[ Scientific Activities - Actividades Científicas ]
Honorary Committee Lecture
Usefulness of Plasma Levels of the Cardiac Hormones Atrial Natriuretic Factors (ANF) and Brain Natriuretic Peptide (BNP) in Clinical Cardiology
Adolfo J. de Bold, PhD
University of Ottawa Heart Institute
The cardiac natriuretic peptides (NP) atrial natriuretic factor (ANF) (3;4) and brain natriuretic peptide (BNP) (10) are polypeptide hormones synthesized, stored and released by cardiac muscle cells (cardiocytes) in the atria of mammals. Under pathophysiological conditions, both ANF and BNP may also be produced by ventricular cardiocytes. Indeed, production of ANF and BNP by the mammalian ventricle is a hallmark of cardiac hypertrophy (14;16).
ANF and BNP are released from the heart at a basal rate that increases following mechanical (hemodynamic) or neuroendocrine, often combined, stimuli. The many biological properties of ANF and BNP allow these hormones to interact with fast responding as well as slow onset mechanisms involved in cardiovascular homeostasis. In many ways, the endocrine heart is as a modulator of several systems such as the sympathetic nervous system, the renin-angiotensin-aldosterone system and other determinants of vascular tone, extracellular fluid volume and renal function (3).
Both ANF and BNP are synthesized by cardiocytes as preprohormones that are enzymatically processed to yield prohormones and, ultimately, hormones that are released into the circulation. In humans, the prohormone proANF is a polypeptide that contains 126 amino acids (ANF1-126) that is processed to ANF1-98 (N-terminal ANF = NT-ANF) and ANF99-126 (C-terminal ANF = CT-ANF). The latter is the biologically active portion. Human proBNP, on the other hand, is 108 amino acids long and it is processed to BNP1-76 (N-terminal BNP = NT-BNP) and BNP 77-108 (C-terminal BNP = CT-BNP). Similarly to CT-ANF, CT-BNP is the biologically active peptide.
Natriuretic Peptides and Cardiovascular Disease
Given the physiological properties and the variables that influence the release of ANF and BNP, it may be expected that their release ant therefore, their plasma levels would be altered in pathophysiological states involving the cardiovascular system and even provide an indirect measure of the hemodynamic burden on the heart under a variety of conditions. Indeed, it has been shown that several pathological conditions affecting the heart are accompanied by altered plasma NP levels (15). Moreover, it has become apparent that ANF and BNP plasma levels can provide surprisingly useful insights that complement and provide unique information applicable to current diagnostic and prognostic parameters in cardiovascular disease. As discussed below, it is now known that measurement of the circulating levels of different fragments of these hormones in plasma is a powerful means to, for example, identify elderly subjects at risk of heart failure (2), to establish long term prognosis after MI (6), to stratify patients in terms of response to ACE inhibition post MI (11) and can demonstrate asymptomatic left ventricular dysfunction (1;9). These results are particularly encouraging for environment where LV function assessment is not available. Even in the cases where it is available, several authors have proposed that the measurement of ANF or BNP may be an independent predictor of cardiovascular outcome.
Soon after the development of specific and sensitive radioimmunoassays for ANF in plasma, it was apparent that the circulating levels of this hormone were significantly elevated in a variety of clinical conditions, all of which were underlain by an increase in pressure or volume load on the heart. It was further established that the plasma levels of CT-ANF were, in general, proportional to the degree and duration of this overload. Thus the highest circulating plasma ANF levels were associated with long-standing essential hypertension and in chronic congestive heart failure classes III and IV. A similar historical development may be found for BNP, for which the development of a radioimmunoassay demonstrated that elevated levels of circulating BNP were an indication of long-standing pressure or volume overload. Further, in the case of BNP a strong induction of the gene expression for these peptides was observed in the hypertrophied ventricle resulting in a relatively larger elevation of plasma BNP levels than those of ANF although, in absolute terms, circulating levels of ANF remain larger than those of BNP.
Having been established that important clinical entities are accompanied by changes in ANF and BNP gene expression, several clinical investigators have explored the possibility that ANF and BNP plasma levels may be used as diagnostic or prognostic indicators of cardiovascular disease. Sthruthers et al. (11) have summarized possible uses of plasma natriuretic peptide concentration measurements given what is already in the literature. These possible uses include: 1) detection of early heart failure; 2) replacement of radionuclide ventriculography; 3) identification of left ventricular dysfunction after myocardial infarction; 4) assisting in the diagnosis of myocardial ischemia; 5) detection of left ventricular hypertrophy in hypertension; 6) early diagnosis of hypertrophy in obstructive cardiomyopathy. 7) very recently we demonstrated that BNP plasma levels may be potentially useful to predict and help diagnose cardiac allograft rejection (Masters et al. Circulation, 1999, In Press). Struthers et al. exemplify the potential usefulness of ANF or BNP plasma measurements citing a population study in which 1.6% was found treated for presumed heart failure yet only 0.84% had echocardiographically demonstrable left ventricular systolic dysfunction. The authors argue that it will be less expensive to identify the 0.84% than to treat l.6% of the population with expensive therapeutic agents such as angiotensin converting enzymes inhibitors. The problem is however, that few hospitals have the resources to perform echocardiography on 1.6% of the population. In this situation is when the measurement of plasma ANF or BNP may be a far more inexpensive way of identifying individuals who require treatment.
Similar conclusions on the usefulness of natriuretic peptide plasma levels have been reached by several authors in a variety of clinical situations. Richards et al. (13) found a good correlation between plasma CT-ANF and cardiac output in elderly patients and further showed that plasma CT-ANF was a prognostic indicator of those who would subsequently develop chronic heart failure. Similarly, Davis et al. (2) showed that in the frail and elderly, CT-ANF plasma levels identifies subjects at risk of CHF allowing for appropriate focussing of medical resources for the prevention, early detection and treatment of this syndrome. At the Mayo Clinic, Lerman et al. (9) have shown that NT-ANF is a highly sensitive marker for symptomless left ventricular dysfunction. Motvany et al (11) have shown that CT-BNP plasma levels identify patients with left ventricular dysfunction who have been identified by the SAVE study as likely to benefit from long term ACE treatment after MI. In this study it was found that plasma CT-ANF was not as predictive as CT-BNP. The usefulness of NT-ANF plasma levels to evaluate clinical status has been recently re-emphasized by a study of Dickstein et al (5) who showed that NT-ANF correlated better than other variables with New York Heart Association functional class and was more closely associated with non invasive measurements than New York Heart Association functional class. Odds ratio measurement demonstrated substantially increased risk of left ventricular dysfunction and dilatation, pulmonary hypertension and New York Heart Association functional class 3 or 4 with increasing NT-ANF value. The authors concluded that the data clearly indicated that the concentration of proANF is related to the degree of clinical heart failure. They further state that since the analysis is simple, it should be of practical value as a supplement in the routine assessment of cardiac status given the heterogeneous nature of the heart failure population. Finally, a recently published study dramatically shows the importance of NP plasma level measurement: Hall et al. (6), reporting for the Thrombolysis in Myocardial Infarction II investigators, showed that NT-ANF, when measured during the first 12 h after the onset of chest pain, is related to 1-year mortality after MI.
It is to be noted that the studies that have compared measurements of NP with other neuroendocrine variables such as norepinephrine circulating levels, ANF and BNP have shown the closest correlation to cardiovascular functional status (5;8). In the early stages of heart failure, plasma renin activity and sympathetic activity are normal because they are suppressed by the increased circulating levels of NP. This explains why NP are elevated before other neuroendocrine variables and explains why the determination of NP plasma levels are an early and sensitive indicator of ventricular dysfunction.
It is also important to note the changes in NP secretion, as for example in studies demonstrating an elevation of plasma NP in subclinical systolic or diastolic dysfunction (1), are distinctive enough to make the measurements useful in diagnosis and prognosis beyond the group statistics, that is, they are useful adjuncts in the evaluation of the individual patient.
Critical Appraisal of the Measurement of Natriuretic Peptide Plasma Levels within a Clinical Environment
It is evident that all literature relating to the clinical application of measurement of NP plasma levels coincide in that in that these measurements have a high, established prognostic and diagnostic value in the management of cardiovascular disease. Further research is needed however, to resolve logistic as well as strategic aspects of this new technology. These are listed below.
1. Measurement of ANF and BNP. The measurement of ANF and BNP is still restricted to tertiary care centers because it needs sophisticated extraction procedures, including prompt centrifugation at 4 oC , pre-purification with chromatography, very low temperature storage of samples (except for claims of high stability of N-terminal ANF, not confirmed) and the handling of radioactivity and the appropriate equipment to perform counting and reduction of data by radioimmunoassay which by itself, requires highly trained personnel and fairly sophisticated laboratory facilities. This restricts the use of these procedures to specially trained personnel in specialized centres. It is probably because of this reason that the measurement of ANF and BNP have been restricted mainly to clinical investigations studies and no transference has been made to less sophisticated environments.
2. Populations studies. There are restrictions in several of the studies mentioned above. Some criticisms apply because of the small number of patients recruited and others apply because the patient populations are not comparable. For example, the recent study of Omland et al. (12) have proposed that BNP and not ANF or N-ANF is an important predictor independent of ejection fraction while N-ANF does not provide information beyond of what ejection fraction can provide. This is in contrast to the results reported by Hall et al (7) that found that N-ANF was a stronger predictor of death than ejection fraction and various other variables studied. It has been pointed out that the differences between these two studies may be due to differences between baseline LV function with the former studies including patients with ejection fractions of <45% while the latter included ejection fractions of #40%. In addition, the study of Omland et al. (12) is small compared to the study of Hall et al. and the number of patients with echocardiographic data was even smaller. Finally, no attempt was made in the former study to define the influence of sampling timing or number of samples taken after the initial event.
In conclusion, it may be said that the measurement of different fragments of the cardiac hormones ANF or BNP is emerging as uniquely valuable new technology that may be particularly useful for the assessment of cardiac patients outside of sophisticated medical environments. In addition, even in these environments, the new test might provide useful insight for stratification of the cardiac patient. However, further work is needed to refine the technology and the definition of patient populations where the new technology is relevant.
1. Arad, M., E. Elazar, A. Shotan, R. Klein, and B. Rabinowits. Brain and atrial natriuretic peptides in patients with ischemic heart disease with and without heart failure. Cardiology 87: 12-17, 1996.
2. Davis, K.M., L.C. Fish, D. Elahi, B.A. Clark, and K.L. Minaker. Atrial natriuretic peptide levels in the prediction of congestive heart failure risk in frail elderly. JAMA 267: 2625-2629, 1992.
3. de Bold, A.J. Atrial natriuretic factor: a hormone produced by the heart. Science 230: 767-770, 1985.
4. de Bold, A.J., H.B. Borenstein, A.T. Veress, and H. Sonnenberg. A rapid and potent natriuretic response to intravenous injection of atrial myocardial extracts in rats. Life Sci. 28: 89-94, 1981.
5. Dickstein, K., A.I. Larsen, V. Bonarjee, M. Thoresen, T. Aarsland, and C. Hall. Plasma proatrial natriuretic factor is predictive of clinical status in patients with congestive heart failure. Am.J.Cardiol. 76: 679-683, 1995.
6. Hall, C., C.P. Cannon, S. Forman, and E. Braunwald. Prognostic value of N-terminal proatrial natriuretic factor plasma levels measured within the first 12 hours after myocardial infarction. J.Am.Coll.Cardiol. 26(6): 1452-1456, 1995.
7. Hall, C., J.L. Rouleau, L. Moye, J. de Champlain, D. Bichet, M. Klein, B. Sussex, M. Packer, J. Rouleau, M.O. Arnold, and et al. N-terminal proatrial natriuretic factor. An independent predictor of long-term prognosis after myocardial infarction. Circulation 89: 1934-1942, 1994.
8. Hall, C., J.L. Rouleau, L. Moyé, J. de Champlain, D. Bichet, M. Klein, B. Sussex, M. Packer, J. Rouleau, M.O. Arnold, G.A. Lamas, F. Sestier, S.S. Gottlieb, C.-C.C. Wun, and M.A. Pfeffer. N-Terminal Proatrial Natriuretic Factor. Circulation 89: 1934-1942, 1994.
9. Lerman, A., R.J. Gibbons, R.J. Rodeheffer, K.R. Bailey, L.J. McKinley, D.M. Heublein, and J.C. Burnett, Jr. Circulating N-terminal atrial natriuretic peptide as a marker for symptomless left-ventricular dysfunction. Lancet 341: 1105-1109, 1993.
10.Maekawa, K., T. Sudoh, M. Furusawa, N. Minamino, K. Kangawa, H. Ohkubo, S. Nakanishi, and H. Matsuo. Cloning and sequence analysis of cDNA encoding a precursor for porcine brain natriuretic peptide. Biochem.Biophys.Res.Commun. 157: 410-416, 1988.
11.Motwani, J.G., H. McAlpine, N. Kennedy, and A.D. Struthers. Plasma brain natriuretic peptide as an indicator for angiotensin- converting-enzyme inhibition after myocardial infarction . Lancet 341: 1109-1113, 1993.
12.Nakazato, M., H. Yamaguchi, H. Kinoshita, K. Kangawa, H. Matsuo, N. Chino, and S. Matsukura. Identification of biologically active and inactive human uroguanylins in plasma and urine and their increases in renal insufficiency. Biochem.Biophys.Res.Commun. 220: 586-593, 1996.
13.Richards, A.M., I.G. Crozier, T.G. Yandle, E.A. Espiner, H. Ikram, and M.G. Nicholls. Brain natriuretic factor: regional plasma concentrations and correlations with haemodynamic state in cardiac disease. Br.Heart J. 69: 414-417, 1993.
14.Sadoshima, J. and S. Izumo. Mechanical stretch rapidly activates multiple signal transduction pathways in cardiac myocytes: potential involvement of an autocrine/paracrine mechanism. EMBO J. 12: 1681-1692, 1993.
15. Struthers, A.D. Ten years of natriuretic peptide research: a new dawn for their diagnostic and therapeutic use?. [Review]. BMJ 308: 1615-1619, 1994.
16. Yokota, N., B.G. Bruneau, B.E. Fernandez, M.L. Kuroski de Bold, L.A. Piazza, H. Eid, and A.J. de Bold. Dissociation of cardiac hypertrophy, myosin heavy chain isoform expression, and natriuretic peptide production in DOCA-salt rats. Am.J.Hypertens. 8: 301-310, 1995.
Adolfo J. de Bold, PhD
University of Ottawa Heart Institute
40 Ruskin St., Ottawa, ON K1Y 4W7, Canada