Prognosis of patients with stable angina pectoris on antianginal drug therapy

Prognosis Pectoris

of Patients with Stable Angina on Antianginal Drug Therapy

Paul Hiemdahl,

MD, PhD, Sven V. Eriksson,

and

Nina

Rehnqvist,

MD, PhD, Claes MD, PhD

Held,

MD,

Antianginal drug treatment reduces symptoms and ischemia but may also influence the prognosis of patients with stable angina pectoris. The Atenolol Silent lschemia Study (ASIST) compared atenolol and placebo treatment (about 140 patient-years on each) in patients with mainly silent ischemia and found less aggravation of angina and a tendency toward fewer cardiac complications with atenolol treatment. The Total lschaemic Burden European Trial (TIBET) compared slow release nifedipine, atenolol, or the combination (about 450 patienbyears on each) and found no significant differences with regard to cardiac complications, a nonsignificant trend toward better prognosis on combined treatment, and more side effects on nifedipine alone compared with the other treatments. The Angina Prognosis Study in Stockholm (APSIS) compared metoprolol and verapamil (about 1,400 patient-years on each) and found similar effects on cardiovascular endpoints, tolerability, and psychosocial variables with the 2 treatments. Hypothe-

sis-generating subgroup analyses in APSIS suggest that treatment effects may differ in hypertensive and diabetic subgroups. Beneficial effects in primary and secondary prevention, together with data from ASIST, suggest that PI blockade influences prognosis favorably. The safety of short-acting nifedipine in ischemic heart disease is questioned, but TIBETdata suggest that slow release nifedipine may be safe. Verapamil has beneficial effects alter myocardial infarction (Danish Vempamil Infarction Trial II) and shows similar efficacy as metoprolol in the APSIS study. The paucity of placebo data (antianginal treatment cannot be withheld during long periods of time in symptomatic patients) precludes firm conclusions regarding effects of drug treatment on prognosis. It is argued that patients with stable angina pectoris do well on medical treatment, and that /.?I blockers, verapamil, and, possibly, slow-release nifedipine may influence their prognosis favorably. (Am J Cardiol 1996;77:6D15D)

ngina pectoris is a symptom-based diagnosis, in A which the symptoms are considered to be caused by myocardial ischemia. The underlying pathophys-

changes in coronary tone or other mechanisms leading to reduced oxygen delivery to a region within the heart7 Exercise-induced angina is more closely related to coronary arteriosclerosis with fixed stenoses and is elicited when cardiac demands at a certain level of exercise exceed the coronary reserve. However, it is well known that many patients have variable thresholds for angina during stress due to a combination of stable flow restriction (arteriosclerotic plaques) and superimposed vasoconstriction, which is especially pronounced in arteriosclerotic segments and may vary from time to time.’ Considerable attention is also paid to signs of asymptomatic myocardial ischemia and the total ischemic burden of angina patients during ambulatory conditions.7-9

iology is usually related to coronary arteriosclerosis, but angina pectoris may also occur in patients with normal coronary arteries. The strong relation to coronary artery disease leads to an increased risk of cardiac events, such as acute myocardial infarction (MI) or sudden death.le4 As a reflection of the arteriosclerotic nature of the disease, the risk of other vascular complications, such as stroke, is also increased.3s5 Angina pectoris may manifest itself in several ways, and the severity and type of symptoms may vary considerably, from occasional chest pain or discomfort of the characteristic type6 to more or less intractable symptoms and threatening MI as in unstable angina pectoris. Symptoms may be elicited by stress or exercise, i.e., with increased cardiac demands, or occur at rest. Angina pectoris without obvious triggering factors is considered to be related to From the Departmentof Clinical Pharmacology(P.H.), Karolinska Hospital, and Department of Medicine (S.V.E., C.H., N.R.), Danderyd’s Hospital, Stockholm, Sweden. Acknowled ments: Work referred to in this article was sup orted by ASTRA-H&se 7 AB (Molndal, Sweden], Knoll AG (Ludwigs R afen, Germany], the Swedish Heart-Lung Foundation, and the Swedish Medical Research Council (5930). We are very grateful to all collaborators in the APSIS study, and to professor Kim Fox and the TIBET investigators for access to results that were in press when the manuscript was written. Address for re rints: Paul H‘emdahl, MD, PhD, Department of Clinical Pharmacoogy,P Karolins I,a Hospital, S-l 71 76 Stockholm, Sweden.

6D

0 1996 by Excerpta All rights reserved.

Medico,

Inc.

INCIDENCE AND PROGNOSIS ANGINA PECTORIS

OF

The varying nature of the disease, and its historybased diagnosis, has made evaluations of incidence and prevalence difficult. The prevalence is also dependent on treatment modalities and criteria. In Sweden, the prevalence of chest pain judged by a cardiologist to be of cardiac origin is estimated to be around 5% in males aged 50-57 years.3 A recent study in the United Kingdom showed an annual incidence of angina pectoris of 1.1 per thousand male inhabitants and 0.5 per thousand female inhabitants aged 31-70 years.” The prognosis of patients with angina pectoris is poorer than that of the general population. Before the introduction of modern treatment modalities, P.D. White reported that 66 of 200 patients died an 0002-9 149/96/$15.00 PII SOOO2-9149(96)00301-3

average of 3.4 years after the diagnosis.” Later, a 25year follow-up of a larger number of patients revealed an average survival of 9.4 years, an adverse impact of signs of more severe disease, and a poorer prognosis in male compared with female patients.’ The annual excess mortality was estimated to be 7% in males and 5.3% in females in this study from 1956.’ Patients with relatively benign prognosis were, however, already identified in the early studies, and later studies have shown that the pro nosis in stable angina pectoris may be rather good.3s5 Clinical characteristics generally considered to be adverse from a prognostic point of view are older age, male sex, previous MI, and multiple vessel or left main stem stenosis of significant magnitude. Gandhi et al” concluded that a new diagnosis of angina pectoris was not benign, as 4% died, 7% sustained a nonfatal MI, and 19% underwent revascularization procedures during a 16-month follow-up. Only 11% of the patients experienced spontaneous remission during this short-term follow-up. In an unselected population of patients with angina pectoris, there is thus every reason to search for patients at high risk of complications, and to intervene effectively. The recent U.K. study lo included patients with unstable angina pectoris (who have a high risk of developing cardiac complications) and is thus not representative of a patient population with stable angina pectoris. The previously mentioned Swedish study followed 427 men with definite or suspected angina pectoris for a mean of 5.8 years and found that 15% had died, 55% still had angina pectoris, and 14% were free from chest pain at follow-up.3 Female patients with chest pain have normal coronary angiograms more frequently than male patients, l2 and a normal angiogram is associated with a good prognosis. 13-15 It is thus not surprising that the prognosis generally is better for female compared with male patients with stable angina pectoris.1*5*16 However, this gender difference with regard to prognosis is not present after an ML5,16 at least not if the infarction was recognized.5 Thus, female patients with angina pectoris may require treatment different from that of male patients, even if little is known about this. The prognosis of patients with stable angina pectoris may thus be relatively benign, but there is excess cardiovascular morbidity and mortality that require attention. Knowledge about how treatment influences the prognosis of patients with stable angina pectoris has been limited. P.D. White could not “emphasize enough the value of proper therapy in prolonging life.” ” Proper therapy has changed much since his statement in 1926 and now includes various coronary interventions and medical therapies that have improved the situation for patients with angina pectoris. This presentation discusses the influence of medical treatment on the prognosis of patients with stable angina pectoris, with an emphasis on calcium channel antagonist treatment. There are so far no mortality studies, which would have required extremely large patient materials. However,

studies that allow some conclusions regarding morbidity have recently appeared.

TREATMENT

MODALITIES

Stable angina pectoris may be treated medically or interventionally. Treatment may be aimed at reducing symptoms and improving the patients’ quality of life, at improving the patients’ prognosis, or, preferably, both. The type of treatment is thus chosen on the basis of symptoms and an evaluation of the individual patients’ likelihood of having a complication, such as MI. Coronary interventions: Coronary artery bypass grafting (CABG) and percutaneous transluminal coronary angioplasty (PTCA) procedures offer possibilities to bypass or remove anatomic obstructions and thus to restore blood flow to the myocardium. However, they also entail interventional risks and considerable costs. Further, long-term results after revascularization are influenced by recurrence of obstruction (graft occlusion or restenosis), which is a problem not yet resolved. Long-term results after CABG have improved with time. Provided that appropriate (high-risk) patients are selected for revascularization, it is possible to obtain better long-term results with CABG than with medical treatment.“*‘* Knowledge about the long-term prognosis after PTCA is less extensive. Symptomatic relief can be better with PTCA compared with medical therapy, l9 but initial costs are high, repeated procedures are frequently needed, and the long-term prognosis after PTCA has not been established.“-*’ For patients at low risk, medical treatment offers symptom relief without apparent worsening of prognosis.18,20 The majority of patients with stable angina are not considered for revascularization for prognostic reasons, even if many patients could be considered for symptomatic reasons. Medical treatment: Antianginal drug therapy may reduce or abolish symptoms of angina pectoris via different mechanisms of action. The main classes of such drugs are the organic nitrates (short or long acting), P-adrenoceptor antagonists, and calcium channel antagonists.21 Their pharmacology will not be discussed in detail here. In brief, organic nitrates and calcium channel antagonists are vasodilators and counteract vasospasm. The nitrates possess venous selectivity (resulting in reduced preload and intraventricular pressure), whereas calcium channel antagonists are arterial vasodilators (resulting in coronary dilation and reduced afterload due to peripheral vasodilation) . Beta blockers reduce myocardial oxygen demands by reducing heart rate and contractility, but may also enhance oxygen delivery by prolonging the diastolic perfusion time and by causing reverse coronary steal. The latter phenomenon may be explained by attenuation of metabolic vasodilation (and perhaps unopposed a-mediated vasoconstriction) in the myocardium, leading to redistribution of flow to compromised areas. These 3 types of drugs may be used singly or in combination. 21,22However, triple therapy may A SYMPOSIUM:

CALCIUM

ANTAGONISM

IN CAD

7D

produce less benefit than the use of only 2 drugs in many patients.23 The calcium channel antagonists are a heterogeneous group of drugs, among which dihydropyridines (nifedipine and others) possess varying degrees of vascular selectivity, whereas verapamil and diltiazem also produce mild negative inotropic effects and some depression of the sinoatrial and atrioventricular nodes.*l,** From a practical point of view, these 2 types of drugs are sometimes referred to as heart rate increasing and heart rate decreasing calcium antagnonists, respectively. Vasodilator effects of dihydropyridines may, especially in the short term, lead to reflexogenic tachycardia and sympathetic activation. Proischemic effects, possibly related to reduced perfusion pressure, reflexogenic cardiac stimulation, or coronary steal, have been noted with dihydropyridine treatment in patients with stable angina pectoris.24 Verapamil and diltiazem, on the other hand, may worsen heart failure and cause conduction disturbances in predisposed patients.*‘*** The importance of thrombosis in the context of ischemic heart disease is now not controversial. Antiplatelet therapy is beneficial in ischemic heart disease, 25 and patients with stable angina pectoris have benefited from aspirin treatment in a retrospective subgroup analysis26 and a prospective study.27 The studies were not large enough to show effects on mortality, but they clearly suggest favorable effects of antiplatelet therapy on prognosis in stable angina pectoris. It is unclear if such therapy also reduces symptoms in stable angina.28 However, a small study with low molecular weight heparin suggests that hemostatic activity may indeed contribute to symptoms in stable angina pectoris.29 Clearly, antithrombotic effects of antianginal drug treatment would be beneficial in patients with angina pectoris. The prospective aspirin study in stable angina pectoris27 is included in the later presentation, for comparison with the available long-term studies of antianginal drugs. Does antianginal drug therapy influence the prognosis of patients with stable angina pectoris? One line of evidence might be influences of drug treatment on the atherosclerotic process as such. Both calcium antagonists 30,3’ and /?-adrenoceptor antagonists 32 have demonstrated antiatherosclerotic effects in animal studies, and the International Nifedipine Trial on Antiatherosclerotic Therapy (INTACT) showed fewer new coronary lesions in patients treated with nifedipine.33 However, this was not accompanied by a favorable effeet on prognosis, as mortality was increased among nifedipine-treated patients in that study.33,34 The treatmentof acute coronary syndromes, such as unstable angina pectoris or MI, has been more thoroughly studied from the prognostic point of view than has the treatment of stable coronary artery disease. Results from such studies may also be of value in this context. 8D

THE AMERICAN

JOURNAL

OF

CARDIOLOGY@’

VOL.

77

LESSONS FROM STUDIES OF ACUTE CORONARY SYNDROMES AND SECONDARY PREVENTION Several /?-adrenoceptor antagonists have been shown to reduce or delay complications, i.e., mortality and reinfarction, after MI.35,36 There is probably consensus that P1-selective agents are as effective as nonselective agents and that p blockers without intrinsic sympathomimetic activity are to be preferred from the prognostic point of view. A relation between the degree of heart rate reduction at rest and effects on the prognosis after an MI has been noted, 37 but findings in this respect are not unanimous.36 There is clinical benefit from the use of p blockers also in unstable angina, as illustrated by metoprolol treatment in the Holland Interuniversity Nifedipine/Metoprolol Trial (HINT), 38 although their influence on rognosis in this patient category is less well studied. P6 The beneficial effects of p blockade after MI may to a large extent be related to reduction of sudden death caused by cardiac rupture or arrhythmias.39 A reduced risk of rupture is logical if reduced contractility in noninfarcted areas reduce stress on the infarcted area but is of limited relevance for patients with . . stable angina pectoris without such a loC’s mmons* Further support for a beneficial effect of p blockade may be derived from subgroup analyses of patients with angina pectoris after MI. In the Norwegian timolol study, p blockade seemed to reduce both total and sudden death in such patients (see Table 19a in the article by Pedersen4’), but statistical analyses are not presented. Subgroup analyses in the Beta-Blocker Pooling Project (BBPP) showed that the relative efficacy of p blockade was similar in patients with and without prior angina pectoris.41 Thus, there was no additional benefit in patients with angina pectoris, but the increased risk in this subgroup did not reduce the relative benefit of treatment. In the Stockholm metoprolol study, the prognostic impact of ischemia on exercise at 6 weeks after the MI was analyzed .42 Multivariate analyses indicated that metoprolol had beneficial effects in patients with ischemia on exercise after MI. The effects of calcium channel antagonist treatment on the prognosis after MI are more complicated, in part due to the heterogeneity of this class of drugs. M onotherapy with nifedipine resulted in an unfavorable outcome in unstable angina.38 Metaanalyses ofplacebo-controlled studies by Furberg et a143.44have shown that nifedipine treatment tends to worsen the prognosis of patients after MI. The most recent metaanalysis of nifedipine trials in coronary artery disease, including the INTACT study in stable angina, claims a significantly increased mortality with shofl-acthg nifedipine at a dosage of,80 mg/day.34 Other dihydropyridines are less well studied in this respect. Based on these findings, there is an intense, on-going debate concerning the use of calcium channel antagonists, which will be further commented on later. The heart rate reducing (or nonselective) calcium antagonists, diltiazem and verapamil, have had more JUNE

20,

1996

favorable effects than nifedipine in post-MI trials. Diltiazem treatment was neutral from the prognostic point of view, as benefits in patients without heart failure were counterbalanced by adverse effects in patients with heart failure.45 With regard to verapamil, combined early and late intervention showed a similar influence of heart failure,46 whereas late intervention only was associated with neutral effects in patients with heart failure and benefits in those without heart failure and overa11.47 Heart rate reducing calcium channel antagonists have also been shown to reduce the rate of reinfarction.36 Long-term treatment with organic nitrates has been tested in 2 very large studies. However, there was neither benefit nor harm from nitrate treatment in acute MI patients in the third Gruppo Italian0 per lo Studio della Sopravvivenza (GISSI-3) and fourth International Study of Infarct Survival (ISIS-4).48,49 Thus, the safety of long-acting nitrate treatment is well documented, but such treatment should be for symptomatic rather than prognostic reasons after MI. The possible inferences with regard to stable angina pectoris that may be drawn from studies in acute coronary syndromes are that heart rate reducing and sympatholytic treatment with a p blocker is the principle of first choice and that verapamil treatment (which also slows heart rate) is the best alternative to p blockade.

STUDIES OF PROGNOSIS ON MEDICAL TREATMENT IN STABLE ANGINA PECTORIS During recent years, 4 rather large, controlled long-term studies of medical treatment in stable angina pectoris have appeared. The treatment principles studied were /? blockade, calcium antagonism, and antiplatelet therapy. Effects of organic nitrates have not been studied in this manner. The studies were not statistically powered to evaluate effects on mortality in stable angina. Such studies would require very large patient numbers due to the relatively benign prognosis of this condition (see above) and are unlikely to appear. The designs, patients, and objectives of the trials differed, which must be considered when interpreting them. The studies are presented briefly and are summarized in Tables I and II. Swedish Angina Pectoris Aspirin Trial: This is a multicenter trial (SAPAT) comparing low-dose aspirin (75 mg daily) and placebo in patients with a history of stable exercise-induced angina pectoris.27 No diagnostic tests were required for inclusion. All patients were treated with the nonselective /I-adrenoceptor antagonist sotalol (and thus tolerated this treatment) during at least 3 weeks before randomization to aspirin or placebo. Thus, the antianginal or prognostic effects of sotalol were not evaluated, but the trial is nonetheless included here for comparison. Altogether 2,035 patients were included and followed for >4 years, on average. Some of their characteristics are given in Table I.

SAPAT was performed in the general practice setting, with endpoint examination by cardiologists. Primary endpoints were fatal or nonfatal MI or sudden death. Secondary endpoints included fatal or nonfatal vascular events (including stroke) and allcause mortality.27 Atenolol Silent lschemia Study: This is a multicenter trial (ASIST) comparing treatment with atenolol (50 to 100 mg daily) or placebo in patients with documented coronary artery disease and ambulatory ischemia during a single-blind run-in period on placebo. 5o The patients studied in ASIST had no or mild symptoms of angina pectoris at the outset. The 306 patients included in the study were followed for < 1 year. Some of their characteristics are given in Table I. The ASIST study focuses especially on the prognostic importance of ambulatory ischemia and excludes patients with moderate-to-severe angina. Thus, the patients were carefully selected from 2,037 patients screened for possible participation. Outcome was assessed in terms of death or resuscitation from ventricular tachycardia or fibrillation (hence no mortality figure in Table II), nonfatal MI, unstable angina, aggravation of angina, or need for revascularization.5 Total Ischaemic Burden European Trial: This is a multicenter trial (TIBET), the design of which was presented earlier.” Study drugs were atenolol and slow release nifedipine, and the patients were divided into 3 treatment groups given atenolol (50 mg twice daily), nifedipine SR (20-40 mg twice daily), or the combination. All patients were tested for tolerability to full dose combination treatment before entering the study. TIBET inclusion criteria were stable angina pectoris and a positive exercise test after a single-blind run-in period on placebo (the latter criterion was fulfilled in 89% of the patients randomized, but all patients were included in the intention-to-treat analysis). The 682 patients were followed for 2 years, on average. Some of their characteristics are given in Table I. The TIBET study focuses on the prognostic importance of ischemia during ambulatory monitoring, but also examines the importance of ischemia on exercise. Outcome was assessed in terms of a combination of endpoints considered to be either “hard” (cardiac mortality, myocardial infarction, or unstable angina) or “soft” (revascularization or treatment failure). Secondary endpoints relate to exercise-induced and ambulatory ischemia.52 Angina Prognosis Study In Stockholm: As the TIBET investigators, we wished to assess the effects of /? blockade and calcium channel antagonism on the prognosis of patients with stable angina pectoris. Therefore in this study (APSIS), we chose to compare treatment with metoprolol and verapamil, since these drugs had stood the test of time, and available data from post-MI studies in the mid-1980s suggested beneficial effects of metoprolol,53*54 and at least neutral effects of verapamil.46 Furthermore, 1 of the post-MI studies with metoprolol was perA SYMPOSIUM:

CALCIUM

ANTAGONISM

IN CAD

9D

TABLE

I Comparison

of tong-Term,

Double-Blind

Trials

SAPAT

in Stable

ASIST

Angina

Pectoris:

TIBET

Patients,

2,035

306

682

809

No. centers Follow-up (yr; median) Treatments compared

94 4.2 Aspirin Placebo

=32 0.9 Atenolol Placebo

69 2 Atenoloi Nifedipine Combination

3.4 Metoprolol Verapomil

67 52 41 0 7 11 aspirin 10 placebo

64 87 30 37 31 24 -

60 86 21 33 7 5 27 atenolol 40 nifedipine$ 29 combination -

PI Withdrawals

(other

reasons)

-

8

and

Withdrawals* Comments

APSIS

No. patients

Patient characteristics Mean age (yr) Mole sex (%) Hypertension (%) Prior Ml (%) Prior revascularization (%) Diabetes (%) Withdrawals (side effects)’

Treatments,

TIBET: positive exercise 89% of patients]

tolerance

test (in

1 TIBET: mean follow-up SAPAT: all patients on sotalol

59 69 27 16 6 9 1 1 metoprolol 15 verapamil

ASIST and TIBET: >3 mo; APSIS: 13 yr ASIST: >3 mo; APSIS > 1 yr SAPAT: tolerability to sotalol tested before randomization to aspirin or placebo TIBET: tolerability to combination pretested Mainly administrative reasons (e.g., moving)

5

PI l SAPAT = Swedish Angina Pechxis Aspirin Trial; ASIST = Atenolol Silent lschemia Trial; TIBET = Total lschaemic Burden European Trial; APSIS = Angina Prognosis Study in Stockholm; MI = myocardiol infarction. t Reporting differs between trials. * p 1.001 compared to atenolol or combination.

TABLE

II Comparison

of Long-Term,

Double-Blind

Trials

SAPAT

ASIST

No. patients/treatments Patient-years per treatment

2,035/2 =4,300

Total deaths (no.; % per year) Cardiac deaths (no.; % per year) Nonfotol Ml, no. (% per year) Revascularized, no. (% per year) All primory endpoints (no.) Primary endpoints in treahnent

188 (2.2) 80 (0.9) 125 (1.5) 205

5 (1.8) 1 (0.4) 56

ASA: 8 1 + Plac: 124

ASA: 2.9 Plac: 1.9

VaUPS Numbers

Cardiac death and nonfatal % per year of follow-up

306/2 =140

in Stable

Angina

TIBET

Pectoris:

Endpoints

Results* Comments

FJ 1,400

13 (1.0) 36 (2.6) 18 (1.3) 124

47 34 31 102 251

Aten: 17$ Plac: 39

Aten: 47 Nif: 46 Comb: 31

Met: 128 Ver: 123

Aten: 2.3 Plac: 5.2

Aten: 3.8 Nif: 4.5 Comb: 2.5

Met: 2.6 Ver: 2.4

-

Main

APSIS

682/3 =450

4

and

809/2 Assuming groups

similar

follow-up

in different

treatment

(1.6)

(1.2) (1.0) (3.5)

Sudden

death/Ml

(ASIST includes

resuscitatibn)

CABG (mostly) or PTCA Specified for each study in footnote

MI ASIST includes

resuscitation

from VT/F

ASA = aspirin; Abn = atenolol; CABG = coronary artery bypass grafting; Comb = combination; Met = metoprolol; MI = myocardial infarction; Nif = nifedipine; Plac = placebo; PTCA = percutaneous transluminol coronary angioplasty; VW = verapomil; VF = ventricular fibrillation; VT = ventricular tachycardia. For study abbreviations see Table I. l Figures regarding cardiac

events from the TIBET and APSIS studies represent severest event, whereas

SAPAT and ASIST have presented first event. Definitions of

primary events differed between studies, and are: SAPAT: nonfatal or fatcrl Ml, sudden death; ASIST: death/resuscitation, nonfatal MI, unstable angina, aggravation of angina (need for unblinded therapy), need for revasculorization; TIBET: cardiac death, nonfatal Ml, unstable angina, revasculorizotion, treatment failure [need for unblinded therapy); APSIS: death, nonfatal Ml, incapacitating or unstable angina, cerebrovascular events (stroke or transient ischemic occident), peripheral vascular evenk. t p = 0.003. * p = 0.001.

formed by us.54 APSIS differs from the other studies, as it is a relatively large single-center study (Table I). The reason for this design was our aim to study not only outcome, but also clinical features and mechanisms influencing this outcome. Thus, extensive clinical and laboratory examinations were performed by a small group of doctors, nurses, and technicians on several occasions during the study. Inclusion criteria were a history of angina pectoris, and a thorough medical examination by a car10D

THE AMERICAN

JOURNAL

OF CARDIOLOGY@

VOL.

77

diologist supporting this diagnosis. There was no requirement for a positive exercise test or a coronary angiogram, as we wished to maintain generalizability of our results to patients with stable angina pectoris seen by any well trained physician. Among the exclusion criteria were MI within 3 years or revascularization during the last year before entry. Double-blind treatment consisted of either controlled release metoprolol (Seloken ZOC, 100-200 mg daily) or slow release verapamil (Isoptin Retard, 120-240 JUNE

20,

1996

mg twice daily). At the end of the study, half of the patients were on full-dose treatment in both groups. The 809 patients included in the APSIS study were followed >3 years, on average. Some of their characteristics are given in Table I. The main prognostic outcome measure in APSIS was a combination of morbidity and mortality, i.e. death (total and cardiovascular) and nonfatal cardiovascular complications. The latter included MI, revascularization, stroke, and peripheral vascular events, but also the “softer” endpoint of treatment failure (worsening angina) with indication for revascularization, as judged by a cardiologist, whether this eventually took place or not. The main results of the APSIS study have recently been presented.55

TOLERABILITY OF TREATMENT IN THE STABLE ANGINA PECTORIS TRIALS Side effects and tolerability in the 4 trials are difficult to compare, as criteria apparently differed. Withdrawals related to side effects are summarized in Table I. The SAPAT study had similar withdrawal rates due to side effects in low-dose aspirin and placebo-treated patients, indicating good tolerability of low-dose aspirin. ASIST reports no withdrawals, but more frequent down-titration of dosages in atenolol compared with placebo-treated patients; symptomatic bradycardia occurred in 6.6% on atenolol versus none on placebo.50 Withdrawals were relatively frequent in the TIBET study (Table I), even though tolerability to the combination of nifedipine and atenolol had been tested before entry into the study. In TIBET, monotherapy with slow release nifedipine resulted in significantly more withdrawals than treatment with atenolol alone or with the combination.52 In the APSIS study, the tolerability to metoprolol and verapamil was judged to be good, with a small increase in withdrawals due to gastrointestinal side effects in the verapamil group, but no overall difference between the 2 drugs.55 Results from a psychosocial inventory, including psychosomatic symptoms, sleep disturbances, and an evaluation of life satisfaction, showed no difference between the treatment groups after 3 years of treatment or at the end of the study.55

MAIN OUTCOMES IN THE STABLE ANGINA PECTORIS TRIALS Table II summarizes the prognostic results of the 4 studies discussed. The rates of events (percent per year) were estimated on the basis of the median or mean follow-up time stated by the authors. In APSIS 2,766 patient-years were studied on an intention-totreat basis.55 The total mortality rate was somewhat higher in SAPAT than in APSIS, probably due to the &year difference in mean age (Table I). Cardiac deaths occurred at similar rates in SAPAT, TIBET, and APSIS, but nonfatal MIS differed between the studies, with extremes of 2.6% per year in TIBET and 1.0% per year in APSIS. The latter may be due to different

requirements for time elapsed since a previous infarction (3 months in TIBET and ASIST, 3 years in APSIS, and no history of MI in SAPAT). Revascularizations also differed, with the lowest rate in ASIST (which had a high proportion of previously revascularized patients, and open-label antianginal therapy as first option on treatment failure) and the highest rate in APSIS. The primary endpoints of the studies differed substantially (see footnote to Table II). SAPAT showed that aspirin was superior to placebo with regard to cardiac death or nonfatal MI. ASIST showed significantly less aggravation of angina on atenolol compared with placebo treatment, and a nonsignificant trend toward fewer cardiac complications (death, ventricular tachycardia or fibrillation, MI, or hospitalization for unstable angina) on atenolol treatment (a relative risk of 0.55, with a confidence interval [CI] of 0.22-1.33).50 The ASIST study thus lacked statistical power to demonstrate if atenolol reduces cardiac complications. Neither of the 2 studies comparing calcium channel antagonism and p, blockade showed any significant difference with regard to primary endpoints, which were a combination of “hard” and “soft” endpoints in both studies (Table II; Figures 1 and 2). TIBET had less statistical power than APSIS, with half as many primary endpoints and 3 treatment groups instead of 2. APSIS was designed to detect a 30% treatment difference, and this was fulfilled according to post hoc calculations.55 The time courses for events in TIBET are illustrated in Figure 1. In the APSIS study the odds ratios for primary endpoints (total death and nonfatal cardiovascular events) with metoprolol versus verapamil treatment were 1.05 (CI: 0.78 1.41) at the end of study, and 1.08 (CI: 0.84-1.38) when calculated by Cox regression analysis and adjusted for sex and smoking.55 The difference between these 2 analyses relates to the slightly different time courses for events in the 2 treatment groups (Figure 2). To compare treatment effects on “hard” endpoints in the 4 studies, we estimated the yearly incidence of cardiac death or nonfatal MI (Table II). This combination of endpoints has only been statistically analyzed in the SAPAT study, which was much larger than the other studies and did not concern antianginal treatment. In the APSIS study there were 34 cardiac deaths or nonfatal MIS on metopro101 and 31 on verapamil treatment.55 In the TIBET study the corresponding figures were 17 on atenolol, 21 on nifedipine, and 11 on combination treatment.52 The rates of cardiac death or MI might differ between atenolol and placebo, and between nifedipine and combination treatment if larger trials were performed, whereas the difference between metoprolol and verapamil appears to be marginal (Table II). Thus, there appears to be no major difference between /I blockade and calcium channel antagonism with regard to “hard” endpoints, but these data must be interpreted with great caution, as neither study had the statistical power to detect such a difference. A SYMPOSIUM: CALCIUM ANTAGONISM

IN CAD

11D

400 Days

Female patients: As expected (see above), female patients with stable angina pectoris had a better prognosis than male patients also in the APSIS study. Table III illustrates that there were very few cardiovascular deaths among women, and that the combined cardiovascular event rate was lower among women than among men. There was no trend toward any gender difference in treatment effects with metoprolol versus verapamil. Hypertensive patients: Approximately 25% of the APSIS patients were treated for hypertension at inclusion into the study. These patients had a clearly poorer prognosis than nonhypertensive patients ( 18 of 2 19 vs 20 of 590 cardiovascular deaths), and there was a nonsignificant trend toward fewer combined events in verapamil compared with metoprololtreated patients with stable angina pectoris and hypertension (Table III). Diabetic patients: A small subgroup of APSIS patients had diabetes (mainly type II), but these patients had a much poorer prognosis than nondiabetic APSIS patients (9 of 69 vs 29 of 740 cardiovascular deaths). There were nonsignificant trends toward fewer combined events in metoprolol- compared with verapamil-treated patients with stable angina pectoris and diabetes (Table III).

600 in study

FKNRE 1. Event-free survival during treatment with atenalol I-), slow release nifedipine (-.-), or the combination (---I in the Total lschaemic Burden European Trial (TlBEl’).52 Evenk were constituted by all endpoints in the study.

DOES ANTIANGINAL DRUG TREATMENT INFLUENCE PROGNOSIS IN STABLE ANGINA?

60 -

50 i 0 n=809

I 1 n=785

I 2 n=650

I 3 n=477

I 4 n=296

I 5 n=145

FIGURE 2. Event-free survival during treatment with controlled release metoprolol (---) or slow release vempamil (--) in the Angina Prognosis Study in Stockholm (APSlS).55 Evenk were constituted by all endpoints in the study.

APSIS SUBGROUP ANALYSES: INFLUENCE OF GENDER, HYPERTENSION, AND DIABETES The APSIS study showed a neutral main result when comparing metoprolol and verapamil treatment in the entire material. However, we specifically intended to investigate relations between various clinical and laboratory parameters and outcome to gain knowledge concerning which subgroups of patients might benefit more from one treatment or the other. Such subgroup analyses should, nonetheless, be considered as hypothesis generating rather than hypothesis testing. To give some more depth to the APSIS data in this context, we have compared outcomes in 3 subgroups of patients with prognoses expected to differ from that of the entire group, i.e., female patients and patients with concomitant hypertension or diabetes. These subgroup analyses are confined to cardiovascular events (excluding the 3 deaths due to malignancy in the metoprolol group and 6 in the verapamil group) and are presented in Table III. 12D

THE AMERICAN

JOURNAL

OF CARDIOLOGY”

VOL.

77

A major problem when addressing this question is the lack of sufficient data comparing antianginal and placebo treatment. We did not include a placebo group in the APSIS study, as we were interested in disease progression and atherosclerotic complications, and we anticipated that withholding prophylactic antianginal treatment would not be possible during a long period of time. The situation with symptomatic angina patients is clearly different from that in the post-MI setting. Similar judgments were apparently made by the investigators in the TIBET study. The SAPAT study concerned antiplatelet treatment and had symptomatic background treatment with sotalol. The ASIST investigators included a placebo comparison but had to terminate the study prematurely due to marked differences in symptorns5’ even though less symptomatic patients were studied. Thus, there is very limited information on the prognosis of patients with stable angina pectoris comparing antianginal treatment to placebo or no treatment, but such comparisons are not feasible and would not be ethically acceptable. Firm conclusions regarding the effects of antianginal treatment on prognosis in stable angina pectoris cannot be drawn in the absence of sufficient placebo comparisons, but results from the available studies may be interpreted in view of other available documentation. With regard to p blockade, the ASIST study clearly demonstrated less aggravation of angina on atenolol compared with placebo treatment,50 but the study was too small and short-lasting to evaluate effects on cardiac complications. One may speculate JUNE

20,

1996

TABLE Versus

III Angina Prognosis Female Patients, and

Study in Stockholm Main Patients with Hypertension Treatment Subgroup

Patient

Group

patients

Female

patients

Hypertensive Diabetic

patients

patients

Verapamil Metoprolol Verapamil Metoprolol Verapamil Metoprolol Veropamil Metoprolol Verapamil Metoprolol

to Cardiovascular II) at Entry*

Cardiovascular No.

(4

All patients Male

and Sizes

Results with Regard or Diabetes (Type

(403) (406) (266) (295) (137) (1 1 1) (104) (1 15) (38) (3 1)

Deaths

Events,

Combined

(%)

No.

19 (4.7)

Analyses

16 (6.0) 19 (6.4) 3 (2.2)

88 104 29 21 35 53 21 13

PI (5.8) (10.4) (18.4) (6.5)

of Male

Events (%)

p Value+

1 17 (29.0) 125 (30.8)

19 (4.7)

0 6 12 7 2

and Subgroup

0.23

(33.1) (35.3) (21.2) (18.9) (33.7) (46.1) (55.3) (41.9)

0.19 0.29 0.08 0.08

* Cordiovasculor deaths included sudden deaths (within 2 h; n = 1 l), acute myocordial infarctions (n = 23), and vascular deaths (II = 4). In addition, there were 9 cancer deaths. Combined events include fatal and nonfatal cardiovascular events. The latter were acute myocardial infarction (n = 31). revasculorizotion (85 by CABG and 17 by PTCA], angiographies without revasculcrizotion (n = 37), other unstable angina (n = 5). cerebrovasculor events (n = 241, and peripheral vosculor events (n = 5). Statistical analyses were made on the intention-t*treat basis, and only on combined events, os this was not o mortality study. t p values ore derived from Cox regression analyses, taking the time courses for events into account. CABG = coronary artery bypass grafting; PTCA = percutaneous transluminal coronary angioplasty.

that the beneficial effects of p blockade in acute coronary syndromes and secondary prevention after MI (see above) should have some relevance for patients with stable angina pectoris, even if cardiac rupture is not a major problem in this patient population. In addition, p blocker-based treatment reduces cardioin hypertensive pavascular complications tients 36s6,57Thus, primary and secondary preventive effects of ,8 blockade, together with the statistically nonsignificant findings regarding “harder” endpoints in ASIST, may be argued to support the idea that p blockade has a favorable influence on prognosis in stable angina pectoris. The APSIS and TIBET studies suggest that also calcium channel antagonism may influence the prognosis of patients with stable angina favorably. The APSIS study allows more firm conclusions in this respect than the TIBET study, for reasons of statistical power. The results of TIBET are also more heterogeneous than those of APSIS, as there may be some difference between nifedipine alone and combined treatment (Table II; Figures 1 and 2). The trend toward fewer events on combined treatment in TIBET suggests that effects of the 2 treatment principles might be additive. This may be due to greater blood pressure reductions,52 but atenolol may also have blocked reflexogenic cardiac stimulation by nifedipine. If so, nifedipine might exert both positive and negative effects from the prognostic point of view, and the balance between these effects might differ between patient populations and with different concomitant treatments. If p blockade influences prognosis favorably, as already argued, results from these 2 trials would indicate favorable effects of treatment with verapamil and slow release nifedipine (the latter at least when combined with a p blocker) on the prognosis of patients with stable angina pectoris. Verapamil had secondary preventive effects in the Danish Verapamil Infarction Trial II (DAVIT II) ,4’ whereas short-acting nifedipine is associated

with adverse effects in acute coronary syndromes and after MI.34,43 It is not known which characteristics of nifedipine have led to these untoward effects. Discussants in the recent and on-going calcium channel antagonist debate favor an importance of the pharmacokinetic profile of the nifedipine formulation used in the early studies, 58*59and the formulation of nifedipine does influence its antianginal effects.60 However, it cannot be excluded that the pharmacodynamic effect profile of nifedipine is of importance, regardless of pharmacokinetic aspects. Slow release nifedipine and the newer dihydropyridines remain to be documented with regard to prognosis in post-MI patients. There are no prospective studies with any calcium channel antagonist regarding primary prevention in hypertension, 36*56357 and preliminary results from moderately sized trials do not favor the calcium antagonists used in those studies.61 Retrospective case-control studies of antihypertensive treatment have yielded conflicting results, 62-64 and may suffer from methodologic shortcomings.57v65 Thus, primary preventive effects of calcium channel antagonism remain undocumented. When comparing slow release nifedipine and verapamil treatment in stable angina (the TIBET and APSIS studies), the differences in study size should be considered. Taken together, the available data suggest an advantage of verapamil over slow release nifedipine. With the limitations mentioned;it is tempting to speculate that both p, blockade and heart rate reducing calcium channel antagonism influence the prognosis of patients with stable angina pectoris favorably, i.e., that the APSIS study compared 2 active treatment principles with similar efficacy. The neutral main results,.in APSIS with regard to metoprolol versus verapamll treatment were accompanied by tendencies toward different outcomes in different subgroups, which suggests that all angina patients may not be equally suited for both treatments. We found no importance of gender with respect to treatment effects of the 2 drugs. However, in higher risk A SYMPOSIUM:

CALCIUM

ANTAGONISM

IN

CAD

13D

patients with stable angina pectoris combined with hypertension or diabetes the 2 treatments tended to differ. The tendency toward fewer combined events with verapamil treatment in patients with hypertension and stable angina pectoris is of interest in view of the lack of documentation for calcium channel antagonism in primary prevention (see before) and the current calcium channel antagonist debate. An unpublished subgroup meta-analysis of post-MI trials with the heart rate lowering calcium antagonists verapamil and diltiazem45-47 shows greater treatment benefit in hypertensive patients than in nonhypertensive patients after an MI (Messerli FH, personal communication). This might be taken to support the present findings. However, similar benefits have been noted for p blockade versus l$aFebo in hypertensive post-MI patients in some, * but not a11,40 subgroup analyses. The findings in hypertensive subgroups of patients with ischemic heart disease are of interest and support the use of heart rate lowering calcium antagonists, especially verapamil, in such patients. It should be stressed, however, that hypotheses generated by these subgroup analyses should be tested further. The trend toward more favorable effects of metoprolol compared with verapamil treatment in diabetic patients with stable angina pectoris is also interesting and agrees well with previous results from post-MI studies. 66*67The negative “metabolic’ ’ effects of p blockade, which are frequently invoked in discussions concerning choice of therapy in hypertension, are apparently not so important in ischemic heart disease. A recent survey of long-term trials in hypertension concluded that antihypertensive drug effects on lipid and carbohydrate metabolism may be transient and of minor importance.57 The mechanisms underlying the favorable effects of metopro101 in diabetic patients after an MI* or with stable angina pectoris (present results) remain to be elucidated. Again, hypotheses based on subgroup analyses should be tested in specifically designed studies.

CONCLUSIONS The recent calcium channel antagonist debate 34*58*59,61*62 has drawn attention to the need for documentation concerning effects of treatment-not only with calcium channel antagonists and not only for drug therapy-on the prognosis of patients with cardiovascular disease. Unexpected adverse treatment outcomes have been found with, for example, antiarrhythmic drug treatment.39 Therefore, the studies reviewed here provide valuable information on the safety of treatment and outcomes in stable angina pectoris. The long-term studies discussed herein do not document effects of antianginal drug treatment on mortality; such studies may never be performed due to the need for sample size, as well as other reasons already discussed. However, the morbidity data accumulated so far do provide a reasonably good basis for treatment of stable angina pectoris with either a 14D

THE AMERICAN

JOURNAL

OF CARDIOLOGY@

VOL.

77

pi blocker or verapamil as monotherapy or in combination with a long-acting organic nitrate. Questions still remain as to why short-acting nifedipine treatment has been associated with increased mortality in ischemic heart disease. TIBET data lend support to its use in a slow release formulation, especially in combination with a ,8 blocker. The tolerability of verapamil, however, seems to be better than that of slow release nifedipine. Organic nitrates are safe to use for symptom relief in ischemic heart disease, but there are at present no suggestions of prognostic benefit. Revascularization improves the prognosis of high-risk patients, but the majority of patients with stable angina pectoris have a relatively good prognosis. If there is no symptomatic indication for revascularization, these patients will do well on antianginal drug treatment. Further long-term studies on drug treatment and more information on possible differences in treatment efficacy in subgroups of patients, as well as on how prognostic factors leading to adverse outcome might modify the effects of antianginal drug treatment in stable angina pectoris, are of interest.

1. Richards DW, Bland EF, White PD. A completed twenty-five-year follow-up study of 456 patients with angina pectoris. J C/won Dis 1956;4:423-433. 2. Goldman GJ, Pichard AD. The natural history of coronary artery disease: does medical therapy improve the prognosis? Prog Cardiovasc Dis 1983;25:513-552. 3. Rosengren A, Hagman M, Penner? K, Wilhelmsen L. Clinical course and symptomatology of angina pectoris in a population study. Acfa Med Scand 1986;220:117-126. 4. Murabito JM, Anderson KM, Kannel WB, Evans JC, Levy D. Risk of coronary heart disease in subjects with chest discomfort: the Framingham heart study. Am J Med 1990;89:297-302, 5. Murabito JM, Evans JC, Larson MG, Levy D. Prognosis after the onset of coronary heart disease-an investigation of differences in outcome between the sexes according to initial coronary disease presentation. Circulation 1993;88:2548-2555. 6. Heberden W. Some account of a disorder of the breast. Med Transact Co11 Physicians 7. Maseri

1778;2:59-67.

A. Medical therapy of chronic stable angina pectoris. Circulation 1990;82:2258-2262. 8. Deedwania PC, Carbajal EV. Silent ischemia during daily life is an independent predictor of mortality in stable angina.. Circulation 1990;81:748-756. 9. Mnlcahy D, Fox K. Therapeutic implications of ischemia in the ambulatory setting. Prog Cardiovasc Dis 1992;34:413-428. 10. Gandhi MM, Lampe FC, Wood DA. Incidence, clinical characteristics, and short-term prognosis of angina pectoris. Br Hearf J 1995;73:193-198. 11. White PD. The prognosis of angina pectoris and of coronary thrombosis. JAMA 1926;87:1525-1530. 12. Sullivan AK, Holdright DR, Wright CA, Sparrow JL, Cunningham D, Fox KM. Chest pain in women: clinical, investigative, and prognostic features. Br Med .I 1994;308:883-886. 13. Chambers J, Bass C. Chest pain with normal coronary anatomy: a review of natural history and possible etiologic factors. Prog Cardiovasc Dis 1990:33:161-184. 14. Cannon RO. Angina pectoris with normal coronary angiograms. Cardiol Clin 1991:9:157-166. 15. Lichtlen PR, Bargheer K, Wenzlaf P. Long-term prognosis of patients with anginalike chest pain and normal coronary angiographic findings. J Am CON Cardiol 1995;25:1013-1018. 16. Chiencia A, Bailey K, Yawn BP, Kottke TE. Effect of gender on long-term outcome of angina pectoris and myocardial infarction/sudden unexpected death. JAMA 1993;269:2392-2397. 17. Emend M, Mock MB, Davis KB, Fisher LD, Holmes DR. Chainnan, BR, Kaiser

GC, Alderman E, Kihip T JR, and Participants in the Coronary Artery Surgery Study (CASS). Long-term survival of medically treated patients in the Coronary Artery Surgery Study (CASS) registty. Cimdation 1994;90:2645-2657. 18. Yusuf S, Zucker D, Peduzzi P, Fischer L, Takaro T, Kennedy W, Davis K, Killip T, Passamam E, Norris R, Morris C, Marhur V, Varnauskas E, Chalmers T. Effect of coronary artery bypass graft surgery on survival: overview of IOJUNE

20,

1996

year results from randomised trials by the Coronary Artery Bypass Surgety Trialists Collaboration. Lancer 1994;344563-570. 19. Parisi AF, Folland ED, Hattigan P. A comparison of angioplasty with medical therapy in the treatment of single-vessel coronary artery disease. Veterans Affairs Investigators. N Engl .I Med 1992;326:10-16. 20. The Swedish Council on Technology Assessment in Health Care (SBU ) Coronary artery bypass graft and percutaneous transluminal coronary angioplasty. SBU report no. 120E. 21. Cohn PF. Concomitant use of nitrates, calcium channel blockers, and beta blockers for optimal antianginal therapy. Clin Cardiol 1994;17:415-421. 22. Weiner DA. Calcium antagonists in the treatment of ischemic heart disease: angina pectoris. Coron Artery Dis 1994;5:14-20. 23. Tolins M, Weir EK, Chesler E, Pierpont GL. “Maximal” drug therapy is not necessarily optimal in chronic angina pectoris. .I Am Coil Cardiol 1984;3:1051-1057. 24. Waters D. Proischemic complications of dihydropyridine calcium channel blockers, Circulation 1991;84:2598-2600. 25. Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet treatment--l: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Br Med J 1994;308:81-106. 26. Ridker PM, Manson JE, Gaziano JM, Buring JE, Hennekens CH. Low dose aspirin therapy for chronic stable angina. A randomized placebo-controlled clinical trial. Ann Intern Med 1991;114:835-839. 27. Juul-Moller S, Edvardsson N, Jahnmatz B, Rosen A, Serensen S, Gmblus R, for the Swedish Angina Pectoris Aspirin Trial (SAPAT) Group, Doubleblind trial of aspirin in primary prevention of myocardial infarction in patients with stable chronic angina pectoris. Lmcet 1992:340:1421-1425. 28. Hjemdahl P. Platelet reactivity, exercise, and stable coronary artery disease. Eur Hem? J 1995;16:1017-1019. 29. Melandti G, Semprini F, Cervi V, Candotti N, Palazzini E, Branzi A, Magnani B. Benefit of adding low molecular weight heparin to the conventional treatment of stable angina pectori--a double-blind, randomized, placebo-controlled trial. Circularion 1993;88:2517-2523. 30. Henry PD. Atherosclerosis, calcium, and calcium antagonists. Circulation 1985;72:456-459. 31. Parmley WW. Vascular protection from atherosclerosis: potential of calcium antagonists. Am J Cardiol 1990;66(suppl):161-221. 32. Kaplan JR, Manuck SB, Adams MR, Clarkson TB. The effects of betaadrenergic blocking agents on atherosclerosis and its complications, Eur Heart J 1987;8:928-944. 33. Lichtlen PR, Hugenholtz PG. Raffler&u1 W, Hecker H, Jost S, Deckers JW, on behalf of the INTACT group. Retardation of angiographic progression of coronary artery disease by nifedipine: results of the International Nifedipine Trial on Antiatherosclerotic Therapy (INTACT). Lancer 1990;335: I 109- 1113. 34. Furberg CD, Psaty BM. Meyer JV. Niiedipine dose-related increase in mortality in patients with coronary heart disease. Circulation 1995;92: 1326- 133 1. 35. Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta-blockade during and after myocardial infarction: an overview of the randomized trials. Prog Cardiovasc Dis 1985;27:335-371. 36. Yusuf S, Lessem J, Jha P, Lonn E. Primary and secondary prevention of myocardial infarction and strokes: an update of randomly allocated, controlled trials. J Hypemns 1993;ll (suppl4):S61-S73. 37. Kjekshus JK. Importance of heart rate in determining beta-blocker efficacy in acute and long-term acute myocardial infarction trials. Am J Cardiol 1986;57(suppl 12):43F-49F. 38. Holland Interuniversity NifedipineMetoproloI Trial (HINT) Research Group. Early treatment of unstable angina in the coronary care unit: a randomized, double-blind, placebo controlled comparison of recurrent ischaemia in patients treated with nifedipine or metoprolol or both. Br Heart J 1986;56:400413. 39. Koon KT, Yusuf S, Furberg CD. Effects of prophylactic antiarrhythmic drug therapy in acute myocardial infarction. JAMA 1993;270:15891595. 40. Pedersen TR, for the Norwegian Multicenter Study Group. Timolol-induced reduction in mortality and reinfarction-a multicenter study on timolol in secondary prevention after infarction. Acta Med Sand 1983;(suppl679): I- 129. 41. The Beta-Blocker Pooling Project Research Group. The Beta-Blocker Pooling Project (BBPP): subgroup findings from randomized trials in post infarction patients. Eur Heart J 1988;9:8-16. 42. Olsson G, Rehnqvist N, Freyschuss II, Zetterquist S. Influence of long-term metoprolol treatment on early and late exercise test performance after myocardial infarction. Am J Cardiol 1988;61:519-523. 43. Held PH, Yusuf S, Furberg CD. Calcium channel blockers in acute myocardial infarction and unstable angina: an overview. Br Med J 1989;299: 1187- 1192. 44. Yusuf S, Held P, Furberg C. Update of effects of calcium antagonists in myocardial infarction or angina in light of the second Danish Verapamil In-

far&on Trial (DAVIT-II) and other recent studies. Am J Cardiol 1992;67:1295-1297. 45. The Multicenter Diltiazem Postinfarction Trial Research Group. The effect of diltiazem on mortality and reinfarction after myocardial infarction, N Engl J Med 1988;319:385-392. 46. The Danish Study Group on Verapamil in Myocardial Infarction, Verapamil in acute myocardial infarction. Eur Hean J 1984;5:516-528. 47. The Danish Study Group on VerapamiI in Myocardial Infarction. Effect of verapamil on mortality and major events after acute myocardiaI infarction (The Danish Vempamil Infarction Trial II-DAVIT II). Am J Cardioll990:66:779-785. 48. Gruppo Italian0 per lo Studio della Sopravvivenza nell’lnfarto miocardico. GISSI-3: effects of lisinopril and transdermal glyceryl trinitrate singly and together on 6-week mortality and ventricular function after acute myocardial infarction. Lmcet 1994;343:1115-1122. 49. ISIS-4 Collaborative Group. ISIS-4 a randomized factorial trial assessing early oral captopril, oral mononitrate. and intravenous magnesium sulphate in 58050 patients with suspected acute myocardial infarction. Lancer 1995;345: 669-685. 50. Pepine CJ, Cohn PF, Deedwania PC, Gibson RS, Handberg E, Hill JA, Miller E, Marks RG, Thadani U, for the ASIST Study Group. Effects of treatment on outcome in mildly symptomatic patients with ischemia during daily life-the Atenolol Silent lschemia Study (ASIST). Circulation 1994;90:762768. 51. The TIBET Study Group. The To& Ischemic Burden European Trial (TlBET): design, methodology and management. Cardiovasc Drugs Ther 1992;6:379-386. 52. Dargie HJ, Ford I, Fox KM, on behalf of the TIBET Study Group. Total lschaemic Burden European Trial (TIBET): effects of ischaemia and treatment with atenolol, nifedipine SR and their combination on outcome. Eur Hean J 1996;17:104-112. 53. Hjalmarsson A, Elmfeldt D, Herlitz J, Holmberg S, Mdlek I, Nyberg G, et al. Effect on mortality of metoprolol in acute myocardial infarction: a double-blind randomized trial. Lancer 1981;ii:823-827. 54. Olsson G, Rehnqvist N, Sjiigren A, Erhardt L, Lundman T. Long-term treatment with metoprolol after myocardial infarction. Report on 3 year mortality and morbidity. JAm Coil Cardiol 1985;5:1428- 1437. 55. Rehnqvist N, Hjemdahl P, Billing E, Bjorkander I, Eriksson SV, Forslund L, Held C, N&man P, Wall&t NH. Effects of metoprolol versus verapamil treatment in patients with stable angina pectoris-the Angina Prognosis Study In Stockholm (APSIS). Eur Heart J 1996;17:76-81. (Erratum p.483) 56. The Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. The fifth report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (JNC V). Arch Intern Med 1993;153:154-183. 57. Swedish Council on Technology Assessment in Health Care. Moderately elevated blood pressure. J Intern Med 1995;238 (suppl737):1-225. 58. Opie LH, Messerli FH. Nifedipine and mortality-grave defects in the dossier. Circulation 1995;92:10681073. 59. Kloner RA. Nifedipine in ischemic heart disease. Circulation 1995;92:1074-1078. 60. Karlsson BW, Emanuelsson H, Herlitz J, Nilsson J-E, Olsson G. Evaluation of the antianginal effect of nifedipine: influence of formulation dependent pharmacokinetics. Ear J Clin Pharmacol 1991;40:501-506. 61. Yusuf S. Calcium antagonists in coronary artery disease and hypertensiontime for reevaluation? Circulation 1995;92: 1079-1082. 62. Psaty BM, Heckbert SR, Koepsell TD, Siscovick DS, Raghunathan TE, Weiss NS, Rosendaal FR, Lemaitre RN, Smith NL, Wahl PW, Wagner EH, Furberg CD. The risk of myocardial infarction associated with antihypertensive drug therapies. JAMA 1995;274:620-625. 63. Hoes AW, Grobbee DE, Lubsen J, Man in’t Veld AJ, van der Does E, Hofman A. Diuretics, &blockers, and the risk for sudden cardiac death in hypertensive patients. Ann Intern Med 1995;123:481-487. 64. Aursnes I, Litleskare I, Frayland H, Abdelnoor M. Association between various drugs used for hypertension and risk of acute myocardial infarction. Blood Press 1995;4:157-163. 65. Buring JE, Glynn RJ, Hennekens CH. Calcium channel blockers and myocardial infarction-a hypothesis formulated but not yet tested. JAMA 1995;274:654-655. 66. Malmberg K, Herlitz J, Hjalmarsson A, RydCn L. Effects of metoprolol on mortality and late infarction in diabetics with suspected acute myocardial infarction. Retrospective data from two large studies. Eur Heart J 1989;10:423428. 67. Tse WY, Kendall M. Is there a role for beta-blockers in hypertensive diabetic patients? Diabet Med 1994;11:137- 144.

A SYMPOSIUM:

CALCIUM

ANTAGONISM

IN CAD

15D