| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
REVIEW ARTICLES |
From the Columbia College of Physicians and Surgeons, New York, NY (K.W.D.); the Cardiovascular Institute, Mount Sinai School of Medicine, New York, NY (K.W.D.); the Department of Psychiatry, Mount Sinai School of Medicine, New York, NY (N.R.); and the Department of Psychiatry, University of Montreal, Montreal, Canada (F.L.).
Address reprint requests to: Karina W. Davidson, PhD, Medicine, Columbia University College of Physicians and Surgeons, 622 W. 168th St, PH9 Center, Room 941, New York, NY 10032. E-mail: kd2124{at}columbia.edu
ABSTRACT
OBJECTIVE: The natural course of elevated depressive symptoms or subthreshold depression in patients with an acute coronary syndrome (ACS) is presented, as is the prognostic impact. Safe and effective psychological treatment options are desirable for subthreshold depression in patients with ACS, should they prove tolerable, efficacious, and cost-effective to cardiologists and their patients. To achieve this long-term goal, we propose focusing on 3 intermediate goals. First, we need to understand which symptoms or patterns of symptoms (eg, fatigue, anhedonia, guilt feelings) are specifically predictive of ACS recurrence. Second, the prevalence of known psychosocial vulnerabilities (proximal causes) of depressive disorders should be assessed in patients with ACS, to understand better the etiology of these symptoms in these patients. Third, randomized controlled trials of vulnerability-related, evidence-based psychological depression interventions in cardiac patients are needed. The ways in which psychological proximal cause theories are relevant—or irrelevant—for both the treatment of depressive symptoms in post-ACS patients and the prevention of ACS recurrence are discussed.
Key Words: depression, • acute coronary syndromes, • psychosocial intervention.
Abbreviations: CHD = coronary heart disease;; ACS = acute coronary syndrome;; BDI = Beck Depression Inventory.
INTRODUCTION
Even minor elevations in depressive symptoms (subthreshold depression, defined to include mood, somatic, and interpersonal symptoms of depression, but not necessarily a diagnosable depressive disorder) significantly increase the risk of incidence of coronary heart disease (CHD) among previously healthy participants or worsen the cardiac prognosis in patients with established CHD. The scientific evidence is strongest for patients who have been hospitalized for an acute coronary syndrome (ACS)—that is, myocardial infarction or unstable angina (1–15).
Well-established theories of the psychosocial causes of depression and their theory-related, evidence-based treatments exist (16–20). However, these theories were formulated and tested primarily on younger, psychiatric (depressive disordered), and treatment-seeking people rather than older, subthreshold depressed, post-ACS patients. Furthermore, as has been previously found in psychiatric populations, there is likely no single proximal depression cause that is inevitably or necessarily present in post-ACS patients with depressive symptoms. We suggest that the usefulness of various psychological depression treatments for post-ACS subthreshold depression will differ depending on the extent to which each of these proximal causes presents in this specific population.
In this article, we outline current evidence for the natural course of depressive symptoms after an ACS event and their prognostic impact for ACS recurrence and cardiac death. We then review 3 accepted depression etiologic theories (cognitive theory, behavioral theory, and interpersonal theory) that all fulfill the following criteria: they have been tested for the presence of their putative depression proximal cause in depressed patients, they have an associated evidence-based treatment protocol (21–24), and they have been tested in elderly patients—of relevance for the majority of patients with ACS. Each theory asserts that their proximal causes should apply only to a subset of people with depression symptoms, because depression itself is a heterogeneous disease (25). The applicability of the existing psychosocial depression theories and their established treatment protocols to the population of ACS patients is then discussed in light of the effects that such treatments may have on preventing ACS recurrence and cardiac death, in addition to the effects on a general reduction in depressive symptoms and related functional impairments (26). Finally, alternative conceptualizations of subthreshold depressive symptoms in this patient population are discussed.
PROGNOSTIC VALUE OF DEPRESSIVE SYMPTOMS FOR ACUTE CORONARY SYNDROME RECURRENCE AND MORTALITY
Numerous studies have established that depression predicts the incidence of CHD in previously healthy people. A recent meta-analysis by Rugulies (15) reported that depression increases the risk of myocardial infarction or coronary death (relative risk = 1.64, 95% confidence interval = 1.29–2.08, p < .001) and that the risk is not limited to patients with clinical depression.
Similar findings have been reported in patients with established CHD. It appears that not only a psychiatric disorder history but also the presence of mildly elevated depressive symptoms, such as a Beck Depression Inventory (BDI) score of 10 or higher, are associated with worse prognosis in patients recovering from ACS (9, 27). Indeed, there appears to be a linear association between depressive symptoms at the time of a myocardial infarction and the risk of subsequent cardiac morbidity and mortality (6, 13) . However, when accounting for naturalistic changes in the course of depressive symptoms after the ACS event, the prognostic importance of mild vs. severe depressive symptoms differs. A recent follow-up study by Lespérance et al. (14) reported the prognostic importance of naturalistic improvement or deterioration of depression symptoms over the period of 1 year postmyocardial infarction. In this study, improvements in depressive symptoms had only an independent positive impact on prognosis (5-year cardiac morbidity, and cardiac and noncardiac mortality) in patients with initial BDI scores from 10 to 18 (mild depression symptoms). Among these patients, the greater the improvement in depression symptoms, the better the long-term prognosis. However, 1-year depression symptom improvement in those with higher initial depressive symptoms (BDI 18) did not result in decreased 5-year cardiac outcome risk. This finding provides preliminary evidence for the potential cardiac benefit of reducing depressive symptoms in a substantial proportion of patients—those with subthreshold depression.
To conclude, even minor symptoms of depression at the time of admission to the hospital predict post-ACS morbidity and mortality. Furthermore, there is some suggestion that previous history of major depressive episodes may further enhance the strong prognostic impact of these symptoms (28). Clearly, further studies should examine factors that potentially moderate the impact of depression besides its severity at baseline, such as the number, the timing and duration of previous episodes, and previous depression treatment history.
Although an enticing prognostic prediction is available from depression inventories, what is assessed by a depressive symptom inventory in post-ACS patients is not entirely clear. Some have argued that the symptoms endorsed by these patients (fatigue, loss of concentration, sadness, pessimism) may result from more severe cardiac illness that is not properly controlled for by assessing the number of diseased vessels or other medical severity measures in published studies (29). The meaning of depressive symptoms and depressive disorder diagnosis post-ACS and possible treatments are presented after reviewing the course of depressive symptoms in this patient population.
NATURAL COURSE OF DEPRESSIVE SYMPTOMS IN POST-ACUTE CORONARY SYNDROME PATIENTS
The course of elevated depressive symptoms in postmyocardial infarction patients has been characterized by Schleifer et al. (30). Of those admitted to the cardiology unit, 27% met criteria for minor depression as assessed by the Schedule for Affective Disorders and Schizophrenia (31), and an additional 18% met criteria for major depression when all were assessed within 10 days of their event. Schleifer et al. (30) reassessed their patients 3 to 4 months after admission. Seventy-seven percent of those with major depression still met criteria for this disorder, but only 35% of those with minor depression continued to have elevated depressive symptoms.
Another study examined the natural course of depressive symptoms in patients with CHD who had not experienced a recent major cardiac event (32). At baseline, 17% of 200 patients met the DSM-IV criteria for major depression, and an additional 17% met the DSM-IV criteria for minor depression. Seventy-five percent of the depressed patients were followed up over a period of 1 year. Of these, 42% of patients with initial minor depression had progressed to major depression 1 year later, and 31% of patients with initial major depression persisted in that category. Full remission was observed in 23% of the major depression group and 50% in the minor depression group. A similar pattern of naturalistic waning of postmyocardial infarction depressive symptoms was reported by Lespérance et al. (28).
Thus, a substantial portion of post-ACS patients may show spontaneous depression symptom remission, especially those who did not initially meet criteria for major depression. Moreover, patients who do remit spontaneously from subthreshold depression have a better cardiac prognosis (14). Consequently, the identification of predictors that are able to differentiate sensitively and specifically between the remitting and the nonremitting subthreshold depressed patients will be a crucial component for successfully identifying those most at risk and those most in need of depression treatment. This intermediate goal can be accomplished in 2 very different ways. First, it is critical to examine the types or patterns of depressive symptoms that predict whose subthreshold depression will remit spontaneously. Irvine et al. (33) examined somatic and cognitive-affective depressive symptoms separately for ACS recurrence prognosis, and only the latter were (marginally significantly) predictive for sudden cardiac death in a sample of 671 postmyocardial infarction patients. This kind of disentanglement of the different types of depressive symptoms typically surveyed within self-report instruments may aid us in identifying patients with ACS at risk for recurrence, and those whose depressive symptoms will persist.
Second, we believe that cognitive, interpersonal, and behavioral proximal depression causes are potential candidates for the differentiation of subthreshold depression remitters from nonremitters. These proximal causes have been shown to increase the likelihood of depressive episode presence, they may differentiate the less and more depressed patients from each other, and they have theoretically derived, evidence-based treatments. However, their prevalence and predictive value in the ACS population remain unknown.
BECK COGNITIVE THEORY OF DEPRESSION
The cognitive theory described by Beck (16, 34) evolved from his empirical observation of depressed patients’ descriptions of their thought content through verbalization. Cognitive psychotherapy takes the form of helping patients become aware of their cognitive distortions or cognitive errors and the underlying assumptions of these thoughts. The patient is then encouraged to seek evidence by which to support or refute these cognitive assumptions and to modify assumptions based on a more balanced view of all available information. There is a well-operationalized, manualized treatment procedure designed to replace maladaptive cognitive processes with more adaptive cognitions (18). Cognitive therapy of depression has been widely studied, and meta-analyses of its efficacy have found it to satisfy clearly the criteria for evidence-based treatment (35, 36). Several randomized controlled trials have shown that, when implemented by experienced therapists, it is as effective as pharmacotherapy (37), regardless of the severity of depressive symptoms, and may be even more effective than pharmacotherapy in preventing relapse of depressive episodes (38). The cognitive theory underlying this treatment, however, has been less well studied (39). Although there are many components of cognition that are important to cognitive depression theory, we focus on the 2 cognitive constructs with well-validated measures: cognitive errors and dysfunctional attitudes. Jointly, we label these cognitive distortions.
Cognitive Distortions in the Etiology of Depression
Cognitive distortions in the absence of a major life stressor are thought to be quiescent. Without specific cognitive intervention, they are also thought to be relatively stable. The conjoint presence of cognitive distortions and a major life stressor, however, is expected to increase the likelihood of a depressive episode. Following from this conceptualization, cognitive distortions should be higher in depressed compared with normal control post-ACS patients, both when a depressive episode is present and when it is absent. This latter point has not yet been established in the literature.
For cognitive distortions to be proximal causes for depression, they must precede the depressive episode; otherwise, they would be better conceptualized as epiphenomena. To address this question, Smith et al. (40) conducted a prospective study with a sample of 72 (nondepressed) patients with rheumatoid arthritis followed for 4 years. Cognitive distortions assessed at baseline by the Cognitive Error Questionnaire predicted significant increases in depressive symptoms 4 years later, controlling for baseline depressive symptoms. Moreover, baseline depressive symptoms did not predict changes in cognitive distortions across the study. Interestingly, increased helplessness (41) was also assessed at baseline and did not predict depressive symptom increase across the 4 years, suggesting that helplessness was not a depression proximal cause in this sample.
Support for the cognitive theory of depression would also be bolstered if cognitive depression interventions led to less cognitive distortion, such that the level of distortion for successfully treated patients were normalized or closer to the level of nondepressed people. Using the Dysfunctional Attitude Scale, depressed patients have been found to have significantly higher initial levels of dysfunctional attitudes than normal controls, and after cognitive therapy, those whose depression remitted had Dysfunctional Attitude Scale scores close to the levels of normal controls in 1 study (42). In contrast, the National Institute of Mental Health Treatment of Depression Collaborative Research Program did not find that cognitive distortions were lowered most by cognitive (behavioral) therapy (43–45) compared with a drug or interpersonal therapy arm.
Beck (46) and Ruehlman et al. (47) have cautioned that the correlates and causes of mildly elevated depressive symptoms or dysphoria cannot be safely assumed to be the same as those for major depression. Specifically, people with subthreshold depression are generally thought to have accurate cognitive perceptions of the stressful life event, resembling adjustment disorder, whereas those with major depression are not (48).
Cognitive theory would predict that post-ACS patients with high dysfunctional attitude scores are more likely to have more severe depressive symptoms compared with post-ACS patients with low dysfunctional attitudes. Furthermore, given that cognitive distortions appear relatively stable and increase a person’s vulnerability to a major depressive episode when combined with a major life event, depressive symptoms in the presence of cognitive distortions would not be expected to spontaneously remit in this patient population.
In summary, there is some evidence from a number of different designs suggesting the possibility that cognitive distortions may be a proximal cause of depression, but there is other evidence such as that from the National Institute of Mental Health trial that is troubling. From the lack of convincing empirical data on the postulates of cognitive theory, some have expressed more global concerns about the theory; they propose that all diathesis-stress models of depression have formidable conceptual and methodological challenges that have not yet been met (49). These challenges compromise the potential usefulness of these theories to depression treatment. However, the presence of cognitive distortions in post-ACS patients, regardless of current depressive episode status, could be 1 of the markers of excess ACS and depression nonremittance risk, because in theory it should mark previous and future vulnerability to depression.
INTERPERSONAL THEORY OF DEPRESSION
The interpersonal theory of depression is based on theories emanating from the interpersonal school of psychiatry (50) and empirical data related to attachment theory and social roles (51). Interpersonal psychotherapy, developed by Klerman et al. (19), is a focused, short-term, time-limited therapy that emphasizes the current interpersonal relations of the depressed patient. The efficacy of interpersonal psychotherapy treatment for major depression has been demonstrated in several controlled comparative depression treatment trials (45). For example, in a depressed geriatric population, interpersonal psychotherapy has shown some advantages over tricyclic antidepressant therapy because of attrition in the latter intervention, in part because of medication side effects (52). A brief psychosocial intervention, based on interpersonal psychotherapy, to treat medical patients in primary care has also demonstrated a reduction in depressive symptoms (19). Interpersonal psychotherapy has been successfully modified and used with older patients (53) and with subsyndromally depressed hospitalized elderly patients (54). Frank et al. (55) also demonstrated the efficacy of maintenance interpersonal psychotherapy, a modified interpersonal psychotherapy version focusing on the prevention of depression in remitted patients, in a long-term randomized controlled trial. Interestingly, they also showed that treatment specificity (the ability of patients and therapists to focus consistently on interpersonal concerns and the techniques of interpersonal psychotherapy to the exclusion of other treatment elements) was essential for successful depression relapse prevention.
Interpersonal psychotherapy focuses on the intuitively appealing concept that events in one’s psychosocial environment affect one’s mood and vice versa. When major events occur, mood worsens, and depression may result. Conversely, depressed mood compromises one’s ability to handle one’s social roles, generally leading to further negative events and ongoing interpersonal distress (56). The crux of interpersonal psychotherapy is to demonstrate empirically a link between mood and interpersonal issues that appear temporally and thematically related to the onset and maintenance of depression (19). Once a patient understands this link and identifies the specific interpersonal area that is currently problematic, the therapist and patient work together to alter the interpersonal environment so that the depression will lift (56). The patient and therapist agree on 1 of the following 4 interpersonal problem areas that will be the focus of the depression treatment: a) grief or complicated bereavement, b) role dispute or ongoing disagreements with a significant person in the patient’s life, c) a recent role transition that results in major interpersonal role changes or alterations (eg, retirement, moving, being diagnosed with a major medical illness), and d) interpersonal deficits (recurrent difficulties in social interactions, in their extreme form classified as personality disorders). Although depressed patients may fit into several or all of the 4 interpersonal problem areas, the treatment demands that 1 area (or occasionally 2) be chosen as the primary target for intervention.
Presence of Interpersonal Problems in the Etiology of Depression
Numerous studies have found interpersonal problems as reflected in divorce, marital problems, and negative partner and child interactions to be significantly more prevalent or elevated in depressed people (57, 58). However, it is still unclear whether interpersonal problems precede depressive episodes rather than co-occur with or even result from depressive mood states. On a variety of interpersonal functioning indicators, Hammen and Brennan (58) compared 83 women with unipolar major depression or dysthymic disorder, 271 women who were not currently in a depressive episode but who had past histories of either DSM-IV depression diagnosis, and 458 never-depressed women. As expected, the currently depressed women had the worst scores on all interpersonal indicators. Compared with the never-depressed group, and controlling for current subclinical depressive symptoms and socioeconomic status, the past depression women reported a more frequent use of coercive interpersonal tactics, more interpersonal conflicts, less secure attachment representations of relationships, and more dysfunctional personality traits. Moreover, their close relationship functioning was found to be more dysfunctional by interviewers, and the marital satisfaction of the women’s spouses was worse. Although cross-sectional, at least this 1 set of results suggests that impoverished relational skills and dysfunctional representations of relationships are proximal causes of depression that persist even in the absence of clinically relevant depressed mood.
Interpersonal psychotherapy explicitly focuses on the here and now of these interpersonal problems, and thus appears applicable to a wide range of patient populations with interpersonal problems. However, little research is available demonstrating that the tenets and postulates of interpersonal theory are supported. For example, it is not yet clear that the reduction or prevention of depressive symptoms is mediated through changes in 1 of the 4 interpersonal problem areas, such as relational functioning. Finally, the efficacy of this therapy in medically ill, mildly depressed patients has yet to be clearly established.
To conclude, the interpersonal theory has not been extensively tested for either increased presence of interpersonal problems before the depressive symptoms or for evidence that interpersonal psychotherapy reduces the identified interpersonal problem. The beneficial effect of interpersonal psychotherapy on depressive symptomatology is well established. Based on this theory, post-ACS patients with the presence of a role transition, or loss, or interpersonal deficits will be more likely to show elevated depressive symptoms. A second group at risk may be patients for whom the cardiac event itself represents a major loss or role transition. Patients with this depressogenic vulnerability may not be as severely depressed as those with cognitive distortions, but their depressive symptoms are also unlikely to remit spontaneously without intervention, because the defined interpersonal problems and interpersonal functioning deficits are considered relatively stable.
BEHAVIORAL THEORY OF DEPRESSION
The predominant behavioral theory of depression postulates that major life stressors can result in a depressive episode because they disrupt normal behavior reinforcement patterns (59). Originating from an operant conditioning paradigm, this theory views depression as the consequence of a lack of or decrease in the efficiency of positively reinforced behavior and perhaps overt punishment for behavioral initiation. This may be a result of a decrease in the availability of reinforcing events, one’s personal skills to act on the environment, the impact of certain types of events, or a combination of these. In addition, the mobilization of support from family and other social networks may result in a negative feedback loop of social reinforcement for depressive behaviors (eg, social withdrawal, positive social reinforcement for withdrawal, further withdrawal). In other words, in times of major stress from unexpected events, people may experience a low rate of positive reinforcement for mood-enhancing behavior and a higher rate of positive reinforcement for depressive behavior.
The behavioral treatment that derives from this theory of depression involves helping patients increase their frequency and quality of pleasant activities. It has been found that depressed patients have low rates of pleasant activities and obtained pleasure; their mood covaries positively with rates of pleasant activities and inversely with rates of aversive activities (60). Finally, behavioral treatment for depression has been shown to be efficacious—to reduce depression—in multiple randomized controlled trials (59, 61).
Lewinsohn (20) discussed dysphoria (defined by him as the presence of low levels of negative mood symptoms) as the affective state that results when few environmental positive reinforcers are available for a person’s behavior. Older people, and particularly patients hospitalized for a life-threatening event, are at greater risk than people at other points in the life cycle, with the possible exception of children, for being placed in situations in which their own behavior has little effect on the environment or the behavior of others (62). Thus, the disruption of the pattern of reinforcement for self-initiated behaviors that occurs when a patient experiences an ACS event is pertinent to the behavioral theory of depression. Frequently, ACS risk factor management recommendations compound this problematic behavior pattern. For example, patients who experience smoking or eating a saturated fat-rich meal as positively reinforcing frequently report that most of the pleasant or pleasurable activities in their lives have now been restricted or removed because of their health-damaging consequences.
Although therapeutic approaches to depression more often involve both cognitive and behavioral components than either 1 alone, component analyses have shown that behavioral activation appears to be as effective as cognitive therapy for altering negative thinking and dysfunctional attributional styles and for producing change in depressive symptoms and course of relapse (63).
Lack of Pleasant Events in the Etiology of Depression
The behavioral theory of depression posits that the specific proximal cause for depression is the interaction of the patient behavior and the reinforcement schedule of the environment. The frequency of pleasant events is lower in depressed compared with normal controls (64, 65), mood covaries with this frequency (60), and behavioral depression treatment increases pleasant event occurrence in those who have their depression successfully treated. For example, Wierzbicki and Rexford (65) studied the frequency and pleasantness or unpleasantness of positive and negative events (assessed with the Pleasant Events Schedule and the Unpleasant Events Schedule) in a clinical sample of 60 people diagnosed with any DSM-III depressive disorder (major affective disorder, dysthymic depression, or atypical depression), and in a nonclinical sample of 143 undergraduate students. They found that in both samples, depression scores on the BDI were negatively related to the frequency and pleasantness of pleasant events and were positively related to the frequency of unpleasant events. Importantly, Grosscup and Lewinsohn (60) demonstrated in depressed patients that scores on daily ratings of the Unpleasant Events Schedule and Pleasant Events Schedule were associated with daily fluctuations in mood level. Moreover, during the course of a specific treatment targeted at increasing pleasant activities, a decrease in the subjective aversiveness of events was associated with clinical improvement in depression symptoms.
In summary, the behavioral theory of depression postulates that low rates of positive events and therefore the absence of positive reinforcement are central to the induction and maintenance of depressive symptoms. Depressed people compared with nondepressed people have fewer pleasant events, and behavioral treatment aimed at increasing pleasant events successfully decreases depression. It is less clear whether the behavioral theory has a prediction about the course of depression in patients with ACS. One possibility is that as the time from the ACS event lengthens, the frequency of pleasant events occurrence and behavior reinforcement will naturalistically increase, and depressive symptoms will decrease. As a consequence, post-ACS patients with a behavioral reinforcement disruption as the proximal cause for their depressive symptoms will be more likely to remit spontaneously than those with the other 2 depression proximal causes.
INTERRELATEDNESS OF THE PROXIMAL CAUSES
In real-life clinical settings, the specific proximal causes that each theory focuses on are not treated as distinct dimensions of functioning, but rather as interrelated domains. This is supported by empirical evidence. For example, Whisman and Friedman (66) demonstrated that in a sample of 390 undergraduate students, higher levels of dysfunctional attitudes were associated with higher levels of interpersonal problems, even when controlling for negative affect. In the study of Wierzbicki and Rexford (65), higher dysfunctional attitude scores were related to less positive behavioral events and more unpleasant behavioral events in both a clinical, treatment-seeking sample and an undergraduate student sample. More importantly, both pleasant events and dysfunctional attitudes were independently associated with depressive symptoms in both samples. Although not directly tested in the Wierzbicki study, this also suggests an independent, additive effect of the presence of more than 1 proximal cause on depressive levels.
PROXIMAL CAUSES IN POST-ACUTE CORONARY SYNDROME PATIENTS
For the specific population of post-ACS patients, the prevalence of the proposed proximal causes and their relationship with depressive symptoms has yet to be determined. Theoretically, one would expect a higher incidence of depressogenic vulnerabilities in this population in comparison with the normal population, given that the overall rate of subthreshold and clinically diagnosed depressive disorder is higher (1, 4, 9). Furthermore, an ACS constitutes a severe life event with considerable threat potential and disruptiveness in people’s daily routines and social lives. Under these circumstances, all 3 vulnerabilities are likely to elicit depressive symptoms temporarily, exacerbate existing subthreshold depression, and even provoke the onset of a new episode of major depression. Interpersonal problems may arise from major lifestyle changes that are prescribed to the patients (eg, abstaining from shared activities, engaging in special diets). In the case that physically demanding activities have to be restricted (eg, caregiving for other family members), or partial or full work retirement is recommended, the ACS event coincides with a major, irreversible role transition. Pre-existing, quiescent cognitive distortions may be reinforced or activated through the experience of the uncontrollable, adverse, life-threatening event (eg, "bad things always happen to me," "I will never be able to lead a fulfilling life again"). Because of the restrictions and demands of lifestyle changes, patients may experience a decrease of pleasant activities and increase of unpleasant activities (eg, abstaining from smoking, drinking, adherence to aversive medication and attendance at routine, but anxiety-provoking medical checkups).
TREATING DEPRESSION IN PATIENTS WITH ACUTE CORONARY SYNDROME: WHAT IS REQUIRED?
For the application of existing psychosocial depression treatments to post-ACS patients, a number of parameters on which this population differs from psychiatric populations studied by the major depression theories must be considered. The original populations were generally self-identified; they were usually without major medical comorbidity; they were mostly young or middle-aged; they met full criteria for formal psychiatric or depressive disorders; and finally and importantly, they sought and accepted psychological treatment. In contrast, the majority of patients with ACS will not self-identify as being depressed, they all have an acute medical disease, many do not meet full criteria for a depressive disorder, and they may be unwilling to seek or accept depression treatment. Moreover, most of these patients are older than 50 years, with a substantial proportion of these patients elderly or very old.
These characteristics pose specific challenges for the implementation of psychosocial depression treatments. In patients with medical comorbidity, depressive symptoms are typically underdiagnosed (67). This applies especially to subthreshold depression (68) and depressive symptoms in the elderly population (69). Although depression is the most common mental health problem from which older adults suffer (69), they underutilize mental health services, they are less likely to self-identify as being depressed, and, once interested in seeking treatment, they must overcome many practical barriers to obtain mental health treatment (70). Even when referred to a mental health professional, older patients are less likely than younger patients to seek specialized mental health care (71); this is probably in part because of the stigma perceived by this patient population in seeking mental health treatment (72).
As Coyne and Thompson (73) have pointed out, evidence for successful treatment of depressive symptoms in populations of primary care patients is sparse and based on heterogeneous populations and interventions. Further, even when treatments reduce depressive symptoms, improvements in event-free survival rates are not necessarily achieved, as recent results of 1 medium-sized antidepressant and 1 large-sized cognitive-behavioral intervention trial suggest (74, 75). For women, psychosocial interventions targeted at loneliness or distress reduction even showed a trend toward higher post-ACS event risk (75, 76). Coyne and Thompson (73) argue that, when implementing and evaluating psychological interventions in primary care settings, the specificity of the conditions to be treated (eg, general distress, subthreshold depression, dysthymia, or major depression) and the specificity of the treatment protocols must be considered (73). These concerns clearly extend to post-ACS, subthreshold depressed patients.
For all of these reasons, it would be premature to introduce a theoretically derived treatment intervention to post-ACS dysphoric patients without first demonstrating that the presumed proximal causes (for example, cognitive distortions) occur at relatively high rates in this population. That is, given the differences between the population on which these theories were derived and tested and the post-ACS patient population, we need to test the generalizability of these depression theories by determining whether their postulated proximal causes are useful for characterizing the subthreshold depressed ACS population.
Should that be the case, the evidence-based depression treatment protocols will have to be tested for their acceptability, safety, and efficacy in this population. We hypothesize that a better matching between the primary proximal cause of depressive symptoms for each patient to the evidence-based treatment specifically designed to modify that proximal cause will improve patient and cardiologist acceptance of the treatment. Further documentation of sufficient efficacy in reducing subthreshold depressive symptoms in post-ACS patients—a desirable outcome in itself—along with a demonstration of pathophysiological risk reduction will aid in the planning of the next outcome-based depression intervention trial in this population. If typical depressogenic proximal causes appear to be absent or only minimally present in this patient population, alternative depression theories and alternative depression treatments must be explored for their potential application. Finally, the reasons why these types of symptoms may exist in these patients and the meaning of these symptoms must also be carefully considered. Thus, it may be that a third common factor—such as genetics—is driving the association noted between depressive symptoms and ACS recurrence.
ALTERNATIVE CONCEPTUALIZATIONS OF SUBTHRESHOLD DEPRESSIVE SYMPTOMS IN POST-ACUTE CORONARY SYNDROME PATIENTS
Link Between Subthreshold Depression and Acute Coronary Syndrome Recurrence Risk Factors Is Based on Common Genetic Factors
Both depression and coronary artery disease are highly prevalent, heterogeneous chronic conditions that are the product of multiple gene by environment and gene by gene interactions. That is, these conditions tend to run in families, but no single gene causes depression or coronary artery disease. For depression and coronary artery disease, a gene could be a risk factor, similar to an environmental factor such as diet or smoking. The pathophysiology of these conditions involves many interconnected pathways of otherwise distinct regulatory systems. For example, dysregulation of the coagulation, inflammatory, hormonal, metabolic (lipid and carbohydrate), autonomic, or endothelial systems may contribute to CHD. Interestingly, the genetic information coding for some basic element (ie, a protein) is used in different organs for different functions. For example, the serotonin transporter protein in the brain is thought to be involved in the regulation of mood, and the same protein on blood platelets participates in blood clot formation. The glucocorticoid receptors regulating the corticotropin-releasing factor in the brain are also present on lymphocytes to downregulate the inflammatory response. Therefore, it is reasonable to argue that frequent and normal variation of a gene, whether or not necessitating the interaction of an environmental factor or another gene, could be associated with an increased risk of depression and CHD (77). However, data confirming a specific genetic contribution common to both conditions are lacking. For example, for the serotonin transporter-linked promoter region polymorphism, the data are so far inconclusive. Research suggests that patients with the long allele (l/l) have a more efficacious serotonin transporter, leading to better reuptake of serotonin inside platelets, higher secretion of platelet factor 4 (78), and heightened platelet activation (79). Interestingly, a large study has found that the l/l allele predicts the incidence of myocardial infarction among previously healthy participants (80). In addition, the l/l allele is associated with a better (81) or a more rapid (82) antidepressant response to selective serotonin reuptake inhibitor. In contrast, it is the short allele (s/s), interacting with stressful life events, that predicts the incidence and the severity of depression (83). Clearly, a further explication is needed for the interaction between this allele and the environment, and this allele and pathophysiological mechanisms implicated in both depression and CHD.
Subthreshold Depression Is an Index of Illness
Even though empirical data strongly support the notion that subthreshold depression contributes to post-ACS recurrence and mortality risk independent of other known risk factors, we cannot rule out the possibility that these symptoms are merely an expression of more severe underlying medical problems. One possibility is that patients reporting minor depressive symptoms are actually reporting feeling more medically ill than nondepressed patients as a result of more severe underlying disease processes that are not captured by the known medical risk factors.
Link Between Subthreshold Depression and Acute Coronary Syndrome Recurrence Risk Factors Is Based on Common Inflammatory Processes
Another third variable or confounder possibility that has received much attention recently focuses on inflammatory processes, which are known to promote the progression of atherosclerosis (84) and also cause depressive symptoms. Studies of cell-mediated immunity have frequently (although not uniformly) demonstrated immunosuppression in depressed patients compared with normal controls (85). In animals, the experimental induction of infection and inflammation through the administration of proinflammatory cytokines results in a pattern of depression-resembling sickness behaviors (eg, reductions in activity, social and sexual interaction, and food and water intake; increasing anhedonia; and altered sleep;86). In humans, there is some evidence that immune activation through viral infections may precede depressive symptoms, and the administration of proinflammatory cytokines has been shown to produce depressive symptoms in healthy, nondepressed adults (87). However, a recent study comparing markers of systemic inflammation and severity of depressive symptoms in clinically depressed and nondepressed healthy adults failed to find evidence for the sickness behavior model (88). Also, successful treatment of depressive symptoms is associated with reductions in the magnitude of inflammation (89), suggesting that the relationship between depressive symptoms and inflammation may be bidirectional. It has yet to be determined whether the higher risk associated with elevated depressive symptoms in post-ACS patients is epiphenomenal because proinflammatory processes are driving the excess risk, or whether inflammation is the mediator of the causal effect of depressive symptoms on ACS recurrence.
VALIDATION OF DEPRESSIVE SYNDROMES IN POST-ACUTE CORONARY SYNDROME PATIENTS
From the available evidence, it appears safe to conclude that the more symptoms of what is commonly thought to be depression patients have at the time of their ACS hospital admission, the greater their risk for subsequent cardiac events. However, the meaning of these symptoms—whether they indicate a subclinical depressive episode, a history of depressive illness, or a chronic untreated inflammatory state, or reflect a more severe medical illness—has not yet been established, and clarification is sorely needed. Most studies have not considered the role of a previous history of depression, and this also needs to be considered carefully in determining the meaning of these depressive symptoms during a medical crisis. Before we know the meaning of these symptoms, treatment both for the relief of suffering and for the possible prevention of additional ACS events remains a formidable challenge.
SUMMARY
Post-ACS patients with a BDI of 10 or greater assessed during their cardiac event hospitalization are at independent, increased risk for sudden cardiac mortality and ACS recurrence. Unknown are the levels or rates of psychosocial depression proximal causes present in these patients and whether the proximal causes predict distinct courses (remittance vs. nonremittance) of depressive symptoms. Cognitive distortions and role transitions or losses are predicted to identify those who do not show depressive symptom improvement naturalistically, whereas absence of positive events may identify those who will be more likely to remit spontaneously. Studying the point prevalence of the differing depression theory proximal causes will allow us a window into the meaning of these elevated depression symptoms in post-ACS patients. Should depressogenic vulnerabilities be highly prevalent and predictive of nonremission and ACS recurrence, the selection of evidence-based depression treatments most likely to treat depressive symptoms successfully—and possibly prevent ACS recurrence—in these patients becomes clear. Alternatively, should the commonly studied depressogenic vulnerabilities be rare or nonpredictive in these patients, pursuing alternative conceptualizations of subthreshold depression symptoms becomes urgent.
ACKNOWLEDGMENTS
Preparation of this article was financially supported by #HC-25197 and #HL-04458 from the National Heart, Lung, and Blood Institute.
Received for publication May 12, 2003.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  J. A Blumenthal, A. Sherwood, S. D. Rogers, M. A. Babyak, P. Murali Doraiswamy, L. Watkins, B. M. Hoffman, C. O'Connell, J. J. Johnson, S. M. Patidar, et al. Understanding prognostic benefits of exercise and antidepressant therapy for persons with depression and heart disease: the UPBEAT study rationale, design, and methodological issues Clinical Trials, October 1, 2007; 4(5): 548 - 559. [Abstract] [PDF]
|
|
|
|
 |
|
 A. Sherwood, J. A. Blumenthal, R. Trivedi, K. S. Johnson, C. M. O'Connor, K. F. Adams Jr, C. S. Dupree, R. A. Waugh, D. R. Bensimhon, L. Gaulden, et al. Relationship of Depression to Death or Hospitalization in Patients With Heart Failure Arch Intern Med, February 26, 2007; 167(4): 367 - 373. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rutledge, V. A. Reis, S. E. Linke, B. H. Greenberg, and P. J. Mills Depression in Heart Failure: A Meta-Analytic Review of Prevalence, Intervention Effects, and Associations With Clinical Outcomes J. Am. Coll. Cardiol., October 17, 2006; 48(8): 1527 - 1537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rutledge, S. E. Reis, M. B. Olson, J. Owens, S. F. Kelsey, C. J. Pepine, S. Mankad, W. J. Rogers, C. N. B. Merz, G. Sopko, et al. Depression Symptom Severity and Reported Treatment History in the Prediction of Cardiac Risk in Women With Suspected Myocardial Ischemia: The NHLBI-Sponsored WISE Study. Arch Gen Psychiatry, August 1, 2006; 63(8): 874 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Sherwood, A. L. Hinderliter, L. L. Watkins, R. A. Waugh, and J. A. Blumenthal Impaired Endothelial Function in Coronary Heart Disease Patients With Depressive Symptomatology J. Am. Coll. Cardiol., August 16, 2005; 46(4): 656 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Blumenthal, A. Sherwood, M. A. Babyak, L. L. Watkins, R. Waugh, A. Georgiades, S. L. Bacon, J. Hayano, R. E. Coleman, and A. Hinderliter Effects of Exercise and Stress Management Training on Markers of Cardiovascular Risk in Patients With Ischemic Heart Disease: A Randomized Controlled Trial JAMA, April 6, 2005; 293(13): 1626 - 1634. [Abstract] [Full Text] [PDF]
|
|
| <||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
