Author + information
- Received August 2, 2001
- Revision received February 13, 2002
- Accepted February 19, 2002
- Published online May 15, 2002.
- Tomoki Nakamura, MD*,* (, )
- Kenzo Sakamoto, MD†,
- Tetsuhiro Yamano, MD†,
- Masayuki Kikkawa, MD*,
- Kan Zen, MD†,
- Takato Hikosaka, MD†,
- Takao Kubota, MD*,
- Akihiro Azuma, MD, PhD† and
- Tsunehiko Nishimura, MD, PhD*
- ↵*Reprint requests and correspondence:
Dr. Tomoki Nakamura, Department of Radiology, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamigyo-Ku, Kyoto 602-8566, Japan.
Objectives We measured plasma atrial/brain natriuretic peptide (ANP/BNP) levels at rest and during exercise and correlated the results with various clinical findings, particularly with myocardial ischemia, in asymptomatic hypertrophic cardiomyopathy (HCM).
Background In patients with HCM, ANP and BNP levels are elevated and exercise-induced myocardial ischemia is common. However, it has not yet been elucidated how these levels at rest and their change with dynamic exercise are related to ischemia.
Methods Levels of ANP and BNP were measured at rest and at peak exercise during 99mTc-tetrofosmin scintigraphy in 31 asymptomatic patients with non-obstructive HCM and in 10 control subjects.
Results Levels of ANP and BNP at rest and the change of ANP and BNP levels (pg/ml) from rest to exercise were significantly greater in HCM than in control subjects (ANP: rest, 53.2 ± 31.8 vs. 11.6 ± 6.1; exercise, 114.5 ± 74.8 vs. 28.3 ± 23.4. BNP: rest, 156.7 ± 104.1 vs. 9.8 ± 9.6; exercise, 201.6 ± 131.5 vs. 13.2 ± 14.5). Septal perforator compression (SPC) and exercise-induced ischemia were observed, respectively, in 20 (64.5%) and in 19 (61.3%) patients with HCM. The increment of ANP during exercise was similar between HCM subgroups with or without inducible ischemia. However, BNP levels at rest and BNP increments during exercise were significantly greater in the HCM subgroup with inducible ischemia than in the subgroup without (rest, 190.5 ± 116.2 vs. 103.1 ± 48.3; exercise, 250.5 ± 142.2 vs. 124.2 ± 58.6). Multiple logistic regression analysis revealed that SPC and BNP levels at rest were independently associated with exercise-induced ischemia.
Conclusions Measurement of plasma BNP levels at rest may be useful in predicting silent myocardial ischemia in HCM.
Atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) play roles in fluid-electrolyte homeostasis (1,2), the main secretion sites being the atria and ventricles, respectively (2,3). Plasma ANP and BNP levels are elevated in proportion to the severity of left ventricular (LV) systolic and diastolic dysfunction and heart failure, and are useful in evaluating prognosis (4–7). Recent studies clarified that BNP levels were increased in hypertensive heart disease and aortic stenosis, suggesting that BNP elevation is related to LV hypertrophy (8,9). In addition, increased ANP and BNP levels during dynamic exercise are associated with myocardial ischemia in patients with angina pectoris (10).
The prognosis of asymptomatic patients with hypertrophic cardiomyopathy (HCM) in adults is relatively good, compared with that of symptomatic patients (11). However, exercise-induced myocardial ischemia has been demonstrated in more than 50% of patients with asymptomatic HCM (12), sometimes resulting in unfavorable events (13,14). Brain natriuretic peptide levels in patients with HCM, even if asymptomatic, were markedly elevated when compared with various cardiac diseases, although little is known concerning pathophysiologic significance of elevated BNP levels (15,16). In the present study, we measured ANP and BNP levels at rest and during exercise and correlated the results with various clinical parameters including myocardial ischemia.
We studied 31 patients with non-obstructive HCM who were free from any cardiac symptoms (24 men, 7 women; mean age 54 ± 10 years) and 10 age-matched control subjects (5 men, 5 women; mean age 54 ± 5 years). The diagnosis of HCM was based on the typical clinical, electrocardiographic and hemodynamic features, with echocardiographic demonstration of a nondilated, asymmetrically hypertrophied LV in the absence of other cardiac or systemic diseases that can produce LV hypertrophy (17). Patients with HCM were excluded from this study when they had organic coronary stenosis, valvular heart disease or systemic hypertension. In all the HCM patients studied, cardiac catheterization, echocardiography, and exercise 99mTc-tetrofosmin scintigraphy were performed. The control group consisted of asymptomatic subjects who were consulted for a survey of minimal electrocardiographic abnormality, but all of them demonstrated normal echocardiograms and 99mTc-tetrofosmin scintigrams. All the study subjects signed informed consent forms before the study, and the study protocol was approved by the research council of our institution.
Cardioactive medications (verapamil or diltiazem) were discontinued for at least five half-lives before the present study, in accordance with previous reports (18,19).
An Advanced Technology Laboratory HDI 5000 ultrasound system was used for echocardiographic analysis. After an overnight fast, patients were subjected to evaluation of wall thickness, cavity size and wall motion from multiple windows. Left ventricular diastolic function was assessed by E/A ratio and isovolumetric relaxation time (IRT).
99mTc-tetrofosmin scintigraphic study and measurement of plasma ANP and BNP levels
Soon after the echocardiographic study, 99mTc-tetrofosmin scintigraphy was performed by a multistage bicycle ergometer exercise test. At peak exercise, each subject intravenously received 370 MBq of 99mTc-tetrofosmin (Nihon Medi-Physics Co., Ltd, Nishinomiya, Japan) from the antecubital vein, and a blood sample was simultaneously drawn from the contralateral antecubital vein. Exercise was continued for an additional 1 min to allow adequate circulation of the isotope. The exercise test was terminated when there was ischemic ST segment depression ≥1 mm, excessive leg fatigue or achievement of 100% of the maximal predicted heart rate. Those who complained of chest pain or discomfort were excluded from this study. Four hours later, blood samples were collected and 740 MBq of 99mTc-tetrofosmin was intravenously injected after 30 min bed rest. Image acquisition at exercise and rest was each begun 30 min after the injection of a tracer. A digital gamma camera (PRISM 2000XP, Marconi Medical Systems, Inc., Cleveland, Ohio) equipped with a low-energy, high resolution parallel-hole collimator was used for single photon emission computed tomographic (SPECT) imaging.
The collected blood samples were transferred to disposable tubes containing aprotinin (500 kallikrein inactivator units/ml) and centrifuged at 3,000 rpm at 3°C for 10 min. Plasma ANP and BNP levels were measured with a specific immunoradiometric assay for alpha-human ANP and human BNP respectively using commercial kits (Shionoria kit, Shionogi and Co., Ltd., Japan). Interassay coefficients of variation for the ANP and BNP assays were 5.2% and 2.3%, respectively.
SPECT images were analyzed from two viewpoints by three experienced nuclear cardiologists who were unaware of angiographic and other clinical data: regional perfusion abnormality and apparent LV cavity dilation (LVCD). First, LV tomograms were divided into 13 segments. Short-axis slices at the basal and midventricular levels were each separated into six segments: anteroseptal, septal, posteroseptal, posterolateral, lateral and anterolateral. A vertical long-axis slice was used to evaluate the apical portion of one segment. The degree of the tracer uptake in these segments was semi-quantified using a four-point scoring system: 0 = normal, 1 = slightly decreased, 2 = moderately decreased, 3 = complete defect. Differences of opinion among the observers were resolved by consensus. Scores for each region were averaged, and a change of 1 or more from exercise to rest was defined as reversible regional perfusion abnormality (18). Second, apparent LVCD was quantified by the software we developed previously (20,21). Thirty-six radii were generated at 10-degree intervals from the center of the middle myocardial images of the midventricular short-axis view. The area surrounded by the 36 points of maximal count on each radius was calculated in the exercise and rest images. The exercise/rest ratio of the surface area was defined as the transient dilation index (TDI). Abnormal apparent LVCD was determined as >mean + 2 SD of the TDI in control subjects. The intraobserver and interobserver analyses were also examined for the reproducibility of TDI by two out of three nuclear cardiologists. Exercise-induced myocardial ischemia was defined as the presence of reversible regional perfusion abnormality and/or abnormal apparent LVCD.
Coronary angiography and transcatheter measurement of hemodynamics were simultaneously evaluated within two weeks of the scintigraphic study. Cineangiographic films were reviewed independently by two cardiologists. Subjects showing atherosclerotic coronary stenosis were excluded from this study. We also investigated the prevalence and location of the systolic compression of the septal branches. The degree of septal perforator compression (SPC) of the left anterior descending artery was further classified according to the report by Pichard et al. (22): grade 0 = no systolic compression, grade 1 = systolic disappearance of the distal third, grade 2 = disappearance of the distal two-thirds, grade 3 = disappearance of the entire septal perforator.
Data are expressed as mean ± SD. Differences between patients and control subjects or between the HCM subgroups were compared by the Mann-Whitney U rank-sum test for unpaired data or by the chi-square test for discrete variables. Two-way repeated measures analysis of variance was used to evaluate the increment of rate-pressure products and ANP and BNP levels during exercise between patients with HCM and control subjects or between the HCM subgroups. Linear regression analysis was used to determine the correlations of ANP and BNP levels at rest with echocardiographic and hemodynamic variables. A multiple logistic regression analysis was performed to detect the influencing factors on exercise-induced myocardial ischemia. The following variables were used as possible influencing factors: left atrial dimension, LV wall thickness and cavity size, systolic and diastolic functional indexes, the increase of systolic blood pressure (ΔsBP), SPC, electrocardiographic ST segment depression and ANP and BNP levels at rest. Statistical significance was defined as p < 0.05.
Clinical features and plasma ANP and BNP levels
Echocardiographically, ventricular septal thickness, posterior wall thickness and left atrial diameter were significantly greater in patients with HCM than in control subjects. However, LV end-diastolic diameter and LV systolic and diastolic indexes did not differ significantly. Exercise tolerance, the increment of rate-pressure products and ΔsBP did not differ significantly between the two groups. Levels of ANP and BNP at rest were significantly greater in patients with HCM than in control subjects (both p < 0.0001). The changes of both ANP and BNP levels during dynamic exercise were more prominent in patients with HCM than in control subjects (Table 1). In patients with HCM, both ANP and BNP levels at rest did not show any significant correlations with echocardiographic and hemodynamic variables (Table 2).
Scintigraphic findings (fig. 1).
Reversible regional perfusion abnormalities were observed in 16 patients with HCM (51.6%). The prevalence (%) of ischemic changes in each segment was 41.9 in anteroseptal, 9.7 in septal, 32.3 in posteroseptal, 6.5 in posterolateral, 0 in lateral, 3.2 in anterolateral and 16.1 in apical wall. Transient dilation index was significantly greater in HCM patients than in control subjects (1.21 ± 0.27 vs. 1.00 ± 0.06, p = 0.004) (Table 1). Abnormal apparent LVCD was identified in 14 patients with HCM (45.2%). Overall, exercise-induced myocardial ischemia was found in 19 patients (61.3%). The reproducibility for calculating TDI was excellent (r = 0.99 for both intraobserver and interobserver analyses).
Septal perforator compression (fig. 2).
Septal perforator compression in the left anterior descending artery was identified in 20 (64.5%) HCM patients: nine cases with grade 1, eight with grade 2 and three with grade 3. Three of these 20 patients with HCM (15.0%) demonstrated the concurrence of SPC in the right coronary artery. Bridging of the epicardial coronary artery was not found in any study patient.
Clinical features in patients with HCM with or without inducible myocardial ischemia (table 3).
We also compared the clinical findings of HCM patients in terms of the presence or absence of exercise-induced myocardial ischemia. Age, family history of HCM, echocardiographic findings, rate-pressure products, ΔsBP and ST depression did not differ significantly between the two subgroups. However, LV end-diastolic pressure (LVEDP) was greater, and SPC was more frequent, in patients with HCM having inducible ischemia. The increment of ANP during exercise was similar between the two subgroups. However, the HCM subgroup with inducible ischemia showed higher BNP levels at rest (p = 0.02) and more marked increment of BNP levels during exercise (p = 0.02) than did those without.
Multiple logistic regression analysis
By multiple logistic regression analysis, SPC and BNP levels at rest were independently associated with exercise-induced myocardial ischemia (p = 0.03 and p = 0.02, respectively).
Myocardial ischemia and SPC in HCM
Apparent LVCD and regional perfusion abnormality have been reported to be the characteristic SPECT findings of ischemia in HCM (12,20,21,23,24). The present apparent LVCD is considered to demonstrate exercise-induced subendocardial ischemia, because this scintigraphic abnormality occurs in the absence of any significant changes in end-diastolic volume detected by radionuclide angiography, as previously confirmed (20). Our software had excellent reproducibility, and TDI enables us to quantify subendocardial ischemia more reliably than does the usual method of judging visually the presence or absence of apparent LVCD (18,24).
Subendocardial ischemia is related to elevated LV filling pressures and impaired myocardial relaxation (23,25), and LVEDP was actually higher in our HCM patients with exercise-induced myocardial ischemia than in those without. Unlike the results in a previous report (18), exercise-induced myocardial ischemia in this study is not considered related to abnormal blood pressure responses, because ΔsBP did not differ between HCM patients with exercise-induced ischemia and those without. The difference may be attributed to the fact that our HCM patients were free from cardiac symptoms and had relatively mild cardiac dysfunctions.
Septal perforator compression was significantly more frequent in HCM patients with exercise-induced myocardial ischemia; this association was independent of other factors. There are also several studies suggesting the contribution of SPC to myocardial ischemia (19,26,27). Hirasaki et al. (27)demonstrated that the septal perforators were deviated toward the LV side in HCM and suggested that SPC is caused by high intramural pressure. However, this mechanism does not fully explain the discrepancy between widespread 99mTc-tetrofosmin defects and the perfusion territory of the septal perforators, as has been already pointed out (12,19).
Myocardial disarray is associated with myocardial ischemia (28). Asynchrony of contraction and tension due to disarray will impose abnormal mechanical force on the muscle fibers and capillary beds, leading to perfusion disorder and ischemia at rest and to their further intensification during exercise. Disarray is often observed in anteroseptal, posteroseptal and apical walls of the LV (29). Marked fascicle disarray is present in the anterior and posterior areas of junction of the LV free wall and ventricular septum, with fiber disarray extending to the adjacent portion of the free walls and septum; however, disarray extends to the middle portion of the septum to a far lesser degree (30). The predominant area of 99mTc-tetrofosmin defects corresponds well with the distribution of fascicle disarray and fiber disarray. The rationale for the coincidence of disarray and leftward deviation of the septal perforators has been described elsewhere (27).
ANP and BNP in HCM
High ANP and BNP levels are associated with diastolic dysfunction (6,7). In our HCM patients also, diastolic dysfunction as suggested by impaired LV relaxation and elevated LVEDP, though mild, may have participated in ventricular wall stretch and BNP secretion. Greater left atrial size, reflecting atrial wall stretch due to diastolic dysfunction, may have promoted ANP release. Cardiac hypertrophy was reported to be a strong stimulus to the elevation of BNP levels (8,9,16,31). However, these contributions are not considered large in the present study, because both ANP and BNP levels at rest failed to demonstrate any significant correlations with echocardiographic and hemodynamic variables.
Immunohistochemical analysis disclosed that BNP immunoactivity in the ventricular wall was related to myocardial disarray, myocyte hypertrophy and fibrosis in HCM (32). 99mTc-tetrofosmin defects were predominantly located in the area where fascicle or fiber disarray, myocyte hypertrophy and fibrosis are marked (30), suggesting that myocardial ischemia due to these histological changes may participate in the accelerated BNP secretion (28). Furthermore, in parallel with ischemia, abnormal stretch of some muscle fibers due to the asynchrony of contraction and tension in the disarrayed myocardium may stimulate the ventricular BNP secretion at rest and during dynamic exercise.
Unlike ANP levels, BNP levels at rest were higher, and the increment of BNP levels on dynamic exercise was more marked, in HCM patients with inducible myocardial ischemia than in those without. Together with SPC, a high BNP level at rest was independently associated with exercise-induced myocardial ischemia. Although the precise mechanism explaining the relationship between the BNP levels and myocardial ischemia remains obscure, these results present a significant clinical implication that BNP levels at rest may be useful in predicting silent myocardial ischemia.
We evaluated LV diastolic functions by Doppler echocardiography just before the ergometer exercise. There is, however, little evidence as to whether Doppler indexes reflect diastolic function as accurately as the invasive evaluation. (33). Trans-catheter measurement of hemodynamics was not performed during ergometer exercise, but we had obtained LVEDP during angiographic study. Because the time lag was small, we may safely consider that a high LVEDP was partly related to subendocardial ischemia and accelerated BNP secretion.
In patients with HCM, exercise-induced myocardial ischemia was associated with higher BNP levels at rest and more remarkable increase of BNP levels during exercise. Evaluation of plasma BNP levels at rest may be useful in predicting silent myocardial ischemia noninvasively.
The authors are grateful to Toshiro Kuribayashi, MD, PhD (Fukuoka, Japan), for reviewing the manuscript and providing invaluable suggestions. We would also like to thank Shionogi and Co., Ltd. (Osaka, Japan) for generously measuring plasma ANP and BNP levels and Chieko Okuda, PhD (Louis Pasteur Center for Medical Research, Kyoto, Japan), for helpful suggestions in statistical analyses.
- atrial natriuretic peptide
- brain natriuretic peptide
- hypertrophic cardiomyopathy
- isovolumetric relaxation time
- left ventricular/ventricle
- left ventricular cavity dilation
- left ventricular end-diastolic pressure
- increase of systolic blood pressure
- septal perforator compression
- single photon emission computed tomography or tomographic
- transient dilation index
- Received August 2, 2001.
- Revision received February 13, 2002.
- Accepted February 19, 2002.
- American College of Cardiology Foundation
- Yasue H.,
- Yoshimura M.,
- Sumida H.,
- et al.
- Davidson N.C.,
- Naas A.A.,
- Hanson J.K.,
- Kennedy N.S.J.,
- Coutie W.J.,
- Struthers A.D.
- Tsutamoto T.,
- Wada A.,
- Maeda K.,
- et al.
- Yu C.M.,
- Sanderson J.E.,
- Shum I.O.L.,
- et al.
- Yasumoto K.,
- Takata M.,
- Ueno H.,
- Tomita S.,
- Tomoda F.,
- Inoue H.
- Takagi E.,
- Yamakado T.,
- Nakano T.
- O’Gara P.T.,
- Bonow R.O.,
- Maron B.J.,
- et al.
- Dilsizian V.,
- Bonow R.O.,
- Epstein S.E.,
- Fananapazir L.
- Hirata Y.,
- Matsumoto A.,
- Aoyagi T.,
- et al.
- Yoshida N.,
- Ikeda H.,
- Wada T.,
- et al.
- Mohiddin S.A.,
- Begley D.,
- Shih J.,
- Fananapazir L.
- Sugihara H.,
- Shiga K.,
- Umamoto I.,
- et al.
- Udelson J.E.,
- Bonow R.O.,
- O’Gara P.T.,
- et al.
- Cannon R.O. III.,
- Dilsizian V.,
- O’Gara P.T.,
- et al.
- Cannon R.O. III.,
- Rosing D.R.,
- Maron B.J.,
- et al.
- Cannon R.O. III.
- Varnava A.M.,
- Elliott P.M.,
- Sharma S.,
- McKenna W.J.,
- Davies M.J.
- Hasegawa K.,
- Fujiwara H.,
- Doyama K.,
- et al.
- Nishimura R.A.,
- Appleton C.P.,
- Redfield M.M.,
- Ilstrup D.M.,
- Holmes D.R.,
- Tajik A.J.
- Study protocol
- Echocardiographic study
- 99mTc-tetrofosmin scintigraphic study and measurement of plasma ANP and BNP levels
- Image analysis
- Angiographic assessment
- Statistical analysis
- Scintigraphic findings (fig. 1).
- Septal perforator compression (fig. 2).
- Clinical features in patients with HCM with or without inducible myocardial ischemia (table 3).
- Multiple logistic regression analysis
- ANP and BNP in HCM
- Study limitations