|Home | About | Journals | Submit | Contact Us | Français|
The purpose of the current paper is to review research on the relationship between smoking cessation and depressed mood post-Acute Coronary Syndrome (Myocardial infarction or unstable angina; ACS). Emerging evidence regarding the effect of anhedonia—a specific subcomponent of mood disturbance characterized by reduced pleasure—on post-ACS outcomes is also discussed. There is strong evidence that depression prospectively predicts post-ACS relapse to smoking. Weaker evidence suggests that smoking at the time of ACS is related to post-ACS depression. Bupropion is a particularly promising treatment for this population because of its smoking cessation and anti-depressant effects. Future research should focus on the relative risk of using nicotine replacement therapies post-ACS, the efficacy of bupropion for smoking cessation and depression reduction in post-ACS patients, the role of anhedonia in post-ACS smoking cessation, and the development and testing of integrated behavioral treatments (smoking cessation plus interventions targeting mood management) for ACS patients.
Cigarette smoking and depression are independently linked to both the development and exacerbation of cardiovascular disease (CVD) . Recent reviews have discussed the relationships among depression, smoking, and CVD in general [1, . The current paper will focus on of the relationship of depression and smoking specifically among those hospitalized for Acute Coronary Syndrome (myocardial infarction or unstable angina; ACS). Hospitalization for ACS is a unique opportunity to initiate smoking cessation treatment, but smoking cessation may be complicated by depressed mood. In addition, recent studies indicate that anhedonia, a specific component of depressed mood that involves lack of positive affect, is an independent risk factor for failure to quit smoking and predicts post-ACS morbidity and mortality. Anhedonia has not been included in previous reviews. Thus this review will also highlight the effect of anhedonia on both quitting smoking and poor health outcomes post-ACS.
Smokers have a higher incidence of depressed mood, major depression, and history of major depression than non-smokers . Smokers with high levels of current depressed mood are significantly less likely to quit , and even very low levels of depression can have an adverse impact on maintenance of abstinence . Smokers with depression appear to experience stronger nicotine withdrawal symptoms than smokers without depression . Social support (which is also relevant to post-ACS outcomes ) may buffer the effect of depressed mood on smoking cessation . The incidence of depression is high among medically ill smokers, but the relationship between depression and quitting smoking may be complicated in this population. For example, one recent study suggests that depression has greater influence on cessation among smokers with low physical quality of life than among those with higher physical quality of life .
There is debate regarding whether smoking cessation increases, decreases, or has no effect on depressed mood   . This issue is highly relevant to the post-cardiac event population as both depression and continued smoking predict poor cardiac outcomes. Thus, quitting smoking at the cost of triggering a depressive episode would be “trading” one risk factor (smoking) for another, perhaps less potent, risk factor (depression). A 2007 review was inconclusive regarding the issue of whether or not successful smoking cessation increased symptoms of depression . Recent research using more sophisticated statistical modeling suggests that successful smoking cessation may even improve mood in some populations of smokers .
It is often overlooked that depression is a multifaceted syndrome that includes multiple subdimensions . One common method of subdivision splits depression into independent dimensions of negative affect and anhedonia/low positive affect. Negative affect is associated with the experience of aversive emotions (e.g., sadness, irritability, anxiety, and agitation) and hyper-responsiveness to aversive stimuli. In contrast, anhedonia involves deficient levels of positive emotions (e.g., feelings of joy, interest, and alertness) and a lack of responsiveness to pleasant stimuli. These two dimensions are psychometrically distinct  associated with different neural underpinnings , and have unique psychosocial correlates .
A recent review  indicated that anhedonia and negative affect may have similar relationships with smoking heaviness (number of cigarettes per day), chronicity, and nicotine dependence. Regarding cessation success, anhedonia and negative affect exhibit somewhat divergent findings. Relations between negative affect and relapse propensity have been demonstrated in some studies (e.g., ), but not others (e.g., ). In contrast, several smoking cessation studies have found that higher levels of anhedonia prospectively predict poorer cessation outcomes [5, [18, [20, [21,  and more rapid relapse [5, [20, . Furthermore, Leventhal et al. found that precessation anhedonia significantly predicts reduced odds of abstinence after controlling for baseline negative affect, other depressive symptoms, history of major depression, and nicotine dependence severity, whereas negative affect did not independently predict outcomes . Thus, anhedonia may play a unique role over and above other depressive symptoms in smoking relapse vulnerability.
Cigarette smoking is well established as a risk factor for both cardiovascular disease in general and ACS specifically. ACS patients are at significantly higher risk for recurrent ACS and death if they continue to smoke , and smoking cessation following ACS reduces risk for all cause mortality and subsequent coronary events by 1/3rd . While the occurrence of ACS motivates some patients to quit, approximately half return to smoking in the next 6 months (e.g., ).
First line treatments for smoking cessation include behavioral treatments, five forms of nicotine replacement therapy (patch, gum, lozenge, nasal spray and inhaler; the first three are available over the counter), and two oral medications (varenicline and bupropion) available by prescription only . Although nicotine replacement therapy is not an independent risk factor for cardiac events and appears to be safe for most patients with stable CVD , there is limited data on the safety of nicotine replacement therapy immediately following ACS . The FDA package inserts recommend that nicotine replacement should be “used with caution” among some cardiac patients, including those in the immediate (within 2 weeks) post-myocardial infarction (MI) period and those with unstable angina pectoris . The American College of Cardiology/American Heart Association guidelines for the management of patients with ST-elevation MI  do not recommend routine use of nicotine replacement in the hospital, but do allow for its use with select patients at discharge. To our knowledge there are no published data on the efficacy of nicotine replacement post-ACS. When providers consider using nicotine replacement with this population, the unknown risk of nicotine replacement therapy at the time of ACS hospitalization must be balanced with the known high level of risk associated with continued smoking.
One study suggests that varenicline is effective for patients with stable CVD . However, based on the data from this trial, the FDA recently released a drug safety communication  indicating that varenicline “may be associated with a small increased risk of certain cardiovascular adverse events” (e.g., non-fatal MI) in patients with CVD. Furthermore, a recent meta-analysis  concluded that Varenicline increases risk for cardiac events among patients without pre-existing CVD. The safety profile of varenicline has not been specifically tested in post-ACS patients.
Bupropion appears to be safe and effective for patients with stable CVD . Bupropion also appears to be safe for ACS patients and has shown some efficacy at 3 month follow-up in this population . However, bupropion has yet to be shown to be more effective than placebo in influencing 1-year abstinence rates among ACS patients also receiving intensive counseling interventions [32, .
Clinical practice guidelines recommend that all physicians take the time to strongly advise every smoking patient to quit, and as little as 3 minutes of brief advice can effect cessation rates . Given these general population findings, providers should advise smoking cessation in all post-ACS smokers. However, two recent reviews suggest that, ideally, post-ACS smoking cessation counseling would begin in the hospital and continue for at least 1 month and possibly 3 months post-discharge [34, . In general populations of smokers, a combination of medication and behavioral smoking counseling produces higher cessation rates ; therefore, combination treatment should be considered in ACS patients when feasible.
Approximately 31% of ACS patients have clinically significant depression , and there is a clear relationship between depressed mood and poor medical outcomes following ACS. A 2004 meta-analyses  found 22 studies exploring this relationship and found that depression within 3 months post-MI was predictive of all cause mortality, cardiac mortality, and new cardiovascular events up to 2 years later. This study concluded that post-MI depression is associated with a 2–2.5× greater risk of these poor outcomes. Depression has also been linked to cardiac mortality at longer follow-ups .
Several studies have found that depression remains predictive of poor post-ACS health outcomes when entered into multivariate models that control for potential confounds, such as medication and prior depression . Given the relationship between depression and smoking discussed above, it is particularly important that smoking is controlled for when assessing whether depression predicts medical outcomes post-ACS. Several studies have shown a significant relationship between post-ACS depression and poor health outcomes after controlling for baseline smoking status , suggesting that depression is not related to later cardiac outcomes solely through baseline co-morbidity with smoking.
There is some controversy regarding the importance of the timing of post-ACS depression measurement. Some data indicates that depression symptoms developing in the weeks after ACS (rather than preexisting or developing at the time of ACS) are particularly predictive of poor cardiac outcomes , while other evidence suggests that depressive symptoms present at the time of ACS that persist after discharge are more important . Some researchers are also concerned that assessment of depression among patients who are still hospitalized identifies too many patients with transient depression symptoms related to their cardiac event that resolve naturally in a matter of weeks .
Biological mechanisms that may mediate the relationship between depression and post-ACS event outcomes include heart rate variability, platelet activation, endothelial dysfunction, hypothalamic-pituitary-adrenocortical system activation, and inflammatory processes . There are also several possible behavioral mechanisms underlying the relationship between depression and post-ACS outcomes, including adherence to recommended lifestyle modifications (e.g., diet, physical activity, smoking cessation)  and delay in treatment seeking following onset of ACS symptoms .
Some behavioral and pharmacological treatments appear efficacious for reducing post-ACS symptoms of depression [6, . However, despite the strong evidence for a depression/post-ACS outcome relationship, there is no convincing evidence from randomized trials that behavioral or pharmacological treatment for depression following ACS improves cardiac outcomes. A recent meta-analysis of five RCTs comparing selective serotonin reuptake inhibitors (SSRIs) to placebo by Mazza et al.,  further complicates this issue. They found that SSRIs are safe for post-ACS patients and that SSRI treatment may reduce re-hospitalization in post-ACS patents. However, they also found that SSRI treatment does not significantly reduce depression symptoms in post-ACS patients.
As discussed above, the utility of dividing depression into unique dimensions of anhedonia and negative affect has become clear in recent smoking cessation research. Recent studies have assessed whether anhedonia prospectively predicts post-ACS health outcomes. *Davidson et al.  conducted a post-ACS observational study using a sample of 453 post-ACS patients. They assessed for the presence of anhedonia and depressed mood using a structured clinician interview within one week of hospital admission. They found that the presence of anhedonia, but not depressed mood, prospectively predicted major cardiac events and mortality (combined) over the next year after controlling for multiple co-variates, including age, co-morbidity, and cardiac functioning (smoking was not included as a covariate). When entered into the same model with depressed mood, anhedonia continued to predict major cardiac events and mortality while depressed mood did not.
*Leroy et al  assessed 291 patients for self-reported depression and anhedonia 1–4 days following ACS hospital admission and tracked occurrence of “severe cardiac events” (mortality or MI) and “clinical events” (severe events plus, recurrence of ACS, hospital readmission, and heart failure) for three years. They found that both dimensional and categorically defined anhedonia predicted occurrence of severe cardiac and clinical events over 3 years. Categorically defined “anhedonics” were twice as likely to experience a severe cardiac event. In multivariate analyses, dimensional anhedonia was an independent predictor of poor cardiac outcomes while depressed mood was not.
Doyle et al  followed 408 ACS patients for a median of 67 weeks post-event. They found that anhedonia predicted “major cardiovascular events” (cardiac related mortality, recurrent ACS, or unplanned revascularization) after controlling for age, sex, history of cardiac disease, and length of hospital stay. However, anhedonia was no longer predictive of major cardiovascular events after a scale of sadness/fatigue was entered into the model.
All three above studies suggest that anhedonia predicts adverse cardiac outcomes and is worthy of more research. The question of whether anhedonia is a better predictor than negative affect or overall depression awaits further work. However, 2 of the 3 studies reviewed above and a previous study of patients following percutaneous coronary intervention with stent implantation  suggest that anhedonia may predict cardiac outcomes over and above negative affect post-ACS.
There is clear evidence that depressed mood and smoking each have an independent influence on morbidity and mortality following ACS. Depression may also affect morbidity and mortality by interfering with smoking cessation. However, intensive counseling focused on depression does not appear to have a serendipitous effect on smoking outcomes among most ACS patients . Table 1 presents several studies linking depression following ACS to smoking status at 3 months to 13 years follow-up.
Some of the most important data on the interaction of depression and smoking post-ACS come from Thorndike et al.  et al. who conducted a secondary analysis (n = 245) of their previously-published RCT testing bupropion vs. placebo for smoking cessation in hospitalized CVD patients . All patents also received counseling for 12 weeks post-discharge. Approximately 90% of patients in this trial had ACS. Depression was measured by self-report during hospitalization. Smoking status was assessed at 4 weeks (self-report of 4 week continuous absence), 3 months (biochemically verified), and one year (biochemically verified). Results indicated that moderately to severely depressed patients were more likely to relapse in the 4 weeks post discharge and to be smoking at 3 months and 1 year post discharge. Depressed patients experienced more nicotine withdrawal symptoms than non-depressed patients while in the hospital, which mediated the effect of depression on 3 month smoking outcomes. Authors estimate that greater nicotine withdrawal symptoms accounted for an estimated 27% of the effect of depression on cessation outcomes. This study has the major advantages that long-term abstinence was bio-chemically verified and a mediator of the effect of depression on smoking outcomes in this population was identified.
*Gerber et al.  recently linked depression at the time of ACS to continued smoking up to 13 years later. Their sample included 330 MI patients smoking at the time of hospitalization. Self reported depression was assessed within a week of discharge for MI hospitalization, and self-reported smoking was assessed at 3–6 months, 1–2 years, 5 years, and 10–13 years post-MI. Participants were categorized as “persistent smokers” (i.e., smoked at all follow-up time points), “intermittent smokers” (i.e., smoked at some follow-up time points), or “continuous abstainers” (i.e., did not smoke at any follow-up time points). Depression was a significant predictor of persistent smoking after controlling for demographic measures, smoking behavior, CVD risk factors, MI characteristics, acute management, and co-morbidities.
There is less research regarding the effect of smoking at the time of ACS on depression symptoms during ACS recovery. There is some evidence indicating that smokers are more likely to be depressed up to 5 years following ACS  and to develop a new episode of major depression in the weeks following ACS  (See Table 2).
Some of the most relevant data on this issue come from a study by Kronish et al  who assessed self-reported depression at the time of ACS and at 3 month follow-up. Adherence to secondary prevention behaviors, including smoking cessation, was assessed by self-report at 3 months. Patients were categorized as “persistently non-depressed” (non-depressed at both time points), “persistently depressed” (depressed at both time points) and “remitted depressed” (depressed in hospital and non-depressed at 3 month follow up). While the study included 492 patients, only 88 were smoking at baseline. Patients with persistent depression were less likely to be abstinent at 3 months than those who were persistently non-depressed. However, more interestingly, among the 88 baseline smokers, those patients whose depression had remitted at 3 months were more likely to be abstinent at 3 months (67%) than those whose depression was persistent (26%). To our knowledge, these are the only findings that smoking cessation among depressed ACS patients is associated with concurrent improvement in depression, suggesting that smoking cessation following ACS may improve depressed mood. This finding coheres with results in non-ACS smokers attempting to quit (e.g., ).
To our knowledge, there is limited data on the interaction of anhedonia and smoking in ACS patients. However, one study reported that anhedonic ACS patients were more likely to be smoking at the time of ACS admission . In a study including both ACS and non-ACS patients, those who were anhedonic when entering cardiac rehabilitation were more likely to be smokers .
ACS patients who are smoking and already depressed at the time of their hospitalization may need additional support for smoking cessation. Clinicians should consider using smoking cessation medication that also has beneficial effects on depressed mood. Clinicians treating post-ACS patents who smoked at the time of their event should screen for and be vigilant of post-ACS development of depression symptoms.
Clinical practice guidelines state that nicotine replacement products should be used with caution in the 2 weeks post-ACS, and there is a lack of comprehensive safety data for using nicotine replacement therapy in this population. This has significant implications for ACS patients that are depressed and smoking. Both depressed smokers in general and depressed smokers with ACS appear to experience stronger nicotine withdrawal than non-depressed smokers [5, . Thus, reluctance to recommend nicotine replacement to ACS patients may be particularly harmful for depressed smokers with ACS. Thus, establishing the safety of nicotine replacement therapy or finding an alternative may be particularly important for this population. Prescribing varenicline currently has similar limitations; There is concern regarding increased cardiac events in both stable CVD and non-CVD samples, it has not been tested for safety in ACS patients, and there is some concern that it may have negative psychiatric side effects  . Thus, even if varenicline is found to be generally safe in post-ACS smokers, it may not be an appropriate 1st line treatment for post-ACS smokers with depression or with a significant psychiatric history.
It appears that treatments or combinations of treatments that effect both depression and smoking may be needed for those smokers who are depressed or at risk for depression (e.g., history of major depression) at the time of ACS hospitalization. Bupropion has promise as such a treatment as it 1) is an antidepressant, 2) is effective for smoking cessation and recommended for depressed smokers in the general population , and 3) reduces nicotine withdrawal . In addition, a recent post-hoc analysis of a large RCT indicates that bupropion may increase long-term cessation among depressed, post-ACS smokers . However, these results are preliminary and await replication. Note that another antidepressant, Nortriptyline, has been shown to be efficacious for smoking cessation but is not considered a first line treatment and may have cardiovascular side effects .
Counseling treatments that have added mood management strategies to smoking cessation treatment have been developed . However, results have been surprisingly mixed, indicating that these treatments are only more effective than standard treatments in smokers with a history of recurrent major depression . Thus the development of novel treatments for depressed, post-ACS smokers is needed. Any new treatment for this population should be integrated, to the extent possible, into existing medical visits. For example, newly developed intensive treatments might be integrated into the cardiac rehabilitation setting.
Behavioral Activation (BA) is a well-established empirically supported treatment for depression that has recently shown promise in a pilot trial for both reducing depressed mood and smoking cessation in a sample of depressed smokers . BA purports to improve mood and facilitate smoking cessation by increasing pleasant, goal directed, and valued activities through repeated between appointment homework assignments. There are several reasons BA may be ideal for smoking cessation and mood improvement post-ACS: 1) Although BA is not an exercise treatment, completing between appointment goals is often incompatible with remaining sedentary, and sedentary lifestyle is predictive of mortality in CVD patients ; 2) BA has been successfully integrated with other health behavior change interventions ; 3) BA is thought to be simpler and easier to train than other empirically supported counseling treatment packages ; and 4) BA directly seeks to increase positive affect rather than targeting reduction of negative affect; thus, BA may be particularly effective for anhedonic patients, which is notable given the role of anhedonia in both smoking cessation and post-ACS health outcomes  .
Depressed smokers with ACS are a high risk population. The most urgent research issues in this population include testing the relative risk of nicotine replacement therapies post-ACS, exploration of the efficacy of bupropion for smoking cessation and depression reduction in depressed, post-ACS smokers, and the development and testing of integrated (smoking cessation plus mood management) counseling treatments. In addition, the relevance of anhedonia in smokers with ACS is clearly worthy of future study. For example, a trial examining whether a treatment focused specifically on increasing positive affect (rather than decreasing depression in general) improves depression, cessation rates, or cardiac outcomes would be worthwhile. One weakness of the smoking/depression/post-ACS outcome literature is the over reliance on self-report for the assessment of both depression and smoking status. Longitudinal studies that assess both depression (ideally using a structured clinical interview) and smoking status (ideally using bio-chemical verification) at multiple time-points following ACS would allow for more nuanced conclusions regarding causality and the specific trajectories of depression in smokers with ACS.
The preparation of this paper was supported by K23-HL107391 to A. Busch, R01 CA137616 to B. Borrelli, and K08-DA025041 and R01-DA026831 to A. Leventhal.
Andrew M. Busch, Centers for Behavioral and Preventive Medicine, Program in Nicotine and Tobacco, Alpert Medical School of Brown University and The Miriam Hospital.
Belinda Borrelli, Centers for Behavioral and Preventive Medicine, Program in Nicotine and Tobacco, Alpert Medical School of Brown University and The Miriam Hospital.
Adam M. Leventhal, Departments of Preventive Medicine and Psychology, University of Sothern California Keck School of Medicine.