B-type natriuretic peptides: Looking to the future

Abstract

Whereas the role of the cardiac natriuretic peptides, ANP and BNP, in some aspects of physiology and pathophysiology is clear, their potential in diagnosis, prognosis, and therapeutics in many clinical disorders remains uncertain. We predict that circulating levels of these peptides will find increasing diagnostic utility in patients presenting with dyspnoea, in guiding the complex pharmacotherapy in heart failure, and may likewise be useful in guiding the management of patients on chronic maintenance renal dialysis. We predict also that levels of these peptides will be of practical use as prognostic indicators in ‘at-risk’ populations (such as those with diabetes, coronary heart disease, hypertension, thalassaemia, etc.) but probably not in the general population. It appears likely that administration of these peptides will find a place in the therapeutics of acute myocardial infarction, but this is less clear for heart failure. We describe the presence of a segment of the signal peptide for BNP within the circulation and discuss its potential clinical utility.

Key words::

• Atrial natriuretic peptide
• B-type natriuretic peptide
• diabetes mellitus
• heart failure
• hypertension
• myocardial infarction
• renal failure
• signal peptide

Abbreviations
ACE	=	angiotensin-converting enzyme
ANP	=	atrial natriuretic peptide
BNP	=	B-type natriuretic peptide
BNPsp	=	B-type natriuretic peptide signal peptide
ESRD	=	end-stage renal disease
NEP	=	neutral endopeptidase 24.11
NPR-A	=	type A natriuretic peptide receptor
NT-proBNP	=	N-terminal B-type natriuretic peptide

Key messages

• Plasma levels of the cardiac natriuretic peptides are likely to find increasing clinical use in the differential diagnosis of patients with dyspnoea, and in guiding the management of patients with chronic heart failure and patients on maintenance renal dialysis.

• Plasma levels of the cardiac natriuretic peptides will probably be used increasingly as prognostic indicators in ‘at-risk’ populations.

• The therapeutic potential of administered cardiac natriuretic peptides appears good after acute myocardial infarction but remains uncertain in heart failure.

Introduction

It is widely appreciated that the heart secretes atrial and B-type natriuretic peptides (ANP and BNP) into the circulation where they are important contributors to circulatory homeostasis through multiple actions with a wide spectrum in time frames. Broadly speaking, in the short term they protect against fluid volume overload, and in the longer term they counteract the tendency to overgrowth of cardiovascular tissues as well as any trend toward fluid volume overload. Under many circumstances they act as a counterpoise to the level of activity and actions of the renin-angiotensin system and aldosterone which protect against fluid volume depletion and tend to promote growth of cardiovascular tissues.

Whereas there is clear understanding of many aspects of the role played by these peptides in the fields of physiology and pathophysiology, there are numerous areas that have defied research attempts to date or remain to be investigated. We here have selected a few topics where the sum of available investigations has given uncertain or conflicting results, and some areas of clinical relevance that have yet to be studied. We chose these topics since we believe they hold considerable potential to increase our understanding of the role of these peptides in health and disease or have diagnostic or therapeutic potential. At the end of each topic we have added a section ‘predicting the future’. In doing so, we have purposely been controversial in order to stimulate discussion and research. We hope that with the passage of time clear, scientifically based outcomes regarding these selected issues are achieved in the interests of improved understanding of mechanisms underlying health and disease—and management of the latter. The National Academy of Clinical Biochemistry guidelines for the clinical use of BNP/N-terminal B-type natriuretic peptide (NT-proBNP) in situations other than acute coronary syndromes and heart failure were published in 2007 (1) and discuss some of the topics we have chosen here.

The literature on the cardiac natriuretic peptides is enormous—hence our bibliography is selective rather than comprehensive. Although earlier studies were focused on ANP since it was discovered before the B-type peptides, the latter have, in recent years and for various reasons (including the wide-spread availability of robust commercial assays), often taken centre stage. In this review we concentrate largely on the two B-type peptides, BNP and NT-proBNP, assays for which have provided broadly similar information despite that fact that absolute levels of the former are considerably lower than levels of the latter under most circumstances.

Diagnostic use of cardiac natriuretic peptide levels in dyspnoeic patients

Although there have been contrary reports, most studies have shown circulating BNP/NT-proBNP levels to be useful in the diagnosis of acute heart failure in dyspnoeic patients. This has been particularly so in the setting of the emergency department where their use, in conjunction with other clinical information, has improved evaluation and treatment whilst also reducing costs of treatment (2–6). The other situation is primary care where, again, most though not all studies point to the clinical usefulness of BNP/NT-proBNP levels in the diagnosis or, in particular, the exclusion of heart failure in dyspnoeic patients (7,8). In both the emergency department and in primary care, patients with known or suspected chronic airways disease presenting with dyspnoea are a diagnostic dilemma. Under these circumstances, a normal BNP/NT-proBNP level is particularly useful in ruling out a diagnosis of heart failure (9–11). Interpretation of circulating BNP/NT-proBNP levels can be complicated, however, by the fact that a number of variables beyond indices of cardiac function (including gender, race, body mass index, renal function, and anaemia), can contribute to the secretion and/or clearance of both peptides.

Predicting the future:

1. We see it likely that BNP/NT-proBNP levels will be used as a routine in most emergency departments to assist ruling out and diagnosing heart failure in patients presenting with dyspnoea.

2. We see it probable that most primary care practices will ultimately make use of BNP/NT-proBNP levels as a routine in ruling out and in diagnosing heart failure in their dyspnoeic patients—especially those with known or suspected chronic airways disease. Point-of-care assays are likely to prove especially helpful in primary care.

3. The usefulness of BNP/NT-proBNP levels to clinicians in both the emergency department and primary care will be enhanced by laboratories providing ‘adjusted’ values according to clinical covariates known to alter circulating levels of the cardiac natriuretic peptides including age, gender, body mass index, renal function, and a history of atrial fibrillation—as proposed by Rogers and colleagues (12).

Cardiac natriuretic peptide levels as a guide to management in heart failure

There is robust evidence to support the use of BNP/NT-proBNP levels in the diagnosis of heart failure and in providing a prognosis. More recently their levels have been proposed as a guide to pharmacotherapy. It is not a new concept that the intensity of drug treatment for patients with chronic heart failure might prove more successful when guided in each individual according to an objective index rather than aiming for the same target drug dose in all patients based on results of randomized clinical trials. For example, Anderson and colleagues (13) suggested in 1988 that plasma levels of ANP might provide a clinically useful method of monitoring the response to treatment in cardiac failure. There were subsequent reports that tailoring of pharmacotherapy in order to reduce cardiac filling pressures and cardiac afterload in patients with severe congestive heart failure had sustained beneficial effects and might decrease the need for cardiac transplantation (14,15). Rather than using haemodynamic indices, Murdoch and colleagues (16) utilized plasma BNP levels as the basis for optimizing angiotensin-converting enzyme (ACE) inhibitor therapy and showed that this approach reduced heart rate and suppressed the renin-angiotensin system and aldosterone levels compared with optimal empirical therapy.

In the study by Troughton et al. (17) in 2000, patients with symptomatic heart failure and impaired left ventricular systolic function whose drug therapy was guided by circulating levels of NT-proBNP had statistically significantly fewer cardiovascular events and delayed time to first event compared to those randomized to management according to standardized clinical assessment. This, the first prospective, randomized trial of hormone-guided versus ‘usual’ therapy was carried out prior to beta-blocker therapy and was viewed by some commentators as being of short duration (median 9.5 months) and in a small cohort (n = 69), although, as noted above, statistically significant predetermined end-points were observed in favour of hormone-guided management. Furthermore, 9.5 months, rather than being short-term, might be viewed as medium- or even long-term for elderly patients with heart failure. In 2009, a meta-analysis of six prospective, randomized controlled studies of biomarker (BNP or NT-proBNP)-guided versus control therapy, including the Troughton trial and involving 1,627 patients, reported a sizeable and statistically significant reduction in all-cause mortality in the former group compared with the latter (hazard ratio 0.69; 95% CI 0.55–0.86) (Figure 1). Two studies within this meta-analysis (TIME-CHF (19) and BATTLESCARRED (20)) noted that the benefits from natriuretic peptide-guided therapy resided largely in patients aged < 75 years. As mentioned by the authors of the meta-analysis, there are a number of reasons why those > 75 years might benefit little, if at all, from hormone-guided treatment. The possibilities include the fact that BNP/NT-proBNP levels increase with age (raising the question of whether target hormone levels should vary according to the patient's age), and the very old often have heart failure associated primarily with diastolic left ventricular dysfunction which responds poorly to conventional pharmacotherapy (21). A meta-analysis subsequent to that of Felker et al. (18) which included eight (rather than six) studies confirmed that, compared with usual care, BNP-guided therapy reduced all-cause mortality in patients with chronic heart failure (hazard ratio 0.76; 95% CI 0.63–0.91), especially for those aged < 75 years (hazard ratio 0.52; 95% CI 0.33–0.82) (22).

Figure 1. Forest plot of all-cause mortality among patients with chronic heart failure randomized to biomarker-guided therapy versus control. The size of the marker for the point estimate (diamond) is proportional to the sample size for each study. Horizontal lines show 95% confidence intervals. From Felker et al., American Heart Journal, 2009, with permission (18).

This topic is of wide-spread theoretical and practical interest. One study from Austria involving 278 patients has been completed recently (23) and is not included in either of the meta-analyses (18,22). This study reported that NT-proBNP-guided management of drug therapy improved clinical outcomes compared to multidisciplinary care or ‘usual care’ (23). At least one other study, involving 1,250 patients with chronic heart failure, is underway to assess the efficacy of NT-proBNP monitoring (24).

Predicting the future:

1. With regard chronic heart failure, we think it likely that a meta-analysis based on individual patient (rather than grouped) data using results from studies included in the meta-analysis of Felker and colleagues (18), and possibly additional studies, will demonstrate a robust beneficial effect of hormone-guided therapy, versus usual care—at least for patients with impaired left ventricular systolic function. Such a meta-analysis is underway.

2. Our suspicion is that circulating BNP/NT-proBNP levels will soon be used as a routine, both in primary and secondary care, in the drug and non-drug (dietary sodium and fluid (25)) management of patients with chronic heart failure and impaired left ventricular systolic function. It will need to be emphasized repeatedly that the hormone-guided approach must complement, rather than replace careful routine clinical assessment and routine investigations (plasma electrolytes, renal function, etc.). We suspect that guidelines for the management of patients with chronic heart failure will, before long, recommend the use of BNP/NT-proBNP levels as a routine in guiding treatment in these patients.

3. For patients in whom chronic heart failure results primarily from impaired left ventricular diastolic function, and for those without access to evaluation of left ventricular function aged > 75 years, any benefit from BNP-guided management will require either a new treatment algorithm (perhaps with more flexible targets for BNP/NT-proBNP) and/or the introduction of new pharmacotherapeutic agents which demonstrably improve the outcome in such patients.

4. For patients with acute heart failure, it is likely that plasma BNP/NT-proBNP levels will be used widely and routinely as one (but never the only) index of response to treatment.

Cardiac natriuretic peptide levels as a guide in the management of patients on chronic renal dialysis

The mortality and morbidity from cardiovascular disease is extremely high in patients with end-stage renal disease, especially for those on maintenance dialysis (26,27). Circulating levels of the cardiac natriuretic peptides have, in most if not all studies, been shown to reflect the degree of haemodynamic impairment and cardiac pathology in such patients (28). Furthermore, as demonstrated by Zoccali and colleagues (29) in 246 patients on regular dialysis treatment, they are robust predictors of total and cardiovascular mortality. Likewise, Gutierrez et al. (30) reported that elevated base-line NT-proBNP levels were predictive of all-cause and cardiovascular mortality over 90 days and 1 year in 2,990 patients receiving chronic haemodialysis. They observed also that an increase in NT-proBNP after 3 months of dialysis was associated with increased mortality. Since changes in haemodynamic load in patients with end-stage renal disease (whether induced by dialysis or medications) appear to be reflected in parallel changes in cardiac natriuretic peptide levels, one can hypothesize that when current management methods are combined with treatment aimed at lowering elevated levels of BNP/NT-proBNP (i.e. hormone-guided therapy), this might prove superior to current management methods alone, especially with regard cardiovascular outcomes (28). In other words, just as hormone-guided treatment in chronic heart failure has proven benefits beyond usual care, as discussed above, a similar approach to the management of patients on chronic maintenance dialysis may likewise improve outcomes. Such a hypothesis was indeed proposed by Zoccali et al. (29), elaborated upon by Dastoor and colleagues (28), and echoed by Gutierrez and colleagues (30). In that the kidneys play a role in the clearance of natriuretic peptides from the circulation, further studies are required to determine the contribution of decreased renal function per se versus cardiac dysfunction in patients with end-stage renal disease.

Predicting the future:

1. We suspect that prospective, randomized studies will reveal that BNP/NT-proBNP-guided management of patients on chronic maintenance dialysis combined with current best practice will be superior regarding cardiovascular (and perhaps total) morbidity and mortality compared to current management methods alone.

2. Presuming that the above prediction turns out to be correct, we think it likely that circulating levels of BNP/NT-proBNP will be used routinely to guide clinicians in the management of their patients on maintenance dialysis—to complement rather than replace currently used indices including clinical symptoms and signs, body weight, and routine investigations (blood biochemistry, chest radiology, etc.).

Cardiac natriuretic peptide levels as prognostic indicators

B-type natriuretic peptides and cardiovascular risk in general and at-risk populations

When the data from 40 long-term prospective studies involving 87,474 subjects and 10,625 incident cardiovascular disease outcomes were reviewed in a meta-analysis, there were statistically robust associations between base-line circulating BNP/NT-proBNP levels and subsequent risk of cardiovascular disease (31). This was so for studies in the general population (mean age 40s–80s), for populations with elevated cardiovascular risk factors, and for populations with manifest stable cardiovascular disease. Whilst superficially these data might be taken to support the wide-spread use of BNP/NT-proBNP levels as predictors of cardiovascular disease, the authors of the meta-analysis are appropriately cautious citing possible publication bias, the relatively small data base from low- and intermediate-risk populations, and their use of published grouped data, whereas results based on individual participant results would have provided more statistically robust information.

Predicting the future:

1. We believe that circulating levels of BNP/NT-proBNP, alone or (more likely) in combination with other biomarkers such as C-reactive protein, will prove to be clinically useful and cost-effective in providing predictive information regarding the likelihood of cardiovascular disease in selected at-risk populations—especially those with diabetes mellitus, primary (‘essential’) hypertension, impaired renal function, thalassaemia, Chaga's disease, and possibly also in patients undergoing cardiac or elective non-cardiac surgery. Although there is already some information regarding the utilization of BNP/NT-proBNP in these clinical situations (32–46), many studies relate these levels to surrogate end-points (left ventricular hypertrophy or diastolic left ventricular dysfunction, for example) rather than to ‘hard’ end-points (stroke, myocardial infarction, heart failure, over-all or cardiovascular mortality, etc.) and/or involve small numbers of patients. Furthermore, for most of these clinical situations, there exist contrary reports which found no clear association between BNP/NT-proBNP levels and cardiovascular abnormalities or outcome. Nonetheless, we hold a buoyant view of the potential clinical usefulness of BNP/NT-proBNP levels in these ‘at-risk’ patient groups.

2. Our suspicion is that in healthy young to middle-aged populations BNP/NT-proBNP levels will not prove to be clinically useful or cost-effective as a marker of future cardiovascular disease or overall mortality—certainly not when taken in isolation and probably not even when used in conjunction with currently accepted risk markers (plasma glucose, lipids, blood pressure, etc.).

Coronary heart disease/acute coronary syndromes

Circulating levels of the cardiac natriuretic peptides, including BNP and NT-proBNP, increase after acute myocardial infarction (47). According to most though not all reports, circulating levels of BNP/NT-proBNP drawn within days of symptom onset in patients with acute coronary syndromes provide a prognostic index which, especially when combined with other biomarkers and/or clinical scoring systems, allows risk stratification over both the short and long term for major cardiac events (48–51). More surprising perhaps are observations that NT-proBNP levels are predictive of adverse outcomes in patients with stable coronary heart disease (52–54). More surprising again, B-type natriuretic peptide levels achieved during exercise or dobutamine testing have been used successfully in the detection of coronary artery disease (55,56).

Predicting the future:

1. We think it likely that circulating B-type natriuretic peptide levels recorded within days of the onset of an acute coronary syndrome will receive wide-spread use, in combination with other risk markers, to provide a prognostic index and a guide to follow-up investigations and treatment.

2. Our suspicion is that BNP/NT-proBNP measurements will be requested commonly for patients with stable coronary heart disease and used as an (among others) index of prognosis—and as a guide to the urgency or otherwise of additional investigations and therapy.

3. We think it probable that BNP/NT-proBNP levels will not be used in routine clinical practice to assist in the detection of myocardial ischaemia due to coronary artery disease.

Cardiac natriuretic peptides as therapeutic agents

In theory, an agent with the demonstrated actions of bioactive ANP and BNP should have therapeutic potential in a number of disorders, heart failure in particular but also acute myocardial infarction, renal failure, and primary (‘essential’) hypertension. Although circulating concentrations of BNP and NT-proBNP as measured by commercially available assays are grossly elevated in patients with severe grades of chronic heart failure, levels of authentic bioactive BNP 1–32 are extremely low or absent (57,58). This being so, one might anticipate that patients should respond extremely well to administered BNP and ANP—or other agents which activate the type A natriuretic peptide receptor (NPR-A). In fact, for whatever reasons, the renal effects of short-term intravenous infusion of ANP or BNP are attenuated (59) although beneficial haemodynamic responses have generally been observed. Whereas indeed infusion of BNP (as nesiritide, recombinant BNP, licensed in 2001 in the USA for the treatment of acute, decompensated heart failure) over a period of hours usually improves haemodynamic indices (60), the possibility that in the longer term renal function deteriorates and mortality actually increases (61) has caused concern.

Animal data suggest that the cardiac natriuretic peptides are cardioprotective in acute myocardial infarction (62). Preliminary studies in patients with acute myocardial infarction have shown that 60–72-hour infusions of nesiritide tended to improve left ventricular structure and function (63,64). Furthermore, a 3-day intravenous infusion of ANP (n = 277 patients) reduced infarct size and improved left ventricular ejection fraction compared with placebo administration (n = 292 patients) (65).

Innovative approaches with treatment potential include the development of a chimeric peptide combining the beneficial properties of C-type natriuretic peptide and Dendroaspis natriuretic peptide (66), design of a renal-protective peptide based on alternative splicing of BNP (67), development of an orally active form of BNP through its conjugation with a short amphiphilic oligomer (68), and the establishment of ANP-secreting human skin grafts (69). Most of this work, however, has been in animals or in vitro, hence the potential therapeutic benefits in humans remain to be determined.

Another approach to utilizing the potential therapeutic benefits of ANP and BNP has been inhibition of their break-down by blockade of the enzyme neutral endopeptidase 24.11 (NEP) (neprilysin). Whereas such ‘NEP inhibitors’ alone induced relatively minor responses, major antihypertensive and haemodynamic effects were observed when they were combined with an ACE inhibitor, omapatrilat being the most studied example. Sadly but unsurprisingly angio-oedema proved to be unacceptably common with these agents, the result, presumably, of inhibition of bradykinin break-down by both the ACE and the NEP inhibitors. Retaining the concept of simultaneously blocking the renin-angiotensin system and enhancing levels of the cardiac natriuretic peptides, a single molecule, LCZ696, has been developed which contains an angiotensin 11 type-1 receptor-blocking moiety and a NEP-inhibiting moiety. This agent has substantial antihypertensive effects and reduced pulse pressure in patients with essential hypertension and, from preliminary observations, did not cause angio-oedema (70).

Predicting the future:

1. We suspect future studies will show that the intravenous administration of ANP and BNP in patients with heart failure, whether acute, chronic, or acute-on-chronic, should have little or no place in therapeutics because the balance of benefit over risk (and cost) will be neutral or disadvantageous.

2. The cardiac natriuretic peptides will be confirmed as having beneficial short-term and long-term effects in patients suffering acute myocardial infarction.

3. Designer compounds with natriuretic peptide-like activity, described above, may find a place in the treatment of acute myocardial infarction but are unlikely to do so in patients with heart failure.

4. Agents, such as LCZ696, with dual actions to block angiotensin 11-type 1 receptors and increase circulating levels of ANP and BNP through blockade of NEP, are likely to find a place in the treatment of patients with essential hypertension and possibly also in heart failure.

BNP-signal peptide is present in the human circulation

The secretion of proteins from cells is directed by signal peptides (signal sequences) which are amino acid sequences at the N terminus of pre-proproteins. They guide the pre-proprotein through the process of translation into the endoplasmic reticulum and thence to secretory vesicle packaging in the Golgi apparatus ready for extracellular secretion. Upon completion of this task, signal peptides undergo proteolytic cleavage from their pre-proprotein, after which, according to current dogma, they are destroyed by intracellular degradation mechanisms. Research by our group has demonstrated that a 10-amino acid fragment of BNP signal peptide (BNPsp) is present not only in cytosolic extracts of explant human heart tissue but also in the circulation of healthy human subjects (71). Furthermore, not only are plasma levels of BNPsp elevated three times the upper limit of normal in patients with ST-segment acute myocardial infarction, but they achieve peak values significantly earlier than currently used biomarkers myoglobin, CK-MB, and troponin I, and very much earlier than NT-proBNP levels (Figure 2).

Figure 2. Plasma levels of BNPsp, NT-proBNP, myoglobin, CK-MB, and troponin I in 25 patients with acute myocardial infarction and ST-segment elevation presenting within 4 hours of the onset of symptoms. Data are shown as mean ± SEM. From Siriwardena et al., Circulation, 2010, with permission (71).

Predicting the future:

1. We think it likely that signal peptides (or fragments thereof) for hormones other than BNP, will be shown to exist in the human circulation. Our group has already shown the presence of immunoreactive signal peptides associated with a number of peptide hormones in human plasma (unpublished data), but definitive evidence that such immunoreactivity indeed reflects the presence of signal peptides is awaited.

2. It is probable, in our view, that plasma levels of signal peptides associated with a number of hormones, including BNPsp, will provide information complementary to that from measurements of the bioactive hormone itself, and such information will prove useful to physiologists and clinicians.

Overview

We have here selected a few topics relating to the cardiac natriuretic peptides we felt to be of particular interest since they have potential to assist in the diagnosis or management of large patient populations. Many other areas of interest involving these peptides (their circulating levels in patients with cardiac valve disease, and in the early detection of rejection after cardiac transplantation, for example) exist and deserve further study. We have largely ignored ANP and NT-proANP in this discussion, yet some consider ANP to be as or more important physiologically than BNP. Furthermore, the diagnostic and prognostic utility of NT-proANP is seen by some to be superior to that of BNP and NT-proBNP. Our observations, that BNPsp is present in the human circulation and its levels are increased earlier than currently used biomarkers after acute myocardial infarction, likely herald a whole new field of research into the presence and significance of signal peptides in the circulation in man.

Declaration of interest: The authors state no conflict of interest and have received no payment in preparation of this manuscript.