in porcine brain. Nature 1988;332:78-81. 3. de Bold AJ, Borenstein HB, Veress AT. A rapid and potent natriuretic response to intravenous injection of atrial myocardial extract in rats. Life Sci 1981;28:89-94. 4. Saito Y, Nakao K, Itoh H. Brain natriuretic peptide is a novel cardiac hormone. Biochem Biophys Res Commun 1989;158:360-8.
732 Qi et al
5. Burnett JC Jr, Kao PC, Hu DC. Atrial natriuretic peptide elevation in congestive heart failure in the human. Science 1986;231:1145-7. 6. Yoshimura M, Yasue H, Okumura K. Different secretion patterns of atrial natriuretic peptide and brain natriuretic peptide in patients with congestive heart failure. Circulation 1993;87:464-9. 7. Sundsfjord JA, Thibault G, Larochelle P. Identification and plasma concentrations of the N-terminal fragment of proatrial natriuretic factor in man. J Clin Endocrinol Metab 1988;66:605-10. 8. Hall C, Aaberge L, Stokke O. In vitro stability of N-terminal proatrial natriuretic factor in unfrozen samples: an important prerequisite for its use as a biochemical parameter of atrial pressure in clinical routine [letter]. Circulation 1995;91:911. 9. Hall C, Rouleau JL, Moye L. N-terminal proatrial natriuretic factor: an independent predictor of long-term prognosis after myocardial infarction. Circulation 1994;89:1934-42. 10. Hunt PJ, Yandle TG, Nicholls MG. The amino-terminal portion of pro-brain natriuretic peptide (Pro-BNP) circulates in human plasma. Biochem Biophys Res Commun 1995;214:1175-83. 11. Braunwald E. Aortic stenosis. In: Heart disease: a textbook of cardiovascular medicine. Philadelphia: WB Saunders Co; 1997. p. 1035-45. 12. Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation 1977;55:613-8. 13. Ihlen H, Molstad P. Cardiac output measured by Doppler echocardiography in patients with aortic prosthetic valves. Eur Heart J 1990;11:399-402.
American Heart Journal October 2001
14. Sadoshima J, Jahn L, Takahashi T. Molecular characterization of the stretch-induced adaptation of cultured cardiac cells: an in vitro model of load-induced cardiac hypertrophy. J Biol Chem 1992; 267:10551-60. 15. Cameron VA, Aitken GD, Ellmers LJ. The sites of gene expression of atrial, brain, and C-type natriuretic peptides in mouse fetal development: temporal changes in embryos and placenta. Endocrinology 1996;137:817-24. 16. Brown LA, Nunez DJ, Wilkins MR. Differential regulation of natriuretic peptide receptor messenger RNAs during the development of cardiac hypertrophy in the rat. J Clin Invest 1993;92:270212. 17. Kohno M, Horio T, Yokokawa K. Brain natriuretic peptide as a cardiac hormone in essential hypertension. Am J Med 1992;92: 29-34. 18. Yamamoto K, Burnett JC Jr, Jougasaki M. Superiority of brain natriuretic peptide as a hormonal marker of ventricular systolic and diastolic dysfunction and ventricular hypertrophy. Hypertension 1996; 28:988-94. 19. Kohno M, Horio T, Yokokawa K. Brain natriuretic peptide as a marker for hypertensive left ventricular hypertrophy: changes during 1-year antihypertensive therapy with angiotensin-converting enzyme inhibitor. Am J Med 1995;98:257-65. 20. Prasad N, Bridges AB, Lang CC. Brain natriuretic peptide concentrations in patients with aortic stenosis. Am Heart J 1997;133:477-9. 21. Frank S, Johnson A, Ross J Jr. Natural history of valvular aortic stenosis. Br Heart J 1973;35:41-6.
Natriuretic peptides in patients with aortic stenosis Wei Qi, MD,a Per Mathisen, MD,b John Kjekshus, MD, PhD,b Svein Simonsen, MD, PhD,b Reidar Bjørnerheim, MD, PhD,b Knut Endresen, MD, PhD,b and Christian Hall, MD, PhDa Oslo, Norway
Background Whereas atrial natriuretic peptide (ANP) is secreted mainly from cardiac atria, brain natriuretic peptide (BNP) is produced to a larger extent in ventricles. Their relative importance as markers of cardiac function and myocardial hypertrophy is not yet clarified. This study evaluated circulating BNP and ANP and the N-terminal part of their propeptides (NT-proBNP and NT-proANP) as markers of left ventricular hypertrophy and atrial pressure increase in patients with aortic stenosis.
Methods The plasma concentrations of BNP, NT-proBNP, ANP, and NT-proANP were measured by radioimmunoassay in 67 patients with aortic stenosis. Peptide plasma concentrations were related to measurements obtained by cardiac catheterization and echocardiography.
Results Receiver operating characteristic curves indicated that BNP and NT-proBNP performed best in the detection of increased left ventricular mass and NT-proANP in the detection of increased left atrial pressure. NT-proBNP was significantly increased in mild left ventricular hypertrophy (left ventricular mass index, 78 to 139 g/m2), whereas NT-proANP was not increased until left ventricular mass index was 141 to 180 g/m2. Conclusions Plasma BNP and NT-proBNP may serve as early markers of left ventricular hypertrophy, whereas ANP and NT-proANP reflect left atrial pressure increase. The repeated and combined measurements of natriuretic peptides might provide diagnostic information relevant to the evaluation of the stage of aortic stenosis. (Am Heart J 2001; 142:725-32.)
Brain natriuretic peptide (BNP) and atrial natriuretic peptide (ANP) constitute a cardiac hormone system mediating natriuresis, diuresis, and vasodilation.1-4 The circulating levels of these peptides are elevated in states of increased cardiac wall stress. In this respect, they may serve as markers of cardiac function and hypertrophy.5,6 Although BNP has a remarkable sequence homology to ANP, the intracardiac distribution of BNP is different from ANP, its production being capable of upregulation not only in atrial but to a larger extent also in ventricular tissue. The relative importance of BNP and ANP with regard to their utility as cardiac markers is currently under discussion. On secretion, ANP prohormone is cleaved into ANP and N-terminal proANP (NT-proANP).7 Because of its longer half-life and in vitro stability, NT-proANP has
From aResearch Institute for Internal Medicine and bMedical Department B, the National Hospital, the University of Oslo, Norway. Supported by Norwegian State Educational Fund and Medinnova SF, Norway. Submitted November 7, 2000; accepted April 16, 2001. Reprint requests: Christian Hall, Research Institute for Internal Medicine, The National Hospital, The University of Oslo, N-0027 Oslo, Norway. E-mail: [email protected] Copyright © 2001 by Mosby, Inc. 0002-8703/2001/$35.00 + 0 4/1/117131 doi:10.1067/mhj.2001.117131
been shown to be a more reliable marker than ANP.8,9 In 1995, Hunt et al10 first reported the measurement of a large molecular N-terminal form of BNP prohormone (1–108) thought to represent N-terminal proBNP (1–76). In comparison to ANP and NT-proANP, the relation between BNP and N-terminal proBNP (NT-proBNP) remains less studied. In aortic stenosis (AS), the left ventricle adapts to increased afterload by myocardial hypertrophy. In the early course of the disease, the enhanced “booster pump” function of the left atrium raises left ventricular end-diastolic pressure without producing a concomitant elevation of mean atrial pressure.11 Later, however, the left ventricle may decompensate with a reduction in the ratio between wall thickness and left ventricular cavity size, atrial and pulmonary wedge capillary pressures rise, and cardiac output decreases.11 The current study was designed to evaluate the relation of A-type (ANP and NT-proANP) and B-type natriuretic peptides (BNP and NT-proBNP) to hemodynamic changes and left ventricular hypertrophy in patients with AS. Because of the differences in intracardiac distribution of the peptides, we hypothesized that they might differ in their abilities to detect left ventricular hypertrophy and to discriminate between hypertrophy and left atrial distention.
American Heart Journal October 2001
726 Qi et al
Table I. Characteristics of patients with aortic stenosis Values Age (y) 70 ± 1 Sex (M/F) 28/39 New York Heart Association classification I 4 II 32 III 28 IV 3 Systolic blood pressure (mm Hg) 148 ± 23 Diastolic blood pressure (mm Hg) 86 ± 2 Serum creatinine (µmol/L) 91 ± 2 LVSF (%) 35 ± 2 LAD (cm) 4.0 ± 0.1 LVMI (g/m2) 189 ± 8 AVAI (cm2/m2) 0.33 ± 0.01 Mean aortic valve gradient (mm Hg) 56 ± 2 PCWP (mm Hg) 12 ± 1 RAP (mm Hg) 4 ± 0.3 LVEDP (mm Hg) 17 ± 1 CI (L/min per m2) 2.7 ± 0.1 BNP (pmol/L) 80 ± 14 NT-proBNP (pmol/L) 237 ± 34 ANP (pmol/L) 48 ± 6 NT-proANP (pmol/L) 1519 ± 122
(Range)
Table II. Comparison between patients with PCWP ≤12 mm Hg and >12 mm Hg
(36–82)
(89–220) (50–120) (67–136) (5–58) (2.2–6.1) (78–336) (0.12–0.66) (19–111) (3–38) (–1 to 14) (4–34) (1–5.7) (3–547) (24–1664) (9–245) (479–5020)
Values are mean ± SE.
Methods Patients
Age (y) LVSF (%) LVMI (g/m2) AVAI (cm2/m2) Mean aortic valve gradient (mm Hg) PCWP (mm Hg) RAP (mm Hg) LVEDP (mm Hg) CI (L/min per m2) BNP (pmol/L) NT-proBNP (pmol/L) ANP (pmol/L) NT-proANP (pmol/L)
PCWP ≤12 mm Hg (n = 41)
PCWP >12 mm Hg (n = 26)
69 ± 1 37 ± 2 173 ± 10 0.34 ± 0.02 56 ± 3
72 ± 2 29 ± 2† 213 ± 13* 0.31 ± 0.02 55 ± 3
8 ± 0.4 3 ± 0.3 15 ± 1 2.8 ± 0.1 49 ± 12 152 ± 24 31 ± 5 952 ± 80
19 ± 1.1‡ 5 ± 0.6‡ 20 ± 2† 2.4 ± 0.1* 130 ± 29* 371 ± 73† 74 ± 11‡ 2156 ± 221‡
Values are mean ± SE. *P < .05. †P ≤ .01. ‡P ≤ .001.
Left ventricular and atrial measurements were obtained from M-mode recordings, which included interventricular septal thickness at end diastole (IVSTd), posterior wall thickness at end diastole (PWTd), left ventricular internal dimension at end systole (LVDs) and at end diastole (LVDd), and left atrial diameter (LAD). Left ventricular mass (LVM) was calculated by the equation of Devereux and Reichek12: LVM (g) = 1.04 ([LVDd + IVSTd + PWTd]3 – LVDd3) – 14 and normalized to body surface area ( LVM index, LVMI). LVSF was calculated as (LVDd – LVDs)/LVDd. The pressure gradient across the aortic valve was estimated by the simplified Bernoulli equation from flow velocity detected by continuous wave Doppler integrated throughout systole. Aortic valve area was calculated by the continuity equation according to Ihlen and Molstad13 and divided by body surface area to give the aortic valve area index (AVAI). Mitral and aortic regurgitation was evaluated by Doppler color-flow mapping.
From June 1995 to September 1996, patients evaluated for aortic valve surgery for AS were consecutively enrolled into the study as they were undergoing diagnostic cardiac catheterization. Only patients with a confirmed diagnosis of AS were included. The exclusion criteria were severe (grade III) aortic (disturbing aortic valve gradient calculation) or mitral (atrial enlargement influencing secretion of natriuretic peptides) regurgitation or serum creatinine >150 µmol/L. Sixty-seven patients (28 men, 39 women) with a mean age of 70 years (range 36 to 82 years) constituted the study population (Table I). For subgroup analysis, the patients were divided into 2 groups: those with pulmonary capillary wedge pressure ≤12 mm Hg and those with pulmonary capillary wedge pressure >12 mm Hg (Table II). Among all patients, 53 had mild to moderate aortic regurgitation, 46 had mild to moderate mitral regurgitation (grade I to grade II), and 35 had significant (≥50%) coronary artery stenosis. Thirty-two percent (18 of 56) of the patients had evidence of left ventricular systolic dysfunction (left ventricular shortening fraction [LVSF] <29%). The patients were treated with the following cardiovascular drugs: digitalis (n = 13), diuretics (n = 28), angiotensin-converting enzyme inhibitors (n = 7), and β-blockers (n = 10). Informed consent was obtained from each patient. The study protocol was approved by the regional ethics committee.
Cardiac catheterization was performed to measure pulmonary capillary wedge pressure (PCWP), right atrial pressure (RAP), systemic arterial pressure, and left ventricular end-diastolic pressure (LVEDP), by using the fourth intercostal space in the anterior axillary line as zero reference level. Cardiac output, normalized to body surface area (cardiac index, CI), was determined according to the direct Fick principle, calculating O2 consumption from expired gas volume and O2 concentration. Coronary angiography was performed with the use of a standard technique.
Echocardiography
Measurement of natriuretic peptides
Echocardiography was performed with a Vingmed CFM 800, Vingmed Sound (Vingmed Sound AS, Horten, Norway).
During catheterization, arterial blood (10 mL) was sampled and transferred into prechilled EDTA Vacutainers for
Cardiac catheterization
American Heart Journal Volume 142, Number 4
analysis of BNP, NT-proBNP, ANP, and NT-proANP. Immediately after sampling, the tubes were placed on ice and centrifuged at 4°C before plasma aliquots were frozen at –70°C for later analysis. The plasma BNP concentration was measured by an IRMA kit for human BNP (Shionoria BNP, Shionogi & Co, Ltd, Osaka, Japan). The detection limit was 1.2 pmol/L and the between- and within-assay coefficients of variation were 9.3% and 3.3%, respectively. The plasma NT-proBNP concentration was measured by an in-house–developed radioimmunoassay directly in plasma with polyclonal antiserum raised in a rabbit immunized with N-terminal proBNP (1–21) (Peninsula Laboratories Inc, Belmont, Calif) as antigen. The detection limit was 9.7 pmol/L, and the between- and within-assay coefficients of variation were 9.0% and 7.3%, respectively. The plasma ANP (irANP 99–126) concentration was measured in plasma by an IRMA kit for α-human ANP (Shionoria ANP, Shionogi & Co, Ltd). The detection limit for ANP was 1.40 pmol/L, and the between- and within-assay coefficients of variation were 6% and 4.8%, respectively. The plasma NT-proANP (irANP 1–98) concentration was measured in unextracted plasma according to Sundsfjord et al.7 The detection limit for NT-proANP was 185 pmol/L, and the between- and within-assay coefficients of variation were 4.1% and 6.3%, respectively. The recovery of synthetic peptide added to plasma from healthy subjects was as follows: BNP: 107% (n =7), NTproBNP: 82% (n = 10), ANP: 75% (n = 9), and NT-ANP: 85% (n = 10). For control, the plasma concentrations of BNP, NTproBNP, ANP, and NT-proANP were measured in 19 volunteers with a mean age of 66 years (range 54 to 83 years; 9 men and 10 women). These control subjects were recruited from an outpatient practice and judged to be free from cardiopulmonary, renal, and infectious disease according to questionnaire and clinical examination. In these subjects, peptide levels were BNP, 7.0 ± 1.2 pmol/L; NT-proBNP, 41 ± 3 pmol/L; ANP, 7.6 ± 0.9 pmol/L; and NT-proANP, 701 ± 105 pmol/L.
Statistical analysis Continuous variables were presented by mean ± standard error (SE). The natriuretic peptide data were transformed by natural logarithms to fit the normal distribution (except for Figure 1). A 2-sample t test was used to examine the differences between groups and 1-sample t test to examine the differences from 0. Pearson correlation and univariate linear regression analyses were performed to examine the relation between continuous variables. Correlation between natriuretic peptides was analyzed by the Spearman rank method. To compare ability of peptide levels to predict the severity of aortic stenosis, natriuretic peptides were incorporated in a stepwise multivariate regression analysis. The area under the receiver operating characteristic (ROC) curve, which represents an index of overall test accuracy, was assessed by the program GraphROC (version 2.0, Turku, Finland). A probability value of <.05 was considered statistically significant. All analysis unless otherwise noted was performed with SPSS statistical analysis package (version 8.0, SPSS, Chicago, Ill).
Qi et al 727
Figure 1
Scatterplots showing relation between plasma concentrations of BNP, ANP, and their respective N-terminal propeptides, NTproBNP and NT-proANP, in 67 patients with AS.
Results Characteristics of patients As shown in Table I, the patients had obstruction to the left ventricular outflow with an AVAI of 0.33 ± 0.01 cm2/m2 (0.12 to 0.66 cm2/m2) and a mean aortic valve gradient of 56 ± 2 mm Hg (19 to 111 mm Hg). The two subgroups of patients with PCWP below and above 12 mm Hg did not differ with regard to age, AVAI, and aortic valve gradient. Patients with elevated PCWP had significantly higher LVMI and LVEDP and lower LVSF and CI (Table II).
Plasma levels of natriuretic peptides The plasma levels of BNP, NT-proBNP, ANP, and NTproANP in the total study population are presented in Table I. A close correlation was found between BNP and NT-proBNP (r = 0.93, P < .001) and ANP and NTproANP (r = 0.81, P < .001) (Figure 1). All peptides were elevated compared with control subjects in the
American Heart Journal October 2001
728 Qi et al
Figure 2
Plasma concentrations of NT-proBNP and NT-proANP in control subjects and in patients with AS, without and with elevated left atrial pressure, represented by PCWP ≤12 and >12 mm Hg. *P < .05 vs control subjects, ‡P < .001 vs control subjects.
group of patients with PCWP ≤12 mm Hg and more so in the group of patients with PCWP >12 mm Hg (Table II). In the control group, the molar concentration of BNP was comparable to that of ANP, whereas BNP was higher than ANP in patients with AS regardless of PCWP elevation. The relation between B-type and Atype natriuretic peptides is further illustrated by the Nterminal propeptides, NT-proBNP and NT-proANP (Figure 2). In the group of patients with PCWP ≤12 mm Hg, the molar concentration of NT-proBNP was increased 3.7-fold (P < .001) compared with control subjects, whereas NT-proANP was increased only 1.4-fold (P < .05). In the group of patients with PCWP >12 mm Hg, NT-proBNP was increased 9.1-fold (P < .001), and NTproANP was increased 3.1-fold (P < .001) compared with control subjects. A similar pattern was evident when patients were divided according to ventricular function (LVSF > 29%: NT-proBNP increase, 3.3-fold [P < .001]; NT-proANP increase, 1.7-fold [P < .005]; LVSF <29%: NTproBNP increase, 11.2-fold [P < .001]; NT-proANP increase, 3.3-fold [P < .001]). Figure 3 illustrates the relative increase of both N-terminal peptides with increasing LVMI. When the patients were grouped according to LVMI quartiles, NT-proBNP was significantly increased relative to control subjects already in the lowest LVMI quartile (78 to 139 g/m2), and it increased further as LVMI increased. In comparison, NT-proANP was increased over control values only from the second quartile (141 to 180 g/m2) and above. When the patients were subgrouped according to the presence of coronary
artery disease, none of the peptides were significantly different across groups.
Correlation of plasma levels of natriuretic peptides with variables of cardiac function and structure When all patients were considered together, all 4 peptides correlated with AVAI, LVMI, LVEDP, LVSF, PCWP, RAP, and LAD (Table III). Of the peptides, only BNP and NT-proBNP were correlated, although weakly, to mean aortic valve gradient. In the group of patients without elevated PCWP, BNP and NT-proBNP (each r = 0.47, P < .01) but not ANP or NT-proANP correlated with LVMI (Figure 4, BNP and ANP data not shown). Furthermore, BNP, NT-proBNP, and NT-proANP (r = –0.62, –0.60, and –0.54, respectively, each P ≤ .001) but not ANP correlated to AVAI (Figure 4). All 4 peptides correlated to mean aortic valve gradient (BNP and NT-proBNP: r = 0.58 and 0.52, P < .001; ANP and NT-proANP: r = 0.42 and 0.39, P < .01). In contrast, in the group of patients with elevated PCWP, none of the peptides correlated to either AVAI or mean aortic valve gradient.
Comparison of ability of peptide levels to predict severity of AS In multivariate analysis, the ability of peptides to predict LVMI, AVAI, and mean aortic valve gradient was examined. When all peptides except NT-proBNP were entered into the model, only BNP remained independently predictive of these parameters (P < .001). The
American Heart Journal Volume 142, Number 4
Qi et al 729
Figure 3
Plot showing relation of NT-proBNP and NT-proANP to LVMI. *P < .05 vs control subjects.
same analysis that included NT-proBNP but not BNP indicated the same finding for NT-proBNP (P < .001). In neither case was there any significant contribution from ANP and NT-proANP to the prediction. The areas under ROC curves for these peptides in the detection of an increased LVMI (>170 g/m2) and elevated PCWP (>12 mm Hg) are presented in Figure 5. BNP and NT-proBNP were the strongest markers of LVMI, whereas NT-proANP performed best in the diagnosis of PCWP elevation.
Discussion In the current study of patients with AS, we found that the concentrations of natriuretic peptides were increased compared with control subjects of the same age range. The increases were related to reduction in AVAI and increases in LVM, LVEDP, and left atrial pressure (represented by PCWP). A stronger association was found for BNP and NT-proBNP with AVAI and LVMI, whereas NTproANP was most closely associated with left atrial pressure. In the subgroup of patients without elevated left atrial pressure, a preferential increase of BNP and NTproBNP in the circulation reflected the presence of left ventricular hypertrophy. Our data showed that NT-proBNP was closely correlated with BNP in plasma, suggesting that the two peptides might be cosecreted into circulation. Probably because of a lower clearance from circulation, NT-
proBNP level in plasma was 3-fold higher than BNP, a relation resembling that of NT-proANP and ANP. In the current study, the association of NT-proBNP with left ventricular parameters was comparable with that of BNP. In AS, the elevated afterload induces left ventricular systolic wall stress, acting as a stimulus for myocardial cell hypertrophy characterized by the reexpression of fetal isogenes, including increased expression of natriuretic peptides in the ventricle.14,15 During the development of cardiac hypertrophy, a gradual disappearance of natriuretic peptide clearance receptor (NPR-C) mRNA was found in the rat heart.16 Thus downregulation of NPR-C might be an additional mechanism contributing to the observed increase of ANP and BNP plasma levels in patients with AS. The relation observed between the concentrations of natriuretic peptides and LVMI was closer for BNP and NT-proBNP than for ANP and NTproANP. This result corroborates the studies by Kohno et al17 and Yamamoto et al,18 who found that BNP and ANP plasma concentration was higher in hypertensive patients with left ventricular hypertrophy than in those without and that BNP was superior to ANP and NTproANP as a marker of ventricular hypertrophy. The natriuretic peptide increase in AS may result not only from left ventricular hypertrophy but also from increased left atrial pressure caused by left ventricular systolic dysfunction. In previous studies, no attempts were made to dissociate between these mechanisms.17-19 The observation that fewer than
American Heart Journal October 2001
730 Qi et al
Figure 4
Relations of NT-proBNP and NT-proANP (log scale) to AVAI and LVMI in patients with normal PCWP.
Table III. Correlation coefficient between plasma levels of natriuretic peptides and variables of left ventricle and atria
PCWP RAP LAD LVEDP LVMI Mean aortic valve gradient AVAI LVSF
BNP
NTproBNP
ANP
NTproANP
0.58‡ 0.31* 0.34† 0.47† 0.60‡ 0.32†
0.62‡ 0.42‡ 0.40† 0.45† 0.57‡ 0.25*
0.65‡ 0.36† 0.42‡ 0.44† 0.44‡ 0.15
0.76‡ 0.40‡ 0.48‡ 0.46† 0.52‡ 0.09
–0.56‡ –0.51‡
–0.52‡ –0.57‡
–0.33* –0.41†
–0.47‡ –0.49‡
PCWP (n = 67); RAP (n = 67); LAD (n = 56); LVEDP (n = 42); LVMI (n = 56); mean aortic valve gradient (n = 66); AVAI (n = 53); LVSF (n = 56). *P < .05. †P ≤ .01. ‡P ≤ .001.
50% of our patients were receiving diuretic therapy illustrates that hemodynamic derangement as well as clinical symptoms were variable and not always severe. The association of BNP and NT-proBNP to LVMI in patients with normal left ventricular systolic function and left atrial pressure suggests that in this setting, BNP secretion is primarily related to the process of left ventricular hypertrophy. ANP concentration was increased only when LVMI was elevated to a larger degree, possibly as a result of concomitant increase in atrial wall stretch. Corroborating this, the ROC analyses showed that BNP and NT-proBNP were most accurate in the detection of increased LVMI, whereas NT-proANP performed best in the detection of increased left atrial pressure. In the natural history of aortic stenosis, a latent period may exist in which left ventricular hypertrophy compensates for the increase in afterload with-
American Heart Journal Volume 142, Number 4
Qi et al 731
Figure 5
ROC curves for plasma concentrations of natriuretic peptides in predicting LVMI >170 g/m2 and PCWP >12 mm Hg. *Significantly different compared with BNP; ‡significantly different compared with NT-proANP.
out preload increase. During progression, decompensation may occur and atrial pressures rise. Thus an elevation of B-type natriuretic peptide levels in the circulation, with little or no increase in A-type natriuretic peptides, may indicate a relatively early stage of AS, whereas the combined elevation of both natriuretic peptides may indicate a more advanced stage. Prasad et al20 found a correlation between BNP, ANP, and mean aortic valve gradient in 30 patients with AS. However, no data on cardiac preload were reported. In our total study group of patients, BNP and NT-proBNP were only weakly related to aortic valve gradient, whereas it was considerably stronger in the subgroup without elevated left atrial pressure. The aortic valve gradient, being dependent on cardiac output, is not linearly related to the aortic valve area. Late in the course of AS, patients may have left ventricular pump failure with increased left atrial pressure, whereas the aortic valve gradient does not further increase or even decreases. Thus markedly increased natriuretic peptides with no or moderate aortic valve gradient elevation might indicate advanced disease with a decompensated cardiac condition. The clinical prognosis of patients with untreated AS, with symptoms of heart failure and angina, is poor.21 Irrespective of symptoms, progressive left ventricular dysfunction and cardiomegaly are considered indications for valve replacement. Conceivably, monitoring AS patients by repeated measurements of natriuretic
peptides may help in determining the stage of disease and thus provide information useful in deciding the timing of surgical intervention. In summary, ventricular hypertrophy and atrial pressure increase contribute to the specific secretion of natriuretic peptides. We suggest that the elevation of plasma levels of BNP and NT-proBNP, in isolation, may reflect early hypertrophic changes of the left ventricle, whereas further elevation of these two peptides combined with increase in NT-proANP indicates decompensation of cardiac function at a later stage of the disease. Thus repeated and combined measurements of natriuretic peptides may provide diagnostic information with regard to evaluation of the stage of aortic stenosis. We thank Ellen Lund Sagen and Hanne Schulz Jensen for technical assistance and the doctors and nurses at the catheterization laboratory for cooperation.
References 1. Ballermann BJ, Brenner BM. George E. Brown memorial lecture: role of atrial peptides in body fluid homeostasis. Circ Res 1986; 58:619-30. 2. Sudoh T, Kangawa K, Minamino N. A new natriuretic peptide in porcine brain. Nature 1988;332:78-81. 3. de Bold AJ, Borenstein HB, Veress AT. A rapid and potent natriuretic response to intravenous injection of atrial myocardial extract in rats. Life Sci 1981;28:89-94. 4. Saito Y, Nakao K, Itoh H. Brain natriuretic peptide is a novel cardiac hormone. Biochem Biophys Res Commun 1989;158:360-8.
732 Qi et al
5. Burnett JC Jr, Kao PC, Hu DC. Atrial natriuretic peptide elevation in congestive heart failure in the human. Science 1986;231:1145-7. 6. Yoshimura M, Yasue H, Okumura K. Different secretion patterns of atrial natriuretic peptide and brain natriuretic peptide in patients with congestive heart failure. Circulation 1993;87:464-9. 7. Sundsfjord JA, Thibault G, Larochelle P. Identification and plasma concentrations of the N-terminal fragment of proatrial natriuretic factor in man. J Clin Endocrinol Metab 1988;66:605-10. 8. Hall C, Aaberge L, Stokke O. In vitro stability of N-terminal proatrial natriuretic factor in unfrozen samples: an important prerequisite for its use as a biochemical parameter of atrial pressure in clinical routine [letter]. Circulation 1995;91:911. 9. Hall C, Rouleau JL, Moye L. N-terminal proatrial natriuretic factor: an independent predictor of long-term prognosis after myocardial infarction. Circulation 1994;89:1934-42. 10. Hunt PJ, Yandle TG, Nicholls MG. The amino-terminal portion of pro-brain natriuretic peptide (Pro-BNP) circulates in human plasma. Biochem Biophys Res Commun 1995;214:1175-83. 11. Braunwald E. Aortic stenosis. In: Heart disease: a textbook of cardiovascular medicine. Philadelphia: WB Saunders Co; 1997. p. 1035-45. 12. Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation 1977;55:613-8. 13. Ihlen H, Molstad P. Cardiac output measured by Doppler echocardiography in patients with aortic prosthetic valves. Eur Heart J 1990;11:399-402.
American Heart Journal October 2001
14. Sadoshima J, Jahn L, Takahashi T. Molecular characterization of the stretch-induced adaptation of cultured cardiac cells: an in vitro model of load-induced cardiac hypertrophy. J Biol Chem 1992; 267:10551-60. 15. Cameron VA, Aitken GD, Ellmers LJ. The sites of gene expression of atrial, brain, and C-type natriuretic peptides in mouse fetal development: temporal changes in embryos and placenta. Endocrinology 1996;137:817-24. 16. Brown LA, Nunez DJ, Wilkins MR. Differential regulation of natriuretic peptide receptor messenger RNAs during the development of cardiac hypertrophy in the rat. J Clin Invest 1993;92:270212. 17. Kohno M, Horio T, Yokokawa K. Brain natriuretic peptide as a cardiac hormone in essential hypertension. Am J Med 1992;92: 29-34. 18. Yamamoto K, Burnett JC Jr, Jougasaki M. Superiority of brain natriuretic peptide as a hormonal marker of ventricular systolic and diastolic dysfunction and ventricular hypertrophy. Hypertension 1996; 28:988-94. 19. Kohno M, Horio T, Yokokawa K. Brain natriuretic peptide as a marker for hypertensive left ventricular hypertrophy: changes during 1-year antihypertensive therapy with angiotensin-converting enzyme inhibitor. Am J Med 1995;98:257-65. 20. Prasad N, Bridges AB, Lang CC. Brain natriuretic peptide concentrations in patients with aortic stenosis. Am Heart J 1997;133:477-9. 21. Frank S, Johnson A, Ross J Jr. Natural history of valvular aortic stenosis. Br Heart J 1973;35:41-6.