Many advances in the understanding of the relationship between depression and cardiovascular disease (CVD) were reported in 2009. As the study of this relationship encompasses cardiology, psychiatry, behavioral medicine, and many other fields, it is difficult to keep abreast of new developments. Relevant papers are found in a variety of journals. Therefore, we systematically searched the empirical research on depression and CVD published in English in 2009. Our search yielded nearly 500 articles. We review here a few of the most provocative and potentially influential findings. We begin with an overview of the methodology of our systematic review and then summarize the key findings and controversies we identified from the 2009 literature before exploring each key finding in detail.
METHODOLOGY OF OUR SYSTEMATIC SEARCH AND RATIONALE FOR FINDINGS REVIEWED
Previous evidence demonstrated that the presence of depressive symptoms or a diagnosis of a depressive disorder predicts poor prognosis and reduced survival rates after any coronary artery disease diagnosis, including myocardial infarction (MI) and unstable angina, as well as after coronary artery bypass graft surgery (CABG).1
We aimed to see how this evidence base was expanded in 2009 by using a systematic search strategy to retrieve the most relevant articles about depression and coronary heart disease (CHD) from the MEDLINE and Psyc-INFO (Ovid interface) databases. The most relevant subject headings and free text terms were identified and combined with “or.” The two sets were then combined with “and.” Terms included “depression,” “depressive disorder,” “depress$,” “coronary artery disease” (CAD) “coronary disease,” “acute coronary syndrome” (ACS), “cardiovascular disease” (CVD), “coronary heart disease,” and “heart diseas$.” The final set was limited to the English-language literature and identified 494 unique articles published during 2009.
Closer inspection of titles and abstracts revealed well more than 100 articles directly relevant to the science and management of patients with CVD and depression or pronounced depressive symptoms. In light of this quantity, a thorough review of all new findings, editorials, and reviews is not feasible. Thus, we review here a few exciting articles in several distinct topical areas that could influence views of the relationship between depression and CVD as well as how we screen for and treat depression in patients with CVD. As with any review that is not strictly evidence-based, our choice of articles is subjective and incomplete, but we hope it will stimulate discussion and further exploration.
SUMMARY OF KEY FINDINGS AND CONTROVERSIES
The following numbered topics emerged as common themes or controversies from our survey of the 2009 literature. The remainder of this article will review findings in each of these areas in detail and discuss important clinical implications as appropriate.
1. Antidepressant use and adverse cardiac events. In surprising findings, antidepressant use was associated with increased risk of incident stroke, CVD, and sudden cardiac death in multiple large observational cohort studies. It is not known if the association is caused by unmeasured confounders or depressive symptom severity, although controlling for symptom severity did not reduce this elevated risk in some studies. Another interesting possibility is that the treatment-resistant depressed patient is at particularly high risk of CVD and death.
2. Effects of depression intervention in CVD patients. Four exciting randomized controlled trials on depression intervention in patients with CVD reported important efficacy results and suggested future directions for larger, definitive trials of depression treatment for these patients.
3. Depression screening and treatment in CVD patients. In the absence of large randomized controlled trials, the debate continues on whether depression screening or any type of depression treatment is beneficial, harmless, or harmful to patients with CVD. Continuation of this debate does not serve the health and well-being of patients or the public. Less controversial—and thus less discussed—is the important insight that psychiatric patients with depression should be routinely screened for cardiac disease and risk factors, as they are clearly at risk of CVD. We await clinical trials in this area to ensure that screening leads to improved CVD outcomes.
OBSERVATIONAL EVIDENCE ON ANTIDEPRESSANT USE AND CVD OUTCOMES
In an analysis of the Nurses’ Health Study, Whang et al examined the relationship of depressive symptoms and antidepressant use with sudden cardiac death and adverse cardiac events in 63,469 women without CVD.2 Depressive symptoms were assessed using the Mental Health Index, a five-item subscale of the Short Form-36 Health Survey. Among women who reported antidepressant use, most (61%) were taking a selective serotonin reuptake inhibitor (SSRI; sertraline, fluoxetine, paroxetine, or citalopram), while 39% reported use of other antidepressants. Women taking antidepressants were more likely to suffer sudden cardiac death, with a fully adjusted hazard ratio (HR) of 3.34 (95% confidence interval [CI], 2.03–5.50).
Krantz et al examined psychotropic medication use and risk of adverse cardiovascular events in 519 women from the Women’s Ischemia Syndrome Evaluation.3 Enrolled women underwent coronary angiography, were separated into four groups according to their psychotropic medication use (none, anxiolytics only, antidepressants only, or both anxiolytics and antidepressants), and were observed for a median of 5.9 years. Results revealed that women who received both medications had a higher risk for adverse cardiovascular events and higher all-cause mortality compared with those using neither medication, even after controlling for anxious and depressive symptoms. In addition, whereas the use of antidepressant medication was associated with a doubling of risk for subsequent CVD events (HR = 2.16; 95% CI, 1.21–3.93) and all-cause mortality (HR = 2.15; 95% CI, 1.16–3.98), use of anxiolytic medication alone was not. While this study did not examine cause of death, cardiac death is likely to have constituted a large proportion of total mortality in this cohort selected for likelihood of CAD.
Although CVD and death have long been outcomes of interest for those studying the effects of depression, additional end points have recently been investigated as well. In a prospective cohort study of 136,293 community-dwelling postmenopausal women in the Women’s Health Initiative, Smoller et al found that new antidepressant use was significantly associated with increased incidence of stroke and all-cause mortality but not with incidence of CHD.4 The rate of stroke per 1,000 person-years was 2.99 for subjects with no antidepressant use versus 4.16 for patients with new SSRI use; the rate of all-cause death was 7.79 versus 12.77, respectively. The rate of all-cause death for subjects with new tricyclic antidepressant use was 14.14 per 1,000 person-years. To address potential confounding by indication, the researchers obtained a propensity score from a logistic regression model to predict any new antidepressant use from demographic, lifestyle, risk factor, and comorbidity variables measured at baseline. New SSRI use was associated with a doubling of the risk of incident hemorrhagic stroke as well as fatal stroke. There were no significant interactions between use of SSRIs and use of statins or aspirin in terms of risk of hemorrhagic stroke.
In an interesting observational study of 7,709 patients with confirmed CAD but without a diagnosis of heart failure or depression (and without current antidepressant use), May et al found that a subsequent diagnosis of depression was associated with a significant 50% increase in the risk of heart failure.5 There was no difference, however, between depressed patients who were using antidepressants and those who were not.
Increased risk of bleeding with SSRI use, particularly in patients with CAD, has also been a concern. Kim et al evaluated 1,380 adults who received any anti depressant before CABG for in-hospital mortality or any bleeding events.6 After controlling for the percentage of patients taking SSRIs (78%), there were no significant differences between those taking SSRIs and those taking non-SSRIs in the rate of any bleeding events (6.5% vs 7.2%; odds ratio [OR] = 0.93; 95% CI, 0.50–1.76) or in-hospital mortality (3.1% vs 2.3%; OR = 0.88; 95% CI, 0.47–1.65). There was no increased risk of bleeding associated with SSRI use when the analysis was restricted to patients who received antiplatelet and anticoagulant therapy. Thus, compared with patients who received non-SSRI antidepressants, patients who received SSRIs preoperatively had no increased risk of bleeding or in-hospital mortality after CABG; however, this study did not evaluate the effect of no antidepressant use.
Another study hypothesized that the use of any drug with the potential to prolong cardiac repolarization would be associated with an increased risk of sudden death.7 Use of individual drugs was analyzed among 1,010 cases of sudden unexplained death and 3,030 living primary care controls, all from the community. SSRI use was associated with a doubling of risk of sudden death (OR = 2.21; 95% CI, 1.61–3.05), and tricyclic antidepressant use was associated with a nonsignificant trend toward increased risk (OR = 1.44; 95% CI, 0.96–2.13). Further analysis that stratified patients according to prior CVD showed that most of the association of SSRIs with sudden death was in those with existing CVD and not in those without CVD. Other drugs found to raise sudden death risk included the typical and atypical antipsychotics.