Glen L. Xiong Department of Psychiatry and Behavioral Sciences, Department of Medicine, University of California, Davis, CA, USA
The relationship between mood states and the heart has been known since antiquity. Across various cultures, statements in present-day languages such as 'my heart aches' are used to communicate depressive emotions. Over the past three decades, a large body of evidence has emerged that documents the adverse impact of depressive disorders on cardiovascular disease. This confirms the early suspicion of astute clinicians that psychological factors play a significant role in the genesis and the course of heart disease, as well as the ancient belief in a mind–body connection in general and human moods and the heart in particular.
This chapter examines epidemiological studies that have investigated the relationship between depression and heart disease, with specific focus on the prevalence of depression in different populations with cardiovascular disease and the adverse effects of depression on clinical outcomes.
We reviewed the literature through MEDLINE searches on English-language articles published between 1966 and September 2009 with the terms 'heart disease', 'ischaemic heart disease', 'myocardial infarction', 'coronary heart disease', 'heart failure', 'depression', 'depressive disorder' and 'antidepressants'. Because the heart diseases being investigated are almost exclusively cardiovascular, we use the term coronary artery disease (CAD) as a general descriptor unless otherwise specified. Depression is used as a general term for all depressive disorders and symptoms of depression.
DEPRESSION AFTER MYOCARDIAL INFARCTION OR UNSTABLE ANGINA
The incidence of major depressive disorder, as defined by DSM-III criteria, after myocardial infarction (MI) has been reported to be 16% by both Frasure-Smith et al. and Schleifer et al. Other studies using similar diagnostic approaches have found rates up to 20%. The rate of depression based on self-administered questionnaires has been generally higher and has varied among studies. In a study by Denollet and Brutsaert, the reported incidence of depression after MI was as high as 50.5%.
As noted in Table 1.1, the follow-up period to examine the relationship of depression and prognosis after MI commonly lasted 6–12 months. The mortality rates among clinically depressed patients were always significantly higher than those of clinically non-depressed patients. The relative risk ratio for death within 6 months among post-MI patients with versus without major depressive disorder was reported to be 3.1 by both Schleifer et al. and Frasure-Smith et al. The results of 1-year follow-up varied (relative risk ratios from 2.3 to 7.5 across studies), possibly reflecting the different methods used to evaluate depression.
Frasure-Smith et al. demonstrated that patients diagnosed with depression post MI were more than five times more likely to die from cardiac causes by 6 months than those without major depression. At 18 months, cardiac mortality had reached 20% in patients with major depression, compared with only 3% in non-depressed patients. Recent work has confirmed and extended these findings. A meta-analysis of 22 studies of post-MI subjects found that post-MI depression was associated with a 2.0–2.5 increased risk of negative cardiovascular outcomes. Another meta-analysis examining 20 studies of subjects with MI, coronary artery bypass graft (CABG), angioplasty or angiographically documented CAD found a twofold increased risk of death among depressed compared with nondepressed patients. Though studies included in these meta-analyses had substantial methodological variability, the overall results were quite similar.
The impact of subclinical or minor depression (notable depressive symptoms, defined as a Beck Depression Inventory (BDI) score ≥ 10 but not meeting diagnostic criteria for major depressive disorder) on mortality after MI is no less than that of major depression. Frasure-Smith's group demonstrated that post-MI patients with a BDI score ≥ 10 were almost seven times more likely to die in the 18 months after their acute MI than were patients whose BDI score was <10. Such impact was independent of cardiac function, previous MI and frequency of premature ventricular arrhythmia, which are known risk factors for mortality in this population.
The time course of depression may be relevant in its relationship to CAD and cardiac morbidity. Post-MI depression includes both depression that pre-dated MI (non-incident depression) and depression that developed after MI (incident depression). In the Depression after Myocardial Infarction study, 25.4% of patients experienced depression during the year following MI, with 55.4% being non-incident episodes of depression. Interestingly, those with incident depression had a trend towards a lower ejection fraction, increased disability at 12 months and a statistically higher risk of revascularisation. Incident post-MI depression was associated with new cardiovascular events in follow-up with a hazard ratio (HR) of 1.65 (95% CI 1.02, 2.65), whereas non-incident depression was not (HR 1.12, 95% CI 0.61, 2.06).
In another sample of patients (N = 489) hospitalised with acute coronary syndrome (ACS), the odds of being readmitted or dying were seven times higher for those with incident (post-ACS) depression. Nevertheless, both incident and non-incident depression were found to increase the risks of cardiac death in the Enhancing Recovery In Coronary Heart Disease (ENRICHD) cohort with 1328 patients over 29 months. This study demonstrated that incident depression had higher risks of cardiac mortality (HR 3.1; 95% CI 1.6, 6.1) than recurrent depression (HR 2.2; 95% CI 1.1, 4).
Of interest, a study from the UK by Lane et al. reported that depression, defined as a BDI score ≥ 10, was not associated by logistic regression with increased mortality during 12-month follow-up, although it was associated with a further decline in quality of life. The reliability of a self-administered questionnaire to assess true depression in a British population is unknown. Another recent study from the UK, using the Hospital Anxiety and Depression scale to measure depressive symptoms after MI, had similar results, that is no association with mortality but a negative association with poor quality of life. Since the treatment of depression is not generally examined in these prospective studies, the impact of depression treatment on CAD outcomes may have contributed to different results in the various cohorts, with notable differences by country of study.
Irvine et al., on the other hand, found that depression, defined as a BDI score ≥ 10, was associated with increased sudden cardiac death only in the placebo arm (RR 2.45; 95% CI 1.14, 5.35) during a 2-year follow-up, and not in the patients who received amiodarone for arrhythmia after MI. The size of the patient sample on amiodarone might have precluded the ability to detect a difference in this subgroup of patients.
Few studies have reported the relationship of depression with reinfarction or other outcomes such as unstable angina, heart failure or repeat hospitalisations after MI. Ladwig et al. found that patients with depression after MI had an almost threefold higher risk of chest pain than non-depressed patients during 6-month follow-up. Although chest pain does not always reflect myocardial ischaemia, and depression has been considered to be associated with non-cardiac chest pain, any recurrent chest pain after MI warrants further evaluation.
The long-term (> 1 year) impact of major depression on mortality after MI has not been as well studied as the short-term one. Two recent studies report odds ratios of mortality at 5 years of 2.53 and 1.87. There were two longitudinal follow-up studies examining the relationship of depression and survival after MI for 7.9 and 8 years, respectively. Denollet and Brutsaert followed 87 Belgian patients for a mean 7.9 years (range 6–10 years) after MI. Depression was measured by the Millon Behavioural Health Inventory. How this inventory compares with those more frequently used in the literature is unknown. The relative risk ratio for cardiac death in the depressed patients was 4.3. The primary purpose of this study was to investigate the association of type D personality (tendency to suppress negative emotions) with mortality after MI; they did not report the independent ability of depression to predict survival. Parakh et al. studied 284 patients using the BDI and found a prevalence of depression of 43% at Johns Hopkins Bayview Medical Center in Baltimore. They found that depression at the time of MI was not associated with mortality at 8 years (HR 0.76; 95% CI 0.47, 1.24). They concluded that the effect of depression after MI in increasing mortality seems to wane over time. However, this negative result might be due to the relatively small sample size. The most recent study involved 369 patients from the Sertraline AntiDepressant Heart Attack Trial (SADHART) cohort after 7 years of follow-up. Glassman et al., using Cox proportional hazards regression models, found that baseline depression severity and failure of depression to improve substantially during treatment with either sertraline or placebo were strongly and independently associated withlong-term mortality(HR 2.30; 95% CI 1.28, 4.14, and HR 2.39; 95% CI 1.39, 2.44, respectively).
DEPRESSION AND STABLE CORONARY ARTERY DISEASE
Of the patients with known CAD but no recent MI, 12–23% have major depressive disorder by DSM-III or DSM-IV criteria. Two studies have examined the prognostic association of depression in patients whose CAD was confirmed by angiography. Carney et al. pioneered the studies in this area. In their study, a diagnosis of major depression by DSM-III criteria was the best predictor of cardiac events (MI, bypass surgery or death) at 1 year, more potent than other clinical risk factors such as impaired left ventricular function, severity of coronary disease and smoking among the 52 patients. The relative risk of a cardiac event was 2.2 times higher in patients with major depression than those with no depression. Mortality was not analysed separately from morbidity, most likely because of the small sample size.
Barefoot et al. provided a larger sample size and longer follow-up duration in their study of 1250 patients who had undergone their first angiogram. They tested patients with the Zung Self-Rating Depression Scale (SDS) after angiography, then followed those with significant CAD (≥ 75% diameter stenosis of at least one of the three major coronary arteries) for a mean of 19.4 years. A high SDS score was significantly associated with increased risks of cardiac death and all-cause mortality. Compared with non-depressed patients, those who were moderately to severely depressed had 69% higher odds of cardiac death and 78% higher odds of all-cause mortality. The mildly depressed had a 38% higher risk of cardiac death and a 57% higher risk of all-cause mortality than non-depressed patients. Of particular interest in this study is how the effect of depression on prognosis changed over time. During the first year of follow-up, the moderately to severely depressed patients had a 66% higher mortality rate than the non-depressed. This effect became somewhat weaker during years 25 of follow-up, but then re-emerged with an even greater association. The moderately to severely depressed patients had an 84% higher risk of mortality in years 6–10 than non-depressed patients, and a 72% higher risk beyond year 10 during the follow-up period. The persistent adverse impact of depression in CAD patients is consistent with the nature of depression, which is chronic, recurrent and fluctuating in severity. Unfortunately, the investigators did not report comorbidities such as reinfarction, revascularisations or repeat hospitalisations during follow-up.
More recently, studies have been extended to examine the prognostic impact of depression in other CAD populations. Lesperance et al. studied 430 patients hospitalised for unstable angina without requirement for coronary artery bypass surgery. Among patients with a BDI score ≥ 10, the rate of death or MI 1 year after depression assessment was almost five times higher than their non-depressed counterparts. The association remained after controlling for other significant prognostic factors, including baseline electrocardiographic evidence of ischaemia, left ventricular ejection fraction and number of diseased coronary arteries (adjusted OR 6.73). The depressed patients were also more likely to be readmitted for unstable angina, although this finding did not reach statistical significance. There was no interaction between depression and percutaneous transluminal coronary angioplasty.
Although depression is less frequent among elderly people, the association of depression with adverse prognosis in CAD remains in old age. The Systolic Hypertension in the Elderly Program followed 4367 hypertensive subjects aged over 60 years for an average of 4.5 years and found an increased risk of death (RR 1.3), MI or stroke (RR 1.2) associated with the presence of significant depressive symptoms. Furthermore, a study of 3701 subjects 70 years or older, followed for an average of 4 years, revealed that men with a new diagnosis of depression were about twice as likely to experience a cardiovascular event (RR 2.1) or die from cardiovascular causes (RR 1.8) than those without a history of depression.
DEPRESSION AND CORONARY ARTERY BYPASS GRAFTING
CABG surgery is frequently performed in patients with severe CAD. After a median follow-up of 9.4 years, surgical management compared favourably with medical treatment with respect to psychosocial outcomes, including lower depression and higher social functioning scores. Nevertheless, adverse surgical outcomes following CABG in patients with depression were found in several studies. Patients with peri-operative depression or higher depressive symptoms experienced more recurrent angina and non-fatal cardiac events and higher mortality.
Although a large number of studies have examined the role of depression in post-MI and CAD outcomes, fewer studies have examined the role of incident and persistent depression on clinical outcomes after CABG (Table 1.2).
Early studies examined the role of emotional distress and depressive symptoms on CABG outcomes. Scheier et al. used the Nottingham Health Profile and identified subjects who were in the 75th percentile in their severity of emotional distress before they underwent CABG. Over a follow-up period of 3 years, they found that 33 of 147 of those subjects had elevated rates of composite cardiac events, including CABG death, fatal MI, non-fatal MI, need for angioplasty and unstable angina. Another early study examined optimism (10-item Revised Life Orientation Test) and found that it was associated with decreased hospital readmission after CABG. Saur et al. studied 416 subjects before CABG who reported the presence of depressive symptoms (using the Short-Form 36 Health Survey) and found that 'feeling down in the dumps' and overall mental health score correlated with hospital readmission within 6 months of CABG, but not with death. The studies cited above examined only depressive symptoms and did not explore whether depression was a disorder or episode (e.g. 2 weeks of continuous symptoms to be considered an episode as defined by the DSM-IV). In another more recent study, Tully et al. found that anxiety symptoms seemed to be more predictive of mortality than depressive symptoms using the Depression Anxiety and Stress Scale in 440 pre-operative CABG patients. This study awaits further confirmation and is limited by the lack of a more rigorous diagnostic assessment.
(Continues...)
Excerpted from Depression and Heart Disease Copyright © 2011 by John Wiley & Sons, Ltd. Excerpted by permission of John Wiley & Sons. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.