Role of Nitrates
in Unstable
Angina Pectoris
John D. Horowitz
Unstable angma pectoris is a clinically heterogeneous process with patient symptoms varying between reduced threshold for exertional angina and the occuwence of multiple episodes of rest paln. The major factors in the pathogenesis of unstable angina appear to be intracoronary platelet aggregation and thrombus formation secondary to fissuring or rupture of atheromatous plaques, with associated coronary vasoconstrlctiin due to release of constrictor materials from aggregating platelets and defiwcy of endothelium-related vasodilator actii. The latter factor is of particular interest in view of the slmilar biochemical mechanisms of actiin of nitroglycerin (NTG) and endothelium-derived relaxing factor (EDRF). The efficacy of NTG In limiting platelet aggregation is also of particular interest In thii condition. Medbl therapy In patients with unstable angina usually requires use of muttiple agents. In the short term, there is a strong case for the use of intravenous heparin both to relieve paln and to reduce the risk of acute myocardiil infarction. Aspirin is perhaps less effective in the short term, but very useful in long-term treatment of such patients. Despite their wklespread cllnical use, (3.adrenoceptor antagonists are probably only marginally beneficial, whereas dihydropyridine calcium antagonists such as nifedipine are potentiilly harmful as monotherapy and of questionable use in combination with other drugs. Other agents that are effectiie in relieving ischemR symptoms are the nondihydropyridine calcium antagonists verapamil and dlltiiem and the oxygen-sparing agent perhexiline maleate. Despite a paucity of controlled trial data, nitrates are used in the vast majority of patients with unstable alrgina. In patients with severe symptoms, intravenous iniusion is preferable, and because of the need for 24-hour prevention From the Cardiology Unit, The Queen Elizabeth Hospital, University of Adelaide, Woodville, Australia. Supported in part by grants from the National Health and Medical Research Council of Australia. Address for reprints: John D. Horowitz, Cardiology Unit, The Queen Elizabeth Hospital, University of Adelaide, Woodville, SA 5011, Australia.
64B
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
of ischemia, the development of nitrate tolerance is a major risk that cannot be reduced by intermittent nitrate admlnistration. There is strong evidence that hrfuslon of NTG at rates MO pg/ min induces the development of extensive hemodynamic tolerance, affecUng both the systeml and coronary circulatii, in most patients. NTG infusion rates shoukl therefore be kept as low as possible in such patients. An atternatiie approach is the combined infusion of NTG and Nacetyk?ysteine (NM) in patients with “high risk” unstable angina. Thls combination may augment the antieory effects of NTG and prevent nitrate tolerance. In a controlled clinical trial, incidence of acute myocardiil infarction was slgnlficantly lawer with NTG/NAC than with NT0 alone. Continuous rather than intermittent adminlstration of NAC is preferable to avoid development of symptomatii hypotension. (Am J CardiillGG2;7O:G4S718)
T
he role of nitrate therapy in unstable angina pectoris must be examined in relation to somewhat heterogeneous manifestations of a disease process the nature of which remains incompletely understood at present. Current definitions of unstable angina pectoriG2 encompass a range of clinical syndromes extending from acceleration of preexistent exertional angina to the occurrence of recurrent prolonged episodes of ischemic pain at rest. The implications of prior myocardial infarction and of progression in the severity of symptoms also complicate categorization of individual patients.2 Because of this heterogeneity, results of uncontrolled studies of patients with unstable angina pectoris are particularly difficult to interpret. Various investigations of natural history have, however, succeeded in partially characterizing subgroups of patients both at relatively low3 and at high& risk for the occurrence of acute myocardial infarction against a background of unstable angina pectoris. Persistent occurrence of chest pain at rest, associated marked ST-segment depression on electroenSEPTEMBER
24, 1992
cephalography and development of transient hemodynamic abnormalities associated with &hernia are all markers of increased risk.“-6 UNSTABLE ANGINA PECTORIS: AND CUNICAL PRESENTATION
PATHOGENESIS
Recent investigations have delineated the pivotal role of coronary thrombosis in the pathogenesis of unstable angina pectoris7-lo Evidence in favor of the occurrence of intracoronary thrombi has been obtained both via results of coronary angioscopy7 and coronary angiography.9 Further, evidence of platelet activation as an antecedent of coronary thrombosis may be obtained via studies demonstrating increased generation of thromboxane in patients with unstable angina.‘OJl Utilizing the optimal form of investigation, involving specific examination of transcoronary thromboxane concentration gradients in patients with ischemic heart disease, Hirsh et all’ demonstrated significant elevation of thromboxane A2 in the coronary sinus only in patients in whom spontaneous ischemic pain had occurred in the preceding 24 hours. Less information is available regarding the nature of events initiating coronary thrombosis in such patients. However, pathologic studies suggest that fissuring or rupture of atheromatous plaques initiates the process of platelet aggregation via exposure of circulating blood to proaggregatory materials, such as collagen and fatty acids.8J2 Similarly, clinical evidence supports a relation between “complex” coronary stenosis morphology-defined on the basis of irregularity and/or ulceration of the relevant stenosis-and unfavorable clinical course in patients with unstable angina pectoris.9 If it is assumed that the major events initiating the clinical presentation of unstable angina pectoris are plaque fissuring or rupture, followed by platelet aggregation leading to thrombus formation, therapeutic interventions in this condition should be directed toward limiting and reversing this process, with the particular objective of preventing thrombus progression to the stage of total coronary occlusion and associated development of infarction. However, it is likely that other factors are also involved. First, platelets from patients with previous myocardial infarction are hyperaggregable in response to adenosine diphosphate (ADP), raising the possibility that a primary abnormality of platelet function may play a more important role.13 Further, the issue of potential reactivity of the coronaryvasculature to the effects of both constrictor and dilator agents remains of fundamental significance. It is possible that in some cases the
initiating event in patients with unstable angina pectoris and also myocardial infarction14 is coronary vasospasm associated with abnormalities of coronary endothelial function; demonstrations of abnormal constrictor reactivity of coronary arteries to adjacent atheromatous plaque@ provide some support for this concept, which would possibly invoke thrombus formation as a “secondary” event. Many of the intervention studies carried out during the period 1975-1980 in patients with unstable angina pectoris were significantly influenced by the potentially “primary” role of coronary vasospasm. On the other hand, it seems more likely that in the majority of cases, changes in coronary vasomotor tone must be viewed as being largely secondary to platelet aggregation and incipient thrombus formation. This in no way diminishes the potentiial role of disordered endothelial function as a modulating factor in patients with unstable angina pectoris. Furthermore, there is considerable evidence that production of endothelium-derived relaxing factor (EDRF) is diminished in association with coronary atherogenesis. Is-r7 This is of particular interest in view of the identification of EDRF as either free nitric oxide radical (NO-) or a chemically bound form of NO, such as an S-nitrosothiol,18J9 thus establishing a close link between EDRF and the bioconversion products of nitroglycerin (NTG) and other nitrovasodilators.20,21 It is also possible that reduced availability of EDRF in patients with unstable angina pectoris may increase susceptibility of coronary vascular smooth muscle not only to the constrictor effects of thromboxane A2 or 5-hydroxytryptamine (serotonin) released from aggregating platelets, but also to those of endothelin, a vasoconstrictor peptide that is synthesized by endothelial cells in response to exposure to thrombin. There is some evidence to suggest that NO may both inhibit the release of endothelin2* and reverse endothelin-induced vasconstriction.23 Current evidence therefore favors the existence of multiple and interdependent factors affecting the process of intracoronary thrombus formation in patients with unstable angina pectoris (Figure 1). The potential for complex interactions between endothelium, coronary vasculature, and aggregating platelets% suggests that many forms of therapeutic intervention may be beneficial, but that an optimal approach to the management of this condition may involve inhibition of several steps in the process of thrombus formation. A SYMPOSIUM:
NITROGLYCERIN
THERAPY
656
INTERVENTION S’mrDIES IN UNSlARti ANGINA PECTORIS: CONSIDERATIONS OF TRlAL DESIGN
AND ENDPOINTS
Given the heterogeneous nature of patients with unstable angina pectoris, appropriate endpoints and optimal sample size for trial design will depend greatly on the precise criteria for admission to the study. For example, selection of patients at relatively high risk for the occurrence of infarction will facilitate use of the incidence of infarction as an endpoint, but this is likely to be complicated by the need for early intervention with coronary angioplasty or surgery in a large proportion of these patients. Possible endpoints (summarized in Table I) also include examination of thrombus morphology in the affected coronary artery as well as assessment of the extent and consequences of resultant myocardial ischemia. Since there is now considerable evidence demonstrating the occurrence of silent myocardial ischemia in this group of patients, the issue arises whether determination of total frequency of ischemia may permit more accurate prediction of the risk of infarction. However, it appears from a recent study25 that detection of asymptomatic ischemia is a relatively poor predictor of serious cardiac events in this group of patients. This is controversial, as other data suggest that persisting or protracted silent ischemia is an adverse prognostic factor. On the other hand, studies requiring a mortality endpoint in patients presenting with unstable symptoms are also problematic, not only by a high frequency of intervention (with a consequent bias toward inclusion of low-risk patients), but also by the need to enroll very large numbers of patients. PLAQUE
FISSURE
I RUPTURE
ENDOTHELIUM
PLATELETS
/ ENDOTHELIN DEFICIENT
fDL2CtW&iGEN
J
EDRF TiRdYBd
\
I PLATELET AGGREGATION
I VASOCONSTRICTION I THROYBUS
FORMATION
Fl6URE 1. Algorithm of the factors likely to infkmnce tbrombus fonnatlon In patients with unstable anglna pactoris. ADP = adenoslne diphospbate; EDRF = endoths lium-darhfad rekudng factor; IliT = sewotonin; m = thromboxane AZ. 666
TABLE Unstable
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
l
I Possible Angina Resolution
Endpoints Pectoris
in Studies
of intracoronary
thrombosis
0 Prevention
of total ischemia-symptomatic
0 Prevention
of symptomatik
l
Prevention
of acute
l
Prevention
of mortality
niyocardial
Involving
Patients
with
and asymptomatic
ischemia infarction
In many of the larger relevant studies, combined endpoints (e.g., recurrent ischemia/infarction; infarction/death) have been utilized to limit this problem. CUNICAL EFFICACY OF PHARMACOTHERAPY IN UNSTABLE ANGINA PECTORtS Studies involving drug therapy in patients with unstable angina may be classified as short-term (generally involving enrollment of patients with severe and frequent symptoms, often awaiting interventions) or long-term (generally involving patients whose initial symptoms have settled during initial hospitalization). Studies involving agents that limit thrombus formation or progression have fallen into both categories (Table II). Among patients with severe ischemic symptoms, the use of continuous intravenous infusion of heparin has been associated with both relief of chest pain and reduction in the incidence of acute myocardial infarction.26 In this setting, heparin appeared to be more effective than aspirin,26T27 intermittently administered heparin,” alteplase,27 or atenolol.28 On the other hand, aspirin alone significantly reduced the incidence of acute myocardial infarction, whereas the combination of heparin and aspirin appeared to confer no incremental therapeutic effect.26 Most physicians use aspirin both acutely and long-term, as well as heparin in the early hospital setting. On the other hand, 2 well-designed large longterm studies2990 have demonstrated efficacy of aspirin therapy in reducing incidence of acute myocardial infarction and cardiac death. Similar effects were observed with ticlopidine,31 another inhibitor of platelet aggregation. Although all of these studies involved patients with a low incidence of ongoing angina1 symptoms requiring angioplasty or coronary surgery, their results establish clearly a valuable therapeutic role for inhibitors of platelet aggregation when medical management of such patients is undertaken (Table II). A number of short-term studies28,32 have examined the effects of various p-adrenoceptor antagonists in patients with severe unstable angina pectoris. These tend to suggest that, despite the SEPTEMBER
24,
1992
TABLE II Some Short- and Long-Term Drugs Other Than Nitrates Reference
Drugs Examined
32
Nifedipine Metoprolol
33
Verapamil Propranolol
26
Aspirin Heparin
28
Heparin Atenolol
29
Aspirin
31
Ticlopidine
30
Aspirin Sulfinpyrazone
Placebo-Controlled
No. 338
Intervention
Studies
Duration of Study (mean)
in Patients
with Unstable
Angina
Pectoris,
Major Endpoints
Results
2 days
Recurrent Infarction
4 days
Angina1
479
6 days
Refractory Infarction
214
7 days
Infarction
1. Heparin 2. Atenolol
12 weeks
Infarction Death
Approximately 50% both endpoints
18
1,266 652 555
6 months 18 months
widespread clinical use of these agents, P-adrenoceptor antagonists are relatively ineffective. The most extensive investigation to date involved the use of metoprolol (and nifedipine) in the Holland Interuniversity Nifedipine/Metoprolol Trial (HINT),32 the results of which suggest a possible beneficial effect for metoprolol regarding the endpoint of recurrent ischemia/infarction. However, this trend did not reach statistical significance. The HINT study32 is also the most extensive investigation to date of the effects of the dihydropyridine calcium antagonist nifedipine in pa.tients with unstable angina pectoris. The results of this study raise major concerns about the use of nifedipine as a primary mode of therapy in this group of patients in view of a strong trend toward an increased risk of recurrent ischemia. A meta-analysis of the effects of various treatments on risk of progression of unstable angina to acute myocardial infarction33 suggested that overall calcium antagonists reduce risk insignificantly (95% confidence limits for change, +32% to -30%) and P-adrenoceptor antagonists reduce risk by approximately 13%, compared with 40-44% reductions for aspirin and heparin, respectively. However, such comparisons are hazardous for 2 major reasons. First, the majority of the patients in studies involving use of aspirin, rather than those treated with other agents, were followed in the long term. Further, analysis of studies involving calcium antagonists should take into consideration
Involving
ischemia
1. Nifedipine monotherapy possibly harmful 2. Metoprolol possibly beneficial 3. Nifedipine beneficial with long-standing p-blocker therapy
frequency
Verapamil more effective than propranolol Propranolol similar to placebo
angina
1, Heparin most effective in relieving angina 2. Both aspirin and heparin prevented infarction
Vascular mortality Acute myocardial
prevented ineffective
Both reduced
infarction reduction
in
40-50%
infarction
Infarction Death
Aspirin effective Sulfinpyrazone ineffective
the major differences in effects between the dihydropyridines (such as nifedipine) and agents related to the nondihydropyridine agents such as verapamil and diltiazem. To date, the only studies carried out involving such agents in patients with severe unstable angina pectoris have been relatively small34a35 but have suggested clinical efficacy with regard to the endpoint of suppression of ischemic symptoms. It appears that verapamil, diltiazem, and related drugs may be less prone than nifedipine to induce aggravation of ischemia3’j; the favorable results of recent investigations involving the role of verapamiP7 and diltiazem38 in secondary prevention after uncomplicated myocardial infarction raises the possibility that beneficial effects on prognosis may extend to patients with unstable angina pectoris. In the interim, the frequent utilizaition of verapamil and diltiazem in patients with unstable angina is supported primarily by their greater efficacy on progression of symptoms than that of adrenoceptor antagonists.34 Combination therapy with a l3 blocker and calcium antagonist is common in this setting. A further potential approach to the management of this group of patients is the utilization of agents that reduce myocardial oxygen consumption without exerting negative inotropic effects or causing significiant coronary vasomotor effects. Perhexiline maleate, one such agent, may be utilized successfully in patients with unstable angina,39,40 but, because of unfavorable absorption pharmacoA SYMPOSIUM:
NITROGLYCERIN
THERAPY
671)
TABLE Nitrates
III
Potentially in Patients
Important with Unstable
Mechanism
Issues Regarding Angina Pectoris
Potential
for induction
Effects of platelet Necessity
critically
stenosed
of coronary
steal
aggregation
for 24.hour
arteries
and disaggregation
effect
Prevention
(and detection)
Prevention
of “rebound”
of tolerance on withdrawal
of administration
Intravenous
vs. oral/topical
Appropriate
dosage
Potential
to minimize
adverse
effects
interactions
Heparin
(with
Calcium
antagonists
Titration
of
of effect
Effects on flow through
Route
Utilization
high dose intravenous
nitroglycerin)
of dosage-basis
kinetics, it tends to be relatively first 1-2 days of therapy.
ineffective
for the
NITRATE UTILIZATION IN UNSTABLE ANGINA: FUNDAMENTAL ISSUES AND CLINICAL DATA Although administration of nitrates to patients with angina has been part of routine clinical practice for many years, virtually no controlled clinical trials have been carried out involving such therapy. One of the major difficulties in performing such studies would be withholding acutely administered nitrates for relief of episodes of ischemia. Further, the common utilization of multiple modes of therapy causes difficulty in analysis of results, particularly in view of potential interactions between nitrates and calcium antagonists4r or heparin.42 Nitrate therapy in unstable angina should theoretically provide 24-hour prophylaxis against ischemia and reduce the risk of acute myocardial infarction. A number of potentially important issues arise, however (Table III). Of particular concern is the risk of nitrate tolerance, which increases with both duration of therapy and serum nitrate concentration.43 Further, since patients with unstable angina may be at particular risk for severe aggravation of ischemic symptoms following withdrawal of nitrate therapy, the use of a nitrate-free period to limit tolerance is relatively impractical in this group. The potential importance of nitrate tolerance in patients with unstable angina extends not only to the vasodilator actions of nitrates, but also to their effects on platelet aggregation.44 Insufficient data are available at present to address this major issue. 666
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
The results of clinical studies of nitrate therapy in patients with unstable angina (summarized in Table IV) generally have involved initial utilization of intravenous NTG, with eventual transition to other modes of therapy, including oral and transderma1 nitrates. A randomized study by Curfman et a145compared the effects of intravenous NTG with those of oral isosorbide dinitrate/topical NTG ointment in 40 patients. Neither treatment regimen was optimal with regard to prevention of nitrate tolerance. Although control of ischemic symptoms was more consistent with intravenous NTG, results of the study were not conclusive. Other studies involving the use of intravenous NTG have varied considerably with regard to infusion rates and the utilization of nonadsorbent polyethylene tubing in infusion sets to ensure predictable delivery of NTG.46-48 Although these factors and the lack of a randomized control group limit interpretation of the results of these studies, it appears that intravenous nitrate therapy was generally associated with amelioration of symptomatic status. On the other hand, there is increasing evidence, despite long-standing clinical practice to the contrary, that infusion rates of intravenous NTG should be limited wherever possible to
in Unstable
Angina Pectoris
John D. Horowitz
Unstable angma pectoris is a clinically heterogeneous process with patient symptoms varying between reduced threshold for exertional angina and the occuwence of multiple episodes of rest paln. The major factors in the pathogenesis of unstable angina appear to be intracoronary platelet aggregation and thrombus formation secondary to fissuring or rupture of atheromatous plaques, with associated coronary vasoconstrlctiin due to release of constrictor materials from aggregating platelets and defiwcy of endothelium-related vasodilator actii. The latter factor is of particular interest in view of the slmilar biochemical mechanisms of actiin of nitroglycerin (NTG) and endothelium-derived relaxing factor (EDRF). The efficacy of NTG In limiting platelet aggregation is also of particular interest In thii condition. Medbl therapy In patients with unstable angina usually requires use of muttiple agents. In the short term, there is a strong case for the use of intravenous heparin both to relieve paln and to reduce the risk of acute myocardiil infarction. Aspirin is perhaps less effective in the short term, but very useful in long-term treatment of such patients. Despite their wklespread cllnical use, (3.adrenoceptor antagonists are probably only marginally beneficial, whereas dihydropyridine calcium antagonists such as nifedipine are potentiilly harmful as monotherapy and of questionable use in combination with other drugs. Other agents that are effectiie in relieving ischemR symptoms are the nondihydropyridine calcium antagonists verapamil and dlltiiem and the oxygen-sparing agent perhexiline maleate. Despite a paucity of controlled trial data, nitrates are used in the vast majority of patients with unstable alrgina. In patients with severe symptoms, intravenous iniusion is preferable, and because of the need for 24-hour prevention From the Cardiology Unit, The Queen Elizabeth Hospital, University of Adelaide, Woodville, Australia. Supported in part by grants from the National Health and Medical Research Council of Australia. Address for reprints: John D. Horowitz, Cardiology Unit, The Queen Elizabeth Hospital, University of Adelaide, Woodville, SA 5011, Australia.
64B
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
of ischemia, the development of nitrate tolerance is a major risk that cannot be reduced by intermittent nitrate admlnistration. There is strong evidence that hrfuslon of NTG at rates MO pg/ min induces the development of extensive hemodynamic tolerance, affecUng both the systeml and coronary circulatii, in most patients. NTG infusion rates shoukl therefore be kept as low as possible in such patients. An atternatiie approach is the combined infusion of NTG and Nacetyk?ysteine (NM) in patients with “high risk” unstable angina. Thls combination may augment the antieory effects of NTG and prevent nitrate tolerance. In a controlled clinical trial, incidence of acute myocardiil infarction was slgnlficantly lawer with NTG/NAC than with NT0 alone. Continuous rather than intermittent adminlstration of NAC is preferable to avoid development of symptomatii hypotension. (Am J CardiillGG2;7O:G4S718)
T
he role of nitrate therapy in unstable angina pectoris must be examined in relation to somewhat heterogeneous manifestations of a disease process the nature of which remains incompletely understood at present. Current definitions of unstable angina pectoriG2 encompass a range of clinical syndromes extending from acceleration of preexistent exertional angina to the occurrence of recurrent prolonged episodes of ischemic pain at rest. The implications of prior myocardial infarction and of progression in the severity of symptoms also complicate categorization of individual patients.2 Because of this heterogeneity, results of uncontrolled studies of patients with unstable angina pectoris are particularly difficult to interpret. Various investigations of natural history have, however, succeeded in partially characterizing subgroups of patients both at relatively low3 and at high& risk for the occurrence of acute myocardial infarction against a background of unstable angina pectoris. Persistent occurrence of chest pain at rest, associated marked ST-segment depression on electroenSEPTEMBER
24, 1992
cephalography and development of transient hemodynamic abnormalities associated with &hernia are all markers of increased risk.“-6 UNSTABLE ANGINA PECTORIS: AND CUNICAL PRESENTATION
PATHOGENESIS
Recent investigations have delineated the pivotal role of coronary thrombosis in the pathogenesis of unstable angina pectoris7-lo Evidence in favor of the occurrence of intracoronary thrombi has been obtained both via results of coronary angioscopy7 and coronary angiography.9 Further, evidence of platelet activation as an antecedent of coronary thrombosis may be obtained via studies demonstrating increased generation of thromboxane in patients with unstable angina.‘OJl Utilizing the optimal form of investigation, involving specific examination of transcoronary thromboxane concentration gradients in patients with ischemic heart disease, Hirsh et all’ demonstrated significant elevation of thromboxane A2 in the coronary sinus only in patients in whom spontaneous ischemic pain had occurred in the preceding 24 hours. Less information is available regarding the nature of events initiating coronary thrombosis in such patients. However, pathologic studies suggest that fissuring or rupture of atheromatous plaques initiates the process of platelet aggregation via exposure of circulating blood to proaggregatory materials, such as collagen and fatty acids.8J2 Similarly, clinical evidence supports a relation between “complex” coronary stenosis morphology-defined on the basis of irregularity and/or ulceration of the relevant stenosis-and unfavorable clinical course in patients with unstable angina pectoris.9 If it is assumed that the major events initiating the clinical presentation of unstable angina pectoris are plaque fissuring or rupture, followed by platelet aggregation leading to thrombus formation, therapeutic interventions in this condition should be directed toward limiting and reversing this process, with the particular objective of preventing thrombus progression to the stage of total coronary occlusion and associated development of infarction. However, it is likely that other factors are also involved. First, platelets from patients with previous myocardial infarction are hyperaggregable in response to adenosine diphosphate (ADP), raising the possibility that a primary abnormality of platelet function may play a more important role.13 Further, the issue of potential reactivity of the coronaryvasculature to the effects of both constrictor and dilator agents remains of fundamental significance. It is possible that in some cases the
initiating event in patients with unstable angina pectoris and also myocardial infarction14 is coronary vasospasm associated with abnormalities of coronary endothelial function; demonstrations of abnormal constrictor reactivity of coronary arteries to adjacent atheromatous plaque@ provide some support for this concept, which would possibly invoke thrombus formation as a “secondary” event. Many of the intervention studies carried out during the period 1975-1980 in patients with unstable angina pectoris were significantly influenced by the potentially “primary” role of coronary vasospasm. On the other hand, it seems more likely that in the majority of cases, changes in coronary vasomotor tone must be viewed as being largely secondary to platelet aggregation and incipient thrombus formation. This in no way diminishes the potentiial role of disordered endothelial function as a modulating factor in patients with unstable angina pectoris. Furthermore, there is considerable evidence that production of endothelium-derived relaxing factor (EDRF) is diminished in association with coronary atherogenesis. Is-r7 This is of particular interest in view of the identification of EDRF as either free nitric oxide radical (NO-) or a chemically bound form of NO, such as an S-nitrosothiol,18J9 thus establishing a close link between EDRF and the bioconversion products of nitroglycerin (NTG) and other nitrovasodilators.20,21 It is also possible that reduced availability of EDRF in patients with unstable angina pectoris may increase susceptibility of coronary vascular smooth muscle not only to the constrictor effects of thromboxane A2 or 5-hydroxytryptamine (serotonin) released from aggregating platelets, but also to those of endothelin, a vasoconstrictor peptide that is synthesized by endothelial cells in response to exposure to thrombin. There is some evidence to suggest that NO may both inhibit the release of endothelin2* and reverse endothelin-induced vasconstriction.23 Current evidence therefore favors the existence of multiple and interdependent factors affecting the process of intracoronary thrombus formation in patients with unstable angina pectoris (Figure 1). The potential for complex interactions between endothelium, coronary vasculature, and aggregating platelets% suggests that many forms of therapeutic intervention may be beneficial, but that an optimal approach to the management of this condition may involve inhibition of several steps in the process of thrombus formation. A SYMPOSIUM:
NITROGLYCERIN
THERAPY
656
INTERVENTION S’mrDIES IN UNSlARti ANGINA PECTORIS: CONSIDERATIONS OF TRlAL DESIGN
AND ENDPOINTS
Given the heterogeneous nature of patients with unstable angina pectoris, appropriate endpoints and optimal sample size for trial design will depend greatly on the precise criteria for admission to the study. For example, selection of patients at relatively high risk for the occurrence of infarction will facilitate use of the incidence of infarction as an endpoint, but this is likely to be complicated by the need for early intervention with coronary angioplasty or surgery in a large proportion of these patients. Possible endpoints (summarized in Table I) also include examination of thrombus morphology in the affected coronary artery as well as assessment of the extent and consequences of resultant myocardial ischemia. Since there is now considerable evidence demonstrating the occurrence of silent myocardial ischemia in this group of patients, the issue arises whether determination of total frequency of ischemia may permit more accurate prediction of the risk of infarction. However, it appears from a recent study25 that detection of asymptomatic ischemia is a relatively poor predictor of serious cardiac events in this group of patients. This is controversial, as other data suggest that persisting or protracted silent ischemia is an adverse prognostic factor. On the other hand, studies requiring a mortality endpoint in patients presenting with unstable symptoms are also problematic, not only by a high frequency of intervention (with a consequent bias toward inclusion of low-risk patients), but also by the need to enroll very large numbers of patients. PLAQUE
FISSURE
I RUPTURE
ENDOTHELIUM
PLATELETS
/ ENDOTHELIN DEFICIENT
fDL2CtW&iGEN
J
EDRF TiRdYBd
\
I PLATELET AGGREGATION
I VASOCONSTRICTION I THROYBUS
FORMATION
Fl6URE 1. Algorithm of the factors likely to infkmnce tbrombus fonnatlon In patients with unstable anglna pactoris. ADP = adenoslne diphospbate; EDRF = endoths lium-darhfad rekudng factor; IliT = sewotonin; m = thromboxane AZ. 666
TABLE Unstable
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
l
I Possible Angina Resolution
Endpoints Pectoris
in Studies
of intracoronary
thrombosis
0 Prevention
of total ischemia-symptomatic
0 Prevention
of symptomatik
l
Prevention
of acute
l
Prevention
of mortality
niyocardial
Involving
Patients
with
and asymptomatic
ischemia infarction
In many of the larger relevant studies, combined endpoints (e.g., recurrent ischemia/infarction; infarction/death) have been utilized to limit this problem. CUNICAL EFFICACY OF PHARMACOTHERAPY IN UNSTABLE ANGINA PECTORtS Studies involving drug therapy in patients with unstable angina may be classified as short-term (generally involving enrollment of patients with severe and frequent symptoms, often awaiting interventions) or long-term (generally involving patients whose initial symptoms have settled during initial hospitalization). Studies involving agents that limit thrombus formation or progression have fallen into both categories (Table II). Among patients with severe ischemic symptoms, the use of continuous intravenous infusion of heparin has been associated with both relief of chest pain and reduction in the incidence of acute myocardial infarction.26 In this setting, heparin appeared to be more effective than aspirin,26T27 intermittently administered heparin,” alteplase,27 or atenolol.28 On the other hand, aspirin alone significantly reduced the incidence of acute myocardial infarction, whereas the combination of heparin and aspirin appeared to confer no incremental therapeutic effect.26 Most physicians use aspirin both acutely and long-term, as well as heparin in the early hospital setting. On the other hand, 2 well-designed large longterm studies2990 have demonstrated efficacy of aspirin therapy in reducing incidence of acute myocardial infarction and cardiac death. Similar effects were observed with ticlopidine,31 another inhibitor of platelet aggregation. Although all of these studies involved patients with a low incidence of ongoing angina1 symptoms requiring angioplasty or coronary surgery, their results establish clearly a valuable therapeutic role for inhibitors of platelet aggregation when medical management of such patients is undertaken (Table II). A number of short-term studies28,32 have examined the effects of various p-adrenoceptor antagonists in patients with severe unstable angina pectoris. These tend to suggest that, despite the SEPTEMBER
24,
1992
TABLE II Some Short- and Long-Term Drugs Other Than Nitrates Reference
Drugs Examined
32
Nifedipine Metoprolol
33
Verapamil Propranolol
26
Aspirin Heparin
28
Heparin Atenolol
29
Aspirin
31
Ticlopidine
30
Aspirin Sulfinpyrazone
Placebo-Controlled
No. 338
Intervention
Studies
Duration of Study (mean)
in Patients
with Unstable
Angina
Pectoris,
Major Endpoints
Results
2 days
Recurrent Infarction
4 days
Angina1
479
6 days
Refractory Infarction
214
7 days
Infarction
1. Heparin 2. Atenolol
12 weeks
Infarction Death
Approximately 50% both endpoints
18
1,266 652 555
6 months 18 months
widespread clinical use of these agents, P-adrenoceptor antagonists are relatively ineffective. The most extensive investigation to date involved the use of metoprolol (and nifedipine) in the Holland Interuniversity Nifedipine/Metoprolol Trial (HINT),32 the results of which suggest a possible beneficial effect for metoprolol regarding the endpoint of recurrent ischemia/infarction. However, this trend did not reach statistical significance. The HINT study32 is also the most extensive investigation to date of the effects of the dihydropyridine calcium antagonist nifedipine in pa.tients with unstable angina pectoris. The results of this study raise major concerns about the use of nifedipine as a primary mode of therapy in this group of patients in view of a strong trend toward an increased risk of recurrent ischemia. A meta-analysis of the effects of various treatments on risk of progression of unstable angina to acute myocardial infarction33 suggested that overall calcium antagonists reduce risk insignificantly (95% confidence limits for change, +32% to -30%) and P-adrenoceptor antagonists reduce risk by approximately 13%, compared with 40-44% reductions for aspirin and heparin, respectively. However, such comparisons are hazardous for 2 major reasons. First, the majority of the patients in studies involving use of aspirin, rather than those treated with other agents, were followed in the long term. Further, analysis of studies involving calcium antagonists should take into consideration
Involving
ischemia
1. Nifedipine monotherapy possibly harmful 2. Metoprolol possibly beneficial 3. Nifedipine beneficial with long-standing p-blocker therapy
frequency
Verapamil more effective than propranolol Propranolol similar to placebo
angina
1, Heparin most effective in relieving angina 2. Both aspirin and heparin prevented infarction
Vascular mortality Acute myocardial
prevented ineffective
Both reduced
infarction reduction
in
40-50%
infarction
Infarction Death
Aspirin effective Sulfinpyrazone ineffective
the major differences in effects between the dihydropyridines (such as nifedipine) and agents related to the nondihydropyridine agents such as verapamil and diltiazem. To date, the only studies carried out involving such agents in patients with severe unstable angina pectoris have been relatively small34a35 but have suggested clinical efficacy with regard to the endpoint of suppression of ischemic symptoms. It appears that verapamil, diltiazem, and related drugs may be less prone than nifedipine to induce aggravation of ischemia3’j; the favorable results of recent investigations involving the role of verapamiP7 and diltiazem38 in secondary prevention after uncomplicated myocardial infarction raises the possibility that beneficial effects on prognosis may extend to patients with unstable angina pectoris. In the interim, the frequent utilizaition of verapamil and diltiazem in patients with unstable angina is supported primarily by their greater efficacy on progression of symptoms than that of adrenoceptor antagonists.34 Combination therapy with a l3 blocker and calcium antagonist is common in this setting. A further potential approach to the management of this group of patients is the utilization of agents that reduce myocardial oxygen consumption without exerting negative inotropic effects or causing significiant coronary vasomotor effects. Perhexiline maleate, one such agent, may be utilized successfully in patients with unstable angina,39,40 but, because of unfavorable absorption pharmacoA SYMPOSIUM:
NITROGLYCERIN
THERAPY
671)
TABLE Nitrates
III
Potentially in Patients
Important with Unstable
Mechanism
Issues Regarding Angina Pectoris
Potential
for induction
Effects of platelet Necessity
critically
stenosed
of coronary
steal
aggregation
for 24.hour
arteries
and disaggregation
effect
Prevention
(and detection)
Prevention
of “rebound”
of tolerance on withdrawal
of administration
Intravenous
vs. oral/topical
Appropriate
dosage
Potential
to minimize
adverse
effects
interactions
Heparin
(with
Calcium
antagonists
Titration
of
of effect
Effects on flow through
Route
Utilization
high dose intravenous
nitroglycerin)
of dosage-basis
kinetics, it tends to be relatively first 1-2 days of therapy.
ineffective
for the
NITRATE UTILIZATION IN UNSTABLE ANGINA: FUNDAMENTAL ISSUES AND CLINICAL DATA Although administration of nitrates to patients with angina has been part of routine clinical practice for many years, virtually no controlled clinical trials have been carried out involving such therapy. One of the major difficulties in performing such studies would be withholding acutely administered nitrates for relief of episodes of ischemia. Further, the common utilization of multiple modes of therapy causes difficulty in analysis of results, particularly in view of potential interactions between nitrates and calcium antagonists4r or heparin.42 Nitrate therapy in unstable angina should theoretically provide 24-hour prophylaxis against ischemia and reduce the risk of acute myocardial infarction. A number of potentially important issues arise, however (Table III). Of particular concern is the risk of nitrate tolerance, which increases with both duration of therapy and serum nitrate concentration.43 Further, since patients with unstable angina may be at particular risk for severe aggravation of ischemic symptoms following withdrawal of nitrate therapy, the use of a nitrate-free period to limit tolerance is relatively impractical in this group. The potential importance of nitrate tolerance in patients with unstable angina extends not only to the vasodilator actions of nitrates, but also to their effects on platelet aggregation.44 Insufficient data are available at present to address this major issue. 666
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
The results of clinical studies of nitrate therapy in patients with unstable angina (summarized in Table IV) generally have involved initial utilization of intravenous NTG, with eventual transition to other modes of therapy, including oral and transderma1 nitrates. A randomized study by Curfman et a145compared the effects of intravenous NTG with those of oral isosorbide dinitrate/topical NTG ointment in 40 patients. Neither treatment regimen was optimal with regard to prevention of nitrate tolerance. Although control of ischemic symptoms was more consistent with intravenous NTG, results of the study were not conclusive. Other studies involving the use of intravenous NTG have varied considerably with regard to infusion rates and the utilization of nonadsorbent polyethylene tubing in infusion sets to ensure predictable delivery of NTG.46-48 Although these factors and the lack of a randomized control group limit interpretation of the results of these studies, it appears that intravenous nitrate therapy was generally associated with amelioration of symptomatic status. On the other hand, there is increasing evidence, despite long-standing clinical practice to the contrary, that infusion rates of intravenous NTG should be limited wherever possible to
