35 Psychological Factors and Heart Disease

Patients diagnosed with heart disease often experience emotional distress due to the life-threatening nature of their disease and are faced with functional impairments that may influence quality of life. In turn, these psychological manifestations interfere with adherence to treatment and increase the risk for mortality and morbidity. The risk incurred by psychological factors is of equal magnitude to that of standard risk factors, including somatic indicators of disease severity, such as left ventricular dysfunction.

This chapter focuses on the psychological impact of heart disease and consequences of psychological manifestations for prognosis, with emphasis on depression, anxiety, social isolation, health status, and Type D personality. Mechanisms, both biological and behavioural, that may be responsible for the link between psychological factors and cardiac prognosis are discussed. Results of recent behavioural and pharmacological intervention trials targeting psychological factors are also presented.

In order to enhance secondary prevention in patients with established heart disease, patients should be screened for psychological risk factors in clinical practice. When seeing patients, it is important that cardiologists take the time to listen carefully to patients, use clear and succinct communication, and make specific and simple recommendations. Time should also be allocated to follow-up patient adherence with medication and lifestyle changes. Patients with psychological comorbidity may need to be referred to healthcare professionals that are specialists in the area of psychological management, as they may require more intensive monitoring and treatment of a behavioural and psychological nature.

In 2003, behavioural and psychological factors were introduced for the first time in the official European guidelines on cardiovascular disease (CVD) prevention [1, 2]. In 2004, the landmark case–control INTERHEART study, based on 15,152 myocardial infarction (MI) patients and 14,820 controls recruited from 52 countries, helped further cement their role in heart disease, as psychological factors were shown to incur a twofold increased risk of incident MI independent of standard risk factors [3]. The European guidelines on CVD prevention ( Chapter 12) have since been updated, with psychological factors as potential risk factors in heart disease having earned a solid place in the guidelines [4].

Research into psychological factors in the context of heart disease has grown considerably in the last decades, with increasing recognition that a patient’s psychological profile not only plays a major role in the adaptation to the disease, but also impacts on mortality, morbidity, and quality of life [4–6]. Psychological factors also impede adaptation of lifestyle changes, moderate treatment outcomes, and serve as obstacles for treatment adherence—including participation in cardiac rehabilitation (CR) ( Chapter 25) [4]. Importantly, psychological risk factors tend to cluster together within individuals, with patients with several psychological risk factors having an increased risk of poor health outcomes [6, 7]. Therefore, psychological risk factors should be studied in concert rather than using a ‘risk factor of the month’ approach. In this regard, personality traits play an important role as determinants of substantial individual differences in vulnerability to psychological risk factors.

Another important issue is how and through which pathways psychological factors may exert their deleterious effect on health outcomes. Although accumulating evidence suggests a number of physiological and behavioural pathways that may explain the association between psychological factors and CVD, there are still many unanswered questions. This may in part be attributed to the complex relationships between psychological factors and heart disease, as many different bodily systems and pathways are involved. It is important to increase our knowledge of these pathways as well as factors that may moderate the effect of psychological risk factors on health outcomes. This may help to optimize secondary prevention trials that target psychological risk factors.

In terms of managing and treating psychological risk factors, some information is available from pharmacological, behavioural, and stress management trials. Nevertheless, there still remain several unresolved issues, including: the optimal time point(s) for screening for psychological factors; which screening instrument(s) to use; which treatment to choose; the timing of the intervention to reduce the impact of psychological factors on health outcomes; and which intervention works best for which patient.

This chapter will focus on the psychological impact of heart disease and the ensuing consequences of psychological manifestations for cardiovascular health, with emphasis on depression, anxiety, social isolation, health status, and Type D personality. Mechanisms, both biological and behavioural, that may be responsible for the link between psychological factors and cardiovascular prognosis are discussed together with results of recent behavioural and pharmacological intervention trials that target psychological risk factors. Suggestions are provided as to how clinicians may identify and help patients who have a psychological profile that may increase their risk of adverse cardiovascular health.

The diagnosis and treatment of CVD may have a major psychological impact. This has been documented in patients with coronary artery disease (CAD) ( Chapter 16 and 17), chronic heart failure (CHF) ( Chapter 23), and peripheral arterial disease (PAD) ( Chapter 36), and in the context of invasive cardiac treatment, including percutaneous coronary intervention (PCI) ( Chapter 16), coronary artery bypass graft surgery (CABG) ( Chapter 17), heart transplantation, and implantable cardioverter defibrillator (ICD) therapy ( Chapter 30). Importantly, there are large individual differences in the manifestation of psychological distress and poor health status.

Depression is one of the most studied psychological factors in CVD [5, 8, 9]. Clinical depression is characterized by the presence of a depressed mood or markedly decreased interest in daily activities (persisting for at least 2 weeks), in combination with at least four of the following additional symptoms: unintentional changes in weight, sleep problems, fatigue, psychomotor retardation or agitation, feelings of guilt or worthlessness, problems concentrating, and suicidal thoughts [10]. Apart from clinical depression as a psychiatric disorder, patients may also experience symptoms of depression without crossing the threshold for depressive disorder. However, typical symptoms of depression such as sadness and guilt are not frequently reported by cardiac patients; rather, they may complain of atypical symptoms like worries or feelings of malaise [11].

The prevalence of clinical depression in CAD is higher than in the general population, ranging between 15–25% [12], and is estimated to be slightly higher (i.e. 30%) for depressive symptoms [13]. Depression is also a common issue in patients with CHF [14]. While depression may be a reactive, transient phenomenon in some patients, it may be more persistent in others. Both standardized interviews and self-report questionnaires can be used to assess depressive disorder and symptoms of depression, respectively [15]. However, two recent reports have formulated opposite recommendations regarding depression screening: one recommends routine screening for depression in patients with CAD [5], whereas the other concludes that there is little evidence to indicate that depression screening would improve cardiovascular outcomes [8].

Apart from depression, vital exhaustion is a related construct that is characterized by feelings of fatigue, irritability, and demoralization, and which may also be associated with adverse clinical outcomes [16]. Within the broad spectrum of depressive symptoms, clinicians should also be aware of the role of specific depressive symptoms, such as feelings of hopelessness [17] and anhedonia, or the relative absence of positive mood states [18]. Both these specific symptoms may impact negatively on cardiovascular health [17, 18].

Recent evidence suggests that depression does not involve a homogeneous diagnostic category in cardiovascular patients, but may comprise distinctly different subtypes. Suggestions have also been made that some manifestations of depression may actually reflect the severity of cardiac disease [19]. Further, recent studies failed to find a relationship between depression and prognosis following MI [20], or found that this relationship may be limited to first-ever depression [21] and somatic symptoms of depression [19], but not to recurrent depression or affective depression symptoms. Depression, as well as CHF, is associated with symptoms of fatigue and malaise, and this overlap in symptoms represents an extra diagnostic challenge. Finally, there is some evidence to suggest that depression in post-MI patients is not accompanied by the distorted depressive cognitions that are typical of psychiatric depressive disorder [22]. As will be discussed later in this chapter, depression does have biological and behavioural effects, such as increased blood platelet reactivity, decreased heart rate variability, and poor adherence to treatment, which may have an adverse effect on prognosis.

In contrast to depression, anxiety in cardiac patients is frequently under-recognized and ignored by healthcare providers [23–26]. Anxiety is a negative emotion that occurs in response to perceived threats and is characterized by a perceived inability to predict or control the threatening situation [23]. Anxious patients are more likely to perceive upcoming situations as threatening, and may fear that they will not be able to control these situations. Although anxiety is a normal reaction to an acute cardiac event and prompts an individual to seek appropriate medical care, persistent anxiety has adverse consequences, including difficulty adhering to treatment [23], and an increase in cardiac symptoms [24]. Anxiety and depression tend to co-occur in post-MI patients [11, 15, 24, 27, 28], and screening for anxiety may help to identify patients at risk for post-MI depression [11].

Anxiety is common among cardiac patients; its prevalence ranges between 70–80% in patients who have experienced an acute cardiac event, and between 20–25% in cardiac patients with persistent anxiety over the long term [23]. More than 30% of patients experience anxiety following an MI [24], and anxiety is also a common negative emotional condition in patients with CHF [14], and in patients treated with ICD therapy [26].

There is some evidence to suggest that the anxious apprehension of some cardiovascular patients may result in physiological arousal and an associated increase in cardiac risk. In patients who survive an acute coronary event ( Chapter 16), anxiety has been associated with increased in-hospital complications, such as arrhythmias and continued myocardial ischaemia [29], and with rehospitalizations and outpatient visits to cardiologists after discharge from the hospital [25].

Panic disorder is a more specific manifestation of anxiety, and qualifies as a form of mental disorder that is associated with periods of intense anxiety. One study found that panic disorder was prevalent in 9% of outpatients with CHF [30], while another study showed that as many as 38% of patients referred to an outpatient chest pain clinic suffered from panic disorder [31]. Female patients and patients with a lower level of education may be at increased risk of comorbid panic disorder [30, 31]. Panic disorder has a significant adverse effect on quality of life in patients with CHF, also after adjustment for age, gender, and New York Heart Association (NYHA) functional class [30, 31]. After long-term follow-up, panic has also been associated with more chest pain intensity and psychological distress [31].

Cardiac patients may be prone to develop post-traumatic stress disorder (PTSD). A cardiac event is potentially life threatening and likely involves a psychological response that is characterized by hyperarousal, fear and helplessness, which, together with symptoms of intrusion and avoidance, provide sufficient criteria to qualify for a diagnosis [32]. Not surprisingly, PTSD may evolve in the aftermath of an acute cardiac event such as MI or cardiac arrest [33, 34]. However, a chronic cardiac condition, such as CHF, may also qualify as traumatic because it may incur a continuous risk of sudden death, thereby constituting a chronic traumatic stressor. Further, treatment with an ICD may also be psychologically traumatic [35]. The prevalence rate of PTSD ranges between 8–32% in patients with MI, 5–38% in survivors of sudden cardiac arrest, and 8–18% following cardiac surgery [32].

Often, these patients experience intrusive and avoidant symptoms, as well as physiological hyperarousal. In fact, sympathetic hyperactivity and reduced parasympathetic cardiac control are hallmarks of PTSD, thus linking PTSD to atherosclerotic progression [32]. Avoidance behaviour, which is a common feature in PTSD, may also induce non-adherence to medication, because ingestion of medication may serve as a reminder of the traumatic cardiac event [36]. This non-adherence to medication, in turn, may increase the risk of adverse clinical events in cardiac patients with comorbid PTSD. Therefore, it is unfortunate that PTSD is generally overlooked in these patients [35].

Social factors that may affect a patient’s psychological status include size and frequency of everyday contacts (social networks), the existence and type of relationships (social relationships), and the quality of support provided by others (social support) [37]. Lack of supportive social networks and relationships may lead to social isolation. Several years ago, Ruberman and colleagues noted the potential importance of social isolation for the clinical course of post-MI patients [38].

Social inhibition (the tendency to inhibit self-expression in social interaction) and social avoidance (the tendency to avoid social contact) are personality traits that may predispose to social isolation. Both traits have been linked to an increased cardiovascular risk [39, 40]. The notion that social isolation may have an adverse effect in patients with CAD is supported by animal [41] and human [42] research showing that isolation is associated with increased physiological stress reactivity. Of note, CR may provide an opportunity to offer social interaction and peer support to socially isolated patients [37].

Health status refers to the impact of disease on patient functioning, as reported by the patient [43]. Given that health status involves a range of manifestations of disease including symptoms, functional limitations, and (discrepancies between actual and desired) quality of life, health status is often used as a patient-centred outcome [44, 45]. Since there is a large discrepancy between physician-rated and patient-rated symptom burden and functional status, physicians need to rely on standardized health status measures in order to accurately estimate patients’ health status [43, 46]. An example of a frequently used generic measure is the Short Form Health Survey 36 (SF-36; a measure of overall physical and mental health status) [47]. The Seattle Angina Questionnaire (SAQ), the Kansas City Cardiomyopathy Questionnaire (KCCQ), and the Minnesota Living with Heart Failure Questionnaire (MLHFQ) comprise examples of frequently used disease-specific measures of health status in CAD (SAQ) and CHF (KCCQ, MLHFQ) patients [44, 48].

Standardized assessment of health status has been proposed as a complementary to current cardiac assessment as a means of improving the quality of care of patients [43, 49]. Importantly, health status measures are not merely surrogate markers for standard cardiac diagnostic assessments, but rather comprise important tools for monitoring patients in routine clinical practice, examining outcomes in clinical trials, and, eventually, a means by which to improve medical decision making [44, 50].

The patient-report of subjectively experienced physical and mental health is not only a function of the limitations incurred by the severity of cardiac disorder, but is also closely related to the patient’s psychological status. Hence, not surprisingly, depression has been associated with poor health status in cardiac patients [51, 52]. In 2005, the Working Group on Outcomes Research in Cardiovascular Disease of the National Heart, Lung, and Blood Institute in the US stated that the first goal in promoting ‘patient-centred’ care is to identify determinants of health status [45]. With reference to this issue, the personality of the patient is a major determinant of large individual differences in health status and psychological distress. This will be discussed in more detail in the following sections.

Patients differ substantially in their vulnerability to psychological distress following a cardiovascular event, and personality traits may account for a large part of these individual differences in physical and mental health status. Across the years, various personality traits have been studied in relation to the risk of physical illness [53]. With reference to CVD, research initially focused on the type A behaviour pattern (characterized by time urgency and hostility), but due to mixed findings on type A research it became outdated to study personality types [54]. In the aftermath of type A research, studies on the role of personality mainly focused on specific traits such as hostility [55]. However, with the introduction of the distressed personality or Type D personality [56] in recent years, there is now a renewed interest in the role of personality [54, 57].

Type D personality is defined as the combination of two normal and stable personality traits, with Type D patients characterized by an elevated score on both negative affectivity (tendency to experience negative emotions) and social inhibition (tendency to inhibit self-expression) at the same time ( Table 35.1). These Type D traits can be assessed with the standardized and validated 14-item Type D Scale (DS14) that consists of seven items measuring negative affectivity (e.g. ‘I often feel unhappy’) and seven items measuring social inhibition (e.g. ‘I am a closed kind of person’) [58]. The DS14 is presented in the section on clinical implications (see Table 35.9). Patients with a Type D personality have a score of ≥10 on both traits; they tend to experience increased negative emotions, such as worrying, feeling down in the dumps, and being irritable, while at the same time not sharing these emotions with others due to fear of negative reactions.

The prevalence of Type D ranges from 25–33% across different types of CVDs, including CAD, CHF, PAD, heart transplantation recipients, and patients with life-threatening arrhythmias treated with ICD therapy [59].

Type D personality is conceptually different from depression and other measures of negative affect, despite some overlap [59, 60]. The most distinctive features of Type D compared with other measures of negative affect comprise its chronicity, and that social inhibition is embedded within the Type D construct. In other words, the construct also stipulates how patients cope with their negative emotions, that is, that they do not disclose their emotions, whereas this is not contained within depression [58, 59]. In a substudy of the Myocardial INfarction and Depression–Intervention Trial (MIND-IT), 206 out of 1205 post-MI patients (17%) met criteria for depressive disorder, while 224 (19%) had a Type D personality. Of note, only one out of four distressed patients displayed both depression and Type D personality, but as many as 74% displayed one form of distress—depression or Type D—but not the other ( Fig. 35.1).

Psychological risk factors often cluster together within individuals [6], and the Type D construct was specifically designed to identify patients who are at risk of this clustering of risk factors. Accordingly, patients with a Type D personality have an increased vulnerability to experience an amalgam of negative outcomes, including increased depression and anxiety. They are also likely to report poor health status and quality of life, which cannot be accounted for by indicators of disease severity, such as left ventricular dysfunction, multi-vessel disease, and NYHA functional class [59]. These results are consistent across studies and cardiovascular diagnosis (including CAD, CHF, and PAD) and despite state-of-the art treatment such as PCI with drug-eluting stenting and ICD therapy [59]. An overview of studies [61–70] on the impact of Type D on health status and quality of life is shown in Table 35.2, indicating that the associated independent risk ranges from two- to sevenfold.

The influence of psychological factors on cardiovascular mortality and morbidity is well documented, with the associated risk being at least of an equal magnitude to that of demographic and clinical risk factors, and independent thereof [9, 59]. Hence, the addition of psychological factors to traditional biomedical risk factors might help enhance risk stratification in cardiac patients. Nevertheless, there are still several gaps in our understanding and conceptualization of psychological risk factors, including whether they should be viewed as true risk factors or as risk markers, the most opportune time to screen, and which interventions might help moderate the effect of these psychological factors on health outcomes [9]. The most pertinent psychological factors influencing cardiovascular health outcomes are discussed in the following sections.

Depression is a common comorbid disorder of CAD [71] that has been associated with increased mortality, morbidity, rehospitalizations, healthcare consumption, and poor quality of life and adherence [9, 72–77]. Not only major depressive disorder but also subclinical levels of depression may increase the risk of adverse clinical events [9, 19, 77], with the associated risk being twofold [9, 72–77]. In addition, there is accumulating evidence that specific depressive symptoms may exert differential effects on prognosis [17, 19], with the quest for the identification of the most cardio-toxic symptoms currently ongoing. The relative importance of new onset versus persistent depression as a prognostic marker is also as yet undetermined, with results to date being mixed [21, 78–80]. Although it is generally assumed that depression in cardiac patients has the same form as that found in psychiatric patients, preliminary evidence indicates that depression in cardiac patients is both quantitatively and qualitatively different [22]. Post-MI patients with clinical depression seem to report lower mean levels of depressive cognitions and also less cognitive/affective but more somatic symptoms of depression than psychiatric patients with clinical depression [22].

Depression as a prognostic marker in CVD has primarily been studied in patients with MI, but seems to be associated with adverse clinical events across different types of cardiac populations, including unstable angina [77], PCI with drug-eluting stenting [17], CABG [78], PAD [81], and CHF [14, 82]. Depression has also been shown to precipitate ventricular arrhythmias in patients implanted with an ICD [83]. Examples of the most recent reviews and meta-analyses on the impact of depression on prognosis are listed in Table 35.3 [9, 14, 50, 73, 84–86].

Anxiety has received far less attention in cardiovascular research than depression, as also reflected in the paucity of reviews and meta-analyses available on anxiety and prognosis ( Table 35.3). Findings on the role of anxiety as a prognostic factor are mixed, with most but not all studies confirming a relationship between anxiety and adverse clinical events [14, 24, 25, 28, 86, 87], and with the associated risk being around twofold in the positive studies [86]. In a recent study, anxiety was found to be an independent predictor of mortality following CABG [88]. There is also evidence to show that anxiety is associated with in-hospital ischaemic and arrhythmic complications [29], increased health care consumption [25], poor health status and quality of life [87], and that anxiety may also enhance the detrimental effect of depression on health status [89]. These studies have primarily been conducted in patients with acute coronary syndrome (ACS) ( Chapter 16) or CHF ( Chapter 23), whereas there is a lack of studies on the prognostic role of anxiety in ICD patients ( Chapter 30) and PAD patients ( Chapter 36).

Beginning evidence shows that panic disorder may also be associated with increased mortality and morbidity in patients with CVD, with results being more convincing for an impact on quality of life [30, 31] than mortality [31]. This was demonstrated in two recent studies of chest pain patients [31] and patients with CHF [30].

The influence of symptoms of PTSD on cardiovascular prognosis has also received little attention, although available evidence indicates that PTSD is associated with increased morbidity, impaired health status and quality of life, and likely also with a greater risk for mortality [33, 34]. A recent study of ICD-treated cardiac event survivors provides more definite evidence that PTSD symptoms significantly increase the risk of long-term mortality, independent of disease severity [35].

Several years ago, Ruberman and colleagues noted that post-MI patients who were socially isolated had a poor clinical course as compared to patients with low levels of isolation [38]. In more recent years, social isolation has been associated with poor prognosis among patients with heart disease in some, but not all studies [37, 90]. On balance, social isolation incurs a two- to threefold risk of mortality, although the impact seems to be greatest in those patients who are most isolated [37]. A recent meta-analysis showing that having a partner increases the chance of participating in CR [91] provides evidence that this may comprise one of the pathways through which social isolation influences cardiovascular health.

Patient-rated health status is becoming an increasingly important outcome measure in cardiovascular research and a performance measure in clinical practice [45]. Cumulative evidence also shows that poor patient-rated health status predicts prognosis both in CAD and CHF [50], and in patients treated with ICD therapy [92]. This risk is independent of traditional biomedical risk factors [50, 92]. The evidence for an impact of health status is stronger for physical health status than for mental health status in both CAD and CHF [50]. Health status assessed with a disease-specific questionnaire seems to have greater prognostic value compared to health status assessed with a generic measure [50]. Given this evidence and that there is often a discrepancy between the physician’s evaluation of patients’ health status compared to patients’ own report [46], assessment of patient-rated health has incremental value in clinical practice and may help in risk stratification and enhance secondary prevention [44].

Personality characteristics have been shown to explain individual differences in health outcomes in patients with somatic disease, including the onset and course of heart disease [53, 59]. Although findings associating global type A behaviour with mortality have been inconsistent, there is some evidence that facets of type A behaviour, in particular hostility, exerts an adverse effect on cardiovascular health and helps accelerate the atherosclerotic process [53]. Especially, with the introduction of the Type D personality construct in recent years, there is a renewed interest in the role of personality in CVD [57]; 35.1.

Type D personality is associated with increased mortality and morbidity [27, 39, 56, 63, 93–97], with the mortality risk being around fourfold ( Table 35.4). As shown in a substudy of the Rapamycin-Eluting Stent Evaluated At Rotterdam Cardiology Hospital (RESEARCH) registry that included PCI patients treated with sirolimus-eluting versus bare metal stenting, Type D personality was associated with mortality and non-fatal MI already at 9 months (i.e. 15 months post index PCI), despite optimal treatment using state-of-the-art in interventional cardiology ( Fig. 35.2) [94]. A later RESEARCH registry substudy showed that the impact of Type D on prognosis can be attributed to the combination of the two subcomponents of Type D,—negative affectivity and social inhibition—rather than the single traits [39]. The risk incurred by Type D is independent of standard risk factors, including left ventricular dysfunction and multi-vessel disease, indicating that the increased risk of mortality and morbidity in patients with this personality disposition cannot be attributed to these patients having more severe underlying cardiac disease [56, 59].

Recently, Type D was also shown to have incremental value above and beyond other, single personality traits, such as neuroticism, in predicting mortality: ‘This raises the possibility that perhaps this type [i.e. referring to Type D personality] embodies unique intra-individual information relevant to health that is not captured by multiple trait ratings’ [98]. Importantly, Type D personality has also been shown to have incremental value above depression as a predictor of poor prognosis [99].

Despite an association between psychological factors and mortality and morbidity in CAD, there is some debate whether these factors exert an independent effect [9], or whether their influence may be confounded by somatic symptoms, including indices of disease severity such as left ventricular dysfunction. If this is true, the implication is that psychological factors predict prognosis largely due to their overlap with the underlying cardiac disease, and that some of these psychological manifestations (e.g. depression) may arise secondary to CAD due to these patients experiencing a more severe disease burden.

The results to date are somewhat mixed, with preliminary evidence showing that, in particular, depression may be confounded by somatic symptoms [19, 100–102], whereas personality factors, such as Type D, seem to be less prone to such confounding [100, 102, 103]. What is puzzling, however, is that generally psychological factors predict cardiac prognosis despite adjustment for measures of disease severity. One plausible explanation is that despite confounding there is a unique psychological component that incurs an increased risk of adverse health outcomes that is independent of disease severity and somatic symptoms [19].

A related issue pertains to whether psychological factors fulfil the criterion of being risk factors for the progression of CAD in the true sense of the word, or whether they constitute risk markers that lie on the causal pathway between a third variable and cardiac prognosis. To be a risk factor, a psychological factor would at a minimum have to satisfy the following criteria [59, 104]:

Given these criteria, even depression, the psychological factor that has received the most attention in cardiovascular research, does not warrant risk factor status, in particular given that randomized controlled behavioural and pharmacologic intervention trials, such as the Enhancing Recovery in Coronary Heart Disease clinical trial (ENRICHD) and Sertraline AntiDepressant Heart Attack Randomized Trial (SADHART), have not been able to modify the impact of depression on survival despite some reduction in depressive morbidity [105, 106].

Mechanisms linking psychological factors such as (chronic) stress, depression, anxiety, and Type D personality to cardiac prognosis are not yet well understood. However, a number of plausible mechanisms, both biological and behavioural, have been identified that may explain the link between psychological factors and prognosis.

The multiple biological and behavioural mechanisms that may serve as candidate intermediaries between psychological factors and progression of cardiac disease are for a large part shared between the psychological factors—although most studies have been done on depression. Therefore, each following section will discuss the individual plausible mechanisms successively, instead of dealing with each psychological factor separately. An overview of all suggested mechanisms is given in Fig. 35.3.

Cortisol, the end-product of the hypothalamus–pituitary–adrenal (HPA) axis, is an important steroid hormone in the regulation of normal physiology, and plays a pivotal role in the body’s stress response. In depression, or as a consequence of continued exposure to stress, the HPA axis may become deregulated, resulting in a chronically excessive secretion of cortisol. A number of studies have implicated HPA axis dysregulation as a potential mechanism for explaining the link between depression and poor prognosis in CAD, as HPA axis dysregulation is associated with cardiovascular risk factors such as coronary artery stenosis, visceral obesity, high blood pressure and heart rate, and hypercholesterolemia [107]. Studies in CAD have shown that depressed patients have impaired feedback control and consequent HPA axis hyperactivity [108]. Similar findings of HPA axis hyperactivity have been found in cardiac patients with a Type D personality [109, 110].

Current literature testifies to the importance of autonomic nervous system imbalance, favouring sympathetic activation, as a risk factor for cardiac prognosis and death (e.g. 111). One of the reasons that depression increases the risk for cardiovascular morbidity and mortality may be decreased heart rate variability, due to vagal withdrawal, which in turn may increase the risk for ventricular arrhythmias. Similarly, sympathetic hyperactivity has been implicated in inducing high blood pressure and myocardial ischaemia, due to coronary vasoconstriction [112]. Depressed post-MI patients are characterized by decreased heart rate variability [113, 114], with one study suggesting heart rate variability to be one of the mediating mechanisms between depression and mortality [115]. More research is needed for a more complete understanding, as a recent study specified that heart rate variability is only related to somatic symptoms but not cognitive symptoms of depression [114]. Another measure of vagal control, baroreceptor sensitivity, was found to be associated with high levels of anxiety, but not depression in one study of post-MI patients [116].

Both anxiety and depression have been linked to abnormalities in the duration of ventricular repolarization (i.e. long QT intervals) and QT variability (QTV), which also have been associated with increased risk of sudden death in cardiac patients (see also Chapter 30). Hypertensive patients with anxiety have been shown to have longer QT intervals, corrected for heart rate, as compared to hypertensive patients without anxiety [117]. Furthermore, there is consistent evidence that links depression to larger QT intervals and QTV in post-MI patients [118] and psychiatric populations [119]. People characterized by high levels of neuroticism also seem to have longer QT intervals [120], placing them at increased risk for life-threatening arrhythmias. Increased arrhythmic activity, i.e. 24-hour ventricular ectopy, has been related to increased levels of state anxiety and self-reported stress in post-MI patients during routine daily activities [121].

Another explanation for the increased morbidity and mortality in patients with comorbid CAD and depression concerns platelet abnormalities, such as increased platelet activation, or a disturbed serotonin metabolism in the platelet, which leaves the platelet more likely to degranulate to certain triggers, in turn leading to thrombosis [122]. Experimental research has shown that depressive and anxious symptoms are related to increased platelet P-selectin expression in response to acute stress, and a delayed recovery [123]. Decreased nitric oxide synthase (NOS) activity and plasma levels of nitric oxide (NO) comprise other abnormalities found in depressed cardiac patients. These abnormalities may contribute to increased platelet aggregation and coronary vasoconstriction, thereby increasing the risk of angina and ischaemia [124]. There is also pre-clinical data to suggest that brain NO is markedly increased in stress, anxiety and anxiety-related disorders [125]. No information to date is present on plasma NO levels in patients with comorbid anxiety and CAD.

Endothelial dysfunction, comprising vasoconstriction, leucocyte adhesion, thrombosis and cellular proliferation of the vessel wall, is a hallmark of early atherosclerosis development [126] ( Chapter 16). Depression seems to be associated with endothelial dysfunction, as several studies have reported impaired arterial flow-mediated dilatation in depressed psychiatric patients [127] as well as in CAD patients with depressive symptomatology [128]. Furthermore, increased levels of biomarkers for endothelial dysfunction such as soluble tissue factor, von Willebrand factor, and soluble intercellular adhesion molecule-1 have been associated with PTSD [129].

Atherosclerosis has been identified as an inflammatory process [130], with cardiac patients with increased levels of pro-inflammatory cytokines, such as tumour necrosis factor (TNF)-α and interleukin (IL)-6, having an increased risk of adverse clinical events [131]. Increased pro-inflammatory cytokine levels are also found in CHF patients, which have been suggested to be due to cardiac remodelling. CAD and CHF patients with depression [108], Type D personality [132], or PTSD [34] have all been characterized by an increased pro-inflammatory state. Although increased inflammation is linked to adverse prognosis, it is yet elusive whether this mediates the association between the psychological factors and prognosis.

A significant percentage of CAD patients experience myocardial ischaemia in response to mental stress and in a small portion of patients, mental stress-induced ischaemia may occur in the absence of exercise- or adenosine-induced ischaemia. Mental stress-induced ischaemia is an independent predictor of poor prognosis in CAD patients, related to incident MI and mortality [133]. Anxiety has been related to an increased incidence of in-hospital complications after acute MI, including myocardial ischaemia [29]. Future studies should examine whether stress-related constructs such as depression and Type D personality are also associated with mental stress-induced ischaemia.

Emotional stress may also lead to a specific type of cardiomyopathy, i.e. the Takotsubo cardiomyopathy (TC) ( Chapter 18), also referred to as the apical ballooning syndrome. TC is a reversible cardiomyopathy that is precipitated by acute emotional stress, such as the death of a family member, fierce argument, or natural disaster, and is characterized by symptoms of ACS and transient apical and midventricular wall motion abnormalities in the absence of obstructive CAD. TC is most common in older postmenopausal women [134]. Research findings reveal several possible pathophysiologic mechanisms such as multivessel coronary vasospasm, elevated plasma catecholamine levels, elevated sympathetic tone, and impaired coronary microcirculation, although the exact aetiology and pathophysiology remain unknown [135].

From a genetic perspective, diseases such as CAD and factors like depression or personality are all considered to be genetically complex, influenced by multiple genes exerting small effects, as well as by interactions between genes, and between genes and the environment. Behavioural genetics studies have indicated moderate to substantial heritability for Type D personality [136], depression, and anxiety [137]. Whether common genetic mechanisms underlie the co-occurrence of heart disease and psychological factors has only been tested for depression, as a large twin study revealed that shared genetic factors contribute substantially to the covariation of depression and CAD [138].

In addition to the biological mechanisms reviewed in earlier sections, there are several behavioural mechanisms linking psychological factors to morbidity and mortality in cardiac patients. These include poor self-management, poor compliance with treatment and CR, and lifestyle factors such as continued smoking, unhealthy eating habits, and lack of physical exercise. The most prominent findings are presented in the following sections, and summarized in Fig. 35.4.

World Health Organization (WHO) statistics show that in patients with chronic conditions, about 50% of patients are not compliant with recommendations on prevention or treatment. Patients may be non compliant for multiple reasons, including side effects of medication, but psychological factors such as depression, anxiety, phobias, or inhibition may also interfere with adherence. It is essential for a patient’s compliance to have positive expectations and beliefs in the benefits of treatment. Depression may have an adverse effect on these expectations and beliefs. Evidence shows that depression is associated with a threefold increased risk of non compliance to treatment and with a low participation rate in CR [139]. PTSD has also been associated with poor adherence to treatment in acute MI patients, resulting in an increased risk of rehospitalization [36]. In patients with obstructive sleep apnoea (a risk factor for CVD), adherence to treatment was significantly lower in Type D patients compared to non-Type D patients [140]. Hence, non compliance is an important behavioural mechanism for patients with combined cardiac disease and one of these psychological factors.

Common means of behavioural secondary prevention in CAD include lifestyle modifications ( Chapter 17), such as smoking cessation and dietary intervention, but also the targeting of psychological risk factors. It is often difficult to implement the appropriate lifestyle changes after a cardiac event. Reasons for reluctance to make these changes may include lack of social support and educational support [141]. Changes in mental health due to the medical illness, i.e. increased symptoms of anxiety and depression, may add to the unwillingness or inability to make lifestyle adjustments [139] (see also Table 35.5).

Type D patients experience negative emotions, which they refrain from sharing with others. A recent study showed that these characteristics negatively affect consultation behaviour of CHF patients. Even though patients with a Type D personality experienced more cardiac symptoms and appraised these symptoms as more worrisome compared to non-Type D patients, they were less likely to consult their cardiologist or heart failure nurse for these symptoms, potentially leading to under-treatment [142]. Poor consultation behaviour has only been examined in relation to Type D personality, and not anxiety or depression.

Benefits of CR are undisputed, although only a minority of eligible patients seem to attend. There may be multiple reasons for this non-attendance, including: socio-demographic factors, like low income, living alone, and living far away from CR facilities; lifestyle factors, such as lack of regular exercise habits; and clinical factors, such as more severe illness and lack of active physician endorsement [143, 144]. Psychological factors are also involved, as research shows that patients who experience more symptoms of depression and anxiety are less likely to participate in rehabilitation [143]. Hence, in clinical practice, it would be advisable to discuss patient barriers to participation in rehabilitation with the patient, as these may include psychological motives.

Pharmacological interventions aim to treat psychiatric disorders such as depression, while behavioural interventions encompass various forms of psychotherapy and behavioural therapy to treat, for example, depressed mood, stress or anxieties. Lifestyle interventions often take place in the context of cardiac rehabilitation and target eating habits, smoking, alcohol consumption and exercise (see Chapter 25). Below, a short overview is given of the effects of these interventions on psychological and cardiovascular health. Table 35.6 summarizes all clinical trials that are discussed in this section.

Among the available antidepressant agents, tricyclic antidepressants (TCAs) and selective serotonin reuptake inhibitors (SSRIs) are used most often to treat depression. In cardiac patients, however, TCAs are not well tolerated at a cardiovascular level, and therefore should be avoided. SSRIs on the other hand do not have undesirable cardiovascular effects in MI, unstable angina, and CHF, and have been used in several clinical trials to examine their effect on depression in these patient groups. An overview of these trials and their main results are summarized in Table 35.6. In general, the two SSRI trials (SADHART and CREATE) show that although SSRI treatment is safe, it is not very efficacious in treating depression in cardiac patients [106, 145]. MIND-IT is another pharmacological trial that evaluated the effects of mirtazapine, a non-tricyclic antidepressant, on adverse clinical events and long-term depression post-MI. The findings of MIND-IT were essentially similar to those of SADHART and CREATE; although mirtazapine was shown to be safe in cardiac patients, it had no effect on recurrent cardiac events, nor were there significant differences between patients assigned to the intervention versus usual care with respect to depressive and somatic symptoms, or quality of life [146].

During the past decades, it became apparent that psychological factors may have a negative impact on cardiac prognosis. More recently, a number of clinical trials have also examined whether behavioral treatment of these factors would lead to an improvement in cardiac prognosis.

Two behavioural intervention trials have examined the effects of cognitive behavioural therapy (CBT) or interpersonal psychotherapy (IPT) on depressive symptoms and adverse cardiac events or death in cardiac patients [105, 145]. The results of these trials are mixed at best. In the ENRICHD trial [105], the intervention significantly reduced depression, while in the CREATE trial [145] interpersonal therapy was not superior to clinical management in terms of reducing depression. Both trials failed to postively affect the incidence of major adverse cardiac events or death (see Table 35.6 for an overview).

Vital exhaustion, characterized by a triad of symptoms, i.e. feelings of excessive fatigue, irritability, and demoralization, has been targeted in the Exhaustion Intervention Trial (EXIT) with the intervention focusing on stress reduction and hostility treatment. Results showed that vital exhaustion and the occurrence of de novo lesions were reduced in the treated PCI patients without a history of CAD, but not in patients with a previous history. There were no treatment effects on new coronary events [148].

Intervention studies that have focused on the treatment of anxiety have primarily been conducted in ICD patients, employing either CBT, CR, or telephone support as the mainstays of treatment or in combination. These intervention studies were small scale and only focused on patient-centred outcomes, with no data available to show whether the intervention also had an effect on hard medical outcomes. All anxiety interventions in ICD patients seem to be successful in reducing anxiety, although CBT and CR have shown the strongest and most stable reductions [149].

In summary, behavioural interventions seem to be somewhat effective in reducing the level of psychological symptoms, such as depression and exhaustion. Although the interventions do not seem to affect cardiac outcomes in the overall group, recent subgroup analyses do reveal some significant relations between, for example, more severe, persistent depression and mortality [150]. It is therefore important to learn from these trials, instead of discarding them because of unexpected findings.

Stress management uses specific cognitive behavioural strategies to help patients reduce stress levels. Strategies include relaxation training, cognitive techniques and cognitive challenge, and/or consideration of specific coping strategies to be used at times of stress. Stress management may be applied on its own, or in the context of CR, and aims to improve patients’ morale and functioning, and decrease suffering.

Multiple studies have examined the efficacy of relaxation therapy, and these are comprehensively reviewed in several reviews and meta-analyses [151, 152]. Results of these meta-analyses show that, if an intervention is successful in reducing psychological stress, the risk of <2-year mortality is reduced by 28%. However, when taking sex differences into account, there was only a significant benefit for men, not women. Increasingly, stress management programmes are tracking markers of cardiovascular risk, showing that stress management reduces cardiac risk by reducing heart rate and stress-induced wall motion abnormalities, and by increasing heart rate variability, and flow-mediated dilatation [e.g. 153]. Blood pressure seems not to be affected by stress reduction [151].

Biofeedback is increasingly used in addition to stress management programmes in CR. Essentially, biofeedback tries to regulate the input of the autonomic nervous system to the heart. There have not yet been any randomized controlled trials assessing the effectiveness of biofeedback to reduce psychological stress and cardiovascular risk [154].

A multi-factorial CR programme that also included psychological intervention has been shown to be successful in improving both emotional functioning and long-term prognosis at 9-year follow-up [155]. This intervention programme focused on stress reduction, coping with stress, assertiveness training, and, depending on the needs of the patient, individual psychological therapy, using CBT to deal with chronic stress and tension, depression, anxiety, non-expression of emotions, hostility and irritability, and partner issues.

The role of psychological factors in patients with CAD has implications for clinical practice and the management of patients. We may as yet be dealing with a number of unknown factors, with the two most important being: 1) psychological factors may not be risk factors in the true sense of the word, but risk markers that lie on the causal pathway between other variables and major adverse cardiac events; and 2) psychological factors may be difficult to modify, at least to the extent of influencing survival, as shown from mixed results from clinical trials targeting depression [105, 106, 145]. In addition, evidence is mixed as to the usefulness of performing routine screening for depression in patients with CAD, at least when it comes to improving cardiovascular outcomes [5, 8]. Irrespectively, there is sufficient and sound evidence to show that psychological factors are related to a wide range of adverse health outcomes in patients with CAD ( Table 35.7). In addition, disorders such as depression and anxiety are serious and debilitating, and influence quality of life, which comprise sufficient reasons for warranting that psychological factors be taken seriously and managed in clinical practice.

Despite the fact that anxiety is common among cardiovascular patients and may have adverse health consequences if untreated, it is infrequently assessed or managed appropriately [23]. It has been estimated that among post-MI patients, only one out of three anxious patients are asked about such symptoms [24]. Importantly, treatment of anxiety with anxiolytic medication may protect against the triggering of arrhythmias [156]. Hence, there is an urgent need to detect anxiety in these patients, and to improve patients’ outcomes by placing anxiety in the forefront of clinical cardiac practice [23].

Screening for psychological factors can easily be made an integral part of patient clinical care, with the use of brief, standardized, and validated measures. Recently, an advisory of a consortium of councils of the American Heart Association advocated routine screening for depression of all cardiac patients seen in clinical practice, using the two-item Patient Health Questionnaire (PHQ) followed by the nine-item PHQ if screening positive on one or both items of the PHQ-2 [5]. Depending on the score on the PHQ-9, patients may subsequently be referred to a mental health professional for a more thorough evaluation [5]. However, this two-stepped approach seems to have no greater value in terms of identifying major depression in cardiac patients compared to the use of the PHQ-2 or PHQ-9 alone [157]. Hence, given its brevity, the PHQ-2 may be the preferred instrument to use in clinical practice ( Table 35.8).

There are other viable alternatives to the PHQ-2 and the PHQ-9 as screening instruments for depression in clinical practice, including the 21-item Beck Depression Inventory [158] and the 14-item Hospital Anxiety and Depression Scale [159]. The Hospital Anxiety and Depression Scale provides the opportunity to identify not only symptoms of depression but also anxiety. Patients with a Type D personality disposition can be identified with the Type D Scale (DS14), a brief 14-item standardized and validated measure that is easy to administer ( Table 35.9) [58]. Following an update of the European guidelines on CVD prevention in clinical practice by the Third Joint Task Force of European and other Societies on Cardiovascular Disease Prevention [1], an international committee provided recommendations for the assessment of psychosocial risk factors, and included the Beck Depression Inventory, the Hospital Anxiety and Depression Scale, and the Type D Scale (DS14) among the measures to use in clinical cardiology practice to identify high-risk patients [160]. Personality measures may comprise particularly good candidates as screening tools in clinical cardiology practice, as they assess

a predisposition to experience a wide variety of emotions across time and situations. Due to the stability of personality characteristics, such measures may also be less prone to be influenced by acute events, such as MI, and the underlying disease severity [100].

Currently, there is no consensus as to the most optimal time point for screening for psychological factors. Evidence from the depression literature indicates that assessment close to an acute cardiac event may be more likely to tap the physical ill health associated with the acute event than actual depression [9, 161]. Evidence on the importance of new onset versus persistent depression as a prognostic marker is also conflicting [21, 78–80]. Hence, pending further research, it seems better to screen patients at serial time points when patients are seen in clinical practice at follow-up visits.

Despite unanswered questions in relation to the role of psychological factors in patients with heart disease, these factors should be taken into account in the clinical management of patients. When seeing patients in clinical practice, it is important that cardiologists take the time to listen carefully to patients, use clear and succinct communication, and make specific and simple recommendations. It may be necessary to spend some time on investigating whether the patient is adhering to his/her medication and goals set for lifestyle changes, in which case involving and questioning the partner may provide added information. When setting goals for lifestyle changes, the chance of success is likely to be enhanced, when the patient is involved in setting the goals rather than the cardiologist dictating what the patient should do [6]. However, the cardiologist can help the patient develop personal, and preferably emotional, reasons for changing behaviour, set goals that are specific and sufficiently small such that they are realistic, and promote self-management and autonomy in terms of looking into, together with the patient, which barriers (e.g. stress) may impede attainment of the set goals and how to deal with these barriers (e.g. engaging in physical exercise or relaxation training to deal with stress) [6]. If the results from screening for psychological factors indicate that these risk indicators are present, it may be necessary to refer the patient to other healthcare professionals, such as a medical psychologist or a psychiatrist, as the patient may then require more intensive monitoring and treatment of a behavioural and psychological nature [160].