ann. behav. med. (2013) 45:99–109 DOI 10.1007/s12160-012-9414-1
ORIGINAL ARTICLE
Psychosocial Correlates of Atrial Natriuretic Peptide: A Marker of Vascular Health Patrick Smith, Ph.D. & Martti T. Tuomisto, Ph.D. & James Blumenthal, Ph.D. & Andrew Sherwood, Ph.D. & Lauri Parkkinen, MS & Mika Kähönen, MD & Ilkka Pörsti, MD & Silja Majahalme, MD & Väinö Turjanmaa, MD
Published online: 21 September 2012 # The Society of Behavioral Medicine 2012
Abstract Background Psychosocial factors have been associated with cardiovascular outcomes, but few studies have examined the association between psychosocial function and natriuretic peptides. Purpose The purpose of this study is to determine the predictive value of hostility, anger, and social support in relation to atrial natriuretic peptide (ANP), a marker of vascular health, among middle-aged men. Methods One hundred twenty-one men (mean age 0 39.8 years, SD04.1) underwent assessments of ANP and completed the Cook–Medley Hostility Scale, the Spielberger
State–Trait Anger Scale, and the Interview Schedule for Social Interaction. Results Higher levels of hostility (β00.22 [95 % CI 0.04, 0.40], P00.032) and trait anger (β00.18 [95 % CI 0.01, 0.37], P00.044) were associated with greater ANP levels. In contrast, higher perceived social support was also associated with lower ANP levels, (β0−0.19 [95 % CI −0.05, −0.41], P00.010). Conclusions Psychosocial factors, including hostility, anger, and social support, are associated with varying ANP levels among middle-aged men, independent of cardiovascular and behavioral risk factors.
P. Smith (*) : J. Blumenthal : A. Sherwood Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC, USA e-mail:
[email protected]
Keywords Social support . Hostility . Anger . Atrial natriuretic peptide . Vascular health
M. T. Tuomisto : L. Parkkinen School of Social Sciences and Humanities (Psychology), University of Tampere, Tampere, Finland
Introduction
M. T. Tuomisto : L. Parkkinen Department of Psychiatry, Tampere University Hospital, Tampere, Finland M. Kähönen : I. Pörsti : V. Turjanmaa School of Medicine, University of Tampere, Tampere, Finland M. Kähönen : I. Pörsti : V. Turjanmaa Tampere University Hospital, Tampere, Finland S. Majahalme Appleton Cardiology Thedacare, Appleton, WI, USA
Cardiovascular disease (CVD) is a major public health burden, affecting more than 79 million individuals in the USA alone and one billion individuals worldwide [1]. Increasing evidence over the past three decades has shown that psychosocial factors may influence the risk of CVD and accelerate disease progression among individuals with cardiac disease [2, 3]. “Negative” psychosocial factors, such as hostility, anger, and psychological distress, have all been associated with an increased incidence of CVD [4–7] while “positive” psychosocial factors, such as greater social support, may serve a protective function[3, 6, 8–10]. Several biobehavioral mechanisms have been proposed to explain the relationship between psychological factors and CVD, including inflammation [11, 12], platelet activation [13], and autonomic nervous system dysfunction [14].
100
Despite the extensive evidence linking psychological factors and CVD risk factors, few studies have examined the relationship between psychosocial variables and natriuretic peptides as biomarkers of CVD risk, which may play an important role in the pathogenesis of CVD. Natriuretic peptides participate in the regulation of salt and water homeostasis and play an important role in blood pressure regulation [15] through their vasodilatory and diuretic effects [16]. Both atrial natriuretic peptide (ANP) and brain natriuretic peptides (BNP) have been examined as prognostic markers in cardiovascular disease [17], although BNP has received greater attention in recent years. Natriuretic peptides, which serve as biomarkers of systemic pressure, volume overload, and ventricular wall stress [18–20], have been associated with disease severity among individuals with congestive heart failure (CHF) [21] and are elevated among individuals with essential hypertension [22]. Natriuretic peptides have also been shown to prospectively predict the development of high blood pressure among men participating in the Framingham study and experimental studies among animals have demonstrated that natriuretic peptide expression precedes the development of hypertension and left ventricular dysfunction [20, 23]. In addition, recent studies have shown that natriuretic peptides may be predictive of subclinical vascular health, including vascular stiffness [24, 25] and incident CVD, after accounting for traditional CVD risk factors, suggesting that they may be an important indicator of cardiovascular risk among adults with hypertension [21, 26]. Recent studies have extended these findings, demonstrating that increased natriuretic peptide levels are associated with greater vascular stiffness in the general population [25], independent of background characteristics and cardiac risk factors. ANP is produced by cardiac atrial myocyte cells and has important vasodilatory, diuretic, and natriuretic effects [27] within the cardiovascular system. ANP increases in response to the diastolic stretching of cardiac myocytes [28] and has been used clinically as a diagnostic marker of occult ventricular dysfunction in patients suspected to have CHF [29]. ANP has gained increasing attention as a potential biomarker of cardiovascular health, as it has been shown to predict incident CVD [21], stroke [30], CHF, and atrial fibrillation [26]. In addition to the prognostic characteristics of ANP in clinical samples, population studies have demonstrated that natriuretic peptides are sensitive to subtle changes in left ventricular functioning, particularly among men [31]. For example, a recent study demonstrated that ANP is associated with greater systolic blood pressure, pulse pressure, and hypertension severity among AfricanAmericans after controlling for background factors and multiple indices of cardiovascular health [32]. Several recent studies have shown that psychosocial factors may be associated with ANP levels after accounting for
ann. behav. med. (2013) 45:99–109
traditional markers of cardiovascular risk [33]. In patients with CHF, higher ANP levels were associated with more pronounced levels of depression, as well as lower ejection fraction, New York Heart Association functional class, and physical quality of life [33]. Preliminary evidence also suggests that ANP may be elevated among healthy adults with depression and that ANP recovery following acute bouts of exercise may be attenuated in these individuals, suggesting that depressed mood may impair vascular recovery following exercise [34]. Krogh and colleagues [34] examined this relationship by comparing 132 depressed participants with 44 healthy controls subjects in an experimental paradigm in which ANP release was assessed while participants engaged in acute bicyle exercises. ANP levels were collected before, during, and after exercise. Baseline assessments of ANP obtained during a resting period prior to exercise showed higher ANP levels among unmedicated, depressed participants compared to healthy controls. Interestingly, antidepressant treatment appeared to be associated with lower ANP levels. Consistent with previous findings, ANP increased from baseline to maximum exercise in both groups, but depressed participants showed an attenuated ANP response, which was lower than that observed in health controls. Because age, sex, and body composition were used to match depressed and non-depressed participants in this study, the differential ANP response was not explained by background differences in these factors. Despite these findings, the relationship between natriuretic peptides and other psychosocial characteristics have not been studied, and no studies have examined psychosocial correlates of natriuretic peptides among middle-aged men. Among psychosocial factors associated with CVD, hostility [35], anger [36], and social support [37] have been shown to predict cardiovascular health [9]. Higher levels of hostility have been shown to increase the risk of incident CVD and overall mortality in population-based studies of healthy adults, as well as accelerating CVD progression in cardiac patients [2]. Prospective studies have also shown increased incidence of CVD and angina among individuals with higher levels of anger [38]. Similarly, lower levels of social support are associated with an approximate 60 % increase in the incidence of heart disease, even after accounting for traditional risk factors of CVD risk [39]. Existing evidence suggests that hostility, anger, and social support may interact in predicting CVD outcomes. Higher levels of hostility have been associated with higher blood pressure (BP) as well as decreasing the quality of social interactions [40], suggesting that hostility may moderate the influence of social support on BP. Vella and colleagues [6], for example, found that higher perceived intimacy in social interactions was associated with lower ambulatory BP among individuals with low levels of hostility, but that ambulatory BP actually increased during these
ann. behav. med. (2013) 45:99–109
interactions among individuals with higher levels of hostility, suggesting that hostility moderated the effect of perceived intimacy on BP. Conversely, social support may moderate the hostility, anger, and CVD relationship, with higher levels of social support serving as a buffer for individuals at risk for CVD. For example, Knox and colleagues [41] found that a combination of high hostility and low social support was associated with an increased incidence of heart disease among participants in the Family Heart Study. Similar findings have been suggested for anger, as individuals with greater levels of anger report poorer social support [42] in addition to having an elevated risk of CVD [43]. Angerer and colleagues [44] found that patients with coronary artery disease and low social support who express anger outwardly are at significantly increased risk of disease progression, even after controlling for medical and background factors. In contrast, recent theories suggest that the impact of hostility and anger on cardiovascular outcomes may be partially explained by interpersonal relationships, which are an important determinant of social support [42], suggesting a mediating role. For example, higher levels of hostility may influence blood pressure, a determinant of CVD, by negatively impacting the presence and quality of social relationships [6, 7, 10, 40, 45, 46]. Although hostility, anger, and social support have been linked to poorer cardiovascular outcomes, no studies, to our knowledge, have examined the relationship between these factors and ANP levels. In addition, very few studies have examined the relationship between psychosocial factors and ANP among individuals without CHF. We therefore examined the relationship between hostility, anger, social support, and ANP among 121 middle-aged men with varying levels of blood pressure. We hypothesized that greater levels of hostility and anger would be associated with higher levels of ANP and that higher levels of social support would be associated with lower levels of ANP. We also hypothesized that social support would moderate the influence of hostility and anger on ANP, such that higher hostility and anger would be related to higher ANP only in the presence of low social support. Finally, we also examined whether the relationship between hostility, anger, and ANP was mediated by social support.
Methods Participants Individuals in the present study were participants in a previous experiment examining the relationship between cardiovascular reactivity to behavioral stress among persons with varying BP levels [46, 47], including both normotensive and hypertensive
101
(>140 mmHg systolic blood pressure [SBP] or >90 mmHg diastolic blood pressure [DBP]) men. ANP samples and psychosocial data were collected in the context of this previous experimental study [47]. Participants were recruited from 14,215 men invited to routine health examinations through City of Tampere primary health care from 1987 to 1991. In order to be included, participants had to be healthy on conventional health measures, with the exception on elevated BP, and could not be currently taking medications. The conventional health measures included normal result in physical examination, hematological and biochemical screening tests, chest Xray, and electrocardiogram. In the previous experiment, the participants were classified into normotensive (SBP < 140 mmHg and DBP 160 mmHg or DBP>95 mmHg), or borderline hypertensive (an SBP or DBP between normotensive and hypertensive) groups using World Health Organization (WHO) criteria [48]. Participants then participated in a series of behavioral laboratory interventions to assess changes in blood pressure reactivity to psychological challenges (e.g., cold pressor test, arithmetic task, etc.). Both the mean diagnostic systolic BP and diastolic BP for the whole group of participants were normally distributed in the initial experiment. Consequently, BPs could be treated as continuous variables and the groups as one group. The data for the present analyses were taken from baseline assessments taken before the experimental protocol of the larger study. Of the participants of the health examinations, 2,950 initially met these criteria. The participants were recruited by health care nurses who gave an information leaflet about the study. Those interested could register for the study immediately at the health examination or later by sending the registration form to the researchers. One hundred twenty-one unmedicated male volunteers agreed to participate from three cohorts of ages 35, 40, and 45 years. All volunteers who met the inclusion criteria were included in the study in chronological order. Markers of Cardiovascular Health Atrial Natriuretic Peptide The blood samples for the atrial natriuretic peptide analysis were drawn with other blood samples into chilled tubes containing EDTA (1 mg/ml) and protease inhibitor (Apronin, 500–1,000 KIU/10-ml tube) after the patient had been in a sitting position for at least 10 min at 8 a.m. Whole blood was centrifuged and plasma immediately frozen and stored at −70 ° C. Atrial natriuretic peptide was extracted from plasma using Sep-Pak C18 cartridges (Waters, Millipore). Concentrations of atrial natriuretic peptides were measured by radioimmunoassay (Human alfa-ANP RIA; Amersham). The sensitivity of the assay was 1 pmol/l and the intra- and interassay coefficients of variation
