Quick viewing(Text Mode)

Depression in Patients Recovering from a Myocardial Infarction

Depression in Patients Recovering from a Myocardial Infarction

GRAND ROUNDS At The Johns Hopkins Bayview Medical Center

Depression in Patients Recovering From a Myocardial

Roy C. Ziegelstein, MD Depression is common among patients recovering from a myocardial infarc- CASE PRESENTATION tion (MI). Approximately 1 in 6 patients with MI experience major depres- A 56-year-old man was admitted to the sion and at least twice as many as that have significant symptoms of de- hospital with an non–Q-wave pression soon after the event. Post-MI depression is an independent factor myocardial infarction (MI). His course for increased mortality. Although the mechanism responsible for this asso- was complicated by postinfarction an- ciation has not yet been defined, depression is clearly associated with poor gina and 3-vessel coronary dis- compliance with risk-reducing recommendations, with abnormalities in au- ease was found at angiography. He un- tonomic tone that may make patients more susceptible to ventricular ar- derwent uncomplicated coronary rhythmias, and with increased platelet activation. Coronary revasculariza- bypass graft surgery and was dis- charged 5 days later. tion procedures also appear to be used less often in those with post-MI The patient had a history of hyper- depression than in comparable patients without . Ongoing tension and . He had no research will address whether treating depression improves prognosis. Un- other significant past medical or psy- til this question is answered, efforts should focus on enhancing adherence chiatric history. For the last 35 years, to treatment regimens in this group of patients, who are clearly at risk for he worked 14 to 18 hours a day as a noncompliance. programs and increasing levels of so- steel cutter doing “bull work,” which cial support may help improve symptoms and should be recommended to he said was “a constant strain on you.” all patients. Treatment of depression itself should be individualized until safety The patient’s wife says he is a stub- born man who is “bullheaded.” Four- and efficacy are determined for antidepressant therapy in patients who re- teen months before admission, he was cently have had an MI. wading with his wife by the seashore, JAMA. 2001;286:1621-1627 www.jama.com complained of chest , slumped over, and had to be dragged ashore. He now present most of the day almost ev- low mood is common after an MI and was brought by to the hos- ery day since hospital discharge. The pa- may adversely affect a person’s quality pital, and shortly after arrival he left the tient said that he frequently cried now, of life, overall sense of well-being, and hospital against medical advice be- often for no obvious reason. He re- ability to actively participate in recov- cause, “I always gotta’ go. I can’t sit still.” ported feeling very unhappy and be- ery. It was noted that the importance of Three weeks after his recent hospi- coming disgusted with himself when he symptoms of depression after an MI is talization, the patient was seen for rou- reflected on some aspects of his life that dependent primarily on their severity tine follow-up. He reported feeling well now appeared to him as a series of fail- and denied chest discomfort or short- ures. He also experienced guilt about Author Affiliation: Department of Medicine, Divi- sion of , Johns Hopkins Bayview Medical ness of breath. His only complaint was some of the possible effects his lifestyle Center, Johns Hopkins University School of Medi- feeling “very bored.” Although he had had had on his health. He specifically cine, Baltimore, Md. Financial Disclosure: Dr Ziegelstein served as an in- a great deal of social contact with fam- said that his attack was his “fault vestigator for the Sertraline Antidepressant Heart At- ily and friends, he was spending most because I didn’t take care of myself.” The tack Randomized Trial (SADHART). of the day in his . He complained patient did not report any change in ap- Corresponding Author and Reprints: Roy C. Ziegel- stein, MD, Department of Medicine, Division of Car- of being restless and bored because he petite, weight, or sleep habits. He de- diology, Johns Hopkins Bayview Medical Center, 4940 thought he could not do his usual ac- nied thoughts of harming himself or feel- Eastern Ave, Baltimore, MD 21224-2780 (e-mail: [email protected]). tivities. He said that for many years he ing that he would be better off dead. Grand Rounds at The Johns Hopkins Medical Insti- would experience low mood around After these symptoms were elicited, tutions Section Editors: David B. Hellmann, MD, D. William Schlott, MD, Stephen D. Sisson, MD, The Johns holidays or anniversaries of deaths of their importance was discussed with the Hopkins Hospital, Baltimore, Md; David S. Cooper, MD, family members but that this feeling was patient and his wife. They were told that Contributing Editor, JAMA.

©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, October 3, 2001—Vol 286, No. 13 1621 DEPRESSION AFTER MYOCARDIAL INFARCTION and duration. The patient and his wife (7) difficulty concentrating, (8) psycho- with physical illness, including MI.5 were told that if his depressed mood be- motor retardation or agitation, and (9) During the office visit 3 weeks after dis- gan to affect his eating and sleeping hab- suicidal ideation. Major depression is di- charge from the hospital, the patient its or to impair his ability to concen- agnosed when 5 or more of these symp- presented herein scored 14 on the BDI. trate, then specific treatment and referral toms have been present most of the day He had several symptoms of depres- to a mental health provider would be rec- nearly every day for more than 2 weeks. sion, including depressed mood and ommended. He and his wife were told To diagnose major depression, at least low self-esteem. Although there had that if he began to have thoughts of 1 of the symptoms has to be either de- been times in the past when he expe- harming himself or wished he were dead, pressed mood, diminished interest, or di- rienced these symptoms for brief peri- referral and treatment should be imme- minished pleasure.1 Dysthymia is char- ods, they had been persistent since his diate and hospitalization may be indi- acterized by 2 or more of these symptoms MI. The patient’s wife noted that he cated. They were told that they would present for more days than not for at least even cried the previous evening while be called to reassess the status of 2 years. they were watching The Highlander,a his mood disturbance in 1 week. The The Beck Depression Inventory movie “about a man getting his head cut patient was informed that it was (BDI)2 may be used to determine the off.” The patient’s wife said he was sit- possible that his symptoms would im- presence and severity of symptoms of ting quietly and suddenly started cry- prove over the next few weeks but that depression but is not appropriate for ing at the theme song. Indeed, the pa- antidepressant therapy might be indi- making the diagnosis of the syndrome tient cried during the office visit when cated if they did not. of major depression. The BDI is a 21- he explained that the song “Who Wants Participation in a cardiac rehabilita- item self-report questionnaire scored to Live Forever?” is played as the main tion program was strongly recom- from 0 to 3 on each item. The total score character, who is immortal, grieves over mended as being potentially helpful to ranges from 0 to 63, with higher scores the death of the woman he loves. his physical and emotional health. The indicating greater levels of distress. Al- Post-MI depression is common.6 Ma- patient enrolled in a phase 2 cardiac re- though a patient who scores 10 to 20 jor depression is found in about 1 in 6 habilitation program, and his symp- points on the BDI would be consid- patients soon after MI and at least twice toms of depression were monitored by ered to have symptoms of at least mild as many have significant symptoms of de- telephone calls and by the supervisor to moderate depression, such an indi- pression (TABLE 1).3,7-12 Most patients of the rehabilitation program. Both the vidual would probably not meet crite- with major depression soon after MI will patient and his wife noted improve- ria for a diagnosis of the syndrome of remain depressed months afterwards. ment in his symptoms over the next few major depression. Nevertheless, even More than 3 out of 4 individuals who are weeks. without a diagnosis of major depres- found to have major depression in the sion, a patient with a BDI score of 10 hospital soon after MI are still de- DISCUSSION or greater is at increased risk of post-MI pressed 3 months later.7 By contrast, only Depression is manifested by a number of mortality during the first 18 months af- about 1 of every 3 patients with less se- symptoms including (1) depressed ter hospital discharge.3 Other easy-to- vere symptoms soon after MI, like the pa- mood, (2) diminished interest or plea- use rating scales are also available to as- tient presented, are still depressed 3 sure in all, or almost all, activities, (3) low sess depression, including the Hospital months later.7 This does not diminish the self-esteem, (4) sleep disturbance, (5) Anxiety and Depression Scale,4 which potential importance of depressive symp- changes in appetite, (6) loss of energy, may be particularly useful for patients toms after MI because even minor de-

Table 1. of Depression After Myocardial Infarction (MI)* Assessment Patients of Depression % With Symptoms % With Major Study, y No. Age, y† Sex After MI, d of Depression Depression Lane et al,10,13 2000; 288 64 Men and women 2-15 30.9 . . . Lane et al14 2001 Ziegelstein et al,9 2000 204 64 Men and women 3-5 17.2 15.2§ Frasure-Smith et al,3 1995; 222 60 Men and women 5-15 30.6 16.0 Frasure-Smith et al,8 1993 Schleifer et al,7 1989 283 64 Men and women 8-10 45.0‡ 18.0 Ladwig et al,12 1991 560 54 Men 17-21 . . . 14.5 Denollet et al,11 1995 105 54 Men 21-42 46.7 . . . *Ellipses indicate not applicable. †Average age in study. ‡Total with major and minor depression. §Major depression and/or dysthymia.

1622 JAMA, October 3, 2001—Vol 286, No. 13 (Reprinted) ©2001 American Medical Association. All rights reserved. DEPRESSION AFTER MYOCARDIAL INFARCTION pressive symptoms may persist in some Table 2. Possible Mechanisms Linking Depression After Myocardial Infarction (MI) and individuals, and it is unclear whether Increased Mortality resolution of symptoms alters the asso- Possible Mechanism Specific Abnormal Finding References ciation of even minor depressive symp- Lifestyle and behavior Decreased adherence to risk-reducing 9, 21-23 toms with increased mortality. recommendations The patient presented also had fre- Neurocardiogenic Increased susceptibility to ventricular ; 3, 24, 25 quent contact with close family and decreased variability friends, a factor that should be consid- Platelet function Increased platelet activation 26-29 ered when deciding how to approach a Treatment Decreased use of cardiovascular procedures 30 patient recovering from an MI. Post-MI depression resolves more rapidly when significant increase in 6-month mortal- increased cardiac risk. Possible factors patients have good . A re- ity rates among patients who were de- responsible for the association include cent study showed that, among those pa- pressed soon after MI, even after con- poor compliance with risk-reducing rec- tients with depression (BDI Ͼ10) soon trolling for left ventricular dysfunction ommendations, abnormalities in auto- after MI who had high baseline social and previous MI.8 Indeed, the mortality nomic tone that may make depressed pa- support, symptoms of depression im- risk of post-MI depression reported in tients more susceptible to ventricular proved to a greater than expected ex- this study was as great as that associ- , and an association with in- tent when reassessed 1 year later.15 De- ated with other recognized prognostic in- creased platelet activation (TABLE 2). pressed patients who were unmarried, dicators, including left ventricular dys- who lived alone, or who reported no function, mellitus, age, and Possible Mechanisms Linking close relatives or little contact with close ventricular . Welin et al17 Post-MI Depression and friends and relatives experienced more found that the association of depres- Increased Mortality symptoms of depression 1 year after the sion soon after an MI and increased mor- Depression is a for noncom- MI than would have been expected based tality was still present 10 years later. Both pliance in many medical conditions. A on their baseline BDI score. Welin et al17 and Frasure-Smith et al3,8 recent meta-analysis indicates that pa- Thus, the patient presented herein, found that the association of baseline de- tients with depression are 3 times as who is married, lives with someone, and pression and post-MI mortality was pre- likely to be noncompliant with medi- reports frequent contact with close sent even after adjusting for the effects cal treatment regimens than patients friends and relatives would be ex- of other established predictors of mor- without depression.31 Several psycho- pected to have a more rapid improve- tality in a multivariate analysis. logical factors, including depression, ment in his symptoms of depression than On the other hand, some authors have been associated with poor compli- if he were socially isolated. It is impor- have found that the association of de- ance with advice to adopt an tant to note, however, that although high pression and post-MI mortality loses program in patients recovering from an levels of social support predict a more significance when adjusting for other acute coronary event.21 Other studies rapid improvement, patients with predictors of mortality12,18 or for po- have shown that patients with depres- post-MI depression with good social sup- tentially confounding symptoms, like sion have a greater dropout rate from port may still require treatment for de- , that may be common to de- cardiac rehabilitation programs22 and pression. The patient presented experi- pression and heart .19 Two other demonstrate poor adherence to pre- enced fairly rapid improvement in his groups failed to find an association be- scribed therapy for secondary symptoms without antidepressant tween depression and mortality at 4 prevention in the treatment of coro- therapy. One month later, he reported months,13 6 months,20 12 months,10,14 nary artery disease.23 few symptoms of depression and had a or 18 months.20 Some have therefore When interviewed 4 months after dis- BDI score of 3. concluded that it is premature to place charge from a hospitalization for an Although post-MI depression is im- depression alongside other indepen- acute MI, patients with baseline de- portant to recognize because of its effect dent risk factors for mortality after an pression reported that they followed on quality of life, its presence takes on MI without additional studies that rig- recommendations to reduce cardiac risk greater significance given its reported as- orously control for disease severity and less often than patients who were not sociation with increased morbidity and accurately identify and measure poten- depressed.9 Compared with those with mortality. Several groups have now docu- tial confounders of this relation- a BDI score of less than 10, patients with mented an association between depres- .10,13,14 a BDI score of 10 or higher soon after sion soon after an MI and mortality at 4 Although future studies may further MI indicated 4 months later that they months,16 6 months,8 12 months,15 and clarify this relationship, the risk of de- were less often following a low- , 18 months.3 In a study of 222 patients pression reported in many recent stud- exercising regularly, or making at- interviewed approximately 1 week af- ies has led to speculation about pos- tempts to reduce their levels of stress ter an MI, Frasure-Smith et al8 found a sible mechanisms linking depression and or increase their social support. Those

©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, October 3, 2001—Vol 286, No. 13 1623 DEPRESSION AFTER MYOCARDIAL INFARCTION with major depression and/or dysthy- susceptible to cardiac arrhythmia or to demonstrated by increased platelet bind- mia during the initial hospitalization re- myocardial . There appears to ing of the monoclonal antibodies anti– ported lower adherence to these rec- be an interplay between emotional state ligand-induced binding site and the ommendations and also reported taking and ventricular irritability after an MI. monoclonal antibody GA6 and in- as prescribed less often Although the presence of frequent pre- creased plasma concentrations of plate- than those without major depression mature ventricular contractions (PVCs) let factor 4. Antidepressant therapy us- and/or dysthymia. Patients with ma- after an MI is generally considered to ing the selective serotonin reuptake jor depression and/or dysthymia who be a predictor of increased mortality inhibitor (SSRI) paroxetine was associ- had diabetes mellitus indicated that they risk, the 18-month post-MI mortality ated with normalization of these mea- followed a diet for patients with diabe- rates of patients with frequent PVCs sures of platelet activation such that af- tes less often than those without de- (Ն10 per hour) on Holter monitoring ter 6 weeks of drug treatment, these pression. and who were not depressed was simi- values were no different from the base- The potential effect of these findings lar to those without frequent PVCs.3 On line values of control subjects. Another on post-MI mortality should be viewed the other hand, patients with a BDI recent study showed that platelet acti- in the context of a recent review article score of 10 or more and frequent ven- vation, as assessed by in vitro platelet se- that indicated that nonad- tricular ectopy were found to have a cretion in response to , was in- herence has a negative effect on mor- mortality risk that was significantly creased in depressed subjects compared bidity and mortality among patients with greater than the rest of the population with those from age- and sex-matched or at risk for or studied (odds ratio, 29.1; 95% confi- control subjects without depression.27 congestive .32 In the Beta- dence interval, 6.97-122.07; P Ͻ.001). The increased platelet activation in de- Blocker Heart Attack Trial, which stud- The effect of depression on auto- pressed subjects was significantly re- ied the effect of on mortal- nomic tone may also contribute to the duced after 6 weeks of treatment with ity in survivors of acute MI, patients who increased mortality risk in post-MI pa- the SSRI sertraline. These findings may did not adhere well to their prescribed tients. Low (HRV) reflect an abnormality of agonist- treatment regimen (defined as those who indicates abnormally high sympathetic receptor coupling in circulating plate- took Յ75% of their prescribed medica- tone with or without abnormally low lets of patients with depression. Plate- tion) were almost 3 times as likely to die parasympathetic neurological input to lets from depressed subjects demonstrate in the first year of follow-up than those the heart. Low HRV has been estab- a greater increase in intracellular cal- who were considered good adherers.33 lished as an independent predictor of in- cium following agonist stimulation than The finding that depression decreases creased mortality in patients after an do platelets from those without depres- the likelihood that patients will make at- MI.34,35 In a recent study, depressed pa- sion.28 tempts to increase their social support9 tients with coronary disease were found This enhanced response in intracel- is important in light of recent findings to have significantly lower HRV than lular signaling was attenuated in the that social support may buffer the ef- age- and sex-matched patients with coro- platelets isolated from depressed pa- fects of depression on prognosis. In a nary disease who were not depressed.24 tients receiving an SSRI.28 The differ- study of 887 patients who completed Interestingly, treatment of patients with ences in intracellular calcium levels may both the BDI and the Perceived Social coronary artery disease who also had be secondary to changes in agonist- Support Scale approximately 1 week af- moderate to severe depression (BDI receptor interactions on the platelet cell ter an MI, the relationship between de- score Ն20) with cognitive behavioral membrane. As is the case in many types pression (BDI score Ն10) and 1-year car- therapy lowered the average resting heart of cells, receptor stimulation in plate- diac mortality decreased with increasing rate and increased the daytime root mean lets is followed by hydrolysis of mem- levels of social support.15 The 1-year car- square successive difference of normal- brane phospholipids and by increases diac mortality of depressed patients with to-normal intervals, an index of HRV.25 in levels of inositol phosphates that then the lowest levels of perceived social sup- Individuals with depression have also mediate the actions of the stimulus in- port was at least 5 times that of de- been found to have evidence of in- side the cell. One of the major path- pressed patients with the highest levels creased platelet activation. If these ab- ways stimulated in this manner in- of perceived social support. Indeed, normalities are present in patients with volves the second messenger inositol among patients in the highest quartile post-MI depression, then depressed pa- 1,4,5-trisphosphate, which increases in- of perceived social support, the associa- tients recovering from an MI may be tracellular calcium levels by releasing tion between depression and increased more susceptible to recurrent myocar- calcium from intracellular stores. Of mortality was no longer present. dial ischemia than those patients with- note, levels of inositol phosphates in the In addition to this effect of post-MI out depression. In a recent study,26 plate- platelets of drug-free patients with de- depression on behavior, there may be let activation under basal conditions was pression were found to be higher after an effect of depression on biological fac- increased in subjects with major depres- receptor stimulation by either epineph- tors that may make the patient more sion compared with control subjects as rine or thrombin than they were in the

1624 JAMA, October 3, 2001—Vol 286, No. 13 (Reprinted) ©2001 American Medical Association. All rights reserved. DEPRESSION AFTER MYOCARDIAL INFARCTION platelets of age- and sex-matched con- fatal MI in patients with an acute MI cant increases in heart rate and reduc- trol subjects without depression.29 On who are depressed or have low per- tions in HRV. Those patients treated the other hand, when platelet inositol ceived social support.36 The safety and with also experienced sig- phosphate levels were increased by efficacy of antidepressant drug treat- nificantly more adverse cardiac events stimulation with sodium fluoride, ment in patients who have recently ex- than those treated with paroxetine. In which does not bind to a receptor, no perienced an acute MI is still un- a double-blind, placebo-controlled differences between platelets from de- known. The multicenter Sertraline study,44 54 patients with major depres- pressed patients and those from con- Antidepressant Heart Attack Random- sion soon after MI were treated with the trol subjects were found. This sug- ized Trial (SADHART) is examining this SSRI fluoxetine. Patients were random- gests that there is a defect in receptor- question.37 Although this study will pro- ized to receive fluoxetine or placebo for agonist coupling in the platelets of vide important information on the ef- the first 9 weeks of the study and those individuals with depression that may fects of antidepressant drug therapy who wished to continue participating make these platelets more likely to be soon after an acute MI, the existing lit- and did not have serious adverse events activated. erature suggests that the SSRIs are safer were given fluoxetine or placebo for an In addition to the abnormalities that in patients with heart disease than the additional 16 weeks. Fluoxetine re- may be found in patients with depres- tricyclic antidepressants (TCAs). sulted in improvement in symptoms of sion that place them at greater risk, it The TCAs have several cardiovascu- depression, particularly in those with also appears that patients with post-MI lar effects that are undesirable in pa- mild depression at baseline, without depression do not undergo potentially tients with heart disease, including their producing changes in cardiac func- beneficial treatments as often as indi- tendency to increase resting heart rate, tion as assessed by electrocardio- viduals without depression. A recent produce orthostatic , and graphy and . In an- study examined the association be- alter intracardiac conduction and the other study,37 sertraline was used to tween the presence of mental disor- susceptibility to ventricular arrhyth- treat 26 patients with major depres- ders and the use of cardiac catheteriza- mias.38 The effects of TCAs on cardiac sion soon after an MI. During treat- tion and coronary revascularization rhythm and conduction are similar to ment with sertraline, depression im- procedures in a national cohort of the type 1 antiarrhythmics quinidine proved and no significant changes in 113653 patients 65 years or older with and moricizine. There is therefore con- heart rate, , cardiac con- an acute MI.30 This study found that pa- cern that the TCAs may increase mor- duction, left ventricular ejection frac- tients with mental disorders were less tality in patients with heart disease, as tion, or ventricular ectopy were ob- likely to undergo percutaneous coro- did moricizine in the Cardiac Arrhyth- served. Taken together, these studies nary intervention or coronary artery by- mia Suppression Trial II (CAST II)39 and indicate that compared with TCAs, the pass graft (CABG) surgery than those quinidine in other studies.40,41 It has SSRIs are the preferred class of antide- without mental disorders. Among pa- therefore been suggested that the TCAs pressant medications in patients with tients with major affective disorder, should be avoided in patients with ven- post-MI depression. which includes those with major de- tricular arrhythmias or ischemic heart Although antidepressant therapy is of- pression, the unadjusted use of percu- disease.42 ten required to treat post-MI depres- taneous coronary intervention was 9.2% By contrast, the early experience in sion sufficiently and prevent relapse, car- and that of CABG surgery was 7.9% patients with SSRIs and ischemic heart diac rehabilitation is also an important compared with 16.8% and 12.6% in disease, some of whom were treated part of the care of patients after an MI, those with no mental disorder. soon after MI, suggests that these agents whether they are depressed or not. Many are safe and effective. In a study of pa- aspects of psychological health are im- Treatment Strategies tients with ischemic heart disease and proved by cardiac rehabilitation, includ- for Depression in Patients major depression, the effects of 6 weeks ing symptoms of depression. In a study Recovering From MI of antidepressant therapy with the SSRI of 338 patients, 75 of whom had re- Whether treating depression can re- paroxetine were compared with the ef- cently experienced an MI, the preva- duce the morbidity and mortality risk fects of the TCA nortriptyline for the lence of depressive symptoms was sig- among patients recovering from an MI same duration of treatment.43 Of note, nificantly reduced after a 12-week phase is unknown. To address this question, patients were specifically excluded from 2 cardiac rehabilitation program among the National Heart, , and Blood In- this study if they had sustained an MI the 69 patients (20%) who were de- stitute has sponsored the Enhancing Re- within the past 3 months. Paroxetine pressed at the time of referral shortly af- covery in Coronary Heart Disease Pa- and nortriptyline were both associ- ter a major cardiac event.45 In this study, tients (ENRICHD) study, which is a ated with improvements in symptoms depression was defined as a score of multicenter, randomized clinical trial of depression. Although the SSRI did greater than 6 on the Symptom Ques- investigating the effects of a psychoso- not affect resting heart rate or HRV, pa- tionnaire.46 On this questionnaire, a cial intervention on mortality and non- tients treated with the TCA had signifi- score of 7 to 9 is considered to repre-

©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, October 3, 2001—Vol 286, No. 13 1625 DEPRESSION AFTER MYOCARDIAL INFARCTION sent moderate depressive symptoms and sures, many patients do not improve The benefits of strategies treating de- a score of 10 or greater is considered se- without antidepressant therapy. It also pression itself or of strategies target- vere. Among the depressed patients, the must be remembered that even mild ing the effects of depression on plate- score on the Symptom Questionnaire symptoms of depression are associ- let activation or ventricular arrhythmia improved from an average of 11.2 at ated with increased post-MI morbid- will be clearly defined. However, if the baseline to 5.1 after rehabilitation ity and mortality.3 This patient’s rapid compliance problems remain unad- (P=.001). Although there was no con- improvement persisted, and he is still dressed, the benefits of this research trol group in this study to permit the ef- doing well now 10 months later. may not reach the very population they fects of the rehabilitation program to be are intended for and the relationship of specifically determined, the rehabilita- The Importance of Enhancing post-MI depression to increased mor- tion program had other positive effects Compliance tality will not have been affected. in the group with depression that were There are 2 critical, but as yet unan- at least as significant as the effects noted swered, questions raised by this dis- REFERENCES in patients without depression. Before cussion: (1) what is the mechanism 1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edi- entry into the rehabilitation program, de- linking depression and increased tion. Washington, DC: American Psychiatric Associa- pressed patients had significantly lower post-MI mortality? and (2) will treat- tion; 1994:317-391. exercise capacity than those patients ing depression affect this association? 2. Beck AT, Ward CH, Mendelson M, et al. An in- ventory for measuring depression. Arch Gen Psychia- without depression (4.9 vs 5.8 meta- Unfortunately, the answers to these try. 1961;4:561-571. bolic equivalents [METs], P=.01). Al- questions will not likely be known for 3. Frasure-Smith N, Lesperance F, Talajic M. Depres- sion and 18-month prognosis after myocardial infarc- though individuals without depression at least several years. While we are tion. Circulation. 1995;91:999-1005. experienced a 36% increase in exercise awaiting the results of the clinical tri- 4. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scand. 1983;67: capacity after the 12-week program als that address these questions, it is im- 361-370. (from 5.8 to 7.9 METs), individuals with portant to use what we know to opti- 5. Johnston M, Pollard B, Hennessey P. Construct vali- depression experienced a 51% increase mally manage high-risk patients with dation of the hospital anxiety and depression scale with clinical populations. J Psychosom Res. 2000;48:579- (from 4.9 to 7.4 METs, P=.001 for each). post-MI depression. It is very clear from 584. In another study,47 268 consecutive pa- the evidence that depression should be 6. Ziegelstein RC. Depression after myocardial infarc- tion. Cardiol Rev. 2001;9:45-51. tients older than 65 years with heart dis- considered a risk factor for noncom- 7. Schleifer SJ, Macari-Hinson MM, Coyle DA, et al. ease were prospectively studied after en- pliance in patients with heart disease. The nature and course of depression following myo- cardial infarction. Arch Intern Med. 1989;149:1785- rollment in a cardiac rehabilitation The American Heart Association has 1789. program. Of these, 64 were post-MI and emphasized the importance of improv- 8. Frasure-Smith N, Lesperance F, Talajic M. Depres- sion following myocardial infarction. JAMA. 1993; 48 (18%) were depressed. Among the ing compliance to achieve better clini- 270:1819-1825. 48 depressed patients, the score on the cal outcomes. In her address to the as- 9. Ziegelstein RC, Fauerbach JA, Stevens SS, et al. Pa- Symptom Questionnaire improved from sociation’s 70th Scientific Session, past tients with depression are less likely to follow recom- mendations to reduce cardiac risk during recovery from an average of 10.4 at baseline to 4.5 president Martha Hill, PhD, RN, spoke a myocardial infarction. Arch Intern Med. 2000;160: after rehabilitation (P=.001). of the gaps between the potential ben- 1818-1823. 10. Lane D, Carroll D, Ring C, et al. Do depression efits to health suggested by the results and anxiety predict recurrent coronary events 12 CASE RESOLUTION of clinical trials and the actual ben- months after myocardial infarction? QJM. 2000;93: 739-744. The patient presented herein was efits achieved in practice. Hill empha- 11. Denollet J, Sys SU, Brutsaert DL. Personality and strongly encouraged to enter a phase 2 sized that these gaps could be nar- mortality after myocardial infarction. Psychosom Med. cardiac rehabilitation program and par- rowed by a between the social 1995;57:582-591. 12. Ladwig KH, Kieser M, Konig J, et al. Affective dis- ticipated for several weeks before he re- and behavioral sciences and the bio- orders and survival after acute myocardial infarction: turned to work. The patient’s physical logical sciences.49 The studies indicat- results from the post-infarction late potential study. Eur Heart J. 1991;12:959-964. and emotional health were closely as- ing that depression adversely affects the 13. Lane D, Carroll D, Ring C, et al. Effects of de- sessed and followed up in the cardiac compliance with therapy of patients pression and anxiety on mortality and quality-of-life 9,21-23,31 4 months after myocardial infarction. J Psychosom Res. rehabilitation program, and his progress with coronary heart disease sug- 2000;49:229-238. was also monitored on a weekly basis gest that compliance-enhancing initia- 14. Lane D, Carroll D, Ring C, et al. Mortality and qual- ity of life 12 months after myocardial infarction: ef- by telephone contact. Since his symp- tives targeted to this group may, if suc- fects of depression and anxiety. Psychosom Med. 2001; toms rapidly improved, antidepres- cessful, favorably affect the relationship 63:221-230. sant therapy was not recommended. between post-MI depression and poor 15. Frasure-Smith N, Lespe´ rance F, Gravel G, et al. Social support, depression, and mortality during the Such close monitoring and observa- outcomes. first year after myocardial infarction. Circulation. 2000; tion are not always possible, particu- Several years from now, many of the 101:1919-1924. 16. Bush DE, Ziegelstein RC, Tayback M, et al. Even larly since many patients do not enter questions about post-MI depression will minimal symptoms of depression increase mortality risk cardiac rehabilitation programs and de- be answered. The safety and efficacy of after acute myocardial infarction. Am J Cardiol. 2001; 88:337-341. pressed patients have a particularly high antidepressant drug therapy in pa- 17. Welin C, Lappas G, Wilhelmsen L. Independent dropout rate.22 Even with these mea- tients early after an MI will be known. importance of psychosocial factors for prognosis

1626 JAMA, October 3, 2001—Vol 286, No. 13 (Reprinted) ©2001 American Medical Association. All rights reserved. DEPRESSION AFTER MYOCARDIAL INFARCTION after myocardial infarction. J Intern Med. 2000;247: let cytosolic calcium responses to serotonin in de- and without heart disease. J Clin Psychiatry. 1989; 629-639. pressed patients and controls: relationship to symp- 50;(suppl):1-18. 18. Kaufmann MW, Fitzgibbons JP, Sussman EJ, et al. tomatology and medication. Biol Psychiatry. 1998; 39. Cardiac Arrhythmia Suppression Trial II Investi- Relation between myocardial infarction, depression, hos- 43:327-334. gators. Effect of the moricizine tility, and death. Am Heart J. 1999;138:549-554. 29. Karege F, Bovier P, Rudolph W, et al. Platelet phos- on survival after myocardial infarction. N Engl J Med. 19. Irvine J, Basinski A, Baker B, et al. Depression and phoinositide signaling system: an overstimulated 1992;327:227-233. risk of sudden cardiac death after acute myocardial pathway in depression. Biol Psychiatry. 1996;39:697- 40. Morganroth J, Goin JE. Quinidine-related mor- infarction: testing for the confounding effects of fa- 702. tality in the short-to-medium-term treatment of ven- tigue. Psychosom Med. 1999;61:729-737. 30. Druss BG, Bradford DW, Rosenheck RA, et al. tricular arrhythmias: a meta-analysis. Circulation 1991; 20. Mayou RA, Gill D, Thompson DR, et al. Depres- Mental disorders and use of cardiovascular proce- 84:1977-1983. sion and anxiety as predictors of outcome after myo- dures after myocardial infarction. JAMA. 2000;283: 41. Coplen SE, Antman EM, Berlin JA, et al. Efficacy and cardial infarction. Psychosom Med. 2000;62:212-219. 506-511. safety of quinidine therapy for maintenance of sinus 21. Guiry E, Conroy RM, Hickey N, et al. Psychologi- 31. DiMatteo MR, Lepper HS, Croghan TW. Depres- rhythm after cardioversion: a meta-analysis of random- cal response to an acute coronary event and its effect sion is a risk factor for noncompliance with medical ized control trials. Circulation. 1990;82:1106-1116. on subsequent rehabilitation and lifestyle change. Clin treatment: meta-analysis of the effects of anxiety and 42. Glassman AH, Roose SP, Bigger JT Jr. The safety Cardiol. 1987;10:256-260. depression on patient adherence. Arch Intern Med. of tricyclic antidepressants in cardiac patients: risk- 22. Blumenthal JA, Williams RS, Wallace AG, et al. 2000;160:2101-2107. benefit reconsidered. JAMA. 1993;269:2673-2675. Physiological and psychological variables predict com- 32. McDermott MM, Schmitt B, Wallner E. Impact of 43. Roose SP, Laghrissi-Thode F, Kennedy JS, et al. pliance to prescribed exercise therapy in patients re- medication nonadherence on coronary heart disease Comparison of paroxetine and nortriptyline in de- covering from myocardial infarction. Psychosom Med. outcomes. Arch Intern Med. 1997;157:1921-1929. pressed patients with ischemic heart disease. JAMA. 1982;44:519-527. 33. Horwitz RI, Viscoli CM, Berkman L, et al. Treat- 1998;279:287-291. 23. Carney RM, Freedland KE, Eisen SA, et al. Major ment adherence and risk of death after a myocardial 44. Strik JJ, Honig A, Lousberg R, et al. Efficacy and depression and medication adherence in elderly pa- infarction. Lancet. 1990;336:542-545. safety of fluoxetine in the treatment of patients with tients with coronary artery disease. Health Psychol. 34. Kleiger RE, Miller JP, Bigger JT, Jr, et al. De- major depression after first myocardial infarction: find- 1995;14:88-90. creased heart rate variability and its association with ings from a double-blind, placebo-controlled trial. Psy- 24. Stein PK, Carney RM, Freedland KE, et al. Severe mortality after myocardial infarction. Am J Cardiol. chosom Med. 2000;62:783-789. depression is associated with markedly reduced heart 1987;59:256-262. 45. Milani RV, Lavie CJ, Cassidy MM. Effects of car- rate variability in patients with stable coronary heart 35. Bigger JT Jr, Fleiss JL, Rolnitzky LM, et al. Fre- diac rehabilitation and exercise training programs on disease. J Psychosom Res. 2000;48:493-500. quency domain measures of heart period variability depression in patients after major coronary events. Am 25. Carney RM, Freedland KE, Stein PK, et al. Change to assess risk late after myocardial infarction. J Am Coll Heart J. 1996;132:726-732. in heart rate and heart rate variability during treat- Cardiol. 1993;21:729-736. 46. Kellner R. A symptom questionnaire. J Clin Psy- ment for depression in patients with coronary heart 36. The ENRICHD Investigators. Enhancing recov- chiatry. 1987;48:268-274. disease. Psychosom Med. 2000;62:639-647. ery in coronary heart disease patients (ENRICHD): study 47. Milani RV, Lavie CJ. Prevalence and effects of car- 26. Musselman DL, Marzec UM, Manatunga A, et al. design and methods. Am Heart J. 2000;139:1-9. diac rehabilitation on depression in the elderly with Platelet reactivity in depressed patients treated with 37. Shapiro PA, Lesperance F, Frasure-Smith N, et al. coronary heart disease. Am J Cardiol. 1998;81:1233- paroxetine: preliminary findings. Arch Gen Psychia- An open-label preliminary trial of sertraline for treat- 1236. try. 2000;57:875-882. ment of major depression after acute myocardial in- 48. Miller NH, Hill M, Kottke T, et al. The multilevel 27. Markovitz JH, Shuster JL, Chitwood WS, et al. farction (the SADHAT Trial): Sertraline Anti- compliance challenge: recommendations for a call to Platelet activation in depression and effects of sertra- Depressant Heart Attack Trial. Am Heart J. 1999;137: action, a statement for healthcare professionals. Cir- line treatment: an open-label study. Am J Psychiatry. 1100-1106. culation. 1997;95:1085-1090. 2000;157:1006-1008. 38. Roose SP, Glassman AH. Cardiovascular effects 49. Fox R. Compliance, adherence, concordance. Cir- 28. Delisi SM, Konopka LM, O’Connor FL, et al. Plate- of tricyclic antidepressants in depressed patients with culation. 1998;97:127.

©2001 American Medical Association. All rights reserved. (Reprinted) JAMA, October 3, 2001—Vol 286, No. 13 1627

Top View