Intended for healthcare professionals

Clinical Review

Fortnightly review: Secondary prevention in acute myocardial infarction

BMJ 1998; 316 doi: (Published 14 March 1998) Cite this as: BMJ 1998;316:838
  1. Rajendra H Mehta, cardiology fellow,
  2. Kim A Eagle, chief of clinical cardiology
  1. University of Michigan Hospital, Division of Cardiology, Taubman Center 3910, 1500 East Medical Center Drive, Ann Arbor, MI 48109-0366, USA
  1. Correspondence to: Dr Eagle

    Acute myocardial infarction affects hundreds of thousands of people each year. Around a quarter die, half of them before reaching a hospital. Survivors are at increased risk of recurrent myocardial infarctions or cardiac death, with a 10% death rate in the first year after discharge and a subsequent annual death rate of 5%—six times that in people of the same age who do not have coronary artery disease. Whereas aggressive management of the acute infarction has been enthusiastically adopted, far less attention has been given to preventive strategies. Though most doctors agree with the importance of secondary prevention, the results of studies suggest that many patients are currently not being given optimal preventive care. Since most patients after acute myocardial infarction are routinely followed up in primary care, general practitioners must be fully informed and participate in treatment strategies designed for the secondary prevention of coronary artery disease.


    We conducted a Medline search for all articles on acute myocardial infarction dating from January 1980 to August 1997, with particular emphasis on secondary prevention in acute myocardial infarction. We scanned all of these reports, which numbered more than 3000. In this review we have also incorporated current standard practice at the University of Michigan, Ann Arbor, for managing myocardial infarction.

    β Blockers

    Several controlled trials in more than 35 000 survivors of myocardial infarction have shown the benefit of long term treatment with β blockers in reducing the incidence of recurrent myocardial infarction, sudden death, and all cause mortality (table 1).1-8 β Blockers reduce myocardial workload and oxygen consumption by reducing the heart rate, blood pressure, and contractility, and they increase the threshold for ventricular fibrillation. A meta-analysis of such treatment in patients who have had myocardial infarctions shows a 20% reduction in long term mortality and a 34% reduction in sudden cardiac death. Patients at high risk without contraindications benefit most from such treatment—that is, those who have had a myocardial infarction, those who have had a large anterior infarction, those with congestive heart failure that is compensated, those with a reduced ejection fraction, those with ventricular arrhythmias, elderly people, and patients with evidence of cardiac ischaemia. β Blockers should be avoided in patients with symptomatic heart failure, clinically significant bradyarrhythmias, hypotension, and reactive airways disease.2-7

    Summary points

    All patients with myocardial infarction (or at least those with congestive heart failure or left ventricular ejection fraction <40%) should be given aspirin 160-325 mg, β blockers, and angiotensin converting enzyme inhibitors

    Diet counselling (low fat diet) should be offered to everybody, and patients with a low density lipoprotein cholesterol concentration >3.37 mmol/l should be treated with lipid lowering agents

    Current smokers should be offered help to stop smoking

    All discharged patients should be referred for outpatient cardiac rehabilitation

    Secondary risk factors such as diabetes and hypertension should be aggressively controlled


    A more extensive version of this article is available on our website,

    Treatment should be started within 24 hours of a myocardial infarction. The size of the infarct can be reduced by intravenous metoprolol or atenolol followed by oral β blockers. This regimen also reduces the incidence of reinfarction, ventricular fibrillation, cardiac rupture, and intracranial haemorrhage in hospital.5-8 Treatment should be continued for at least two or three years and for longer if well tolerated. 2 9 The role of β blockers in non-Q wave infarction is less clear, but they should be used in all cases if not contraindicated.9

    The beneficial effects of β blockers seem to be a class effect, but those with agonist activity do not show a beneficial effect on mortality, and their use cannot be recommended at present.

    Despite unequivocal evidence favouring benefits over risks, only 36-42% of patients who were enrolled in the United States National Registry of Myocardial Infarction from 1990 to 1993 received β blockers. This treatment is particularly underused in patients cared for by general practitioners, who may be less familiar than cardiovascular specialists with the results from clinical trials.

    Table 1

    β Blockers in myocardial infarction: results of randomised trials

    View this table:

    Calcium channel blockers

    The evidence for beneficial effects of calcium channel blockers is far less compelling. Nifedipine treatment is associated with a trend towards increased mortality and reinfarction (table 2).10-13 Verapamil does not reduce mortality or reinfarction. 14 15 Diltiazem increases cardiac events (death or non-fatal myocardial infarction) in patients with congestive heart failure as manifested by a left ventricular ejection fraction of <40% but decreases their incidence in patients with preserved left ventricular function.16 Diltiazem in patients with non-Q wave infarction seems to reduce the reinfarction rate during the first six months after the infarction, but the incidence of late reinfarction is similar to that with placebo. On the basis of these and other studies, the routine use of calcium channel blockers for secondary prevention after Q wave infarction cannot be recommended.

    Table 2

    Calcium channel blockers in myocardial infarction: results of randomised trials

    View this table:

    Angiotensin converting enzyme inhibitors

    Many clinical trials have evaluated angiotensin converting enzyme inhibitors in patients with myocardial infarction (table 3). Patients with large myocardial infarctions, clinical signs of heart failure, or objective evidence of having a left ventricular ejection fraction <40% benefit most, with significant reduction in mortality and risk of heart failure.17-20 Modest benefits from angiotensin converting enzyme inhibitors have also been shown in trials enrolling a wide cross section of patients, supporting its use for all patients with myocardial infarction who have no contraindications to their use.21-24 Angiotensin converting enzyme inhibitors are also effective in reducing ischaemic events after myocardial infarction (risk of recurrent infarction, unstable angina, and death from recurrent myocardial infarction). 17 25

    The consensus is that treatment with angiotensin converting enzyme inhibitors should be instituted in all patients with acute myocardial infarction complicated by symptomatic or asymptomatic left ventricular dysfunction. Additionally, there is some rationale for their use in all patients because of their effect in reducing ischaemic events.

    The beneficial effects of angiotensin converting enzyme inhibitors seem to be a class effect. Treatment should ideally be started within the first 24-48 hours and should be continued indefinitely, especially in patients with left ventricular dysfunction. It should be avoided in patients who have hypotension, bilateral renal artery stenosis, or a history of cough or angio-oedema with angiotensin converting enzyme inhibitors.9

    Table 3

    Angiotensin converting enzyme inhibitor in myocardial infarction: results of randomised trials

    View this table:


    Aspirin's benefit in secondary prevention is well established. In 19 791 patients who had had myocardial infarctions reviewed by the Antiplatelet Therapy Trialists, aspirin led to a 12% reduction in death, a 31% reduction in reinfarction, and a 42% reduction in non-fatal stroke.26 Low to medium doses (75-325 mg/day) seem to be as effective as high doses (1200 mg/day), with fewer gastrointestinal side effects. The use of other antiplatelet agents such as sulphinpyrazone and dipyridamole are not supported by the literature, except in patients with allergy to aspirin who are poor candidates for treatment with oral anticoagulants. No data yet exist to support the use of ticlopidine for secondary prevention after myocardial infarction.


    Studies of anticoagulant treatment after myocardial infarction indicate reduction in death, recurrent myocardial infarction, and thromboembolic complications. Trials comparing warfarin with aspirin for secondary prevention, however, show no difference in recurrent infarction or death between the two groups. Gastrointestinal side effects were more common with aspirin whereas haemorrhagic complications were more common with warfarin. Combinations of warfarin with aspirin show no advantage over aspirin alone.

    Currently, aspirin is recommended for all patients with myocardial infarction because of its low cost and ease of administration. Oral anticoagulants are indicated for patients with aspirin intolerance and for those at risk of embolisation from left ventricular or left atrial clot (those with severe left ventricular dysfunction, persistent atrial fibrillation, or left atrial thrombus). The suggested target international normalised ratio for such patients is 2-3.

    Lipid lowering agents

    An association of raised low density lipoprotein cholesterol with increased risk of heart disease is well recognised.27 However, considerable resistance has been observed among clinicians in the routine use of lipid lowering agents for secondary prevention of heart disease despite recent trials showing benefits on mortality.28

    Meta-analysis of secondary prevention trials also suggests that lipid lowering agents produce a reduction in fatal and non-fatal myocardial infarctions and cardiovascular deaths (table 4). The Scandinavian simvastatin survival study randomised 4444 patients with angina or myocardial infarction and raised low density lipoprotein cholesterol concentration to receive simvastatin or placebo.28 During the median follow up period of 5.4 years all cause mortality for the patients given simvastatin was reduced by 30%, which was largely attributed to the 42% reduction in the incidence of fatal coronary events in this group. Treatment led to a 34% reduction in the risk of a major coronary events and a 37% decrease in the need for revascularisation during the study.28 The cholesterol and recurrent events trial used pravastatin in a population of patients with myocardial infarction whose mean total cholesterol concentration (5.42 mmol/l) and low density lipoprotein cholesterol (3.60 mmol/l) were essentially the average for a general population. After a mean follow up of five years patients assigned pravastatin had a 24% reduction in fatal coronary artery disease and non-fatal myocardial infarctions (P=0.003).29 Favourable results have also been seen when cholesterol lowering was achieved by non-pharmacological interventions such as lifestyle modification (dietary treatment, weight management, or exercise) and ileal bypass surgery. Results from angiographic studies suggest that treatment retards progression and in some instances causes regression of coronary atherosclerosis.

    Table 4

    Lipid lowering drugs for secondary prevention

    View this table:

    A rationale also exists to increase high density lipoprotein cholesterol concentration in patients with coronary artery disease whose concentrations are low since reduced high density lipoprotein cholesterol concentration is an independent risk factor for coronary disease. Similarly, moderate to severely raised triglyceride concentrations in patients after myocardial infarction should be treated.27

    In the United States the Second National Cholesterol Education Panel has recommended that a complete lipid profile be obtained in all patients with myocardial infarction at the time of admission or no later than the first 24 hours; otherwise, the cholesterol concentration may be falsely low for up to four weeks after infarction. All patients should be started on a low cholesterol, low saturated fat diet. Treatment with inhibitors of 3-hydroxy-3-methylglutamyl coenzyme A reductase, nicotinic acid, or bile acid sequestrants should be instituted when plasma low density lipoprotein cholesterol concentration exceeds 3.37 mmol/l, with a goal of achieving a concentration <2.57 mmol/l. Use of fibrates would be appropriate for patients with heart disease, a raised triglyceride concentration, and a low high density lipoprotein cholesterol concentration.27

    Antiarrhythmic treatments

    The results of the cardiac arrhythmias suppression trials (CAST) have raised major concerns about the empirical prophylactic use of class I antiarrhythmic drugs to suppress asymptomatic ventricular ectopy after myocardial infarction. Patients enrolled in these trials who had had myocardial infarctions and took encainide, flecainide, or moracizine had an excess mortality compared with patients given placebo. This was the result of an increase in deaths from arrhythmia, even though drug treatment suppressed asymptomatic ventricular ectopy seen on Holter monitoring. A meta-analysis of clinical trials of antiarrhythmic treatment after myocardial infarction did not find any benefits on mortality from class I agents. Results of the survival with oral D-sotalol trial (SWORD) were equally disappointing. The experience with amiodarone is more hopeful. Two recent trials of the use of amiodarone after myocardial infarction in patients with left ventricular dysfunction—the Canadian amiodarone infarction trial (CAMIAT) and the European myocardial infarction trial (EMIAT)—suggest that amiodarone may decrease ventricular fibrillation and death from arrhythmia, although no significant effect on total mortality was seen. Also, amiodarone treatment was discontinued in 40% of patients because of side effects.

    Currently, β blockers are the only drugs that can be recommended for general prophylaxis against sudden death after acute myocardial infarction. Antiarrhythmic drugs such as amiodarone or the use of internal cardioverter defibrillators are probably justified in patients with sustained ventricular tachycardia or ventricular fibrillation after myocardial infarction. Further research is needed to clarify the role in high risk patients of antiarrhythmic drug treatment guided by the results of electrophysiological studies in comparison with implantation of automatic defibrillators.

    Other treatments

    Oestrogen given alone or in combination with progestins improves lipid profile and lowers fibrinogen concentration in postmenopausal women. Limited data support the use of oestrogen supplementation in secondary prevention after myocardial infarction. Small studies have shown that long term survival in women after coronary artery bypass surgery is greater in those taking oestrogen. In postmenopausal women with an intact uterus, oestrogen treatment should be combined with progestin and regular screening for uterine cancer. Women who have had a hysterectomy should receive only oestrogen. Oestrogen is often not used in women with a family history of breast cancer, although the association with breast cancer is less clear.

    Stopping smoking

    Smoking increases the risk of an initial cardiac event and doubles the rate of subsequent infarction and death. The risk declines rapidly after stopping and by three years reaches that of survivors who have never smoked. Having an acute myocardial infarction is a powerful impetus for patients to stop smoking, and doctors should capitalise on this to provide a firm, unequivocal, and recurrent message to stop smoking. The likelihood of stopping smoking increases with increasing participation of doctors. A multifaceted approach to stopping smoking includes repetitive coaching from a doctor, drug treatment, and, in difficult cases, referral to a formal programme of smoking cessation.30

    Lifestyle modification and cardiac rehabilitation

    All other cardiac risk factors should be assessed. Blood pressure control to within normal range improves outcome. Cardiac rehabilitation provides an opportunity to educate patients on modifying coronary risk factors and to give advice about exercise training, weight reduction, and psychosocial support. A formal programme improves functional capacity and reduces cardiovascular mortality and the risk of subsequent ischaemic events. It can also encourage compliance with treatment. Home exercise programmes, although lacking the advantage of supervision and education, are beneficial in patients at low risk. Depression is common in patients after a myocardial infarction and adversely affects mortality. Support from family and friends improves patient survival, and its importance should be emphasised.


    Every effort should be made to prevent the progression of coronary artery disease in patients who have had a myocardial infarction (box). β Blockers and aspirin should be given to all patients who have had a myocardial infarction, and lipid lowering agents should be given to patients with a low density lipoprotein concentration >3.37 mmol/l. Angiotensin converting enzyme inhibitors are likely to be protective in most patients, but especially in those with congestive heart failure or a reduced left ventricular ejection fraction. Treatment with β blockers should be continued for at least three years. Angiotensin converting enzyme inhibitors should be generally given to all patients who are able to tolerate them for 4-6 weeks and should be continued indefinitely in patients with congestive heart failure or a left ventricular ejection fraction <40%. Use of calcium channel blockers and routine prophylactic use of antiarrhythmic drugs to suppress ventricular ectopy should be avoided. Recommendations about diet, stopping smoking, and achieving ideal body weight should be an integral part of management. Hormone replacement treatment should be considered in postmenopausal women. Referral for outpatient rehabilitation should also be strongly encouraged. Adherence to the above recommendations reduces the long term likelihood of death, recurrent myocardial infarction, stroke, and need for coronary revascularisation.

    Discharge goals that improve outcome

    Aspirin 160-325 mg orally once a day Oral anticoagulants for aspirin intolerance (target international normalised ratio 2-3) β Blockers (avoid those with intrinsic sympathomimetic activity) Angiotensin converting enzyme inhibitors in patients with congestive heart failure or left ventricular ejection fraction <40% Avoidance of calcium channel blockers Avoidance of prophylactic use of antiarrhythmic drugs to suppress asymptomatic ventricular ectopy Dietary counselling (low fat diet) Lipid lowering agents when low density lipoprotein cholesterol >3.37 mmol/l Stopping smoking Outpatient rehabilitation Hormone replacement treatment in postmenopausal women Control of other secondary risk factors such as hypertension and diabetes


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    23. 23.
    24. 24.
    25. 25.
    26. 26.
    27. 27.
    28. 28.
    29. 29.
    30. 30.
    View Abstract