Wei Jiang, MD

Depression is a risk factor for heart disease, and in patients with heart disease, it is a risk factor for complications and death. Unfortunately, in the trials performed to date, treating depression in cardiac patients did not lead to lower rates of recurrent cardiovascular events or death. Nevertheless, we recommend that clinicians systematically screen for it in their heart patients, in view of the benefits of antidepressant therapy.

In this article we review key epidemiologic and psychosocial studies, the mechanistic links between depression and heart disease, and recent intervention trials. We also offer practical management advice and address the continued need for guidelines and risk stratification in the treatment of depressed cardiac patients.

After we submitted our review article, the American Heart Association (AHA)¹ released a consensus document recommending that health care providers screen for and treat depression in patients with coronary heart disease. We will discuss the same screening tests that have been recommended by the AHA.

DEPRESSION AND HEART DISEASE: COMMON AND LINKED

Depression and heart disease are very common and often coexist: the prevalence of depression in various heart conditions ranges from 15% to 20%.^1–3 According to data from the World Health Organization, by the year 2020 depression will be the second-leading cause of disability in developed countries (after heart disease).⁴

The World Health Survey⁵ showed that depression worsens health more than angina, arthritis, asthma, or diabetes. Furthermore, patients with severe mental illness have a higher risk of dying from heart disease and stroke.⁶

SOME HEART DISEASE RISK FACTORS ARE PSYCHOSOCIAL

In the 1980s, the “type A” personality (ambitious, aggressive, hostile, and competitive, with a chronic sense of urgency) was linked to heart disease.⁷ Later studies differed as to whether the entire set of features is valid as a collective risk factor for progressive heart disease,⁸ but hostility remains a validated risk factor and a focus of behavior modification.^9,10

Other psychosocial risk factors have been implicated,^11,12 one of which is social isolation.^9,13 Another is the “type D” personality, which includes a tendency to experience negative emotions across time and situations coupled with social inhibition and which is believed to be more valid than the type A personality as a risk factor for cardiac disease.^14,15

The INTERHEART study¹⁶ gathered data about attributable risk in the development of myocardial infarction (MI) in 52 countries in a case-control fashion. Psychosocial factors including stress, low generalized locus of control (ie, the perceived inability to control one’s life), and depression accounted for 32.5% of the attributable risk for an MI.¹⁷ This would mean that they account for slightly less attributable risk than that of lifetime smoking but more than that of hypertension and obesity.

Job stress increases the risk of initial coronary heart disease¹⁸ and also the risk of recurrent cardiac events after a first MI.¹⁹ Even though numerous psychosocial risk factors have been associated with coronary heart disease, including anxiety,^20,21 depression is perhaps the best studied.

PROSPECTIVE STUDIES OF DEPRESSION AND HEART DISEASE

To examine the impact of depression in coronary heart disease, prospective studies have been done in healthy people and in patients with established cardiovascular disease who develop depression.²²

In healthy people, depression increases the risk of coronary disease

The 1996 Epidemiologic Catchment Area study²³ found that people with major depression had a risk of MI four times higher than the norm, and people with 2 weeks of sadness or dysphoria had a risk two times higher.

A subsequent meta-analysis of 11 studies,²⁴ which included 36,000 patients, found that the overall relative risk of developing heart disease in depressed but healthy people was 1.64.

A meta-analysis by Van der Kooy et al²⁵ of 28 epidemiologic studies with nearly 80,000 patients showed depression to be an independent risk factor for cardiovascular disease.

Wulsin and Singal²⁶ performed a systematic review to see if depression increases the risk of coronary disease. In 10 studies with a follow-up of more than 4 years, the relative risk in people with depression was 1.64, which was less than that in active smokers (2.5) but more than that in passive smokers (1.25).

Depression can also exacerbate the classic risk factors for coronary disease, such as smoking, diabetes, obesity, and physical inactivity. ²⁷

A 2007 study from Sweden²⁸ prospectively followed patients who were hospitalized for depression. The odds ratio of developing an acute MI was 2.9, and the risk persisted for decades after the initial hospitalization.

A prospective United Kingdom cohort study of people initially free of heart disease revealed major depression to be associated with a higher rate of death from ischemic heart disease.²⁹ Specifically, patients who had depression currently or in the past 12 months had a 2.7 times higher risk of dying than those who had never had depression or who had had it more than 12 months previously.

In existing heart disease, depression predicts recurrent events, death

Carney et el³⁰ found that patients with major depressive disorder had a higher incidence of new cardiac events in the 12 months after undergoing cardiac catheterization than those without major depressive disorder.

Frasure-Smith et al,³¹ in a landmark study, showed that patients who were depressed at 1 week after an MI were three to four times more likely to die in the next 6 months than nondepressed post-MI patients. Even after 18 months, depression remained an independent risk factor for cardiac-related death.³²

In longer studies (with up to 19.4 years of follow-up), depression was associated with higher rates of death from cardiac and all causes in patients with coronary artery disease.³³ Lespérance et al³⁴ found that in MI patients, the higher the Beck Depression Inventory score at the time of hospital admission, the higher the 5-year death rate.

Using meta-analysis, Barth et al³⁵ found the risk of dying in the first 2 years after initial assessment to be twice as high in depressed cardiac patients as in nondepressed cardiac patients (odds ratio 2.24).

Van Melle et al³⁶ reviewed 22 studies and found that in the 2 years after an MI, depressed patients had a 2 to 2.5 times higher risk of dying of a cardiac or any other cause than did nondepressed patients.

Depression also predicts higher morbidity and mortality rates in patients undergoing coronary artery bypass grafting,^37,38 patients with congestive heart failure,³⁹ and heart transplant recipients.⁴⁰

 

MEDICAL ILLNESS CAN PREDISPOSE TO DEPRESSION, AND VICE VERSA

Medical illnesses can predispose a patient to develop depression. Specifically, compared with healthy people, cardiac patients appear to be at greater risk of developing depression for many years after the initial medical diagnosis is made.⁴¹

Katon et al⁴² reviewed 31 studies involving 16,922 patients, that assessed the impact of depression and anxiety in chronic medical illnesses such as heart disease, diabetes, pulmonary disease, and arthritis. After the severity of the medical disorder was controlled for, patients with depression and anxiety reported a higher number of medical symptoms.

DEPRESSION WORSENS QUALITY OF LIFE AND ADHERENCE TO TREATMENT

Depressed patients perceive their health status and quality of life negatively. In the Heart and Soul study,⁴³ depressive symptoms and low exercise capacity—but not low ejection fraction or ischemia—were significantly associated with perceived deterioration of health in patients with coronary artery disease.

After an MI, patients who take their cardiac drugs properly have a better chance of survival.^44,45 Clinical depression can worsen compliance with cardiac medication regimens,⁴⁶ and reducing depression increases medication adherence overall.⁴⁷ Not surprisingly, depressed patients also adhere less well to other recommendations,⁴⁸ including modifying the diet, exercising, stopping smoking, and attending cardiac rehabilitation programs. ⁴⁹

PLAUSIBLE MECHANISMS LINK DEPRESSION AND HEART DISEASE

Traditional cardiac risk factors such as smoking, high cholesterol, hypertension, diabetes, and obesity tend to cluster in depressed patients. ⁵⁰ Other mechanisms linking depression and heart disease are reviewed below.^51,52

Autonomic imbalance

Excessive sympathetic stimulation or diminished vagal stimulation or both are associated with higher rates of morbidity and death.⁵³

Lack of variability in the heart rate reflects a sympathetic-vagal imbalance and is a risk factor for ventricular arrhythmias and sudden cardiac death in patients with cardiovascular disease.⁵⁴ Carney et al⁵⁵ reported that patients with coronary artery disease and depression had significantly less heart rate variability than nondepressed cardiac patients. Similarly, after an MI, depressed patients had significantly less heart rate variability than nondepressed patients,⁵⁶ implying that low heart rate variability may mediate the adverse effect of depression on survival after an MI.⁵⁷

In the Heart and Soul study, Gehi et al⁵⁸ found no distinct relationship between heart rate variability and depression. However, in the same study, de Jong et al⁵⁹ did find specific somatic symptoms of depression to be associated with lower heart rate variability, although cognitive symptoms were not.

Platelet activation, endothelial dysfunction

Depressed patients have been found to have exaggerated platelet reactivity.⁶⁰ Plasma levels of platelet factor IV and beta-thromboglobulin, markers of platelet activation, are higher in depressed patients with ischemic heart disease than in nondepressed patients with ischemic heart disease and in control patients.⁶¹ This activation of platelets can lead to vascular damage and thrombosis.

In a subset study of the Sertraline Anti-Depressant Heart Attack Randomized Trial (SADHART), depressed MI patients were treated with sertraline (Zoloft), a selective serotonin reuptake inhibitor (SSRI), and had substantially less platelet and endothelial biomarker release.⁶²

Depressed cardiac patients also have impaired flow-mediated dilation of the brachial artery, a sign of endothelial dysfunction.⁶³ Although a recent study did not find coronary endothelial dysfunction in depressed patients who did not have cardiac disease, these patients had more clustering of other cardiac risk factors.⁶⁴

Hypothalamic-pituitary-adrenocortical and sympathetic adrenal medullary activation

High cortisol levels can accelerate the development of hypertension and atherosclerosis and result in endothelial vascular injury. Sympathoadrenal activation in turn can lead to higher levels of catecholamines, predisposing to vasoconstriction, a rapid heart rate, and platelet activation. Depressed patients have more activation of the hypothalamic-pituitary-adrenocortical and sympathetic adrenal medullary systems,^51,65 yet another plausible mechanism for worse clinical outcomes in depressed cardiac patients.

Sudden emotional stress can cause transient left ventricular dysfunction, even in people without coronary disease, an effect that may be mediated by elevated plasma catecholamine levels.⁶⁶

Inflammatory cytokines

Inflammatory cytokines play a key role in the development of atherosclerosis.⁶⁷ C-reactive protein, an acute-phase reactant produced in hepatocytes, can be induced by cytokines such as interleukin 6. Damage to endothelial tissues leads to the release of inflammatory cytokines, including interleukin 1, interleukin 6, and tumor tumor necrosis factor alpha.

Depressed patients have higher levels of these inflammatory markers.^68,69 A prospective study reported direct correlations between depression scores and C-reactive protein levels in post-MI patients.⁷⁰ The Heart and Soul study, however, did not confirm that coronary patients have more inflammation if they have depression,⁷¹ indicating that the relationship is complex and is perhaps more evident in specific types of depression.⁷²

Anticholinergic inflammatory pathway

Tracey⁷³ proposed a theory that vagal tone inhibits the release of inflammatory cytokines. This has important implications for treatment, as exercise, biofeedback, and meditation can stimulate the vagus nerve and therefore have beneficial anti-inflammatory effects.⁷⁴

Polymorphism in the serotonin transport promoter region gene

Research is focusing on the serotonin transport promoter region gene (5-HTTLPR).⁷⁵ The gene exists in two forms, a long one and a less-effective short one that appears to predispose to depression.⁷⁶

Nakatani et al⁷⁷ showed that MI patients were more likely to become depressed and to have subsequent cardiac events if one or both of their alleles of this gene were short. Otte et al,⁷⁸ using Heart and Soul study data, found that patients with a short allele had a higher likelihood of depression, higher perceived levels of stress, and higher urinary norepinephrine secretion. However, the long allele genotype may be associated with a higher risk of developing an MI.⁷⁹

Our knowledge of the genetic interplay of depression and cardiovascular disease is still in its infancy, and further studies are needed to clarify these findings.

 

IN TRIALS, LESS DEPRESSION BUT NO EFFECT ON DEATHS, RECURRENT MI

Major behavioral and drug trials conducted in the last 15 years have focused on how to best treat depression in cardiac patients.^80–85

The Montreal Heart Attack Readjustment Trial (MHART)⁸¹ used telephone calls and home nursing visits to explore and monitor psychological distress for up to 1 year after an MI. The overall trial did not show these interventions to have any impact on survival compared with usual care. In fact, in women receiving the telephone intervention, there was a trend toward higher rates of cardiac and all-cause death, which was quite unexpected. Uncovering stresses and problems without resolving them, rather than encouraging patients to place these on the “back burner,” may partially explain these results.

SADHART⁸² studied the safety of sertraline in depressed post-MI patients. No major differences in cardiac function were noted between the sertraline and placebo groups, showing that sertraline was safe for these patients. The sertraline group had fewer cardiovascular events, but the difference was not statistically significant.

The Enhancing Recovery in Coronary Heart Disease (ENRICHD) study⁸³ was primarily designed to see whether a psychosocial intervention would decrease deaths in depressed cardiac patients. Much to the chagrin of behavioral medicine, the group undergoing cognitive behavioral therapy did not have a higher rate of event-free survival, although the intervention had a favorable impact on depression and social support.

The Myocardial Infarction Depression Intervention Trial (MIND-IT)⁸⁴ looked at whether the antidepressant mirtazapine (Remeron) would improve long-term depression and cardiovascular outcomes in depressed post-MI patients. In 18 months of follow-up, neither objective was obtained.

The Cardiac Randomized Evaluation of Antidepressant and Psychotherapy Efficacy (CREATE) trial⁸⁵ tested the efficacy of the SSRI citalopram (Celexa) and interpersonal therapy in a short-term intervention. Here, the antidepressant was superior to placebo in the primary outcome of treating depression, but interpersonal therapy had no advantage over “clinical management,” ie, a shorter, 20-minute supportive intervention.

Common threads in these studies.

• In ENRICHD and MIND-IT, patients whose depression did not respond to treatment were at higher risk of cardiac events during follow-up.^86–88
• In SADHART and CREATE, which used drug treatment, the antidepressant response was more robust in patients with a history of depression before their heart attacks, suggesting that a patient with recurrent depression at the time of a cardiac event should receive medication for it.^85,89

CLINICAL RECOMMENDATIONS

Use a depression screening tool

Ziegelstein et al⁹⁰ recently studied the ability of clinical personnel to detect depression in hospitalized MI patients. If a screening tool was not used, the results were abysmal, indicating the need to use formal screening for symptoms of depression in acute MI patients.

Many self-rating scales are available, among which are the Beck Depression Inventory (BDI) and the Hospital Anxiety and Depression Scale (HADS). Others are:

The Patient Health Questionnaire (PHQ-9) is a nine-item tool, easy to administer and score (Table 1). It has been well studied in both screening for and follow-up of depression in primary care.^91,92 It was used in the Heart and Soul study and the Prospective Registry Evaluating Outcomes After Myocardial Infarction: Events and Recovery (PREMIER) study.³ It has also been used to identify and document depressive symptoms in patients with acute coronary syndrome.⁹⁴ A cut-off score of 10 or higher on the PHQ-9 is diagnostic of depression.⁹⁵

The PHQ-2 consists of the two first questions of the PHQ-9, which deal with mood and lack of pleasure. A cut-off score of 3 or higher has a sensitivity of 83% and a specificity of 92%,⁹⁶ fulfilling the need for a quick and reliable depression screening tool. The clinician can also ask for a yes-or-no answer to the two questions of the PHQ-2 (Table 1). A yes to either of the two questions is up to 90% sensitive and 75% specific.^92,97

When to suspect depression in cardiac patients

Cardiac patients may not realize they have the classic symptoms of depression, since they often ascribe somatic symptoms to their heart disease and overlook emotional associations. Lespérance and colleagues⁹⁸ suggest that certain clues should make us suspect depression in cardiac patients (Table 2).

Which type of psychotherapy is best?

The negative results of psychosocial interventions (phone calls and home visits from a nurse) in MHART and of cognitive behavioral therapy in ENRICHD raise questions about which type of psychotherapy is best for depression in heart disease. CREATE found that 50-minute weekly sessions of interpersonal psychotherapy were no more beneficial than clinical management, ie, 20-minute weekly sessions that focused on compliance with treatment and education about depression and overall management. Perhaps a type of therapy akin to “clinical management” in this study or the brief behavior-based and targeted therapy used in the Improving Mood Promoting Access to Collaborative Care Treatment (IMPACT) trials of depression in primary care⁹⁹ could be designed specifically to treat depression in cardiac disease. However, it is also quite possible that treatments that focus on uncovering stresses or problems may not be timely for these patients.

Which therapy is best for women is another area of consideration. In MHART, even after 5 years of follow-up,¹⁰⁰ women who received the psychosocial support intervention did marginally worse. In the ENRICHD study, women did not experience a benefit from cognitive behavioral therapy. Further studies must address sex differences in response to different therapies.

 

 

SSRIs seem to be better than other antidepressants for cardiac patients

Before SSRIs were available, tricyclic antidepressants were the mainstays. Subsequent analysis showed the tricyclics to have an unfavorable risk-benefit profile in cardiac patients,¹⁰¹ and since other types of antidepressants are available, tricyclics should be avoided altogether in cardiac patients.¹⁰²

Whether the SSRIs actually decrease one’s risk of death in heart disease is still an issue of debate, but there are encouraging signs. In SADHART, the rate of death and recurrent nonfatal MI was 20% lower in the patients randomized to receive sertraline, although the difference was not statistically significant.⁸² In ENRICHD, patients who did not respond to cognitive behavioral treatment or had severe depression could receive sertraline or other antidepressant drugs on a nonrandomized basis, and those who did had a 42% lower incidence of death or recurrent MI.¹⁰³

The SADHART and CREATE trials provide convincing evidence of the cardiac safety and antidepressant efficacy of two SSRIs (sertraline and citalopram) in depressed cardiac patients. Mirtazapine, studied in MIND-IT, was not effective in treating depression in cardiac patients, although it had a better adverse effect and safety profile than tricyclic antidepressants. ¹⁰⁴

Clinical observations indicate that SSRIs are associated with less risk of MI than non-SSRI drugs.^105,106 During hospitalization for acute coronary syndromes, patients on SSRIs had lower rates of recurrent ischemia and heart failure but higher bleeding rates than patients not taking SSRIs.¹⁰⁷ In a retrospective study of patients undergoing coronary artery bypass grafting, those on an SSRI before surgery had higher rates of death and rehospitalization.¹⁰⁸ Being on antidepressant medication could be interpreted as a surrogate marker of having more severe depression before surgery; this issue clearly requires further study.

Given current observations and recent data from interventional trials coupled with the safe drug-interaction profile of sertraline and citalopram, these two SSRIs are recommended for treating depression in cardiac patients. If the patient is also receiving an anticoagulant, one should monitor for bleeding, as all SSRIs are associated with a prolonged bleeding time. Monitoring for rare cases of hyponatremia and bradycardia should also be part of early follow-up.

Do cardiac drugs have psychiatric effects?

Some concerns have arisen about cardiovascular drugs causing or aggravating psychiatric conditions.

Statins were once suspected of causing clinical depression or even suicide. However, subsequent studies have not substantiated this.^109,110 In fact, long-term statin use has been associated with improved psychological wellbeing. ¹¹¹ Whether the favorable psychological profile is due to an improved lifestyle, a direct noncholesterol effect, or an immunomodulatory effect has yet to be determined.

Beta-blockers have been suspected of increasing depression and fatigue. Robust metaanalyses have shown no increased risk of depressive symptoms but a small increased risk of fatigue and sexual dysfunction.¹¹² Observational trials in the first year post-MI have shown no differences between beta-blocker users and nonusers in depressive symptoms or depressive disorders.¹¹³

Statins and beta-blockers offer both immense cardiac benefit and low risk, and both may be prescribed with confidence in depressed cardiac patients.

Refer patients for cardiac rehabilitation

The American Association of Cardiovascular and Pulmonary Rehabilitation strongly recommends screening cardiac patients for depression and referring them to cardiac rehabilitation programs.¹¹⁴ Typical programs run 12 weeks, affording an opportunity to further listen to and assess the patient and to promote general wellness via nutrition, stress management, and exercise.

These interventions by themselves can favorably affect depression. Blumenthal and colleagues,¹¹⁵ in the Standard Medical Intervention and Long-Term Exercise (SMILE) study, found that exercise was as effective as drug treatment in reducing depression. In addition, stress management as a psychosocial treatment in cardiac rehabilitation can reduce death rates in cardiac patients.¹¹⁶

Unfortunately, many patients who are eligible for cardiac rehabilitation programs do not avail themselves of them.¹¹⁷

Our algorithm

In view of the data outlined in this review, we propose an algorithm for use in depressed cardiac patients (Figure 2), which is similar to the algorithm proposed by the AHA committee,¹ but which we developed independently.

FUTURE DIRECTIONS FOR RESEARCH

Can we predict the course of depression?

We need to identify better which patients will have a spontaneous remission of their depressive symptoms after a cardiac event, which patients will linger with depression, and which patients will best respond to treatment. Risk stratification, using the psychiatric history, symptoms and severity of depression, and genetic predisposition¹¹⁸ might allow improved targeted therapies.

Does depression cause cardiac disease?

The link between depression and heart disease can be seen as merely an association. In the interventional trials performed to date, we have not yet seen a reduction in cardiac deaths when depression was treated, challenging any assumption of a causal relationship between depression and heart disease. The debate about association vs cause is germane to behavioral medicine,¹¹⁹ and the better we understand the mechanistic pathways, the better we can advise patients and treat depression comorbid with heart disease.

Behavioral medicine is currently measuring the aspects of depression associated with cardiac disease, including the spectrum of somatic (body) and affective (mood) symptoms¹²⁰ and specific areas such as sympathetic arousal and early morning insomnia.¹²¹ If we can determine the depression subtype that carries a worse cardiac prognosis, we may untangle the biobehavioral links that bidirectionally bridge clinical depression and cardiac disease.

Another area of interest, emotional vitality (a positive state associated with interest, enthusiasm, excitement, and energy for living) has been shown to protect against coronary heart disease¹²² and holds much promise.

In the plenary lecture of the Academy of Psychosomatic Medicine in 2006, Frasure-Smith spoke of the “pleiotropism” of our antidepressant interventions on the various risk factors in depressed cardiac patients.¹²³ We need behavioral medicine studies that elucidate these mechanisms, guiding more precise treatments as well as novel therapies. Omega-3 fatty acids, which benefit heart disease and clinical depression,¹²⁴ will be used in a randomized controlled trial by Lespérance and colleagues.¹²⁵ We await the results of this exciting research.

 

 

Will treating depression help in other types of heart disease?

The SADHART-CHF trial is examining whether 12 weeks of sertraline therapy is better than placebo in preventing death and improving cardiac outcomes in patients with chronic heart failure and comorbid major depressive disorder. It was to be completed in the fall of 2008. The results and experience of this study will help in designing future interventional trials to reduce the risk of depression in cardiovascular diseases.

We also await the results of a National Heart, Lung, and Blood Institute (NHLBI) trial, “Bypassing the Blues,” which is studying the treatment of depression after cardiac bypass surgery. This study should provide further insights into management of the depressed cardiac patient. Further prognostic studies in cardiac patients are also needed using the PHQ-9 and its shorter version, PHQ-2.

Current and future guidelines

For years our European colleagues have been ahead of us in recognizing depression screening and stress management as key to cardiac disease-prevention strategies.¹²⁶ The NHLBI nicely outlined recommendations on the assessment and treatment of depression in cardiovascular patients.¹²⁷ The just-published AHA Science Advisory should further encourage clinicians to screen and treat depression in the patient population.¹ As our knowledge grows, we look forward to future evidence-based guidelines for depressed cardiac patients.

Depression is a risk factor for heart disease, and in patients with heart disease, it is a risk factor for complications and death. Unfortunately, in the trials performed to date, treating depression in cardiac patients did not lead to lower rates of recurrent cardiovascular events or death. Nevertheless, we recommend that clinicians systematically screen for it in their heart patients, in view of the benefits of antidepressant therapy.

In this article we review key epidemiologic and psychosocial studies, the mechanistic links between depression and heart disease, and recent intervention trials. We also offer practical management advice and address the continued need for guidelines and risk stratification in the treatment of depressed cardiac patients.

After we submitted our review article, the American Heart Association (AHA)¹ released a consensus document recommending that health care providers screen for and treat depression in patients with coronary heart disease. We will discuss the same screening tests that have been recommended by the AHA.

DEPRESSION AND HEART DISEASE: COMMON AND LINKED

Depression and heart disease are very common and often coexist: the prevalence of depression in various heart conditions ranges from 15% to 20%.^1–3 According to data from the World Health Organization, by the year 2020 depression will be the second-leading cause of disability in developed countries (after heart disease).⁴

The World Health Survey⁵ showed that depression worsens health more than angina, arthritis, asthma, or diabetes. Furthermore, patients with severe mental illness have a higher risk of dying from heart disease and stroke.⁶

SOME HEART DISEASE RISK FACTORS ARE PSYCHOSOCIAL

In the 1980s, the “type A” personality (ambitious, aggressive, hostile, and competitive, with a chronic sense of urgency) was linked to heart disease.⁷ Later studies differed as to whether the entire set of features is valid as a collective risk factor for progressive heart disease,⁸ but hostility remains a validated risk factor and a focus of behavior modification.^9,10

Other psychosocial risk factors have been implicated,^11,12 one of which is social isolation.^9,13 Another is the “type D” personality, which includes a tendency to experience negative emotions across time and situations coupled with social inhibition and which is believed to be more valid than the type A personality as a risk factor for cardiac disease.^14,15

The INTERHEART study¹⁶ gathered data about attributable risk in the development of myocardial infarction (MI) in 52 countries in a case-control fashion. Psychosocial factors including stress, low generalized locus of control (ie, the perceived inability to control one’s life), and depression accounted for 32.5% of the attributable risk for an MI.¹⁷ This would mean that they account for slightly less attributable risk than that of lifetime smoking but more than that of hypertension and obesity.

Job stress increases the risk of initial coronary heart disease¹⁸ and also the risk of recurrent cardiac events after a first MI.¹⁹ Even though numerous psychosocial risk factors have been associated with coronary heart disease, including anxiety,^20,21 depression is perhaps the best studied.

PROSPECTIVE STUDIES OF DEPRESSION AND HEART DISEASE

To examine the impact of depression in coronary heart disease, prospective studies have been done in healthy people and in patients with established cardiovascular disease who develop depression.²²

In healthy people, depression increases the risk of coronary disease

The 1996 Epidemiologic Catchment Area study²³ found that people with major depression had a risk of MI four times higher than the norm, and people with 2 weeks of sadness or dysphoria had a risk two times higher.

A subsequent meta-analysis of 11 studies,²⁴ which included 36,000 patients, found that the overall relative risk of developing heart disease in depressed but healthy people was 1.64.

A meta-analysis by Van der Kooy et al²⁵ of 28 epidemiologic studies with nearly 80,000 patients showed depression to be an independent risk factor for cardiovascular disease.

Wulsin and Singal²⁶ performed a systematic review to see if depression increases the risk of coronary disease. In 10 studies with a follow-up of more than 4 years, the relative risk in people with depression was 1.64, which was less than that in active smokers (2.5) but more than that in passive smokers (1.25).

Depression can also exacerbate the classic risk factors for coronary disease, such as smoking, diabetes, obesity, and physical inactivity. ²⁷

A 2007 study from Sweden²⁸ prospectively followed patients who were hospitalized for depression. The odds ratio of developing an acute MI was 2.9, and the risk persisted for decades after the initial hospitalization.

A prospective United Kingdom cohort study of people initially free of heart disease revealed major depression to be associated with a higher rate of death from ischemic heart disease.²⁹ Specifically, patients who had depression currently or in the past 12 months had a 2.7 times higher risk of dying than those who had never had depression or who had had it more than 12 months previously.

In existing heart disease, depression predicts recurrent events, death

Carney et el³⁰ found that patients with major depressive disorder had a higher incidence of new cardiac events in the 12 months after undergoing cardiac catheterization than those without major depressive disorder.

Frasure-Smith et al,³¹ in a landmark study, showed that patients who were depressed at 1 week after an MI were three to four times more likely to die in the next 6 months than nondepressed post-MI patients. Even after 18 months, depression remained an independent risk factor for cardiac-related death.³²

In longer studies (with up to 19.4 years of follow-up), depression was associated with higher rates of death from cardiac and all causes in patients with coronary artery disease.³³ Lespérance et al³⁴ found that in MI patients, the higher the Beck Depression Inventory score at the time of hospital admission, the higher the 5-year death rate.

Using meta-analysis, Barth et al³⁵ found the risk of dying in the first 2 years after initial assessment to be twice as high in depressed cardiac patients as in nondepressed cardiac patients (odds ratio 2.24).

Van Melle et al³⁶ reviewed 22 studies and found that in the 2 years after an MI, depressed patients had a 2 to 2.5 times higher risk of dying of a cardiac or any other cause than did nondepressed patients.

Depression also predicts higher morbidity and mortality rates in patients undergoing coronary artery bypass grafting,^37,38 patients with congestive heart failure,³⁹ and heart transplant recipients.⁴⁰

 

MEDICAL ILLNESS CAN PREDISPOSE TO DEPRESSION, AND VICE VERSA

Medical illnesses can predispose a patient to develop depression. Specifically, compared with healthy people, cardiac patients appear to be at greater risk of developing depression for many years after the initial medical diagnosis is made.⁴¹

Katon et al⁴² reviewed 31 studies involving 16,922 patients, that assessed the impact of depression and anxiety in chronic medical illnesses such as heart disease, diabetes, pulmonary disease, and arthritis. After the severity of the medical disorder was controlled for, patients with depression and anxiety reported a higher number of medical symptoms.

DEPRESSION WORSENS QUALITY OF LIFE AND ADHERENCE TO TREATMENT

Depressed patients perceive their health status and quality of life negatively. In the Heart and Soul study,⁴³ depressive symptoms and low exercise capacity—but not low ejection fraction or ischemia—were significantly associated with perceived deterioration of health in patients with coronary artery disease.

After an MI, patients who take their cardiac drugs properly have a better chance of survival.^44,45 Clinical depression can worsen compliance with cardiac medication regimens,⁴⁶ and reducing depression increases medication adherence overall.⁴⁷ Not surprisingly, depressed patients also adhere less well to other recommendations,⁴⁸ including modifying the diet, exercising, stopping smoking, and attending cardiac rehabilitation programs. ⁴⁹

PLAUSIBLE MECHANISMS LINK DEPRESSION AND HEART DISEASE

Traditional cardiac risk factors such as smoking, high cholesterol, hypertension, diabetes, and obesity tend to cluster in depressed patients. ⁵⁰ Other mechanisms linking depression and heart disease are reviewed below.^51,52

Autonomic imbalance

Excessive sympathetic stimulation or diminished vagal stimulation or both are associated with higher rates of morbidity and death.⁵³

Lack of variability in the heart rate reflects a sympathetic-vagal imbalance and is a risk factor for ventricular arrhythmias and sudden cardiac death in patients with cardiovascular disease.⁵⁴ Carney et al⁵⁵ reported that patients with coronary artery disease and depression had significantly less heart rate variability than nondepressed cardiac patients. Similarly, after an MI, depressed patients had significantly less heart rate variability than nondepressed patients,⁵⁶ implying that low heart rate variability may mediate the adverse effect of depression on survival after an MI.⁵⁷

In the Heart and Soul study, Gehi et al⁵⁸ found no distinct relationship between heart rate variability and depression. However, in the same study, de Jong et al⁵⁹ did find specific somatic symptoms of depression to be associated with lower heart rate variability, although cognitive symptoms were not.

Platelet activation, endothelial dysfunction

Depressed patients have been found to have exaggerated platelet reactivity.⁶⁰ Plasma levels of platelet factor IV and beta-thromboglobulin, markers of platelet activation, are higher in depressed patients with ischemic heart disease than in nondepressed patients with ischemic heart disease and in control patients.⁶¹ This activation of platelets can lead to vascular damage and thrombosis.

In a subset study of the Sertraline Anti-Depressant Heart Attack Randomized Trial (SADHART), depressed MI patients were treated with sertraline (Zoloft), a selective serotonin reuptake inhibitor (SSRI), and had substantially less platelet and endothelial biomarker release.⁶²

Depressed cardiac patients also have impaired flow-mediated dilation of the brachial artery, a sign of endothelial dysfunction.⁶³ Although a recent study did not find coronary endothelial dysfunction in depressed patients who did not have cardiac disease, these patients had more clustering of other cardiac risk factors.⁶⁴

Hypothalamic-pituitary-adrenocortical and sympathetic adrenal medullary activation

High cortisol levels can accelerate the development of hypertension and atherosclerosis and result in endothelial vascular injury. Sympathoadrenal activation in turn can lead to higher levels of catecholamines, predisposing to vasoconstriction, a rapid heart rate, and platelet activation. Depressed patients have more activation of the hypothalamic-pituitary-adrenocortical and sympathetic adrenal medullary systems,^51,65 yet another plausible mechanism for worse clinical outcomes in depressed cardiac patients.

Sudden emotional stress can cause transient left ventricular dysfunction, even in people without coronary disease, an effect that may be mediated by elevated plasma catecholamine levels.⁶⁶

Inflammatory cytokines

Inflammatory cytokines play a key role in the development of atherosclerosis.⁶⁷ C-reactive protein, an acute-phase reactant produced in hepatocytes, can be induced by cytokines such as interleukin 6. Damage to endothelial tissues leads to the release of inflammatory cytokines, including interleukin 1, interleukin 6, and tumor tumor necrosis factor alpha.

Depressed patients have higher levels of these inflammatory markers.^68,69 A prospective study reported direct correlations between depression scores and C-reactive protein levels in post-MI patients.⁷⁰ The Heart and Soul study, however, did not confirm that coronary patients have more inflammation if they have depression,⁷¹ indicating that the relationship is complex and is perhaps more evident in specific types of depression.⁷²

Anticholinergic inflammatory pathway

Tracey⁷³ proposed a theory that vagal tone inhibits the release of inflammatory cytokines. This has important implications for treatment, as exercise, biofeedback, and meditation can stimulate the vagus nerve and therefore have beneficial anti-inflammatory effects.⁷⁴

Polymorphism in the serotonin transport promoter region gene

Research is focusing on the serotonin transport promoter region gene (5-HTTLPR).⁷⁵ The gene exists in two forms, a long one and a less-effective short one that appears to predispose to depression.⁷⁶

Nakatani et al⁷⁷ showed that MI patients were more likely to become depressed and to have subsequent cardiac events if one or both of their alleles of this gene were short. Otte et al,⁷⁸ using Heart and Soul study data, found that patients with a short allele had a higher likelihood of depression, higher perceived levels of stress, and higher urinary norepinephrine secretion. However, the long allele genotype may be associated with a higher risk of developing an MI.⁷⁹

Our knowledge of the genetic interplay of depression and cardiovascular disease is still in its infancy, and further studies are needed to clarify these findings.

 

IN TRIALS, LESS DEPRESSION BUT NO EFFECT ON DEATHS, RECURRENT MI

Major behavioral and drug trials conducted in the last 15 years have focused on how to best treat depression in cardiac patients.^80–85

The Montreal Heart Attack Readjustment Trial (MHART)⁸¹ used telephone calls and home nursing visits to explore and monitor psychological distress for up to 1 year after an MI. The overall trial did not show these interventions to have any impact on survival compared with usual care. In fact, in women receiving the telephone intervention, there was a trend toward higher rates of cardiac and all-cause death, which was quite unexpected. Uncovering stresses and problems without resolving them, rather than encouraging patients to place these on the “back burner,” may partially explain these results.

SADHART⁸² studied the safety of sertraline in depressed post-MI patients. No major differences in cardiac function were noted between the sertraline and placebo groups, showing that sertraline was safe for these patients. The sertraline group had fewer cardiovascular events, but the difference was not statistically significant.

The Enhancing Recovery in Coronary Heart Disease (ENRICHD) study⁸³ was primarily designed to see whether a psychosocial intervention would decrease deaths in depressed cardiac patients. Much to the chagrin of behavioral medicine, the group undergoing cognitive behavioral therapy did not have a higher rate of event-free survival, although the intervention had a favorable impact on depression and social support.

The Myocardial Infarction Depression Intervention Trial (MIND-IT)⁸⁴ looked at whether the antidepressant mirtazapine (Remeron) would improve long-term depression and cardiovascular outcomes in depressed post-MI patients. In 18 months of follow-up, neither objective was obtained.

The Cardiac Randomized Evaluation of Antidepressant and Psychotherapy Efficacy (CREATE) trial⁸⁵ tested the efficacy of the SSRI citalopram (Celexa) and interpersonal therapy in a short-term intervention. Here, the antidepressant was superior to placebo in the primary outcome of treating depression, but interpersonal therapy had no advantage over “clinical management,” ie, a shorter, 20-minute supportive intervention.

Common threads in these studies.

• In ENRICHD and MIND-IT, patients whose depression did not respond to treatment were at higher risk of cardiac events during follow-up.^86–88
• In SADHART and CREATE, which used drug treatment, the antidepressant response was more robust in patients with a history of depression before their heart attacks, suggesting that a patient with recurrent depression at the time of a cardiac event should receive medication for it.^85,89

CLINICAL RECOMMENDATIONS

Use a depression screening tool

Ziegelstein et al⁹⁰ recently studied the ability of clinical personnel to detect depression in hospitalized MI patients. If a screening tool was not used, the results were abysmal, indicating the need to use formal screening for symptoms of depression in acute MI patients.

Many self-rating scales are available, among which are the Beck Depression Inventory (BDI) and the Hospital Anxiety and Depression Scale (HADS). Others are:

The Patient Health Questionnaire (PHQ-9) is a nine-item tool, easy to administer and score (Table 1). It has been well studied in both screening for and follow-up of depression in primary care.^91,92 It was used in the Heart and Soul study and the Prospective Registry Evaluating Outcomes After Myocardial Infarction: Events and Recovery (PREMIER) study.³ It has also been used to identify and document depressive symptoms in patients with acute coronary syndrome.⁹⁴ A cut-off score of 10 or higher on the PHQ-9 is diagnostic of depression.⁹⁵

The PHQ-2 consists of the two first questions of the PHQ-9, which deal with mood and lack of pleasure. A cut-off score of 3 or higher has a sensitivity of 83% and a specificity of 92%,⁹⁶ fulfilling the need for a quick and reliable depression screening tool. The clinician can also ask for a yes-or-no answer to the two questions of the PHQ-2 (Table 1). A yes to either of the two questions is up to 90% sensitive and 75% specific.^92,97

When to suspect depression in cardiac patients

Cardiac patients may not realize they have the classic symptoms of depression, since they often ascribe somatic symptoms to their heart disease and overlook emotional associations. Lespérance and colleagues⁹⁸ suggest that certain clues should make us suspect depression in cardiac patients (Table 2).

Which type of psychotherapy is best?

The negative results of psychosocial interventions (phone calls and home visits from a nurse) in MHART and of cognitive behavioral therapy in ENRICHD raise questions about which type of psychotherapy is best for depression in heart disease. CREATE found that 50-minute weekly sessions of interpersonal psychotherapy were no more beneficial than clinical management, ie, 20-minute weekly sessions that focused on compliance with treatment and education about depression and overall management. Perhaps a type of therapy akin to “clinical management” in this study or the brief behavior-based and targeted therapy used in the Improving Mood Promoting Access to Collaborative Care Treatment (IMPACT) trials of depression in primary care⁹⁹ could be designed specifically to treat depression in cardiac disease. However, it is also quite possible that treatments that focus on uncovering stresses or problems may not be timely for these patients.

Which therapy is best for women is another area of consideration. In MHART, even after 5 years of follow-up,¹⁰⁰ women who received the psychosocial support intervention did marginally worse. In the ENRICHD study, women did not experience a benefit from cognitive behavioral therapy. Further studies must address sex differences in response to different therapies.

 

 

SSRIs seem to be better than other antidepressants for cardiac patients

Before SSRIs were available, tricyclic antidepressants were the mainstays. Subsequent analysis showed the tricyclics to have an unfavorable risk-benefit profile in cardiac patients,¹⁰¹ and since other types of antidepressants are available, tricyclics should be avoided altogether in cardiac patients.¹⁰²

Whether the SSRIs actually decrease one’s risk of death in heart disease is still an issue of debate, but there are encouraging signs. In SADHART, the rate of death and recurrent nonfatal MI was 20% lower in the patients randomized to receive sertraline, although the difference was not statistically significant.⁸² In ENRICHD, patients who did not respond to cognitive behavioral treatment or had severe depression could receive sertraline or other antidepressant drugs on a nonrandomized basis, and those who did had a 42% lower incidence of death or recurrent MI.¹⁰³

The SADHART and CREATE trials provide convincing evidence of the cardiac safety and antidepressant efficacy of two SSRIs (sertraline and citalopram) in depressed cardiac patients. Mirtazapine, studied in MIND-IT, was not effective in treating depression in cardiac patients, although it had a better adverse effect and safety profile than tricyclic antidepressants. ¹⁰⁴

Clinical observations indicate that SSRIs are associated with less risk of MI than non-SSRI drugs.^105,106 During hospitalization for acute coronary syndromes, patients on SSRIs had lower rates of recurrent ischemia and heart failure but higher bleeding rates than patients not taking SSRIs.¹⁰⁷ In a retrospective study of patients undergoing coronary artery bypass grafting, those on an SSRI before surgery had higher rates of death and rehospitalization.¹⁰⁸ Being on antidepressant medication could be interpreted as a surrogate marker of having more severe depression before surgery; this issue clearly requires further study.

Given current observations and recent data from interventional trials coupled with the safe drug-interaction profile of sertraline and citalopram, these two SSRIs are recommended for treating depression in cardiac patients. If the patient is also receiving an anticoagulant, one should monitor for bleeding, as all SSRIs are associated with a prolonged bleeding time. Monitoring for rare cases of hyponatremia and bradycardia should also be part of early follow-up.

Do cardiac drugs have psychiatric effects?

Some concerns have arisen about cardiovascular drugs causing or aggravating psychiatric conditions.

Statins were once suspected of causing clinical depression or even suicide. However, subsequent studies have not substantiated this.^109,110 In fact, long-term statin use has been associated with improved psychological wellbeing. ¹¹¹ Whether the favorable psychological profile is due to an improved lifestyle, a direct noncholesterol effect, or an immunomodulatory effect has yet to be determined.

Beta-blockers have been suspected of increasing depression and fatigue. Robust metaanalyses have shown no increased risk of depressive symptoms but a small increased risk of fatigue and sexual dysfunction.¹¹² Observational trials in the first year post-MI have shown no differences between beta-blocker users and nonusers in depressive symptoms or depressive disorders.¹¹³

Statins and beta-blockers offer both immense cardiac benefit and low risk, and both may be prescribed with confidence in depressed cardiac patients.

Refer patients for cardiac rehabilitation

The American Association of Cardiovascular and Pulmonary Rehabilitation strongly recommends screening cardiac patients for depression and referring them to cardiac rehabilitation programs.¹¹⁴ Typical programs run 12 weeks, affording an opportunity to further listen to and assess the patient and to promote general wellness via nutrition, stress management, and exercise.

These interventions by themselves can favorably affect depression. Blumenthal and colleagues,¹¹⁵ in the Standard Medical Intervention and Long-Term Exercise (SMILE) study, found that exercise was as effective as drug treatment in reducing depression. In addition, stress management as a psychosocial treatment in cardiac rehabilitation can reduce death rates in cardiac patients.¹¹⁶

Unfortunately, many patients who are eligible for cardiac rehabilitation programs do not avail themselves of them.¹¹⁷

Our algorithm

In view of the data outlined in this review, we propose an algorithm for use in depressed cardiac patients (Figure 2), which is similar to the algorithm proposed by the AHA committee,¹ but which we developed independently.

FUTURE DIRECTIONS FOR RESEARCH

Can we predict the course of depression?

We need to identify better which patients will have a spontaneous remission of their depressive symptoms after a cardiac event, which patients will linger with depression, and which patients will best respond to treatment. Risk stratification, using the psychiatric history, symptoms and severity of depression, and genetic predisposition¹¹⁸ might allow improved targeted therapies.

Does depression cause cardiac disease?

The link between depression and heart disease can be seen as merely an association. In the interventional trials performed to date, we have not yet seen a reduction in cardiac deaths when depression was treated, challenging any assumption of a causal relationship between depression and heart disease. The debate about association vs cause is germane to behavioral medicine,¹¹⁹ and the better we understand the mechanistic pathways, the better we can advise patients and treat depression comorbid with heart disease.

Behavioral medicine is currently measuring the aspects of depression associated with cardiac disease, including the spectrum of somatic (body) and affective (mood) symptoms¹²⁰ and specific areas such as sympathetic arousal and early morning insomnia.¹²¹ If we can determine the depression subtype that carries a worse cardiac prognosis, we may untangle the biobehavioral links that bidirectionally bridge clinical depression and cardiac disease.

Another area of interest, emotional vitality (a positive state associated with interest, enthusiasm, excitement, and energy for living) has been shown to protect against coronary heart disease¹²² and holds much promise.

In the plenary lecture of the Academy of Psychosomatic Medicine in 2006, Frasure-Smith spoke of the “pleiotropism” of our antidepressant interventions on the various risk factors in depressed cardiac patients.¹²³ We need behavioral medicine studies that elucidate these mechanisms, guiding more precise treatments as well as novel therapies. Omega-3 fatty acids, which benefit heart disease and clinical depression,¹²⁴ will be used in a randomized controlled trial by Lespérance and colleagues.¹²⁵ We await the results of this exciting research.

 

 

Will treating depression help in other types of heart disease?

The SADHART-CHF trial is examining whether 12 weeks of sertraline therapy is better than placebo in preventing death and improving cardiac outcomes in patients with chronic heart failure and comorbid major depressive disorder. It was to be completed in the fall of 2008. The results and experience of this study will help in designing future interventional trials to reduce the risk of depression in cardiovascular diseases.

We also await the results of a National Heart, Lung, and Blood Institute (NHLBI) trial, “Bypassing the Blues,” which is studying the treatment of depression after cardiac bypass surgery. This study should provide further insights into management of the depressed cardiac patient. Further prognostic studies in cardiac patients are also needed using the PHQ-9 and its shorter version, PHQ-2.

Current and future guidelines

For years our European colleagues have been ahead of us in recognizing depression screening and stress management as key to cardiac disease-prevention strategies.¹²⁶ The NHLBI nicely outlined recommendations on the assessment and treatment of depression in cardiovascular patients.¹²⁷ The just-published AHA Science Advisory should further encourage clinicians to screen and treat depression in the patient population.¹ As our knowledge grows, we look forward to future evidence-based guidelines for depressed cardiac patients.

Depression is a risk factor for heart disease, and in patients with heart disease, it is a risk factor for complications and death. Unfortunately, in the trials performed to date, treating depression in cardiac patients did not lead to lower rates of recurrent cardiovascular events or death. Nevertheless, we recommend that clinicians systematically screen for it in their heart patients, in view of the benefits of antidepressant therapy.

In this article we review key epidemiologic and psychosocial studies, the mechanistic links between depression and heart disease, and recent intervention trials. We also offer practical management advice and address the continued need for guidelines and risk stratification in the treatment of depressed cardiac patients.

After we submitted our review article, the American Heart Association (AHA)¹ released a consensus document recommending that health care providers screen for and treat depression in patients with coronary heart disease. We will discuss the same screening tests that have been recommended by the AHA.

DEPRESSION AND HEART DISEASE: COMMON AND LINKED

Depression and heart disease are very common and often coexist: the prevalence of depression in various heart conditions ranges from 15% to 20%.^1–3 According to data from the World Health Organization, by the year 2020 depression will be the second-leading cause of disability in developed countries (after heart disease).⁴

The World Health Survey⁵ showed that depression worsens health more than angina, arthritis, asthma, or diabetes. Furthermore, patients with severe mental illness have a higher risk of dying from heart disease and stroke.⁶

SOME HEART DISEASE RISK FACTORS ARE PSYCHOSOCIAL

In the 1980s, the “type A” personality (ambitious, aggressive, hostile, and competitive, with a chronic sense of urgency) was linked to heart disease.⁷ Later studies differed as to whether the entire set of features is valid as a collective risk factor for progressive heart disease,⁸ but hostility remains a validated risk factor and a focus of behavior modification.^9,10

Other psychosocial risk factors have been implicated,^11,12 one of which is social isolation.^9,13 Another is the “type D” personality, which includes a tendency to experience negative emotions across time and situations coupled with social inhibition and which is believed to be more valid than the type A personality as a risk factor for cardiac disease.^14,15

The INTERHEART study¹⁶ gathered data about attributable risk in the development of myocardial infarction (MI) in 52 countries in a case-control fashion. Psychosocial factors including stress, low generalized locus of control (ie, the perceived inability to control one’s life), and depression accounted for 32.5% of the attributable risk for an MI.¹⁷ This would mean that they account for slightly less attributable risk than that of lifetime smoking but more than that of hypertension and obesity.

Job stress increases the risk of initial coronary heart disease¹⁸ and also the risk of recurrent cardiac events after a first MI.¹⁹ Even though numerous psychosocial risk factors have been associated with coronary heart disease, including anxiety,^20,21 depression is perhaps the best studied.

PROSPECTIVE STUDIES OF DEPRESSION AND HEART DISEASE

To examine the impact of depression in coronary heart disease, prospective studies have been done in healthy people and in patients with established cardiovascular disease who develop depression.²²

In healthy people, depression increases the risk of coronary disease

The 1996 Epidemiologic Catchment Area study²³ found that people with major depression had a risk of MI four times higher than the norm, and people with 2 weeks of sadness or dysphoria had a risk two times higher.

A subsequent meta-analysis of 11 studies,²⁴ which included 36,000 patients, found that the overall relative risk of developing heart disease in depressed but healthy people was 1.64.

A meta-analysis by Van der Kooy et al²⁵ of 28 epidemiologic studies with nearly 80,000 patients showed depression to be an independent risk factor for cardiovascular disease.

Wulsin and Singal²⁶ performed a systematic review to see if depression increases the risk of coronary disease. In 10 studies with a follow-up of more than 4 years, the relative risk in people with depression was 1.64, which was less than that in active smokers (2.5) but more than that in passive smokers (1.25).

Depression can also exacerbate the classic risk factors for coronary disease, such as smoking, diabetes, obesity, and physical inactivity. ²⁷

A 2007 study from Sweden²⁸ prospectively followed patients who were hospitalized for depression. The odds ratio of developing an acute MI was 2.9, and the risk persisted for decades after the initial hospitalization.

A prospective United Kingdom cohort study of people initially free of heart disease revealed major depression to be associated with a higher rate of death from ischemic heart disease.²⁹ Specifically, patients who had depression currently or in the past 12 months had a 2.7 times higher risk of dying than those who had never had depression or who had had it more than 12 months previously.

In existing heart disease, depression predicts recurrent events, death

Carney et el³⁰ found that patients with major depressive disorder had a higher incidence of new cardiac events in the 12 months after undergoing cardiac catheterization than those without major depressive disorder.

Frasure-Smith et al,³¹ in a landmark study, showed that patients who were depressed at 1 week after an MI were three to four times more likely to die in the next 6 months than nondepressed post-MI patients. Even after 18 months, depression remained an independent risk factor for cardiac-related death.³²

In longer studies (with up to 19.4 years of follow-up), depression was associated with higher rates of death from cardiac and all causes in patients with coronary artery disease.³³ Lespérance et al³⁴ found that in MI patients, the higher the Beck Depression Inventory score at the time of hospital admission, the higher the 5-year death rate.

Using meta-analysis, Barth et al³⁵ found the risk of dying in the first 2 years after initial assessment to be twice as high in depressed cardiac patients as in nondepressed cardiac patients (odds ratio 2.24).

Van Melle et al³⁶ reviewed 22 studies and found that in the 2 years after an MI, depressed patients had a 2 to 2.5 times higher risk of dying of a cardiac or any other cause than did nondepressed patients.

Depression also predicts higher morbidity and mortality rates in patients undergoing coronary artery bypass grafting,^37,38 patients with congestive heart failure,³⁹ and heart transplant recipients.⁴⁰

 

MEDICAL ILLNESS CAN PREDISPOSE TO DEPRESSION, AND VICE VERSA

Medical illnesses can predispose a patient to develop depression. Specifically, compared with healthy people, cardiac patients appear to be at greater risk of developing depression for many years after the initial medical diagnosis is made.⁴¹

Katon et al⁴² reviewed 31 studies involving 16,922 patients, that assessed the impact of depression and anxiety in chronic medical illnesses such as heart disease, diabetes, pulmonary disease, and arthritis. After the severity of the medical disorder was controlled for, patients with depression and anxiety reported a higher number of medical symptoms.

DEPRESSION WORSENS QUALITY OF LIFE AND ADHERENCE TO TREATMENT

Depressed patients perceive their health status and quality of life negatively. In the Heart and Soul study,⁴³ depressive symptoms and low exercise capacity—but not low ejection fraction or ischemia—were significantly associated with perceived deterioration of health in patients with coronary artery disease.

After an MI, patients who take their cardiac drugs properly have a better chance of survival.^44,45 Clinical depression can worsen compliance with cardiac medication regimens,⁴⁶ and reducing depression increases medication adherence overall.⁴⁷ Not surprisingly, depressed patients also adhere less well to other recommendations,⁴⁸ including modifying the diet, exercising, stopping smoking, and attending cardiac rehabilitation programs. ⁴⁹

PLAUSIBLE MECHANISMS LINK DEPRESSION AND HEART DISEASE

Traditional cardiac risk factors such as smoking, high cholesterol, hypertension, diabetes, and obesity tend to cluster in depressed patients. ⁵⁰ Other mechanisms linking depression and heart disease are reviewed below.^51,52

Autonomic imbalance

Excessive sympathetic stimulation or diminished vagal stimulation or both are associated with higher rates of morbidity and death.⁵³

Lack of variability in the heart rate reflects a sympathetic-vagal imbalance and is a risk factor for ventricular arrhythmias and sudden cardiac death in patients with cardiovascular disease.⁵⁴ Carney et al⁵⁵ reported that patients with coronary artery disease and depression had significantly less heart rate variability than nondepressed cardiac patients. Similarly, after an MI, depressed patients had significantly less heart rate variability than nondepressed patients,⁵⁶ implying that low heart rate variability may mediate the adverse effect of depression on survival after an MI.⁵⁷

In the Heart and Soul study, Gehi et al⁵⁸ found no distinct relationship between heart rate variability and depression. However, in the same study, de Jong et al⁵⁹ did find specific somatic symptoms of depression to be associated with lower heart rate variability, although cognitive symptoms were not.

Platelet activation, endothelial dysfunction

Depressed patients have been found to have exaggerated platelet reactivity.⁶⁰ Plasma levels of platelet factor IV and beta-thromboglobulin, markers of platelet activation, are higher in depressed patients with ischemic heart disease than in nondepressed patients with ischemic heart disease and in control patients.⁶¹ This activation of platelets can lead to vascular damage and thrombosis.

In a subset study of the Sertraline Anti-Depressant Heart Attack Randomized Trial (SADHART), depressed MI patients were treated with sertraline (Zoloft), a selective serotonin reuptake inhibitor (SSRI), and had substantially less platelet and endothelial biomarker release.⁶²

Depressed cardiac patients also have impaired flow-mediated dilation of the brachial artery, a sign of endothelial dysfunction.⁶³ Although a recent study did not find coronary endothelial dysfunction in depressed patients who did not have cardiac disease, these patients had more clustering of other cardiac risk factors.⁶⁴

Hypothalamic-pituitary-adrenocortical and sympathetic adrenal medullary activation

High cortisol levels can accelerate the development of hypertension and atherosclerosis and result in endothelial vascular injury. Sympathoadrenal activation in turn can lead to higher levels of catecholamines, predisposing to vasoconstriction, a rapid heart rate, and platelet activation. Depressed patients have more activation of the hypothalamic-pituitary-adrenocortical and sympathetic adrenal medullary systems,^51,65 yet another plausible mechanism for worse clinical outcomes in depressed cardiac patients.

Sudden emotional stress can cause transient left ventricular dysfunction, even in people without coronary disease, an effect that may be mediated by elevated plasma catecholamine levels.⁶⁶

Inflammatory cytokines

Inflammatory cytokines play a key role in the development of atherosclerosis.⁶⁷ C-reactive protein, an acute-phase reactant produced in hepatocytes, can be induced by cytokines such as interleukin 6. Damage to endothelial tissues leads to the release of inflammatory cytokines, including interleukin 1, interleukin 6, and tumor tumor necrosis factor alpha.

Depressed patients have higher levels of these inflammatory markers.^68,69 A prospective study reported direct correlations between depression scores and C-reactive protein levels in post-MI patients.⁷⁰ The Heart and Soul study, however, did not confirm that coronary patients have more inflammation if they have depression,⁷¹ indicating that the relationship is complex and is perhaps more evident in specific types of depression.⁷²

Anticholinergic inflammatory pathway

Tracey⁷³ proposed a theory that vagal tone inhibits the release of inflammatory cytokines. This has important implications for treatment, as exercise, biofeedback, and meditation can stimulate the vagus nerve and therefore have beneficial anti-inflammatory effects.⁷⁴

Polymorphism in the serotonin transport promoter region gene

Research is focusing on the serotonin transport promoter region gene (5-HTTLPR).⁷⁵ The gene exists in two forms, a long one and a less-effective short one that appears to predispose to depression.⁷⁶

Nakatani et al⁷⁷ showed that MI patients were more likely to become depressed and to have subsequent cardiac events if one or both of their alleles of this gene were short. Otte et al,⁷⁸ using Heart and Soul study data, found that patients with a short allele had a higher likelihood of depression, higher perceived levels of stress, and higher urinary norepinephrine secretion. However, the long allele genotype may be associated with a higher risk of developing an MI.⁷⁹

Our knowledge of the genetic interplay of depression and cardiovascular disease is still in its infancy, and further studies are needed to clarify these findings.

 

IN TRIALS, LESS DEPRESSION BUT NO EFFECT ON DEATHS, RECURRENT MI

Major behavioral and drug trials conducted in the last 15 years have focused on how to best treat depression in cardiac patients.^80–85

The Montreal Heart Attack Readjustment Trial (MHART)⁸¹ used telephone calls and home nursing visits to explore and monitor psychological distress for up to 1 year after an MI. The overall trial did not show these interventions to have any impact on survival compared with usual care. In fact, in women receiving the telephone intervention, there was a trend toward higher rates of cardiac and all-cause death, which was quite unexpected. Uncovering stresses and problems without resolving them, rather than encouraging patients to place these on the “back burner,” may partially explain these results.

SADHART⁸² studied the safety of sertraline in depressed post-MI patients. No major differences in cardiac function were noted between the sertraline and placebo groups, showing that sertraline was safe for these patients. The sertraline group had fewer cardiovascular events, but the difference was not statistically significant.

The Enhancing Recovery in Coronary Heart Disease (ENRICHD) study⁸³ was primarily designed to see whether a psychosocial intervention would decrease deaths in depressed cardiac patients. Much to the chagrin of behavioral medicine, the group undergoing cognitive behavioral therapy did not have a higher rate of event-free survival, although the intervention had a favorable impact on depression and social support.

The Myocardial Infarction Depression Intervention Trial (MIND-IT)⁸⁴ looked at whether the antidepressant mirtazapine (Remeron) would improve long-term depression and cardiovascular outcomes in depressed post-MI patients. In 18 months of follow-up, neither objective was obtained.

The Cardiac Randomized Evaluation of Antidepressant and Psychotherapy Efficacy (CREATE) trial⁸⁵ tested the efficacy of the SSRI citalopram (Celexa) and interpersonal therapy in a short-term intervention. Here, the antidepressant was superior to placebo in the primary outcome of treating depression, but interpersonal therapy had no advantage over “clinical management,” ie, a shorter, 20-minute supportive intervention.

Common threads in these studies.

• In ENRICHD and MIND-IT, patients whose depression did not respond to treatment were at higher risk of cardiac events during follow-up.^86–88
• In SADHART and CREATE, which used drug treatment, the antidepressant response was more robust in patients with a history of depression before their heart attacks, suggesting that a patient with recurrent depression at the time of a cardiac event should receive medication for it.^85,89

CLINICAL RECOMMENDATIONS

Use a depression screening tool

Ziegelstein et al⁹⁰ recently studied the ability of clinical personnel to detect depression in hospitalized MI patients. If a screening tool was not used, the results were abysmal, indicating the need to use formal screening for symptoms of depression in acute MI patients.

Many self-rating scales are available, among which are the Beck Depression Inventory (BDI) and the Hospital Anxiety and Depression Scale (HADS). Others are:

The Patient Health Questionnaire (PHQ-9) is a nine-item tool, easy to administer and score (Table 1). It has been well studied in both screening for and follow-up of depression in primary care.^91,92 It was used in the Heart and Soul study and the Prospective Registry Evaluating Outcomes After Myocardial Infarction: Events and Recovery (PREMIER) study.³ It has also been used to identify and document depressive symptoms in patients with acute coronary syndrome.⁹⁴ A cut-off score of 10 or higher on the PHQ-9 is diagnostic of depression.⁹⁵

The PHQ-2 consists of the two first questions of the PHQ-9, which deal with mood and lack of pleasure. A cut-off score of 3 or higher has a sensitivity of 83% and a specificity of 92%,⁹⁶ fulfilling the need for a quick and reliable depression screening tool. The clinician can also ask for a yes-or-no answer to the two questions of the PHQ-2 (Table 1). A yes to either of the two questions is up to 90% sensitive and 75% specific.^92,97

When to suspect depression in cardiac patients

Cardiac patients may not realize they have the classic symptoms of depression, since they often ascribe somatic symptoms to their heart disease and overlook emotional associations. Lespérance and colleagues⁹⁸ suggest that certain clues should make us suspect depression in cardiac patients (Table 2).

Which type of psychotherapy is best?

The negative results of psychosocial interventions (phone calls and home visits from a nurse) in MHART and of cognitive behavioral therapy in ENRICHD raise questions about which type of psychotherapy is best for depression in heart disease. CREATE found that 50-minute weekly sessions of interpersonal psychotherapy were no more beneficial than clinical management, ie, 20-minute weekly sessions that focused on compliance with treatment and education about depression and overall management. Perhaps a type of therapy akin to “clinical management” in this study or the brief behavior-based and targeted therapy used in the Improving Mood Promoting Access to Collaborative Care Treatment (IMPACT) trials of depression in primary care⁹⁹ could be designed specifically to treat depression in cardiac disease. However, it is also quite possible that treatments that focus on uncovering stresses or problems may not be timely for these patients.

Which therapy is best for women is another area of consideration. In MHART, even after 5 years of follow-up,¹⁰⁰ women who received the psychosocial support intervention did marginally worse. In the ENRICHD study, women did not experience a benefit from cognitive behavioral therapy. Further studies must address sex differences in response to different therapies.

 

 

SSRIs seem to be better than other antidepressants for cardiac patients

Before SSRIs were available, tricyclic antidepressants were the mainstays. Subsequent analysis showed the tricyclics to have an unfavorable risk-benefit profile in cardiac patients,¹⁰¹ and since other types of antidepressants are available, tricyclics should be avoided altogether in cardiac patients.¹⁰²

Whether the SSRIs actually decrease one’s risk of death in heart disease is still an issue of debate, but there are encouraging signs. In SADHART, the rate of death and recurrent nonfatal MI was 20% lower in the patients randomized to receive sertraline, although the difference was not statistically significant.⁸² In ENRICHD, patients who did not respond to cognitive behavioral treatment or had severe depression could receive sertraline or other antidepressant drugs on a nonrandomized basis, and those who did had a 42% lower incidence of death or recurrent MI.¹⁰³

The SADHART and CREATE trials provide convincing evidence of the cardiac safety and antidepressant efficacy of two SSRIs (sertraline and citalopram) in depressed cardiac patients. Mirtazapine, studied in MIND-IT, was not effective in treating depression in cardiac patients, although it had a better adverse effect and safety profile than tricyclic antidepressants. ¹⁰⁴

Clinical observations indicate that SSRIs are associated with less risk of MI than non-SSRI drugs.^105,106 During hospitalization for acute coronary syndromes, patients on SSRIs had lower rates of recurrent ischemia and heart failure but higher bleeding rates than patients not taking SSRIs.¹⁰⁷ In a retrospective study of patients undergoing coronary artery bypass grafting, those on an SSRI before surgery had higher rates of death and rehospitalization.¹⁰⁸ Being on antidepressant medication could be interpreted as a surrogate marker of having more severe depression before surgery; this issue clearly requires further study.

Given current observations and recent data from interventional trials coupled with the safe drug-interaction profile of sertraline and citalopram, these two SSRIs are recommended for treating depression in cardiac patients. If the patient is also receiving an anticoagulant, one should monitor for bleeding, as all SSRIs are associated with a prolonged bleeding time. Monitoring for rare cases of hyponatremia and bradycardia should also be part of early follow-up.

Do cardiac drugs have psychiatric effects?

Some concerns have arisen about cardiovascular drugs causing or aggravating psychiatric conditions.

Statins were once suspected of causing clinical depression or even suicide. However, subsequent studies have not substantiated this.^109,110 In fact, long-term statin use has been associated with improved psychological wellbeing. ¹¹¹ Whether the favorable psychological profile is due to an improved lifestyle, a direct noncholesterol effect, or an immunomodulatory effect has yet to be determined.

Beta-blockers have been suspected of increasing depression and fatigue. Robust metaanalyses have shown no increased risk of depressive symptoms but a small increased risk of fatigue and sexual dysfunction.¹¹² Observational trials in the first year post-MI have shown no differences between beta-blocker users and nonusers in depressive symptoms or depressive disorders.¹¹³

Statins and beta-blockers offer both immense cardiac benefit and low risk, and both may be prescribed with confidence in depressed cardiac patients.

Refer patients for cardiac rehabilitation

The American Association of Cardiovascular and Pulmonary Rehabilitation strongly recommends screening cardiac patients for depression and referring them to cardiac rehabilitation programs.¹¹⁴ Typical programs run 12 weeks, affording an opportunity to further listen to and assess the patient and to promote general wellness via nutrition, stress management, and exercise.

These interventions by themselves can favorably affect depression. Blumenthal and colleagues,¹¹⁵ in the Standard Medical Intervention and Long-Term Exercise (SMILE) study, found that exercise was as effective as drug treatment in reducing depression. In addition, stress management as a psychosocial treatment in cardiac rehabilitation can reduce death rates in cardiac patients.¹¹⁶

Unfortunately, many patients who are eligible for cardiac rehabilitation programs do not avail themselves of them.¹¹⁷

Our algorithm

In view of the data outlined in this review, we propose an algorithm for use in depressed cardiac patients (Figure 2), which is similar to the algorithm proposed by the AHA committee,¹ but which we developed independently.

FUTURE DIRECTIONS FOR RESEARCH

Can we predict the course of depression?

We need to identify better which patients will have a spontaneous remission of their depressive symptoms after a cardiac event, which patients will linger with depression, and which patients will best respond to treatment. Risk stratification, using the psychiatric history, symptoms and severity of depression, and genetic predisposition¹¹⁸ might allow improved targeted therapies.

Does depression cause cardiac disease?

The link between depression and heart disease can be seen as merely an association. In the interventional trials performed to date, we have not yet seen a reduction in cardiac deaths when depression was treated, challenging any assumption of a causal relationship between depression and heart disease. The debate about association vs cause is germane to behavioral medicine,¹¹⁹ and the better we understand the mechanistic pathways, the better we can advise patients and treat depression comorbid with heart disease.

Behavioral medicine is currently measuring the aspects of depression associated with cardiac disease, including the spectrum of somatic (body) and affective (mood) symptoms¹²⁰ and specific areas such as sympathetic arousal and early morning insomnia.¹²¹ If we can determine the depression subtype that carries a worse cardiac prognosis, we may untangle the biobehavioral links that bidirectionally bridge clinical depression and cardiac disease.

Another area of interest, emotional vitality (a positive state associated with interest, enthusiasm, excitement, and energy for living) has been shown to protect against coronary heart disease¹²² and holds much promise.

In the plenary lecture of the Academy of Psychosomatic Medicine in 2006, Frasure-Smith spoke of the “pleiotropism” of our antidepressant interventions on the various risk factors in depressed cardiac patients.¹²³ We need behavioral medicine studies that elucidate these mechanisms, guiding more precise treatments as well as novel therapies. Omega-3 fatty acids, which benefit heart disease and clinical depression,¹²⁴ will be used in a randomized controlled trial by Lespérance and colleagues.¹²⁵ We await the results of this exciting research.

 

 

Will treating depression help in other types of heart disease?

The SADHART-CHF trial is examining whether 12 weeks of sertraline therapy is better than placebo in preventing death and improving cardiac outcomes in patients with chronic heart failure and comorbid major depressive disorder. It was to be completed in the fall of 2008. The results and experience of this study will help in designing future interventional trials to reduce the risk of depression in cardiovascular diseases.

We also await the results of a National Heart, Lung, and Blood Institute (NHLBI) trial, “Bypassing the Blues,” which is studying the treatment of depression after cardiac bypass surgery. This study should provide further insights into management of the depressed cardiac patient. Further prognostic studies in cardiac patients are also needed using the PHQ-9 and its shorter version, PHQ-2.

Current and future guidelines

For years our European colleagues have been ahead of us in recognizing depression screening and stress management as key to cardiac disease-prevention strategies.¹²⁶ The NHLBI nicely outlined recommendations on the assessment and treatment of depression in cardiovascular patients.¹²⁷ The just-published AHA Science Advisory should further encourage clinicians to screen and treat depression in the patient population.¹ As our knowledge grows, we look forward to future evidence-based guidelines for depressed cardiac patients.

Over the past several decades, a large body of evidence has emerged demonstrating the adverse impact of depressive disorder on heart disease. This evidence confirms the early suspicion of observant clinicians that psychological factors play a significant role in the genesis and course of heart disease, as well as confirming the ancient belief in a mind-body connection in general and a connection between human moods and the heart in particular. Given the high prevalence of these two disorders, we need a better understanding of the impact of depressive disorder on heart disease, the proposed underlying pathophysiologic mechanisms, and the effects of treating depression in relation to risk reduction in patients with heart disease.

In this article, I will focus on (1) reviewing the results of meta-analyses examining the association of depression with cardiac diseases, (2) discussing the relationship between depression and mental stress–induced myocardial ischemia, (3) reviewing the available studies of the treatment of depression in patients with cardiac disease, and (4) discussing future directions for research in this area.

ASSOCIATION OF DEPRESSION WITH PROGRESSION OF CARDIAC DISEASES

As a disease of the brain, depression is common. The lifetime prevalence of major depressive disorder, a sig­nificant form of depression, is 16.2%.¹ The point prevalence of depression in medically ill patients is much higher, ranging from 20% to 50%, and the prevalence of milder depression is even more common. Despite this substantial prevalence, depression (especially in its milder forms) is rarely recognized. It often occurs insidiously, confusing its sufferer into believing that it is part of his or her character rather than an illness.

An invisible killer

The adverse effects of depression manifest in many aspects of life—from relationships to job performance to compliance with medical treatments—and can be so severe as to render the condition an “invisible killer.” The first evidence of this emerged in the medical literature in 1937 when Malzberg² reported that patients with melancholia had a significantly higher death rate than the general population and that cardiac death occurred in more than 40% of those patients. Although it took another several decades for the field to accelerate, ample data have now been gathered to prove an unshakable association between depression and progression of cardiac diseases. Instead of reviewing results of each study, I will present the results of several meta-analyses.

Prognosis of post–myocardial infarction patients with depression

In a meta-analysis published in 2004, van Melle et al³ examined data derived from the MEDLINE, EMBASE, and PsycINFO databases between 1975 and 2003 on the prognostic association of post–myocardial infarction (MI) depression with mortality and cardiovascular events. Twenty-two studies met the selection criteria (post-MI status with measurement of depression and up to 2 years of follow-up); these studies included a total 6,367 post-MI patients and had an average follow-up of 13.7 months. The analysis revealed that post-MI depression was associated with each of the following:

• All-cause mortality (fixed-effects odds ratio [OR] = 2.38; 95% confidence interval [CI], 1.76 to 3.22; P < .00001)
• Cardiac mortality (fixed-effects OR = 2.59; 95% CI, 1.77 to 3.77; P < .00001)
• Occurrence of cardiovascular events (random-effects OR = 1.95; 95% CI, 1.33 to 2.85; P = .0006).

Prognosis of depressed patients with ischemic heart disease

In another 2004 meta-analysis, Barth et al⁴ examined the association of depression with mortality among patients with other forms of ischemic heart disease (IHD) (ie, beyond just MI) using data derived from English- and German-language databases (MEDLINE, PsycINFO, and PSYNDEX) from 1980 to 2003. A total of 11,905 patients from 20 cohorts were included. Although depression assessment was heterogeneous among the studies included, the unfavorable impact of depression on mortality among IHD patients was consistently observed regardless of whether the depression was self-reported or detected by psychiatric professionals. The risk of dying in the first 2 years after initial assessment was more than two times higher in patients with high depressive symptoms than in those with low depressive symptoms (OR = 2.24; 95% CI, 1.37 to 3.60). This negative prognostic impact remained over the long term and after adjustment for other risk factors (hazard ratio [HR] = 1.76; 95% CI, 1.27 to 2.43). Although clinical depression had no significant effect on mortality within the first 6 months after initial assessment (OR = 2.07; 95% CI, 0.82 to 5.26), after 2 years it was associated with a greater than twofold higher risk of death (OR = 2.61; 95% CI, 1.53 to 4.47).⁴

Prognosis of depressed patients with heart failure

Several studies over the past decade, including one from my research group,⁵ have prospectively examined the impact of depression on outcomes in patients with heart failure (HF). Rutledge et al⁶ used meta-analysis to summarize the findings of eight independent cohort studies that tracked the association between depression and mortality or cardiac events in a total of 1,845 patients with HF; follow-up ranged from 6 months to more than 4 years. They found that those patients who were depressed had higher rates of death and secondary events (relative risk [RR] = 2.1; 95% CI, 1.7 to 2.6) compared with their nondepressed counterparts, as well as trends toward increased health care use and higher rates of hospitalization and emergency room visitation.

Development of ischemic heart disease in depressed patients

To assess depression’s role as a potential predictor of IHD development, Rugulies⁷ reviewed data from MEDLINE (1966 to 2000) and PsycINFO (1887 to 2000), selecting 11 cohort studies based on assessment of patients by standardized psychometric scale (clinical depression or depressed symptoms) and “hard” events (fatal/nonfatal MI, coronary death, or cardiac death). Among the 36,549 individuals in these studies, the overall RR for development of IHD in depressed subjects (as compared with nondepressed subjects) was 1.64 (95% CI, 1.29 to 2.08; P < .001). Sensitivity analysis revealed that clinical depression was a stronger predictor of IHD (RR = 2.69; 95% CI, 1.63 to 4.43; P < .001) than depressive symptoms were (RR = 1.49; 95% CI = 1.16 to 1.92; P = .02).

In summary, individuals with depressive disorder, even mild forms, are more likely to develop IHD than are individuals without depression. The increased likelihood of developing IHD is independent of conventional risk factors. Therefore, depression is a primary risk factor for IHD. Depression is also a secondary risk factor, independent of conventional risk factors, for significantly worse prognosis in patients with MI, other forms of IHD, and HF. Depression’s adverse effect on HF prognosis is independent of the baseline impairment in cardiac function and of the ischemic etiology of HF.

 

 

DEPRESSION AND MENTAL STRESS–INDUCED MYOCARDIAL ISCHEMIA

Of the numerous proposed pathophysiologic mechanisms explaining the adverse impact of depression on cardiac diseases, I would like to emphasize the clinical and research significance of mental stress–induced myocardial ischemia (MSIMI).

Myocardial ischemia is an important measure of the clinical manifestation of IHD. Ambulatory electrocardiographic monitoring yielded the insight that myocardial ischemia occurs frequently and transiently during daily living; it usually occurs in the context of a lower heart rate, is asymptomatic or silent, does not necessarily involve high-intensity physical activity, and commonly occurs in conjunction with increased negative emotions.^8,9

Over the past 2 to 3 decades, several laboratories have consistently demonstrated that mental stress testing elicits myocardial ischemia in patients with documented IHD.^8,10,11 The prevalence of MSIMI, defined by wall motion abnormality and/or significantly reduced ejection fraction, is comparable to that of exercise-induced myocardial ischemia in the laboratory setting.¹²

Differences from exercise-induced ischemia

MSIMI differs from exercise-induced ischemia in several notable ways. It occurs silently most of the time and rarely results in ischemic electrocardiographic changes. Mental stress induces greater frequency and severity of left ventricular dysfunction. Furthermore, mental stress testing causes a greater diastolic blood pressure response but a modest increase in heart rate, whereas exercise testing elicits a smaller elevation in diastolic blood pressure but a several-fold increase in heart rate.

A key mechanism: Transient coronary vasoconstriction

One of the underlying mechanisms by which mental stress induces myocardial ischemia in susceptible patients is transient coronary vasoconstriction. Yueng et al¹³ used an intracoronary Doppler catheter to assess the change in coronary blood flow during mental stress testing and endothelium-dependent vasodilation in a group of patients with IHD. Coronary artery responses varied from 38% constriction to 29% dilation, with changes in coronary blood flow ranging from a decrease of 48% to an increase of 42%. Interestingly, although it has been proposed that mental stress triggers release of catecholamines that induce coronary vasoconstriction, the direction and magnitude of the change were not predicted by changes in heart rate, blood pressure, or plasma norepinephrine level. The change in coronary perfusion was correlated, however, with the response to acetylcholine infusion.¹³

Dakak et al¹⁴ showed that while the coronary microcirculation dilated during mental stress testing in individuals without IHD, it failed to dilate during such testing in IHD patients, a response that is likely mediated by alpha-adrenergic receptor activation. Furthermore, systemic vascular resistance has been found to increase significantly during mental stress and to be positively correlated with increases in plasma epinephrine.¹⁵ In contrast, systemic vascular resistance was reduced significantly during exercise testing, and there was no relationship between the exercise-induced hemodynamic change and the plasma epinephrine level.¹⁵ Compared with exercise-induced ischemia, epinephrine-induced ischemia (which may occur during emotional distress) is marked by smaller increases in heart rate and rate-pressure product and by a marked increase in contractility.¹⁶

MSIMI predicts cardiac events

From a prognostic standpoint, MSIMI consistently predicts an increase in future adverse cardiac events.^11,17–19 In a sample of 132 IHD patients with a recent positive exercise test,¹¹ MSIMI was associated with an increase in cardiac events during 5-year follow-up (OR = 2.8; 95% CI, 1.0 to 7.7; P < .05) independent of patients’ age, history of prior MI, or baseline cardiac function. In contrast, exercise-induced ischemia was not predictive for adverse cardiac events (OR = 1.5; 95% CI, 0.6 to 3.9; P = .39) in this same sample.

Depression correlates with MSIMI occurrence

Adapted from American Heart Journal (Jiang W, et al. Depression and increased myocardial ischemic activity in patients with ischemic heart disease. Am Heart J 2003; 146:55–61.), copyright 2003, with permission from Elsevier.

More relevant is the finding that depression is associ­ated with the occurrence of MSIMI (Figure 1). Following withdrawal of antianginal medication for at least 48 hours, we tested 135 patients with stable IHD using the Center for Epidemiological Studies Depression Scale (CES-D) to evaluate for depressive symptoms and using radionuclide ventriculography to detect the occurrence of wall motion abnormalities during mental stress and exercise testing.²⁰ The mental stress tasks used in this study included mental arithmetic, public speaking, a mirror trace task, reading, and a type A videotaped structured interview.

The mean CES-D score was 8.2 (SD = 7.4; range, 0 to 47) and the median score was 7. Logistic regression models using restricted cubic splines revealed a curvilinear relation between CES-D scores and the probability of ischemia triggered by mental stress testing and exercise testing. For patients with CES-D scores less than or equal to 19 (81.5% of the study population), a 5-point increment in the CES-D score was associated with a roughly twofold increase in the likelihood of MSIMI (Figure 1A). For patients with CES-D scores greater than 19, the relation between scores and ischemia during mental stress tended to be inverse (Figure 1A), but these patients represented a small portion of the study sample (18.5%). In contrast, depression was not related to the occurrence of exercise-induced ischemia (Figure 1B). This finding strongly indicates that MSIMI may be a significant mechanism by which depression increases the risk of mortality and morbidity in patients with IHD. A few patients in this study had severe depressive symptoms (CES-D scores > 19), which makes interpretation of the result very diffi­cult. Because only 18.5% of the patients had CES-D scores greater than 19, this pattern of results needs to be confirmed in a sample with a greater representation of these more severely depressed patients.²⁰

 

 

INSIGHTS FROM STUDIES OF DEPRESSION THERAPY IN CARDIAC PATIENTS

To date, six trials evaluating therapies targeting depression or depression-related problems in patients with cardiac diseases have been completed, and an additional trial is ongoing (Table 1).

Among antidepressants, selective serotonin reuptake inhibitors (SSRIs) have been uniformly demonstrated to be effective in improving depressive symptoms and relatively safe for cardiac patients.^21–24 Not surprisingly, tricyclic antidepressants have been found to cause more cardiac problems.²¹ Mirtazapine, a central nervous system alpha-2 antagonist, failed to improve depressive symptoms in depressed post-MI patients in the Myocardial Infarction and Depression Intervention Trial (MIND-IT),^25,26 but because the results from this study have been presented only in abstract form, more details will be necessary to gain insight into explanations for this failure.

Although psychotherapy has been found to be quite effective among depressed patients without other medical illnesses, its effectiveness among patients with cardiac disease has not been impressive to date (Table 1).

No evidence of prognostic benefit from psychotherapy

Results from evaluations of psychotherapeutic interventions on cardiac prognosis have been rather disappointing (Table 1). The Enhancing Recovery In Coronary Heart Disease Patients (ENRICHD) study,²⁷ which involved randomization of 2,481 post-MI patients with depression and/or low perceived social support to usual care or cognitive behavior therapy, failed to show an impact of cognitive behavior therapy on the combined end point of death or nonfatal MI. Similarly, the Montreal Heart Attack Readjustment Trial (M-HART)²⁸ failed to demonstrate a benefit from home-based psychosocial nursing intervention on cardiac prognosis in IHD patients. These studies suggested that psychotherapeutic intervention might have differing or even opposite effects on the two genders.

Potential prognostic benefit from antidepressant therapy

In theory, adequate treatment of depression could affect dysregulated physiologic factors as well as dysregulated psychosocial factors, thereby leading to improved cardiac outcomes. There is physiologic evidence to support beneficial pleiotropic effects of antidepressant medications in IHD, such as reduced platelet activity^29–31 and improvement in low heart rate variability^32–34 with both sertraline and paroxetine.

The MIND-IT study evaluated mirtazapine for post-MI depression using a randomized placebo-controlled design.²⁵ However, this trial failed to find a significant treatment effect for either depression or cardiac outcomes.²⁶ These results may have been related to a lack of statistical power, as only 209 treated patients were compared with 122 patients receiving usual care. This trial also raises the question whether any nontricyclic antidepressant (other than SSRIs) might have beneficial effects on cardiovascular outcomes, or whether such an effect might be limited to SSRIs alone.

Provocative results emerged from the Sertraline Antidepressant Heart Attack Randomized Trial (SADHART),²³ a randomized, double-blind, placebo-controlled investigation of the safety and efficacy of sertraline for major depressive disorder among 369 patients with recent MI or unstable angina. Patients receiving sertraline had fewer cardiac events (death, MI, stroke, worsened angina, or onset of HF) compared with patients taking placebo. The relative risk ratio for having at least one cardiac event was 0.77 with sertraline therapy, but this reduction in risk was not statistically significant (95% CI, 0.51 to 1.16). Although these findings suggest that sertraline may improve cardiac outcomes, the study was not adequately powered to detect differences on this measure. Power calculations indicate that in order to confirm a 20% reduction in relative risk in a randomized trial, a sample of at least 4,000 depressed patients with acute coronary syndrome would be required.²³ Based on the cost of SADHART, the estimated expense to complete such a study is approximately $200 million.

The SADHART-CHF trial is a randomized, double-blind, placebo-controlled study examining sertraline’s efficacy for major depressive disorder among patients with HF, as well as its effects on mortality and cardiac outcomes. This trial is in its last year of enrollment, and results will be forthcoming in 2008.

FUTURE DIRECTIONS

These recent insights into depression’s impact on cardiac disease give rise to several new questions to consider:

• Expand research to patients with depressive symptoms? To date, investigations into treatment effects have focused only on patients with cardiac disease who have major depressive disorder. However, depressive symptoms as reported on self-administered questionnaires consistently have been shown to be a risk for poor cardiac outcomes. Should we expand our interventional studies to patients with self-reported depressive symptoms?
• How thoroughly to test for differences among antidepressants? Three of the six SSRIs have been studied among depressed cardiac patients. Based on the available findings, can we assume that all SSRIs have the same efficacy and safety profiles and are similarly cardiovascularly protective? Should every antidepressant or SSRI be tested? Should head-to-head comparison studies be conducted? Tricyclic antidepressants are cardiotoxic, and central nervous system alpha-2 antagonists like mirtazapine may not be effective, but what about other types of antidepressants for depressed cardiac patients?
• Is there a role for studying surrogate end points? Studies examining the effects of an intervention on mortality and/or morbidity can be very expensive. As research budgets tighten, can we instead test the effects of depression therapy on some surrogate end points?

Our laboratory has been funded by the National Heart, Lung, and Blood Institute to compare the effects of escitalopram with those of placebo on MSIMI in patients with stable IHD and a score of 5 or greater on the Beck Depression Inventory. This study, the Responses of Myocardial Ischemia to Escitalopram Treatment (REMIT) trial, will provide SSRI therapy to patients with a broad spectrum of depressive symptoms (not just major depressive disorder), assess the ischemic activity induced by mental stress testing as its primary end point, and explore the effects on other hypothesized mechanisms of depression that adversely affect cardiac diseases (platelet aggregation, inflammatory biomarkers, etc). Stay tuned for the results in the near future.

Over the past several decades, a large body of evidence has emerged demonstrating the adverse impact of depressive disorder on heart disease. This evidence confirms the early suspicion of observant clinicians that psychological factors play a significant role in the genesis and course of heart disease, as well as confirming the ancient belief in a mind-body connection in general and a connection between human moods and the heart in particular. Given the high prevalence of these two disorders, we need a better understanding of the impact of depressive disorder on heart disease, the proposed underlying pathophysiologic mechanisms, and the effects of treating depression in relation to risk reduction in patients with heart disease.

In this article, I will focus on (1) reviewing the results of meta-analyses examining the association of depression with cardiac diseases, (2) discussing the relationship between depression and mental stress–induced myocardial ischemia, (3) reviewing the available studies of the treatment of depression in patients with cardiac disease, and (4) discussing future directions for research in this area.

ASSOCIATION OF DEPRESSION WITH PROGRESSION OF CARDIAC DISEASES

As a disease of the brain, depression is common. The lifetime prevalence of major depressive disorder, a sig­nificant form of depression, is 16.2%.¹ The point prevalence of depression in medically ill patients is much higher, ranging from 20% to 50%, and the prevalence of milder depression is even more common. Despite this substantial prevalence, depression (especially in its milder forms) is rarely recognized. It often occurs insidiously, confusing its sufferer into believing that it is part of his or her character rather than an illness.

An invisible killer

The adverse effects of depression manifest in many aspects of life—from relationships to job performance to compliance with medical treatments—and can be so severe as to render the condition an “invisible killer.” The first evidence of this emerged in the medical literature in 1937 when Malzberg² reported that patients with melancholia had a significantly higher death rate than the general population and that cardiac death occurred in more than 40% of those patients. Although it took another several decades for the field to accelerate, ample data have now been gathered to prove an unshakable association between depression and progression of cardiac diseases. Instead of reviewing results of each study, I will present the results of several meta-analyses.

Prognosis of post–myocardial infarction patients with depression

In a meta-analysis published in 2004, van Melle et al³ examined data derived from the MEDLINE, EMBASE, and PsycINFO databases between 1975 and 2003 on the prognostic association of post–myocardial infarction (MI) depression with mortality and cardiovascular events. Twenty-two studies met the selection criteria (post-MI status with measurement of depression and up to 2 years of follow-up); these studies included a total 6,367 post-MI patients and had an average follow-up of 13.7 months. The analysis revealed that post-MI depression was associated with each of the following:

• All-cause mortality (fixed-effects odds ratio [OR] = 2.38; 95% confidence interval [CI], 1.76 to 3.22; P < .00001)
• Cardiac mortality (fixed-effects OR = 2.59; 95% CI, 1.77 to 3.77; P < .00001)
• Occurrence of cardiovascular events (random-effects OR = 1.95; 95% CI, 1.33 to 2.85; P = .0006).

Prognosis of depressed patients with ischemic heart disease

In another 2004 meta-analysis, Barth et al⁴ examined the association of depression with mortality among patients with other forms of ischemic heart disease (IHD) (ie, beyond just MI) using data derived from English- and German-language databases (MEDLINE, PsycINFO, and PSYNDEX) from 1980 to 2003. A total of 11,905 patients from 20 cohorts were included. Although depression assessment was heterogeneous among the studies included, the unfavorable impact of depression on mortality among IHD patients was consistently observed regardless of whether the depression was self-reported or detected by psychiatric professionals. The risk of dying in the first 2 years after initial assessment was more than two times higher in patients with high depressive symptoms than in those with low depressive symptoms (OR = 2.24; 95% CI, 1.37 to 3.60). This negative prognostic impact remained over the long term and after adjustment for other risk factors (hazard ratio [HR] = 1.76; 95% CI, 1.27 to 2.43). Although clinical depression had no significant effect on mortality within the first 6 months after initial assessment (OR = 2.07; 95% CI, 0.82 to 5.26), after 2 years it was associated with a greater than twofold higher risk of death (OR = 2.61; 95% CI, 1.53 to 4.47).⁴

Prognosis of depressed patients with heart failure

Several studies over the past decade, including one from my research group,⁵ have prospectively examined the impact of depression on outcomes in patients with heart failure (HF). Rutledge et al⁶ used meta-analysis to summarize the findings of eight independent cohort studies that tracked the association between depression and mortality or cardiac events in a total of 1,845 patients with HF; follow-up ranged from 6 months to more than 4 years. They found that those patients who were depressed had higher rates of death and secondary events (relative risk [RR] = 2.1; 95% CI, 1.7 to 2.6) compared with their nondepressed counterparts, as well as trends toward increased health care use and higher rates of hospitalization and emergency room visitation.

Development of ischemic heart disease in depressed patients

To assess depression’s role as a potential predictor of IHD development, Rugulies⁷ reviewed data from MEDLINE (1966 to 2000) and PsycINFO (1887 to 2000), selecting 11 cohort studies based on assessment of patients by standardized psychometric scale (clinical depression or depressed symptoms) and “hard” events (fatal/nonfatal MI, coronary death, or cardiac death). Among the 36,549 individuals in these studies, the overall RR for development of IHD in depressed subjects (as compared with nondepressed subjects) was 1.64 (95% CI, 1.29 to 2.08; P < .001). Sensitivity analysis revealed that clinical depression was a stronger predictor of IHD (RR = 2.69; 95% CI, 1.63 to 4.43; P < .001) than depressive symptoms were (RR = 1.49; 95% CI = 1.16 to 1.92; P = .02).

In summary, individuals with depressive disorder, even mild forms, are more likely to develop IHD than are individuals without depression. The increased likelihood of developing IHD is independent of conventional risk factors. Therefore, depression is a primary risk factor for IHD. Depression is also a secondary risk factor, independent of conventional risk factors, for significantly worse prognosis in patients with MI, other forms of IHD, and HF. Depression’s adverse effect on HF prognosis is independent of the baseline impairment in cardiac function and of the ischemic etiology of HF.

 

 

DEPRESSION AND MENTAL STRESS–INDUCED MYOCARDIAL ISCHEMIA

Of the numerous proposed pathophysiologic mechanisms explaining the adverse impact of depression on cardiac diseases, I would like to emphasize the clinical and research significance of mental stress–induced myocardial ischemia (MSIMI).

Myocardial ischemia is an important measure of the clinical manifestation of IHD. Ambulatory electrocardiographic monitoring yielded the insight that myocardial ischemia occurs frequently and transiently during daily living; it usually occurs in the context of a lower heart rate, is asymptomatic or silent, does not necessarily involve high-intensity physical activity, and commonly occurs in conjunction with increased negative emotions.^8,9

Over the past 2 to 3 decades, several laboratories have consistently demonstrated that mental stress testing elicits myocardial ischemia in patients with documented IHD.^8,10,11 The prevalence of MSIMI, defined by wall motion abnormality and/or significantly reduced ejection fraction, is comparable to that of exercise-induced myocardial ischemia in the laboratory setting.¹²

Differences from exercise-induced ischemia

MSIMI differs from exercise-induced ischemia in several notable ways. It occurs silently most of the time and rarely results in ischemic electrocardiographic changes. Mental stress induces greater frequency and severity of left ventricular dysfunction. Furthermore, mental stress testing causes a greater diastolic blood pressure response but a modest increase in heart rate, whereas exercise testing elicits a smaller elevation in diastolic blood pressure but a several-fold increase in heart rate.

A key mechanism: Transient coronary vasoconstriction

One of the underlying mechanisms by which mental stress induces myocardial ischemia in susceptible patients is transient coronary vasoconstriction. Yueng et al¹³ used an intracoronary Doppler catheter to assess the change in coronary blood flow during mental stress testing and endothelium-dependent vasodilation in a group of patients with IHD. Coronary artery responses varied from 38% constriction to 29% dilation, with changes in coronary blood flow ranging from a decrease of 48% to an increase of 42%. Interestingly, although it has been proposed that mental stress triggers release of catecholamines that induce coronary vasoconstriction, the direction and magnitude of the change were not predicted by changes in heart rate, blood pressure, or plasma norepinephrine level. The change in coronary perfusion was correlated, however, with the response to acetylcholine infusion.¹³

Dakak et al¹⁴ showed that while the coronary microcirculation dilated during mental stress testing in individuals without IHD, it failed to dilate during such testing in IHD patients, a response that is likely mediated by alpha-adrenergic receptor activation. Furthermore, systemic vascular resistance has been found to increase significantly during mental stress and to be positively correlated with increases in plasma epinephrine.¹⁵ In contrast, systemic vascular resistance was reduced significantly during exercise testing, and there was no relationship between the exercise-induced hemodynamic change and the plasma epinephrine level.¹⁵ Compared with exercise-induced ischemia, epinephrine-induced ischemia (which may occur during emotional distress) is marked by smaller increases in heart rate and rate-pressure product and by a marked increase in contractility.¹⁶

MSIMI predicts cardiac events

From a prognostic standpoint, MSIMI consistently predicts an increase in future adverse cardiac events.^11,17–19 In a sample of 132 IHD patients with a recent positive exercise test,¹¹ MSIMI was associated with an increase in cardiac events during 5-year follow-up (OR = 2.8; 95% CI, 1.0 to 7.7; P < .05) independent of patients’ age, history of prior MI, or baseline cardiac function. In contrast, exercise-induced ischemia was not predictive for adverse cardiac events (OR = 1.5; 95% CI, 0.6 to 3.9; P = .39) in this same sample.

Depression correlates with MSIMI occurrence

Adapted from American Heart Journal (Jiang W, et al. Depression and increased myocardial ischemic activity in patients with ischemic heart disease. Am Heart J 2003; 146:55–61.), copyright 2003, with permission from Elsevier.

More relevant is the finding that depression is associ­ated with the occurrence of MSIMI (Figure 1). Following withdrawal of antianginal medication for at least 48 hours, we tested 135 patients with stable IHD using the Center for Epidemiological Studies Depression Scale (CES-D) to evaluate for depressive symptoms and using radionuclide ventriculography to detect the occurrence of wall motion abnormalities during mental stress and exercise testing.²⁰ The mental stress tasks used in this study included mental arithmetic, public speaking, a mirror trace task, reading, and a type A videotaped structured interview.

The mean CES-D score was 8.2 (SD = 7.4; range, 0 to 47) and the median score was 7. Logistic regression models using restricted cubic splines revealed a curvilinear relation between CES-D scores and the probability of ischemia triggered by mental stress testing and exercise testing. For patients with CES-D scores less than or equal to 19 (81.5% of the study population), a 5-point increment in the CES-D score was associated with a roughly twofold increase in the likelihood of MSIMI (Figure 1A). For patients with CES-D scores greater than 19, the relation between scores and ischemia during mental stress tended to be inverse (Figure 1A), but these patients represented a small portion of the study sample (18.5%). In contrast, depression was not related to the occurrence of exercise-induced ischemia (Figure 1B). This finding strongly indicates that MSIMI may be a significant mechanism by which depression increases the risk of mortality and morbidity in patients with IHD. A few patients in this study had severe depressive symptoms (CES-D scores > 19), which makes interpretation of the result very diffi­cult. Because only 18.5% of the patients had CES-D scores greater than 19, this pattern of results needs to be confirmed in a sample with a greater representation of these more severely depressed patients.²⁰

 

 

INSIGHTS FROM STUDIES OF DEPRESSION THERAPY IN CARDIAC PATIENTS

To date, six trials evaluating therapies targeting depression or depression-related problems in patients with cardiac diseases have been completed, and an additional trial is ongoing (Table 1).

Among antidepressants, selective serotonin reuptake inhibitors (SSRIs) have been uniformly demonstrated to be effective in improving depressive symptoms and relatively safe for cardiac patients.^21–24 Not surprisingly, tricyclic antidepressants have been found to cause more cardiac problems.²¹ Mirtazapine, a central nervous system alpha-2 antagonist, failed to improve depressive symptoms in depressed post-MI patients in the Myocardial Infarction and Depression Intervention Trial (MIND-IT),^25,26 but because the results from this study have been presented only in abstract form, more details will be necessary to gain insight into explanations for this failure.

Although psychotherapy has been found to be quite effective among depressed patients without other medical illnesses, its effectiveness among patients with cardiac disease has not been impressive to date (Table 1).

No evidence of prognostic benefit from psychotherapy

Results from evaluations of psychotherapeutic interventions on cardiac prognosis have been rather disappointing (Table 1). The Enhancing Recovery In Coronary Heart Disease Patients (ENRICHD) study,²⁷ which involved randomization of 2,481 post-MI patients with depression and/or low perceived social support to usual care or cognitive behavior therapy, failed to show an impact of cognitive behavior therapy on the combined end point of death or nonfatal MI. Similarly, the Montreal Heart Attack Readjustment Trial (M-HART)²⁸ failed to demonstrate a benefit from home-based psychosocial nursing intervention on cardiac prognosis in IHD patients. These studies suggested that psychotherapeutic intervention might have differing or even opposite effects on the two genders.

Potential prognostic benefit from antidepressant therapy

In theory, adequate treatment of depression could affect dysregulated physiologic factors as well as dysregulated psychosocial factors, thereby leading to improved cardiac outcomes. There is physiologic evidence to support beneficial pleiotropic effects of antidepressant medications in IHD, such as reduced platelet activity^29–31 and improvement in low heart rate variability^32–34 with both sertraline and paroxetine.

The MIND-IT study evaluated mirtazapine for post-MI depression using a randomized placebo-controlled design.²⁵ However, this trial failed to find a significant treatment effect for either depression or cardiac outcomes.²⁶ These results may have been related to a lack of statistical power, as only 209 treated patients were compared with 122 patients receiving usual care. This trial also raises the question whether any nontricyclic antidepressant (other than SSRIs) might have beneficial effects on cardiovascular outcomes, or whether such an effect might be limited to SSRIs alone.

Provocative results emerged from the Sertraline Antidepressant Heart Attack Randomized Trial (SADHART),²³ a randomized, double-blind, placebo-controlled investigation of the safety and efficacy of sertraline for major depressive disorder among 369 patients with recent MI or unstable angina. Patients receiving sertraline had fewer cardiac events (death, MI, stroke, worsened angina, or onset of HF) compared with patients taking placebo. The relative risk ratio for having at least one cardiac event was 0.77 with sertraline therapy, but this reduction in risk was not statistically significant (95% CI, 0.51 to 1.16). Although these findings suggest that sertraline may improve cardiac outcomes, the study was not adequately powered to detect differences on this measure. Power calculations indicate that in order to confirm a 20% reduction in relative risk in a randomized trial, a sample of at least 4,000 depressed patients with acute coronary syndrome would be required.²³ Based on the cost of SADHART, the estimated expense to complete such a study is approximately $200 million.

The SADHART-CHF trial is a randomized, double-blind, placebo-controlled study examining sertraline’s efficacy for major depressive disorder among patients with HF, as well as its effects on mortality and cardiac outcomes. This trial is in its last year of enrollment, and results will be forthcoming in 2008.

FUTURE DIRECTIONS

These recent insights into depression’s impact on cardiac disease give rise to several new questions to consider:

• Expand research to patients with depressive symptoms? To date, investigations into treatment effects have focused only on patients with cardiac disease who have major depressive disorder. However, depressive symptoms as reported on self-administered questionnaires consistently have been shown to be a risk for poor cardiac outcomes. Should we expand our interventional studies to patients with self-reported depressive symptoms?
• How thoroughly to test for differences among antidepressants? Three of the six SSRIs have been studied among depressed cardiac patients. Based on the available findings, can we assume that all SSRIs have the same efficacy and safety profiles and are similarly cardiovascularly protective? Should every antidepressant or SSRI be tested? Should head-to-head comparison studies be conducted? Tricyclic antidepressants are cardiotoxic, and central nervous system alpha-2 antagonists like mirtazapine may not be effective, but what about other types of antidepressants for depressed cardiac patients?
• Is there a role for studying surrogate end points? Studies examining the effects of an intervention on mortality and/or morbidity can be very expensive. As research budgets tighten, can we instead test the effects of depression therapy on some surrogate end points?

Our laboratory has been funded by the National Heart, Lung, and Blood Institute to compare the effects of escitalopram with those of placebo on MSIMI in patients with stable IHD and a score of 5 or greater on the Beck Depression Inventory. This study, the Responses of Myocardial Ischemia to Escitalopram Treatment (REMIT) trial, will provide SSRI therapy to patients with a broad spectrum of depressive symptoms (not just major depressive disorder), assess the ischemic activity induced by mental stress testing as its primary end point, and explore the effects on other hypothesized mechanisms of depression that adversely affect cardiac diseases (platelet aggregation, inflammatory biomarkers, etc). Stay tuned for the results in the near future.

Over the past several decades, a large body of evidence has emerged demonstrating the adverse impact of depressive disorder on heart disease. This evidence confirms the early suspicion of observant clinicians that psychological factors play a significant role in the genesis and course of heart disease, as well as confirming the ancient belief in a mind-body connection in general and a connection between human moods and the heart in particular. Given the high prevalence of these two disorders, we need a better understanding of the impact of depressive disorder on heart disease, the proposed underlying pathophysiologic mechanisms, and the effects of treating depression in relation to risk reduction in patients with heart disease.

In this article, I will focus on (1) reviewing the results of meta-analyses examining the association of depression with cardiac diseases, (2) discussing the relationship between depression and mental stress–induced myocardial ischemia, (3) reviewing the available studies of the treatment of depression in patients with cardiac disease, and (4) discussing future directions for research in this area.

ASSOCIATION OF DEPRESSION WITH PROGRESSION OF CARDIAC DISEASES

As a disease of the brain, depression is common. The lifetime prevalence of major depressive disorder, a sig­nificant form of depression, is 16.2%.¹ The point prevalence of depression in medically ill patients is much higher, ranging from 20% to 50%, and the prevalence of milder depression is even more common. Despite this substantial prevalence, depression (especially in its milder forms) is rarely recognized. It often occurs insidiously, confusing its sufferer into believing that it is part of his or her character rather than an illness.

An invisible killer

The adverse effects of depression manifest in many aspects of life—from relationships to job performance to compliance with medical treatments—and can be so severe as to render the condition an “invisible killer.” The first evidence of this emerged in the medical literature in 1937 when Malzberg² reported that patients with melancholia had a significantly higher death rate than the general population and that cardiac death occurred in more than 40% of those patients. Although it took another several decades for the field to accelerate, ample data have now been gathered to prove an unshakable association between depression and progression of cardiac diseases. Instead of reviewing results of each study, I will present the results of several meta-analyses.

Prognosis of post–myocardial infarction patients with depression

In a meta-analysis published in 2004, van Melle et al³ examined data derived from the MEDLINE, EMBASE, and PsycINFO databases between 1975 and 2003 on the prognostic association of post–myocardial infarction (MI) depression with mortality and cardiovascular events. Twenty-two studies met the selection criteria (post-MI status with measurement of depression and up to 2 years of follow-up); these studies included a total 6,367 post-MI patients and had an average follow-up of 13.7 months. The analysis revealed that post-MI depression was associated with each of the following:

• All-cause mortality (fixed-effects odds ratio [OR] = 2.38; 95% confidence interval [CI], 1.76 to 3.22; P < .00001)
• Cardiac mortality (fixed-effects OR = 2.59; 95% CI, 1.77 to 3.77; P < .00001)
• Occurrence of cardiovascular events (random-effects OR = 1.95; 95% CI, 1.33 to 2.85; P = .0006).

Prognosis of depressed patients with ischemic heart disease

In another 2004 meta-analysis, Barth et al⁴ examined the association of depression with mortality among patients with other forms of ischemic heart disease (IHD) (ie, beyond just MI) using data derived from English- and German-language databases (MEDLINE, PsycINFO, and PSYNDEX) from 1980 to 2003. A total of 11,905 patients from 20 cohorts were included. Although depression assessment was heterogeneous among the studies included, the unfavorable impact of depression on mortality among IHD patients was consistently observed regardless of whether the depression was self-reported or detected by psychiatric professionals. The risk of dying in the first 2 years after initial assessment was more than two times higher in patients with high depressive symptoms than in those with low depressive symptoms (OR = 2.24; 95% CI, 1.37 to 3.60). This negative prognostic impact remained over the long term and after adjustment for other risk factors (hazard ratio [HR] = 1.76; 95% CI, 1.27 to 2.43). Although clinical depression had no significant effect on mortality within the first 6 months after initial assessment (OR = 2.07; 95% CI, 0.82 to 5.26), after 2 years it was associated with a greater than twofold higher risk of death (OR = 2.61; 95% CI, 1.53 to 4.47).⁴

Prognosis of depressed patients with heart failure

Several studies over the past decade, including one from my research group,⁵ have prospectively examined the impact of depression on outcomes in patients with heart failure (HF). Rutledge et al⁶ used meta-analysis to summarize the findings of eight independent cohort studies that tracked the association between depression and mortality or cardiac events in a total of 1,845 patients with HF; follow-up ranged from 6 months to more than 4 years. They found that those patients who were depressed had higher rates of death and secondary events (relative risk [RR] = 2.1; 95% CI, 1.7 to 2.6) compared with their nondepressed counterparts, as well as trends toward increased health care use and higher rates of hospitalization and emergency room visitation.

Development of ischemic heart disease in depressed patients

To assess depression’s role as a potential predictor of IHD development, Rugulies⁷ reviewed data from MEDLINE (1966 to 2000) and PsycINFO (1887 to 2000), selecting 11 cohort studies based on assessment of patients by standardized psychometric scale (clinical depression or depressed symptoms) and “hard” events (fatal/nonfatal MI, coronary death, or cardiac death). Among the 36,549 individuals in these studies, the overall RR for development of IHD in depressed subjects (as compared with nondepressed subjects) was 1.64 (95% CI, 1.29 to 2.08; P < .001). Sensitivity analysis revealed that clinical depression was a stronger predictor of IHD (RR = 2.69; 95% CI, 1.63 to 4.43; P < .001) than depressive symptoms were (RR = 1.49; 95% CI = 1.16 to 1.92; P = .02).

In summary, individuals with depressive disorder, even mild forms, are more likely to develop IHD than are individuals without depression. The increased likelihood of developing IHD is independent of conventional risk factors. Therefore, depression is a primary risk factor for IHD. Depression is also a secondary risk factor, independent of conventional risk factors, for significantly worse prognosis in patients with MI, other forms of IHD, and HF. Depression’s adverse effect on HF prognosis is independent of the baseline impairment in cardiac function and of the ischemic etiology of HF.

 

 

DEPRESSION AND MENTAL STRESS–INDUCED MYOCARDIAL ISCHEMIA

Of the numerous proposed pathophysiologic mechanisms explaining the adverse impact of depression on cardiac diseases, I would like to emphasize the clinical and research significance of mental stress–induced myocardial ischemia (MSIMI).

Myocardial ischemia is an important measure of the clinical manifestation of IHD. Ambulatory electrocardiographic monitoring yielded the insight that myocardial ischemia occurs frequently and transiently during daily living; it usually occurs in the context of a lower heart rate, is asymptomatic or silent, does not necessarily involve high-intensity physical activity, and commonly occurs in conjunction with increased negative emotions.^8,9

Over the past 2 to 3 decades, several laboratories have consistently demonstrated that mental stress testing elicits myocardial ischemia in patients with documented IHD.^8,10,11 The prevalence of MSIMI, defined by wall motion abnormality and/or significantly reduced ejection fraction, is comparable to that of exercise-induced myocardial ischemia in the laboratory setting.¹²

Differences from exercise-induced ischemia

MSIMI differs from exercise-induced ischemia in several notable ways. It occurs silently most of the time and rarely results in ischemic electrocardiographic changes. Mental stress induces greater frequency and severity of left ventricular dysfunction. Furthermore, mental stress testing causes a greater diastolic blood pressure response but a modest increase in heart rate, whereas exercise testing elicits a smaller elevation in diastolic blood pressure but a several-fold increase in heart rate.

A key mechanism: Transient coronary vasoconstriction

One of the underlying mechanisms by which mental stress induces myocardial ischemia in susceptible patients is transient coronary vasoconstriction. Yueng et al¹³ used an intracoronary Doppler catheter to assess the change in coronary blood flow during mental stress testing and endothelium-dependent vasodilation in a group of patients with IHD. Coronary artery responses varied from 38% constriction to 29% dilation, with changes in coronary blood flow ranging from a decrease of 48% to an increase of 42%. Interestingly, although it has been proposed that mental stress triggers release of catecholamines that induce coronary vasoconstriction, the direction and magnitude of the change were not predicted by changes in heart rate, blood pressure, or plasma norepinephrine level. The change in coronary perfusion was correlated, however, with the response to acetylcholine infusion.¹³

Dakak et al¹⁴ showed that while the coronary microcirculation dilated during mental stress testing in individuals without IHD, it failed to dilate during such testing in IHD patients, a response that is likely mediated by alpha-adrenergic receptor activation. Furthermore, systemic vascular resistance has been found to increase significantly during mental stress and to be positively correlated with increases in plasma epinephrine.¹⁵ In contrast, systemic vascular resistance was reduced significantly during exercise testing, and there was no relationship between the exercise-induced hemodynamic change and the plasma epinephrine level.¹⁵ Compared with exercise-induced ischemia, epinephrine-induced ischemia (which may occur during emotional distress) is marked by smaller increases in heart rate and rate-pressure product and by a marked increase in contractility.¹⁶

MSIMI predicts cardiac events

From a prognostic standpoint, MSIMI consistently predicts an increase in future adverse cardiac events.^11,17–19 In a sample of 132 IHD patients with a recent positive exercise test,¹¹ MSIMI was associated with an increase in cardiac events during 5-year follow-up (OR = 2.8; 95% CI, 1.0 to 7.7; P < .05) independent of patients’ age, history of prior MI, or baseline cardiac function. In contrast, exercise-induced ischemia was not predictive for adverse cardiac events (OR = 1.5; 95% CI, 0.6 to 3.9; P = .39) in this same sample.

Depression correlates with MSIMI occurrence

Adapted from American Heart Journal (Jiang W, et al. Depression and increased myocardial ischemic activity in patients with ischemic heart disease. Am Heart J 2003; 146:55–61.), copyright 2003, with permission from Elsevier.

More relevant is the finding that depression is associ­ated with the occurrence of MSIMI (Figure 1). Following withdrawal of antianginal medication for at least 48 hours, we tested 135 patients with stable IHD using the Center for Epidemiological Studies Depression Scale (CES-D) to evaluate for depressive symptoms and using radionuclide ventriculography to detect the occurrence of wall motion abnormalities during mental stress and exercise testing.²⁰ The mental stress tasks used in this study included mental arithmetic, public speaking, a mirror trace task, reading, and a type A videotaped structured interview.

The mean CES-D score was 8.2 (SD = 7.4; range, 0 to 47) and the median score was 7. Logistic regression models using restricted cubic splines revealed a curvilinear relation between CES-D scores and the probability of ischemia triggered by mental stress testing and exercise testing. For patients with CES-D scores less than or equal to 19 (81.5% of the study population), a 5-point increment in the CES-D score was associated with a roughly twofold increase in the likelihood of MSIMI (Figure 1A). For patients with CES-D scores greater than 19, the relation between scores and ischemia during mental stress tended to be inverse (Figure 1A), but these patients represented a small portion of the study sample (18.5%). In contrast, depression was not related to the occurrence of exercise-induced ischemia (Figure 1B). This finding strongly indicates that MSIMI may be a significant mechanism by which depression increases the risk of mortality and morbidity in patients with IHD. A few patients in this study had severe depressive symptoms (CES-D scores > 19), which makes interpretation of the result very diffi­cult. Because only 18.5% of the patients had CES-D scores greater than 19, this pattern of results needs to be confirmed in a sample with a greater representation of these more severely depressed patients.²⁰

 

 

INSIGHTS FROM STUDIES OF DEPRESSION THERAPY IN CARDIAC PATIENTS

To date, six trials evaluating therapies targeting depression or depression-related problems in patients with cardiac diseases have been completed, and an additional trial is ongoing (Table 1).

Among antidepressants, selective serotonin reuptake inhibitors (SSRIs) have been uniformly demonstrated to be effective in improving depressive symptoms and relatively safe for cardiac patients.^21–24 Not surprisingly, tricyclic antidepressants have been found to cause more cardiac problems.²¹ Mirtazapine, a central nervous system alpha-2 antagonist, failed to improve depressive symptoms in depressed post-MI patients in the Myocardial Infarction and Depression Intervention Trial (MIND-IT),^25,26 but because the results from this study have been presented only in abstract form, more details will be necessary to gain insight into explanations for this failure.

Although psychotherapy has been found to be quite effective among depressed patients without other medical illnesses, its effectiveness among patients with cardiac disease has not been impressive to date (Table 1).

No evidence of prognostic benefit from psychotherapy

Results from evaluations of psychotherapeutic interventions on cardiac prognosis have been rather disappointing (Table 1). The Enhancing Recovery In Coronary Heart Disease Patients (ENRICHD) study,²⁷ which involved randomization of 2,481 post-MI patients with depression and/or low perceived social support to usual care or cognitive behavior therapy, failed to show an impact of cognitive behavior therapy on the combined end point of death or nonfatal MI. Similarly, the Montreal Heart Attack Readjustment Trial (M-HART)²⁸ failed to demonstrate a benefit from home-based psychosocial nursing intervention on cardiac prognosis in IHD patients. These studies suggested that psychotherapeutic intervention might have differing or even opposite effects on the two genders.

Potential prognostic benefit from antidepressant therapy

In theory, adequate treatment of depression could affect dysregulated physiologic factors as well as dysregulated psychosocial factors, thereby leading to improved cardiac outcomes. There is physiologic evidence to support beneficial pleiotropic effects of antidepressant medications in IHD, such as reduced platelet activity^29–31 and improvement in low heart rate variability^32–34 with both sertraline and paroxetine.

The MIND-IT study evaluated mirtazapine for post-MI depression using a randomized placebo-controlled design.²⁵ However, this trial failed to find a significant treatment effect for either depression or cardiac outcomes.²⁶ These results may have been related to a lack of statistical power, as only 209 treated patients were compared with 122 patients receiving usual care. This trial also raises the question whether any nontricyclic antidepressant (other than SSRIs) might have beneficial effects on cardiovascular outcomes, or whether such an effect might be limited to SSRIs alone.

Provocative results emerged from the Sertraline Antidepressant Heart Attack Randomized Trial (SADHART),²³ a randomized, double-blind, placebo-controlled investigation of the safety and efficacy of sertraline for major depressive disorder among 369 patients with recent MI or unstable angina. Patients receiving sertraline had fewer cardiac events (death, MI, stroke, worsened angina, or onset of HF) compared with patients taking placebo. The relative risk ratio for having at least one cardiac event was 0.77 with sertraline therapy, but this reduction in risk was not statistically significant (95% CI, 0.51 to 1.16). Although these findings suggest that sertraline may improve cardiac outcomes, the study was not adequately powered to detect differences on this measure. Power calculations indicate that in order to confirm a 20% reduction in relative risk in a randomized trial, a sample of at least 4,000 depressed patients with acute coronary syndrome would be required.²³ Based on the cost of SADHART, the estimated expense to complete such a study is approximately $200 million.

The SADHART-CHF trial is a randomized, double-blind, placebo-controlled study examining sertraline’s efficacy for major depressive disorder among patients with HF, as well as its effects on mortality and cardiac outcomes. This trial is in its last year of enrollment, and results will be forthcoming in 2008.

FUTURE DIRECTIONS

These recent insights into depression’s impact on cardiac disease give rise to several new questions to consider:

• Expand research to patients with depressive symptoms? To date, investigations into treatment effects have focused only on patients with cardiac disease who have major depressive disorder. However, depressive symptoms as reported on self-administered questionnaires consistently have been shown to be a risk for poor cardiac outcomes. Should we expand our interventional studies to patients with self-reported depressive symptoms?
• How thoroughly to test for differences among antidepressants? Three of the six SSRIs have been studied among depressed cardiac patients. Based on the available findings, can we assume that all SSRIs have the same efficacy and safety profiles and are similarly cardiovascularly protective? Should every antidepressant or SSRI be tested? Should head-to-head comparison studies be conducted? Tricyclic antidepressants are cardiotoxic, and central nervous system alpha-2 antagonists like mirtazapine may not be effective, but what about other types of antidepressants for depressed cardiac patients?
• Is there a role for studying surrogate end points? Studies examining the effects of an intervention on mortality and/or morbidity can be very expensive. As research budgets tighten, can we instead test the effects of depression therapy on some surrogate end points?

Our laboratory has been funded by the National Heart, Lung, and Blood Institute to compare the effects of escitalopram with those of placebo on MSIMI in patients with stable IHD and a score of 5 or greater on the Beck Depression Inventory. This study, the Responses of Myocardial Ischemia to Escitalopram Treatment (REMIT) trial, will provide SSRI therapy to patients with a broad spectrum of depressive symptoms (not just major depressive disorder), assess the ischemic activity induced by mental stress testing as its primary end point, and explore the effects on other hypothesized mechanisms of depression that adversely affect cardiac diseases (platelet aggregation, inflammatory biomarkers, etc). Stay tuned for the results in the near future.