Arteriographie anatomy and mechanisms of myocardial ischemia in unstable angina

JACC Vol. 9. No.6 June 1987:1397-402

1397

EDITORIALS

Arteriographic Anatomy and Mechanisms of Myocardial Ischemia in Unstable Angina JOHN A. AMBROSE, MD, FACC, CRAIG E. HJEMDAHL-MONSEN, MD New York , New York

Over the past 20 years, there has been considerable research into the pathogenesis of myocardial ischemia in unstable angina. Early data (1-3) suggested that transient increases in myocardial oxygen demand explained episodes of rest pain in patients with unstable angina and severe coronary artery disease. In the 1970s this view changed and the concept of vasospasm became the most favored mechanism to explain episodes of myocardial ischemia and rest pain, thus changing the emphasis from the demand side to the supply side of coronary dynamics (4,5). More recently, serial angiographic studies (6,7) have demonstrated that progression of coronary artery disease is extremely common in patients with stable angina who are restudied after an episode of unstable angina . Angiographic, angioscopic and pathologic data (8-13) have also incriminated thrombus in the pathogenesis of this condition. In this review we explore the relation between the coronary arteriographic anatomy in unstable angina and mechanisms of myocardial ischemia .

stable angina; and 3) all ischemic rest pain without evidence of myocardial infarction. These catagories are not mutually exclusive . The category of rest pain would include some patients with previously mild angina and a marked increase in symptoms (crescendo angina) and also patients with previously severe angina and a slight increase in symptoms. Other proposed classifications for unstable angina include the categories of subendocardial infarction and postinfarction angina. The natural history of patients with these conditions may be different from that of patients classified in the preceding categories (14). More complex classifications have been proposed based on clinical or angiographic criteria , or both (15,16) . For the remainder of this discussion, we will classify unstable angina as we did in the preceding paragraph .

Definition of Unstable Angina

Atherosclerotic disease. Eighty-five to 90 percent of patients with unstable angina have significant coronary artery disease but in about 10 to 15% of patients, normal or nonobstructed coronary arteries are found (17,18). Thus , some patients with normal or nonobstructed coronary arteries have dynamic obstruction and myocardial ischemia related to coronary artery spasm or some other mechanism; other patients have a chest pain syndrome that is noncardiac in origin. The majority ofpatients with unstable angina have multivessel coronary artery disease. However, if one looks at only those patients with recent onset unstable angina, there is a high incidence of one vessel disease (25 to 50%) (19,20) . Significant obstruction of the left main coronary artery is more common in unstable than in stable angina and occurs in up to 25% of patients with rest and exertional chest pain and a previous history of stable angina (21). Ischemia-producing artery. In patients with unstable angina and single vessel disease, it can be assumed that the ischemia-producing artery is that vessel with the significant obstruction . In rare instances, however, ergonovine testing

A major difficulty in unraveling the cause of any condition occurs when that condition is not precisely defined. Such is the case with unstable angina , which is a descriptive term for a spectrum of acute myocardial syndromes that clinically lie between stable angina and myocardial infarction. It generally includes the following categories: I) New onset of angina (1 to 2 months ' duration) occurring at low work loads or at rest; 2) crescendo angina defined as a marked increase in the frequency or severity of previously *Editorials published in Journal ofthe American College ofCardiol ogy rellect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology. From the Department of Medicine. Division of Cardiology. The Mount Sinai School of Medicine and The Mount Sinai Medical Center . New York, New York 10029. This work is supported by the New York Cardiac Center, Englewood Cliffs, New Jersey . Manuscript received September 29. 1986; revised manuscript received December 10, 1986, accepted January 7. 1987. Address for reprints: John A. Ambrose, MD, Division of Cardiology . The Mount Sinai Medical Center , One Gustave L. Levy Place, New York. New York 10029. © 1987 by the American College of Cardiology

Coronary Arteriographic Anatomy in Unstable Angina

0735-1097/87/$3.50

1398

AMBROSE AND HJEMDAHL-MONSEN EDITORIAL

has indicated spasm (with reproducibility of clinical symptoms) of a normal-appearing artery in the presence of significant obstruction in other vessels (22). Whereas multivessel disease is common in unstable angina, it is often possible to identify an ischemia-producing artery. By a careful assessment of both the distribution of lesions within the coronary arteries and segmental wall motion analysis as well as electrocardiographic (ECG) changes during pain, we (8) were able to localize the presumed artery responsible for symptoms in about 40% of patients with either stable or unstable angina and multivessel coronary artery disease. The presence of an "acute" coronary lesion as determined by analysis of coronary morphology will increase the likelihood of localizing the ischemia-producing artery in unstable angina and multivessel disease (see later). Progression of coronary artery disease. In previously catheterized patients with a history of stable angina pectoris, progression of coronary artery disease is seen in approximately 75% who are recatheterized after an episode of unstable angina (6,7). On the other hand, only about 33% of patients recatheterized with little or no increase in symptoms show such progression. This progression can be seen to occur in lesions that at the time of the first angiogram were either significant (>50% diameter obstruction) or "nonobstructive" «50% diameter obstruction) (6,7). We found (7) that in unstable angina, progression in the ischemiaproducing artery usually originated from a previously insignificant lesion or even from a normal-appearing artery. This pattern of progression from previously insignificant disease was seen less commonly in the study of Moise et al. (6). Although quantitative techniques for analyzing coronary stenoses were not utilized in these studies of coronary disease progression, approximately 25% of patients in both studies did not appear to have any progression of disease despite the symptomatic progression to unstable angina. Quantitative analysis of coronary lesions in unstable angina. There are ample data, (18,23) suggesting that the extent and location of coronary artery disease are similar for patients with stable and unstable angina except for the higher incidence of left main coronary involvement in unstable angina. However, there are few data that quantitatively analyze the degree of stenosis or the minimal diameter of the ischemia-producing lesion in unstable angina versus other acute coronary syndromes. Although visual analysis of the degree of coronary stenosis has been shown to be similar in the acute coronary syndromes (23), this method is inaccurate for precise comparison. Raffenbeul et al. (24) found similar degrees of stenosis in patients with either unstable angina or stable angina when lesions were quantitatively measured with a vernier caliper. Utilizing previously developed quantitative techniques, McMahon et al. (25) found more significant stenosis (both a higher percent stenosis and a smaller minimal diameter) in a small number of patients with one vessel disease and subendocardial in-

JACC Vol. 9, No.6 June 1987:1397-402

Figure 1. Left coronary artery angiogram in the leftanterior oblique projection showing a type II eccentric lesion in the left anterior descending artery (arrow).

farction in comparison with patients with one vessel disease and unstable angina. The lack of more quantitative data makes it difficult to assess the role of the degree of stenosis, the minimal diameter or the cross-sectional area of the coronary lesion in the ischemia-producing artery as a sole determinant of the particular clinical syndrome. Therefore, more quantitative data are required to compare unstable angina with other acute coronary syndromes. Qualitative analysis of coronary lesions in unstable angina. In either acute presentations of unstable angina (8), recent non-Q wave infarction (26), acute infarction with ST segment elevation or recent Q wave myocardial infarction (27), a characteristic coronary lesion is found in the majority of cases in which an ischemia-producing artery < 100% occluded is identified. An eccentric stenosis in the form of a convex intraluminal obstruction with a narrow neck due to one or more overhanging edges, or borders that are scalloped (type II eccentric lesion) is found in approximately two-thirds of such patients (Fig. 1). This type II eccentric lesion is also common in coronary artery disease progression. It is rarely found in patients with stable angina. In the remaining one-third of ischemia-producing arteries in these acute syndromes, other types of morphology can be seen. Symmetric and smooth (concentric)or asymmetric and smooth (type I eccentric) narrowings are the most common lesions in those ischemia-producing arteries that contain non-type II eccentric lesions. Other angiographic studies have assessed the importance of intracoronary filling defects in unstable angina (9,10). Filling defects located either within or distal to a lesion have been described in about one-third of patients with unstable angina and probably represent intracoronary thrombi. In patients with an episode of rest pain within 24 hours before angiography, the incidence of intracoronary filling defects is even higher. Bresnahan et al. (9) found that about two-

JACC Vol. 9, No.6

AMBROSE AND HJEMDAHL-MONSEN EDITORIAL

June 1987:1397--402

thirds of filling defects were located distal to the lesion. The recognition of filling defect location is important because the pathogenesis of proximal and distal defects might differ. A filling defect located within a lesion probably represents thrombus forming ar a disruption of the atherosclerotic plaque whereas a filling defect distal to a lesion probably represents thrombus formation due to altered flow patterns in a region of diminished or turbulent blood flow. Pathologic-arteriographic correlations. Postmortem angiographic analysis (12) of patients dying after myocardial infarction or after coronary bypass surgery has indicated that the "complicated lesions" exhibited plaque disruption with partially occlusive or totally occlusive thrombus. These complicated lesions were usually eccentric with irregular borders. Careful histopathologic sectioning of the coronary arteries in patients dying suddenly after an episode of unstable angina (13) has shown that thrombi are commonly present which are anchored to a disruption of the atherosclerotic plaque. Furthermore, these thrombi are usually layered, indicating that such events are not necessarily acute. Almost all plaque disruptions occur at the proximal side of the atherosclerotic lesion, suggesting the importance of rheologic factors in its formation (Davies M, personal communication, 1986). Calculations based on an arterial model indicate that this is the point of maximal shear forces (28). It is likely that the type II eccentric lesion represents a disrupted plaque or a partially occlusive thrombus. or both. Undoubtedly, this lesion is similar in appearance to that described by other groups as intracoronary filling defects at the site of an angiographic stenosis. Angioscopic examination of the coronary arteries at the time of coronary bypass surgery has also emphasized the importance of these lesions in unstable angina. Preliminary data with this technique (11) have shown that plaque rupture and thrombus are even more frequent than suggested by angiography.

Mechanisms of Ischemia in Unstable Angina Supply versus demand. In establishing the origin of ischemia in unstable angina, it is important to ascertain whether myocardial ischemia is due to an increase in oxygen demand relative; to supply or to a reduction in supply relative to demand. Roughgarden (1) determined that in cases of spontaneous angina, increases in blood pressure or heart rate, or both (rate-pressure product), preceded the occurrence of pain. Other investigators (2,3,29) also noted increases in rate-pressure product before the onset of subjective pain sensation. This associationmight lead one to conclude that changes in oxygen demand are the cause of myocardial ischemia. However, this is not necessarily the case because chest pain may be delayed by minutes after the onset of myocardial ischemia as determined by 5T segment changes (5). Several investigators (5,30,31) studied patients with rest angina and found that ECG changes occurred before

1399

the onset of changes in rate-pressure product during an ischemic episode; they concluded that, in patients with rest angina, a primary reduction in myocardial perfusion, rather than an increase in demand, was the cause of ischemia. Any proposed mechanism for ischemia in unstable angina must explain the variable threshold that exists for the occurrence of ischemic pain. Patients with unstable angina may be exercised or their heart artificially paced to a higher heart rate and blood pressure without pain than during spontaneous episodes of pain (32,33). Furthermore, the activity required to elicit pain may vary during the day. These observations also suggest that decreases in supply rather than increases in demand are primarily responsible for ischemic symptoms in patients with unstable angina. Although the visual assessment of the severity of coronary disease has been found to be similar in patients with stable and unstable angina (23), transient decreases in coronary artery diameter related to vasospasm or intermittent thrombosis occurring at the time of an ischemic episode can account for the unstable angina. Fluid dynamics of coronary stenoses. Dynamic changes in coronary diameter can explain episodes of ischemia in unstable angina. To understand how small changes in coronary stenoses affect myocardial blood supply, we must consider the fluid dynamics of blood flow through a coronary stenosis. Pressure loss in terms of flow through a coronary stenosis is mathematically represented by the following equation (neglecting inertial losses) (34): dP =

81T~L ) Q + -pk (I- - -I) ( -2As 2 As An

2

Q2,

where ilP is the pressure loss across the lesion, p. is the blood viscosity, L is the stenosis length, As is the crosssectional area of the stenosis, An is the cross-sectional area of the normal segment, p is the blood density, k is a constant related to the shape of the distal portion of the stenosis and Q is blood flow. The first term in the equation represents pressure lost through viscous or frictional effects whereas the second term represents pressure lost as blood flow exits a lesion by flow separation and turbulence. If one expresses area of the stenotic segment in terms of diameter, then changes in pressure vary as the fourth power of diameter. There are several important consequences to the fluid dynamic equation. Viscous losses are dependent on the blood viscosity, stenosis length and minimal cross-sectional area of the stenosis. Turbulent losses are also dependent on the area of the stenosis, as well as the area of the normal segment. The relation between pressure loss and flow is not linear. As flow increases, turbulent losses increase as the square of the flow. Except for extremely long lesions, the first term in the equation has little impact on the pressure gradient due to a stenosis. The percent stenosis, which is still widely used to assess the severity of a coronary lesion, does not appear in the equation. Rather, the minimal cross-

1400

AMBROSE AND HJEMDAHL·MONSEN EDITORIAL

sectional area of the stenosis is the primary geometric factor that determines the hemodynamic significance of a lesion. In a study by McMahon et al. (25), patients with unstable angina had a minimal area of stenosis of 0.63 mrrr'. At this range, changes in stenosis severity of as little as 0.1 mm in diameter can greatly increase the pressure loss across the lesion , and account for an ischemic episode. This small change can be related, as mentioned, to vasospasm or a small intraluminal thrombus. Causes of altered supply. There are several possible causes for a change in stenosis severity that can account for rest ischemia in unstable angina. Active vasomotion. Coronary arteries are subjectto changes in smooth muscle tone that are under the control of various humoral and autonomic factors. Variation in these factors can result in changes in caliber of the coronary vessels. This active vasomotion or spasm has been proposed by Maseri et al. (35) as an explanation for rest ischemia in patients with unstable angina. Increases in sympathetic or parasympathetic activity can also alter the stimulation of alphaadrenergic receptors in the coronary arteries causing vasoconstriction or dilation. Among the humoral factors that cause active coronary vasoconstriction are angiotensin, serotonin and acetylcholine. The role of some of these and other platelet-related factors are considered later. It also is possible that normal daily variation in the level of these factors causing mild constriction of the epicardial vessel adjacent to an atheroma can alter coronary hemodynamics and account for spontaneous episodes of myocardial ischemia (36) . Passive vasomotion. Changes in coronary artery perfusion pressure can alter the geometry of a coronary stenosis as a result of passive changes in intraluminal distending pressure. Thus, passive vasomotion can also cause increased coronary stenosis resistance and episodes of ischemia in unstable angina. Role of platelets. Platelet attachment changes the geometry of the stenosis and alters coronary flow in several ways . Platelets release vasoactive substances , including thromboxane , serotonin and platelet-derived growth factor, which are known to be powerful vasoconstrictors (37,38) . These substances are released when platelets attach to injured vascular endothelium and become activated. In an experimental pig model (39), the degree of vasospasm of a normal carotid artery after endothelial injury was proportional to the degree of platelet deposition at the site of injury . Also, platelet aggregates might plug the coronary stenosis and lead to further reduction in stenosis diameter during the episodes of myocardial ischemia . In some cases, these changes in luminal caliber needed to significantly affect coronary flow might be small enough to be beyond the resolving power of cineangiographic systems used for cardiac catheterization . It may, therefore, be difficult to identify these changes in coronary stenoses an-

JACCVol. 9, No.6 June 1987:1397-402

giographically, thus accounting in some instances for the apparent lack of progression of disease reported in some patients with unstable angina. In concentrically narrowed canine coronary arteries, cyclic reductions in coronary flow have been demonstrated (40,41) distal to the stenosis and appear related to platelet aggregation alternating with dislodgment of thrombus and embolization. Thromboxane and serotonin inhibitors are usually effective in abolishing these cyclical variations in flow (42,43); in these studies alpha-2-adrenergic antagonists partially reduced these cyclical flow reductions whereas alpha1- and beta-antagonists did not. An eccentrically narrowed experimental model does not exist although most human coronary lesions are asymmetric. In such patients plateletmediated release of these vasoactive prostanoid and nonprostanoid substances at the site of a disrupted plaque can cause active constriction of the ring of normal tissue that partially surrounds the eccentric stenosis (44,45). In addition, it has been observed (46) that concentric lesions in fresh segments of coronary arteries from heart transplant recipients are just as capable of contraction in response to vasomotor stimuli as are eccentric lesions. When flow is reduced during an episode of myocardial ischemia and rest pain, conditions are further enhanced for extension of this platelet-rich thrombus at the site of a disrupted plaque with formation of a red thrombus that may further occlude the lumen. A vicious cycle can result with platelet-induced vasospasm and thrombus formation leading to further platelet aggregation and thrombus formation and ultimately to myocardial infarction . Counteracting this tendency to clotting and vasoconstriction are various platelet, endothelial and humoral factors that cause dilation of the lumen through relaxation of smooth muscle and endogenous thrombolysis. Platelet aggregates forming on the atherosclerotic plaque may subsequently dislodge and embolize to the distal coronary bed. This latter view is supported by pathologic studies of Falk (13) in which platelet-rich thrombi were seen distal to a disrupted plaque in patients dying with a history of unstable angina . Antiplatelet drugs , which can prevent platelet aggregation , do not prevent the attachment of the single layer of platelets (39) to the damaged endothelium of a disrupted atherosclerotic plaque . Therefore, these drugs may not be protective against episodes of chest pain, although they might protect against thrombotic occlusion. In a recent clinical trial using aspirin for unstable angina (47), myocardial infarction and death were reduced by aspirin therapy, but the frequency of anginal attacks was not reduced . Experimentally (39) , pretreatment with aspirin decreases but does not eliminate arterial vasospasm after balloon dilation of the carotid artery. This is related to decreased platelet deposition with aspirin at the site of endothelial injury and might partially explain the inability of aspirin to decrease the incidence of anginal pain in unstable angina .

AMBROSE AND HJEMDAHL-MONSEN EDITORIAL

JACC Vol. 9, No.6 June 1987:1397-402

Anatomic and Pathophysiologic Correlates Incorporating the arteriographic, pathologic and angioscopic anatomy with the potential mechanisms of myocardial ischemia in unstable angina, one can suggest hypotheses that begin to explain the pathophysiology of this syndrome (48). In patients with unstable angina, plaque disruption with resultant platelet deposition or thrombus formation, or both appears to be the initial event in most patients presenting with this acute syndrome. The formation of this "complicated plaque" may significantly reduce the lumen of the coronary artery to a critical point. Angiographically, this stage appears to be represented by the eccentric and irregular lesions seen in most patients with unstable angina (7). It is unclear how much of this progression is due to thrombus formation within or distal to the lesion. However, angioscopic data (11) suggest that it may be a ubiquitous finding. Pathologic examination of eccentric atherosclerotic plaques (44,45) demonstrates a ring of relatively normal vessel wall that is capable of responding to changes in arterial tone mediated by vasoactive substances or changes in autonomic tone. Vasomotion in this arc of normal wall or lysis of platelet aggregates may be all that is necessary to produce small changes in coronary artery diameter. Thus, spontaneous occurrences of myocardial ischemia are usually followed by spontaneous resolution of the ischemia and reestablishment of flow through the coronary artery. Whether these same mechanisms are operative in concentric stenoses is unknown although, as mentioned, concentric lesions respond to vasomotor stimuli in similar fashion as do eccentric lesions. In patients in whom there is no obvious angiographic progression of coronary disease despite the new appearance of unstable angina, there may be other mechanisms leading to ischemia. Increases in myocardial demand may play a role in some of these patients. In addition, as mentioned earlier, the changes caused by active vasomotion or plateletinduced changes (plugging, vasoactive mediators and initiation of thrombus formation at the site of a small plaque disruption, and so forth) may be so subtle as not to be recognized by the usual angiographic techniques.

Conclusions Further research is needed to understand the exact interplay between arteriographic anatomy and myocardial ischemia in unstable angina. Although it appears that most episodes of rest ischemia in unstable angina are related to a decrease in myocardial perfusion relative to myocardial demand, the sequence of events remains unclear. Unraveling the relations among the role of the platelet, active and passive vasomotion and the possibility of embolization of thrombus distal to the lesion should enable us to better understand this condition.

1401

Observations before and during spontaneous episodes of myocardial ischemia and chest pain may be necessary to resolve certain questions concerning the causes of ischemia in unstable angina. Analysis of patients before stabilization with medication may be necessary. Many important issues need to be resolved. The exact role of platelets and humoral mediators in the genesis of ischemia needs to be precisely determined. The influence, if any, of the autonomic nervous system on spontaneous changes in vasomotor tone needs to be evaluated. The importance of thrombus and plaque disruption in eccentric and concentric lesions as well as quantitative analysis of the ischemia-producing coronary lesions during episodes of spontaneous ischemia should also be studied. A randomized, placebo-controlled trial of thrombolytic agents appears warranted. These and other areas will have to be examined to link anatomy to ischemia in unstable angina. It is likely that there is more than one mechanism for ischemia in unstable angina. Given the varied presentation and coronary anatomy, multiple mechanisms are probably responsible. A better understanding of myocardial ischemia in unstable angina should result in more effective therapy, We thank Richard Gorlin, MD, and Valentin Fuster, MD, for continued support and helpful suggestions in reviewing this manuscript.

References I. Roughgarden JW. Circulatory changes associated with spontaneous angina pectoris. Am J Med 1966;41:947-61. 2. Cannon OS, Harrison DC, Schroeder JS. Hemodynamic observations in patients with unstable angina pectoris. Am J Cardiol 1974;33:17-22. 3. Scheidt S, Wolk M, Killip J. Unstable angina pectoris: natural history, hemodynamics, uncertainties of treatment and the ethics of clinical study. Am J Med 1976;60:409-17. 4. Maseri A, L' Abbate A, Chierchia S. Significance of spasm in the pathogenesis of ischemic heart disease. Am J Cardiol 1979;44:788-92. 5. Chierchia S, Brunelli C, Simonette 0, Lazzari M, Maseri A. Sequence of events in angina at rest: primary reduction in coronary flow. Circulation 1980;61:759-68. 6. Moise A, Theroux P, Taeymans Y, et al. Unstable angina and progression of coronary atherosclerosis. N Engl J Med 1983;309:685-9. 7. Ambrose JA, Winters SL, Arora RR, et al. Angiographic evolution of coronary artery morphology in unstable angina. J Am Coli Cardiol 1986;7:472-8. 8. Ambrose JA, Winters SL, Stem A, et al. Angiographic morphology and the pathogenesis of unstable angina pectoris. J Am Coli Cardiol 1985;5:609-16. 9. Bresnahan DR, Davis DR, Holmes DR Jr, Smith HC. Angiographic occurrence and clinical correlates of intraluminal coronary artery thrombus: role of unstable angina. J Am Coli Cardiol 1985;6:285-9. 10. Capone G, Wolf NM, Meyer B, Meister SG. Frequency of intracoronary filling defects by angiography in angina pectoris at rest. Am J Cardiol 1985;56:403-6.

II. Shennan CT, Litvack F, Grundfest W, et al. Fiberoptic coronary angioscopy Identifies thrombus in all patients with unstable angina (abstr). Circulation 1985;72(suppl 1II):III-1l2. 12. Levin DC, Fallon JT. Significance of the angiographic morphology

1402

AMBROSE AND HJEMDAHL-MONSEN EDITORIAL

of localized coronary stenosis. Histopathologic correlations. Circulation 1982;66:316-20. 13. Falk E. Unstable angina with fatal outcome: dynamic coronary thrombus leading to infarction and or sudden death. Circulation 1985;71: 699-708. 14. Farhi Jl, Cohen M, Fuster V. The broad spectrum of unstable angina pectoris and its implications for future controlled trials. Am J Cardiol 1986;58:547-50. 15. Plotnick G. Unstable angina: a clinical approach. New York: Futura, 1985:11-17. 16. Maseri A. Pathogenetic classifications of unstable angina as a guideline to individual patient management and prognosis. Am J Med 1986;80(suppl 4C):48-55. 17. Plotnick GD, Greene HL, Carliner NH, Becker LC, Fisher ML. Clinical indicators of left main coronary artery disease in unstable angina. Ann Intern Med 1979;91:149-53. 18. Alison HW, Russell RO Jr, Mantle JA, Kouchoukos NT, Moraski RE, Rackleu CEo Coronary anatomy and arteriography in patients with unstable angina pectoris. Am J Cardiol 1978;41:204-9. 19. Unstable Angina Pectoris Group. Unstable angina pectoris: national cooperative study group to compare medical and surgical therapy. Am J Cardiol 1976;37:896-902.

JACC Vol. 9, No.6 June 1987: 1397--402

dynamics during spontaneous angina pectoris: comparison between angina with ST segment depression and angina with ST segment elevation. Br Heart J 1975;37:401-13. 31. Figueras J, Singh BN, Phil D, Ganz W, Charuzi Y, Swan HJe. Mechanism of rest and nocturnal angina: observations during continuous hemodynamic and electrocardiographic monitoring. Circulation 1979;59:955-68. 32. Berndt TB, Fitzgerald J, Harrison DC, et al. Hemodynamic changes at the onset of spontaneous versus pacing induced angina. Am J Cardiol 1977;39:784-8. 33. Smitherman TC, Hillert MC Jr, Narahara KA, et al. Evidence for transient limitations in coronary blood flow during unstable angina: hemodynamic changes with spontaneous pain at rest versus exerciseinduced ischemia following stabilizationof angina. Clin Cardiol 1980;3: 309-16. 34. Gould KL. Dynamic coronary stenosis. Am J Cardiol 1980;45:286-92. 35. Maseri A, Mimmo R, Chierchia S, et al. Coronary artery spasm as a cause of acute myocardial ischemia in man. Chest 1975;68:625-33. 36. Gorlin R. Dynamic factors in the genesis of myocardial ischemia. J Am Coil Cardiol 1983;1:897-906. 37. Vanhoutte PM, Houston DS. Platelets, endothelium and vasospasm. Circulation 1985;72:728-34.

20. Victor MF, Likoff MJ, Mintz GS, Likoff W. Unstable angina pectoris of new onset: a propective clinical and arteriographic study of 75 patients. Am J Cardiol 1981;47:228-32.

38. Berk BC, Alexander W, Brock TA, et al. Vasoconstriction: a new activity for platelet-derived growth factor. Science 1986;232:87-9.

21. Plotnick GD. Unstable angina: what the symptoms tell you. J Cardiovasc Med 1981;6:499-508.

39. Lam JYT, Chesebro JH, Steele PM, Fuster V. Is vasospasm related to platelet deposition? In vivo relationship in a pig model of arterial injury. Circulation 1987;75:243-8.

22. Mercurio P, Kronzon I, Winer H. Spasm of a normal or minimally narrowed coronary artery in the presence of severe fixed stenoses of the remaining vessels, clinical and angiographic observations. Circulation 1982;65:825-30.

40. Folts JD, Gallagher K, Rowe GG. Blood flow reductions in stenosed canine coronary arteries: vasospasm or platelet aggregation? Circulation 1982;65:248-55.

23. Fuster V, Frye RL, Connolly DC, Danielson MA, Elveback LR, Kurkland LT. Arteriographic patterns early in the onset of the coronary syndromes. Br Heart J 1975;37:1250-5.

41. Folts JD, Crowell EB, Rowe GG. Platelet aggregation in partially obstructed vessels and its elimination with aspirin. Circulation 1976;54: 365-70.

24. Raffenbeul W, Smith LR, Rogers WJ, et al. Quantitative coronary arteriography coronary anatomy of patients without unstable angina pectoris re-examined one year after optimal medical therapy. Am J Cardiol 1979;43:699-707.

42. Bush LR, Campbell WB, Kern K, et al. The effects of alpha 2adrenergic and serotonergic receptor antagonists on cyclic blood flow alterations in stenosed canine coronary arteries. Circ Res 1984;55: 642-52.

2S. McMahon MM, Brown BG, Cukingnan R, et al. Quantitative coronary angiography: measurement of the "critical" stenosis in patients with unstable angina and single vessel disease without collaterals. Circulation 1979;60:106-13.

43. Bush LR, Campbell WB, Buja LM, et al. Effects of the selective thromboxane synthetase inhibitor dazoxiben on variations in cyclic blood flow in stenosed canine coronary arteries. Circulation 1984;69: 1161-70.

26. Ambrose JA, Winters SL, Arora RR, et al. Coronary angiographic morphology in myocardial infarction: a link between the pathogensis of unstable angina and myocardial infarction. J Am Coil Cardiol 1985;6: 1233-8.

44. Freudenberg H, Lichtlen PRo The normal wall segment in coronary stenosis-a postmortem study. Z Kardiol 1981;70:863-9. 45. Brown BG, Bolson EL, Dodge HT. Dynamic mechanisms in human coronary stenosis. Circulation 1984;70:917-22.

27. Ambrose JA, Monsen C, Barrico S, et al. Coronary morphology demonstrates a common link between unstable angina and non-Q wave infarction (abstr). J Am Coil Cardiol 1987;9:196A.

46. Ginsberg R, Bristow MR, Davis K, Dibiase A, Billingham ME. Quantitative pharmacologic responses of normal and atherosclerotic isolated human epicardial coronary arteries. Circulation 1984;69:430-40.

28. Young DF, Tsai FY. How characteristics in models of arterial stenoses. I. Steady flow. J Biomech 1973;6:395-410.

47. Lewis HD, Davis JW, Archibald DG, et al. Protective effects of aspirin against acute myocardial infarction and death in men with unstable angina. Results of a Veteran's Administration cooperative study. N Engl J Med 1983;309:396-403.

29. Fischl SJ, Herman MV, Gorlin, R. The intermediate coronary syndrome. Clinical, angiographic and therapeutic aspects. N Engl J Med 1973;288:1193-8. 30. Guazzi M, Polese A, Fiorentini C, et al. Left and right heart hemo-

48. Fuster V, Chesebro JW. Mechanisms of unstable angina. N Engl J Med 1986;315:1023-5.