![Hypothetical model of neural systems underlying the association between hypertension and panic disorder. According to the model, deficiencies in inhibitory control, either local GABAergic inhibitory mechanisms within the dorsomedial hypothalamus (DMH), or serotonergic inhibitory mechanisms, acting within the dorsal periaqueductal gray (DPAG; 5-HT1A/5-HT2 receptors) or rostral ventrolateral medulla (RVLM; 5-HT1A receptors), would result in vulnerability to both hypertension and panic disorder. Serotonergic neurons in both the ventrolateral part of the dorsal raphe nucleus (DRVL)/VLPAG region and the medullary raphe pallidum (RPa) are directly excited by the panicogenic agent CO2 or decreases in extracellular pH. Normally, this mechanism would serve as a negative feedback system, with increasing concentrations of CO2 or decreasing pH-activating serotonergic neuronal firing rates, preventing an overactivation of both the behavioural and autonomic symptoms of panic. If these serotonergic neurons are compromised, either by changes in their intrinsic properties or changes in the neural input regulating their activity (e.g. changes in executive function in the prefrontal cortex, PFC), behavioural and autonomic responses would continue unchecked. Chronic, reduced activity of serotonergic neurons in the DRVL/VLPAG region would be expected to lead to vulnerability to both hypertension and the behavioural and autonomic symptoms of panic. Chronic, reduced activity of serotonergic neurons in the RPa region would be expected to lead to vulnerability to both hypertension and the autonomic, but not behavioural, symptoms of panic. Indeed tryptophan depletion can exacerbate panic symptoms. In contrast, SSRIs, possibly by increasing serotonergic neurotransmission in these systems, can alleviate panic symptoms. C1 = C1-adrenergic cell group. Reproduced from Davies et al. [29].](https://www.researchgate.net/profile/Simon-Davies-6/publication/265692960/figure/fig1/AS:295429494329355@1447447288536/Hypothetical-model-of-neural-systems-underlying-the-association-between-hypertension-and_Q320.jpg)
Content uploaded by Simon Davies
Author content
All content in this area was uploaded by Simon Davies on Jun 22, 2015
Content may be subject to copyright.
Content uploaded by Simon Davies
Author content
All content in this area was uploaded by Simon Davies on Jun 22, 2015
Content may be subject to copyright.
Content uploaded by Simon Davies
Author content
All content in this area was uploaded by Simon Davies on Jun 22, 2015
Content may be subject to copyright.
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
Abstract
This chapter examines the association of anxiety disorders and anxiety symptoms with cardiovascu-
lar disease, focussing on hypertension (an independent risk factor for myocardial infarction and
stroke) and coronary heart disease. In both cases, epidemiological data linking the cardiovascular
disorder with specific anxiety disorders and anxiety symptoms are examined first, and evidence re-
lating to putative mechanisms that may underlie these associations is explored. For hypertension,
an association with panic attacks and panic disorder has been reported most consistently, but the
literature relating to other forms of anxiety is inconsistent, especially as some studies have reported
an association of anxiety with low blood pressure. Recent work which has attempted to elucidate
this confusing situation is presented. Mechanisms which may be responsible for the link between
hypertension and panic include autonomic nervous system dysfunction (which may be under sero-
tonergic control), respiratory mechanisms, cytokines, platelet dysfunction and behavioural factors.
While an association of depression with coronary heart disease has been studied extensively, the
association with anxiety disorders has been slower to emerge. Studies contributing to this evidence
base are examined, and as for hypertension putative mechanisms are discussed.
Copyright © 2013 S. Karger AG, Basel
In this chapter, the associations of anxiety and anxiety disorders with cardiovascular
disease will be examined. As the bulk of the evidence relates to hypertension and cor-
onary heart disease (CHD), the focus of this paper will be on these common cardio-
vascular problems.
Anxiety and Hypertension
Hypertension is a common disorder with increasing prevalence through the lifespan.
It is a prognostic factor for myocardial infarction, heart failure and stroke [1], but this
risk is modifiable by antihypertensive drugs. The term ‘essential hypertension’ has
Anxiety and Cardiovascular Disease
Simon J.C. Davies
a, b ⋅ Christer Allgulander
c
aGeriatric Psychiatry Division, Centre for Addiction and Mental Health, Department of Psychiatry,
University of Toronto, Toronto, Ont., Canada; bAcademic Unit of Psychiatry, University of Bristol, Bristol, UK;
cSection of Psychiatry, Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
86 Davies · Allgulander
traditionally been applied to high blood pressure (BP) of unknown aetiology as it was
thought that hypertension was commonly a compensatory mechanism to ensure that
adequate perfusion was maintained when arteries became sclerosed [2]. However,
hypertension previously considered ‘essential’ can often now be linked to pathologies
including insulin resistance, salt sensitivity, sleep apnoea and dysfunction of the sym-
pathetic nervous system [3, 4].
Association of Anxiety Disorders with Hypertension
Clinical observation suggests that panic attacks are associated with transient increas-
es in BP [5]. Controlled studies have reported higher rates of panic attacks or panic
disorder with hypertension [6, 7] but had methodological limitations, such as failure
to adjust for confounders other than age and sex.
Meanwhile, rather different conclusions have been drawn about the association be-
tween hypertension and generalized anxiety disorder (GAD) or non-specific measures
of anxiety. In studies employing hypertension as a dichotomous end point, all kinds of
relationships have been reported. For instance, hypertension as a category was associ-
ated with both panic disorder and GAD in a recent study from South Africa [8], and
with anxiety disorders distinct from panic in further studies [9, 10], but in others no
association with anxiety [11, 12] was found. In contrast, large population-based studies,
using continuous BP measures rather than a hypertension diagnosis, have reported as-
sociations between low BP and general psychological morbidity and related symptoms
[13, 14]. These latter findings are in line with clinical practice in some countries where
a syndrome of low BP, tiredness and (generalized) anxiety is well recognized [15].
To summarise, measures of anxiety have been associated both with low and high
BP; and among all the anxiety symptoms and disorders, panic attacks/panic disorder
have most consistently been associated with hypertension. To reconcile these diverse
findings, we have employed data from the population-based HUNT study where all
92,936 individuals aged 20 or more residing in one Norwegian county were invited to
participate and just under 65,000 had BP recorded. We examined the hypothesis that
there is a non-linear relationship between systolic BP and panic and an association of
low BP with generalized anxiety [16]. Both unadjusted (n = 61,408) and adjusted
analyses provided evidence for a non-linear relationship between panic and systolic
BP, represented by a ‘U’-shaped curve with a minimum prevalence of panic around
140 mm Hg (fig.1). The relationship was strengthened after adjustment for multiple
confounders with the quadratic term significantly associated with panic (p = 0.03).
Generalized anxiety symptoms were associated only with low systolic BP (fig.1). The
‘U’-shaped relation between systolic BP and panic provides a unifying explanation for
the separate strands of published literature in this area, as high BP appears to be as-
sociated only with panic symptoms, while several forms of anxiety including panic
and generalized anxiety may be associated with low BP.
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
Anxiety and Cardiovascular Disease 87
Mechanisms in the Association of Hypertension and Panic Attacks/Panic Disorder
What mechanisms might explain why panic attacks and panic disorder, but not other
anxiety disorders, are associated with high BP? A prime candidate is the autonomic ner-
vous system. A study examining panic attack symptoms [17] illustrated that only the
typical autonomic symptoms of sweating and flushes are significantly more common in
attacks reported by hypertensive patients compared with normotensives, and on factor
analysis only the autonomic dominated factor (comprising sweating, flushing, shaking
and the non-specific symptom of nausea) was associated significantly with hypertension.
At the neurochemical level, central or peripheral catecholamine dysfunction has
been described in both panic disorder and hypertension. Despite the majority of cases
of hypertension being classified as ‘essential hypertension’, there has been an acknowl-
edgement that in many cases a dysfunction of the autonomic nervous system may be
the underlying pathology [18]. Evidence from the studies by Esler et al. [19] of clinical
microneurography and direct measurement of noradrenaline spillover from cardiac
nerve terminals suggests sympathetic dysfunction in hypertensives. ‘Spillover’ is the
0
0.005
0.010
0.015
0.020
0.025
Probability of frequent panic
80 100 120 140 160 180 200
Systolic BP (mm Hg)
a
Probalility of HADS generalized
anxiety items total score in
uppermost quartile
0
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
80 100 120 140 160 180 200
Systolic BP (mm Hg)
b
Fig. 1. a Probability of fre-
quent panic by systolic BP af-
ter adjustment for potential
confounding variables (n =
43,974). b Probability of Hos-
pital Anxiety and Depression
Scale-derived generalized
anxiety symptoms (score in
uppermost quartile of sample)
by systolic BP after adjust-
ment for potential confound-
ing variables (n = 39,755). Dot-
ted lines indicate 95% confi-
dence limits. Reproduced
from Davies et al. [16].
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
88 Davies · Allgulander
overflow of a substance to an organ’s venous drainage. Catheterisation, in this case of
the coronary sinus, is required, and catecholamine overflow is measured through an
isotope dilution technique with constant infusion of radiolabelled noradrenaline. Sim-
ilarly, excess adrenaline spillover from the heart has been reported during panic attacks
[20]. There is also evidence of abnormal central catecholamine function in both disor-
ders. Nutt [21] reported altered central α2-adrenoceptor sensitivity in panic disorder,
and excess catecholamine spillover from the brain has been seen in hypertension [22].
There has been interest as to whether serotonin systems and their dysfunction may
contribute to autonomic nervous system dysfunction and thereby to the association
of hypertension and panic disorder. Serotonin-promoting antidepressants are first-
line treatments in anxiety disorders, and transient depletion of serotonin by acute
tryptophan depletion renders treated patients with a history of panic disorder more
vulnerable to panic on stress challenge [23]. Polyak has reported that selective sero-
tonin reuptake inhibitor (SSRI) antidepressants can assist BP control in hypertensive
patients with co-morbid panic [24]. After 3–6 months’ drug treatment, patients with
co-morbid panic disorder and mild hypertension experienced more pronounced BP
reduction if on treatment with the SSRI fluoxetine than those treated with the antihy-
pertensive moxonidine. Reduced heart rate variability has been reported in panic dis-
order [25, 26] which can be rectified by SSRIs [27].
Thus, lowering serotonin concentrations using the acute tryptophan depletion tech-
nique should alter both cardiovascular and psychological parameters relevant to these
conditions. In patients with treated panic disorder or social anxiety disorder, acute tryp-
tophan depletion left participants vulnerable to significantly greater BP and psychologi-
cal responses to stress challenges than that seen under nondepleted conditions [28]. Se-
rotonin may have an anti-stress role in both psychological and cardiovascular domains.
We have previously constructed a model to illustrate possible neuroanatomical
pathways that may be involved in the association of panic and hypertension through
common autonomic dysfunction, illustrating the possibility that failure of serotonin-
modulated control mechanisms may play a role [29]. The model (fig.2) highlights
serotonin-dependent pathways in the ventrolateral periaqueductal gray (VLPAG)
and raphe pallidum which may exert control over the C1 cells of the rostral ventrolat-
eral medulla. The C1 cells are in turn responsible for sympathetic activation. Seroto-
nergic control of the VLPAG may impinge also on behavioural symptoms of panic,
mediated through the dorsal periaqueductal gray.
Other Putative Mechanisms
Although some evidence of cytokine disturbance in panic disorder exists (involving
reports of increased interleukin-1β [30] and IL-2 plasma concentrations [31] in pan-
ic disorder compared with controls), overall the evidence is as yet insufficient to un-
derpin an association with hypertension.
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
Anxiety and Cardiovascular Disease 89
Among other biological mechanisms postulated to underlie the association of hy-
pertension and panic disorder, respiratory mechanisms have attracted interest. Hy-
perventilation is a prominent component of panic attacks, and acutely has a signifi-
cant but short-lived pressor effect, averaging 9/8 mm Hg in normotensive subjects
[32]. Klein [33] divided panic attacks into two subtypes, the first due to ‘false suffoca-
tion alarms’ and characterized by panic with a predominance of respiratory symp-
toms and the second group attributable to sympathetic nervous system or hypothal-
amo-pituitary-adrenal axis deficits. However, in the factor analysis of panic attack
Stress
Sympathetic activation
• BP↑
• Panic
(Autonomic symptoms)
Panic
(Behavioural symptoms)
CO2
CO2
+
+
+
+
+
+
SSRIs Trp depletion
SSRIs Trp depletion
–
–5-HT1A
–
–
–
DPA G
PFC
DMH
RPa
C1 cells
RVLM
VLPAG
5-HT1A
5-HT1A/5-HT2
5-HT Neurones
5-HT Receptors
Fig. 2. Hypothetical model of neural systems underlying the association between hypertension and
panic disorder. According to the model, deficiencies in inhibitory control, either local GABAergic in-
hibitory mechanisms within the dorsomedial hypothalamus (DMH), or serotonergic inhibitory
mechanisms, acting within the dorsal periaqueductal gray (DPAG; 5-HT1A/5-HT2 receptors) or rostral
ventrolateral medulla (RVLM; 5-HT1A receptors), would result in vulnerability to both hypertension
and panic disorder. Serotonergic neurons in both the ventrolateral part of the dorsal raphe nucleus
(DRVL)/VLPAG region and the medullary raphe pallidum (RPa) are directly excited by the panico-
genic agent CO2 or decreases in extracellular pH. Normally, this mechanism would serve as a nega-
tive feedback system, with increasing concentrations of CO2 or decreasing pH-activating serotoner-
gic neuronal firing rates, preventing an overactivation of both the behavioural and autonomic symp-
toms of panic. If these serotonergic neurons are compromised, either by changes in their intrinsic
properties or changes in the neural input regulating their activity (e.g. changes in executive function
in the prefrontal cortex, PFC), behavioural and autonomic responses would continue unchecked.
Chronic, reduced activity of serotonergic neurons in the DRVL/VLPAG region would be expected to
lead to vulnerability to both hypertension and the behavioural and autonomic symptoms of panic.
Chronic, reduced activity of serotonergic neurons in the RPa region would be expected to lead to
vulnerability to both hypertension and the autonomic, but not behavioural, symptoms of panic. In-
deed tryptophan depletion can exacerbate panic symptoms. In contrast, SSRIs, possibly by increas-
ing serotonergic neurotransmission in these systems, can alleviate panic symptoms. C1 = C1-adren-
ergic cell group. Reproduced from Davies et al. [29].
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
90 Davies · Allgulander
symptoms in hypertensive and normotensive patients described earlier [17], the fac-
tor significantly associated with hypertension was that comprising symptoms typical
of sympathetic nervous system dysfunction, whereas respiratory panic symptoms
were no more common in hypertensives than normotensives, and the factor domi-
nated by respiratory symptoms had no association with hypertension.
It remains possible that the association of hypertension with panic attacks and
panic disorder is due to illness behaviour, so-called ‘white coat responses’ or labelling
effects. Psychological symptoms may impair the ability of patients both to tolerate or
adhere to medication regimes and to follow interventions that reduce cardiovascular
risk after myocardial infarction. Panic attacks, anxiety and depression are associated
with episodes of intolerance to antihypertensive agents [34] provided the intoleranc-
es reported are not typical of the drugs implicated (non-specific intolerance). The
number of episodes of non-specific intolerance was significantly associated with poor
outcome in BP control [34].
It has been suggested that reported associations of hypertension and panic disorder
may be that patients with panic disorder appear artefactually to have higher BPs due
to a greater ‘white coat response’ (i.e. medical setting anxiety-induced hypertension)
response compared with patients without panic disorder. Patients who are prone to
panic attacks may perceive a primary care facility or hospital clinic as threatening, and
could have a pressor effect as a conditioned response to these situations [35]. How-
ever in an earlier clinical study, we found no excess ‘white coat effect’ in patients with
panic disorder and panic attacks making this explanation unlikely [36].
Finally, it is conceivable that the association of panic disorder and hypertension
might be due at least in part to a ‘labelling effect’. Patients’ awareness of a diagnosis of
hypertension may lead to subsequent adverse effects on psychological well-being [37]
and to vulnerability to the development of panic disorder. Indeed, in the one study
which examined the temporal relationship of the onset of panic attacks and hyperten-
sion [7], the diagnosis of hypertension preceded panic attacks significantly more often
than vice versa (p < 0.01).
Anxiety and Coronary Heart Disease
CHD includes ischaemic heart disease, angina and cardiac events such as myocardial
infarction. Depression has been recognized as an independent risk factor for CHD in
numerous prospective studies and several meta-analyses [38, 39]. Some studies have
questioned the strength of the relation. A recent systematic review of 48 studies con-
cluded that while the majority of individual studies did report an association and that
there was an association overall, attention was drawn to methodological issues in
many papers, especially in that 85% of studies had not controlled for any measure of
anxiety despite the high rates of co-morbidity between anxiety and depression [40].
This left most studies open to the possibility of residual confounding through the
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
Anxiety and Cardiovascular Disease 91
presence of anxiety symptoms assuming that the association of anxiety and CHD is
itself important.
The evidence for an association of CHD with anxiety disorders has been slower to
emerge. The abundance of literature relating to depression compared with that relat-
ing to anxiety disorders in this area might be seen as indicating that depression is in-
deed more strongly associated with CHD than are anxiety disorders. However, there
are alternative explanations for this focus on depression. Research collaborations de-
signing prospective studies with complex medical outcomes may have been more im-
mediately familiar with depression than anxiety disorders (and therefore included
measures of depression in preference to those of anxiety) or considered anxiety as
insufficiently distinct from depression to merit investigation. The multiplicity of anx-
iety symptoms, existence of several different anxiety disorders, and diagnoses which
have changed markedly in their criteria over time may be further disincentives to
study anxiety in addition to depression as a cardiovascular risk factor.
A meta-analysis [41] has brought together prospective studies examining the link
between anxiety or anxiety disorders and subtypes of CHD. Roest et al. [41] identified
20 studies with varying types of anxiety or anxiety disorder as baseline risk factors and
end points related to CHD as the outcome. Illustrating the heterogeneous nature of
these studies with regard to the type of anxiety considered, one study examined the
impact of panic attacks, one post-traumatic symptoms and a further 2 GAD or its core
symptoms of excess worry. A further 4 employed the ‘phobic anxiety’ subscale of the
Crown-Crisp Experiential Index, the remaining 12 employed end points described as
‘general’ measures of anxiety (as opposed to GAD) using scales which were not de-
signed to differentiate between specific types of anxiety or anxiety disorder. Many of
these ‘general’ scales contain items which predominantly describe symptoms of gen-
eralized anxiety, with questions on episodic panic attacks and the autonomic activa-
tion symptoms typically experienced in panic underrepresented. Overall, anxious
persons were at a significantly higher risk of incident CHD independent of demo-
graphic variables, biological risk factors and health behaviours (hazard ratio 1.21; 95%
CI: 1.15–1.38, p < 0.0001) in a sample of around 250,000 people with a mean follow-
up of 11.2 years.
What cannot be established from the existing evidence base of prospective studies
is whether there is a differential risk of CHD between different types of anxiety. No-
tably, the single study which had panic attacks as the exposure [42] produced the larg-
est hazard ratio of 4.20 (95% CI: 1.76–10.00). This study was undertaken in a popula-
tion of 3,369 post-menopausal women for an outcome of cardiac events (fatal and
nonfatal myocardial infarction and stroke) over a 5.3-year follow-up period. Other
studies not included in the meta-analysis of Roest et al. [41] may shed further light on
the potential importance of panic and panic disorder. Using a managed care database
of 78,000 patients, a 2-fold increased risk for CHD was reported in patients with pan-
ic disorder (based on the ICD-9 diagnostic system), independent of the presence of
major depressive disorder [43]. A large cohort study based on the British General
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
92 Davies · Allgulander
Practice Research database reported that panic attacks/panic disorder conferred a sig-
nificantly increased risk of myocardial infarction if panic onset was before age 50
years, and elevated risk of CHD at all ages [44], while a Taiwanese longitudinal study
reported an increased risk of myocardial infarction in panic disorder of around 1.75-
fold which was independent of hypertension [45]. Most recently, a prospective study
using data from a well-known cohort of young Swedish men examined for military
service, which included a baseline assessment for ‘anxiety neurosis’ has been pub-
lished [46]. This diagnostic forerunner of panic disorder and GAD containing some
aspects of both modern-day diagnoses appeared in ICD-8 which was the predominant
diagnostic system at the time the cohort was established. Over an impressive follow-
up period of 37 years for almost 50,000 participants, anxiety neurosis conferred a haz-
ard ratio of 2.17 (95% CI: 1.28–3.67) for CHD. Finally, strong associations of panic or
anxiety neurosis with CHD and cardiovascular mortality are supported by cross-sec-
tional data in smaller studies [47, 48] and data derived from the Epidemiological
Catchment Area study [6]. A significant excess of cardiovascular conditions including
history of ‘heart attack’ which had an odds ratio of 4.54 in panic disorder compared
to subjects with no psychiatric illness was reported.
Roest cited four studies which examined the association of CHD with ‘phobic anx-
iety’ symptoms by a subscale of the Crown Crisp Experiential Index. This question-
naire [49] dates from the mid-1960s and incorporates eight items, four of which refer
to the concept of agoraphobia, two to specific phobias (claustrophobia and height
phobia) and two to worries about family members and about health which might now
be considered to refer to generalized anxiety symptoms or hypochondriasis. In one
study, phobic anxiety scores of 5 or more had relative risk of 3.77 for fatal CHD (95%
CI: 1.64–8.64) compared to those scoring 0 or 1 on the subscale in white males fol-
lowed up for an average of 6.7 years [50]. In similar studies [51, 52] of men free of
CHD at baseline, phobic anxiety was associated with excess CHD mortality, which
was entirely due to excess sudden deaths. A further study by the same authors [53]
showed an association between baseline phobic anxiety, and increased risk of sudden
cardiac death over 32 years of follow-up. In another prospective study, Watkins et al.
[54] reported associations of phobic anxiety and depressive symptomatology with
ventricular arrhythmias. Fleet and Beitman’s review [55] concluded that panic and
phobic anxiety taken together, which he termed ‘panic-like anxiety’ appeared to be an
independent risk factor for cardiovascular death. However, the heterogeneous nature
of the Crown Crisp Experiential Index phobic anxiety subscale – with half of the mea-
sure being made up of items approximating to agoraphobia, which is known to be
closely linked to panic disorder – leaves questions to be answered as to whether the
anxiety associated with phobias carries a similarly elevated risk of CHD and its con-
sequences as described above with panic disorder.
In contrast, a number of studies have examined the association of cardiovascular
disease with measures of anxiety more closely related to GAD. A meta-analysis of
anxiety as a risk factor for cardiovascular disease in studies up to 2003 concluded that
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
Anxiety and Cardiovascular Disease 93
the evidence for generalized anxiety as a cardiovascular risk factor was relatively
sparse [56], with several studies finding no association. In total, Roest et al. [41] cited
14 studies in which the anxiety measure at baseline was either GAD, its core symptom,
worry, or a ‘general’ measure of anxiety made up predominantly of GAD symptoms
(e.g. HADS-A, Spielberger Trait Inventrory, EDS Anxiety Subscale). Around half of
these studies, including 6 of the 9 published since 2006 reported a significant associa-
tion of the GAD-related measure with incident CHD. Although Roest’s group did not
pool these studies for comparison with the effect size of the single panic study, there
is some suggestion that generalized anxiety symptoms and disorder may be a weaker
risk factor for CHD than panic, mirroring the situation described earlier with panic
disorder. However, only a larger prospective study with robust ascertainment of pan-
ic disorder diagnoses at baseline will allow this to be ascertained with a greater degree
of certainty.
Finally, a recent study [57] highlights an interesting paradox relating to the impact
of GAD on cardiovascular outcome. In a sample of patients with acute coronary syn-
drome followed over 5 years, GAD at baseline was associated with significantly better
cardiovascular outcomes after adjustment for depression and medical comorbidity.
The authors suggested that GAD may increase the likelihood of seeking treatment and
of adhering to cardiac rehabilitation, and described this phenomenon as ‘constructive
worrying’.
Mechanisms in the Association of Coronary Heart Disease and Anxiety
Just as the epidemiological evidence for the association with anxiety and CHD has
lagged behind that for depression, elucidation of the biological causes of an asso-
ciation is not yet well developed. Mechanisms which have been put forward to ex-
plain why depression may be an independent risk factor for CHD include auto-
nomic nervous system dysfunction, inflammatory mediators, platelet dysfunction
and atherosclerosis. Some of these ideas have been applied to the arena of anxiety
and CHD.
We have discussed mechanisms relating to autonomic dysfunction in examining
the association of panic and hypertension. Many of the same concepts have been in-
voked to explain an association of CHD primarily with panic disorder. Of course, this
association may merely be based on hypertension being on the causal pathway as an
important risk factor for CHD, but some markers of autonomic dysfunction are ab-
normal in the presence of CHD irrespective of hypertension, for example reduced
heart rate variability [58]. As with hypertension, serotonin may have a role in moder-
ating autonomic nervous system dysfunction in ischaemic heart disease. The
SADHART study [59] reported a trend towards SSRI treatment reducing further isch-
aemic heart disease-related morbidity in people who had experienced myocardial in-
farction or unstable angina, while in depressed patients who had experienced stroke
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
94 Davies · Allgulander
cardiovascular outcomes were significantly better in patients randomized to an SSRI
[60]. It is possible that the association could act through anxiety disorders causing
cardiac arrhythmias, and increased serotonin availability is known to reduce vulner-
ability to ventricular fibrillation in the cat [61].
There is evidence of changes in concentrations of inflammatory markers such as
IL-1, IL-2, IL-6, TNFα and CRP in cardiovascular disease [62], and indeed in depres-
sion [62; see also chapter by Hou et al., pp. 67–84]. Increased anxiety symptoms have
also been correlated with raised CRP, IL-6 and TNFα levels in one large study of 853
people free of cardiovascular disease [63]. In more controlled conditions, the Trier
social stress test, which is an acute psychological stressor designed as a laboratory
probe of social anxiety symptoms, has been consistently reported to produce a tran-
sient increase in plasma IL-6 [64] and IL-1β [65]. As noted earlier, some evidence ex-
ists for elevated IL-6 and IL-2 in panic disorder [30, 31]. However, although the pos-
sibility that cytokines may be responsible for the association of anxiety disorders with
ischemic heart disease is still to be fully investigated, there have been some recent
intriguing reports which suggest that immune functioning may link risk factors for
cardiovascular disease and anxiety (or at least psychological stress). For example, an
increased body mass index confers an excessively large proinflammatory cytokine re-
sponse to experimentally induced anxiety [66], and the combination of immune stim-
ulus and experimental anxiety challenge produces a synergistic increase in proinflam-
matory cytokine production, negative mood and BP [67].
Recent studies examining potential platelet dysfunction in panic disorder have ad-
dressed ‘second messenger’ systems. Initial studies reported decreased serotonin re-
ceptor coupling [68], decreased platelet cyclic adenosine monophosphate (cAMP)
concentrations [69] and altered subunit ratios of protein kinase A [70] in panic dis-
order patients. As elevated platelet cAMP concentrations are known to inhibit platelet
activation [71], it is possible that platelets are more aggregable in panic disorder.
However, some previous studies have shown that platelets show less aggregation in
response to serotonin challenge in panic disorder patients than controls [72]. Overall,
it remains unclear as to the degree of involvement platelets have in mediating the link
between panic disorder and cardiovascular disease. A further study used an anxiety
measure more closely linked to GAD than to panic (the anxiety subscale of the Hos-
pital Anxiety and Depression Scale). Anxiety was associated with excess platelet ag-
gregation in response to serotonin and ephedrine combinations in people with CHD
[73] and was a better predictor of this abnormal response than was depression.
One study [74] has reported an association of subclinical atherosclerosis with anx-
iety disorders, and it has been postulated that the impact of anxiety on CHD risk may
act through this pathway. The authors suggested several ways in which subclinical
atherosclerosis could be linked to anxiety disorders; these include (a) activation of im-
mune system pathways, (b) hypothalamic pituitary adrenal axis dysfunction, (c) pro-
atherogenic metabolic abnormalities, such as abdominal obesity and dyslipidemia,
and (d) sympathetic nervous overactivity in anxious persons.
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
Anxiety and Cardiovascular Disease 95
Acknowledgement
The authors would like to thank Dr. David M. Christmas for his assistance especially in the area of
cytokines/inflammatory markers.
References
1 Lewington S, Clarke R, Qizilbash N, Peto R, Collins
R: Age-specific relevance of usual blood pressure to
vascular mortality: a meta-analysis of individual data
for one million adults in 61 prospective studies. Lan-
cet 2002; 360: 1903–1913.
2 White GW: The New Hampshire Academy of Sci-
ence. Science 1931; 74: 98.
3 Mancia G, Grassi G, Parati G, Zanchetti A: The sym-
pathetic nervous system in human hypertension.
Acta Physiol Scand Suppl 1997; 640: 117–121.
4 Esler M: The sympathetic system and hypertension.
Am J Hypertens 2000; 13: 99S-105S.
5 White WB, Baker LH: Episodic hypertension sec-
ondary to panic disorder. Arch Int Med 1986; 146:
1129–1130.
6 Weissman MM, Markowitz JS, Ouellette R, Green-
wald S, Kahn JP: Panic disorder and cardiovascular/
cerebrovascular problems: results from a communi-
ty survey. Am J Psychiatry 1990; 147: 1504–1508.
7 Davies SJC, Ghahramani P, Jackson PR, Noble TW,
Hardy P, Hippisley-Cox J, Yeo WW, Ramsay LE: As-
sociation of panic disorder and panic attacks with
hypertension. Am J Med 1999; 107: 310–316.
8 Grimsrud A, Stein DJ, Seedat S, Williams D, Myer L:
The association between hypertension and depres-
sion and anxiety disorders: results from a nationally-
representative sample of South African adults. PLoS
One 2009; 4:e5552.
9 Jonas BS, Franks P, Ingram DD: Are symptoms of
anxiety and depression risk factors for hypertension?
Longitudinal evidence from the National Health and
Nutrition Examination Survey I Epidemiologic Fol-
low-up Study. Arch Fam Med 1997; 6: 43–49.
10 Paterniti S, Alperovitch A, Ducimetiere P, Dealberto
MJ, Lepine JP, Bisserbe JC: Anxiety but not depres-
sion is associated with elevated blood pressure in a
community group of French elderly. Psychosom
Med 1999; 61: 77–83.
11 Shinn EH, Poston WS, Kimball KT, St Jeor ST, Fo-
reyt JP: Blood pressure and symptoms of depression
and anxiety a prospective study. Am J Hypertens
2001; 14: 660–664.
12 Sparrow D, Garvey AJ, Rosner B, Thomas HE Jr:
Factors in predicting blood pressure change. Circu-
lation 1982; 65: 789–794.
13 Pilgrim JA, Stansfeld S, Marmot M: Low blood pres-
sure, low mood? BMJ 1992; 304: 75–78.
14 Hildrum B, Mykletun A, Stordal E, Bjelland I, Dahl
AA, Holmen J: Association of low blood pressure
with anxiety and depression: the Nord-Trondelag
Health Study. J Epidemiol Community Health 2007;
61: 53–58.
15 Pemberton J: Does constitutional hypotension exist?
BMJ 1989; 298: 660–662.
16 Davies SJC, Bjerkeset O, Nutt DJ, Lewis G: A U-
shaped relationship between systolic blood pressure
and panic symptoms: the HUNT study. Psychol Med
2012; 17: 1–8.
17 Davies SJC, Jackson PR, Lewis G, Hood SD, Nutt DJ,
Potokar JP: Is the association of hypertension and
panic disorder explained by the clustering of auto-
nomic panic symptoms in hypertensive patients? J
Affect Disord 2008; 111: 344–350.
18 Mann SJ: Neurogenic essential hypertension revisit-
ed: the case for increased clinical and research atten-
tion. Am J Hypertens 2003; 16: 881–888.
19 Esler M, Rumantir M, Kaye D, Jennings G, Hastings
J, Socratous F, et al: Sympathetic nerve biology in es-
sential hypertension. Clin Exp Pharmacol Physiol
2001; 28: 986–989.
20 Wilkinson DJ, Thompson JM, Lambert GW, Jen-
nings GL, Schwarz RG, Jefferys D, et al: Sympathetic
activity in patients with panic disorder at rest, under
laboratory mental stress, and during panic attacks.
Arch Gen Psychiatry 1998; 55: 511–520.
21 Nutt DJ: Altered central alpha 2-adrenoceptor sensi-
tivity in panic disorder. Arch Gen Psychiatry 1989;
46: 165–169.
22 Ferrier C, Cox H, Esler M: Elevated total body nor-
adrenaline spillover in normotensive members of
hypertensive families. Clin Sci (Lond) 1993; 84: 225–
230.
23 Bell C, Forshall S, Adrover M, Nash J, Hood S, Argy-
ropoulos S, et al: Does 5-HT restrain panic? A tryp-
tophan depletion study in panic disorder patients re-
covered on paroxetine. J Psychopharmacol (Oxf)
2002; 16: 5–14.
24 Polyák J: How should we manage cardiovascular
panic disorder accompanied by hypertension? J Hy-
pertens 2001; 19:S64.
25 Yeragani VK, Sobolewski E, Igel G, Johnson C, Jam-
pala VC, Kay J, et al: Decreased heart-period variabil-
ity in patients with panic disorder: a study of Holter
ECG records. Psychiatry Res 1998; 20; 78: 89–99.
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
96 Davies · Allgulander
26 Friedman BH, Thayer JF: Autonomic balance revis-
ited: panic anxiety and heart rate variability. J Psy-
chosom Res 1998; 44: 133–151.
27 Yeragani VK, Jampala VC, Sobelewski E, Kay J, Igel
G: Effects of paroxetine on heart period variability in
patients with panic disorder: a study of holter ECG
records. Neuropsychobiology 1999; 40: 124–128.
28 Davies SJ, Hood SD, Argyropoulos SV, Morris K,
Bell C, Witchel HJ, et al: Depleting serotonin en-
hances both cardiovascular and psychological stress
reactivity in recovered patients with anxiety disor-
ders. J Clin Psychopharmacol 2006; 26: 414–418.
29 Davies SJ, Lowry CA, Nutt DJ: Panic and hyperten-
sion: brothers in arms through 5-HT? J Psychophar-
macol 2007; 21: 563–566.
30 Weizman R, Laor N, Wiener Z, Wolmer L, Bessler H:
Cytokine production in panic disorder patients. Clin
Neuropharmacol 1999; 22: 107–109.
31 Norman TR, Judd FK, Gregory M, James RH, Kim-
ber NM, McIntyre IM, et al: Platelet serotonin up-
take in panic disorder. J Affect Disord 1986; 11: 69–
72.
32 Kaplan NM: Anxiety-induced hyperventilation: a
common cause of symptoms in patients with hyper-
tension. Arch Intern Med 1997; 157: 945–948.
33 Klein DF: False suffocation alarms, spontaneous
panics, and related conditions. Arch Gen Psychiatry
1993,50: 306–317.
34 Davies SJC, Jackson PR, Ramsay LE, Ghahramani P:
Drug intolerance due to non-specific side effects re-
lated to psychiatric morbidity in hypertensive pa-
tients. Arch Intern Med 2003; 163: 592–600.
35 Pickering TG, Devereux RB, Gerin W, James GD,
Pieper C, Schlussel YR, et al: The role of behavioral
factors in white coat and sustained hypertension. J
Hypertens Suppl 1990; 8:S141–S147.
36 Davies SJ, Jackson PR, Ramsay LE, Ghahramani P,
Palmer RL, Hippisley-Cox J: No evidence that panic
attacks are associated with the white coat effect in
hypertension. J Clin Hypertens (Greenwich) 2003; 5:
145–152.
37 Macdonald LA, Sackett DL, Haynes RB, Taylor DW:
Labelling in hypertension: a review of the behav-
ioural and psychological consequences. J Chronic
Dis 1984; 37: 933–942.
38 Van der Kooy K, van Hout H, Marwijk H, Marten H,
Stehouwer C, Beekman A: Depression and the risk
for cardiovascular diseases: systematic review and
meta-analysis. Int J Geriatr Psychiatry 2007; 22: 613–
626.
39 Frasure-Smith N, Lespérance F: Recent evidence
linking coronary heart disease and depression. Can J
Psychiatry 2006; 51: 730–737.
40 Stamfer HG, Hince DA, Dimmett SB: Depression as
a risk factor for coronary heart disease – How strong
is the evidence? Open J Psychiatry 2012; 2: 284–291.
41 Roest AM, Martens EJ, de Jone P, et al: Anxiety and
risk of incident coronary heart disease: A meta-anal-
ysis. J Am Coll Cardiol 2010; 56: 38–46.
42 Smoller JW, Pollack MH, Wassertheil-Smoller S,
Jackson RD, Oberman A, Wong ND, et al: Panic at-
tacks and risk of incident cardiovascular events
among postmenopausal women in the Women’s
Health Initiative Observational Study. Arch Gen
Psychiatry 2007; 64: 1153–1160.
43 Gomez-Caminero A, Blumentals WA, Russo LJ,
Brown RR, Castilla-Puentes R: Does panic disorder
increase the risk of coronary heart disease? A cohort
study of a national managed care database. Psycho-
som Med 2005; 67: 688–691.
44 Chen YH, Tsai SY, Lee HC, Lin HC: Increased risk of
acute myocardial infarction for patients with panic
disorder: a nationwide population-based study. Psy-
chosom Med 2009; 71: 798–804.
45 Walters K, Rait G, Petersen I, Williams R, Nazareth I:
Panic disorder and risk of new onset coronary heart
disease, acute myocardial infarction, and cardiac
mortality: cohort study using the general practice re-
search database. Eur Heart J 2008; 29: 2981–2988.
46 Janszky I, Ahnve S, Lundberg I, et al: Early-onset de-
pression, anxiety and risk of subsequent coronary
heart disease. 37-year follow-up of 49,321 young
Swedish men. J Am Coll Cardiol 2010; 56: 31–37.
47 Coryell W, Noyes R, Clancy J: Excess mortality in
panic disorder: a comparison with primary unipolar
depression. Arch Gen Psychiatry 1982; 39: 701–703.
48 Coryell W, Noyes R Jr, House JD: Mortality among
outpatients with anxiety disorders. Am J Psychiatry
1986; 143: 508–510.
49 Crown S, Crisp AH: A short clinical diagnostic self-
rating scale for psychoneurotic patients. The Mid-
dlesex Hospital Questionnaire (M.H.Q.). Br J Psy-
chiatry 1966; 112: 917–923.
50 Haines AP, Imeson JD, Meade TW: Phobic anxiety
and ischaemic heart disease. BMJ (Clin Res Ed) 1987;
295: 297–299.
51 Kawachi I, Sparrow D, Vokonas PS, Weiss ST:
Symptoms of anxiety and risk of coronary heart dis-
ease: the Normative Aging Study. Circulation 1994;
90: 2225–2229.
52 Albert CM, Chae CU, Rexrode KM, Manson JE,
Kawachi I: Phobic anxiety and risk of coronary heart
disease and sudden cardiac death among women.
Circulation 2005; 111: 480–487.
53 Kawachi I, Colditz GA, Ascherio A, Rimm EB,
Giovannucci E, Stampfer MJ, et al: Prospective study
of phobic anxiety and risk of coronary heart disease
in men. Circulation 1994; 89: 1992–1997.
54 Watkins LL, Blumenthal JA, Davidson JR, Babyak
MA, McCants CB Jr, Sketch MH Jr: Phobic anxiety,
depression, and risk of ventricular arrhythmias in
patients with coronary heart disease. Psychosom
Med 2006; 68: 651–656.
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
Baldwin DS, Leonard BE (eds): Anxiety Disorders.
Mod Trends Pharmacopsychiatry. Basel, Karger, 2013, vol 29, pp 85–97 (DOI: 10.1159/000351945)
Anxiety and Cardiovascular Disease 97
55 Fleet RP, Beitman BD: Cardiovascular death from
panic disorder and panic-like anxiety: a critical re-
view of the literature. J Psychosom Res 1998; 44: 71–
80.
56 Suls J, Bunde J: Anger, anxiety, and depression as risk
factors for cardiovascular disease: the problems and
implications of overlapping affective dispositions.
Psychol Bull 2005; 131: 260–300.
57 Parker G, Hyett M, Hadzi-Pavlovic D, Brotchie H,
Walsh W: GAD is good? Generalized anxiety disor-
der predicts a superior five-year outcome following
an acute coronary syndrome. Psychiatry Res 2011;
188: 383–389.
58 Kleiger RE, Miller JP, Bigger JT Jr, et al: Decreased
heart rate variability and its association with in-
creased mortality after acute myocardial infarction.
Am J Cardiol 1987; 59: 256–262.
59 Glassman AH, O’Connor CM, Califf RM, Swedberg
K, Schwartz P, Bigger JT Jr, et al: Sertraline treatment
of major depression in patients with acute MI or un-
stable angina. JAMA 2002; 288: 701–709.
60 Rasmussen A, Lunde M, Poulsen DL, Sørensen K,
Qvitzau S, Bech P: A double-blind, placebo-con-
trolled study of sertraline in the prevention of de-
pression in stroke patients. Psychosomatics 2003; 44:
216–221.
61 Lehnert H, Lombardi F, Raeder EA, Lorenzo AV,
Verrier RL, Lown B, et al: Increased release of brain
serotonin reduces vulnerability to ventricular fibril-
lation in the cat. J Cardiovasc Pharmacol 1987; 10:
389–397.
62 Miller GE, Stetler CA, Carney RM, Freedland KE,
Banks WA: Clinical depression and inflammatory
risk markers for coronary heart disease. Am J Car-
diol 2002; 90: 1279–1283.
63 Pitsavos C, Panagiotakos DB, Papageorgiou C, Tset-
sekou E, Soldatos C, Stefanadis C: Anxiety in relation
to inflammation and coagulation markers, among
healthy adults: the ATTICA study. Atherosclerosis
2006; 185: 320–326.
64 von Känel R, Kudielka BM, Metzenthin P, Helfricht
S, Preckel D, Haeberli A, Stutz M, Fischer JE: Aspirin,
but not propranolol, attenuates the acute stress-in-
duced increase in circulating levels of interleukin-6:
a randomized, double-blind, placebo-controlled
study. Brain Behav Immun 2008; 22: 150–157.
65 Yamakawa K, Matsunaga M, Isowa T, Kimura K, Ka-
sugai K, Yoneda M, Kaneko H, Ohira H: Transient
responses of inflammatory cytokines in acute stress.
Biol Psychol 2009; 82: 25–32.
66 Wirtz PH, Ehlert U, Emini L, Suter T: Higher body
mass index (BMI) is associated with reduced gluco-
corticoid inhibition of inflammatory cytokine pro-
duction following acute psychosocial stress in men.
Psychoneuroendocrinology 2008; 33: 1102–1110.
67 Brydon L, Walker C, Wawrzyniak A, Whitehead D,
Okamura H, Yajima J, Tsuda A, Steptoe A: Synergis-
tic effects of psychological and immune stressors on
inflammatory cytokine and sickness responses in
humans. Brain Behav Immun 2009; 23: 217–224.
68 Dell’Osso L, Carmassi C, Palego L, Trincavelli ML,
Tuscano D, Montali M, et al: Serotonin-mediated cy-
clic AMP inhibitory pathway in platelets of patients
affected by panic disorder. Neuropsychobiology
2004; 50: 28–36.
69 Marcourakis T, Gorenstein C, Brandao de Almeida
PE, Ramos RT, Glezer I, Bernardes CS, et al: Panic
disorder patients have reduced cyclic AMP in plate-
lets. J Psychiatr Res 2002; 36: 105–110.
70 Tardito D, Zanardi R, Racagni G, Manzoni T, Perez
J: The protein kinase A in platelets from patients
with panic disorder. Eur Neuropsychopharmacol
2002; 12: 483–487.
71 Schwartz UR, Walter U, Eigenthaler M: Taming
platelets with cyclic nucleotides. Biochem Pharma-
col 2001; 62: 1153–1161.
72 Butler J, O’Halloran A, Leonard BE: The Galway
Study of Panic Disorder. II: Changes in some periph-
eral markers of noradrenergic and serotonergic
function in DSM III-R panic disorder. J Affect Dis-
ord 1992; 26: 89–99.
73 Zafar MU, Paz-Yepes M, Shimbo D, Vilahur G, Burg
MM, Chaplin W, Fuster V, Davidson KW, Badimon
JJ: Anxiety is a better predictor of platelet reactivity
in coronary artery disease patients than depression.
Eur Heart J 2010; 31: 1573–1582.
74 Seldenrijk A, Vogelzangs N, van Hout HP, van Mar-
wijk HW, Diamant M, Penninx BW: Depressive and
anxiety disorders and risk of subclinical atheroscle-
rosis. Findings from the Netherlands Study of De-
pression and Anxiety (NESDA). J Psychosom Res
2010; 69: 203–210.
Simon J.C. Davies
Geriatric Psychiatry Division, Centre for Addiction and Mental Health
Department of Psychiatry, University of Toronto
Toronto, ON M6J 1H4 (Canada)
E-Mail simon_davies@camh.net
Downloaded by:
Centre for Addiction and Mental Health
192.197.205.95 - 3/11/2015 4:16:27 PM
