Targeting lifestyle change in patients with depression

Depression is widely recognized as being important for both mental and physical health. More than one in five of adults suffer from major depression (MDD) at some point in their lifetimes, and estimates are even higher for patients with coronary heart disease (CHD). Approximately 20% of patients suffer from MDD following a myocardial infarction and an additional 20% may experience elevated depressive symptoms without meeting diagnostic criteria for MDD. Depression is associated with poorer quality of life and increased medical expenditures and healthcare utilization. In addition, depression is one of the leading sources of disability in the United States and is associated with greater absenteeism and missed time at work. According to the World Health Organization, depression is the second leading cause of disability adjusted life years, following only CHD.

There also is a growing literature that has documented the association of depression with adverse health outcomes, particularly in patients with CHD. Numerous studies from multiple research teams have reported that depression is associated with a 2- to 4-fold increase in risk for death and non-fatal CHD events among patients with a variety of cardiac conditions including recent myocardial infarction, coronary disease requiring surgical revascularization, stable and unstable angina, and chronic heart failure^1–3. As a result of these findings, the American Heart Association commissioned a Scientific Advisory Panel to evaluate the risk with depression in CHD patients⁴. While recognizing that there was no direct evidence that screening for depression leads to improved outcomes, the Panel concluded that because of the overwhelming evidence for the association of depression with worse prognosis, assessment of depression was important with the goal of “targeting those most in need of treatment and support services.”

One of the most critical issues now facing clinicians is how to best treat depressed patients. Current recommendations for depression treatment typically involve pharmacotherapy, usually with a selective serotonin reuptake inhibitor (SSRI), although treatments with tricyclic antidepressants, benzodiazepines, and combined therapies are also prescribed. It has been estimated that approximately 50% of patients will have a clinical ‘response’ to treatment (i.e., a 50% reduction in symptoms), whereas many patients will require augmented treatment with more than one antidepressant agent⁵. However, many patients do not respond to antidepressant medications or experience untoward side effects. Indeed, although data are limited, results from a several randomized controlled trials of antidepressants in cardiac patients have provided negative or equivocal findings. For example, the SADHART trial was a randomized, double-blind, placebo-controlled, 24-week trial of sertraline for MDD among patients hospitalized for acute myocardial infarction⁶. Improvement in depressive symptoms among participants treated with sertraline was observed only in a subset of patients with more severe depression; there was no difference between sertraline and placebo in the full sample. The study was not powered to examine clinical outcomes. A second study (the SADHART-CHF trial) confirmed the safety of sertraline among patients with chronic heart failure, but sertraline did not reduce depressive symptoms any more than placebo⁷. Moreover, participants treated with sertraline had no better clinical outcomes compared to those receiving placebo. The MIND-IT study failed to demonstrate improvement in either depressive symptoms or cardiac outcomes with antidepressant treatment⁸. In one of the few positive trials, the CREATE study⁹ reported that depressive symptoms and remission rates were improved with citalopram compared to placebo, whereas no differences in depressive symptoms were observed between patients receiving interpersonal therapy and clinical management compared to patients receiving clinical management alone. The sample was too small to examine clinical outcomes, however. Meta analyses of randomized, controlled trials of anti-depressant medications in cardiac patients generally have observed no significant differences between antidepressant medication and placebo controls, but the authors encouraged the use of antidepressant drugs because non-randomized trials showed greater benefits with medication^10,11.

One of the problems in developing effective interventions for depression in cardiac patients is that the mechanisms underlying the association between depression and worse outcomes are not fully understood. A number of potentially interrelated of biobehavioral pathways have been studied, albeit often individually. Depressive symptoms have been linked to autonomic nervous system dysregulation, vascular endothelial dysfunction, abnormal inflammatory processes, and maladaptive health behaviors¹². The Heart and Soul study noted that among a variety of health behaviors, physical inactivity was especially important¹³. Furthermore, there is mounting evidence that patients who exhibit a worsening of depression over time appear to be at especially increased risk for death or other adverse outcomes including non-fatal MI or cardiac-related hospitalizations^14,15. It remains unclear whether depression is causally related to worse outcomes or whether depression is simply a marker for more severe CHD or some other factor associated with worse outcomes.

In this issue of the Journal, Ye and colleagues¹⁶ seek to examine this question by studying prospectively the relationship between depressive symptoms and non-fatal myocardial infarction or all-cause mortality in a subset of participants in the REGARDS (Reasons for the Geographic And Racial Differences in Stroke) study. REGARDS is a population-based, longitudinal study of 30,000 African-American and white men and women ≥ 45 years designed to determine the reasons for the higher incidence of stroke mortality in the southeastern “stroke belt” region of the United States and among African Americans¹⁷. Individuals were identified from commercially available lists of residents and were recruited using an initial mailing followed by telephone screening calls. The present study was limited to 4,676 participants with a history of CHD at baseline and complete data. Over a median follow up period of 3.8 years, 125 of 638 (19.6%) participants with elevated depressive symptoms experienced a clinical event compared to 657 of 4038 (16.3%) participants without elevated depressive symptoms. After adjustment for medical comorbidities, medications, and lifestyle factors, the association of depression and worse clinical outcomes was no longer statistically significant, and the authors emphasized the importance of lifestyle factors in explaining the prognostic value of depression and CHD, and in guiding intervention efforts to reduce the elevated risk of depression in cardiac patients.

The authors are to be congratulated for their contribution to the literature with this interesting analysis of a large and important dataset. Several important limitations of the study should be acknowledged, however. As with many substudies in which the variables under investigation are of secondary importance to the parent study, the assessment measures were not optimal and only a limited number of potential mediators were examined. Depression was assessed by a seldom used, 4-item self-report measure of depressive symptoms that may not accurately classify depression in all patients, while the four behavioral factors relied on brief measures of self-reported alcohol use, medication adherence, smoking, and physical activity. This lack of precision may help to explain the lower-than-expected rates of depression in this CHD population and also some unexpected results, such as trends for better outcomes in patients with high alcohol consumption and poor medication adherence. Smoking and physical activity explained a substantial proportion of the association of depression and the combined endpoint of MI and death, leading the authors to recommend smoking cessation and increasing physical activity in this population.

These lifestyle recommendations should come as little surprise to readers of the Journal. Cigarette smoking is not uncommon in patients with MDD¹⁸, and patients who continue to smoke, despite having CHD, may be engaged in a highly maladaptive effort to treat their depression. Similarly, a number of cross sectional and observational studies have shown that physical inactivity is associated with depression. For example, in the Alameda County Study, Camacho and colleagues¹⁹ measured subjects’ activity level and depressive symptoms in 1965, 1974, and 1983. Compared to men and women who reported higher activity levels, those who were inactive at baseline were at greater risk for higher depression scores at the first follow-up. Participants who increased their physical activity level between 1965 and 1974, however, were at no greater risk for depression in 1983 than those who were active throughout the period. Conversely, those who became more inactive by 1974 were more likely to have higher depression scores in 1983 than those maintaining a high level of physical activity.

The effect of exercise on depression now has been examined in over two dozen randomized trials, and has been the topic of several meta-analyses and systematic reviews^20,21. Existing studies vary substantially in size, type of control group, methodological rigor, length of follow-up, and even the type of exercise modality. In a recent Cochrane review²⁰, the primary analyses were limited to those trials comparing exercise treatment with no treatment or a control intervention (n = 25). There were large, clinically meaningful improvements in depressive symptoms associated with exercise in comparison with controls (standardized mean difference [SMD] = −0.82 (95% CI −1.12 to −0.51), which was somewhat diminished when the analyses were further limited to those trials using intention-to-treat analyses and blinded outcome assessments. In addition, when analyses were conducted among the five trials that collected long-term follow-up data, the effects of treatment were again slightly weaker, consistent with a moderate clinical improvement (SMD −0.44, 95% CI −0.71 to −0.18). However, in three randomized trials^22–24, including a recent trial of 101 CHD patients²⁴, exercise training resulted in significant reductions in depressive symptoms that were greater than placebo and comparable, or even superior, to antidepressant medications. A recent substudy from the HF-ACTION trial²⁵ also found exercise to be superior to guideline-based, usual care in reducing depressive symptoms in over 2,300 HF patients.

So what is the take home message? The term "cardiovascular vulnerable patient" has been used to describe patients susceptible to acute coronary events based upon plaque, blood, or myocardial characteristics.^26,27 It has become clear that depression is one of those characteristics that also increases the vulnerability of CHD patients to adverse events. In their report, Ye and colleagues¹⁶ provide support for the prognostic importance of elevated depressive symptoms in CHD patients. Moreover, by identifying smoking and physical inactivity as mediators of this relationship, they remind us of the importance of promoting healthy lifestyle changes, and specifically smoking cessation and exercise training, to improve clinical outcomes in this vulnerable population.