Author + information
- Received June 8, 2004
- Revision received August 25, 2004
- Accepted September 2, 2004
- Published online January 4, 2005.
- Patrick T. Ellinor, MD, PhD,
- Adrian F. Low, MB, BS,
- Kristen K. Patton, MD,
- Marisa A. Shea, RN and
- Calum A. MacRae, MB, ChB, PhD* ()
- ↵*Reprint requests and correspondence:
Dr. Calum A. MacRae, Cardiovascular Research Center, Massachusetts General Hospital, 149 13th Street, Charlestown, Massachusetts 02129
Objectives We sought to characterize natriuretic peptide levels in a cohort of rigorously characterized subjects with lone atrial fibrillation (AF).
Background Atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) are sensitive biomarkers of cardiac contractile dysfunction. Both peptides have been reported to be elevated in cohorts with AF, but previous studies have included subjects with underlying structural heart disease. We studied these hormones in 150 subjects with lone AF.
Methods Study subjects had electrocardiographic evidence of at least one episode of AF and a structurally normal heart on echocardiography. Subjects were excluded if they had a history of a myocardial infarction, rheumatic heart disease, cardiomyopathy, significant valvular disease, hyperthyroidism, or hypertension that preceded the onset of AF. Control subjects were obtained from a healthy outpatient primary care population. Plasma pro-ANP and N-terminal pro-BNP (nt-pro-BNP) levels were determined using commercially available immunoassays.
Results A total of 150 serial subjects with lone AF were enrolled and studied, the majority during normal sinus rhythm. Median levels of nt-pro-BNP were significantly elevated in subjects with lone AF as compared with control subjects (166 vs. 133 fmol/ml, p = 0.0003). There was no significant difference in pro-ANP levels between subjects with lone AF and control subjects (1,730 vs. 1,625 fmol/ml, p = 0.90).
Conclusions Discordant natriuretic peptide levels were observed in this homogeneous population of subjects with lone AF. This biomarker pattern, which is present even in sinus rhythm, may represent an underlying subclinical predisposition to this common arrhythmia.
Atrial fibrillation (AF) is the most common clinical arrhythmia, as well as an important source of morbidity and mortality (1). The primary pathophysiology of AF is obscured by the secondary effects of the arrhythmia itself (2). Heritable forms of AF suggest that some individuals have an intrinsically lower threshold for developing this arrhythmia (3). Such a diathesis might take the form of pulmonary venous abnormalities (4), subclinical myocardial disease (5), or isolated disturbances of atrial electrophysiology (2).
In searching for biomarkers of a predisposition to AF, we hypothesized that a diathesis toward this arrhythmia might be reflected in abnormalities of the natriuretic peptide (NP) axis, as atrial structure and function are central to both the afferent and efferent limbs of this endocrine pathway (6). The NP hormones are secreted in response to several stimuli, including intracardiac pressure, and have become popular candidates for the noninvasive assessment of cardiac dysfunction (6,7). Previous studies of the NP axis in AF have included subjects with a wide variety of etiologies for their arrhythmia (8–10). To limit the role of potential confounders, we studied pro-atrial natriuretic peptide (ANP) and N-terminal-pro-brain natriuretic peptide (BNP) in subjects with rigorously defined lone AF.
We have identified a significant elevation in plasma nt-pro-BNP levels in subjects with lone AF. This elevation is present when subjects are in sinus rhythm. Interestingly, pro-ANP levels were not elevated, discordant with the nt-pro-BNP results. These data suggest that the combination of elevated nt-pro-BNP and normal pro-ANP levels is associated with lone AF and may prove to be a marker of an underlying predisposition to this arrhythmia.
All studies were performed with Institutional Review Board approval at Massachusetts General Hospital and with written, informed consent from each subject. Individuals were eligible for enrollment if they had at least one documented electrocardiogram (ECG) with AF and a structurally normal heart on echocardiogram. Subjects were excluded if they had a history of myocardial infarction, rheumatic heart disease, cardiomyopathy, significant valvular disease, or hyperthyroidism. Subjects reporting a diagnosis of hypertension within two years of the onset of AF also were excluded.
One hundred and fifty-eight subjects with lone AF were enrolled between July 5, 2001, and May 14, 2003. Seventy-five control subjects, matched for age, gender, and ethnicity, were selected from a healthy primary care population (from the Global Repository, Genomics Collaborative, Inc., Cambridge, Massachusetts).
Each subject underwent a physical examination and a structured interview to elicit a detailed medical history, medications, symptoms, and possible triggers for AF. The medical history of all first-degree relatives was obtained using a standardized questionnaire. All subjects with lone AF had an ECG and echocardiogram within 90 days of enrollment. Echocardiography included a comprehensive two-dimensional, M-mode, and Doppler evaluation; ejection fraction was estimated using the modified Quinones method.
Blood sampling and natriuretic hormone assays
Blood samples were obtained in EDTA from each subject at enrollment. Samples were centrifuged, and plasma was extracted, aliquoted, and stored at −80°C until analysis. Plasma pro-ANP and nt-pro-BNP levels were determined using commercially available enzyme immunoassays without extraction (manufactured by Biomedica Gruppe, Vienna, Austria), according to the manufacturer's instructions, with a Victor-3 plate reader (Perkin-Elmer, Wellesley, Massachusetts).
The immunoassay for nt-pro-BNP employs an immunoaffinity-purified sheep antibody specific for nt-pro-BNP (amino acids 8–29). The pro-ANP assay consists of a capture antibody specific for pro-ANP (amino acids 10–19) and a detection antibody specific for pro-ANP (amino acids 85–90). For each assay, the cross reactivity with other NP epitopes is <1%. Assays were performed in duplicate in a single run and normalized to a standard curve. Intra-assay and inter-assay variances for pro-ANP and nt-pro-BNP were all ≤7%.
Baseline and transformed values of pro-ANP and nt-pro-BNP were highly skewed in patients and control subjects. The NP levels were therefore compared in patients and control subjects by using the Wilcoxon rank-sum test. Normally distributed values are displayed as mean values with standard deviations, whereas for pro-ANP and nt-pro-BNP, the median values are provided. Data were analyzed using Intercooled Stata 8.0 (Stata Corp., College Station, Texas).
During the study period, 158 subjects with lone AF were enrolled. Eight subjects who developed structural heart disease after the onset of AF but before study enrollment were excluded from analysis. Seventy-five control subjects were matched to subjects with lone AF based on age, gender, race, and ethnicity. Body mass index and systolic and diastolic blood pressures were similar between patients and control subjects (Table 1).
The mean age at diagnosis with AF was 44.8 ± 11.9 years, and the mean age at enrollment was 53.8 ± 10.7 years. As observed in other cohorts with lone AF, subjects were predominately male (80.8%) (11). At the time of enrollment, 130 subjects had paroxysmal AF and 20 subjects were in permanent AF (Table 2).Of those subjects with paroxysmal AF, 111 (85%) were in sinus rhythm, 8 (6%) were in AF, 7 (5%) were in atrial flutter, and the remaining 4 (3%) were either paced or in another rhythm. Subjects with lone AF were receiving a variety of medications for management of AF, whereas control subjects were not taking any cardiovascular medications.
Although 23% of subjects in sinus rhythm had evidence of left atrial enlargement on the ECG, there were no other electrocardiographic abnormalities. Echocardiography was notable for a mean left atrial diameter at the upper limits of normal. The mean values of all other echocardiographic parameters, including chamber dimensions, wall thicknesses, and functional indexes, were normal in the study cohort. Control subjects had no significant medical history.
Natriuretic hormone levels
Median nt-pro-BNP levels were significantly elevated in subjects with lone AF, as compared with matched control subjects (166 vs. 133 fmol/ml, p = 0.0003) (Fig. 1A).The nt-pro-BNP levels in those with permanent AF were also significantly higher than those in subjects whose arrhythmia remained paroxysmal (189 vs. 157 fmol/ml, p = 0.0015) (Fig. 2A).Among subjects who presented with paroxysmal AF, nt-pro-BNP levels were higher in those individuals who progressed from paroxysmal to permanent AF (197 vs. 163 fmol/ml, p = 0.05).
There was no significant difference in pro-ANP levels between patients with lone AF and healthy control subjects (1,730 vs. 1,625 fmol/ml, p = 0.90) (Figs. 1B and 2B). However, pro-ANP levels were significantly higher in subjects who developed hypertension after the onset of AF (3,764 vs. 1,622 fmol/ml, p = 0.01). A negative correlation was noted between pro-ANP and diastolic blood pressure (r = −0.22, p = 0.01).
A discordant pattern of nt-pro-BNP elevation (p = 0.004) with normal pro-ANP levels (p = 0.57) was present, even when all subjects in AF or in paced rhythm at the time of measurement were excluded (Figs. 2A and 2B). The results were also unaffected by the exclusion of those subjects taking beta-blockers (data not shown).
Natriuretic peptides in structural heart disease
The natriuretic hormone axis is the best characterized cardiac endocrine pathway. This family of short peptides share several physiologic actions, including natriuresis, vasodilation, and the modulation of central and peripheral baroreflexes (6). Each active peptide is cleaved from the C-terminus of a longer precursor, and although the biologic roles of the N-terminal fragments are unknown, the kinetics of these pro-peptides make them more suitable for use as biomarkers (12). Both ANP and BNP are coordinately regulated in a wide range of clinical conditions associated with abnormalities of systolic or diastolic function. Elevated circulating levels of these peptides have been correlated with symptoms and adverse clinical outcomes in congestive heart failure (7). The utility of the NPs as a screening tool for left ventricular function (13), a diagnostic adjunct in the evaluation of dyspnea, or as a therapeutic end point in heart failure, is under investigation (7).
Natriuretic peptides in AF
Elevated levels of both ANP and BNP have been described in several different AF cohorts. In congestive heart failure, where NP levels are already markedly elevated, the onset of AF is associated with a further increase (8,14). ANP and BNP each have been reported as elevated both in paroxysmal and permanent AF (6,8). Although in several studies, NP levels decreased after return to sinus rhythm, there also seems to be a slow attenuation of the initial elevation of ANP over time with sustained AF (15). This pattern of release is consistent with the known response characteristics of ANP, which is only transiently elevated after acute stimuli but is persistently elevated with chronic stimuli. This is thought to be a result of rapid release from a finite pool of storage vesicles in the acute setting, whereas chronic disease states induce increased messenger ribonucleic acid and peptide synthesis expanding the storage pool (6). The late attenuation in permanent AF has been attributed to the trophic consequences of persistent arrhythmia, but, so far, there are few data on the nature of such endocrine remodeling (16). All of these observations are potentially confounded by the known effects of structural heart disease on the NPs. Atrial fibrillation is a highly heterogeneous condition, and it has proven difficult to dissociate the effects of the arrhythmia from those of the underlying disorders (1).
Approximately 20% to 30% of AF occurs in the absence of any objective evidence of structural heart disease and is known as lone AF (11). To avoid the confounding effects of underlying structural heart disease, we compared the levels of pro-ANP and nt-pro-BNP in subjects with rigorously defined lone AF and in healthy control subjects. Our data suggest that lone AF, unlike other forms of the arrhythmia, is associated with a unique pattern of ANP and BNP release. In contrast to previous studies of the NPs, we identified discordant regulation of the BNP and ANP axes, with a significant elevation of nt-pro-BNP in the presence of normal pro-ANP levels in subjects with a history of lone AF. These findings were unrelated to the current rhythm of the study subjects (Fig. 2A).
Possible mechanisms of discordant regulation of NPs
Release of ANP and BNP may be uncoupled in some experimental situations. Hypertrophic stimuli, including endothelin-1 and phenylephrine, differentially induce ANP and BNP in in vitro models (17), but whether these stimuli result in abnormal circulating NP levels before any hemodynamic perturbation is not known. Secretion of ANP and BNP is distinctively regulated in different tissues, and the potential contributions of various sources to circulating plasma levels under pathologic conditions may vary (6). Finally, the NP pathways intersect at many levels with other neuroendocrine, paracrine, or autocrine pathways, whose cardiac effects are less well defined and may have differential effects on ANP and BNP (6). It remains to be seen whether the source of BNP in subjects with lone AF is atrial, ventricular, or extracardiac.
Although the fundamental mechanisms of isolated BNP elevation remain uncertain, recent population data have identified BNP as a predictor of specific cardiovascular end points. Levels of nt-pro-BNP may predict the subsequent development of AF, hypertension, stroke, and congestive heart failure (18); however, the temporal relationship between these disorders varies, raising the possibility that these phenotypes represent distinct manifestations of a common etiology characterized by abnormal BNP levels. Our data support this hypothesis. Theoretically, dysregulation of the NP axis may be upstream of AF, as BNP is known to potentiate vagal efferent activity, thus sustaining fibrillatory rotors (19). Brain natriuretic peptide is also coupled to several ionic currents, but there is no evidence of increased automaticity with this peptide, as there is with ANP (6). A more likely unifying hypothesis is that elevated BNP is simply a marker for a subset of AF patients with a primary disorder predisposing not only to the maintenance of AF, but also to subsequent hypertension, stroke, and heart failure. Detailed longitudinal investigation of subjects with lone AF will help clarify these potential relationships and explore the precise mechanisms for uncoupling ANP and BNP.
The current study is based in a tertiary-care arrhythmia clinic and may be prone to referral bias. The cohort demographics are comparable to those of previous lone AF studies, and these individuals are of largely Northern European descent. It will be important to replicate our findings in ethnically, racially, and geographically distinct populations.
We have identified a particular pattern of discordant ANP and BNP levels that is associated with lone AF. This discordance may reflect the differential sensitivities of the ANP and BNP pathways to the primary process in AF. Further work will be required to dissect the link between abnormal expression and secretion of these peptides and the risk of AF. The definition of specific biomarker profiles, possibly incorporating the NPs, will be an important step in dissecting the etiology of AF. These findings emphasize the need for detailed clinical investigation in the evaluation of emerging biomarkers.
The authors thank Drs. David J. Milan, Laura Mauri, and Tommy J. Wang for their insightful suggestions on the manuscript. We also thank Drs. Jeremy N. Ruskin and G. William Dec for their continued support.
This work was supported by grants from the National Institutes of Health (HL-71632) and the Smith Family Foundation (to Dr. Ellinor) and the National Medical Research Council of Singapore (to Dr. Low).
- Abbreviations and acronyms
- atrial fibrillation
- atrial natriuretic peptide
- brain natriuretic peptide
- natriuretic peptide
- Received June 8, 2004.
- Revision received August 25, 2004.
- Accepted September 2, 2004.
- American College of Cardiology Foundation
- Chugh S.S.,
- Blackshear J.L.,
- Shen W.K.,
- Hammill S.C.,
- Gersh B.J.
- Darbar D.,
- Herron K.J.,
- Ballew J.D.,
- et al.
- Frustaci A.,
- Chimenti C.,
- Bellocci F.,
- Morgante E.,
- Russo M.A.,
- Maseri A.
- Rossi A.,
- Enriquez-Sarano M.,
- Burnett J.C. Jr.,
- Lerman A.,
- Abel M.D.,
- Seward J.B.
- Clerico A.,
- Del Ry S.,
- Giannessi D.
- Wozakowska-Kaplon B.,
- Opolski G.
- Wang T.J.,
- Larson M.G.,
- Levy D.,
- et al.
- Herring N.,
- Zaman J.A.,
- Paterson D.J.