Drugs used in secondary prevention after myocardial infarction: Case presentation

Introduction

Part I of this case presentation focused on the emergency management of a patient with an acute anterior myocardial infarction who presented with sudden onset chest pain [1]. In part II, we will deal with the subsequent management of the case as from the early hours after admission to the Coronary Care Unit.

Case presentation – Part II

A 57 year old man was rushed to his local hospital with a 30 min history of severe chest pain associated with dizziness and nausea [1]. On examination he was pale, sweaty and tachycardic and an electrocardiogram demonstrated changes consistent with an acute anterior myocardial infarction. He was treated immediately with aspirin, opiate analgesia, oxygen and was given early intravenous streptokinase therapy in the absence of any contraindication (the potential benefits of tissue plasminogen activator were considered). The importance of initiating treatment as early as possible, the importance of the dedicated coronary care unit setting and the potential uses of heparin, magnesium, intravenous β-adrenoceptor blockade and insulin in the acute, early setting were also discussed [1].

We now contemplate his further management and consider therapeutic measures that will reduce his morbidity and increase his chances of survival in the coming months and years (Table 1).

Table 1.

Secondary coronary prevention measures following myocardial infarction.

Established efficacy following myocardial infarction
Aspirin	Dose 75–300 mg daily. Lower doses are equally efficacious but have less gastrotoxicity. Alternative antiplatelet drugs are also effective.
β-adrenoceptor blocker	Recognized cautions and contra-indications include severe cardiac failure, reversible airways obstruction and peripheral vascular insufficiency.
ACE inhibitor	Indicated for those with evidence of impaired left ventricular function.
Statins	Indicated for those in whom diet has failed or is unlikely to achieve target cholesterol reductions total cholesterol < 5.0 mmol l−1 or LDL cholesterol < 3.0 mmol l−1.
Lifestyle factors	Smoking cessation, dietary modification, aerobic exercise.

Potentially beneficial interventions following myocardial infarction
Anticoagulation	Superiority over antiplatelet therapy alone has not been established. May be indicated for other reasons in postinfarct patients e.g. dysrrhythmia or ventricular aneurysm.
Amiodarone	May offer benefits in subgroups at high risk of arrhythmic death postinfarct.
Revascularization	CABG offers mortality benefits over medical therapy in left mainstem disease and patients with triple vessel disease and impaired left ventricular function. PTCA not of proven benefit largely because of the high rates of restenosis.
Hormone replacement therapy	Exogenous oestrogens have apparently beneficial effects and seem to offer cardioprotective benefits in observational studies, alone and in combination with progestogens. No proven benefit demonstrated in prospective studies.
Antioxidants	Promising epidemiological data but no evidence of a mortality benefit in prospective trials.

Antiplatelet therapy

The benefits of aspirin therapy given daily in the first 5 weeks on early mortality after myocardial infarction are well established and have been reviewed [1]. Aspirin inhibits the enzyme cyclooxygenase I in platelets reducing aggregation and the platelet release reaction at much lower doses than those required for its anti-inflammatory actions [2]. Although aspirin is rapidly cleared by renal excretion (plasma half-life 30 min [3]) doses as low as 75 mg lead to irreversible inhibition of cyclooxygenase I and its antiplatelet effects that can persist for several days depending on platelet turnover. A variety of doses have been used in randomized trials examining the impact of regular prophylactic aspirin in secondary prevention following myocardial infarction. The ISIS 2 study demonstrated a significant 23% reduction in vascular mortality at 35 days in patients randomized to receive 162 mg enteric coated aspirin daily following a myocardial infarction or an absolute benefit of 24 vascular deaths per 1000 patients treated [4]. Ten year mortality data show persistence of the early mortality benefit [5]. It is less clear whether any additive benefit accrues from continuing aspirin beyond the first month. The Antiplatelet Trialists' Collaboration provided an overview of antiplatelet therapy that included 45 controlled trials of ‘prolonged’ (at least 1 month) aspirin therapy 75 mg to 1500 mg daily in patients at high risk of vascular disease [6]. A meta-analysis of the 11 largest antiplatelet trials in patients with prior myocardial infarction suggested that 36 (s.d. 6) vascular events were avoided for every 1000 survivors treated for a mean of 27 months. However, a smaller number of trials have included patients randomized beyond this period and the totality of the data suggest that the proportional benefit of treatment on vascular events is reduced in year 3 of treatment and beyond. All aspirin doses between 75 mg and 325 mg seem to be similarly efficacious and aspirin alone seems to be as efficacious as any other antiplatelet regimen.

The commonest adverse effect of chronic aspirin therapy is gastrointestinal toxicity with an increased risk of bleeding and ulceration despite doses as low as 75 mg daily [7]. A recent review concluded that long-term aspirin prophylaxis is associated with approximately 2.5 major and 7 minor gastrointestinal bleeding episodes per 1000 patient-years [8]. Another overview of randomized trials involving over 75 000 person years of exposure to aspirin suggested that there was a dose-related increase in gastrointestinal bleeding (2.5–7.7 bleeds per 1000 person years) and peptic ulcer (0.5–3.3 per 1000 person years) but that fatal bleeds were extremely rare [9]. These risks are heavily outweighed by the above benefit in most survivors of a myocardial infarction. Although the more expensive enteric-coated aspirin preparations have been associated with less peptic ulceration both in an endoscopic study [10] and a case-control study [11], a recent review of gastrointestinal bleeding episodes failed to show a clear benefit [12].

Another option for aspirin intolerant patients is its combination with antiulcer therapy. An endoscopic study of subjects taking 300 mg aspirin daily randomized to receive placebo or omeprazole (20 mg/40 mg) treatment showed a reduction in ulceration at 14 days from 12% to 3%/2% [13]. Ranitidine (a histamine type 2 receptor antagonist) treatment reduced duodenal ulceration in the setting of long-term aspirin therapy but was not as efficacious as proton pump inhibition [14]. Other antiulcer agents such as misoprostol and sucralfate also reduce aspirin-induced toxicity [15, 16]. In each case the cost associated with dual prescribing is a significant drawback. Ticlopidine is an alternative antiplatelet drug which inhibits ADP-induced platelet aggregation and shows similar efficacy and adverse event rates to aspirin therapy [17], but has the potential to cause blood disorders in a small number of patients. Clopidogrel has a similar action and has been shown to be at least equivalent to aspirin in secondary prevention following a variety of primary events [18]. On the basis of the available information we recommend that aspirin should be continued indefinitely after a myocardial infarction in doses of between 75 mg and 150 mg. Patients who cannot tolerate aspirin should be prescribed an alternative antiplatelet agent.

β-adrenoceptror blockers

β-adrenoceptor antagonists effectively reduce resting and exercise-induced heart rate, blood pressure, myocardial contractility and hence myocardial oxygen demand. In addition, they possess antiarrhythmic activity by antagonizing sympathetic outflow to the heart and increase the threshold for ventricular fibrillation in animal models [19]. When used in the setting of acute myocardial infarction they are associated with reduced mortality, fewer dysrhythmias, limitation of infarct size and lower incidence of ventricular septal rupture [1]. Beneficial pharmacological actions of this group are matched by well-established reductions of mortality in secondary coronary prevention (Table 2).

Table 2.

Trials of long-term β-adrenoceptor blockade in secondary prevention after myocardial infarction.

Trial	Drug/dose	Patients/duration	Reinfarction	Mortality
Norwegian Multicentre	Timolol	1884	14.4% vs 20.1%	10.6% vs 17.4%
1985 [20]	10 mg bd	17 months	RR 28.4%	RR 39% (P = 0.0005)
BHAT	Propranolol	3837	4.4% vs 5.3%	7.2% vs 9.8%
1982 [21]	60–80 mg tds	25 months	RR 16% (NS)	RR 26% (P < 0.05)
ISIS-1	Atenolol	16027		10.7% vs 12.0% ⋆
1986 [22]	100 mg od#	12 months		RR 11% (2P < 0.01)
Lopressor Intervention	Metoprolol	2395		5.6 vs 5.2%
Trial 1987 [23]	100 mg bd	12 months		RR −5% (NS)
EIS Group	Oxprenolol	1741	6.2% vs 5.1%	2.9% vs 2.7%
1984 [24]	160 mg bd	12 months	RR −22% (ns)	RR −7.4% (NS)
Australian/Swedish	Pindolol	529	14.1% vs 15.4%	10.6% vs 11.7%
1983 [25]	15 mg od	24 months	RR 8% (NS)	RR 5% (NS)
APSI Study	Acebutolol	607		5.8% vs 12%
1997 [26]	200 mg bd	12 months		RR 48% (P < 0.01)

^⋆vascular mortality

^#open control/atenolol for 7 days postinfarct.

NS = nonsignificant.

Three large prospective randomized trials have shown long-term mortality benefits from β-adrenoceptor blockade with timolol [20], propranolol [21], and atenolol [22] in patients who have sustained a myocardial infarction. In the Norwegian Multicentre Study Group trial timolol 10 mg twice daily was shown to significantly reduce total mortality in postinfarction patients (10.3% vs 16.2% P = 0.0003) over a mean 17 month follow-up, and at 6 years this benefit was maintained (26.4% vs 32.3% P = 0.0028) [20]. The Beta-Blocker Heart Attack Trial (BHAT) randomized 3837 postinfarction patients to receive placebo or propranolol 60–80 mg three times daily (administered according to plasma drug level) and demonstrated a significant mortality reduction at 25 months (7.2% vs 9.8%, P < 0.005) in the treated group [21]. Both studies demonstrated significant reductions in sudden cardiac death. The ISIS 1 study demonstrated a significant reduction in mortality due to early intervention with 5–10 mg intravenous atenolol followed by 100 mg oral atenolol daily for 7 days. Oral atenolol was continued in 35% of treated patients and 25% of controls at discharge, and after 1 year reduction of vascular mortality was maintained in the original treatment group (10.7% vs 12.0%, P < 0.005) [22]. Several studies of metoprolol in postinfarct patients have shown nonsignificant reductions in overall mortality in patients followed for up to 3 years [23, 27–29]. A meta-analysis of the metoprolol studies suggested a significantly reduced overall mortality [30].

No evidence is available from randomized controlled trials regarding the benefits of initiation of β-adrenoceptor blockade more than 1 year postinfarct. The duration of benefit from β-adrenoceptor blockade is uncertain but it seems prudent to continue treatment indefinitely since discontinuation after 3 years treatment is associated with an excess of angina symptoms, a nonsignificant increase in total mortality and the need to recommence β-adrenoceptor blockade in at least one third of patients [31]. Follow-up of patients randomized in the Norwegian multicentre trial has shown parallel survival curves suggesting continued benefit for at least 6 years postinfarction [20].

Some have questioned whether the benefit of β-adrenoceptor blockade in postinfarct patients is a ‘class effect’ and suggest that ancillary properties of individual drugs may be important. Pindolol and oxprenolol possess significant intrinsic sympathomimetic activity (partial agonist activity) and failed to demonstrate significant mortality reduction when used for secondary coronary prophylaxis [24, 25]. However, acebutolol possesses similar activity and 200 mg administered twice daily after myocardial infarction led to a significant reduction in total mortality at 1 year postinfarct (5.8% vs 12.0%, P < 0.01), but no significant effect was observed at 6 years postinfarction [26, 32]. Animal models have suggested that those β-adrenoceptor blockers which have greater lipid solubility such as metoprolol or propranolol may have an important central action on vagal tone and heart rate variability which could have a significant impact on postinfarct dysrhythmia and mortality [33]. Clinical studies comparing treatment with atenolol (hydrophilic) or metoprolol (lipophilic) found no significant difference in heart rate variability and failed to support the findings from the animal models [34].

The use of β-adrenoceptor blockers poses little hazard in appropriately selected patients [35]. They should be avoided in patients with hypotension, heart block, severe heart failure or cardiogenic shock. Mild or asymptomatic heart failure should no longer be seen as a contra-indication to β-adrenoceptor blockade. Subgroup analysis of early postinfarction clinical trials such as BHAT and ISIS 1 and more recent trials in patients with chronic heart failure have conclusively demonstrated that such patients have most to gain from β-adrenoceptor blockade [22, 36–39]. Although β-adrenoceptor blockade may worsen claudication in patients with peripheral vascular disease and in some cases exacerbate chronic reversible airways disease, these are not absolute contraindications to treatment. Treatment may mask symptoms of hypoglycaemic awareness but this should not unduly influence the decision to prescribe a β-adrenoceptor blocking agent in diabetic patients who are at particularly high risk of recurrent vascular events and also have much to gain from β-adrenoceptor blockade.

A recent observational study of 6581 unselected postinfarct patients found that only 38% were prescribed a β-adrenoceptor blocker and that 48% received a lower dosage than suggested by the large trials [40]; follow-up showed improved survival among the treated vs the untreated patients irrespective of which dosage was employed. β-adrenoceptor blockers reduce mortality, particularly sudden death, following myocardial infarction especially in those patients with poor left ventricular function and they are currently under-prescribed. The therapeutic benefits are probably shared by most drugs in this class. Although much of the above data comes from the prethrombolytic era, re-infarction and sudden cardiac death remain major issues of secondary prevention suggesting that β-adrenoceptor blockers continue to have an important role.

The patient is pain free and comfortable following initial management of the acute infarction. He has received aspirin but has not yet been initiated on a β-adrenoceptor blocker. Three hours later you are called urgently by the CCU staff nurse because he has ‘gone off’ and become hypotensive. You consider possible causes for his deteriorating condition prior to your arrival on the CCU including acute left ventricular failure secondary to the infarct, ruptured papillary muscle, acute dysrhythmia, acute bleeding following thrombolysis, a re-infarction, acute pulmonary embolus or a cardiac rupture. You find the patient distressed and speaking in half sentences. The ECG shows occasional runs of nonsustained ventricular tachycardia. He is not in any pain but is very dyspnoeic. On examination he is centrally cyanosed and sweaty. BP is 110/70 mm Hg and pulse rate 120 beats min⁻¹in sinus rhythm. Auscultation reveals an added 3rd heart sound and bilateral basal inspiratory crackles to midzones. A chest X-ray has been arranged and shows a large heart, upper lobe blood flow diversion and pulmonary oedema. After careful consideration of remedial causes such as valvular failure, septal rupture, dysrhythmia and persisting ischaemia you diagnose acute left ventricular failure secondary to myocardial infarction. He is receiving oxygen and you administer intravenous diamorphine, frusemide and commence an intravenous nitrate infusion. You consider inotropic support but his haemodynamic state gradually improves without further intervention.

Use of ACE inhibitors after myocardial infarction

Angiotensin-converting enzyme (ACE) inhibitors competitively antagonize the enzymatic conversion of angiotensin I to angiotensin II reducing systemic vascular resistance and cardiac afterload. These drugs also reduce aldosterone release with consequent reduction of circulating fluid load and lower cardiac preload. ACE inhibitors have been widely studied in chronic heart failure and their superiority over other vasodilating regimens has been proven unequivocally [41, 42]. These trials demonstrated a significant reduction in mortality, nonfatal myocardial infarction and the need for revascularization procedures suggesting that the benefit of ACE inhibitors may be extended to patients who have recently sustained a myocardial infarction (Table 3).

Table 3.

Studies of the initiation of ACE inhibitor therapy after acute myocardial infarction. * In GISSI-3, lisinopril had an open control rather than placebo and CONSENSUS II followed initial intravenous treatment.^#

Trial	Drug/ Dosage	Patients	Commenced/ Duration	Mortality
Unselected postinfarct patients
CONSENSUS-II*	Enalapril	6090	Immediate	11% vs 10.2%
[43]	20 mg od	ST↑/Qs/↑enz	6 months	RR −10% (P = 0.26)
GISSI-3*	Lisinopril	18 895	Up to 24 h	6.3% vs 7.1%
[44]	10 mg od	ST↑	6 weeks	RR 12% (2p = 0.03)
ISIS-4	Captopril	58 050	Up to 24 h	7.2% vs 7.7%
[45]	50 mg bd	suspected MI	5 weeks	RR 6% (2p = 0.02)
CCS-1	Captopril	13 634	Up to 36 h	9.1% vs 9.6%
[46]	12.5 mg tds	suspected MI	4 weeks	RR 6% (2p = 0.3)
Heart failure/high risk postinfarct patients
SAVE	Captopril	2231	3–16 days	20% vs 25%
[47]	50 mg tds	EF < 40%	24–60 months	RR 19% (P = 0.02)
AIRE	Ramipril	2006	3–10 days	17% v. 23%
[48]	5 mg bd	clinical heart failure	6–30 months	RR 27% (P = 0.002)
TRACE	Trandolapril	1749	3–7 days	35% vs 42%
[49]	4 mg od	EF < 35%	24–50 months	RR 22% (P = 0.001)
SMILE	Zofenopril	1556	Within 24 h	10.0% vs 14.1%
[50]	30 mg bd	Anterior MI	12 months #	RR 29% (P = 0.011)

RR = relative risk reduction. EF = left ventricular ejection fraction.

^#Zofenopril treatment for 6 weeks only.

The CONSENSUS-II study examined the effect of very early administration of intravenous enalaprilat followed by oral enalapril in all patients presenting with an acute myocardial infarction [43]. The 6 month mortality was nonsignificantly increased by 9% in the enalapril group and the trial was terminated early. In the ISIS-4 and GISSI-3 studies all patients who had sustained a suspected myocardial infarction were randomized within 24 h to an ACE inhibitor (captopril/lisinopril) or control (placebo/open) [44, 45]. Both studies demonstrated significant but small relative reductions in mortality after short-term follow-up (7% after 5 weeks and 12% after 6 weeks, respectively). Retrospective subgroup analysis of patients in the GISSI-3 study revealed a 3% absolute mortality reduction in those with (Killip class 2 or above) heart failure while only a 0.3% reduction in those without heart failure (Killip class 1). Similarly, the relative mortality benefit of captopril in the ISIS-4 study was 9.4% in those with heart failure but only 4.8% in those free of heart failure.

Later studies have examined the role of ACE inhibitors used selectively in postinfarct patients with reduced left ventricular function and found significant treatment benefits. Post-infarct patients with clinical, ventriculographic or echocardiographic evidence of cardiac failure treated with long-term captopril (SAVE), ramipril (AIRE) and trandolopril (TRACE) had a significantly reduced mortality (19% at 42 months, 27% at 15 months and 22% at 26 months, respectively) when compared with placebo [47–49]. Furthermore, the mortality reduction appears to be sustained with continued treatment. The participants of the AIRE study had a 36% relative reduction in mortality (27.5% vs 38.9%, P = 0.002) after an average of 59 months of follow-up, equivalent to 114 additional survivors at 5 years for each 1000 patients treated for an average 12.4 months [51]. The SMILE study randomized patients with anterior myocardial infarction who had not received thrombolysis and who were therefore at high risk of developing heart failure to zofenopril or placebo [50]. The treatment was given for 6 weeks at which point there was a nonsignificant 25% reduction in mortality (4.9% vs 6.5%; P = 0.19). When the patients were re-evaluated at 1 year those assigned zofenopril had a 29% reduction in total mortality (10.1% vs 14.1%; P = 0.011).

Hypotension was a common feature of the trials involving early randomization to ACE inhibitors. This deleterious effect may, in part, explain why benefits of treatment were much diminished among unselected groups of postinfarct patients many of whom did not have cardiac failure. The more recent studies of patients selected for poor left ventricular function did not initiate therapy until day 3 or beyond. On the basis of the available evidence, any patient with clinical or investigational evidence of cardiac failure after myocardial infarction should have an ACE inhibitor initiated within 24–48 h if the patient is haemodynamically stable. It is important that the dose should be increased subsequently to the dose used in clinical trials. Benefit of ACE inhibition in postinfarct patients is probably a class effect. However, the choice of agent will be influenced by patient compliance, drug half-life, cost and available evidence from clinical trials.

There has been some concern that the benefits of ACE inhibitors may be diminished in patients with heart failure who are also taking aspirin. A retrospective subgroup analysis of the CONSENSUS II study suggested that patients concomitantly using aspirin derived little survival benefit from enalapril [52]. The interaction may be explained by the potential for aspirin to inhibit bradykinin-induced synthesis of vasodilating prostaglandins which is normally potentiated by ACE inhibitors. A similar but less marked reduction in the survival benefit attributable to ramipril was also seen in the AIRE study [48]. Although the aspirin–ACE inhibitor interaction is of therapeutic interest use of both agents is recommended where indicated by the above criteria. Recent trial evidence has also demonstrated that the presence of chronic heart failure, whether or not it occurs in the postinfarct period, is also an indication for the initiation of β-adrenoceptor blockade in addition to ACE inhibition [37–39]. Patients should be clinically stable and started on a low dose before subsequent dose-titration. The benefits of β-adrenoceptor blockade appear to be a class-effect having been observed in agents with varying pharmacological properties.

The indications for ACE inhibition have also been recently extended by the results of the HOPE study [53]. In this factorial design study (with vitamin E), 9297 patients at high risk of coronary disease (previous coronary disease, stroke, peripheral vascular disease or diabetes with another risk factor) were randomly assigned to receive either ramipril 10 mg daily or placebo once daily for a mean of 5 years. Treatment with ramipril significantly reduced the risk of cardiovascular death (14%), myocardial infarction (20%), revascularization (15%), heart failure (13%) and total mortality (16%). The benefits were broadly equivalent across all entry criteria subgroups suggesting that the use of ACE inhibitors should be employed in secondary prevention more widely than merely for patients with evidence heart failure. To what extent these benefits were independent of and beyond the small reduction in blood pressure observed in the ramipril group remains unclear.

The patient improves over the next 24 h and is gradually mobilized. After a further 48 h he is transferred to the general ward where he makes an uneventful recovery over the subsequent 4 days. His current medications include aspirin, an ACE inhibitor and a low dose of frusemide. He has no symptoms or signs of cardiac failure.

Cholesterol-lowering drugs

Hypercholesterolaemia has long been recognized as an important risk factor for coronary artery disease but there was considerable controversy about the benefits or otherwise of intervention with drug treatment. Although there were apparent reductions in cardiovascular endpoints these appeared to be offset by an excess of noncardiac events raising fears that cholesterol-lowering or the drugs themselves may be harmful. These concerns have now been addressed by large prospective randomized controlled trials of lipid lowering therapy for primary and secondary coronary prevention (Table 4).

Table 4.

Cholesterol-lowering trials in postmyocardial infarction patients.

Trial	4S [54]	CARE [55]	LIPID [56]
Patients	4444	4159	9,014
MI/Angina-only (%)	79/21	100/0	64/36
Age (years)	35–70	21–75	31–75
M/F (%)	81/19	86/14	83/17
Lipid criteria
Total cholesterol	5.5–8.0	< 6.2	4.0–7.0
LDL-cholesterol	3.0–4.5
Triglycerides	< 2.5	< 4.0	< 5.0
Drug	Simvastatin	Pravastatin	Pravastatin
(mean dose)	27.4 mg	40 mg	40 mg
Impact on lipids
Total cholesterol	↓25%	↓18%
LDL-cholesterol	↓35%	↓32%	↓25%
Triglycerides	↓10%	↓14%	↓11%
HDL-cholesterol	↑8%	↑5%	↑5%
Duration (median, range, years)	5.4 (4.9–6.3)	5.0 (4.0–6.2)	Av. 6.1 years
Annual cardiac event rate	7.2%	2.6%	8.3%
Clinical events (% reduction, 95%CI)
Death	30 (15–42) 0.0003	9 (−12–26) 0.37	24 (12–35) < 0.001
Coronary death	42 (27–54)	20 (−5–39) 0.10	25 (13–35) < 0.001
Non-fatal MI	37(20–34)	23 (4–39) 0.02
Revascularization	37 (26–46) 0.00001	27 (15–37) < 0.001	20 (10–28) < 0.001
Stroke	30 (4–48) 0.024	31 (3–52) 0.03	19 (0–34) 0.048
Side effects (%, statin/placebo)
transaminases > 3x	2.2/1.5	3.2/3.5	2.1/1.9
creatine kinase > 10x	0.3/0.04	0.6/0.3
myositis	0.04/0	0/0.2	0.2/0.2
cancers	4.1/4.3	8.3/7.7	2.8/3.1

The Scandinavian Simvastatin Survival Study (4S study) was a large prospective randomized control study of simvastatin 20–40 mg daily (mean 27 mg) in 4444 patients with elevated serum cholesterol (5.5–8.0 mmol l⁻¹) and established coronary disease, 79% of whom had sustained a previous myocardial infarction [54]. Patients with a triglyceride level greater than 2.5 mmol l⁻¹ and those with cardiac failure were excluded. Intervention with simvastatin produced a 28% reduction of total cholesterol and 38% reduction of LDL cholesterol which resulted in a significant 30% mortality reduction (8% vs 12%, P = 0.0003) during a median 5.4 years follow up. The number needed to treat during this follow-up period was 15 to prevent one coronary event and 30 to prevent one death.

The Cholesterol and Recurrent Events (CARE) study was a randomized controlled study which examined the effect of pravastatin in 4159 postinfarct patients who had essentially average serum cholesterol levels (less than 6.2 mmol l⁻¹) and LDL cholesterol levels (3.0–4.5 mmol l⁻¹) [55]. The primary end point (a fatal coronary event or a nonfatal myocardial infarction) was reduced by 24% in the treated group (10.2% vs 13.2%, P = 0.003) during a median 5 years follow-up period. Most benefit was seen in those with higher baseline LDL cholesterol levels and no significant benefit of treatment was apparent in those with pre-existing LDL cholesterol less than 3.2 mmol l⁻¹. There was a nonsignificant reduction of total mortality in the treated group (8.6% vs 9.4%, P = 0.37). In this lower risk group, the number needed to treat during this follow-up period was 33 to prevent one coronary event and 91 to prevent one coronary death.

In the LIPID study, 9014 patients with a previous myocardial infarction or unstable angina across 84 centres in Austrialia and New Zealand were randomized to receive 40 mg pravastatin daily or placebo [56]. After an average 6.1 years follow-up period there was a significant reduction of total mortality (11.1% vs 14.1%, P < 0.001) and the postinfarct subgroup had a significant reduction of combined death and nonfatal myocardial infarction (13.8% vs 17.4%). Therefore, the number needed to treat during this follow-up period was 28 to prevent one coronary event and 33 to prevent one death.

The mortality benefit of statin therapy in these high risk groups was apparent from as early as 1 year of treatment. A further benefit was a significant reduction in the rate of stroke (19–31%) and the need for coronary revascularization procedures (20–37%). Serum cholesterol levels fall after a myocardial infarction or other major physical illness and may not return to baseline level until at least 6 weeks after the event [57]. Any measurement during this period is likely to be unreliable and underestimate risk. Although most of the trials have included a ‘run in’ period of diet before selecting patients for treatment, in the clinical setting, this is likely to mean a significant delay in initiating proven cardioprotective therapy and increase the likelihood that the treatment will not be initiated at all. We believe that cholesterol concentration should be determined from an admission blood sample and an early decision to treat should be based on this result. Directives concerning statin usage for postinfarct patients in the United Kingdom have been issued by the Standing Medical Advisory Committee and address the importance of global risk reduction in coronary heart disease management but suggested a threshold for treatment of 4.8 mmol l⁻¹[58]. More recent, joint recommendations from the British Cardiac Society, British Hyperlipidaemia Society, British Hypertension Society and the British Diabetic Association advocate that the lipid-lowering target in this group should be to lower total cholesterol and LDL cholesterol below 5.0 mmol l⁻¹ and 3.0 mmol l⁻¹, respectively, whether by dietary modification or statin therapy [59].

The major side-effects of HMG-CoA reductase inhibitors are disturbance of liver function tests and myositis, both of which occur more frequently when they are combined with resin therapy. It is recommended that liver biochemistry is checked before and infrequently during treatment and creatinine phosphokinase should be measured if the patient becomes symptomatic [60]. Statins should be used more cautiously in the setting of pre-existing liver disease or liver biochemistry disturbance and physicians should be aware of the increased potential for myositis when statin and fibrate therapy are combined. However, statins are generally very well tolerated and the incidence of the above side-effects in the large trials was reassuringly low (Table 4).

Despite the strong evidence of mortality benefit conferred by statin therapy in postinfarct patients, the use of lipid lowering treatment is still not widely applied amongst those who fulfil the above criteria, perhaps reflecting concerns about financial cost of widespread use. Although the cost-effectiveness of statins is likely to remain a contentious issue in the setting of primary prevention it is optimal in patients identified to be at the highest risk because of a previous coronary event. Although dietary modification may have some role in the reduction of circulating LDL cholesterol [61] the benefits have been at best modest [62] and the vast majority of postinfarct patients will require a statin.

Calcium channel blockers

The use of calcium channel blocking drugs following a myocardial infarction is controversial. The SPRINT I study randomized 2276 postinfarct patients to receive 30 mg nifedipine daily or placebo and over a 10 month follow-up period found no significant effect on re-infarction rate (6.4% vs 6.5%) or mortality rate (5.8% vs 5.7%) [63]. A further study randomized 1373 postinfarct patients to receive 60 mg nifedipine daily or placebo and after a 6 month follow-up period demonstrated a nonsignificant excess of re-infarctions (5.1% vs 4.2%) and mortality (18.7% vs 15.6%) among the treated patients [64]. The Multicentre Diltiazem Post-Infarction Trial research group randomized 2466 postinfarct patients to receive 120–240 mg diltiazem daily or placebo [65]. During a 25-month follow-up period, no significant reduction in reinfarction (8.0% vs 9.4%) nor overall mortality (13.5% vs 13.5%) was noted. The DAVIT II study assigned randomly 1775 postinfarction patients to receive 360 mg verapamil daily or placebo [66]. During a mean 16 month follow-up period a significant reduction in re-infarction was noted (11.0% vs 13.2%, P < 0.04) but there was no significant impact on overall mortality (11.1% vs 13.8%). On the basis of the available evidence routine use of any calcium channel blocker for postinfarct patients is not justified. However, a rate-limiting agent (diltiazem or verapamil) is an option in patients unable to tolerate a β-adrenoceptor blocker.

Anti-arrhythmic drugs

After myocardial infarction the presence of frequent ventricular extrasystoles or ventricular dysrhythmias on an exercising ECG is associated with a poor prognosis and sudden cardiac death, especially in the setting of impaired left ventricular function [67, 68]. This has stimulated interest in the potential use of antiarrhythmic drugs to reduce postinfarct mortality. The Cardiac Dysrhythmias Suppression Trial randomized postinfarction patients with asymptomatic ventricular extrasystoles to flecainide, encainide, moracizine or placebo [69]. The study was terminated early because of an excess of deaths in the patients receiving flecainide and encainide, the pro-arrhythmic effects being most noteable in the setting of myocardial ischaemia. A trial of mexilitine also failed to demonstrate clinical benefit in spite of proven suppression of ventricular dysrhythmias on ambulatory monitoring [70]. An overview of the randomized controlled trials of Class I antiarrhythmic drugs given to survivors of acute myocardial infarction found an overall excess mortality in those assigned to active treatment [71].

There has been speculation that the apparently deleterious effects may have been due to the negative inotropic activity associated with many antiarrhythmic drugs. Amiodarone is a Class III antiarrhythmic drug which has less impact on ventricular function. The Polish amiodarone study, a small randomized trial of amiodarone in 613 postinfarct patients who could not tolerate β-adrenoceptor blockers, showed a significant reduction in cardiac mortality (6.2% vs 10.7%) and ventricular dysrhythmias (7.5% vs 19.7%) in the treated group after a 1 year follow-up [72]. EMIAT randomized 1486 survivors of a myocardial infarction with an ejection fraction of 40% or less to amiodarone or placebo and showed a 35% reduction of arrhythmic deaths (4.4% vs 6.7; P = 0.05) but no significant reduction of cardiac (11.4% vs 12.0%) or all-cause (13.9% vs 13.7%) mortality during a median 21 month follow-up [73]. CAMIAT randomized 1202 survivors of myocardial infarction who had frequent ventricular extrasystoles or a salvo of ventricular tachycardia to receive amiodarone or placebo. The trial showed a significant reduction of ventricular fibrillation and arrhythmic death (1.77% vs 3.38% per annum, P = 0.016) with greatest reduction in patients with congestive cardiac failure [74]. However, 64% of the patients assigned amiodarone withdrew from treatment and the benefit was no longer evident when intention-to-treat analysis was applied. A meta-analysis of 13 randomized controlled trials of amiodarone involving 6553 patients with previous myocardial infarction (78%) or cardiac failure demonstrated an overall mortality reduction of 13% (1–22% P = 0.03) and reduction of dysrhythmia or sudden death by 29% (15–44% P = 0.0003) [75]. The strongest predictor of sudden death was symptomatic heart failure and amiodarone was found to have greatest impact on postinfarct patients with cardiac failure. Adverse effects occurring in the amiodarone group (odds ratios compared to control) over a mean follow-up of 1.1 years were hypothyroidism 7% (7.3), hyperthyroidism 1.4% (2.5), peripheral neuropathy 0.5% (2.8), lung infiltrates 1.6% (3.1), bradycardia 2.4% (2.6) and altered liver function 1.0% (2.7).

(+)-Sotalol is a potassium channel blocker (with little β-adrenoceptor blocking activity) and a class III antiarrhythmic drug that is generally better tolerated than amiodarone. The SWORD trial randomized 3121 postinfarct or cardiac failure patients to receive 200–400 mg of (+)-sotalol daily or placebo. The trial was discontinued because of an increase in fatality rate among the treated group (5.0% vs 3.1%) due to arrhythmic deaths, most often in patients with ejection fraction less than or equal to 30% [76].

On the basis of the above data, no antiarrhythmic (β-adrenoceptor blockers aside) can be broadly recommended for secondary prevention in postinfarct patients. Amiodarone may usefully prevent dysrhythmia or sudden death in appropriately selected very high risk patients with symptomatic heart failure and pre-existing dysrhythmias. More information about the specific role that antiarrhythmic drugs might play should be forthcoming from further trials or subgroup meta-analyses. A promising alternative approach for postinfarct patients at high risk for arrhythmic death is the use of implantable defibrillators. Although there is now good evidence of superior efficacy when compared with antiarrhythmic drugs [77], the cost per life year gained is significantly greater than that with most other secondary preventative measures [78].

Anticoagulation

A number of studies have randomised patients to receive routine anticoagulation with warfarin (or other coumarins) or placebo after a myocardial infarction. The ASPECT trial (using nicoumalone or phenprocoumon vs placebo) followed up patients for a mean of 37 months and found a significant reduction in reinfarction (6.7% vs 14.2%), cardiovascular events (14.1% vs 21.5%) and stroke (2.2% vs 3.6%) and nonsignificant reduction of all-cause mortality (10% vs 11.1%) [79]. Major bleeding was increased by fourfold in the treatment group although the target INR of 2.8–4.8 was relatively aggressive. The WARIS trial randomised survivors of acute myocardial infarction to warfarin or placebo and found a significant reduction in recurrent infarction, cerebrovascular events and all cause mortality with an excess of major extracranial bleeding [80]. The reduction in recurrent vascular events and all-cause mortality compared favourably with studies of antiplatelet therapy vs placebo although oral anticoagulation was associated with significantly more major extracranial haemorrhages [81]. The excess of haemorrhagic strokes in patients assigned to anticoagulation was more than offset by a reduction in ischaemic stroke.

Direct comparison of aspirin and oral anticoagulation has been made in three trials. The German-Austrian trial showed a reduction in all-cause mortality of 26% (NS) and coronary mortality of 46% (NS) in patients treated with aspirin vs phenprocoumon over 2 years of follow-up [82]. Follow-up for 29 months of patients randomized in the EPSIM trial showed mortality was 11.1% with aspirin and 10.3% with oral anticoagulants (NS) [83]. The Coumadin Aspirin Reinfarction Study (CARS) randomly assigned 8803 postinfarct patients to daily doses of either 160 mg aspirin, 3 mg warfarin plus 80 mg aspirin or 1 mg warfarin plus 80 mg aspirin over a median follow-up of 14 months [84]. Neither combination showed any benefit over aspirin monotherapy.

It can be concluded that both oral anticoagulants and aspirin are more efficacious than placebo in the postinfarct patient. For survivors of uncomplicated myocardial infarction lack of proven superiority and the greater cost, complexity and hazards of oral anticoagulation favour the administration of aspirin. Special subgroups such as those with atrial fibrillation, severe congestive cardiac failure and mobile mural thrombus are at high risk of systemic embolization and are likely to derive greater benefit from oral anticoagulation which should replace aspirin.

Other treatments

Hormone replacement therapy. The risk of coronary heart disease is low in premenopausal women but increases rapidly as endogenous oestrogen levels fall following the menopause prompting speculation that oestrogens have a significant cardioprotective effect. Their potential cardiovascular benefits include lowering LDL-cholesterol and lipoprotein (a), increasing HDL-cholesterol, an antioxidant action and inhibition of platelet adhesiveness. Concerns that the more popular combined oestrogen-progesterone preparations might blunt the beneficial effects on lipoprotein metabolism seem unfounded [85], and combination therapy has been associated with significant protection in cohort studies [86, 87]. A recent multicentre prospective trial randomised 2763 postmenopausal women aged 66.7 years (mean) with known coronary heart disease to a combined oestrogen and progestogen preparation or placebo [88]. Over a mean 4.1 years follow-up period, 11% reduction of LDL cholesterol and 10% elevation of HDL cholesterol was noted but no overall mortality reduction was reported. An early increase of deaths due to thromboembolic and gall-bladder disease were found to offset later benefits in cardiovascular mortality. This result should temper enthusiasm for widespread use of oestrogens as a secondary prevention measure and illustrate the need for further randomized trials in high risk patients.

Antioxidant therapy. Many of the important risk factors for the development of coronary disease are associated with increased oxidative stresses which may, in turn, facilitate the accumulation of lipids in the vascular wall and inactivate nitric oxide, a major protective factor. Epidemiological studies have supported the possible cardioprotective effects of natural antioxidants, notably vitamin E [89]. The recent CHAOS study randomized 2002 patients with angiographically proven coronary disease to vitamin E (400IU or 800IU) or placebo and followed them up for a mean of 510 days [90]. Patients receiving vitamin E had a significant reduction in major cardiovascular events (4.0% vs 6.4%; P = 0.015) and nonfatal infarction (1.4% vs4.2%; P = 0.0001) but there was no effect on cardiovascular death (2.6% vs 2.4%). The recent HOPE study failed to demonstrate any benefit of 400IU vitamin E in patients at high risk of coronary disease for a variety of reasons [91]. Further information about the impact of antioxidant therapy will soon be available from ongoing studies.

Lifestyle factors. Patients who have suffered a myocardial infarction are usually particularly receptive to advice about changes in lifestyle. Smoking cessation is associated with a significant reduction in mortality [92] and delayed onset of postinfarct angina [93]. Regular aerobic exercise is associated with a variety of benefits in postinfarct patients including a reduction in blood pressure and weight, increased HDL concentrations, and increased sense of well-being. An overview of the randomised trials of exercise programs suggests that long-term mortality may be reduced [94, 95]. The role of dietary modification in preventing coronary disease has been controversial but the results of recent trials of dietary modification after a myocardial infarction have been more promising. These have focused on qualitative changes in the diet such as increasing consumption of fruits, vegetables, fish and monounsaturates rather than lipid-lowering and have demonstrated significant reductions in further coronary events [96–98].

Coronary revascularization. Revascularization procedures can undoubtedly relieve anginal symptoms poorly responsive to medical therapy and improve quality of life. Coronary artery bypass grafting (CABG) has been unequivocally demonstrated to be superior to medical therapy alone in prolonging survival in subgroups of patients with left main stem disease and those with triple vessel disease and impaired left ventricular function [99]. Aggressive lipid-lowering regimens delay the subsequent rate of graft occlusion [100, 101]. No study has yet been able to show long-term mortality benefits of percutaneous transluminal angioplasty (PTCA) over medical therapy in chronic stable angina due to single vessel disease [102, 103]. A major problem is the high early restenosis rate following PTCA although this is now being challenged by the use of intravascular stents [104] and drugs such as ticlopidine (a specific GpIIbIIIa antagonist) to prevent re-occlusion [105]. Trials comparing PTCA and CABG excluded many of the patients most likely to benefit from surgery and showed no superiority of either technique [106, 107]. However, those receiving PTCA were more likely to require another revascularization procedure within the first year and more likely to be suffering from angina. Routine medical follow-up of postinfarct patients should be offered in order to identify those at highest risk and most likely to benefit from intervention. Secondary preventative strategies such as antiplatelet and lipid-lowering therapy continue to offer benefits even after revascularization.

The patient was discharged from hospital 6 days following admission. His discharge medications were aspirin 150 mg once daily, an ACE inhibitor and a statin (his cholesterol level was 6.7 mmol l⁻¹on the admission). At this point there was no clinical evidence of heart failure and he had a good exercise tolerance with no evidence of postinfarct angina. An exercise tolerance test had been arranged for 4 weeks after discharge. He had also received advice about diet and exercise and been encouraged to take part in the hospital's cardiac rehabilitation programme. An outpatient appointment was made for 6 weeks after discharge to assess his recovery, review the result of his exercise test, ensure adequate titration of his drugs and consider the possibility of introducing a β-adrenoceptor blocker.