The terminology of reinfarction can be confusing: the term reinfarction is used when there is new myocardial necrosis, infarct extension when an area of infarction extends to involve adjacent ischemic tissue⁴ and infarct expansion when the infarcted tissue expands and contributes to hemodynamic deterioration. Infarct expansion does not require revascularization, but is treated by measures to control left ventricular dysfunction (see below). Reinfarction in patients with either non-ST segment elevation myocardial infarction (NSTEMI)⁵ or ST elevation MI (STEMI)⁶ may represent an extension of the initial necrosis or a new episode of infarction. The new universal definition of myocardial infarction recommends a diagnosis of reinfarction based on typical clinical features with a troponin elevation of >20% above earlier levels.³ The evidence that the risk of reinfarction in NSTEMI is lessened by early angiography and revascularization is now strong. In a meta-analysis of seven trials with 8375 patients, all -cause mortality at one month was reduced by 25% at two years (relative risk (RR) 0.75, 95% confidence interval (CI) 0.63–0.90, P = 0.001) and the incidence of non-fatal myocardial infarction was reduced by 17% (RR 0.83, 95% CI 0.72–0.96, P = 0.012) in the early invasive group compared with the conservative group.⁷ In STEMI treated with fibrinolysis, reinfarction can occur in 4–8% of patients. The reinfarction rate is significantly lower when primary percutaneous coronary intervention (PCI) is the initial treatment.⁸

Coronary reperfusion with fibrinolysis or PCI is strongly recommended and reduces the risk of reinfarction in STEMI (Class I, Level A).

In STEMI, PCI achieves better rates of reperfusion and is preferable to the use of fibrinolysis when available (Class I, Level A).

The risk of reinfarction in NSTEMI is reduced with early invasive investigation and revascularization where appropriate (Class I, Level A).

Clinical features and prognosis

Pathophysiology

Adverse remodeling of the ventricle can occur immediately after coronary occlusion⁹ and continues over the ensuing months and years, leading to an increase in end-diastolic and end-systolic volumes, an increase in the sphericity of the ventricle, and systolic bulging and thinning of the infarct zone, without necessarily any extension of the infarcted zone.¹⁰ Results from autopsy studies suggested that MIs that involved greater than 40% of the left ventricle were usually fatal.¹¹ However, a more recent prospective study conducted in the reperfusion era showed that out of 16 patients with infarcts involving >40% of the myocardial mass and followed for 13 months, only one had persistent heart failure and subsequently died.¹² Extensive damage can occur as a consequence of one large infarction or multiple smaller ones. NSTEMI may also cause left ventricular dysfunction if there has been prior cumulative myocardial damage.¹³

Prognostic markers based on left ventricular dysfunction

The extent of LV dysfunction is a strong predictor of short -and long-term prognosis after MI. The Killip and Kimball¹⁴ classification stratifies MI patients from low to very high risk based upon clinical signs of heart failure. It remains a reasonably accurate predictor of short-term survival. In patients undergoing primary percutaneous transluminal coronary angioplasty (PTCA), the in-hospital mortality was 2.4%, 7%, and 19% for Killip class I, II, and III respectively and six-month mortality was 4%, 10%, and 28% for class I, II, and III, respectively.¹⁵ The Forrester classification, comprising four categories defined according to the presence or absence of pulmonary congestion and peripheral hypoperfusion, requires measurement of the pulmonary artery pressure using a balloon flotation catheter.¹⁶ In postinfarction patients,¹⁷ abnormal hemodynamic variables determined from right heart catheterization correlate strongly with a higher mortality even after adjusting for other prognostic variables. However, insertion of balloon flotation catheters, while safe in experienced hands, has a recognized risk of adverse events, including ventricular tachyarrhythmias and pulmonary hemorrhage or infarction.¹⁸ Meta-analyses of trials using pulmonary artery cath-eterization have failed to show any benefit on outcomes,¹⁹ while increasing length of hospital stay.²⁰ Recent guidelines recommend the use of balloon flotation catheters only in severe or progressive congestive heart failure (CHF) or pulmonary edema, cardiogenic shock or progressive hypotension or suspected mechanical complications of acute infarction, i.e. ventricular septal defect (VSD), papillary muscle rupture or pericardial tamponade.⁶

Late (>30 days) postinfarction mortality was 3% in patients with an ejection fraction (EF) above 0.40, 12% when the EF was between 0.20 and 0.40, and 47% when it was below 0.20.²¹ The presence of clinical signs of left ventricular failure is a strong indicator of a poor long-term prognosis but in most patients, more detailed assessment is necessary and the use of echocardiography or radionu-clide assessment may provide information that cannot be obtained clinically.²² Approximately two-thirds of patients with an ejection fraction low enough to indicate a poor long-term prognosis, e.g. < 0.40, have no radiologic evidence of left ventricular failure.²³ The information obtained from assessing left ventricular function by echocardiography, radionuclide imaging or cardiac catheterization has been found to be of equivalent value in predicting one-year prognosis²⁴ and the choice of modality for assessment of left ventricular function depends on local availability and expertise. The left ventricular chamber volume has long been established as an important marker of long -term prognosis after myocardial infarction,²⁵ and the prevention of adverse remodeling by early coronary reperfusion is an important therapeutic aim in current treatment of myocardial infarction.²⁶ However, despite successful restoration of epicardial blood flow, left ventricular remodeling is still seen in a substantial number of patients, probably as a result of microvascular dysfunction in the coronary vessels.²⁷ Various methods of documenting this phenomenon have included measurements of the extent of recovery of ST segment elevation,²⁸ the myocardial blush grade on coronary angiography,²⁹ an index of microcirculatory resistance measured at angiography,³⁰ cardiac magnetic resistance imaging (MRI)³¹ and myocardial contrast echo-cardiography.³² The latter technique has been shown in a multicenter study to be superior to measurements of ST segment resolution and myocardial blush grade in predicting adverse remodeling.³² Simple bedside scores based on readily available clinical data can also accurately stratify the risks of death and ischemic events in STEMI patients, as shown in large clinical trials such as the GUSTO,³³ GISSI³⁴ and TIMI trials,³⁵ in NSTEMI, as shown in the TIMI trials,³⁶ and in non-STE acute coronary syndromes as shown in the Global Registry of Acute Coronary Events (GRACE).³⁷

Measurements of left ventricular function and diastolic volume predict short-and long-term prognosis after STEMI (Level A).

Clinical markers of prognosis can be used for risk stratification after STEMI and NSTEMI with a high degree of accuracy (Level A).

Biochemical markers

Biochemical markers of necrosis provide an index of the extent of left ventricular infarction, which in turn is correlated with the extent of left ventricular dysfunction. Creatine kinase was shown in the pre-reperfusion era to predict short-and long-term prognosis³⁸ but the introduction of reperfusion into routine clinical practice has reduced the utility of creatine kinase (CK) or CK-MB to reflect the extent of left ventricular dysfunction because of early, direct release of the myocardial enzymes into the plasma during reperfusion and high, early peaking of the serum levels. The use of newer markers such as troponins is now widespread. Both troponin-I³⁹ and troponin-T⁴⁰ correlate well with prognosis, and 96-hour troponin levels correlate well with infarct size.⁴¹ B-type natriuretic peptide⁴² and high sensitivity C-reactive protein⁴³ also correlate with cardiac failure and predict long-term prognosis after acute MI (AMI). A multimarker approach to using these bio-markers has been suggested, as patients with more than three abnormal biomarker levels have a worse prognosis than those with normal levels of biomarkers (hazard ratio (HR) 3.8 after correction for standard prognostic factors).⁴⁴

Ninety-six hour troponin levels provide a quantitative index of the extent of infarction (Level B).

Management

Reperfusion therapy

While the role of early coronary reperfusion in improving outcomes has been well established, the relationship between improvements in left ventricular function and prognosis after reperfusion therapy has been surprisingly difficult to demonstrate. Although some of the early studies demonstrated clear benefits on left ventricular function from coronary thrombolysis,^45,46 the evidence since then has been conflicting, with some groups showing a worse left ventricular function despite an improved prognosis. In a meta-analysis of 10 studies enrolling 4088 patients treated with thrombolytic therapy versus control, only a modest improvement in left ventricular function was demonstrated after thrombolytic therapy.⁴⁷ By four days, mean LV ejection fraction was 53% versus 47% (thrombolytic versus control therapy, P < 0.01); by 10–28 days it was 54.1% and 51.5%, respectively. The reason for the discrepancy in the marked improvement in survival and the limited benefit on left ventricular function is not clear but is at least in part due to the fact that more patients not receiving fibrinolytic therapy die before LV function can be assessed post MI. Patients who have had coronary reperfusion after MI may have myocardium that is stunned⁴⁸ or even hibernating,⁴⁹ phenomena that may affect the assessment of ventricular function, and the preinfarction left ventricular function obviously impacts on the LV function during and after myocardial infarction. Stunned myocardium has been successfully reperfused but has not regained its normal contractile function, but has the potential to recover without any further reperfusion or revascularization. A study of 352 patients with anterior MI found that out of the 252 patients with abnormal LV function on day 1, 22% had complete and 36% had partial recovery of function by day 90.⁵⁰ Hibernating myocardium is underperfused and non-contractile because of persisting ischemia, but is not infarcted and may gradually improve its function with revascularization. The degree of success in achieving coronary patency with thrombolysis is an obvious confounding factor.⁵¹ A detailed analysis correcting for these confounding factors concluded that left ventricular function is improved by successful coronary reperfusion.⁵²

Overall, reperfusion therapy results in a modest improvement in systolic LV function (Level B).

Hemodynamic therapy

Improvements in hemodynamic status after myocardial infarction have not translated into better outcomes. For example, furosemide has been shown to reduce elevated L V filling pressures without adversely affecting cardiac output⁵³ but there is no evidence of improvement in outcomes with diuretic therapy in AMI. Nitrates have been shown to improve the hemodynamic status and adverse remodeling post AMI⁵⁴ but large clinical trials have not shown improved prognosis.^55,56 The calcium-sensitizing agent levosimendan has been evaluated in post-MI patients with cardiac failure and showed only a non-statistically significant trend towards improved outcome.⁵⁷ Infusion of the B-type natriuretic peptide nesiritide for acutely decompensated heart failure, including some patients with acute myocardial infarction, was initially reported to be safer than inotropic agents⁵⁸ but a meta-analysis of all trials showed an apparent adverse effect on outcomes.⁵⁹ In contrast, aldosterone antagonists have been shown to be ben-eficial in post-MI patients with left ventricular dysfunction and pulmonary congestion. The Randomized Aldactone Evaluation Study (RALES) studied patients with New York Heart Association class III–IV heart failure, many with remote myocardial infarction. Spironolactone treatment compared with placebo was associated with an 11% absolute risk reduction and 30% relative risk reduction in all-cause mortality over 24 months of follow-up.⁶⁰ The Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS) demonstrated that in post-MI patients with an ejection fraction < 0.40 and heart failure or diabetes that compared with placebo, eplerenone significantly reduced overall mortality, cardiovascular mortality, and cardiac hospitalizations.⁶¹

Diuretics and nitrates can be used for relief of pulmonary congestion in acute MI, but there is no clinical trial evidence of benefit on outcomes (Class I, Level B).

Aldosterone antagonists (spironolactone and eplere-none) are indicated in post-MI patients with left ventricular dysfunction or pulmonary congestion (Class I, Level A).

Levosimendan may be considered as an alternative treatment in patients with severe cardiac failure following MI (Class IIa, Level B).

Nesiritide should not be used in patients with pulmonary congestion complicating myocardial infarction because of concerns about its safety profile (Class III, Level A).

ACE inhibitors for post-MI patients with left ventricular dysfunction

The beneficial effects of angiotensin-converting enzyme (ACE) inhibitors in the treatment of patients with left ventricular dysfunction complicating myocardial infarction have been striking. Eight large randomized, placebo -controlled trials have assessed the effect of an ACE inhibitor on mortality after MI. ACE inhibitors unequivocally reduce mortality overall, and the benefit appears to be greatest among patients with depressed LV function, overt heart failure or anterior infarction.^62–69

In a meta-analysis of data from all the randomized trials involving more than 1000 patients in whom ACE inhibitor treatment was started within 36 hours of onset of myocardial infarction, there were results available on 98 496 patients from four eligible trials.⁷⁰ Among patients allocated to ACE inhibitors there was a 7% (95% CI 2–11%; 2P < 0.004) proportional reduction in early mortality, an absolute reduction of five (SD, 2) deaths per 1000 patients. The absolute benefit was greatest in those patients with evidence of left ventricular dysfunction (i.e. Killip class 2–3, heart rate >= 100 bpm at entry) and in anterior MI. ACE inhibitor therapy also reduced the incidence of non-fatal manifestations of left ventricular dysfunction. During longer-term follow-up of patients enrolled in randomized controlled trials, ACE inhibitors have also been shown to be effective. In three long-term follow-up trials involving 5966 postinfarction patients, mortality was significantly lower with ACE inhibitors than with placebo (odds ratio (OR) 0.74, 95% CI 0.66 –0.83).⁷¹ Despite concerns that aspirin may blunt the effect of ACE inhibitors,⁷² a detailed analysis of six long-term trials showed no adverse interaction between aspirin and ACE inhibitors.⁷³ The optimum timing of initiation of ACE inhibitor therapy has been studied in only a small number of direct comparative trials. In a direct comparison of early versus delayed administration, 352 patients with acute anterior myocardial infarction were randomized to early (1–14 days) or late (14–19 days) post -MI treatment with the ACE inhibitor ramipril and were followed by echocardiography. Those receiving early treatment had a greater improvement in ejection fraction, suggesting that such patients should be commenced on ACE inhibitor therapy early in their course of infarction.⁷⁴

In considering treatment for left ventricular dysfunction, the hemodynamic benefits need to be balanced against the possible adverse effect. In the only ACE inhibitor trial that did not show a mortality benefit, CONSENSUS-II, treatment was begun early with an intravenous ACE inhibitor.⁶²

All patients with post-MI LV dysfunction should be administered ACE inhibitors (Class I, Level A).

ACE inhibitors for post-MI patients without left ventricular dysfunction

Whether low-risk postinfarction patients without LV dysfunction derive benefit from ACE inhibitors is still controversial. The clear-cut benefits demonstrated in the HOPE⁷⁵ trial of ramipril in high-risk patients with coronary artery disease and the EUROPA trial of perindopril in post-AMI patients without left ventricular dysfunction⁷⁶ were not supported by the PEACE trial of trandolopril in lower risk patients.⁷⁷ A subsequent meta-analysis of the three trials supported the use of ACE inhibitors in the absence of LV dysfunction⁷⁸ and the totality of the evidence was examined in a meta-analysis of six trials including 16 772 patients randomized to ACE inhibitors and 16 728 patients randomized to placebo, which concluded that the benefits were favorable though modest,⁷⁹ consistent with the conclusion that the absolute benefit is proportional to the risk, with those at lowest risk benefiting least.⁵

ACE inhibitors are indicated for post-MI patients even in the absence of left ventricular dysfunction, but are more beneficial in patients at higher risk (Class IIa, Level A).

Clinical features and prognosis

Cardiogenic shock is a syndrome characterized by hypotension and peripheral hypoperfusion, usually accompanied by high LV filling pressures. The common clinical manifestations of these hemodynamic derangements include mental obtundation or confusion, cold and clammy skin, and oliguria or anuria. Cardiogenic shock is the most common cause of in-hospital mortality after MI.⁸⁰ When cardiogenic shock is not secondary to a correctable cause, such as arrhythmia, bradycardia, hypovolemia or a mechanical defect, short-term mortality is 80% or higher, depending upon the strictness of the definition. Despite the major improvements in treatment in the past two decades, the in-hospital mortality in a recent international registry for patients with cardiogenic shock treated with modern therapy in the late 1990s was 66%.⁸¹ Old age, diabetes, previous infarction and extensive infarction as assessed either by enzymatic or electrocardiographic criteria are factors commonly associated with cardiogenic shock. Among patients receiving fibrinolytic therapy, the risk of cardiogenic shock increased by 47% with each decade increase in age.⁸² Patients who survive to 30 days after cardiogenic shock have an excellent late survival, almost as good as patients without shock.⁸³

Management

Inotropic drugs have been subjected to detailed study and widespread use in cardiogenic shock, but no benefit on mortality has been demonstrated.⁸⁴

Newer drugs such as levosimendan may improve myocardial efficiency by their calcium-sensitizing effect, but levosimendan also has vasodilating properties, which make it unsuitable for patients in cardiogenic shock with significant hypotension.⁸⁵ Nesiritide has been shown to improve pulmonary capillary wedge pressure in patients with decompensated heart failure⁸⁶ but was subsequently shown to increase mortality.⁸⁷

Intra-aortic balloon pumping has been used to stabilize patients with cardiogenic shock. Clear-cut benefits on hemodynamic status and coronary blood flow have been shown, but benefits on survival have not; in-hospital mortality remained at 83% despite the use of balloon pumping in a co-operative clinical trial.⁸⁸ Nevertheless, intra-aortic balloon pumping has a clear place in stabilizing the unstable cardiogenic shock patient for more definitive treatment such as coronary angioplasty or bypass surgery,⁸⁹ as has been demonstrated to improve coronary patency in a randomized trial in the setting of rescue angioplasty.⁹⁰ Newer methods of circulatory support have shown highly encouraging results^91,92 but benefits on survival remain to be established.

Fibrinolysis

Although the outcome of cardiogenic shock has been shown to be dependent on the patency of the infarct -related artery, clinical trials of fibrinolytic therapy have not shown a benefit in patients with established cardiogenic shock.⁹³ Alternative antithrombotic strategies may improve outcomes, but data are limited to observational studies.⁹⁴

Fibrinolysis could be considered for the patient with car-diogenic shock if access to PCI is not readily available (Class IIb, Level B).

Percutaneous coronary intervention or CABG

Observational studies on the use of PCI in cardiogenic shock suggest that an aggressive approach with early revascularization reduces the mortality of patients with cardiogenic shock after MI,⁹⁵ and a registry report has suggested that an aggressive approach with reperfusion therapy and intra-aortic balloon pulsation treatment of patients in cardiogenic shock due to predominant LV failure is associated with lower in-hospital mortality rates than standard medical therapy.⁹⁶ In a controlled clinical trial of an aggressive approach involving early catheterization with revascularization and intra-aortic balloon pumping, in cardiogenic shock patients (the SHOCK trial),⁹⁷ 87% of patients in the invasive arm underwent revascularization (surgical or percutaneous). There was a clear trend at 30 days towards reduced mortality in the invasive group compared with the medical therapy group (46.7% vs 56.0%); however, this difference did not reach statistical significance. There was an early hazard in the first five days after assignment to the invasive approach, which was likely associated with procedure-related complications. However, after the first five days there was a statistically significant survival benefit in favor of the revascularization group, which persisted at one year, when survival in the early revascularization group was 46.7% compared with 33.6% in those treated with early medical stabilization (RR for death 0.7, 95% CI 0.54–0.95).⁹⁸

Evidence from clinical trials and registries supports invasive intervention in patients with cardiogenic shock post MI. These patients should undergo urgent coronary angiography with a view to coronary angioplasty or, in selected patients, coronary bypass surgery (Class I, Level A).

Clinical features and prognosis

Right ventricular (RV) infarction typically occurs in association with inferior or posterior MI, as a consequence of total occlusion of the right coronary artery proximal to its marginal branches⁹⁹ or of the proximal circumflex in patients with a dominant left coronary system. RV infarction was present in 54% of patients with inferior MI in one series, although clinical manifestations are usually evident in only 10–15%.¹⁰⁰ Right ventricular involvement in inferior infarction has been reported to increase the mortality by fivefold. A meta-analysis of six studies including 1198 patients con-firmed that RV myocardial involvement was associated with an increased risk of death (OR 3.2, 95% CI 2.4–4.1), shock (OR 3.2, 95% CI 2.4–3.5), and arrhythmic complications.¹⁰¹ Patients in the SHOCK trial registry with cardiogenic shock due to RV infarction had the same serious prognosis (hospital mortality in excess of 50%) as patients with shock due to predominant LV infarction.¹⁰² Early reperfusion may improve the prognosis: in patients who have primary PCI for treatment of their cardiogenic shock, those with right ventricular infarction fared better than patients with cardio-genic shock due to left ventricular infarction.^103,104

The clinical features of RV infarction complicating inferior MI include hypotension, an elevated jugular venous pressure and clear lung fields. Jugular venous distension on inspiration (Kussmaul’s sign) has been reported to be a sensitive and specific sign of RV infarction.¹⁰⁵ The hemodynamic features of RV infarction may disappear with volume depletion or may emerge only after volume loading, making the clinical diagnosis elusive in some cases.

ST segment elevation in a right precordial lead (V _4R) has been reported to have a sensitivity of 70% and a specificity of nearly 100% for the diagnosis of RV infarction when the electrocardiogram is recorded within the first hours after the onset of symptoms. Echocardiography commonly reveals wall motion abnormalities of the right ventricle and interventricular septum. Bowing of the interatrial septum toward the left atrium indicates that the right atrial pressure exceeds the left atrial pressure,¹⁰⁶ and bowing of the interventricular septum into the right ventricle, compounding the dysfunction of the right ventricle,¹⁰⁷ both indicate a poor prognosis. Detection of a low RV EF and a segmental wall motion abnormality by radionuclide right ventriculography had a sensitivity of 92% and a specificity of 82% for identifying hemodynamically significant RV infarction in one study.¹⁰⁵ Assessment of RV function can help assess long-term prognosis, with patients with right ventricular ejection fraction having a risk of death four times that of patients with normal RV function.¹⁰⁸

Patients with RV infarction complicating inferior MI have three times the risk of death of patients without RV infarction (Level A).

Patients with cardiogenic shock due to RV infarction have the same poor prognosis (>50% hospital mortality) as patients with LV infarction (Level A).

Management

Volume loading can normalize blood pressure and increase cardiac output.¹⁰⁹ Earlier studies of RV infarction demonstrated a marked response to volume loading.¹¹⁰ Many of these patients were volume depleted secondary to aggressive diuresis in response to a raised venous pressure.

Volume loading can achieve hemodynamic improvement but has not been shown to improve outcomes (Class I, Level B).

Inotropic agents are often used in the treatment of right ventricular infarction when volume loading fails to improve cardiac output, but the effect of this on prognosis is unclear. The maintenance of atrioventricular synchrony is often critical to the maintenance of a satisfactory cardiac output; atrioventricular pacing has been shown to improve hemo-dynamics.¹¹¹ Successful reperfusion with fibrinolysis¹¹² or PCI¹¹³ appears to reduce the incidence of RV infarction and is associated with dramatic recovery of right ventricular function and reduced mortality. In contrast, unsuccessful right coronary artery reperfusion was associated with a high mortality.¹¹⁴

Reperfusion therapy in right ventricular infarction has not been studied in randomized trials but appears to be effective (Level B).

Clinical features and prognosis

Left ventricular aneurysms develop most commonly after large transmural anterior MIs, although in 5–15% of cases the site is inferior or posterior.¹¹⁵ The diagnosis of aneurysm is less frequent in the reperfusion era, although documented trends have not been published. The coronary anatomy is an important determinant of the development of left ventricular aneurysm. Total occlusion of the left anterior descending artery in association with poor collateral blood supply is a significant determinant of aneurysm formation in anterior MI. Multivessel disease with either good collateral circulation or a patent left anterior descending artery is uncommonly associated with the development of left ventricular aneurysm.¹¹⁶ Coronary patency also determines the likelihood of developing aneurysm.¹¹⁷ A ventricular aneurysm can often be palpated as a dyskinetic region adjacent to the apical impulse. A third heart sound and signs of heart failure may also be detected. A non-specific marker of an aneurysm is ST segment elevation that persists weeks after the acute phase of infarction. Echocardiography can delineate LV aneurysms as well as left ventriculography and has a higher sensitivity in the detection of thrombus.¹¹⁸ A left ventricular aneurysm may cause no problems, but may be associated with heart failure because the left ventricle functions at a mechanical disadvantage. Ventricular tachycardia late after infarction is commonly associated with an aneurysm, but its incidence may be reduced in patients receiving thrombolysis. In a non-randomized study of patients who developed a ventricular aneurysm after myocardial infarction, inducible ventricular tachycardia was less likely in patients who had received fibrinolytic therapy than those who had not (8% vs 88%; P = 0.0008) and there was a reduced incidence of sudden death on subsequent follow -up (0% vs 50%; P = 0.002).¹¹⁹ A ventricular aneurysm also provides a nidus for the development of an intracavitary thrombus. The risk of a clinical embolic event, based on four observational studies, is approximately 5%¹²⁰ and is greatest within the first few weeks post infarction.

Management

Surgical removal of a left ventricular aneurysm is indicated in patients with heart failure that is difficult to control medically, in patients with recurrent ventricular tachycardia not controlled by other means, and in patients with embolic episodes in spite of adequate anticoagulation.¹²¹ The earlier technique of linear excision of the aneurysm was associated with a high mortality, and has been supplanted by geometric reconstruction surgery in which the aneurysm is excluded from the left ventricular cavity, first described by Vincent Dor.¹²² An overview of the reported results of the newer approach suggested substantially better outcomes,¹²³ but recent RCT revealed no benefit.^123a

Surgical treatment of post-MI left ventricular aneurysms can help in the control of intractable cardiac failure (Class IIB, Level C).

A pseudoaneurysm is a rare complication of MI that develops when a myocardial rupture is sealed off by surrounding adherent pericardium. The aneurysmal sac may progressively enlarge but maintains a narrow neck, in contrast to a true ventricular aneurysm. In a series of 290 patients with LV pseudoaneurysms, congestive heart failure, chest pain and dyspnea were the most frequently reported symptoms, but >10% of patients were asymptomatic.¹²⁴ Physical examination revealed a murmur in 70% of patients. Almost all patients had electrocardiographic abnormalities, but only 20% of patients had ST segment elevation. Radiographic findings were frequently non -specific but a mass was detected in more than one half of patients. Differentiation of left ventricular pseudoaneurysms from true aneurysms may be difficult, and can be enhanced with echocardiography¹²⁵ or magnetic resonance imaging.¹²⁶ Regardless of treatment, patients with LV pseu-doaneurysms have a high mortality rate, but especially those who are managed non-surgically.¹²⁷ Surgical repair can be achieved, albeit with a high mortality.¹²⁸

Surgery should be considered in all patients with LV pseudoaneurysms (Class IIB, Level B).

Clinical features and prognosis

Rupture of the free wall of the left ventricle is an almost uniformly fatal complication of MI that now probably accounts for 10–20% of in-hospital deaths.¹²⁹ In the GISSI trial, cardiac rupture was the cause of 19% of the deaths among patients aged 60 years or younger and 86% of deaths among those more than 70 years old.¹³⁰ Rupture occurs most frequently in elderly women.¹³¹ Anterior infarctions, hypertension on admission and marked or persistent ST elevation are also risk factors for rupture.¹³² The usual presentation is sudden collapse, associated electrical -mechanical dissociation, and failure to respond to cardiopulmonary resuscitation, although in some patients ventricular rupture is subacute, allowing time for ante -mortem diagnosis.¹³³ This clinical entity is probably under -recognized and may be amenable to surgical repair. Premonitory symptoms of chest discomfort, a sense of impending doom and intermittent bradycardia signal impending myocardial rupture in many cases¹³⁴ and, if recognized, can lead to life -saving surgery¹³⁵ although, in one report¹³⁶ of 81 consecutive patients presenting with acute hypotension with electrical–mechanical dissociation, 19 survived with medical management alone. Urgent echocardiography is invaluable in the assessment of a patient who develops the above clinical features.¹³⁷

Overall, reperfusion reduces the frequency of rupture¹³⁸ but a meta-analysis of 58 cases of rupture involving 1638 patients from four trials showed that the odds ratio (treated/control) of cardiac rupture was directly correlated with time to treatment (P = 0.01); late administration of fibrinolytic therapy may increase the risk of cardiac rupture.¹³⁹

Management

Urgent surgical repair is mandatory for acute rupture¹⁴⁰ (Class I, Level B).

Clinical features and prognosis

Pericarditis occurs in approximately 25% of patients with Q-wave infarctions and 9% of patients with non-Q wave infarctions;¹⁴¹ it usually occurs within the first week.¹⁴² A pericardial friction rub may be present but is not found in half of patients with typical symptoms and is not required for diagnosis or treatment. On the other hand, the only evidence of pericarditis in many patients is a transient pericardial rub, with no symptoms. Pericarditis following myocardial infarction is associated with a higher risk of death in the year post infarction, possibly due to the associated large effusion.¹⁴¹

Management

High-dose aspirin and non-steroidal anti-inflammatory drugs (NSAIDs) are recommended to treat the symptoms of postinfarction pericarditis, although no randomized studies have been done to document their efficacy. Prolonged administration of NSAIDs should be avoided. A serial echocardiographic study of patients with postinfarction pericarditis showed that patients treated with indo-methacin or ibuprofen showed a greater tendency for infarct expansion, but it was not clear if the infarct expansion was due to the NSAIDs or to the selection for treatment of those with larger infarctions.¹⁴³ There is evidence that the adverse effects of the COX-2 inhibitors may affect endothelial function and enhance thrombosis by inhibiting the production of prostacyclin.¹⁴⁴ Recent reports indicate that the risks may be as high with all NSAIDs.^145,146 Until more observations become available, limited use of COX-2 inhibitors and NSAIDs in acute coronary syndromes is a sensible precaution.

A single dose or short-term treatment with a non-steroi-dal agent may be very effective for postinfarction pericarditis, but long-term therapy should be avoided (Class IIb, Level B).

Pericardial effusion and tamponade

A pericardial effusion can be detected by echocardiography in one-quarter of patients with acute Q-wave MI.¹⁴⁷ This finding correlates with the presence of heart failure and a poor prognosis. Cardiac tamponade is a rare complication of fibrinolytic therapy for acute MI, being reported in four of 392 consecutively treated patients in one series.¹⁴⁸ A large effusion and persistent pericarditis may be a sign of subacute rupture and should initiate consideration of surgical repair.

Dressler’s syndrome

A form of postinfarction pericarditis, occurring 2–11 weeks after the acute event, was described in 1956 by Dressler.¹⁴⁹ The full syndrome includes prolonged or recurrent pleuritic chest pain, a pericardial friction rub, fever, pulmonary infiltrates or a small pulmonary effusion, and an increased sedimentation rate. There has been a striking reduction in the incidence of this postinfarction complication since the introduction of reperfusion into clinical practice.¹⁵⁰

Non-steroidal anti-inflammatory drugs may be required for control of Dressler’s syndrome, but there are no randomized trials to confirm their efficacy (Class IIb, Level B).

Clinical features and prognosis

The overall risk of stroke after MI estimated from published community-based studies is 11.1 ischemic strokes during hospitalization per 1000 MI, 12.2 at 30 days and 21.4 at one year.¹⁵¹ In patients with large anterior STEMIs, left ventricular thrombi develop in up to 40%.¹⁵² If left untreated, up to 15% of thrombi will dislodge and result in a symptomatic embolic event¹⁵³ This risk is higher in patients with large anterior infarctions^154,155 and patients with atrial fibrillation.¹⁵⁶ Emboli are more common within the first few months after infarction than later, and with large, irregular shaped thrombi, particularly those with frond-like appendages.¹⁵⁷

When thrombus is visualized by echocardiography, the risk ratio for embolization is 5.45 (95% CI 3.0–9.8) according to a meta-analysis.¹⁵⁸ In NSTEMI, stroke is relatively uncommon but carries a high mortality. In a pooled analysis from six trials (n = 31 402) which included patients randomized to Gp IIb/IIIa inhibitors, there were 228 (0.7%) strokes: 155 (0.5%) non-hemorrhagic, 20 (0.06%) hemor-rhagic, and 53 without computed tomography (CT) confir-mation.¹⁵⁹ Older age, prior stroke and elevated heart rate were the strongest predictors of 30-day all-cause stroke. The risk of dying after a post-STEMI stroke is approximately 40%¹⁶⁰ and 25% after a NSTEMI stroke.¹⁶¹ Fibrino-lytic therapy is associated with an excess of stroke of four extra strokes on day one compared with placebo.¹⁶² This risk is reduced if angioplasty is used instead of fibrinolytic therapy. In a meta-analysis comparing the effects of angio-plasty with fibrinolysis, angioplasty was associated with a significant reduction in total stroke (0.7% vs 2.0%; P = 0.007) primarily due to a reduction in hemorrhagic stroke (0.1% vs 1.1%; P < 0.001).¹⁶³ In a Cochrane meta-analysis, the use of PCI compared with fibrinolysis significantly decreased the frequency of strokes of any cause by 66% (95% CI 28–84%) and no significant difference was observed for the incidence of major bleeding (RR 1.18, 95% CI 0.73–1.90) but the confidence intervals were large.¹⁶⁴

PCI results in a lower risk of post-MI stroke compared with fibrinolysis (Level B).