- Email: [email protected]
Circadian variation and triggering of acute coronary events
Supplement to
AHJ American Heart Journal
Copyright © 1999 by Mosby, Inc.
April 1999 Volume 137, Number 4, Part 2
Circadian variation and triggering of acute coronary events James E. Muller, MD Lexington, Ky
The recognition that the onset of cardiovascular events follows a circadian periodicity and is frequently triggered by physical or mental stresses has created new possibilities for disease prevention. Morning peaks in occurrence are now well-documented for acute myocardial infarction, sudden cardiac death, transient myocardial ischemia, and ischemic stroke. The morning increase in events begins after subjects assume an upright posture and start the day’s activities, during a time of sympathetic nervous system activation. Additional triggers of onset include heavy physical exertion, sexual activity, and anger, the risks of which have been quantified in the Determinants of Myocardial Infarction Onset Study. A general hypothesis of the triggering of coronary thrombosis has been proposed. The process begins with the development of a vulnerable atherosclerotic plaque, which may become disrupted by internal forces or by external hemodynamic or vasoconstrictive changes. Once disrupted, the plaque becomes a thrombogenic focus. An occlusive thrombus is more likely to form if other factors come into play to increase coagulability and vasoconstriction. From a clinical standpoint these findings provide theoretical support for the use of longacting agents to provide adequate anti-ischemic protection during the higher risk morning hours in patients already taking antiischemic medications. From a research standpoint this new information on triggering provides clues to a mechanism of onset that might lead to more effective preventive therapy. Because most deaths from coronary artery disease occur before any type of acute therapy can be given, further efforts to explore this new field are warranted. (Am Heart J 1999;137:S1-S8)
Despite new insights into long-term cardiovascular risk factors and striking advances in the acute in-hospital treatment of patients with myocardial infarction (MI), the coronary artery disease (CAD) mortality rate remains persistently high. Indeed, each year in the United States alone, CAD accounts for 500,000 deaths, half of which occur outside the hospital and spoil suddenly.1 Moreover, 600,000 acute MIs occur each year in individuals who either had no symptoms or had only mild symptoms.2 These grim statistics underscore the need to broaden our concept of risk factor modification to encompass acute and chronic risk factors for acute MI and sudden death.1 In the past decade epidemiologic studies have objecFrom the Gill Heart Institute. Reprint requests: James E. Muller, MD, Gill Heart Institute, 800 Rose St, University of Kentucky Medical Center, Lexington, KY 40502. Copyright © 1999 by Mosby, Inc. 0002-8703/99/$8.00 + 0 4/0/96909
tively confirmed the circadian variation in the onset of cardiac events and have established the role of physical and mental stressors as triggers of onset. Although the possibility of a positive reporting bias in the medical literature cannot be completely excluded, the overwhelming majority of studies find that most types of cardiovascular events show a morning increase. The identification of these triggers and the physiologic or pathophysiologic changes that they induce in the hours and minutes before the onset of a cardiac event has opened up broad new horizons for prevention. The major challenge for the coming decade will be to devise therapies that delay the transition from chronic CAD to an acute event. It may prove possible to accomplish this delay by interrupting the links between such triggers as awakening, physical exertion, and mental stress and such acute risk factors as an increase in blood pressure (BP), an increase in coagulability, or vasoconstriction. After a review of the evidence that supports circadian
American Heart Journal April 1999
S2 Muller
Figure 1
periodicity and triggering stresses in the onset of cardiac events, this article will discuss the ramifications of these findings for patient management.
Circadian variation and cardiovascular events Myocardial infarction
image not available
Bar graphs of time of day of onset of myocardial infarction, sudden cardiac death, thrombotic stroke, and transient myocardial ischemia in 4 different groups of patients. Reproduced with permission from Muller JE, Tofler GH, Stone PH. Circadian variation and triggers of onset of acute cardiovascular disease. Circulation 1989;79:733-43.
Reports in the world literature going back as far as 1960 have described a circadian periodicity in the onset of MI pain and have documented a peak incidence in the hours between 6 AM and noon. However, because these early studies relied on subjective descriptions of pain, they were unable to rule out the possibility that pain experienced during sleep might have gone unnoticed. Not until 1985 was an attempt made to quantify the temporal distribution of acute MI onset on the basis of objective evidence. The records of nearly 1000 patients enrolled in the Multicenter Investigation of Limitation of Infarct Size trial were analyzed to determine the time of infarct onset according to serial measurements of plasma creatine kinase MB activity.2 Consistent with previous reports, the Multicenter Investigation of Limitation of Infarct Size investigators observed that the frequency of pain onset increased significantly between 6 AM and noon and that the incidence of MI during this 6-hour period was 1.28 times greater than during the other three 6-hour intervals of the day (P < .01).2 When we analyzed the distribution of MI onset according to enzyme measurements, we noted a peak between the hours of 5 AM and 2 PM.2 These data demonstrated that the incidence of enzymatically confirmed infarct onset between 6 AM and noon was 1.43 times higher than during the other three 6-hour time intervals. A subsequent study has shown that the increase in the incidence of MI was not a consequence of time per se but rather could be narrowed down to the first 4 hours after awakening.3 The characteristic circadian rhythm was as prominent in patients who did not smoke or drink coffee as in those who did, indicating that these habits were unlikely to have accounted for the morning peak.2 Circadian periodicity was not observed in patients who were under treatment with β-blockers at the time of their acute MI.2 The emergence of objective evidence linking the onset of infarction to the hours after awakening raised a number of interesting questions about the biologic underpinnings of this relationship. Particular attention has been focused on the role of certain endogenous
American Heart Journal Volume 137, Number 4, Part 2
daily rhythms such as the morning trough in body temperature, the rise in plasma catecholamine levels and heart rate (HR), the increase in arterial pressure stimulated by increased sympathetic activity, and the trough in fibrinolytic activity coupled with an increase in platelet aggregation.2 Most significant was the realization that identification and modification of the processes that drive the circadian rhythm of MI onset could potentially represent a strategy for delaying or preventing such catastrophic events.2
Muller S3
Figure 2
Sudden death Two large-scale analyses, the Massachusetts Death Certificate Study4 and the Framingham Heart Study,5 have established a circadian variation in the incidence of sudden cardiac death that parallels the pattern reported for acute MI. Examination of the records of more than 2000 subjects who died suddenly in Massachusetts in 1983 has identified a primary peak between 7 AM and 11 AM.4 This analysis revealed that the frequency of sudden cardiac death dropped between the hours of 11 AM and 5 PM and reached its nadir during sleep between the hours of 11 PM and 6 AM.4 These findings were bolstered by data from more than 5000 subjects in the original Framingham cohort, of whom 429 had sudden cardiac death. According to the Framingham analysis, the hourly risk of sudden death was at least 70% greater between 7 AM and 9 AM than during the rest of the day. Analogous to the situation with acute MI onset, significantly more sudden deaths occurred between 6 AM and noon than during the other three 6-hour intervals of the day.5 The morning peak documented in these 2 studies is compatible with 2 prevailing theories of the cause of sudden death.4,5 The morning increase in sympathetic activity could increase electrical instability, leading to fatal arrhythmias; alternatively, the rise in BP on awakening could increase the odds of atherosclerotic plaque rupture, the pathophysiologic significance of which would be compounded by morning increases in coronary artery tone and platelet aggregability.4,5
image not available
Stroke In a study of patients with stroke, neurologists at 4 academic hospital centers collected information on more than 1200 patients with ischemic strokes during a 3-year period.6 Their analysis demonstrated that in awake patients, more strokes occurred between 10 AM and noon than during any other 2-hour interval. The incidence of ischemic strokes was at its maximum from 8 AM to noon, declining during the afternoon
Bar graphs of variation during 24-hour period of 5 physiologic processes possibly contributing to increase in morning frequency of disease onset. Reproduced with permission from Muller JE, Tofler GH, Stone PH. Circadian variation and triggers of onset of acute cardiovascular disease. Circulation 1989;79:733-43.
American Heart Journal April 1999
S4 Muller
Figure 3
image not available
Illustration of hypothetical method by which daily activities may trigger coronary thrombosis. Three triggering mechanisms are presented: (1) physical or mental stress producing hemodynamic changes leading to plaque rupture, (2) activities causing increase in coagulability, and (3) stimuli leading to vasoconstriction. Reproduced with permission of the American College of Cardiology from Muller JE, Abela GS, Nesto RW, Tofler GH. Triggers, acute risk factors and vulnerable plaques: the lexicon of a new frontier. J Am Coll Cardiol 1994;23:809-13.
and early evening.6 The circadian pattern was not altered by therapy with aspirin, dipyridamole, or warfarin, by a history of hypertension, or by BP level on admission.6 This study demonstrated that, contrary to traditional belief, the onset of ischemic stroke peaks when BP is at its highest for the day. The authors suggested that the preponderance of morning strokes may have also been related to morning peaks in plasma viscosity, hematocrit, and platelet aggregability.6
Transient ischemia Studies have consistently shown that the incidence of transient episodes of myocardial ischemia with STsegment depression is also at a high point between the hours of 6 AM and noon.3 Adjustment for the time of awakening indicates that, as with acute MI, the peak incidence of transient myocardial ischemia occurs in the first 4 hours after awakening.3
Daily activities as triggers Although there is a confluence of neurohormonal, hemostatic, and hemodynamic changes in the morning
that could precipitate a cardiovascular event in susceptible individuals, similar processes occur throughout the day in response to a variety of seemingly innocuous activities. Indeed, the morning peak in the incidence represents only 1 special circumstance in which triggering mechanisms come into play. The relationship between acute MI and triggering activities was recognized as early as 1910 by Obraztsov and Strazhesko in Russia, who colorfully described the onset of acute MI in the context of stressful activities: “Direct events often precipitated the disease; the infarct began in one case on climbing a high staircase, in another during an unpleasant conversation, and in a third during emotional distress associated with a heated card game.”3 After nearly a century of controversy, recent new insights have highlighted the role of daily activities in precipitating many cases of acute MI and sudden death.
A new epidemiologic method Although descriptive studies have indicated that the onset of acute MI is preceded by heavy physical exertion in approximately 5% of patients7 and by emotional
American Heart Journal Volume 137, Number 4, Part 2
stress such as anger in 4% to 18% of cases,8 these studies have been hampered by the lack of control data. To more rigorously clarify the role of triggering factors in the development of acute MI, we developed and used a novel case-crossover method in the Determinants of Myocardial Infarction Onset Study. The method offered the unique advantage of having each patient serve as his or her own control.7 Trained examiners interviewed patients an average of 4 days after acute MI to ascertain the patients’ exposure to triggering factors during a brief “hazard period” before infarction. Exposure during the hazard period was then compared with the patient’s exposure to the possible trigger during a comparable period the day before infarction, and with his or her usual frequency of exposure during the year preceding infarction.
Muller S5
Figure 4
Physical exertion and acute MI By means of interviews with more than 1200 survivors of acute MI in 45 hospitals across the country, we first quantified the relative risk of infarction after heavy physical exertion and assessed whether this risk could be modified by regular physical activity.7 Our findings revealed that 4.4% of patients had a trigger of heavy physical exertion in the hour before their infarction. The most frequently implicated activities were mixed isotonic and isometric exercise such as gardening or splitting wood (52%) followed by isotonic exercise such as jogging or racquet sports (30%) and isometric exercise such as lifting or pushing (18%).7 The risk of having an acute MI was nearly 6 times higher in the hour after heavy exertion than during periods of light or no exertion. Perhaps most important, the risk of infarction after heavy exertion was more than 100 times greater for patients who were habitually sedentary but was only 2.4 times greater for those who regularly exercised. These findings suggested that a regular program of physical activity not only lowers the baseline risks of a cardiac event but also decreases the relative risk (RR) of MI after heavy exertion. It should also be stressed that the Onset Study demonstrated no excess RR for patients who had a history of CAD before their infarction, although such patients would be expected to have a greater baseline risk.
Anger and acute MI We then applied the same epidemiologic technique to clarify the role of anger in triggering acute MI in more than 1600 patients.8 With a specially designed self-report onset anger scale of 1 (calm) to 7 (enraged), we found that 8% of patients reported episodes of
Relative risk of acute myocardial infarction after identifiable triggering activities. Dotted line represents baseline risk.7-9,12
anger at level 5 (very angry) and above during the 24 hours before acute MI and that 2.4% reported such episodes in the 2-hour period preceding the infarction.8 Comparison with the frequency of angry episodes during the day before and the year preceding infarction revealed that the odds of an acute MI were more than doubled in the 2 hours after an episode of anger.8 For regular users of aspirin, however, the RR was only 1.4 compared with 2.9 among nonusers.8 Thus, aspirin appeared to reduce the excess risk of MI posed by discrete episodes of anger.8
Other cardiac events A considerable number of descriptive reports have also linked sudden cardiac death to physical or mental stress.9 However, studies with case-control and casecrossover designs will be necessary to quantify the RRs of these stresses. In other investigations, it has been shown that more than half of episodes of transient myocardial ischemia are preceded by possible triggering activities such as increased physical or mental stress.3
Mechanisms of triggering The circadian rhythm of onset of acute MI and sudden death suggests a hypothetical sequence of events that may culminate in a cardiac event. The common shared factor in most cases of MI and sudden cardiac
American Heart Journal April 1999
S6 Muller
death is coronary artery thrombosis. Occlusive thrombosis is the culprit in virtually all patients who have Q-wave infarctions and in at least one half of patients who have a sudden cardiac death. Moreover, even in victims of sudden death who do not have an occlusive thrombosis, nonocclusive thrombosis is often present.3
namic changes all affect the activation and release of platelets, leading to a thrombogenic tendency. Intensified sympathetic activity could give rise to paradoxical coronary vasoconstriction, which has been demonstrated in diseased arterial walls and may also predispose to complete occlusion.9
Vulnerable plaque
Triggering factors
Pathologic and angiographic studies have indicated that coronary thrombus most frequently forms on the surface of a ruptured atherosclerotic plaque.3 Although the exact mechanisms by which plaque becomes vulnerable to rupture are not fully known, several suggestive clues have been identified. Plaque disruption frequently begins in lesions previously causing a <50% reduction in lumen diameter. Disrupted plaques have been found to contain higher concentrations of lipids and macrophage foam cells than have nondisrupted plaques.10 This accumulation of lipids forms the core of the plaque, which is covered by a thin fibrous cap. The integrity of the fibrous cap may be a key determinant of the transformation from stable to vulnerable plaque. The degradation of collagen and elastin in the cap by collagenase, gelatinases, stromelysin, and other protease enzymes may leave the plaque more susceptible to rupture.11
Although these physiologic changes occur in tandem in the morning, they may occur at random times throughout the day in response to any of a variety of physical and mental triggers.3 We have hypothesized that the first precipitating step in the chain of events leading to coronary thrombosis is an acute physical or mental stress that triggers a hemodynamic change sufficient to rupture a vulnerable atherosclerotic plaque (Fig 3).10 Whether the ruptured plaque leads to total or partial occlusion of the arterial lumen or to no symptoms at all will depend not only on the severity of the rupture but also on the contribution of external factors that may heighten coagulability, thereby promoting thrombus growth, or that may increase vasoconstriction, thereby occluding an already narrowed artery. A hypothetical model can also be created to explain how external triggers may lead to sudden cardiac death.9 An external trigger, whether the act of awakening, heavy exertion, or mental stress, results in an increase in sympathetic activity. Hemodynamic forces may then cause the rupture of vulnerable plaque. During the critical phase of plaque rupture and thrombus formation, platelet characteristics and arterial wall function appear to be involved in the pathogenesis of sudden death.9 Exposed collagen in the fibrous cap of the plaque may lead directly to the development of an occlusive thrombus and acute MI. Alternatively, the thrombus may grow gradually, depending on blood coagulability and vasoconstriction. This growth leads to reduced flow, microemboli, and ischemia or even small areas of necrosis, both of which have been linked to a decreased threshold for ventricular fibrillation. Damage caused by necrosis, fibrosis, and hypertrophy may increase the hazard of acute cardiac electric instability. Episodes of ischemia in the context of underlying CAD have been associated with an increased risk of sudden death.9 Sympathetic activation may also lead directly to sudden death by causing a primary arrhythmia. It is also possible that plaques rupture because of processes intrinsic to the plaque, and the trigger simply adds thrombotic forces to a previously ruptured plaque.
Acute hemodynamic, hemostatic, and vasoconstrictive risk factors A number of physiologic processes, many of which are amplified during the morning hours, may act in concert as acute risk factors for a cardiac event. The morning peaks in acute MI and sudden death may reflect the synchronization of the population for these risk factors in the hours after awakening. The most important of these processes are the increases in systemic BP and HR, the decrease in coronary blood flow, increases in blood viscosity and platelet aggregability, decrease in tissue-type plasminogen activator activity, and increases in plasma cortisol and sympathetic activity (Fig 2).3 The morning rise in arterial pressure and HR could precipitate rupture of a vulnerable atherosclerotic plaque. The decrease in flow caused by a fixed stenosis could be exacerbated by the morning increase in coronary arterial tone. Increases in blood viscosity and platelet aggregability, coupled with a drop in endogenous tissue-type plasminogen activator activity, could lead to a hypercoagulable state that sets the stage for thrombus growth. Acute arterial wall injury, endothelial cell dysfunction, and hemody-
American Heart Journal Volume 137, Number 4, Part 2
Clinical ramifications In practical terms the key question is how to translate our newly enhanced understanding of triggering mechanisms into clinical recommendations. It should be emphasized that identification of a possible triggering activity does not automatically mean that the patient must be advised to avoid this activity.1 It is clear that awakening in the morning and unanticipated physical and psychologic stresses are unavoidable aspects of daily living. In addition, it is important to recall that even a dramatic increase in RR may translate into a very small absolute risk if the baseline risk is low to begin with.7 For example, the odds that a healthy 50-year-old nonsmoking man will have an acute MI during any given 1-hour period are 1 in a million. If he regularly engages in exercise, the odds of his having a heart attack during an hour of exercise are more than doubled, but this means an absolute risk of only 1 in 400,000. If the same man led a more sedentary life, his RR of MI during an hour of exercise would be 4100 times greater than during rest, but his absolute risk would still be only 1 in 10,000. Although the relative risks of morning exertion versus evening exertion have not yet been clarified, the absolute risk at either time is so low, and the benefits of exercise so well established, that avoidance of morning exertion is not justified. Of great value to MI survivors and those with cardiac disease will be the recent findings of the Determinants of Myocardial Infarction Onset Study Investigators regarding the RRs of nonfatal MI triggered by sexual activity, both among the general population and in patients with previous coronary heart disease.12 A casecrossover analysis of 858 sexually active patients indicates that although sexual activity can trigger the onset of MI, the RR of acute MI related to sexual intercourse is quite low, and the absolute risk is extremely low (1 in 1,000,000 for a healthy individual). Moreover, the RR is not increased in patients with a history of cardiac disease. Thus, physicians can actually use data on triggering mechanisms to reassure their patients that the odds of having a heart attack during, for example, sexual activity, are very slim indeed (Fig 4).7-9,12 Particularly noteworthy is the Onset Study’s observation that MI induced by sexual activity was most common among sedentary individuals and least common in active individuals.13 It has been demonstrated that at any specific level of submaximal effort associated with exertion such as sexual activity, HR, BP, and level of sympathetic activation are lower in physically active
Muller S7
persons than in sedentary ones.14 Thus the Onset Study results strongly support the premise that regular exercise can prevent triggering of MI.12 Data from studies of triggering factors can be relevant to the treatment of patients who repeatedly and regularly have a particular avoidable trigger. Repeated regular exposure to a trigger such as anger may markedly increase the cumulative risk and may therefore warrant intervention. For example, stress management instruction may be helpful for the patient who has anger repeatedly throughout the day.1 The understanding that anger is an acute risk factor for cardiac events may help clarify some of the confusion surrounding the relationship between personality and cardiovascular disease. Although triggering information has only limited importance for an individual considering a particular activity, it has great public health value from the perspective of the coronary care unit, because a trigger can be identified in at least 17% of patients who have an acute MI. Recognition of circadian peaks in onset raises the possibility of providing pharmacologic protection during the morning hours for patients already receiving anti-ischemic therapy.3 Although it is clearly impossible to completely eliminate triggering activities, it may eventually prove possible to pharmacologically interrupt the connection between triggering activities and cardiac events.3 For example, the Multicenter Investigation of Limitation of Infarct Size study suggested that β-blockade blunted the morning peak in acute MI presumably by blocking sympathetic activation,2 and the Physicians’ Health Study demonstrated a similar blunting in subjects taking aspirin.15
Research directions To elucidate the mechanisms by which triggers cause cardiac events, it would be useful to study the relationship between the daily activities that are triggers and the physiologic or pathophysiologic processes they induce. The molecular mechanisms of these relationships also should be explored, with the goal of designing therapies that suppress unnecessary and hazardous changes in hemodynamics, hemostasis, and vasoconstriction.3 The factors that render atherosclerotic plaque vulnerable to rupture remain obscure and necessitate urgent clarification.3 It is likely that the techniques used to study the fatty streak that typifies the early stage of atherosclerosis can also be used to determine the mechanisms by which more advanced plaques become vulnerable and susceptible to disruption.10 Future questions would
American Heart Journal April 1999
S8 Muller
include whether it might be possible to detect and treat vulnerable plaques before they rupture. Intracoronary angioscopy and near-infrared spectroscopy are promising approaches in this regard. On the molecular level the identification of the growth factors and chemotactic agents that regulate macrophage infiltration of plaques and control production of collagenase might ultimately enable us to prevent the development of vulnerable plaques.
7.
8.
9.
References 1. Willich SN, Muller JE. Introduction. In: Willich SN, Muller JE, editors. Triggering of acute CHD: implications for prevention. Dordrecht: Kluwer Academic Publishers; 1995. p. 17-20. 2. Muller JE, Stone PH, Turi ZG, Rutherford JD, Cziesler CA, Parker C, et al and the MILIS Study Group. Circadian variation in the frequency of onset of acute myocardial infarction. N Engl J Med 1985;313:1315-22. 3. Muller JE, Tofler GH, Stone PH. Circadian variation and triggers of onset of acute cardiovascular disease. Circulation 1989;79:733-43. 4. Muller JE, Ludmer PL, Willich SN, Tofler GH, Aylmer G, Klangos I, et al. Circadian variation in the frequency of sudden cardiac death. Circulation 1987;75:131-8. 5. Willich SN, Levy D, Rocco MB, Tofler GH, Stone PH, Muller JE. Circadian variation in the incidence of sudden cardiac death in the Framingham Heart Study population. Am J Cardiol 1987; 60:801-6. 6. Marler JR, Price TR, Clark GL, Muller JE, Robertson T, Mohr JP, et
10.
11. 12.
13. 14. 15.
al. Morning increase in onset of ischemic stroke. Stroke 1989;20: 473-6. Mittleman MA, Maclure M, Tofler GH, Sherwood JB, Goldberg RJ, Muller JE, for the Determinants of Myocardial Infarction Onset Study Investigators. Triggering of acute myocardial infarction by heavy physical exertion. Protection against triggering by regular exertion. N Engl J Med 1993;329:1677-83. Mittleman MA, Maclure M, Sherwood JB, Mulry RP, Tofler GH, Jacobs SC, et al, for the Determinants of Myocardial Infacrtion Onset Study Investigators. Triggering of acute myocardial infarction onset by episodes of anger. Circulation 1995;92:1720-25. Willich SN, Maclure M, Mittleman MA, Arntz H-R, Muller JE. Sudden cardiac death. Support for a role of triggering in causation. Circulation 1993;87:1442-50. Muller JE, Abela GS, Nesto RW, Tofler GH. Triggers, acute risk factors and vulnerable plaques: the lexicon of a new frontier. J Am Coll Cardiol 1994;23:809-13. Libby P. Molecular bases of the acute coronary syndromes. Circulation 1995;91:2844-50. Muller JE, Mittleman MA, Maclure M, Sherwood JB, Tofler GH, for the Determinants of Myocardial Infarction Onset Study Investigators. Triggering of myocardial infarction by sexual activity. Low absolute risk and prevention by regular physical exertion. JAMA 1996;275:1405-9. DeBusk RF. Sexual activity triggering myocardial infarction. One less thing to worry about (editorial). JAMA 1996;275:1447-8. Clausen JP. Effect of physical training on cardiovascular adjustments to exercise in man. Physiol Review 1977;57:779-815. Steering Committee of the Physicians’ Health Study Research Group. Final report on the aspirin component of the ongoing Physicians’ Health Study. N Engl J Med 1989; 321:129-35.
AHJ American Heart Journal
Copyright © 1999 by Mosby, Inc.
April 1999 Volume 137, Number 4, Part 2
Circadian variation and triggering of acute coronary events James E. Muller, MD Lexington, Ky
The recognition that the onset of cardiovascular events follows a circadian periodicity and is frequently triggered by physical or mental stresses has created new possibilities for disease prevention. Morning peaks in occurrence are now well-documented for acute myocardial infarction, sudden cardiac death, transient myocardial ischemia, and ischemic stroke. The morning increase in events begins after subjects assume an upright posture and start the day’s activities, during a time of sympathetic nervous system activation. Additional triggers of onset include heavy physical exertion, sexual activity, and anger, the risks of which have been quantified in the Determinants of Myocardial Infarction Onset Study. A general hypothesis of the triggering of coronary thrombosis has been proposed. The process begins with the development of a vulnerable atherosclerotic plaque, which may become disrupted by internal forces or by external hemodynamic or vasoconstrictive changes. Once disrupted, the plaque becomes a thrombogenic focus. An occlusive thrombus is more likely to form if other factors come into play to increase coagulability and vasoconstriction. From a clinical standpoint these findings provide theoretical support for the use of longacting agents to provide adequate anti-ischemic protection during the higher risk morning hours in patients already taking antiischemic medications. From a research standpoint this new information on triggering provides clues to a mechanism of onset that might lead to more effective preventive therapy. Because most deaths from coronary artery disease occur before any type of acute therapy can be given, further efforts to explore this new field are warranted. (Am Heart J 1999;137:S1-S8)
Despite new insights into long-term cardiovascular risk factors and striking advances in the acute in-hospital treatment of patients with myocardial infarction (MI), the coronary artery disease (CAD) mortality rate remains persistently high. Indeed, each year in the United States alone, CAD accounts for 500,000 deaths, half of which occur outside the hospital and spoil suddenly.1 Moreover, 600,000 acute MIs occur each year in individuals who either had no symptoms or had only mild symptoms.2 These grim statistics underscore the need to broaden our concept of risk factor modification to encompass acute and chronic risk factors for acute MI and sudden death.1 In the past decade epidemiologic studies have objecFrom the Gill Heart Institute. Reprint requests: James E. Muller, MD, Gill Heart Institute, 800 Rose St, University of Kentucky Medical Center, Lexington, KY 40502. Copyright © 1999 by Mosby, Inc. 0002-8703/99/$8.00 + 0 4/0/96909
tively confirmed the circadian variation in the onset of cardiac events and have established the role of physical and mental stressors as triggers of onset. Although the possibility of a positive reporting bias in the medical literature cannot be completely excluded, the overwhelming majority of studies find that most types of cardiovascular events show a morning increase. The identification of these triggers and the physiologic or pathophysiologic changes that they induce in the hours and minutes before the onset of a cardiac event has opened up broad new horizons for prevention. The major challenge for the coming decade will be to devise therapies that delay the transition from chronic CAD to an acute event. It may prove possible to accomplish this delay by interrupting the links between such triggers as awakening, physical exertion, and mental stress and such acute risk factors as an increase in blood pressure (BP), an increase in coagulability, or vasoconstriction. After a review of the evidence that supports circadian
American Heart Journal April 1999
S2 Muller
Figure 1
periodicity and triggering stresses in the onset of cardiac events, this article will discuss the ramifications of these findings for patient management.
Circadian variation and cardiovascular events Myocardial infarction
image not available
Bar graphs of time of day of onset of myocardial infarction, sudden cardiac death, thrombotic stroke, and transient myocardial ischemia in 4 different groups of patients. Reproduced with permission from Muller JE, Tofler GH, Stone PH. Circadian variation and triggers of onset of acute cardiovascular disease. Circulation 1989;79:733-43.
Reports in the world literature going back as far as 1960 have described a circadian periodicity in the onset of MI pain and have documented a peak incidence in the hours between 6 AM and noon. However, because these early studies relied on subjective descriptions of pain, they were unable to rule out the possibility that pain experienced during sleep might have gone unnoticed. Not until 1985 was an attempt made to quantify the temporal distribution of acute MI onset on the basis of objective evidence. The records of nearly 1000 patients enrolled in the Multicenter Investigation of Limitation of Infarct Size trial were analyzed to determine the time of infarct onset according to serial measurements of plasma creatine kinase MB activity.2 Consistent with previous reports, the Multicenter Investigation of Limitation of Infarct Size investigators observed that the frequency of pain onset increased significantly between 6 AM and noon and that the incidence of MI during this 6-hour period was 1.28 times greater than during the other three 6-hour intervals of the day (P < .01).2 When we analyzed the distribution of MI onset according to enzyme measurements, we noted a peak between the hours of 5 AM and 2 PM.2 These data demonstrated that the incidence of enzymatically confirmed infarct onset between 6 AM and noon was 1.43 times higher than during the other three 6-hour time intervals. A subsequent study has shown that the increase in the incidence of MI was not a consequence of time per se but rather could be narrowed down to the first 4 hours after awakening.3 The characteristic circadian rhythm was as prominent in patients who did not smoke or drink coffee as in those who did, indicating that these habits were unlikely to have accounted for the morning peak.2 Circadian periodicity was not observed in patients who were under treatment with β-blockers at the time of their acute MI.2 The emergence of objective evidence linking the onset of infarction to the hours after awakening raised a number of interesting questions about the biologic underpinnings of this relationship. Particular attention has been focused on the role of certain endogenous
American Heart Journal Volume 137, Number 4, Part 2
daily rhythms such as the morning trough in body temperature, the rise in plasma catecholamine levels and heart rate (HR), the increase in arterial pressure stimulated by increased sympathetic activity, and the trough in fibrinolytic activity coupled with an increase in platelet aggregation.2 Most significant was the realization that identification and modification of the processes that drive the circadian rhythm of MI onset could potentially represent a strategy for delaying or preventing such catastrophic events.2
Muller S3
Figure 2
Sudden death Two large-scale analyses, the Massachusetts Death Certificate Study4 and the Framingham Heart Study,5 have established a circadian variation in the incidence of sudden cardiac death that parallels the pattern reported for acute MI. Examination of the records of more than 2000 subjects who died suddenly in Massachusetts in 1983 has identified a primary peak between 7 AM and 11 AM.4 This analysis revealed that the frequency of sudden cardiac death dropped between the hours of 11 AM and 5 PM and reached its nadir during sleep between the hours of 11 PM and 6 AM.4 These findings were bolstered by data from more than 5000 subjects in the original Framingham cohort, of whom 429 had sudden cardiac death. According to the Framingham analysis, the hourly risk of sudden death was at least 70% greater between 7 AM and 9 AM than during the rest of the day. Analogous to the situation with acute MI onset, significantly more sudden deaths occurred between 6 AM and noon than during the other three 6-hour intervals of the day.5 The morning peak documented in these 2 studies is compatible with 2 prevailing theories of the cause of sudden death.4,5 The morning increase in sympathetic activity could increase electrical instability, leading to fatal arrhythmias; alternatively, the rise in BP on awakening could increase the odds of atherosclerotic plaque rupture, the pathophysiologic significance of which would be compounded by morning increases in coronary artery tone and platelet aggregability.4,5
image not available
Stroke In a study of patients with stroke, neurologists at 4 academic hospital centers collected information on more than 1200 patients with ischemic strokes during a 3-year period.6 Their analysis demonstrated that in awake patients, more strokes occurred between 10 AM and noon than during any other 2-hour interval. The incidence of ischemic strokes was at its maximum from 8 AM to noon, declining during the afternoon
Bar graphs of variation during 24-hour period of 5 physiologic processes possibly contributing to increase in morning frequency of disease onset. Reproduced with permission from Muller JE, Tofler GH, Stone PH. Circadian variation and triggers of onset of acute cardiovascular disease. Circulation 1989;79:733-43.
American Heart Journal April 1999
S4 Muller
Figure 3
image not available
Illustration of hypothetical method by which daily activities may trigger coronary thrombosis. Three triggering mechanisms are presented: (1) physical or mental stress producing hemodynamic changes leading to plaque rupture, (2) activities causing increase in coagulability, and (3) stimuli leading to vasoconstriction. Reproduced with permission of the American College of Cardiology from Muller JE, Abela GS, Nesto RW, Tofler GH. Triggers, acute risk factors and vulnerable plaques: the lexicon of a new frontier. J Am Coll Cardiol 1994;23:809-13.
and early evening.6 The circadian pattern was not altered by therapy with aspirin, dipyridamole, or warfarin, by a history of hypertension, or by BP level on admission.6 This study demonstrated that, contrary to traditional belief, the onset of ischemic stroke peaks when BP is at its highest for the day. The authors suggested that the preponderance of morning strokes may have also been related to morning peaks in plasma viscosity, hematocrit, and platelet aggregability.6
Transient ischemia Studies have consistently shown that the incidence of transient episodes of myocardial ischemia with STsegment depression is also at a high point between the hours of 6 AM and noon.3 Adjustment for the time of awakening indicates that, as with acute MI, the peak incidence of transient myocardial ischemia occurs in the first 4 hours after awakening.3
Daily activities as triggers Although there is a confluence of neurohormonal, hemostatic, and hemodynamic changes in the morning
that could precipitate a cardiovascular event in susceptible individuals, similar processes occur throughout the day in response to a variety of seemingly innocuous activities. Indeed, the morning peak in the incidence represents only 1 special circumstance in which triggering mechanisms come into play. The relationship between acute MI and triggering activities was recognized as early as 1910 by Obraztsov and Strazhesko in Russia, who colorfully described the onset of acute MI in the context of stressful activities: “Direct events often precipitated the disease; the infarct began in one case on climbing a high staircase, in another during an unpleasant conversation, and in a third during emotional distress associated with a heated card game.”3 After nearly a century of controversy, recent new insights have highlighted the role of daily activities in precipitating many cases of acute MI and sudden death.
A new epidemiologic method Although descriptive studies have indicated that the onset of acute MI is preceded by heavy physical exertion in approximately 5% of patients7 and by emotional
American Heart Journal Volume 137, Number 4, Part 2
stress such as anger in 4% to 18% of cases,8 these studies have been hampered by the lack of control data. To more rigorously clarify the role of triggering factors in the development of acute MI, we developed and used a novel case-crossover method in the Determinants of Myocardial Infarction Onset Study. The method offered the unique advantage of having each patient serve as his or her own control.7 Trained examiners interviewed patients an average of 4 days after acute MI to ascertain the patients’ exposure to triggering factors during a brief “hazard period” before infarction. Exposure during the hazard period was then compared with the patient’s exposure to the possible trigger during a comparable period the day before infarction, and with his or her usual frequency of exposure during the year preceding infarction.
Muller S5
Figure 4
Physical exertion and acute MI By means of interviews with more than 1200 survivors of acute MI in 45 hospitals across the country, we first quantified the relative risk of infarction after heavy physical exertion and assessed whether this risk could be modified by regular physical activity.7 Our findings revealed that 4.4% of patients had a trigger of heavy physical exertion in the hour before their infarction. The most frequently implicated activities were mixed isotonic and isometric exercise such as gardening or splitting wood (52%) followed by isotonic exercise such as jogging or racquet sports (30%) and isometric exercise such as lifting or pushing (18%).7 The risk of having an acute MI was nearly 6 times higher in the hour after heavy exertion than during periods of light or no exertion. Perhaps most important, the risk of infarction after heavy exertion was more than 100 times greater for patients who were habitually sedentary but was only 2.4 times greater for those who regularly exercised. These findings suggested that a regular program of physical activity not only lowers the baseline risks of a cardiac event but also decreases the relative risk (RR) of MI after heavy exertion. It should also be stressed that the Onset Study demonstrated no excess RR for patients who had a history of CAD before their infarction, although such patients would be expected to have a greater baseline risk.
Anger and acute MI We then applied the same epidemiologic technique to clarify the role of anger in triggering acute MI in more than 1600 patients.8 With a specially designed self-report onset anger scale of 1 (calm) to 7 (enraged), we found that 8% of patients reported episodes of
Relative risk of acute myocardial infarction after identifiable triggering activities. Dotted line represents baseline risk.7-9,12
anger at level 5 (very angry) and above during the 24 hours before acute MI and that 2.4% reported such episodes in the 2-hour period preceding the infarction.8 Comparison with the frequency of angry episodes during the day before and the year preceding infarction revealed that the odds of an acute MI were more than doubled in the 2 hours after an episode of anger.8 For regular users of aspirin, however, the RR was only 1.4 compared with 2.9 among nonusers.8 Thus, aspirin appeared to reduce the excess risk of MI posed by discrete episodes of anger.8
Other cardiac events A considerable number of descriptive reports have also linked sudden cardiac death to physical or mental stress.9 However, studies with case-control and casecrossover designs will be necessary to quantify the RRs of these stresses. In other investigations, it has been shown that more than half of episodes of transient myocardial ischemia are preceded by possible triggering activities such as increased physical or mental stress.3
Mechanisms of triggering The circadian rhythm of onset of acute MI and sudden death suggests a hypothetical sequence of events that may culminate in a cardiac event. The common shared factor in most cases of MI and sudden cardiac
American Heart Journal April 1999
S6 Muller
death is coronary artery thrombosis. Occlusive thrombosis is the culprit in virtually all patients who have Q-wave infarctions and in at least one half of patients who have a sudden cardiac death. Moreover, even in victims of sudden death who do not have an occlusive thrombosis, nonocclusive thrombosis is often present.3
namic changes all affect the activation and release of platelets, leading to a thrombogenic tendency. Intensified sympathetic activity could give rise to paradoxical coronary vasoconstriction, which has been demonstrated in diseased arterial walls and may also predispose to complete occlusion.9
Vulnerable plaque
Triggering factors
Pathologic and angiographic studies have indicated that coronary thrombus most frequently forms on the surface of a ruptured atherosclerotic plaque.3 Although the exact mechanisms by which plaque becomes vulnerable to rupture are not fully known, several suggestive clues have been identified. Plaque disruption frequently begins in lesions previously causing a <50% reduction in lumen diameter. Disrupted plaques have been found to contain higher concentrations of lipids and macrophage foam cells than have nondisrupted plaques.10 This accumulation of lipids forms the core of the plaque, which is covered by a thin fibrous cap. The integrity of the fibrous cap may be a key determinant of the transformation from stable to vulnerable plaque. The degradation of collagen and elastin in the cap by collagenase, gelatinases, stromelysin, and other protease enzymes may leave the plaque more susceptible to rupture.11
Although these physiologic changes occur in tandem in the morning, they may occur at random times throughout the day in response to any of a variety of physical and mental triggers.3 We have hypothesized that the first precipitating step in the chain of events leading to coronary thrombosis is an acute physical or mental stress that triggers a hemodynamic change sufficient to rupture a vulnerable atherosclerotic plaque (Fig 3).10 Whether the ruptured plaque leads to total or partial occlusion of the arterial lumen or to no symptoms at all will depend not only on the severity of the rupture but also on the contribution of external factors that may heighten coagulability, thereby promoting thrombus growth, or that may increase vasoconstriction, thereby occluding an already narrowed artery. A hypothetical model can also be created to explain how external triggers may lead to sudden cardiac death.9 An external trigger, whether the act of awakening, heavy exertion, or mental stress, results in an increase in sympathetic activity. Hemodynamic forces may then cause the rupture of vulnerable plaque. During the critical phase of plaque rupture and thrombus formation, platelet characteristics and arterial wall function appear to be involved in the pathogenesis of sudden death.9 Exposed collagen in the fibrous cap of the plaque may lead directly to the development of an occlusive thrombus and acute MI. Alternatively, the thrombus may grow gradually, depending on blood coagulability and vasoconstriction. This growth leads to reduced flow, microemboli, and ischemia or even small areas of necrosis, both of which have been linked to a decreased threshold for ventricular fibrillation. Damage caused by necrosis, fibrosis, and hypertrophy may increase the hazard of acute cardiac electric instability. Episodes of ischemia in the context of underlying CAD have been associated with an increased risk of sudden death.9 Sympathetic activation may also lead directly to sudden death by causing a primary arrhythmia. It is also possible that plaques rupture because of processes intrinsic to the plaque, and the trigger simply adds thrombotic forces to a previously ruptured plaque.
Acute hemodynamic, hemostatic, and vasoconstrictive risk factors A number of physiologic processes, many of which are amplified during the morning hours, may act in concert as acute risk factors for a cardiac event. The morning peaks in acute MI and sudden death may reflect the synchronization of the population for these risk factors in the hours after awakening. The most important of these processes are the increases in systemic BP and HR, the decrease in coronary blood flow, increases in blood viscosity and platelet aggregability, decrease in tissue-type plasminogen activator activity, and increases in plasma cortisol and sympathetic activity (Fig 2).3 The morning rise in arterial pressure and HR could precipitate rupture of a vulnerable atherosclerotic plaque. The decrease in flow caused by a fixed stenosis could be exacerbated by the morning increase in coronary arterial tone. Increases in blood viscosity and platelet aggregability, coupled with a drop in endogenous tissue-type plasminogen activator activity, could lead to a hypercoagulable state that sets the stage for thrombus growth. Acute arterial wall injury, endothelial cell dysfunction, and hemody-
American Heart Journal Volume 137, Number 4, Part 2
Clinical ramifications In practical terms the key question is how to translate our newly enhanced understanding of triggering mechanisms into clinical recommendations. It should be emphasized that identification of a possible triggering activity does not automatically mean that the patient must be advised to avoid this activity.1 It is clear that awakening in the morning and unanticipated physical and psychologic stresses are unavoidable aspects of daily living. In addition, it is important to recall that even a dramatic increase in RR may translate into a very small absolute risk if the baseline risk is low to begin with.7 For example, the odds that a healthy 50-year-old nonsmoking man will have an acute MI during any given 1-hour period are 1 in a million. If he regularly engages in exercise, the odds of his having a heart attack during an hour of exercise are more than doubled, but this means an absolute risk of only 1 in 400,000. If the same man led a more sedentary life, his RR of MI during an hour of exercise would be 4100 times greater than during rest, but his absolute risk would still be only 1 in 10,000. Although the relative risks of morning exertion versus evening exertion have not yet been clarified, the absolute risk at either time is so low, and the benefits of exercise so well established, that avoidance of morning exertion is not justified. Of great value to MI survivors and those with cardiac disease will be the recent findings of the Determinants of Myocardial Infarction Onset Study Investigators regarding the RRs of nonfatal MI triggered by sexual activity, both among the general population and in patients with previous coronary heart disease.12 A casecrossover analysis of 858 sexually active patients indicates that although sexual activity can trigger the onset of MI, the RR of acute MI related to sexual intercourse is quite low, and the absolute risk is extremely low (1 in 1,000,000 for a healthy individual). Moreover, the RR is not increased in patients with a history of cardiac disease. Thus, physicians can actually use data on triggering mechanisms to reassure their patients that the odds of having a heart attack during, for example, sexual activity, are very slim indeed (Fig 4).7-9,12 Particularly noteworthy is the Onset Study’s observation that MI induced by sexual activity was most common among sedentary individuals and least common in active individuals.13 It has been demonstrated that at any specific level of submaximal effort associated with exertion such as sexual activity, HR, BP, and level of sympathetic activation are lower in physically active
Muller S7
persons than in sedentary ones.14 Thus the Onset Study results strongly support the premise that regular exercise can prevent triggering of MI.12 Data from studies of triggering factors can be relevant to the treatment of patients who repeatedly and regularly have a particular avoidable trigger. Repeated regular exposure to a trigger such as anger may markedly increase the cumulative risk and may therefore warrant intervention. For example, stress management instruction may be helpful for the patient who has anger repeatedly throughout the day.1 The understanding that anger is an acute risk factor for cardiac events may help clarify some of the confusion surrounding the relationship between personality and cardiovascular disease. Although triggering information has only limited importance for an individual considering a particular activity, it has great public health value from the perspective of the coronary care unit, because a trigger can be identified in at least 17% of patients who have an acute MI. Recognition of circadian peaks in onset raises the possibility of providing pharmacologic protection during the morning hours for patients already receiving anti-ischemic therapy.3 Although it is clearly impossible to completely eliminate triggering activities, it may eventually prove possible to pharmacologically interrupt the connection between triggering activities and cardiac events.3 For example, the Multicenter Investigation of Limitation of Infarct Size study suggested that β-blockade blunted the morning peak in acute MI presumably by blocking sympathetic activation,2 and the Physicians’ Health Study demonstrated a similar blunting in subjects taking aspirin.15
Research directions To elucidate the mechanisms by which triggers cause cardiac events, it would be useful to study the relationship between the daily activities that are triggers and the physiologic or pathophysiologic processes they induce. The molecular mechanisms of these relationships also should be explored, with the goal of designing therapies that suppress unnecessary and hazardous changes in hemodynamics, hemostasis, and vasoconstriction.3 The factors that render atherosclerotic plaque vulnerable to rupture remain obscure and necessitate urgent clarification.3 It is likely that the techniques used to study the fatty streak that typifies the early stage of atherosclerosis can also be used to determine the mechanisms by which more advanced plaques become vulnerable and susceptible to disruption.10 Future questions would
American Heart Journal April 1999
S8 Muller
include whether it might be possible to detect and treat vulnerable plaques before they rupture. Intracoronary angioscopy and near-infrared spectroscopy are promising approaches in this regard. On the molecular level the identification of the growth factors and chemotactic agents that regulate macrophage infiltration of plaques and control production of collagenase might ultimately enable us to prevent the development of vulnerable plaques.
7.
8.
9.
References 1. Willich SN, Muller JE. Introduction. In: Willich SN, Muller JE, editors. Triggering of acute CHD: implications for prevention. Dordrecht: Kluwer Academic Publishers; 1995. p. 17-20. 2. Muller JE, Stone PH, Turi ZG, Rutherford JD, Cziesler CA, Parker C, et al and the MILIS Study Group. Circadian variation in the frequency of onset of acute myocardial infarction. N Engl J Med 1985;313:1315-22. 3. Muller JE, Tofler GH, Stone PH. Circadian variation and triggers of onset of acute cardiovascular disease. Circulation 1989;79:733-43. 4. Muller JE, Ludmer PL, Willich SN, Tofler GH, Aylmer G, Klangos I, et al. Circadian variation in the frequency of sudden cardiac death. Circulation 1987;75:131-8. 5. Willich SN, Levy D, Rocco MB, Tofler GH, Stone PH, Muller JE. Circadian variation in the incidence of sudden cardiac death in the Framingham Heart Study population. Am J Cardiol 1987; 60:801-6. 6. Marler JR, Price TR, Clark GL, Muller JE, Robertson T, Mohr JP, et
10.
11. 12.
13. 14. 15.
al. Morning increase in onset of ischemic stroke. Stroke 1989;20: 473-6. Mittleman MA, Maclure M, Tofler GH, Sherwood JB, Goldberg RJ, Muller JE, for the Determinants of Myocardial Infarction Onset Study Investigators. Triggering of acute myocardial infarction by heavy physical exertion. Protection against triggering by regular exertion. N Engl J Med 1993;329:1677-83. Mittleman MA, Maclure M, Sherwood JB, Mulry RP, Tofler GH, Jacobs SC, et al, for the Determinants of Myocardial Infacrtion Onset Study Investigators. Triggering of acute myocardial infarction onset by episodes of anger. Circulation 1995;92:1720-25. Willich SN, Maclure M, Mittleman MA, Arntz H-R, Muller JE. Sudden cardiac death. Support for a role of triggering in causation. Circulation 1993;87:1442-50. Muller JE, Abela GS, Nesto RW, Tofler GH. Triggers, acute risk factors and vulnerable plaques: the lexicon of a new frontier. J Am Coll Cardiol 1994;23:809-13. Libby P. Molecular bases of the acute coronary syndromes. Circulation 1995;91:2844-50. Muller JE, Mittleman MA, Maclure M, Sherwood JB, Tofler GH, for the Determinants of Myocardial Infarction Onset Study Investigators. Triggering of myocardial infarction by sexual activity. Low absolute risk and prevention by regular physical exertion. JAMA 1996;275:1405-9. DeBusk RF. Sexual activity triggering myocardial infarction. One less thing to worry about (editorial). JAMA 1996;275:1447-8. Clausen JP. Effect of physical training on cardiovascular adjustments to exercise in man. Physiol Review 1977;57:779-815. Steering Committee of the Physicians’ Health Study Research Group. Final report on the aspirin component of the ongoing Physicians’ Health Study. N Engl J Med 1989; 321:129-35.