CHAPTER 40: ST-SEGMENT ELEVATION MYOCARDIAL INFARCTION

Ischemic heart disease is the leading cause of morbidity and mortality in Western society and is a worldwide epidemic. In 2008, it was estimated that, worldwide, ischemic heart disease was responsible for 12.7% of all deaths (7.25 million total). Low-income countries now account for 80% of all deaths; mortality is falling in high-income countries.¹ Approximately 1,100,000 Americans suffer from an acute coronary syndrome (ACS) per year, nearly one-third of which are caused by an acute ST-segment elevation myocardial infarction (STEMI).² The incidence of acute myocardial infarction (AMI) has declined over the past two decades from 244 per 100,000 population in 1975 to 162 per 100,000 population in 2006.³ The in-hospital mortality also has declined from 18% in 1975 to 10% in 2006.⁴ Despite these improvements, AMI continues to be a major public health problem, and it has been estimated that the number of years of life lost because of an AMI is 17.1 years, and the cost to American society (both direct and indirect) is $165.4 billion per year.²

The management of STEMI patients is complex, multidisciplinary, and involves the following four different stages of care: (1) prehospital care, (2) emergency department, (3) cardiac catheterization laboratory, and (4) coronary care unit. This chapter discusses the diagnosis and management of STEMI patients in each of these four settings. The pathophysiology of disease is discussed in Chap. 37, and the ACSs of unstable angina and non–ST-segment elevation myocardial infarction are discussed in Chap. 39.

Symptoms

The classic symptom of acute myocardial ischemia is precordial or retrosternal discomfort, commonly described as a pressure, crushing, aching, or burning sensation. Radiation of the discomfort to the neck, back, or arms frequently occurs, and the pain is usually persistent rather than fleeting. The discomfort typically achieves maximum intensity over several minutes and can be associated with shortness of breath, nausea, diaphoresis, generalized weakness, and a fear of impending death. Some patients, particular the elderly, may also present with syncope, unexplained nausea and vomiting, acute confusion, agitation, or palpitations. Symptoms in the advanced elderly (> 75 years old) are more likely to be atypical (new or worsening dyspnea) than in younger patients and can lead to a missed diagnosis if a medical professional is not vigilant in the initial assessment.

Approximately 20% of AMI patients are asymptomatic or have atypical symptoms that are not initially recognized. Painless myocardial infarction occurs more frequently in the elderly, women, diabetics, and postoperative patients. These patients tend to present with dyspnea or frank congestive heart failure (HF) as their initial symptom.⁵ Some patients with ST-segment elevation do not have occlusive thrombosis and have a unique regional cardiomyopathy called takotsubo syndrome, which is described at the end of the chapter.

Physical Examination

Patients often appear anxious and uncomfortable. Those with substantial left ventricular (LV) dysfunction at presentation may have tachypnea, tachycardia, pulmonary rales, and a third heart sound. The presence of a systolic murmur suggests ischemic dysfunction of the mitral valve apparatus or papillary muscle or ventricular septal rupture and needs to be distinguished from a pericardial friction rub secondary to pericarditis or possible partial rupture with adhesions.

In patients with right ventricular (RV) infarction, increased jugular venous pressure, Kussmaul sign (rise in jugular venous pressure with inspiration), and an RV third sound may be present. Such patients typically have inferior infarctions as a result of proximal right coronary artery occlusion, usually without evidence of left heart failure, and may have exquisite blood pressure sensitivity to nitrates or hypovolemia. In patients with extensive ongoing ischemia or LV dysfunction, shock is indicated by hypotension, diaphoresis, cool skin and extremities, pallor, oliguria, and possible altered mental status.

Electrocardiogram

The classic initial electrocardiographic manifestations of STEMI are discussed in Chap. 12 and involve an increase in the amplitude of the T wave (peaking), followed within minutes by ST-segment elevation. Diagnostic ST-segment elevation in the absence of LV hypertrophy or left bundle branch block (LBBB) is defined by the European Society of Cardiology (ESC)/American College of Cardiology Foundation (ACCF)/American Heart Association (AHA)/World Heart Federation Task Force for the Universal Definition of Myocardial Infarction as new ST-segment elevation at the J point in at least two contiguous leads of 2 mm (0.2 mV) in men or 1.5 mm (0.15 mV) in women in leads V₂ to V₃ and/or of 1 mm (0.1mV) in other contiguous chest leads or the limb leads.⁶

The R wave may initially increase in height but soon decreases, and often Q waves develop. If the jeopardized myocardium is reperfused, the ST segment may promptly decrease, although T waves can remain inverted, and Q waves may or may not regress. Persistent ST-segment elevation after restoration of flow in the epicardial coronary artery is a marker of abnormal myocardial perfusion and associated with an adverse prognosis. In the absence of reperfusion, the ST segment gradually returns to baseline in several hours to days, and T waves become symmetrically inverted. Failure of the T wave to invert within 24 to 48 hours suggests regional pericarditis.

The specific leads with ST-segment elevation can help localize the infarct (see Chap. 12): ST-segment elevation in the inferior, anterior, or high lateral leads is seen with infarction of the corresponding areas of myocardium. ST-segment elevation in lead aVR is an ominous sign and is more frequent in patients with left main artery occlusion than in patients with left anterior descending coronary artery or right coronary artery occlusion.⁷ In a study of STEMI patients, ST-segment elevation in lead aVR that was greater than or equal to the extent of ST-segment elevation in lead V₁ had 81% accuracy for diagnosing left main occlusion.⁷ ST-segment elevation in lead V₁ in the setting of inferior myocardial infarction suggests RV involvement. Because no leads on the standard 12-lead electrocardiogram (ECG) directly represent the posterior myocardium, isolated infarction of this area may be difficult to diagnose but is typically manifested by ST-segment depression in V₁ to V₃, a mirror image of anterior myocardial infarct.⁸ Depending on the affected artery (right coronary vs circumflex), posterior myocardial infarction is frequently associated with either inferior or lateral injury, with ECG manifestations of STEMI corresponding to those areas of myocardium. Sensitivity of detecting posterior myocardial infarction may be enhanced by the use of additional ECG leads (V₇-V₉) designed to represent the posterolateral myocardium.^9,10 ST-segment elevation in these additional leads is suggestive of posterior involvement.

New-onset (or not known to be old) LBBB in the setting of chest pain is typically considered and treated as an STEMI.¹¹ The diagnosis of STEMI in the setting of old LBBB can be difficult. Findings suggesting STEMI include (1) a pathologic Q wave in leads I, aVL, V₅, or V₆ (two leads); (2) precordial R-wave regression; (3) late notching of the S wave in V₁ to V₄; and (4) deviation of the ST segment in the same direction as that of the major QRS deflection.¹² Similar findings may be expected in RV pacing with LBBB structure of the QRS. In one study, patients with paced rhythm had higher in-hospital and 1-year mortality, in part because of undertreatment, which illustrates the difficulty of AMI diagnosis in this setting.^13,14

Analysis of ECG data from the GUSTO (Global Use of Strategies to Open Occluded Coronary Arteries) I study identified three criteria for diagnosing myocardial infarction in the presence of the LBBB: (1) ST-segment elevation ≥ 1 mm concordant with the QRS complex; (2) ST-segment depression ≥ 1 mm in leads V₁, V₂, or V₃; and (3) ST-segment elevation ≥ 5 mm discordant with the QRS.¹⁴

The ECG, however, has several limitations: (1) Even though ST-segment elevation usually signifies acute coronary occlusion, acute coronary occlusion may not cause ST-segment elevation in certain circumstances, such as a circumflex artery occlusion; (2) in addition, not all ST-segment elevation is caused by AMI, including conditions such as electrolyte abnormalities, myocarditis, and prior LV aneurysm; and (3) finally, an old LBBB limits the usefulness of the ECG for AMI diagnosis in this subset of patients.

Laboratory Studies

Damaged cardiomyocytes release several proteins in the circulation, including myoglobin, creatine kinase (CK) and its myocardial band isoenzyme (CK-MB), troponins (I and T), myoglobin, aspartate aminotransferase, and lactate dehydrogenase (see Chap. 38). Figure 40–1 shows the timing of release.¹⁵ Cardiac troponins are currently the preferred biomarkers for myocardial damage because of their high sensitivity and specificity.¹⁶ CK-MB is the best alternative, if cardiac troponin assays are not available. CK-MB, because of its more rapid appearance and disappearance from the blood, can be used (1) in patients presenting early after symptom onset; (2) to time the onset of injury if the troponin is increased; and (3) to detect reinfarction later in the hospital course. Determinations of total CK, aspartate aminotransferase, and lactate dehydrogenase are no longer recommended. Blood sampling for biomarker determination is recommended at hospital admission, at 6 to 9 hours, and at 12 to 24 hours if the earlier samples were negative and the clinical index of suspicion is high (see Chap. 38).¹⁶

Myoglobin

Myoglobin is a 17.8-kDa protein that is released from injured myocardial cells. As shown in Fig. 40–1, myoglobin release occurs within hours after the onset on infarction, reaches peak levels at 1 to 4 hours, and remains elevated for approximately 24 hours. Although the rapid rise allows for its use as an early marker for STEMI, myoglobin is not specific to myocardial cells and should not be used in isolation as a method for diagnosing myocardial infarction.

Creatine Kinase-MB

The MB isoenzyme of CK is present in largest concentration in myocardium, although small amounts (1%-2%) can be found in skeletal muscle, tongue, small intestine, and diaphragm. CK-MB appears in serum within approximately 3 hours after the onset of infarction, reaches peak levels at 12 to 24 hours, and has a mean duration of activity of 1 to 3 days.¹⁵ Other cardiac, but non-AMI etiologies of increased CK-MB levels can occur after cardioversion, cardiac surgery, myopericarditis, and percutaneous coronary intervention (PCI), and occasionally after rapid tachycardia. Noncardiac causes of increased CK-MB levels may occur with hypothyroidism, extensive skeletal muscle trauma, rhabdomyolysis, the muscular dystrophies, and some other neuromuscular disorders.

Occasionally, the concentration of CK-MB isoenzyme may be increased in the presence of normal total levels of CK enzyme. This finding usually indicates a small amount of myocardial necrosis in a patient whose baseline total CK enzyme level is at the low-normal end of the range (see Chap. 38).

Troponins

The cardiac troponins regulate the interaction of actin and myosin and are more cardiac specific than CK-MB. There are two isoforms of cardiac troponin: T and I. Their levels start to rise 3 to 12 hours after the onset of ischemia, peak at 12 to 24 hours, and may remain elevated for 8 to 21 days (troponin T) or 7 to 14 days (troponin I). Several new high-sensitivity troponin assays can detect lower concentrations of circulating troponins compared with previous assays. These assays were recently evaluated by two large multicenter studies, with corroborating results. The high-sensitivity assays proved more sensitive in the diagnosis of myocardial infarction than older assays (94%-96% vs 85%-90%) at the expense of a reduced specificity (90.2% vs 97.2%).^17,18 The high-sensitivity troponins are an integral part of diagnosis for myocardial infarction, including STEMI. There are emerging data that ultra–high-sensitivity troponin assays with ranges of detection 0.2 pg/mL provide prognostic information in ACS patients with undetectable high-sensitivity troponin levels.¹⁹ In summary, it is clear that elevated troponin levels correlate with pathologically proven myocardial necrosis and indicate poor prognosis in patients with suspected ACSs (see Chap. 36).^{20,21,22,23,24}

Prehospital Diagnosis and Regional Systems of Care

Rapid diagnosis is a pivotal component of the management of STEMI patients. Early diagnosis can be achieved with a prehospital ECG that is obtained in the field by emergency medical personnel and transmitted to an emergency department physician or cardiologist on-call. This allows for early administration of fibrinolytic therapy or activation of the cardiac catheterization team before arrival of the patient and has been demonstrated in several studies to reduce both the door-to-needle and door-to-balloon time.^25,26,27

Specifically, analysis of data from the National Registry of Myocardial Infarction revealed that only 4.5% (1599 of 35,370) of patients receiving thrombolytic therapy and 8% (1696 of 21,277) of patients undergoing primary PCI had a prehospital ECG.²⁶ The analysis also showed that a prehospital ECG resulted in a significantly shorter door-to-needle time (24.6 vs 34.7 minutes; P < .0001) for those receiving fibrinolytic therapy and a shorter door-to-balloon time (94 vs 110 minutes; P < .0001) for primary PCI patients.²⁶

The use of newer automated wireless technologies for the transmission of prehospital ECGs was studied in the STAT-MI (ST-Segment Analysis Using Wireless Technology in Acute Myocardial Infarction) trial.²⁸ In this study, the ECGs of 20 patients with STEMI were transmitted via cellular phones to a tertiary care center. Compared with data obtained in patients who did not receive prehospital ECGs, wireless transmission resulted in the improvement of the following times: door-to-intervention (80.1 vs 145.6 minutes; P < .001), door-to-cardiologist notification (14.6 vs 61.4 minutes; P < .001), door-to-arterial access (47.6 vs 108.1 minutes; P < .001), and time to first angiographic injection (52.8 vs 119.2 minutes; P < .001). These data further illustrate the importance of early evaluation and triage of patients with suspected STEMI to facilitate rapid and appropriate treatment. The availability and cost of this technology, however, may preclude its widespread use.

Based on these data and data from regional systems in both the United States and Europe, the American College of Cardiology (ACC)/AHA²⁹ and the ESC³⁰ guidelines for STEMI emphasize the importance of systems of care for patients with suspected AMI/STEMI. The ACC/AHA have developed regional systems of STEMI care aimed at improving the coordination and likelihood of getting time to treatment goals achieved.

Specifically, the ACC/AHA Guidelines²⁹ recommend the following:

The goal of the regional systems of care is to maximize the opportunity for timely reperfusion for STEMI patients. In that regard, the ACC/AHA STEMI guideline provide prehospital transport recommendations and criteria.

Patients with possible ischemic symptoms are encouraged to use EMS and ambulances rather than private vehicles for transportation to hospital because (1) one out of 300 patients with chest pain transported by private vehicle will suffer a cardiac arrest in transportation,³¹ and (2) as stated above, transportation with EMS is associated with earlier activation of reperfusion services and faster time to reperfusion.

Additionally, the guidelines specifically note that a prehospital ECG can activate the PCI team while in transport to the hospital, that the emergency department physician can also activate the PCI team, and that one call with feedback to the PCI team on time metrics should be the goal for reperfusion centers.

Figure 40–2 provides a schematic from the guidelines for considering transportation to regional hospitals for reperfusion as outlined in the AHA Mission Lifeline program.

The timely reperfusion goals include determining the closest primary PCI center for the preferred primary PCI (within 90 minutes of first medical contact), determining acceptable delays that allow transfer of patients to get primary PCI reperfusion within 120 minutes, and quickly treating patients who are outside of those windows with fibrinolytic therapy followed by possible transfer for invasive care as needed.

These regional system goals are captured in the following recommendations²⁹:

Given the transport time to hospitals and centers for primary PCI, the time delay can be substantial, and many have investigated delivering prehospital fibrinolytic therapy as a reperfusion strategy for patients in areas without primary PCI capabilities.

Multiple randomized trials have shown that this strategy is feasible and would reduce treatment delays, and a meta-analysis of six higher quality randomized controlled trials has demonstrated that a 60-minute reduction in treatment delay is possible with a resultant 17% reduction in mortality.³²

The cornerstone of STEMI therapy is a rapid and accurate evaluation at first medical contact. All patients presenting with complaints of chest discomfort should be rapidly triaged and allowed to bypass the emergency department waiting room. An ECG should be obtained within the first 10 minutes of arrival at the emergency department, and a focused history and physical examination assessing the symptoms and signs described in the diagnosis section of this chapter should be quickly performed.

The physical examination also provides a method for the risk stratification of STEMI patients. As shown in Table 40–1, the Killip classification can be used as a method to stratify patients and predict clinical outcomes, and it is noteworthy that this classification has stood the test of time.³³ With modern therapy, the mortality of those in cardiogenic shock has improved from 83% to approximately 40%.³⁴

In addition to the rapid evaluation of the ECG (within 10 minutes of arriving), guidelines have recommended a targeted physical exam in the emergency department focusing on the cardiovascular and neurologic systems, and repeated ECGs if the initial ECGs are not diagnositic.³⁵

The initial management of patients in the emergency department may be brief because the goal for STEMI is to reduce the time to reperfusion. In fact, over the last several years, the use of adjunctive therapies such as oxygen, β-blockers, and initiation of anticoagulation in the emergency department has decreased.

Oxygen

Low-flow oxygen therapy delivered by nasal cannula was previously recommended to be routinely given during the first 24 to 48 hours and perhaps several days after AMI in most patients. Although routine use of supplemental oxygen is common practice, hard evidence supporting its use during the acute phase of a STEMI is lacking.³⁶ In fact, a Cochrane analysis that pooled the results of three trials found a three-fold higher risk of death in patients getting oxygen with confirmed infarctions compared to patients on room air.³⁷ Supplementary oxygen may also increase coronary vascular resistance.³⁸ Therefore, oxygen is not routinely indicated but is used in patients with clinically significant hypoxemia and evidence of HF.

Aspirin

Aspirin decreases mortality in myocardial infarction and should be administered as early as possible and continued indefinitely in patients with ACSs. In cases of aspirin allergy or major intolerance, clopidogrel should be substituted. In the ISIS-2 (International Study of Infarct Survival 2), 77,187 STEMI patients were randomized to treatment with aspirin, streptokinase, or placebo. The 5-week incidence of vascular death decreased by 23% with aspirin and heparin, by 25% with streptokinase and heparin, and by 41% with the combination of all three agents.³⁹ Thus, the effect of aspirin was surprisingly nearly as great as that of streptokinase alone in that study, and the benefits of each were partially additive. Therefore, 160 to 325 mg of chewable aspirin should be given to patients at presentation, with a subsequent preferred dose of 81 mg daily. For those with a history of a documented significant adverse reaction to aspirin, 300 mg of clopidogrel can be used as an alternative.

The clinical practice guidelines recommend the following aspirin recommendations as part of STEMI care with reperfusion as a planned treatment²⁹:

P2Y₁₂ Inhibitors

Thienopyridines (clopidogrel and prasugrel) and now direct P2Y₁₂ inhibitors, such as ticagrelor, are oral antiplatelet agents that inhibit platelet activation through the adenosine diphosphate (ADP)-mediated pathway. Several factors contribute to the choice of antiplatelet agent (clopidogrel, prasugrel, or ticagrelor), as well as the initial dose. Unless contraindicated, these antiplatelet agents should be initiated upon presentation of STEMI. See the Adjuvant Antiplatelet Therapy section in this chapter for further discussion and the clinical practice guidelines regarding P2Y₁₂ inhibitors.

a-Blockers

The results of clinical trials investigating the use of β-blockers in ACSs have documented a decrease in early and late mortality. In pooled data from 28 trials of β-blockers, the average mortality decrease was 28% at 1 week, with the majority of benefit occurring in the first 48 hours.⁴⁰ Specifically, reinfarction was reduced by 18% and cardiac arrest by 15%.⁴⁰ The long-term effects of β-blockade for the secondary prevention of death after myocardial infarction were established by large-scale randomized trials. In a recent meta-analysis of 82 randomized trials of β-blockers, there was a significant reduction in long-term, but not short-term, mortality.⁴¹ The number of patients needed to treat with a β-blocker to prevent one death over 2 years was 42, compared with 24 for thrombolytics, 94 for standard-dose statins, and 153 for antiplatelet agents.

Traditionally, metoprolol has been the agent of choice and was initially administered intravenously followed by an oral dose. The Thrombolysis in Myocardial Infarction (TIMI) 2B study of early (immediate intravenous load followed by oral dosing) versus delayed (no intravenous load) therapy in STEMI patients who were undergoing reperfusion therapy with thrombolytic therapy showed no difference between the two groups in mortality or ejection fraction at the time of discharge.⁴² Data from COMMIT (Clopidogrel and Metoprolol in Myocardial Infarction Trial), which involved 45,852 patients, demonstrated that β-blockade resulted in a 15% to 20% relative risk reduction in the rate of recurrent infarction, but in a 30% relative increase in the risk of cardiogenic shock.⁴³ Early metoprolol use has also been shown to be cardioprotective and reduce infarct size.^44,45 Therefore, the use of β-blockade in patients with evidence of hemodynamic instability or heart block should be delayed until the patients become stable. Also, the routine initial intravenous administration of β-blockers is no longer considered standard therapy due to the potential risk of worsening status in patients with severe HF or cardiogenic shock. Finally, the benefits of β-blockade in patients undergoing primary PCI remain unclear, and there have been no randomized prospective studies.

Analgesia

Morphine is frequently used for pain relief and is best administered intravenously in boluses of 1 to 2 mg, to a maximum of 10 to 15 mg for a normal adult. Respiratory depression can occur, and care must be taken not to oversedate patients. Morphine should be used with caution in patients with hemodynamic instability because of its effects on reducing cardiac preload. In patients with inferior myocardial infarction and possible RV involvement, preload reduction is poorly tolerated, particularly in patients already volume depleted due to poor intake, vomiting, diaphoresis, and older age, and drugs such as morphine and nitrates should be used with caution.

Several studies have demonstrated an increased risk of adverse cardiovascular events with the administration of nonsteroidal anti-inflammatory drugs (NSAIDS) and cyclooxygenase-2 inhibitors.^46,47 These agents should not be administered for analgesia at any time during presentation with or hospitalization for STEMI.

Nitrates

Nitrates cause non–endothelium-dependent coronary vasodilatation, systemic venodilatation, reduced cardiac preload, and enhanced perfusion of ischemic myocardial zones (see Chap. 43). Animal and human data demonstrate that nitrate use may result in a reduction of infarct size; however, there is no benefit of nitrates in reducing mortality.^48,49,50 For patients with symptomatic ischemia, however, intravenous nitroglycerin may be very effective, although nitroglycerin tolerance has been observed as early as 12 hours after institution of intravenous therapy. Often, hemodynamic intolerance to nitroglycerin may be a problem, especially in patients with inferior infarction, hemodynamically significant RV infarction, or concomitant aortic valve stenosis, or among the elderly, particularly in the setting of preexisting volume depletion. Often the hypotension is accompanied by a bradycardia as a manifestation of the von Bezold-Jarisch reflex, particularly in patients with RV infarction.

Intravenous nitroglycerin should be initiated at 5 to 10 µg/min and gradually increased with a goal of a 10% to 30% reduction in systolic blood pressure and symptomatic pain relief. In most patients, use of this agent will be tapered within 24 to 36 hours. Nitrates should not be administered to patients with recent sildenafil use. In addition, approximately 40% of patients with an inferior STEMI will have involvement of the right ventricle, and nitrates should be used with caution in these patients in order to avoid profound hypotension, which usually responds promptly to intravenous volume replacement.

The management of STEMI patients with adjunctive therapies is presented in Table 40–2 from the ACC/AHA guidelines.²⁹

Anticoagulation

Heparins

Anticoagulation with heparin is standard in the management of STEMI. Currently, two forms of heparin are used: unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH). The use and pharmacology of these agents is explained in detail in Chap. 41.

When bound to antithrombin III, UFH inactivates factor Xa and thrombin. Its use has been widely studied, and UFH is considered a Class I indication for patients with STEMI undergoing primary PCI or receiving fibrin-specific thrombolytic agents.²⁹ An initial bolus of 60 U/kg (4000 U maximum) followed by a 12 U/kg/h (1000 U/h maximum) infusion should be administered promptly. A goal-activated partial thromboplastin time of 1.5 to 2.0 times normal should be achieved. The anticoagulation effects of UFH, however, can be variable as a consequence of unpredictable protein binding.

The LMWHs are glycosaminoglycans consisting of chains of alternating residues of D-glucosamine and uronic acid. When compared with UFH, they have a more predictable anticoagulation effect as a result of a longer half-life, better bioavailability, and dose-independent clearance. Unlike UFH, the LMWHs have a greater activity against factor Xa than thrombin, and their anticoagulation effect cannot be measured with standard laboratory tests.⁵¹

The use of the LMWH enoxaparin in conjunction with fibrinolysis has been studied in the ASSENT 3 (Assessment of the Safety and Efficacy of a New Thrombolytic Regimen) and TIMI-25 EXTRACT (Enoxaparin and Thrombolysis Reperfusion for Acute Myocardial Infarction Treatment) trials.^52,53 In the ASSENT 3 trial, patients treated with enoxaparin plus tenecteplase had a lower combined end point of 30-day mortality, in-hospital reinfarction, and in-hospital refractory ischemia than those treated with UFH plus tenecteplase (11.4% vs 15.4%; P = .0002). The TIMI-25 EXTRACT trial randomized 20,506 patients receiving fibrinolytic therapy for STEMI to anticoagulation with enoxaparin through the entire index hospitalization or UFH for the initial 48 hours. The combined primary end point of death or recurrent myocardial infarction at 30 days occurred in 9.9% of patients in the enoxaparin group and in 12% of those in the heparin group (P < .001).

Because of the delay in the onset of action of subcutaneous administration, an initial intravenous loading dose of 30 mg of enoxaparin followed by the traditional 1 mg/kg subcutaneous dose every 12 hours was used in both trials for patients younger than 75 years of age. Use of enoxaparin, however, was associated with a higher risk of minor and major bleeding when compared with UFH in the TIMI-25 EXTRACT trial.⁴¹ Switching from LMWH to UFH was also associated with an increase in bleeding in the SYNERGY trial.⁵⁴ In general, when this transition is needed, starting the UFH at the time of the next dose of LMWH may be sensible.

Direct Thrombin Inhibitors

Direct thrombin inhibitors bind directly to thrombin and have been used as an alternative to heparin in patients with heparin-induced thrombocytopenia. Several studies have compared direct thrombin inhibitors with heparin for the management of STEMI patients.^55,56 The HERO (Hirulog and Early Reperfusion or Occlusion)-2 trial randomized 17,073 STEMI patients to the direct thrombin inhibitor bivalirudin or heparin. Thirty-day mortality was similar in both groups; the incidence of reinfarction was decreased at the cost of an increased risk of mild and moderate bleeding.⁵⁵ A meta-analysis of 11 randomized trials demonstrated that, compared with heparin, direct thrombin inhibitors were associated with a lower risk of death or myocardial infarction both at the end of treatment (4.3% vs 5.1%; P = .001) and at 30 days (7.4% vs 8.2%; P = .02).⁵⁷ This difference was primarily a result of a reduction in the incidence of reinfarction (2.8% vs 3.5%; P < .001), with no difference in mortality (1.9% vs 2.0%; P = .69). There was no excess in intracranial hemorrhage with direct thrombin inhibitors compared with heparins.⁵⁷

The use of direct thrombin inhibitors in STEMI patients undergoing primary PCI was studied in the HORIZONS-AMI (Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction) trial.⁵⁸ The trial randomized 3602 STEMI patients to bivalirudin alone or heparin plus a glycoprotein (GP) IIb/IIIa inhibitor before undergoing primary PCI. The combined primary end point of major bleeding as well as adverse clinical events (death, reinfarction, target vessel revascularization, or stroke) was lower in those treated with bivalirudin compared with those receiving heparin plus GP IIb/IIIa inhibitor (9.2% vs 12.1%; P = .005). The primary end point, however, was primarily driven by a reduction in major bleeding (4.9% vs 8.3%; P < .001). In addition, further subgroup analysis revealed that bivalirudin reduced the 30-day rate of cardiovascular death (1.8% vs 2.9%; P = .03) and all-cause death (2.1% vs 3.1%; P = .047). Use of bivalirudin, however, resulted in an increase in acute stent thrombosis within the first 24 hours compared with heparin plus GP IIb/IIIa (1.3% vs 0.3%; P < .001). The HEAT PPCI (Unfractionated Heparin Versus Bivalirudin in Primary Percutaneous Coronary Intervention) trial evaluating bivalirudin versus UFH in a single-center all-comers STEMI setting did not show a mortality benefit for bivalirudin, but did show the consistent finding of stent thrombosis.⁵⁹ In fact, a meta-analysis of several randomized trials of bivalirudin versus UFH in patients with STEMI found that a bivalirudin-based regimen increases the risk of myocardial infarction and stent thrombosis but decreases the risk of bleeding, with the magnitude of the reduction depending on concomitant GP IIb/IIIa inhibitor use.⁶⁰

The BRIGHT (Bivalirudin in Acute Myocardial Infarction Versus Glycoprotein IIb/IIIa and Heparin Undergoing Angioplasty) randomized trial comparing generic bivalirudin versus heparin with or without tirofiban found a clinical benefit with bivalrudin.⁶¹ The EUROMAX (European Ambulance Acute Coronary Syndrome Angiography) study showed no benefit in prolonged bivalirudin infusion but did confirm the reduction in bleeding events in STEMI patients.⁶²

Hence, physicians caring for patients with STEMI will have to weigh the risks of bleeding versus thrombosis in their patients to make the decision regarding UFH and bivalirudin.

Factor Xa Inhibitors

The factor Xa inhibitor fondaparinux was investigated in the OASIS-6 (Organization for the Assessment of Strategies for Ischemic Syndromes) trial. In this study, 12,092 STEMI patients were randomized to fondaparinux for 8 days or placebo with standard heparin therapy.⁶³ The primary end point of death or reinfarction at 30 days occurred in 9.7% of patients in the fondaparinux group and 11.2% of those in the placebo group (P = .008). Patients treated with fibrinolytic therapy had the most benefit, and there was no significant increase in the rate of bleeding.

In a subset of the OASIS-5 trial, fondaparinux was compared with enoxaparin in 20,078 patients with ACS, 6238 of whom underwent PCI.⁶⁴ Although similar rates of death, myocardial infarction, and stroke were seen between groups (6.2% vs 6.3% for fondaparinux and enoxaparin, respectively; P = .79), the use of fondaparinux reduced major bleeding by one-half compared with enoxaparin (2.4% vs 5.1%; P < .00001). The rates of catheter-related thrombosis, however, were higher in those receiving fondaparinux alone compared with those receiving enoxaparin alone (0.9% vs 0.4%, respectively; hazard ratio, 2.3), which was completely obviated in both groups by the addition of UFH at the time of PCI without any increased risk of bleeding. Therefore, the use of upstream fondaparinux in patients with ACS undergoing PCI appears reasonable, although adjunctive UFH is recommended at the time of cardiac catheterization to prevent catheter-related thrombosis.

The clinical practice guidelines recommend the following^29,35,65:

The main goal of STEMI management is rapid reperfusion to establish coronary blood flow to ischemic myocardium. Currently, there are three main reperfusion strategies: fibrinolytic therapy, primary PCI, and fibrinolytic-facilitated primary PCI.

Fibrinolytic Therapy

Fibrinolytic therapy for STEMI has been shown to be effective in numerous randomized trials involving more than 100,000 patients.⁶⁶ It is widely available, easily administered, and relatively inexpensive. However, only approximately 50% of STEMI patients are eligible for fibrinolytic therapy (Table 40–3), and only 50% to 60% of patients treated with fibrinolytics will achieve complete reperfusion (TIMI grade 3 flow). In addition, 10% to 20% of patients will experience reocclusion, and 1% will suffer from a stroke caused by intracranial hemorrhage. Fibrinolytic therapy is most effective when given within 3 hours from onset of chest pain.⁶⁷

Although streptokinase is still widely used around the world, fibrin-specific agents are almost exclusively used in the United States. These therapies are summarized in Table 40–4 and are molecular modifications of the tissue plasminogen activator (tPA) molecule. The impetus for their development was the hope that increasing fibrin specificity (tenecteplase) or prolonging the plasma half-life (reteplase, tenecteplase, lanoteplase) would increase TIMI grade 3 flow rates in the infarct-related artery beyond the approximately 50% found at 90-minute angiography for tPA. Phase I and II trials of these agents have shown TIMI grade 3 flow rates in excess of 60%; however, there was no significant reduction in mortality or in the incidence of stroke.^68,69

The clinical practice guidelines focus on getting primary PCI as the mode of reperfusion within 120 minutes, and recommend the following regarding fibrinolysis when the delay to reperfusion is anticipated to be greater than 120 minutes²⁹:

Primary Percutaneous Coronary Intervention

See Chap. 42 for a more detailed discussion. Approximately 95% of patients who are treated with primary PCI obtain complete reperfusion versus 50% to 60% of patients who are treated with fibrinolytics. Primary PCI is also associated with a lower risk of stroke than treatment with fibrinolysis, and diagnostic angiography quickly defines coronary anatomy, LV function, and mechanical complications.

A meta-analysis by Keeley et al⁷⁰ of 23 trials that included 3872 patients treated with primary PCI and 3867 patients treated with fibrinolytics showed that PCI was superior to fibrinolytic therapy. Primary PCI was associated with a lower mortality (7% vs 9%; P = .0002), less reinfarction (3% vs 7%; P = .0001), and fewer strokes (1% vs 2%; P = .0004) at 30 days when compared with fibrinolysis. PCI capability, however, is available at fewer than 50% of hospitals in the United States, and each 30-minute delay from symptom onset to balloon inflation during primary PCI is associated with a 7.5% relative increase in mortality at 1 year.⁷¹ In addition, a meta-analysis by Nallamothu and Bates⁷² of data from 23 trials that included 7739 patients, which compared fibrinolytic therapy with primary PCI, demonstrated that the mortality advantage of primary PCI is lost if the door-to-balloon time is 60 minutes greater than the door-to-needle time for fibrinolytic therapy.

Three large-scale, randomized trials have compared fibrinolysis with transfer for primary PCI in STEMI patients (Table 40–5). The combined end point of death, reinfarction, and disabling stroke at 30 days was significantly lower in the patients treated with primary PCI. Although transfer for primary PCI appears to be the treatment of choice, two important points need to be made. First, the transport time in these trials was extremely short (median time in the DANAMI-2 [Danish Trial in Acute Myocardial Infarction-2] trial was 32 minutes), and these times may not be achievable outside of a clinical trial and in areas in which longer distances and weather may play a substantial role in hindering rapid transport. Second, fibrinolytic therapy still has a critically important role during the “golden hour” (Fig. 40–3) of myocardial infarction or when there is a delay in transfer.⁶⁷ The mortality and infarct size in patients treated with thrombolytic therapy within the first 60 to 90 minutes of symptoms are extremely low, suggesting that fibrinolytic therapy still plays a vital role in the management of patients presenting to hospitals without primary PCI capability. Also, the benefit of primary PCI appears to be mainly in high-risk patients, which were a minority of the patients in the trial (Fig. 40–4).⁷³

The clinical practice guidelines recommend the following^29,65:

Fibrinolytic-Facilitated Percutaneous Coronary Intervention

Facilitated PCI refers to the pretreatment with fibrinolytics in STEMI patients as a bridge to intended routine immediate PCI. This pretreatment was proposed as a method to initiate earlier reperfusion and reduce ischemic time and infarct size in patients who experience a delay before the onset of PCI. Fibrinolytic-facilitated therapy, however, can expose patients to a higher risk of bleeding, including intracranial bleeding. Two large trials of facilitated PCI, ASSENT-4 and FINESSE (Facilitated Intervention with Enhanced Reperfusion Speed to Stop Events), have investigated the use of fibrinolytic-facilitated PCI in STEMI. The ASSENT-4 trial randomized 1666 STEMI patients undergoing immediate PCI to pretreatment with a fibrinolytic (tenecteplase) or placebo. The data safety monitoring board halted the trial before the enrollment of the 4000 anticipated patients because of an increased 30-day mortality in the facilitated PCI group (6.0% vs 3.8%; P = .004).⁷⁷ Patients in the fibrinolytic plus PCI group also had a greater incidence of stroke (1.81% vs 0%; P < .001), bleeding (31.3% vs 23.4%; P < .001), and reinfarction (4.1% vs 1.9%; P = .01).

In the FINESSE trial, 2452 STEMI patients undergoing immediate PCI were randomized to receive one of the following three strategies: half-dose fibrinolytic (reteplase) plus the GP IIb/IIIa inhibitor abciximab (combination-facilitated PCI), early administration of abciximab (abciximab-facilitated PCI), or late administration of abciximab (primary PCI).⁷⁸ A primary composite end point that included death from all causes and complication of myocardial infarction through day 90 showed no significant differences among all three groups (9.8%, 10.5%, and 10.7%, respectively; P = .55). Safety end points indicated a significant difference between the combination-facilitated PCI and primary PCI groups, showing increased nonintracranial TIMI major or minor bleeding (14.5% vs 6.9%; P < .001). Treatment with early abciximab was associated with a trend toward increased bleeding.

Therefore, based on the results of ASSENT-4 and FINESSE, the data do not support the use of facilitated PCI as a routine reperfusion strategy in STEMI.

Pharmacoinvasive Strategies

Unlike facilitated PCI, in which PCI is performed immediately after fibrinolysis, pharmacoinvasive strategies refer to PCI that is routinely performed between 3 and 24 hours after fibrinolysis in hemodynamically stable patients as opposed to “rescue PCI.” The TRANSFER-AMI (Trial of Routine Angioplasty and Stenting After Fibrinolysis to Enhance Reperfusion in Acute Myocardial Infarction) is the largest and most recent among five studies that have evaluated routine early PCI after fibrinolysis.⁷⁹ In this study, 1059 STEMI patients who received fibrinolytic therapy were randomized to either standard treatment (transfer for angiography no less than 24 hours after lysis) or PCI of the infarct-related artery within 6 hours of fibrinolysis (“early PCI”). The primary end point was a composite of death, reinfarction, recurrent ischemia, new or worsening HF, or cardiogenic shock. At 30 days, the primary end point occurred in 11% of the early PCI group versus 17.2% of the standard group (P = .004). The favorable results for the early PCI group were primarily driven by a decreased rate of reinfarction. Notably, groups did not differ in the incidence of major bleeding.

The results of the four smaller studies investigating early PCI after fibrinolysis (CAPITAL AMI [Combined Angioplasty and Pharmacological Intervention Versus Thrombolytics Alone in Acute Myocardial Infarction],⁸⁰ CARESS-in-AMI [Combined Abciximab Reteplase Stent Study in Acute Myocardial Infarction],⁸¹ SIAM III [Southwest German Interventional Study in Acute Myocardial Infarction],⁸² and GRACIA [Grupo de Análisis de la Cardiopatía Isquémica Aguda] 1⁸³) are consistent with the findings of TRANSFER-AMI. However, the optimal window to perform PCI after receiving fibrinolytic therapy is still uncertain. The interval between fibrinolysis and PCI in the aforementioned studies ranged from 2 to 17 hours, with each interval suggesting similar efficacy.⁸⁴ After fibrinolysis, the risk of reocclusion increases within 24 hours. The STREAM (Strategic Reperfusion Early After Myocardial Infarction) trial demonstrated a trend toward improved outcomes at 30 days with a pharmacoinvasive approach with revascularization at 3 hours compared to primary PCI.^85,86 Additionally, studies have demonstrated reductions in HF, shock, and infarct size with early pharmacoinvasive strategies.⁸⁷ Therefore, until further data are available, striving for early PCI between 2 and 24 hours appears to be most reasonable.

In a meta-analysis⁸⁸ including seven randomized controlled trials of early transfer after fibrinolysis for catheterization, a strategy of routine early catheterization fibrinolysis was associated with a statistically significant reduction in the incidence of death or myocardial infarction at 30 days and 1 year.

For STEMI patients, when available, primary PCI is the preferred reperfusion strategy. However, when this therapy is not available, fibrinolysis followed by PCI within 2 to 24 hours appears to be a reasonable alternative.

The clinical practice guidelines recommend the following^35,65:

Delayed Invasive Management: Rescue Percutaneous Coronary Intervention

Approximately 40% to 50% of patients who receive fibrinolytic therapy fail to achieve optimal reperfusion, and 20% suffer reinfarction within the first 12 hours. The optimal management of these patients was investigated in the REACT (Rescue Angioplasty Versus Conservative Treatment or Repeat Thrombolysis) trial, which randomized 427 patients who failed to achieve reperfusion with fibrinolysis to repeat fibrinolysis, conservative therapy, or rescue PCI.⁸⁹

The combined primary end point of death, reinfarction, stroke, or severe HF within 6 months was lower in the rescue PCI group (15.3%) compared with those in the repeat fibrinolysis and conservative management groups (31% and 29.8%, respectively; P < .001). The primary end point was mainly driven by a decrease in recurrent myocardial infarction in the rescue PCI group compared with the repeat fibrinolysis and conservative management groups (2.1% vs 10.6% and 8.5%, respectively; P < .01). There was no difference in major bleeding among groups, although a significant increase in minor bleeding with rescue PCI was observed (P < .001), due mainly to sheath-related bleeding.

Recently, longer-term outcomes of the REACT trial were reported, with the results indicating that rescue PCI remains superior to more conservative, noninvasive strategies.⁹⁰ One-year clinical follow-up was available in 91% of the 427 randomized patients. At 1 year, event-free survival for patients was 81.5% for rescue PCI compared with 64.1% and 67.5% for repeat fibrinolysis and conservative therapy, respectively (P = .004). More importantly, a significant reduction in long-term mortality was seen in the rescue PCI group. At 4.4 years from randomization, there were 77 total deaths; 11.2% were from the rescue PCI group compared with 22.3% and 22.4% from the repeat fibrinolysis and conservative therapy groups, respectively (P = .026).

These and other data support the recommendation that rescue PCI should be considered for patients in whom fibrinolytic therapy fails to achieve reperfusion in STEMI.

The clinical practice guidelines state the following²⁹:

Selection of the Optimal Reperfusion Strategy

Figure 40–2 outlines a systematic, evidence-based framework for selecting reperfusion strategies in STEMI patients.⁹¹ STEMI patients presenting to PCI-capable hospitals should undergo primary PCI with a target door-to-balloon time of less than 90 minutes. If the hospital does not have PCI capability, the clinician must first determine whether the patient is eligible for fibrinolytic therapy (see Table 40–3). Patients ineligible for fibrinolytic therapy should be transferred for primary PCI. For those who are eligible for fibrinolytic therapy, the clinician must then consider two important factors: duration from onset of symptoms (fixed ischemia time) and transport time to the nearest PCI facility (incurred ischemia time). These two factors can be incorporated into a 2 × 3 table to select a reperfusion strategy (Fig. 40–5).

Patients facing a transport time of less than 30 minutes should be transferred for primary PCI. Fibrinolytic-eligible patients who present less than 2 to 3 hours from onset of symptoms and have a transport time precluding primary PCI within 120 minutes of FMC should receive fibrinolytic therapy. Patients presenting more than 2 to 3 hours after the onset of chest pain and who have a transport time of 60 minutes or less should be promptly transported for primary PCI. If the anticipated transport time is more than 60 minutes and anticipated FMC to reperfusion is 120 minutes, then patients can be treated with either fibrinolytic therapy or primary PCI.

Duration of symptoms and time to treatment play a crucial role in the decision to administer fibrinolytics or to proceed with transfer for primary PCI. Specifically, data from the rapid reperfusion project in North Carolina have indicated that when patients were taken directly to PCI hospitals bypassing non-PCI hospitals they had shorter times to reperfusion.⁹² For those eligible to receive fibrinolytics, prehospital administration (as opposed to in-hospital administration) of fibrinolytic therapy is a means to decrease time to reperfusion. The CAPTIM (Comparison of Angioplasty and Prehospital Thrombolysis in Acute Myocardial Infarction) trial is the only study to compare prehospital fibrinolysis and primary PCI as a function of time from onset of symptoms. The trial randomized 840 STEMI patients to prehospital fibrinolysis or primary PCI. Outcomes were assessed according to time between symptom onset (< 2 hours or > 2 hours) and randomization to each group. Results indicated that regardless of time of symptom onset, there was no significant difference between prehospital fibrinolysis and primary PCI in the primary end point of combined death, reinfarction, or disabling stroke (7.4% vs 6.5% [P = .855] in those with < 2 hours from symptom onset and 9.1% vs 5.9% [P = .326] in those with > 2 hours from symptom onset). There was, however, a trend toward increased mortality in those randomized within 2 hours who received primary PCI versus fibrinolytic therapy (5.7% vs 2.2%; P = .058). This trend was almost completely reversed in those randomized after 2 hours (7% for PCI vs 11% for fibrinolysis; P = .470). Although patients randomized early who underwent primary PCI also experienced significantly more cardiogenic shock (5.3% vs 1.3%; P = .032), this secondary end point was driven primarily by transport time.

The results of CAPTIM are consistent with those of PRAGUE-2 (Primary Angioplasty in Patients Transferred From General Community Hospitals to Specialized Percutaneous Transluminal Coronary Angioplasty Units With or Without Emergency Thrombolysis), which showed similar mortality for fibrinolysis and primary PCI when patients presented within 3 hours of symptom onset, but a higher mortality with fibrinolysis when patients were randomized after 3 hours. Long-term mortality data of CAPTIM were recently published, indicating outcomes consistent with those of the initial study.⁹³

These data suggest that duration of symptoms should be a strong consideration in the selection of prehospital fibrinolytic versus primary PCI therapy for STEMI. Barring exclusion criteria, in patients experiencing a symptom duration of less than 2 to 3 hours, prehospital fibrinolysis may be a viable option if there is an anticipated transfer time of greater than 60 minutes.

All patients receiving fibrinolytic therapy should be transferred to a PCI facility for potential failure to achieve reperfusion (ongoing chest pain or < 50% resolution of ST-segment elevation at 90 minutes) and rescue PCI. As mentioned earlier, the meta-analysis of randomized data suggests that all STEMI patients who are treated with fibrinolytic therapy benefit from routine coronary angiography during the index hospitalization.⁹⁴

Clopidogrel

Clopidogrel is an oral thienopyridine prodrug whose active metabolite inhibits the activation of platelets by ADP. Its antiplatelet effects are more potent than aspirin and less potent than the GP IIb/IIIa inhibitors.

Use With Fibrinolytic Therapy

Clopidogrel in combination with fibrinolytic therapy has been studied in the CLARITY-TIMI 28 (Clopidogrel as Adjunctive Reperfusion Therapy–Thrombolysis in Myocardial Infarction 28) trial.⁹⁵ In this trial, 3491 STEMI patients 75 years of age or younger who were treated with thrombolytics were randomized to therapy with aspirin plus placebo or aspirin plus clopidogrel. Clopidogrel was given as a 300-mg loading dose within minutes of thrombolysis and 75 mg daily thereafter. The composite primary end point of death, reinfarction before angiography, or occluded infarct-related artery at angiography occurred in 15% of patients in the clopidogrel group and 22% of patients in the placebo group (P < .001). The use of clopidogrel was not associated with a higher rate of major or minor bleeding.

Although a loading dose of 300 mg of clopidogrel was not associated with a greater risk of bleeding, patients older than 75 years of age were excluded from the study. Patients older than 75 years of age were included in the COMMIT trial⁴³ (a randomized, placebo-controlled trial of adding clopidogrel to aspirin in 46,000 AMI patients), which randomized 45,849 STEMI patients treated with fibrinolysis to therapy with aspirin plus placebo or aspirin plus clopidogrel. Unlike in the CLARITY trial, a 300-mg loading dose was not given. Instead, patients randomized to the clopidogrel arm received 75 mg of clopidogrel at the time of fibrinolysis and then 75 mg daily for the duration of hospitalization. Patients in the clopidogrel arm had a lower rate of the composite end point of death, reinfarction, or stroke (9.3% vs 10.1%; P = .002) and no increase in major or minor bleeding.

The data from these two trials suggest that patients treated with fibrinolysis should receive clopidogrel. For patients older than 75 years of age, 75 mg of clopidogrel without a loading dose should be used. In patients 75 years of age or younger, the current data suggest that a 300-mg loading dose followed by 75 mg daily of clopidogrel is beneficial and safe.

Use With Primary Percutaneous Coronary Intervention

Data from the PCI-CURE (Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic Events) trial (effects of pretreatment with clopidogrel and aspirin followed by long-term therapy in patients undergoing PCI)⁹⁶ demonstrated that the early use of clopidogrel in non–ST-segment elevation ACS patients who eventually undergo PCI is of benefit (see Chap. 42). The effectiveness of clopidogrel in the setting of primary PCI for STEMI is unproven, but extrapolation of the data indicating that upstream clopidogrel dosing improves PCI outcomes in non–ST-segment elevation myocardial infarction and scheduled PCI has led to the recent reasonable recommendation for oral clopidogrel loading at STEMI presentation and long-term maintenance therapy, barring contraindications in those younger than age 75 years.⁶⁵

Prasugrel

The advent of prasugrel, a novel thienopyridine with potent P2Y₁₂ platelet receptor-blocking capability, generated several recent studies comparing the efficacy of this agent to clopidogrel in patients undergoing PCI in both acute and scheduled settings.

To date, the largest trial to study prasugrel use in PCI is the TRITON-TIMI 38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel Thrombolysis in Myocardial Infarction 38).⁹⁷ The study randomized 13,608 patients with ACSs (including 3534 patients with STEMI) undergoing PCI to loading followed by maintenance doses of prasugrel or clopidogrel for 6 to 15 months. The primary end point was cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. Compared with clopidogrel, the prasugrel group had a significantly reduced primary end point (9.9% vs 12.1%; P < .001), as well as significantly reduced ischemic events (7.4% vs 9.7%; P < .001) and stent thromboses (1.1% vs 2.4%; P < .001). The benefits of prasugrel, however, were at the expense of increased non–coronary artery bypass graft–related major bleeding (2.4% vs 1.8%; P = .03), life-threatening bleeding (1.4% vs 0.9%; P = .01), and fatal bleeding (0.4% vs 0.1%; P = .002).

Due to the higher rates of bleeding with prasugrel, a post hoc analysis was performed to identify subgroups who were especially at risk for harm with prasugrel use. The analysis revealed a particular concern for those with a history of transient ischemic attack or stroke. Compared with clopidogrel, prasugrel use in these patients had a higher rate of intracranial hemorrhage (2.3% vs 0%; P = .02) and a strong trend toward a greater rate of TIMI major bleeding (5.0% vs 2.9%; P = .06). In addition, patients older than 75 years and those weighing less than 60 kg were also identified as being at risk for bleeding. Although not statistically significant, the absolute rate of TIMI major bleeding was increased in the latter two groups, and they received no net clinical benefit from prasugrel administration.⁹⁷

A subset analysis of the 3534 STEMI patients in TRITON-TIMI 38 was also performed.⁹⁸ The analysis corroborated the results of the full cohort, revealing the effectiveness of prasugrel versus clopidogrel in reducing the primary end point (6.5% vs 9.5%; P = .0017). In addition, there were fewer cardiovascular events (8.8% vs 6.2%; P < .0042) and stent thromboses (2.4% vs 1.2%; P < .008). This effect was maintained at both 30 days and 15 months. Unlike the increased risk of bleeding seen in the full cohort, STEMI patients exhibited no differences in TIMI major bleeding (2.4% vs 2.1%; P = .6451) or life-threatening bleeding (1.3% vs 1.1%; P = .7500) when comparing prasugrel with clopidogrel. STEMI patients requiring urgent coronary artery bypass grafting, however, exhibited significantly increased TIMI major bleeding in the prasugrel arm compared with the clopidogrel arm (18.8% vs 2.7%; P = .0033).

Thus, the use of prasugrel in STEMI patients undergoing primary PCI appears reasonable and appropriate for select patient subgroups. The risk of increased bleeding, however, particularly in those with a history of cerebrovascular events, needs to be further investigated and may limit its use. The current US Food and Drug Administration label approves prasugrel use for the reduction of thrombotic cardiovascular events in patients with unstable angina or myocardial infarction who undergo PCI. However, a boxed warning regarding bleeding risk emphasizes that prasugrel is not recommended for patients with a history of transient ischemic attack or stroke, patients over age 75 years, or patients who are likely to undergo coronary artery bypass grafting within the ensuing 7 days.⁹⁹

Ticagrelor

Unlike clopidogrel and prasugrel, ticagrelor is a reversible, nonthienopyridine P2Y₁₂ receptor antagonist that does not require metabolic conversion to active drug. This allows a rapid onset of action that has been tested in ACS patients, including those with STEMI.

The PLATO (Platelet Inhibition and Patient Outcomes) study evaluated 18,264 patients with ACS comparing ticagrelor (180-mg loading dose, 90 mg twice daily thereafter) with clopidogrel (300- or 600-mg loading dose, 75 mg daily thereafter) for the prevention of cardiovascular events. It is notable that 35% of the patients in the study had STEMI.¹⁰⁰ Among the patients with STEMI who underwent primary PCI, findings were consistent with the overall trial results. Ticagrelor reduced stent thrombosis, and there were fewer numeric total deaths, although there were more strokes and episodes of intracerebral hemorrhage with ticagrelor.¹⁰¹ It should be noted that a prespecified subgroup analysis in the PLATO trial showed a significant interaction between treatment effect and geographic region, with a smaller ticagrelor effect in North America than in other areas. Although this interaction could have been due to chance alone,¹⁰² a contribution from higher aspirin doses, as more commonly used in the United States, could not be excluded. Therefore, the package insert from the US Food and Drug Administration, consistent with the practice guidelines, recommends aspirin at a dose less than 100 mg when using ticagrelor.

The clinical practice guidelines for thienopyridine use state the following^35,65:

Glycoprotein IIb/IIIa Inhibitors

GP IIb/IIIa inhibitors are potent agents that inhibit the final common pathway for platelet aggregation. There are currently three intravenous agents available in the United States: abciximab, tirofiban, and eptifibatide.

Use With Fibrinolytic Therapy

Fibrinolysis has been shown to be a potent activator of platelets,¹⁰³ and the concomitant use of aspirin has been shown to be beneficial.¹⁰⁴ Two dose-finding studies^105,106 have shown that the GP IIb/IIIa inhibitor abciximab, when used in combination with half-dose thrombolytics, improves coronary artery blood flow in STEMI patients. Three subsequent randomized trials (Table 40–6) have investigated combination therapy with a GP IIb/IIIa inhibitor and half-dose fibrinolytic therapy.^52,107,108 The largest of the trials was the GUSTO-V trial, which included 16,588 patients with STEMI who presented within 6 hours from the onset of chest pain. The primary end point, 30-day mortality, occurred in 5.6% of patients treated with abciximab plus half-dose reteplase and in 5.9% of patients treated with full-dose reteplase (P = .43).¹⁰⁷ Patients treated with the combination therapy had a higher rate of major (1.1% vs 0.5%; P < .0001) and minor bleeding (20% vs 11.4%; P < .0001).¹⁰⁷ This increased risk of bleeding and lack of survival benefit with combination therapy was also seen in the ASSENT-3⁵² and INTEGRITI (Integrilin and Tenecteplase in Acute Myocardial Infarction)¹⁰⁸ trials (see Table 40–6). Based on the results of these three trials, GP IIb/IIIa inhibitors should not be used in combination with fibrinolytic therapy.

Use With Primary Percutaneous Coronary Intervention

The early (before arrival in the catheterization laboratory) versus delayed (at the time of catheterization) use of GP IIb/IIIa inhibitors in STEMI patients was investigated in eight randomized trials (Table 40–7) involving abciximab,^109,110,111 tirofiban,^112,113,114 and eptifibatide.^115,116 A meta-analysis of six of these trials by Montalescot et al¹¹⁷ showed that early administration of GP IIb/IIIa inhibitors in STEMI patients was associated with a greater prevalence of TIMI grade 2 or 3 flow (41.7% vs 29.8%; P < .001) in the infarct-related artery before PCI. Because prior studies have shown that better coronary artery flow after PCI is associated with less in-hospital and 1-year adverse outcomes,^118,119,120 early administration of GP IIb/IIIa inhibitors should probably be encouraged.

Although abciximab has historically been the GP IIb/IIIa inhibitor of choice during primary PCI due to the large number of studies utilizing this agent, there is growing evidence that eptifibatide and tirofiban may be acceptable alternatives in STEMI patients. A meta-analysis of six randomized trials comparing abciximab versus eptifibatide or tirofiban in STEMI patients undergoing primary PCI revealed no difference in outcomes among the three agents.¹²¹ The analysis included 2197 patients randomized to either abciximab or one of the small-molecule agents (tirofiban or eptifibatide). TIMI grade 3 flow was similar between groups (89.8% vs 89.1%; P = .72), as was ST-segment resolution (67.8% vs 68.2%; P = .66). Thirty-day mortality (2.2% vs 2.0%; P = .66), reinfarction (1.2% each; P = .88), and major bleeding complications (1.3% vs 1.9%; P = .27) were also similar between the two classes of agents. These findings are encouraging, as the small-molecule GP IIb/IIIa inhibitors are less expensive and easier to use in the acute care setting than abciximab.

Rapid Transport Protocol

For STEMI patients who present to hospitals without PCI capabilities, a well-developed plan for transfer to a PCI hospital is essential. This plan should be evidence based, easy to activate, and time efficient. It requires a multidisciplinary approach and coordination of personnel at the PCI facility, community hospital, and emergency transport service.

The Mayo Clinic developed a “Fast Track” protocol to facilitate the management and transport of STEMI patients from surrounding community hospitals. The protocol has 23 activated regional hospitals and three helicopters covering a service area extending 100 nautical miles. This rapid transport protocol involves four key concepts:

1. The identification and elimination of inefficient practices involved in the management of STEMI patients from referring hospitals.

2. Developing an evidence-based protocol to standardize the care of STEMI patients.

3. Implicating a communication network to ensure easy activation of the rapid transport protocol system by community hospitals.

4. Reducing the time for activation of the cardiac catheterization laboratory team.

The implication of these processes has resulted in a significant reduction in the median door-to-balloon times from 202 to 107 minutes in all STEMI patients transferred to the Mayo Clinic.¹²²

Primary PCI should be performed by operators who perform at least 75 PCIs per year, 36 of which are primary PCI for STEMI patients in a catheterization laboratory that performs at least 200 PCIs per year.

Primary Percutaneous Coronary Intervention Without On-Site Surgical Backup

Primary PCI at centers without surgical backup has been proposed as a possible method of increasing the availability of PCI to STEMI patients. Aversano et al¹²³ conducted the multicenter, prospective, randomized C-PORT (Atlantic Cardiovascular Patient Outcomes Research Team) trial to investigate the feasibility of off-site primary PCI. In this study, 11 centers with cardiac catheterization facilities, but no surgical backup, randomized 451 fibrinolytic therapy–eligible STEMI patients to fibrinolytic therapy or primary PCI at the community hospital. The composite primary end point of death, recurrent myocardial infarction, and stroke at 6 months occurred in 19.9% of those in the fibrinolytic group and 12.4% of those in the PCI group (P = .03).

The Mayo experience with off-site primary PCI has been positive. Of the 1007 PCIs performed at a Mayo Clinic affiliate hospital between 1999 and 2005, 285 were primary PCIs for STEMI patients.¹²⁴

There was a 93% success rate without any major complications, and no patients required emergency bypass surgery. The survival of primary PCI patients treated at the off-site catheterization laboratory was identical to that of patients treated at the Mayo Clinic.

The ACC/AHA guidelines for PCI recommend the following¹²⁵:

Distal Protection and Thrombectomy Devices

Distal embolization of the intracoronary thrombus at the time of PCI and subsequent no reflow is a concern. To prevent embolization, distal protection devices have been used in STEMI patients. The EMERALD (Enhanced Myocardial Efficacy and Recovery by Aspiration of Liberated Debris) trial investigated the use of the Guidewire Plus distal occlusion device (Boston Scientific, Marlborough, MA) in 501 STEMI patients. Participants were randomized to conventional primary PCI or PCI with the Guidewire Plus device. The primary end point of ST-segment resolution measured 30 minutes after PCI by continuous Holter monitoring and infarct size measured by technetium (Tc)-99m sestamibi imaging between days 5 and 14 was similar between both groups.¹²⁶ Another distal protection device, the Filter Wire-EX (Boston Scientific), was investigated in the PROMISE (Protection Devices in PCI Treatment of Myocardial Infarction for Salvage of Endangered Myocardium) trial. Two hundred AMI patients undergoing PCI were randomized to conventional therapy or PCI with the Filter Wire-EX distal protection device.¹²⁷ The primary end point of maximal adenosine-induced Doppler flow velocity in the recanalized infarct artery was similar in both groups. The secondary end point of infarct size as determined by magnetic resonance imaging was also similar between the two groups.

Mechanical debulking of intracoronary thrombus has also been proposed as a method to reduce distal embolization. The AIMI (AngioJet Rheolytic Thrombectomy in Patients Undergoing Primary Angioplasty for Acute Myocardial Infarction) trial randomized 480 STEMI patients to conventional PCI or PCI plus rheolytic thrombectomy with the AngioJet thrombectomy device (Boston Scientific). Patients in the thrombectomy group had a larger infract size (12.5% vs 9.8%; P = .03) and a lower rate of postprocedure TIMI grade 3 flow (92% vs 96%; P < .02).

In addition to mechanical aspiration, manual aspiration of thrombus material may also be considered to reduce distal embolization and microvascular obstruction. In the TAPAS (Thrombus Aspiration During Primary Percutaneous Coronary Intervention) study, 1071 STEMI patients undergoing primary PCI were randomized to manual aspiration through a 6-Fr Export Catheter (Medtronic, Edgewater, MD) before angioplasty versus conventional therapy.¹²⁸ The primary end point, a TIMI myocardial blush grade of 0 or 1, occurred in 17.1% of the aspiration group and 26.3% of the conventional PCI patients (P < .001). In addition, several secondary end points favored the thrombus-aspiration group compared with the conventional therapy group, including complete ST-segment elevation resolution (56.6% vs 44.2%; P < .001), absence of persistent ST-segment deviation (53.1% vs 40.5%; P < .001), and absence of pathologic Q waves on ECG (24.5% vs 15.9%; P = .001).

The ATTEMPT (Pooled Analysis of Trials on Thrombectomy in Acute Myocardial Infarction Based on Individual Patient Data) study, a recently published pooled analysis of patient data comparing standard PCI with or without thrombectomy, includes results from the TAPAS trial.¹²⁹ In this analysis, data from 2686 STEMI patients enrolled in 11 prospective trials were randomized to standard PCI versus PCI with thrombectomy device. The primary end point was all-cause mortality at 1 year. Results indicated a significantly lower all-cause mortality for the thrombectomy group versus the standard PCI group (P = .049). Subgroup analyses indicated that these results were primarily driven by reduced mortality in those who underwent thrombectomy in the presence of GP IIb/IIIa inhibitors. More importantly, the survival benefit was seen only in patients treated with manual thrombectomy devices.

However, three multicenter randomized trials, two of which were large trials with routine manual thrombectomy in STEMI patients, have not shown a mortality benefit. The TASTE (Thrombus Aspiration During ST-Segment Elevation Myocardial Infarction) trial (n = 7244) and the TOTAL (Trial of Routine Aspiration Thrombectomy With PCI Versus PCI Alone in Patients With STEMI) trial randomized 10,732 patients with STEMI to either aspiration thrombectomy before primary PCI or primary PCI only. The TOTAL trial also found an increase in stroke in patients undergoing routine aspiration.

The clinical practice guidelines recommend the following³⁵:

Drug-Eluting Stents

Several studies have demonstrated that DESs significantly reduce the occurrence of restenosis in patients undergoing elective PCI.^{130,131,132,133} More recently, however, multiple studies have investigated the efficacy and safety of DESs in STEMI patients undergoing primary PCI. In the TYPHOON (Trial to Assess the Use of the Cypher Stent in Acute Myocardial Infarction Treated With Balloon Angioplasty) trial, 712 STEMI patients who were to undergo primary PCI were randomized to receive either a Cypher DES (Cordis, Hialeah, FL) or a BMS.¹³⁴ The combined primary end point of death, reinfarction, or target vessel revascularization at 1 year occurred in 7.3% of the patients in the Cypher group and in 14.3% of patients in the BMS group (P = .0036). The end point, however, was mainly driven by a reduction in target vessel revascularization (3.7% vs 12.6%; P < .0001). Rates of stent thrombosis were similar in both groups (3.4% Cypher vs 3.6% BMS).

The PASSION (Paclitaxel Eluting Stent Versus Conventional Stent in ST-Segment Elevation Myocardial Infarction) trial randomized 619 STEMI patients who were to undergo primary PCI to the Taxus DES (Boston Scientific) or to an Express BMS (Boston Scientific).¹³⁵ Unlike the TYPHOON trial, use of the Taxus DES was not associated with a significant difference in the 1-year combined primary end point of death, reinfarction, or target vessel revascularization (8.7% Taxus vs 12.6% bare metal; P = .12). The rate of target vessel revascularization was 6.2% in the Taxus group and 7.4% in the BMS group (P = .23).

To date, HORIZONS-AMI is the largest trial of DES in STEMI patients undergoing primary PCI.¹³⁶ In this study, after emergent angiography, 3006 STEMI patients were randomized in a 3:1 ratio to receive either paclitaxel-coated (Taxus) stents (2257 patients) or BMSs (749 patients). Two primary end points, one for efficacy and one for safety, were identified: 12-month target lesion revascularization due to ischemia and a composite safety outcome of death, reinfarction, stroke, or stent thrombosis. At 12 months, target lesion revascularization occurred in 4.5% of the paclitaxel group versus 7.5% of the BMS group (P = .002). Importantly, safety outcomes between the groups were essentially equal, occurring in 8.1% and 8.0% of the DES and BMS groups, respectively (P = .92).

A 2009 meta-analysis that included the PASSION, TYPHOON, and HORIZONS-AMI studies identified 13 randomized trials and 18 registries comparing the outcomes of BMS with those of paclitaxel or sirolimus DESs in STEMI.¹³⁷ Among the 13 trials, 7352 STEMI patients were randomized to receive BMS or DES. Specific outcomes included mortality, myocardial infarction, target vessel revascularization, and stent thrombosis. Results showed a reduction in target vessel revascularization for those treated with a DES compared with a BMS (5.3% vs 11.5%; P < .001). More importantly, there was no difference between the DES and BMS in mortality (3.7% vs 4.3%; P = .36), myocardial infarction (3.4% vs 3.8%; P = .12), or stent thrombosis in the first year (2.7% vs 2.6%; P = .81).

The registry data, including 26,521 STEMI patients, were also consistent with those seen in randomized trials. Results indicated a significant reduction in target vessel revascularization in the DES-treated patients (relative risk [RR], 0.54; P < .01) without a difference between groups in myocardial infarction. In addition, the use of DES reduced mortality (RR, 0.68; P < .01) and stent thrombosis (RR, 0.52; P = .01) at 1 year. These effects, however, were not significant at 2 years.¹³⁷ These data imply the use of DES in STEMI to be efficacious and safe, although US Food and Drug Administration approval for their use in this setting is still pending.

For those patients who do receive a DES, the importance of dual antiplatelet therapy (DAPT) must be emphasized. A registry of AMI patients who received DES revealed that 13.6% of patients stop taking clopidogrel after the first month of therapy. These patients had a 1-year mortality of 7.5%, which was significantly greater than the death rate of 0.7% in those who were compliant with DAPT.¹³⁸

The clinical practice guidelines recommend the following³⁵:

Non–Infarct-Related Artery Coronary Intervention at Time of ST-Segment Elevation Myocardial Infarction Care

Prior versions of the STEMI guidelines had recommended against treatment of nonculprit stenoses during the initial STEMI revascularization procedure or during the same hospitalization in the absence of clinical instability or further testing documenting ischemia. The three new randomized studies have challenged this concept, leading to a focused update of the STEMI guideline and the new Class IIb assignment for treatment of nonculprit stenoses in the setting of primary PCI.^74,75,76 The PRAMI (Preventive Angioplasty in Acute Myocardial Infarction) trial randomized patients with > 50% stenosis felt to be treatable by the operator during primary PCI.⁷⁴ The CvLPRIT (Complete Versus Lesion-Only Primary PCI Trial) trial required the nonculprit artery stenosis to be 70% in one view and > 50% in two views.⁷⁶ Although treatment during the primary PCI was encouraged, treatment during the initial hospitalization was possible. Finally, in DANAMI3-PRIMULTI, nonculprit stenosis were required to be > 50% with fractional flow reserve < 0.80 for treatment or > 90%.⁷⁶ These three studies demonstrated varied levels of clinical benefit, mostly as a result of worsening symptoms and need for urgent revascularization. This has led to the change to a Class IIB recommendation and ongoing studies.

Fibrinolytic-Treated Patients

It should be noted that most patients with STEMI at hospitals without PCI capabilities may receive fibrinolysis if primary PCI is not immediately possible (within 60 minutes). These patients in general should be referred for elective angiogram within 3 to 24 hours if logistically possible.

Additional Pharmacotherapy

Inhibition of the Renin-Angiotensin-Aldosterone System

Angiotensin-Converting Enzyme Inhibitors

Nine trials, with cumulative enrollment of more than 100,000 patients, have documented the effects of angiotensin-converting enzyme (ACE) inhibitors on mortality in a prospective randomized fashion. These trials can be conveniently divided into those in which all patients were given the drug and those in which drug administration was limited to only patients who were at higher risk. The four nonselective trials include the CONSENSUS-2 (Cooperative New Scandinavian Enalapril Survival Study-2),¹³⁹ the GISSI-3 (Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico-3),¹⁴⁰ the ISIS-4 (Fourth International Study of Infarct Survival),⁵⁰ and the CCS-1 (Chinese Cardiac Study).¹⁴¹ In CONSENSUS-2, intravenous enalaprilat was administered within 24 hours of presentation followed by oral enalapril.¹³⁹ In the remaining studies, patients did not receive an initial intravenous load, and in three of these four trials, the drug was initiated within 24 hours of presentation. In CONSENSUS-2, mortality was nonsignificantly increased in the treatment group, but in the remaining three trials, a statistically significant reduction in mortality was observed in the treatment group with approximately 5 lives saved per 1000 patients receiving ACE inhibitor therapy.

Five trials studied the selective use of ACE inhibitors after AMI. These included the SAVE (Survival and Ventricular Enlargement) study,¹⁴² the TRACE (Trandolapril Cardiac Evaluation) study,¹⁴³ the AIRE (Acute Infarction Ramipril Efficacy) study,¹⁴⁴ the SMILE (Survival of Myocardial Infarction Long-Term Evaluation) study,¹⁴⁵ and the CATS (Captopril and Thrombolysis Study).¹⁴⁶ In SAVE and TRACE, patients were selected by laboratory evidence of an LV ejection fraction of less than 40% or wall motion abnormality. In AIRE, transient HF was the entrance criterion. Clinically and statistically significant mortality reduction of 40 to 70 lives saved per 1000 patients treated was documented in four of these five trials.

Given the results of these trials, it is reasonable to initiate therapy with ACE inhibitors within the first 24 hours as long as no contraindications exist and the patient is hemodynamically stable. Patients with an ejection fraction greater than 45% and no clinical evidence of HF, significant mitral regurgitation, or hypertension can have therapy discontinued after determination of risk status while still hospitalized. Because captopril has the shortest half-life, overdosing and inadvertent hypotension may be most easily correctable with the use of this agent. In addition, the short half-life allows for more rapid titration. Intravenous administration is unnecessary unless the patient is unable to take oral medication. Duration of treatment is uncertain; however, many patients will be treated indefinitely.

Angiotensin Receptor Blockers

Additional blockade of the renin-angiotensin system with the angiotensin receptor blocker valsartan was investigated in VALIANT (Valsartan in Acute Myocardial Infarction Trial).¹⁴⁷ This was a prospective, randomized, double-blinded trial involving 14,808 AMI patients complicated by impaired LV function (ejection fraction ≤ 0.35 on echocardiography or contrast angiography and ≤ 0.40 on radionuclide ventriculography) who were randomized to one of three groups: captopril only, valsartan only, or both captopril and valsartan. After a median follow-up of 25 months, the primary end point of cardiovascular death was 16.9% in the captopril group, 16.8% in the valsartan group, and 16.9% in the combination group (P = not significant). The combination group, however, experienced a statistically significant greater amount of hypotension and renal dysfunction. The results of VALIANT suggest that valsartan is as effective as captopril in the management of AMI patients with LV dysfunction, but the combination of an ACE inhibitor and an angiotensin receptor blocker should be avoided.

In OPTIMAAL (Optimal Trial in Myocardial Infarction With the Angiotensin II Antagonist Losartan), 5477 AMI patients were randomized to therapy with losartan or captopril.¹⁴⁸ The 3-year primary end point of all-cause mortality occurred in 18% of patients in the losartan group and in 16% of those in the captopril group (P = .07). These studies suggest that ACE inhibitors should be used as primary therapy and that angiotensin receptor blockers should be used in those who cannot tolerate ACE inhibitors.

Aldosterone Antagonists

The use of an aldosterone antagonist in patients with AMI complicated by LV dysfunction was studied in the EPHESUS (Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study) trial.¹⁴⁹ The 6642 patients with AMI and a resulting ejection fraction < 0.40 were randomized to the aldosterone antagonist eplerenone or placebo 3 to 14 days after admission. After a median follow-up of 16 months, the primary end point of death from any cause occurred in 14.4% of patients in the eplerenone group and 16.7% of patients in the placebo group (P = .008). As expected, eplerenone use was associated with a greater amount of hyperkalemia and less hypokalemia. Based on the results of the EPHESUS trial, it is reasonable to use an aldosterone antagonist in conjunction with β-blockade and ACE inhibition in patients with AMI and subsequent LV dysfunction (ejection fraction < 0.40). Diligent monitoring of the serum potassium must occur with patients who are treated with an aldosterone antagonist.

β-Blockade

Initiation of β-blockade therapy in the coronary care unit is essential in the management of STEMI patients. β-Blockers have both acute and long-term benefits in STEMI patients treated with either lysis or primary PCI. A meta-analysis of 24,000 AMI patients demonstrated a significant 14% reduction in acute mortality and a 23% reduction in long-term mortality. Short-acting β-blockade with metoprolol should be initiated as early as possible and rapidly titrated to the maximally tolerated dose. In patients who present with shock or HF, the initiation of β-blockers should be delayed until patients become hemodynamically stable.

The clinical practice guidelines recommend the following²⁹:

Statins

Numerous trials have demonstrated that statins should be used in the secondary prevention of patients with coronary artery disease.¹⁵⁰ In addition to lowering low-density lipoprotein (LDL) cholesterol, statins also improve endothelial function, have antiplatelet effects, and reduce inflammation.¹⁵¹ The benefits of initiating statin therapy during the acute setting in STEMI patients was investigated in the PROVE-IT (Pravastatin or Atorvastatin Evaluation and Infection Therapy) TIMI 22 trial.¹⁵² In this study, 4162 ACS patients (33% with STEMI) were randomized to standard therapy with 40 mg of pravastatin or to intense lipid-lowering therapy with atorvastatin 80 mg daily within the first 10 days of hospitalization for ACS. The combined primary end point of death from any cause, myocardial infarction, documented unstable angina requiring rehospitalization, revascularization (performed at least 30 days after randomization), and stroke occurred in 26.3% of the patients in the pravastatin group and 22.4% of patients in the atorvastatin group (P = .005). The final LDL was 95 mg/dL in the pravastatin group and 62 mg/dL in the atorvastatin group (P < .001).

It is therefore reasonable to initiate statin therapy during the acute setting of STEMI. Although the data are not clear regarding the benefits of early statin use, STEMI patients are more likely to be on statin therapy in the post–myocardial infarction period if treatment is initiated during the index hospitalization. An LDL goal of less than 70 mg/dL should be achieved.

Several studies have investigated the long-term benefits of statin therapy as secondary prevention in patients with coronary artery disease.

Hemodynamic Complications

The most common major mechanical complications of acute STEMI include cardiogenic shock, RV infarction, acute mitral regurgitation, ventricular septum rupture, and free wall rupture.

Cardiogenic shock

Cardiogenic shock as a result of severe LV dysfunction occurs in approximately 7% of patients with myocardial infarction and has a historic mortality of approximately 80%.¹⁵³ A population-based study analyzed temporal trends in cardiogenic shock complicating AMI in 9076 patients between the years 1975 and 1997.¹⁵⁴ The frequency of cardiogenic shock remained relatively stable at 7.1%, and the mortality was approximately 72%. There was a trend toward increased in-hospital survival in the mid to late 1990s, which correlated with the increased application of reperfusion technologies.¹⁵⁵ The SHOCK II (Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock II) trial showed that the cardiogenic shock mortality remains elevated at near 40%.³⁴

The goals for treatment are two-fold: first, hemodynamic stabilization to ensure adequate oxygenation, acid-base balance, and tissue perfusion; and second, rapid investigation of any potentially reversible causes for the patient’s condition. Hypovolemia, which is more common in older patients, those receiving chronic diuretic therapy, and those receiving narcotics or preload-reducing agents, must be corrected. Hemodynamic monitoring by using a balloon-tipped pulmonary artery catheter allows immediate access to valuable hemodynamic information. Forrester and others^156,157 described the treatment of patients on the basis of hemodynamic subsets related to pulmonary artery wedge pressure and cardiac output (Table 40–8). The basic goals of this approach include adjustment of the intravascular volume status to bring the pulmonary artery capillary wedge pressure to 18 to 20 mm Hg and optimization of cardiac output with inotropic and/or vasodilating agents. Severely hypotensive patients can be temporarily aided by intra-aortic balloon pumping or possibly by a ventricular assist device. However, the benefits from these mechanical treatments are often temporary, and there may be a significant risk of complications.

Many observational studies suggest that reperfusion therapy, especially with primary PCI, can reduce the incidence of cardiogenic shock. In one of the largest observational series, 1321 patients with shock in the GUSTO-1 trial were analyzed with respect to revascularization within 30 days versus no revascularization therapy. After multivariable logistic regression analysis, revascularization was independently associated with reduced 1-year mortality (odds ratio [OR], 0.6; 95% CI, 0.4-0.9; P = .007). In-hospital mortality, however, remained high at 56%.¹⁵⁸

The SHOCK trial evaluated the effect of early revascularization in AMI complicated by cardiogenic shock.^159,160 The 302 patients with shock caused by LV dysfunction complicating myocardial infarction were randomly assigned to either emergency revascularization (n = 152) or initial medical stabilization (n = 150). Eighty-six percent of patients in both groups had intra-aortic balloon counterpulsation placement, and revascularization was accomplished by either coronary artery bypass grafting or angioplasty. The median time from symptom onset to revascularization was approximately 12 hours. The revascularization group had lower all-cause mortality at 30 days (46.7% vs 56%; P = .11), 6 months (50.3% vs 63.1%; P = .027), and 1 year (53.3% vs 64.4%; P < .03). There was an interaction between the effect of therapy and age in that only patients younger than age 75 years benefited from revascularization. Although the 20% relative mortality reduction was more modest than expected on the basis of observational studies, only eight patients would have to receive early revascularization to prevent one death at 6 months. This result compares very favorably with benefits achieved by fibrinolytic therapy or primary PCI in patients without shock.

Although vasoconstriction has traditionally been associated with cardiogenic shock, vasodilatation can also be present. The mechanism of vasodilatation occurs from an increase in inducible nitric oxide synthase and subsequent increased production of nitric oxide. Use of the nitric oxide synthase inhibitor N^G-monomethyl-L-arginine had been proposed as a potential therapy for these patients with vasodialtion,¹⁶¹ but definitive randomized clinical trials demonstrated no benefit of this therapy over placebo.¹⁶²

The SHOCK II trial randomized patients to planned intra-aortic balloon pump (IABP) with PCI versus PCI alone. The trial found that there was not a significant reduction in mortality with IABP use. This has led to a change in guideline recommendations for routine IABP use with shock to Class IIA.

The clinical practice guidelines recommend the following²⁹:

Cardiogenic Shock in Patients With ST-Segment Elevation Myocardial Infarction May Be Caused by Right Ventricular Infarction

Involvement of the right ventricle is a common sequela of acute inferior myocardial infarction,¹⁶³ especially after proximal right coronary artery occlusion.¹⁶⁴ However, hemodynamically significant dominant RV dysfunction is much less common, particularly in the reperfusion era, occurring in relatively few patients with RV infarction (RVI).

The diagnosis of hemodynamically significant RVI rests on the clinical triad of hypotension, increased jugular venous pressure, and clear lung fields in a patient with acute inferior myocardial infarction. Additional diagnostic techniques that can document RV involvement include ST-segment elevation in right-sided chest leads (V₃R or V₄R), visualization of RV wall motion abnormalities, and RV dilatation on radionuclide angiography or echocardiography.

Hemodynamic measurements in patients with significant RVI demonstrate elevation of the right atrial pressure, usually more than 10 mm Hg, and often show a right atrial pressure/pulmonary artery wedge pressure ratio of 0.8 or more. However, in cases with significant LV dysfunction and increased wedge pressure, this ratio may be lower and does not exclude the presence of significant RV involvement. Rarely, substantial arterial desaturation may be observed because of opening of a patent foramen ovale as a result of increased right atrial pressure and right-to-left atrial shunting. The incidence of bradyarrhythmias is increased in patients with RVI and can be detrimental because an increased heart rate may be necessary to compensate for the decrease in RV (and as a result LV) stroke volume caused by the RVI. Similarly, high-grade atrioventricular block may lead to loss of the atrioventricular synchrony with RV underfilling and further decrease in the RV and LV stroke volume.¹⁶⁵

Treatment of RVI initially involves volume loading with normal saline to achieve a pulmonary artery wedge pressure of 18 to 20 mm Hg. In some patients, this alone is sufficient to improve cardiac output and systemic pressure. However, some patients will not respond to fluid-loading alone. This may be a result of marked RV enlargement within a relatively noncompliant pericardium, which may result in functional LV compression because of ventricular interaction. In addition to volume loading, use of dobutamine improves cardiac index.¹⁶⁶ Patients requiring temporary pacing for heart block may also benefit from arteriovenous sequential pacing rather than lone ventricular pacing. Reperfusion by thrombolysis,¹⁶⁷ and especially with primary PCI,¹⁶⁸ improves both RV function and clinical outcomes.

The natural history of dominant RVI may be favorable because the right ventricle is very resistant to ischemia and usually recovers.¹⁶³ Most patients, even those with substantial RV dysfunction, spontaneously improve in 48 to 72 hours after the acute event. Patients in shock may benefit from angioplasty of the occluded right coronary artery or from temporary use of an RV assist device; however, many of these patients have associated significant LV dysfunction, which complicates the picture.¹⁶⁹ The overall balance between the extent of RV and LV dysfunction is a major determinant of long-term outcome, and the majority of patients with RVI and significant hemodynamic compromise usually have evidence of extensive biventricular infarction and cardiogenic shock.

The clinical practice guidelines recommend the following²⁹:

Acute Mitral Regurgitation

Severe mitral regurgitation caused by papillary muscle rupture is responsible for approximately 5% of deaths in AMI patients. Rupture may be complete or partial, and it usually involves the posteromedial papillary muscle because its blood supply is derived only from the posterior descending artery, whereas the anterolateral papillary muscle has a dual blood supply from both the left anterior descending and the circumflex coronary arteries. Most patients have relatively small areas of infarction with poor collaterals, and up to half of the patients may have single-vessel disease.

The clinical presentation of papillary muscle rupture is the acute onset of pulmonary edema, usually within 2 to 7 days after inferior myocardial infarction. The characteristics of the murmur vary; as a result of a rapid increase of pressure in the left atrium, no murmur may be audible. Thus a high degree of suspicion, especially in patients with inferior wall infarction, is necessary for diagnosis. Two-dimensional echocardiographic examination demonstrates the partially or completely severed papillary muscle head and a flail segment of the mitral valve. LV function is hyperdynamic as a result of the severe regurgitation into the low-impedance left atrium; this finding alone, in a patient with severe congestive HF, should suggest the diagnosis.

The cornerstones of successful therapy are prompt diagnosis and emergency surgery. Emergent placement of an IABP and blood pressure control may be beneficial. The current approach of emergency surgery accrues an overall operative mortality of 0% to 21%, but this appears to be decreasing, and the late results of this approach can be excellent.^170,171

The clinical practice guidelines recommend the following²⁹:

Ventricular Septal Rupture

Before the reperfusion era, rupture of the ventricular septum occurred in 1% to 3% of acute infarctions and caused approximately 5% of peri-infarction deaths. Data from GUSTO-1, however, indicate that ventricular septal rupture (VSR) was confirmed in only 0.2% of more than 41,000 patients who received lytic therapy for ACS.¹⁷² These data indicate that mechanical complications of myocardial infarction appear to be substantially reduced since the advent of reperfusion therapies. When VSR occurs, however, the substrate is quite similar to that of free wall rupture in terms of number of vessels diseased and infarct size. Typically, VSRs associated with anterior infarction are located in the apical septum, and those associated with inferior infarction are located in the basal inferior septum. The prevalence of anterior and inferior infarctions is approximately equal, unlike papillary muscle rupture.¹⁷³

The diagnosis should be suspected clinically when a new pansystolic murmur is present. As with other cardiac ruptures, surgical management is advocated, although the outcome is not as gratifying as in acute mitral regurgitation, because the extent of myocardial necrosis is generally larger.^172,174 Percutaneous closure using occluding devices has been done in nonsurgical candidates with variable results: three of seven patients survived to hospital discharge in one report.¹⁷⁵

The outcome of 91 patients seen in a single institution with VSR after AMI was reviewed by Lemery et al.¹⁷⁴ Advanced age, cardiogenic shock, and long delay between septal rupture and surgery correlated with adverse outcome. In patients with cardiogenic shock and ventricular septal defect, only those operated on within 48 hours survived; thus the proportion surviving in this group was only 38%. For patients not in shock, mortality was similar for surgery either within 2 to 14 days or after 14 days; however, the clinical course was unpredictable, with rapid deterioration and death in approximately 50% of these patients.

The incidence and outcome of peri-infarction ventricular septal defect in the fibrinolytic era were reported in a study from the GUSTO-1 database.¹⁷² In 84 patients (0.2%), ventricular septal defect developed after fibrinolysis, and the median time from symptom onset to the diagnosis of ventricular septal defect was 1 day. Mortality at 30 days in this group was 73.8%, and in patients selected for surgical repair versus those managed medically, 30-day mortality was 47% versus 94%, respectively. All patients presenting in Killip class III or IV died regardless of therapy; however, 1-year survival was excellent in patients surviving the initial 30 days after infarction.

The clinical practice guidelines recommend the following²⁹:

Free Wall Rupture

Rupture of the free wall of the left ventricle occurs in 1% to 4.5% of AMI patients and accounts for 10% to 15% of early AMI deaths.^{176,177,178,179,180,181} Rupture is more common in women, the elderly, and persons with delayed admission to hospital.^176,178,181 Although any wall may be involved, rupture of the lateral wall is probably most common. Rupture occurs within the first 5 days of infarction in 50% of cases and within 14 days in 87% of cases. The area of rupture always occurs within the area of infarction but usually is eccentrically located near the junction with normal myocardium. The incidence of rupture has decreased in the fibrinolytic era, but the timing is earlier (24-48 hours).¹⁷⁸ Further decrease in the rate of rupture can be achieved by primary PCI.^177,181 Survival to discharge in patients with cardiogenic shock included in the SHOCK trial¹³³ was 39.3%, similar to the survival of the other patients.¹⁸²

The clinical presentation is usually sudden electromechanical dissociation. Most patients die even when rapid resuscitative measures, including pericardiocentesis, IABP insertion, and emergent cardiac surgery, are attempted.¹⁸³ Surgical correction of acute free wall rupture is the treatment of choice, even though surgical mortality can be high.¹⁸⁴ Surgical correction can be challenging because of the friability of the tissue that surrounds the rupture. A sutureless patch technique has shown encouraging initial results.¹⁸⁵ Early diagnosis is critical and can be achieved by a high degree of suspicion in any patient with sudden hemodynamic deterioration, particularly in the absence of evidence for recurrent ischemia or infarct extension.

Alternatively, rupture may be subacute, with periodic small amounts of blood leaking into the pericardial space.^183,186 ECG evidence of regional pericarditis may be a warning of impending rupture. Persistent and severe pericardial pain also may be a manifestation of this phenomenon, and the subsequent inflammatory process may serve to wall off the area of pericardial leakage from the remaining pericardial space, forming a false or pseudoaneurysm of the left ventricle. The entity is easily and reliably detected by two-dimensional echocardiography and, when acute, mandates early surgical intervention because of the risk of further expansion of the false aneurysm or rupture producing tamponade and death. It is critically important to have a heightened level of sensitivity for the possibility of subacute rupture in all patients with evidence of pericarditis and especially if a friction rub is present. Immediate performance of echocardiography is mandatory. Chronic LV false aneurysm has a relatively low risk of rupture. Survival is reduced from multiple causes. In one observational series of 21 patients followed for a mean of 3.6 years, mortality was 64%, although no patient died of rupture.¹⁸⁷

The clinical practice guidelines recommend the following²⁹:

Electrical Complications

Bradyarrhythmias

The bradyarrhythmias are reviewed, but in the modern reperfusion era, the incidence and permanent pacemaker use have declined significantly.¹⁸⁸

Bradyarrhythmias usually result from ischemic injury to the sinus node/conduction system or abnormal reflexes that are vagally mediated or both. The blood supply to the sinus node arises from the proximal right coronary artery in 55% of patients and from the proximal left circumflex in the remainder. The blood supply to the atrioventricular node arises from the distal branches of the right coronary artery in 90% of patients and from the distal portions of the left circumflex artery in the remaining 10% of patients. The right bundle branch is supplied primarily by the septal perforator vessels originating from the left anterior descending artery, as is the distal portion of the anterior left bundle branch. The main left bundle branch has a dual supply from both distal branches of the right coronary and proximal circumflex vessels, and the posterior division of the left bundle branch is supplied from branches of the circumflex coronary artery.

In general, conduction disturbances associated with inferoposterior infarction are related to enhanced vagal activity, tend to be more transient, are often responsive to atropine, and imply a somewhat more benign outcome than those involved in anterior infarction. Conversely, major conduction disturbances associated with anterior infarction usually imply extensive septal necrosis and more significant reduction in LV function.

Sinus Bradycardia

Sinus bradycardia and sinus pauses are usually benign. First-degree atrioventricular block occurs in 4% to 13% of myocardial infarctions. Observation and avoidance of any medications that might prolong atrioventricular conduction are required.

Second-Degree Atrioventricular Block

This usually develops within the first 24 hours of myocardial infarction in 3% to 10% of individuals. With type I second-degree atrioventricular block, progressive prolongation of the PR interval is observed. This is usually seen in inferoposterior infarction and may often respond to atropine. A narrow QRS complex is usually present, and temporary pacing is not needed unless the ventricular rate decreases to less than 45 bpm or symptoms of impaired perfusion develop. Type II second-degree atrioventricular block is identified by intermittent dropped beats in the absence of progressive PR prolongation and implies extensive infranodal conduction system injury. Often the QRS complex is wide, indicating associated bundle branch block, and progression to complete heart block occurs in approximately one-third of these patients. Most patients with anterior infarction and type II second-degree atrioventricular block need temporary transvenous pacing because of the unpredictable risk of complete heart block.

Complete Heart Block

This occurs in 3% to 12% of AMIs.^189,190 In general, patients with inferoposterior infarction progress to third-degree heart block after a period of second-degree heart block and again may demonstrate some responsiveness to atropine or aminophylline.¹⁹¹ A stable junctional escape rhythm is often present, and recovery tends to occur within 3 to 7 days. The occurrence of complete heart block in a patient with inferior infarction confers a 1.5- to 4-fold increase in risk of in-hospital mortality.^189,190 Consequently, these patients should be carefully observed, often with standby temporary transcutaneous pacing patches in place. Complete heart block in the presence of anterior infarction usually indicates an extensive area of myocardial necrosis and has a poor prognosis. Most patients require temporary transvenous pacing, and some physicians advocate permanent transvenous pacing. Late mortality in these patients is usually the result of contractile failure or ventricular fibrillation rather than persistent, high-grade atrioventricular block.

Bundle Branch Block

The occurrence of any new bundle branch block with AMI also identifies patients with extensive infarction who are at higher risk for complications. Unifascicular block, especially left anterior hemiblock, occurs in approximately 5% of patients and has a relatively benign prognosis. Complete right or left bundle branch block occurs in 10% to 15% of patients and most commonly (two-thirds of cases) involves the right bundle branch. Both LBBB and right bundle branch block are associated with higher in-hospital and long-term mortality.¹⁹² In the past, the new occurrence of LBBB or right bundle branch block has been a generally accepted indication for temporary transvenous pacing.

Permanent transvenous pacing is indicated in patients with persistent complete or high-grade atrioventricular block or persistent type II second-degree atrioventricular block after myocardial infarction and in patients with a new bundle branch block and transient but resolved complete heart block during the acute course of infarction. Occasionally, patients may require an electrophysiologic study to determine the site of atrioventricular block and to aid in the decision about whether permanent pacing is indicated. Permanent pacing also may be indicated in the rare patient with profound sinus node dysfunction. However, this complication is rarely related to myocardial infarction.

The clinical practice guidelines recommend the following²⁹:

Tachyarrhythmias

Multiple factors play a role in the genesis of tachyarrhythmias in patients with myocardial infarction. Decreased blood flow leads to anaerobic metabolism, and decreased venous outflow allows accumulation of by-products of this process, resulting in acidosis, increase in extracellular potassium concentration, and an increase in intracellular calcium concentration. In addition to these ionic changes, there may be alterations in sympathetic and vagal tone and increased concentrations of circulating catecholamines. The electrophysiologic correlates of these cellular abnormalities include slowing of conduction and prolongation of refractoriness, which, coupled with the inhomogeneous nature of the infarction process, produces an ideal situation for the occurrence of reentrant arrhythmias. The presence of injury currents may directly enhance phase IV depolarization of Purkinje cells, resulting in increased automaticity. Fiber stretch, resulting from increased atrial and ventricular end-diastolic pressures, is also arrhythmogenic. Finally, reperfusion, possibly caused by intracellular calcium overload or production of free oxygen radicals, may generate reperfusion arrhythmias that may be either automatic or reentrant.

Supraventricular Arrhythmias

Sinus tachycardia may occur in up to 25% of patients with acute infarction and often results from pain, anxiety, and sometimes hypovolemia. Persistent sinus tachycardia may be a marker of severe LV dysfunction and is a poor prognostic sign. After relief of pain and assessment for the presence of pulmonary congestion, it is desirable to decrease the heart rate to less than 70 bpm by intravenous administration of β-adrenergic receptor blockers. A short-acting β-adrenergic receptor blocker, such as esmolol, may be appropriate for patients in whom the extent of LV dysfunction is of particular concern.

Although its incidence in the era of reperfusion with lytics and PCI has declined, atrial fibrillation may occur in approximately 6% to 21% of patients with acute infarction.^193,194,195 Risk factors for the development of atrial fibrillation in AMI include advanced age, HF symptoms, LV dysfunction, and tachycardia on admission.¹⁹⁵ Its early presence signifies atrial ischemia; later, it may also represent atrial stretch caused by increased filling pressures, which is why it is associated with such an adverse prognosis. This adverse prognosis is not only limited to the primary hospitalization. Long-term mortality regardless of the treatment received for the presenting myocardial infarction is also affected. Immediate cardioversion is the best treatment for patients with symptomatic rapid atrial fibrillation or in whom the rapid ventricular response produces ischemia. If the ventricular response is only moderate and the patient is asymptomatic, diltiazem or esmolol (in patients with a normal ejection fraction) given intravenously is useful for control of heart rate, along with digoxin given orally; the latter may take 4 to 8 hours for full effect. Recurrent episodes of atrial fibrillation should be suppressed with an antiarrhythmic agent such as procainamide or intravenous amiodarone. The treatment of atrial flutter is similar to that of atrial fibrillation, except that drug treatment is less effective in controlling the ventricular response. Occasionally, atrial overdrive pacing may be used to terminate atrial flutter without resorting to cardioversion. Atrial fibrillation in the setting of AMI, especially if it occurs after admission, is associated with higher mortality and incidence of stroke.^193,194

The clinical practice guidelines recommend the following²⁹:

Ventricular Tachyarrhythmias

Accelerated idioventricular rhythm may occur in up to 40% of continuously monitored patients and may be a marker of reperfusion in some. This rhythm disturbance is generally considered benign and is usually untreated (see Chap. 85).

The clinical practice guidelines recommend the following²⁹:

Ventricular tachycardia occurs in up to 15% of patients during AMI. The ventricular rate is usually between 140 and 200 bpm, and this rhythm disturbance may degenerate to ventricular fibrillation. The rhythm disturbance usually responds to lidocaine given intravenously. However, procainamide, bretylium, cardioversion, ventricular overdrive pacing, or amiodarone may all be required in the acute stages for resistant cases. The use of lidocaine prophylactically for prevention of ventricular tachycardia is not currently recommended. Pooled results of studies showed that the incidence of ventricular tachycardia was decreased, but fatal asystole was more common and no survival advantage was observed.

Ventricular fibrillation is seen in approximately 8% of patients surviving to hospitalization for acute infarction. It is more frequent in large ST-segment elevation infarcts and may occur with or without warning arrhythmias. The occurrence of ventricular fibrillation within the first 24 hours of hospitalization was previously thought not to confer any long-term risk to patients successfully resuscitated; however, some studies indicate a poorer outcome for patients with ventricular fibrillation at any time during their hospital course. Ventricular fibrillation or tachycardia occurring late in the hospital course may be the result of pump failure, severe electrolyte imbalance, effects of antiarrhythmic medications, or other metabolic derangements; it is usually associated with decreased LV systolic function and portends a poor prognosis. Sustained monomorphic ventricular tachycardia (VT), occurring early or late during the hospital course, is not common but implies a fixed arrhythmogenic substrate and a propensity for recurrence after dismissal. An invasive electrophysiologic study can be justified in these patients before discharge from the hospital.

The clinical practice guidelines recommend the following²⁹:

Implantable Cardiac Defibrillator

Several trials have demonstrated a mortality benefit with the use of implantable cardiac defibrillators (ICDs) as primary prevention for sudden cardiac death in patients with chronic ischemic cardiomyopathy. The DINAMIT (Defibrillator in Acute Myocardial Infarction Trial) investigated the used of ICD therapy early after AMI. In this study, 674 patients within 6 to 40 days after an AMI with subsequent LV dysfunction (ejection fraction < 35%) were randomized to ICD therapy or placebo. After a median follow-up period of 30 months, there was no significant difference in the mortality between the two groups. These results were somewhat surprising given that the highest incidence of sudden cardiac death in the VALIANT trial occurred during the first 3 months after myocardial infarction.¹⁴⁷

AMI patients with LV dysfunction should not receive an ICD during the early period for prophylactic indications. ICD therapy is recommended in those who have recurrent sustained episodes of VT during the post–myocardial infarction period. A repeat assessment of LV dysfunction 30 to 40 days after the acute event should be performed, and those with an ejection fraction less than 35% should receive an ICD. Antiarrhythmic therapy has not been demonstrated to prevent sudden cardiac death in the post–myocardial infarction setting in patients without sustained ventricular arrhythmias.

STEMI patients with abnormal LV function who are not candidates for ICD implantation remain at risk for sudden cardiac death due to arrhythmia. The use of a home automated external defibrillator was studied in the HATS (Home Use of Automated External Defibrillators for Sudden Cardiac Arrest) trial.¹⁹⁶ The study randomized 7001 post-STEMI patients to one of two responses should cardiac arrest occur at home: conventional emergency medical service response with home cardiopulmonary resuscitation versus the use of an automated external defibrillator (AED) followed by the conventional response. Of the 450 patients who died after home cardiac arrest, there was no difference in survival between conventional resuscitation and home AED use (6.5% vs 6.4%; P = .77). At this time, data do not support the recommendation for routine AED access at home. Currently these patients may be considered for wearable defibrillators. However, more data are needed, and an ongoing trial is under way in patients with reduced LV ejection fraction and AMI (VEST; ClinicalTrials.gov identifier: NCT01446965). A scientific statement also provides impetus for more data on use of wearable defibrillators.¹⁹⁷

The clinical practice guidelines recommend the following²⁹:

Cardiac Rehabilitation

Rehabilitation after myocardial infraction is an important aspect of secondary prevention and is described in Chap. 45.

Takotsubo syndrome (stress-related or apical ballooning) is characterized by transient systolic and diastolic dysfunction that is often preceded by physical or emotional triggers. Takotsubo syndrome was first described in Japan in 1990¹⁹⁸ and is derived from a Japanese word that means “octopus pot,” named for the fishing pot with a narrow neck and wide base that is used to trap octopus, to describe the characteristic ballooning of the LV apex. The regional dysfunction of the left ventricle is transient, extends beyond the territory perfused by a single coronary artery, and occurs in the absence of angiographically obstructive coronary artery disease or plaque rupture on imaging or histopathologic evaluation. The left ventricle in its most common typical form demonstrates hypokinesis or akinesis of mid and apical segments, while the basal walls remain hyperkinetic.

Epidemiology

There has been increasing recognition of this syndrome since its initial description. Across multiple series, stress cardiomyopathy occurs in 1% to 2% of patients presenting with STEMI or troponin-positive ACS. Of 2944 patients who underwent left heart catheterization and presented with troponin-positive ACS, 1.2% (35 patients) presented with wall motion abnormalities typical of this syndrome.¹⁹⁹ In a systematic review of 14 studies (from 1996 to 2005), Gianni and colleagues²⁰⁰ described apical ballooning in 2% of patients presenting with STEMI. There were 6837 (0.02%) patients diagnosed with takotsubo syndrome among 33,506,402 hospitalizations in a recent Nationwide Inpatient Sample (NIS).²⁰¹ The incidence of hospitalization with takotsubo syndrome increased more than 4- to 19-fold from 2006 to 2012 in the NIS.^202,203 Similarly, hospitalization rates for Medicare fee-for-service beneficiaries with principal and secondary diagnosis of takotsubo syndrome increased from 5.7 per 100,000 person-years in 2007 to 17.4 in 2012, but whether the increase represents greater recognition or a true epidemiologic trend is unclear.²⁰⁴

Pathogenesis

The underlying mechanisms of takotsubo syndrome remain unknown. The association of Takotsubo syndrome with physical and emotional stressors and its presence in conditions with catecholamine excess (pheochromocytoma, dobutamine stress echocardiography) point to a pathophysiologic role of catecholamine. Plasma catecholamine levels in 13 patients with stress-related cardiomyopathy were elevated when compared to 7 patients with Killip class III myocardial infarction²⁰⁵; however other studies do not report similar increase in the levels of adrenergic hormones.²⁰⁶ In eight patients with takotsubo syndrome who underwent serial endomyocardial biopsies during the acute and recovery phases, Nef and colleagues²⁰⁷ observed contraction bands and increases in cytoplasmic proteins and glycogen without evidence of cell death, suggesting catecholamine-mediated microcirculatory disturbance inducing ischemia and direct catecholamine myocardial toxicity. The mechanisms underlying the association of sympathetic overactivity and myocardial ischemia could be caused by resultant vasoconstriction; however, the results are conflicting. In a study of 30 patients with takotsubo syndrome from Japan, coronary angiography revealed spontaneous coronary artery spasm in 3 patients, and among 14 patients, ergonovine or acetylcholine induced single (n = 4) or multivessel (n = 6) spasms.²⁰⁶ In a meta-analysis, spontaneous multivessel spasm was uncommon (3 of 212 patients; 1.4%), and results of spasm on provocation varied and ranged from 9% to 100%, indicating involvement of other underlying potential mechanisms.²⁰⁰ Reversible myocardial dysfunction can result from multivessel coronary spasm; however, in its absence, diffuse coronary microvascular impairment has been put forward as an explanation for the dynamic nature and more extensive myocardial involvement. Elesber and colleagues²⁰⁸ noted abnormal myocardial perfusion in 29 of 42 patients (69%) presenting with takotsubo syndrome, which correlated with the extent of myocardial injury. Measurement of coronary microcirculation assessed by coronary flow velocity reserve and deceleration time of diastolic velocity were abnormal in patients with takotsubo syndrome and recovered 3 weeks later during remeasurement, suggesting a role of abnormal coronary microcirculation in the pathogenesis of takotsubo syndrome.²⁰⁸ In another study from the Mayo Clinic, 10 patients underwent coronary epicardial and microvascular response to intracoronary acetylcholine. Six patients demonstrated epicardial vasoconstriction, and seven (70%) had < 50% increase in coronary blood flow, indicating abnormal microvascular response.²⁰⁹

There is no proven pathophysiologic mechanism to explain the unique cardiac appearance in takotsubo syndrome. In a rat model, Paur and colleagues²¹⁰ demonstrated apical-basal gradients with higher apical β₂-adrenergic receptor density and sensitivity than basal cardiomyocytes, which can explain mid and apical myocardial predilection. Inability of norepinephrine at similar or higher doses to initiate stress-related cardiomyopathy excluded the possibility of coronary vasospasm or role of β₁-adrenergic–mediated signaling in its pathophysiology.²¹⁰

Presentation

Clinical

Takotsubo syndrome typically affects postmenopausal women. The International Takotsubo Registry reported 89.8% of women (mean age, 66.8 years) among 1750 patients with this diagnosis.²¹¹ In the NIS sample, 90% of patients with Takotsubo syndrome were women, and 90% were older than 50 years.²⁰¹ In the Swedish Coronary Angiography and Angioplasty Register, between 2009 and 2013, among 505 patients diagnosed with Takotsubo syndrome, 87.5% of patients were women, with a mean age of 67 years.²¹² In a cohort of 136 consecutive patients with takotsubo syndrome, about one-third were male, ≤ 50 years old, without a stressor, or with in-hospital death, nonfatal recurrence, embolic stroke, or delayed normalization of ejection fraction, underscoring the clinical and demographic diversity of takotsubo syndrome at presentation.²¹³

Patients with takotsubo syndrome typically present with clinical and ECG features indistinguishable from more common ACS (angina, shortness of breath, palpitations, syncope, or cardiogenic shock) occurring as a consequence of plaque rupture; however, patients with takotsubo syndrome may describe a wave of pressures from the chest to the neck or head, consistent with the catecholamine surge frequently associated with diaphoresis and heightened anxiety.²¹⁴ In the largest takotsubo registry, chest pain (75.9%), dyspnea (46.9%), and syncope (7.7%) were the predominant symptoms.²¹¹

Electrocardiographic

ECG findings are insufficient to differentiate between AMI and takotsubo syndrome.²¹⁵ The ECG changes are transient and resolve within months in most cases. In the International Takotsubo Registry, 90% of patients with takotsubo syndrome had sinus rhythm, and patients with takotsubo syndrome had similar rates of STEMI (31.1%) and lower rates of ST-segment depression compared with patients with ACS.²¹¹ In a study from the Mayo Clinic, anterior STEMI or new LBBB was absent in two-thirds of patients with takotsubo syndrome; these patients presented with deep T-wave inversion (30%; > 3 mm in three contiguous leads) or nonspecific ST-T–wave abnormalities.²¹⁶ Subsequent T-wave inversion and Q-wave formation also are frequently noted with this syndrome. Most series have reported anterior STEMI at the time of presentation in > 90% of patients, which rapidly evolved into deep symmetrical T-wave inversions within 24 to 48 hours associated with prolongation of the QT interval and transient anteroseptal Q waves.²¹⁷

Triggers

Physical or emotional triggers such as unexpected death in the family, abuse, fierce argument, financial loss, and natural disasters (earthquakes), among others, typically precede the presentation of patients with takotsubo syndrome (Table 40–9).^214,218 The incidence of triggers varies among different series and ranges from 14% to 100%.^{199,206,213,219,220,221,222,223}

In the International Takotsubo registry,²¹¹ neurologic or psychiatric disorders were present in more than half of patients with takotsubo syndrome (55.8%), and these disorders were evident in only 25.7% of patients presenting with ACS (P < .001).

The distinction between primary or principal versus secondary diagnosis of takotsubo syndrome is important because outcomes for secondary diagnosis of takotsubo syndrome are likely driven by the primary reason for hospitalization.^7,17 Patients with takotsubo syndrome with principal diagnoses of ACS more likely represent primary coronary presentations, whereas hospitalizations with secondary diagnoses of takotsubo syndrome reflect underlying acute medical illness as the primary reason for admission, and in the latter situation, acute medical illness precipitates takotsubo syndrome. Patients with secondary diagnosis have worse short- and long-term prognosis.²²⁴

Diagnosis

Approach

In a patient presenting with suspected ACS (particularly postmenopausal women), the diagnosis of takotsubo syndrome rests on the reversible nature of the LV dysfunction and its extension beyond single epicardial coronary distribution; absence of any evidence of culprit atherosclerotic coronary artery disease; significant increase in natriuretic peptide; and a positive but relatively small elevation of serum troponin levels (Table 40–10). The clinical manifestations and electrocardiographic abnormalities are generally out of proportion to the degree of elevation in cardiac biomarkers. An identifiable physical or emotional trigger is often, but not always, present.

The diagnosis of takotsubo syndrome generally requires coronary angiography, serial assessment of LV systolic function (initial assessment generally by ventriculography or echocardiography with subsequent assessment by echocardiography), an ECG, and cardiac troponin level. Several diagnostic criteria have been proposed, including those by the Mayo Clinic (modified in 2008), the Japanese Takotsubo Cardiomyopathy Group, the Gothenburg Group, and the Takotsubo Italian Network and, more recently, by the Heart Failure Association of the ESC (see Table 40–10).

Special Considerations

Coronary Artery Disease

Absence of obstructive coronary disease or angiographic evidence of acute plaque rupture is an important criterion; however, the diagnosis of takotsubo syndrome can still be made in the presence of coronary artery disease if the wall motion abnormalities are not in the distribution of the coronary disease. This exception is made because some patients with stress cardiomyopathy have concurrent coronary disease (15.3% in the International Takotsubo Registry, 9.6% in the Takotsubo Italian Network Registry).²²⁵

Pheochromocytoma

Catecholamine surge triggers reversible LV dysfunction and is encountered in about 10% of cases of pheochromocytoma. Most diagnostic criteria for takotsubo syndrome exclude patients with pheochromocytoma, myocarditis, intracranial bleed, or significant coronary artery disease. The pathophysiology and clinical phenotype of pheochromocytoma are identical with takotsubo syndrome, and such patients were included in the recent Heart Failure Association criteria as cases of secondary takotsubo syndrome. Patients with pheochromocytoma presenting with takotsubo syndrome were younger (fourth to fifth decade), had fewer antecedent stressors, had a lower incidence of typical features of pheochromocytoma (headache, palpitations, diaphoresis, and hypertension), and had higher presentation with shock.²²⁶

Myocarditis

Distinction of myocarditis from takotsubo syndrome is sometimes difficult, especially when myocardial edema and inflammation overlap in the anatomic distribution seen in patients with acute takotsubo syndrome.^227,228,229 A careful history of antecedent viral illness, rise in inflammatory markers, diffuse ventricular involvement on echocardiogram or LV angiogram, late gadolinium enhancement with nonischemic distribution, and inflammatory cells on myocardial biopsy help differentiate myocarditis from takotsubo syndrome.²¹⁴

Biomarkers and Takotsubo Syndrome

Cardiac Biomarkers

Serum cardiac troponin levels are elevated in most (> 90%) patients with takotsubo syndrome (eg, median initial troponin of 7.7 times the upper limit of normal with an interquartile range of 2.2 to 24 in the International Takotsubo Registry study); however, the rise in troponin or creatinine kinase is disproportionately low relative to the extent of regional wall abnormality or LV dysfunction.²³⁰ In contrast, levels of brain natriuretic peptide (BNP) or N-terminal pro-BNP (NT-proBNP) are elevated in most patients with stress cardiomyopathy. BNP levels were elevated in 82.9% of patients with takotsubo syndrome in the International Takotsubo Registry study, with a median level 6.12 times the upper limit of normal (interquartile range, 2.12-15.70), and exceeded those seen in a matched cohort of patients with ACS. Best differentiation of takotsubo syndrome and ACS is possible by determining a ratio between peak levels of NT-proBNP (ng/L) to troponin T (μg/L). A cut-off value of NT-proBNP (ng/L)/troponin T (μg/L) ratio of 2889 distinguished takotsubo syndrome from STEMI (sensitivity: 91%; specificity: 95%), whereas an NT-proBNP (ng/L)/troponin T (μg/L) ratio of 5000 distinguished well between takotsubo syndrome and non–ST-segment elevation myocardial infarction (sensitivity: 83%; specificity: 95%).²³¹

Radionuclide Myocardial Perfusion Imaging

Radionuclide myocardial perfusion imaging is generally not indicated in patients presenting with suspected takotsubo syndrome because most have high-risk features for ACS (eg, elderly, elevated cardiac troponin levels) and will require coronary angiography. Patients with suspected non–ST-segment elevation ACS with low- or intermediate-risk features may undergo radionuclide myocardial perfusion imaging. Hypocontractile LV segments demonstrate normal perfusion but reduced uptake of ¹⁸F-fluorodeoxyglucose and ¹²³I-metaiodobenzylguanidine (MIBG).^232,233,234 The extent of the metabolic defects and innervation abnormalities are overlapping, and with improvement in LV function, progressive improvement of both glucose metabolism and sympathetic innervation is noted. A recent study evaluated cardiac sympathetic activity by MIBG scan in 32 patients who presented with initial symptoms and signs closely mimicking those of takotsubo syndrome with suspicion of ST-segment elevation ACS. The major findings were that patients with takotsubo syndrome showed cardiac hypersympathetic activity (decreased heart-to-mediastinal ratio and increased washout rate on MIBG scan) in the subacute phase, compared with their recovery at 4 months.²³⁵

Cardiac Magnetic Resonance Imaging

Cardiac magnetic resonance imaging is helpful in borderline cases, especially when the distinction needs to be made from myocarditis. In a prospective study of 256 cases of takotsubo syndrome, cardiovascular magnetic resonance imaging data (available for 239 patients [93%]) revealed four distinct patterns of regional ventricular ballooning: apical (n = 197 [82%]), biventricular (n = 81 [34%]), midventricular (n = 40 [17%]), and basal (n = 2 [1%]); and accurately identified all cases using specific criteria: a typical pattern of LV dysfunction, myocardial edema, absence of significant necrosis/fibrosis, and markers for myocardial inflammation.²³⁶ Follow-up cardiovascular magnetic resonance imaging showed complete normalization of LV ejection fraction and inflammatory markers in the absence of significant fibrosis in all patients.

Traditionally, the absence of late gadolinium enhancement (LGE) is a common diagnostic criterion for takotsubo syndrome.²³⁶ Subtle fibrosis presenting as LGE may be seen in patients with takotsubo syndrome; however, the myocardial scarring is patchy, and the signal intensity and extent of the late gadolinium enhancement are smaller.²³⁷ Absence of significant late gadolinium enhancement combined with myocardial edema and marked ballooning of the left ventricle is a unique feature of takotsubo syndrome. The inverse flow metabolism mismatch noted with positron emission tomography in patients with takotsubo syndrome has not been well studied.²³⁸

Coronary and Left Ventricular Angiography and Echocardiography

Coronary angiography is indicated in most patients because of their high-risk presentation with ECG abnormalities and elevation of cardiac biomarkers. Incidental coronary artery disease can be present in up to 10% of cases; however, characteristics of coronary angiography and intravascular ultrasound in patients with takotsubo syndrome include absence of plaque rupture, thrombus, and coronary dissection.^239,240,241 In addition, degree and type of LV dysfunction on LV angiography exceed the angiographic findings. LV angiography should be performed as soon as a diagnosis of takotsubo syndrome is suspected. If LV angiography cannot be performed, transthoracic echocardiography is the preferred noninvasive imaging technique. Advantages of echocardiography include confirmation of findings of LV angiography without the need of contrast administration, right ventricle involvement, and mechanical complication (rupture), and detection of mitral regurgitation and LV outflow obstruction.^242,243,244 In the Italian Takotsubo Network, echocardiographic features like low ejection fraction, elevated E/e′ ratio, reversible moderate-to-severe mitral regurgitation, and age ≥ 75 years were independent risk factors for acute HF, cardiogenic shock, and in-hospital death.²⁴⁵ Echocardiogram can be safely repeated to demonstrate timeline and degree of improvement of LV function.

Ventricular Dysfunction and Variants

Ubiquity and reversibility of LV dysfunction are the hallmarks of takotsubo syndrome. Reduced LV ejection fraction (mean 40.7% ± 11.2%) was noted in 86.5% of patients with takotsubo syndrome on admission but in only 54.2% of patients with an ACS in the International Takotsubo Registry. In a systematic review of 14 studies, the LV function ranged from 20% to 49%.²⁰⁰ Lower ejection fraction is also a predictor for cardiogenic shock. This complication is noted in 2.8% to 12.4% of patients with takotsubo syndrome.^{246,247,248,249} The focus of most studies of takotsubo syndrome is on LV dysfunction; however, there is evidence that about one-third of cases involve both right and left ventricles.²³⁶ Based on LV angiography or transthoracic echocardiography, takotsubo syndrome is classified as one of the following five types (Fig. 40–6).²¹¹

Apical Type

The left ventricle regional wall motion abnormality, as seen using the echocardiogram or LV angiography, is one of the characteristic features of takotsubo syndrome. Among the several patterns, the apical type is most common and is characterized by systolic apical ballooning of the left ventricle, reflecting depressed mid and apical segments with hyperkinesis of the basal walls. This type was present in 81.7% of patients in the International Takotsubo Registry study.

Midventricular Type

In this type of takotsubo syndrome, the base and apical portions of the left ventricle are spared, with ventricular hypokinesis restricted to the mid-ventricle. This type was present in 14.6% of patients in the International Takotsubo Registry study.

Basal Type

This type is also referred to as reverse or inverted takotsubo syndrome and is characterized by hypokinesis of the base with sparing of the mid-ventricle and apex. This type was present in 2.2% of patients in the International Takotsubo Registry study.

Focal Type

This rare variant is characterized by isolated dysfunction of the left ventricle, commonly the anterolateral segment. This type was present in 1.5% of patients in the International Takotsubo Registry study.

Global Type

Rarely, patients with takotsubo syndrome have global hypokinesis.

Treatment

Overview

The presentation of patients with takotsubo syndrome overlaps with that of patients presenting with ACS. Clinical suspicion of takotsubo syndrome should not delay the management because presentation in a significant majority of these patients is a result of plaque rupture that would benefit from early reperfusion. Patients with ST-segment elevation on ECG should undergo urgent cardiac catheterization with the aim to perform primary PCI. The diagnosis of takotsubo syndrome should be suspected if coronary angiography fails to demonstrate critical coronary artery disease, especially with features of plaque rupture. LV angiography showing apical ballooning or mid-wall hypokinesis confirms the diagnosis of takotsubo syndrome. In patients treated with fibrinolytic therapy, subsequent angiogram without significant coronary artery disease favors the diagnosis of takotsubo syndrome; however, recanalization of the infarct-related artery with recovery of LV function and ACS with intact fibrous cap blurs the distinction between takotsubo syndrome and ACS.²⁵⁰ Patients without ST-segment elevation with high-risk features (eg, age, positive biomarkers, congestive HF) may merit early coronary angiography that would help the clinicians make the diagnosis of takotsubo syndrome.

Stable or Low-Risk Patients

The major objective of treatment in patients with takotsubo syndrome is to support the left ventricle and manage complications during the acute phase of presentation. Management of HF during acute presentation should follow the standard guidelines and should include β-blockers and ACE inhibitors. Recommendations for anticoagulation to prevent thromboembolism in patients with takotsubo syndrome are similar to those for patients with ventricular thrombus in the setting of ACS.

High-Risk Patients

Heart Failure

HF should be managed in the critical care setting with continuous ECG monitoring. Systolic HF is the most common complication during the acute phase. Independent predictors of acute HF are advanced age, low ejection fraction at presentation, higher admission and peak troponin levels, and a physical stressor preceding the onset of symptoms.²¹⁴ Acute systolic failure, especially in the setting of cardiogenic shock, may require assessment of LV outflow tract. LV outflow obstruction with gradients ranging between 20 and 140 mm Hg has been observed in 10% to 25% of patients.²⁵¹ In addition, dynamic outflow tract obstruction caused by hyperkinesis of the basal LV segments and the resultant systolic anterior motion of the mitral valve may cause and/or worsen hypotension. Patients with takotsubo syndrome who have hypotension or are in shock should get urgent echocardiography to rule out LV outflow obstruction. Increase in preload with judicious use of intravenous fluids and reduction in myocardial contractility with short-acting β-blockers are helpful. Peripheral vasoconstrictors such as phenylephrine can be given if β-blockers or fluid administration are contraindicated or ineffective. Mitral regurgitation and outflow obstruction improve with improvement in the LV function.

Patients with hypotension and shock but without significant LV outflow obstruction should receive cautious fluid resuscitation and mechanical support with temporary left ventricle assist devices and extracorporeal membrane oxygenation.^252,253 The use of inotropes (dopamine, dobutamine, epinephrine, or norepinephrine) is generally contraindicated because these agents further activate the catecholamine receptors and worsen the clinical status and prognosis of these patients.²⁵⁴ IABP may worsen the LV outflow obstruction and should be avoided.²¹⁴ The use of levosimendan, a noncatecholaminergic inodilator, has been studied in small series, and its role in the management of patients with takotsubo syndrome with acute systolic HF is ill defined.²⁵⁵

Arrhythmias

Arrhythmias such as atrial fibrillation (seen in 5%-15%), VT, and ventricular fibrillation (seen in 5%-9%) are not uncommon in takotsubo syndrome.^{248,256,257,258} In a systematic review, life-threatening arrhythmias such as third-degree atrioventricular block, VT, ventricular fibrillation, and cardiac arrest were seen in 37 of 353 patients (14.6%; 95% CI, 10.9%-18.3%).²³⁰ In the International Takotsubo Registry, VT, thrombus, and cardiac rupture were seen in 3.0%, 1.3%, and 0.2% of patients, respectively.²¹¹ Atrial arrhythmias can worsen the cardiac output and HF, whereas ventricular arrhythmias can cause cardiac arrest in 4% to 6% of patients, and the standard guidelines for treatment should be followed.

Left Ventricular Thrombus

Significant LV dysfunction is a nidus for thrombus formation. It is seen in 2% to 8% of patients presenting with takotsubo syndrome and can result in peripheral arterial embolism.²⁴⁸ Most thrombi develop within the first week; however, thrombus formation at 2 weeks after symptom onset has been described.^259,260 Once recognized, patients should be treated with oral anticoagulation for 3 months. Follow-up imaging is recommended to confirm recovery of apical contractile function.

Recovery of Left Ventricular Function

Complete recovery of LV systolic function is necessary to confirm the diagnosis of takotsubo syndrome. The recovery time varies and can be as short as several days or as long as several weeks. In the International Takotsubo Registry, mean LV ejection fraction increased from 41.1% at admission to 51.2% prior to discharge before reaching normal values (59.9%) at 60-day follow-up.²¹¹ It is recommended to follow the same approach to treat LV systolic dysfunction as that for other causes of cardiomyopathy (with ACE inhibitors and β-blockers), at least until LV systolic function recovers.

Recurrence

Recurrence rates following an index episode are approximately 10%. Five-year recurrence rates of 5% to 22% have been noted.^{217,218,257,261} The rate of recurrence in the International Takotsubo Registry was 1.8% per patient-year, with a span of 25 days up to 9.2 years following the index episode.²¹¹ A study from the Mayo Clinic detected a recurrence rate of 11.4% in 100 patients at a mean follow-up of 4.4 ± 4.6 years. In that study, the mortality was 16%, which was similar to an age- and sex-matched local population.²⁴⁷

Prognosis

Previously, the prognosis of takotsubo syndrome was considered benign,²⁶² and a meta-analysis showed that only 1.1% of reported patients died during the index hospitalization.²⁰⁰ In a study from the Mayo Clinic, 4-year survival was not different from that of an age- and gender-matched population.²⁴⁷ These observations could be a result of small sample size of previous studies, selection bias with inclusion of lower risk patients, and rapid recovery of LV function. However, worse short- and long-term prognosis has been recently reported. A Swedish registry study found a 30-day mortality of 4% in 302 patients with takotsubo syndrome, which was similar to patients presenting with non–ST-segment elevation myocardial infarction but lower than patients with STEMI.²⁶³ The International Takotsubo Registry reported a 30-day mortality of 5.9%.²¹¹ The rate of death during long-term follow-up was 5.6% per patient-year, and men had a higher incidence of short- and long-term mortality and morbidity as compared to women. Data from a large, multicenter registry suggested better outcomes for elderly patients with emotional triggers and a worse prognosis for younger patients with physical triggers and emotional or psychiatric diseases. Furthermore, a recent bicentric study in 286 prospectively identified patients with takotsubo syndrome revealed 28-day, 1-year, and long-term mortality of 5.5%, 12.5%, and 24.7%, respectively.²⁶⁴ Long-term mortality was noted to be higher as compared to patients with STEMI (24.7% vs 15.1%; hazard ratio, 1.58; 95% CI, 1.07-2.33; P = .02). These findings challenge the initial notion of a favorable prognosis in patients presenting with takotsubo syndrome. Prognosis of patients presenting with secondary takotsubo syndrome is worse compared with patients presenting with ACS (primary or principal takotsubo syndrome). Murugiah et al²⁰⁴ examined the US Medicare database and reported 30-day and 1-year mortality of 2.5% and 6.9% for patients with principal takotsubo that compared favorably to rates of 4.7% and 11.4% for patients with secondary takotsubo, respectively.²⁰⁴ Excess mortality dominantly occurs in the first 4 years after diagnosis, is noncardiac, and is largely a result of illnesses such as malignancy.

Follow-Up

The follow-up of patients with takotsubo syndrome is no longer considered favorable. The novel insights gleaned from recent studies confirm that patients are prone to severe complications during the acute and subacute phases of the disease, including HF, cardiogenic shock, life-threatening arrhythmias, or thromboembolic complications. Therefore, the latest position statement from the ESC suggests close monitoring in coronary care units, in a similar fashion as recommended for patients with ACS. The ESC recommends documentation of recovery of LV function, and ECG changes should be documented during regular follow-up visits.²¹⁴

Despite the numerous improvements in the management of STEMI, it remains one of the leading causes of morbidity and mortality worldwide. The key steps in the management of STEMI patients include rapid diagnosis, prompt delivery of initial therapeutic agents, immediate reperfusion, and diligent in-hospital management, which includes the aggressive implementation of secondary prevention strategies. Future improvements in each of these categories, some of which are logistical, need to be made if any significant reductions in the morbidity and mortality are to be obtained. The persistently high mortality of myocardial infarction in the elderly provides a formidable but highly important challenge for the future.

We would like to thank Dr. Emily E. Hass, Dr. Eric H. Yang, and the late Dr. Robert A. O’Rourke, who contributed to the previous version of this chapter in the 13th edition.