Psychiatric and Psychosocial Aspects of Cardiovascular Disease

Stress and Emotional Triggers of Acute Cardiovascular Events

Cardiac event rates often surge in populations that are exposed en masse to extremely stressful events. Acute myocardial infarctions (MIs) increased 35% in Los Angeles after the 1994 Northridge earthquake, and sudden cardiac deaths increased from 4.6 per day during the previous week to 24 on the day of the earthquake. The 1995 Great Hanshin (Kobe) earthquake produced a 3.5-fold increase in acute MIs and a significant increase in fatal MIs. In contrast, the 1989 Loma Prieta earthquake did not affect the rate of acute MIs. Whether a disaster triggers cardiac events may depend on the time of day that it strikes—whereas the Northridge earthquake hit at 4:30 am , the Loma Prieta earthquake hit at 5:00 pm . Superimposition of an extremely stressful event on the stress of sudden early morning awakening may be especially dangerous for individuals who are at risk for an acute MI.

Cardiac event rates may also increase during wars and terrorist attacks. For example, CV mortality increased 58% on the first day of Iraqi missile attacks on Israel during the 1991 Persian Gulf War, primarily in the targeted cities. In contrast, CV mortality did not increase in New York City on September 11, 2001. Cardiac event rate changes may depend in part on whether the stressful situation is inescapable. Individuals who were in close proximity to the 9/11 attacks were at higher risk for stress-related disorders than were those who were farther away at the time.

Private stressors can also trigger CV events. In a Danish study, the death of a child increased the risk of a first MI by 31% and of a fatal MI by 58%. In the Determinants of MI Onset Study, the risk of MI increased 21-fold after the death of a loved one. A U.K. study of older primary care patients found elevated 30-day rates of MI (rate ratio 2.14) and stroke (rate ratio 2.40) in bereaved compared to matched non-bereaved patients.

Takotsubo cardiomyopathy mostly affects women and is usually associated with transient left ventricular dysfunction (see also Chapters 37 , 38, and 50 ). The onset is preceded by a stressful event in about four out of five cases. The cardiomyopathy frequently resolves within weeks but recurrences are common and patients (especially older ones) who have had this syndrome are at increased risk for mortality.

Emotional triggers have been studied retrospectively in survivors of cardiac events. The trauma of having a cardiac event can affect a patient’s recall of antecedent emotions, and recall bias can affect the validity of this type of research. Case-crossover designs, in which patients serve as their own controls, can reduce recall bias. The Determinants of MI Onset Study used this design and found a 2.43-fold increase in the incidence of acute MI within two hours after angry outbursts. A meta-analysis of case-crossover studies found that the risk of MI or acute coronary syndrome (ACS) was 4.74 times higher in the 2 hours after an angry outburst than at other times.

Cardiovascular Responses to Everyday Stressors and Emotions

Ambulatory monitoring studies have documented a variety of CV responses to everyday stressors and negative emotions. For example, in patients with implantable cardioverter-defibrillators (ICDs), anger precedes about 15% of shocks compared to only 3% of control periods. Everyday stressors and negative emotions such as anger, tension, frustration, and sadness can reduce heart rate variability and trigger episodes of myocardial ischemia and ventricular ectopy in patients with CHD.

Mechanisms Underlying Acute Stress Effects

Several interrelated biobehavioral mechanisms have been implicated in the CV effects of acute stress. Mental stress-induced myocardial ischemia (MSMI) is accompanied by hemodynamic, neurohormonal, and vascular responses and activation of brain areas involved in stress reactivity and depression ( Fig. 99.1 ). Among 196 patients with coronary disease in the Psychophysiological Investigations of Myocardial Ischemia (PIMI), 58% developed MSMI during mental stress testing. MSMI was accompanied by increases in heart rate, blood pressure, cardiac output, and systemic vascular resistance, a decrease in left ventricular ejection fraction, and wall motion abnormalities. In the more recent Mental Stress Ischemia Prognosis Study (MIPS) of 660 patients with coronary disease, mental stress testing increased the rate-pressure product, arterial stiffness, microvascular constriction, and plasma epinephrine, as well as the inflammatory biomarkers IL-6, MCP-1, and MMP-9. The 106 (16%) patients who developed MSMI had greater hemodynamic and vasoconstrictive responses to mental stress compared to MSMI-negative patients. Normal epicardial coronary arteries and coronary microvessels dilate in response to acute mental stress. Diseased vessels paradoxically constrict in response to mental stress, and resistance vessel dilation is impaired. The coronary microvascular response to mental stress is endothelium-dependent and mediated by nitric oxide.

Childhood Adversity

Adverse childhood experiences (ACEs) include physical, sexual, and emotional abuse, as well as neglect, economic disadvantage, homelessness, exposure to violent crime, bullying, and other forms of victimization. Over 50% of the U.S. population has had at least one ACE. An American Heart Association scientific statement concluded that there is substantial evidence linking ACEs to cardiometabolic diseases later in life, including heart disease, diabetes, and stroke. A recent analysis of Behavioral Risk Factor Surveillance System (BRFSS) data found dose-response relationships between ACE exposure and the incidence of CVD, as well as asthma, arthritis, chronic obstructive pulmonary disease (COPD), and depression. The associations with CVD and COPD were explained in part by smoking, heavy drinking, and obesity. A separate analysis of BRFSS data also found that high exposure to ACEs was associated with CVD, but only in respondents with a history of depression.

Socioeconomic Status

There has been extensive research on the health effects of low SES, also known as socioeconomic position. Most studies of the long-term effects of SES in childhood are based on parental or household income, education, or occupation. There is growing evidence that low childhood SES increases the risk of CVD in adulthood. For example, a longitudinal population-based cohort study in Finland found that growing up in a family with low SES predicts increased left ventricular mass and impaired diastolic function in middle age.

Most studies associating adult SES with health outcomes use education, income, wealth, occupation, or employment status as indicators. Early studies established that adult SES is related to CVD risk, and more recent studies have strengthened the evidence. For example, a recent report from the Atherosclerosis Risk in Communities Study found that over a 24-year follow-up, the lowest SES group had a 1.92-fold higher risk of developing heart failure compared with the highest SES group after adjusting for income, education, deprivation, CV risk factors, and health care access. A recent Medicare Expenditure Panel Survey report found that the lowest income group had the highest prevalence of cardiac risk factors including obesity, diabetes, hypertension, and physical inactivity. The trend in physical inactivity was particularly concerning, with a 71% increase in the lowest income group over a decade. A study from the Swedish National Diabetes Register reported CV mortality hazard ratios (HRs) of 1.87 for the lowest versus highest income quintiles and 0.84 for individuals with college degrees versus those with less than 10 years of education. SES effects have also been found in recent clinical follow-up studies. For example, low SES was associated with a high risk of all-cause mortality over 4.5 years in a retrospective study of 4503 patients who had been hospitalized with atrial fibrillation.

Occupational Stress and Unemployment

A large body of research has linked various forms of occupational or work-related stress to CVD. The job strain or demand-control model has dominated the research on occupational stress. It hypothesizes that demanding jobs in which the worker has little control are highly stressful, especially in socially unsupportive work environments. There is considerable evidence that job strain is a CV risk factor, at least among men. In an individual-level meta-analysis with 47,045 participants, individuals with job strain were more likely to have elevated Framingham Risk Scores (odds ratio, 1.13). A cumulative meta-analysis of 26 prospective cohort studies reported a HR for incident CHD of 1.34 for the presence versus absence of job strain.

The effort-reward imbalance model identifies another source of occupational stress. The rewards of some highly demanding jobs are insufficient in terms of compensation, job security, prospects for advancement, and/or prestige. The evidence for effort-reward imbalance as a CV risk factor in men is more limited than it is for job strain, and little is known about its effects in women. However, it predicted CV and all-cause mortality in men in the Kuopio Ischemic Heart Disease Risk Factor Study and CV mortality over 25 years in a prospective Finnish cohort study. Job strain predicted and effort-reward imbalance marginally predicted incident CHD in the Whitehall II study of male civil servants in London, but only among employees who were frequently subjected to unfair criticism or other forms of occupational injustice. Occupational injustice itself was an independent predictor of incident CHD, even after adjusting for job strain and effort-reward imbalance.

Job insecurity and unemployment have also been identified as contributors to poor CV health. An individual-level meta-analysis of 13 cohort studies found an adjusted relative risk of high versus low job security of 1.32 for incident CHD, with no differences between men and women or younger and older individuals. The CV risks of job insecurity were partly explained by lower SES and higher prevalence of CHD risk factors among job-insecure individuals. An analysis of nationally representative prospective data on adults aged 51 to 75 years in the Health and Retirement Study showed that the risk of having an acute MI was significantly higher among unemployed than consistently employed workers (HR 1.35), and that there was a dose-response relationship with MI risk and the cumulative number of job losses. The first year of unemployment was an especially high-risk period.

Social Discrimination and Stigmatization

There is inconsistent evidence as to whether various forms of social discrimination and stigmatization, including discrimination based on race, age, sex, or sexual orientation, increase the risk of CVD. Some studies have yielded null or paradoxical results, such as findings from the Jackson Heart Study that racial discrimination among African Americans is associated with a lower risk of all-cause mortality and from the Coronary Artery Risk Development in Young Adults (CARDIA) study that racial discrimination is inversely associated with coronary artery calcification. However, other studies do suggest that chronic exposure to discrimination or stigmatization can have adverse CV consequences. For example, when participants in the National Epidemiological Survey on Alcohol and Related Conditions were grouped by state-level indicators of structural racism, there was a significantly greater past-year prevalence of acute MI among blacks living in states with high levels of structural racism compared to those living in low-structural racism states. Conversely, whites were less likely to have had an acute MI if they lived in a high- rather than a low-structural racism state. Among participants in the Multi-Ethnic Study of Atherosclerosis who were initially free of clinical CVD, those who reported high lifetime levels of racial discrimination had a higher 10-year risk of incident CV events (adjusted HR, 1.36). African American participants in the Jackson Heart Study without hypertension at baseline who experienced medium (HR, 1.49) or high (HR, 1.34) levels of racial discrimination were at increased risk for incident hypertension. Thus, there is growing evidence that racial discrimination increases CVD risks in African Americans.

Caregiving

There has been limited research on the CV effects of chronic stress associated with caregiving for a family member with a debilitating chronic illness such as Alzheimer disease, and much of this work has focused on surrogate outcomes. Nevertheless, there is evidence that stressful caregiving over relatively long periods may promote CVD. In the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study, Framingham Stroke Risk scores averaged 23% higher in participants who reported high caregiver strain compared to those with low or no caregiver strain. The risk was especially high in African American men. Caregiver strain did not affect Framingham CHD Risk scores in this study. Other studies have shown that caregiver stress can contribute to endothelial dysfunction, impairment of the cardiovagal baroreflex, the development or worsening of cardiometabolic syndrome, and the development of carotid plaque.

Combinations of Chronic and Acute Stress

Acute stress is often superimposed on a background of chronic stress and other psychological (e.g., depression) and pathophysiological (e.g., unstable plaque) vulnerabilities. The “perfect storm” model proposes that when mental stress triggers an ACS, it does so in concert with these other factors. Few studies have examined whether the CV effects of acute mental stress differ depending on the background level of chronic stress. However, a recent analysis of MIPS data showed that in patients with stable coronary disease, a high level of chronic psychosocial distress is associated with a blunted hemodynamic response to acute mental stress. In prior studies, blunted CV reactivity to mental stress has been associated with obesity, smoking, and other health risk behaviors.

Mechanisms

The search for pathways that link psychiatric disorders to incident cardiac disease and subsequent cardiac events is ongoing, and many candidate mechanisms have been identified. The links between depression and cardiac outcomes have received the most study. Depression is associated with dysregulation of the autonomic nervous system (ANS) and the hypothalamic–pituitary–adrenal (HPA) axis, including higher levels of plasma and urinary catecholamines and cortisol, higher resting and mean 24-hour heart rates, and lower heart rate variability (see Fig. 99.1 ). Other studies have found elevated proinflammatory cytokines, acute-phase proteins, chemokines, and adhesion molecules, including increased levels of C-reactive protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor (TNF). There is also evidence that depressed patients with CHD have elevated markers of coagulation and platelet activity, especially β-thromboglobulin and platelet factor 4. However, none of these abnormalities is present in every depressed patient, and the proportion of the effect of depression on incident CHD or cardiac events (including mortality) explained by these factors is modest. This suggests that several pathways may be involved and that they differ across individuals.

In addition to the physiological mechanisms, there are behavioral characteristics of depressed patients that likely contribute to the increased risk for CVD and adverse cardiac outcomes. Depressed patients are more likely to be sedentary, to smoke, and to engage in other unhealthy behaviors (e.g., poor diet, higher alcohol consumption) (see Fig. 99.1 ). Furthermore, depression predicts poor adherence to medication regimens, risk factor modification interventions including dietary regimens and smoking cessation programs, and cardiac rehabilitation.

Shared Features Across Mental Disorders

As discussed below, there is now compelling evidence that a wide range of negative emotional states and psychiatric disorders increase the risk for incident CVD and for cardiac events in patients with established CVD. These observations have led to growing interest in elucidating common elements that might explain the risks. One such effort has been the attempt to identify underlying personality dimensions and temperaments associated with these psychiatric disorders and with the susceptibility to the effects of acute and chronic stress that might explain their CV effects. Type D (distressed) personality, which consists of a combination of neuroticism and social inhibition, is an example. Neuroticism is a personality trait that in itself has been associated with depression and anxiety disorders. The hypothesis that a single underlying personality type explains much of the effect of different negative affective states on CVD is intuitively appealing. It might also be more efficient to study a single unifying disorder than the growing number of negative affective states and disorders (e.g., distress, anger, hostility, depression, general anxiety, panic disorder, phobias, PTSD, vital exhaustion) that have been identified as risk factors for CVD. These disorders are often comorbid, and they share many of the same symptoms and many of the same putative mechanisms that may explain their effect on CVD (e.g., ANS dysfunction, increased inflammatory activity, poor diet, insufficient exercise, smoking). Furthermore, many drugs considered to be primarily antidepressants, and many forms of psychotherapy including cognitive behavior therapy (CBT), are used to treat depression, anxiety disorders, PTSD, and psychosocial distress associated with stressful situations. Thus, there is both mechanistic and therapeutic overlap between these ostensibly distinct psychiatric disorders, suggesting that an integrative approach to their evaluation and management may be warranted. Nonetheless, most research continues to study these affective states and psychiatric disorders individually as separate albeit related entities.

Anxiety

The 12-month prevalence of anxiety disorders in the United States is about 18%, and the lifetime prevalence is about 30% in women and 19% in men. There is evidence that anxiety is a risk factor for incident CHD, as well as atrial and ventricular arrhythmias. However, most studies of anxiety as a risk factor for cardiac morbidity and mortality have used self-report anxiety symptom questionnaires. There have been fewer studies of clinically diagnosed anxiety disorders.

In summary, there is moderate evidence that anxiety is a risk factor for incident CHD and cardiac events in patients with established CHD. However, there are fewer studies of anxiety than of depression as predictors of cardiac outcomes, and not all studies have found anxiety to be a significant independent predictor of incident CHD or cardiac events. Nevertheless, the consensus among most experts is that anxiety is likely to be a risk factor for incident CHD. More research is needed to determine whether this risk differs by type of anxiety disorder and the extent to which the effects of anxiety are independent from those of depression.