Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Brain natriuretic peptide (BNP) is a cardiac hormone secreted by cardiomyocytes in response to ventricular wall stretch. It counteracts the renin–angiotensin–aldosterone and sympathetic nervous systems by inducing vasodilation, diuresis and natriuresis. BNP has been widely used as a marker of ventricular dysfunction1 2 and was found to be predictive of survival in congestive heart failure3 and idiopathic pulmonary arterial hypertension (PAH).4
In this issue, Diller et al report on their retrospective analysis of a large Eisenmenger population (181 patients) undergoing plasma BNP measurements as part of their routine clinical assessment.5 They found that BNP concentrations were increased in these patients and that elevated BNP concentrations were predictive of morbidity and all-cause mortality independent of Down's syndrome, renal function and 6 min walk test distance. They also observed that treatment of Eisenmenger patients with disease targeting therapies was associated with a significant reduction in BNP concentrations.
Although BNP concentrations have often been used to guide therapy in heart failure,6 few data are available in the setting of PAH.7 8 The conclusions of Diller et al are appealing and support previous studies indicating that targeting therapies are associated with a reduction in BNP concentrations in PAH.
Eisenmenger syndrome is a complex pathological condition that can affect multiple organs, whose function progressively deteriorates over time. In this respect, BNP is just one among the many key features that should be taken into account for clinical decision making in such a composite scenario. Functional class is used to determine a patient's clinical condition, and the 6 min walk test is considered as a measure of exercise tolerance in PAH patients. However, assessing functional class may be difficult in Eisenmenger syndrome patients because of their self-limiting behaviour and, as nicely stated by Diller et al, exercise capacity is also largely influenced by non-cardiac factors such as oxygen carrying capacity of the blood and the degree of desaturation during exercise. The consequences of chronic hypoxaemia include secondary polycythaemia, hyperviscosity, gout, joint and long bone pain due to hypertrophic osteoarthropathy, bleeding (haemoptysis) or thrombosis.9 Syncope and pre-syncope are usually associated with inadequate cardiac output as a result of physical effort, cold or warm exposure, or arrhythmias.9
Renal failure due to cardiac dysfunction, glomerular abnormalities and hyperuricaemia, and hepatic dysfunction, including cholelithiasis and cholecystitis, are frequent.
Considering the complex pathophysiology of Eisenmenger syndrome, it is not surprising that there are different factors relating to deterioration and death in this population (table 1). Daliento et al10 retrospectively analysed the clinical course of 188 patients from three different expert centres in adult congenital heart disease, followed for a median period of 31 years. They observed that complex anatomy, non-cardiac surgery, pregnancy, renal failure, younger age at presentation and overt right ventricular dysfunction were all predictors of mortality. The most frequent cause of death was sudden unexpected death (29.5%), followed by heart failure (22.9%) and massive haemoptysis (11.4%), mainly due to pulmonary artery rupture.
In a retrospective study of 109 adult patients with Eisenmenger syndrome followed for a median of 6.3 years, Cantor et al11 found that younger age at presentation, complex anatomy, supraventricular arrhythmias and voltage evidence of right ventricular hypertrophy were associated with increased mortality.
In another large retrospective single centre study, Diller et al12 assessed the risk factors for mortality in 171 Eisenmenger patients for a median follow-up of 67 months. Similar to previous studies, functional class, signs of heart failure and a history of clinical arrhythmias were found to be predictors of mortality. In addition, several ECG features (ie, QRS duration and QTc interval) and low serum albumin and potassium levels were shown to have a prognostic impact on survival. More recently, in a population of 229 Eisenmenger patients, the same group13 observed that treatment with specific drugs for PAH resulted in a significant reduction in mortality, even after adjustment for baseline clinical differences. The authors concluded that specific drugs for PAH in a large contemporary cohort of adult patients with Eisenmenger syndrome were associated with a lower risk of death.
In patients with Eisenmenger syndrome, it is generally assumed that extensive structural changes in pulmonary resistive vessels may prevent the persistence of significant vasoreactivity.19 Recent studies, however, show that residual pulmonary vasoreactivity is relatively common in patients with Eisenmenger syndrome, and is a significant predictor of clinical outcome.14 15
The clinical utility of BNP measurement in adult patients with congenital heart disease has already been assessed. In a retrospective study of 62 paediatric and adult patients with congenital heart disease,16 BNP levels were significantly higher in patients with ventricular dysfunction. In particular, the degree of BNP elevation was associated with heart failure severity and high ventricular filling pressures. In 54 adults with congenital heart disease, Perlowski et al17 showed a significant correlation of BNP levels with myocardial performance index, an echocardiographic Doppler derived measure of ventricular function dependent on neither geometric shape nor heart rate. This study underscored the ability of both BNP and myocardial performance index in assessing global ventricular function in geometrically complex hearts. Moreover, in a prospective study of 49 patients with adult congenital heart disease, Giannakoulas et al18 found that atrial natriuretic peptide and BNP levels were strong predictors of mortality at mid-term follow-up (median of 7.9 years) and could be used as a simple clinical marker for risk stratification. Taking all of these data together, we are tempted to use a ‘syllogistic approach’. Considering that right ventricular function plays a key role in PAH prognosis (both idiopathic or related to congenital heart disease) and that BNP is a reliable marker of ventricular dysfunction, it appears very reasonable that BNP levels are related to outcome and may reflect the response to targeting therapy. However, we need to be very careful in using a similar approach in medicine. We have already learnt the lesson from the past. For example, inotropic intervention in heart failure, including that based on type 3 phosphodiesterase inhibition, has been shown to deteriorate rather than to improve survival.20 In addition, a variety of confounding factors may lead to increased BNP concentrations in Eisenmenger syndrome. It is well known that BNP levels rise when ventricular hypertrophy develops (which, according to Laplace's law, decreases wall tension), and that ventricular hypertrophy largely varies in different heart diseases and at different stages of the same disease, being initially adaptive and later deleterious. BNP concentrations are also dependent on renal function, which is often impaired in Eisenmenger syndrome patients.
In conclusion, the results of the study of Diller et al suggest that BNP is another piece in the complex puzzle of Eisenmenger syndrome. BNP levels should be periodically measured in Eisenmenger patients because elevated concentrations are associated with increased mortality and thus deserve special medical attention. Finally, treatment of Eisenmenger patients with disease targeted therapies was found to be associated with a significant reduction in BNP concentrations. It is worth remembering that a comprehensive evaluation of such a complex condition requires a multiparametric approach and that clinical decision making should be made face to face by integrating clinical experience, pathophysiology and personal factors.