CHAPTER 39: EVALUATION AND MANAGEMENT OF NON–ST-SEGMENT ELEVATION MYOCARDIAL INFARCTION

As discussed in Chap. 36, acute coronary syndromes (ACS) refer to a pattern of clinical symptoms that are consistent with acute myocardial ischemia (Fig. 39–1)¹; the pathophysiology, findings, and treatment of ACS range along a common spectrum. This chapter discusses two closely related forms of ACS, namely unstable angina (UA) and non–ST-segment elevation myocardial infarction (NSTEMI).

The pathophysiology of ACS classically involves erosion or rupture of an atherosclerotic plaque with thrombus formation that obstructs the coronary artery lumen. Accordingly, patients with an ACS are frequently treated similarly with individual variations in management depending on the classification of patient risk.^1,2

UA/NSTEMI is also termed non–ST-segment elevation ACS (NSTE-ACS). Angiographic, intravascular ultrasound (IVUS), and angioscopic studies indicate that acute UA/NSTEMI usually results from the disruption of an atherosclerotic plaque with a subsequent platelet-rich thrombus that obstructs microvascular blood flow and may transiently or partially obstruct epicardial blood flow.

Initial diagnosis and management are based on information available at the time of presentation and are updated using new information accumulated over time.^1,2 Initial patient evaluation includes rapid efforts to distinguish noncardiac chest pain from myocardial ischemia. The clearest separation between UA and NSTEMI is the absence or presence, respectively, of abnormal concentrations of biomarkers indicative of myocardial necrosis, either the troponins (which are structural proteins) or the MB band of creatine kinase (CK-MB, which is a cardiac enzyme); the clearest separation between this end of the spectrum and ST-segment elevation myocardial infarction (STEMI) is made by the electrocardiogram (ECG).

A patient with symptoms consistent with ACS should have an ECG performed and interpreted within 10 minutes. The most important goal of the early ECG is to identify patients with STEMI who are candidates for immediate reperfusion therapy. Each patient should be given a provisional diagnosis of (1) definite ACS, which should be classified as STEMI, NSTEMI, or UA; (2) possible ACS; (3) a non-ACS cardiac condition (eg, chronic stable angina or heart failure); or (4) a noncardiac diagnosis, which should be as specific as possible. If a provisional diagnosis of ACS is assigned, risk assessment should be performed to determine the probability of major cardiac complications.^1,2 Such risk assessment is not just important among individuals with definite ACS; among patients with possible ACS, risk assessment should be used to determine the contingent probability of an adverse cardiac event if the diagnosis of ACS is confirmed, because this information will guide appropriate triage, medical therapy, and timing of subsequent evaluation, including the use of invasive procedures.

Mismatch between myocardial oxygen supply and demand is the cause of myocardial ischemia. Most cases of UA/NSTEMI are attributable to acute reductions in myocardial oxygen supply, which are caused by a platelet-rich thrombus that develops after rupture or erosion of an atherothrombotic plaque (see Chaps. 33, 36, 37, and 38). The thrombus may be flow limiting, but usually does not completely occlude the epicardial lumen; microembolization of platelet aggregates and components of the disrupted plaque are key components of the acute ischemic syndromes and are likely responsible for the release of markers of myonecrosis.

Other less common causes of UA/NSTEMI include dynamic obstruction, which may be a result of intense focal spasm of a segment of an epicardial coronary artery (Prinzmetal or variant angina, as discussed specifically in the later section Variant Angina). UA/NSTEMI may also occur because of severe fixed narrowing of the epicardial coronary artery without spasm or thrombosis, which may occur because of progressive atherosclerosis or restenosis after percutaneous coronary intervention (PCI). In such cases, the presentation may be more gradual than among those with a de novo plaque rupture and may initially manifest consequent to increase in myocardial oxygen demand beyond what can be supplied by the diseased epicardial coronary arteries. Less commonly, UA/NSTEMI may be caused by coronary dissection or vasculitis; spontaneous dissection of the coronary artery (SCAD) is an increasingly recognized entity, particularly in younger, female patients.³

Secondary UA/NSTEMI occurs when the precipitating condition is extrinsic to the coronary arterial bed. Such patients usually have underlying fixed coronary atherosclerotic narrowing; however, secondary UA/NSTEMI may also be seen in the absence of severe obstructive disease with acute increases in myocardial oxygen demand (eg, fever, tachycardia, hypertensive emergency), reductions in coronary blood flow (eg, hypotension), or diminution in myocardial oxygen delivery (eg, severe anemia), or a combination of these (eg, pulmonary embolism, sepsis). Among individuals with severe left ventricular (LV) hypertrophy, secondary UA/NSTEMI may occur with a lesser degree of imbalance of oxygen supply and demand.

UA may present as angina that is new onset; it may manifest as angina increasing in frequency, duration, or severity; or it may present immediately as rest angina. NSTEMI typically presents with rest symptoms that are more prolonged or intense. The different presentations of UA reflect differences in pathophysiology and short-term prognosis, which have implications for therapy. Whereas rest angina more commonly results from an unstable coronary plaque (as described earlier and in Chaps. 33, 37, and 38), new-onset or progressive exertional angina more commonly is caused by fixed atherosclerotic disease without plaque rupture. Other settings include UA/NSTEMI within 1 year after PCI, which may be caused by restenosis or from acute thrombosis of a drug-eluting stent (DES); the latter scenario typically occurs in the context of suspension of dual antiplatelet therapy (DAPT).

Patients with ACS may present with “atypical” symptoms, which include acute dyspnea, indigestion, unusual locations of pain, agitation, altered mental status, profound weakness, and syncope. Such presentations are more common in women, advanced elderly individuals, and patients with long-standing diabetes mellitus and are associated with higher risk for death and major complications.⁴ In addition, UA/NSTEMI may be present without evident clinical symptoms, particularly among patients in the perioperative state and those with comorbid medical conditions such as diabetes.

Patients with symptoms suggestive of ACS require a rapid evaluation and expedient decision making, with the intensity of monitoring and therapy determined both by the probability of ACS and the associated likelihood of a poor outcome if ACS is subsequently confirmed and/or managed without revascularization. Factors that modulate the development and outcomes of ACS are listed in Table 39–1. Approximately 15% of patients evaluated in US emergency departments with symptoms suggestive of ACS ultimately have an ACS diagnosis confirmed; conversely, patients with unrecognized ACS have a substantially higher mortality risk. Thus, an organized, efficient, and systematic strategy is needed to avoid exposing patients to risk associated with both the diagnostic procedures as well as the therapies used and unnecessary resource utilization, while simultaneously ensuring accuracy of diagnosis. Formal chest pain or ACS protocols or critical pathways are recommended strategies to provide a systematic approach to the evaluation of possible ACS and to provide greater adherence to guideline recommendations.^1,2

Initial Evaluation

The initial evaluation of a patient with possible UA/NSTEMI is focused on addressing two interrelated questions^1,2: (1) How likely are the presenting symptoms and signs to be attributable to an ACS event? (2) If ACS is confirmed, what is the risk to the patient? This early assessment of short-term risk determines the intensity of initial and often subsequent treatment. Determining whether the presentation is attributable to ACS may prove challenging in the absence of ECG changes or typical patterns of cardiac biomarker elevation. In a patient with known coronary artery disease (CAD), typical symptoms are more likely to be caused by myocardial ischemia, especially if the current symptoms are identical to previous episodes when CAD was objectively documented as the cause. However, because the accuracy of descriptive chest pain characteristics is only moderate, a systematic approach to patient evaluation is required.

History

Patients with ACS often experience discomfort typical of angina, but such episodes are more severe, prolonged, and with lower threshold; they often occur at rest or with very low levels of exertion. Also, chest discomfort because of UA/NSTEMI is less likely to be completely relieved by nitroglycerin (NTG) than pain from stable angina. Some patients may have no chest discomfort but present solely with jaw, neck, ear, arm, or epigastric discomfort or with dyspnea in the absence of pain. These can be “anginal equivalents.”

Nausea, perspiring, and/or shortness of breath may accompany episodes of chest discomfort. In elderly or diabetic patients, these symptoms may be the only indication that myocardial ischemia is present. Women also present challenges with regard to diagnosis as they are more likely than men to have ACS with an “atypical” presentation⁴ and diagnoses such as SCAD are much more common in women; because SCAD usually occurs in coronary arteries without substantial atherosclerosis, affected patients are often medically quite healthy (without conventional risk factors for CAD), making consideration for ACS less likely. When UA is suspected in a patient younger than age 50 years, it is particularly important to consider cocaine use. Cocaine can cause coronary vasoconstriction, vasospasm, and thrombosis in addition to its direct effect on altering myocardial oxygen demands through increases in heart rate and blood pressure.⁵

Evaluation of patients with suspected UA/NSTEMI should include the physician's opinion of the probability of the symptoms being caused by myocardial ischemia, categorizing the presentation into high-, intermediate-, or low-probability categories (Table 39–2).

Electrocardiography

The diagnostic yield of the 12-lead ECG is enhanced greatly if it can be recorded during an episode of chest discomfort; consideration for frequently obtaining repeat tracings may increase sensitivity. Although a completely normal ECG during chest pain does not rule out ACS, it reduces the probability significantly and is a favorable prognostic sign. Transient ST-segment depression of at least 0.5 mm (Fig. 39–2) that appears during chest discomfort and disappears after relief provides objective evidence of transient myocardial ischemia. When it is a constant finding with or without chest pain, it is less specific. A common but nonspecific ECG pattern in patients with UA/NSTEMI consists of persistent negative T waves over the involved area (Fig. 39–3). Deeply negative T waves across the precordial (anterior) leads suggest a proximal, severe, left anterior descending coronary artery stenosis as the culprit lesion and are considered a marker of high risk.

ECG abnormalities may appear or progress in the absence of new symptoms or signs in patients with ACS. Accordingly, it is appropriate and diagnostically useful to obtain serial ECGs during the initial observation period, before discharge, and during recurrent episodes of chest pain.

Biochemical Cardiac Markers

Biochemical cardiac markers are useful for both the diagnosis of myocardial necrosis and the estimation of prognosis. Cardiac troponins are the preferred biomarkers for the diagnosis of myocardial infarction (MI), whereas CK-MB is a less preferred option and should only be used when troponin is not available.⁶ Cardiac troponins T and I (cTnT and cTnI) are structural components of the cardiac contractile apparatus (they are not enzymes) and are found not only in the sarcomere of the cardiomyocyte, but also in small concentrations within the cytosol. Both cTnT and cTnI are thought to be nearly entirely cardiac in origin; this confers very high cardiac specificity. Numerous studies have demonstrated that even minor troponin elevations are independently associated with adverse events in populations with NSTE-ACS.⁷ Many of these patients will have normal levels of CK-MB. Coronary angiographic trials have demonstrated that in patients with ACS, troponin elevation is associated with multivessel coronary disease, complex lesion morphology, and visible thrombus, as well as with impairment in microvascular function.⁸ These findings explain the consistent association between troponin concentration, even at low levels, and recurrent ischemic events in patients with ACS.

As troponin assays become more sensitive, the proportion of individuals with elevation continues to increase; with current assays, more than 30% of individuals with a presentation of UA/NSTEMI have detectable levels of cTnI or cTnT, and with more sensitive assays that will soon be clinically available, the proportion is two- to three-fold higher. Such highly sensitive troponin (hsTn) assays are increasingly available in several countries. The main advantages of hsTn assays over conventional methods are superior sensitivity to detect minute quantities in TnT or TnI concentrations and increased analytical precision at these very low concentrations. In this regard, compared to conventional troponin methods, hsTnT and hsTnI have both been shown to be more likely to be abnormal at first draw (Fig. 39–4), to provide incremental prognostic information, and (as noted) to reclassify the diagnosis from UA to NSTEMI in approximately 30% of patients.⁹ The increased sensitivity of hsTn assays also allows for very rapid exclusion of myocardial injury, if extremely low concentrations are noted at presentation¹⁰; such patients could still have UA, however. Lastly, the increased analytical precision (which allows for resolution of small changes in hsTn over short periods of time) facilitates strategies of accelerated diagnostic pathways to exclude NSTEMI in as little as 1 hour.¹¹ With broader availability of hsTnT and hsTnI will come greater experience with these novel assays.

Present recommendations support serial measurement of troponins in patients with suspected NSTE-ACS, seeking a rise and/or fall in troponin concentration and reserving the diagnosis of NSTEMI for patients with clinical signs/symptoms, ECG changes, or loss of myocardial function on noninvasive assessment.⁶ Short of these corroborating elements, caution is necessary. It is clear that patients who have isolated elevation in troponin levels should be classified as having an NSTEMI, provided the clinical syndrome is consistent with ACS. This caveat is critically important and is discussed further in the later sections. A reflex cardiac evaluation based solely on an elevated cardiac biomarker without considering whether the episode is indeed consistent with an ACS is inappropriate and may be harmful. In recent years, several different “rule-in” and “rule-out” protocols have been proposed. The most recent European Society of Cardiology (ESC) guidelines for the management of ACS provide excellent examples of both by using the highest sensitivity troponin assays.²

Although troponins are highly specific for myocardial injury, they are not specific for the cause of cardiac injury. Multiple noncardiac diagnoses may cause cardiac injury and lead to troponin elevation (Table 39–3). Particular consideration should be given to the diagnosis of acute pulmonary embolism, which may present with nonspecific chest symptoms, tachycardia, and ST-segment and T-wave changes along with low-level elevation in cardiac troponins. Moreover, among individuals with chronic cardiovascular (CV) conditions, such as heart failure or chronic CAD, as well as individuals from the general population with diabetes, chronic kidney disease, or asymptomatic LV hypertrophy or LV dysfunction, chronic troponin elevation may be seen, even in the absence of signs and symptoms of ischemia.^12,13,14 Therefore, it is crucial to interpret troponin values in the context of the clinical presentation and available clinical information; this will become all the more important as hsTnT and hsTnI become available.

In most non-ACS conditions associated with cardiac injury, patients with detectable troponin are at increased risk for adverse cardiac outcomes.^14,15 However, it should be recognized that in patients with atypical presentations, elevation in troponin may be caused by a process other than ACS. Interpreting troponins in patients with renal failure presents unique problems. Although persistent troponin elevation frequently occurs in patients on hemodialysis even in the absence of signs of ACS, detectable cTnT or cTnI is still associated with very high cardiac event rates.^16,17,18 Knowledge of prior troponin values may help to assess whether a current troponin elevation is new or changing (ie, potentially caused by ACS) or chronic. If ACS is suspected, troponin elevation should be interpreted similarly in patients with and without chronic kidney disease.

As troponin assays become increasingly sensitive, elevations from causes other than ACS are becoming more common and more frustrating for consultant cardiologists. Indeed, recent estimates suggest that as many as 50% of troponin elevations occurring in the hospital setting may be attributable to disease processes other than ACS, including heart failure.^19,20 Although hsTn assays improve early sensitivity for detection of myocardial infarction,²¹ they will increase the proportion of elevated levels from causes other than ACS.

At present, few data are available to help clinicians distinguish ACS causes of troponin elevation from other causes. Elevations caused by ACS are usually of a greater magnitude and should demonstrate a dynamic pattern characterized by a transient increase or decrease in values. Demonstration of a transient increase or decrease may improve the specificity for ACS. When a cause other than ACS is suspected, treatment should focus on the underlying disease process. An echocardiogram should be considered, given the frequent association of chronic troponin elevation with cardiac structural abnormalities. Empiric treatment with aspirin (ASA) is recommended unless contraindicated, and in selected individuals, empiric treatment with β-blockers and an evaluation for ischemia may be considered after the acute illness has been stabilized.

Chest Pain Units

The evaluation of patients with chest pain who may have an ACS can be difficult and uncertain. Hospitalizing all such patients for an extensive workup is unwise and results in unnecessary tests with patient risk and expense. However, missing the diagnosis of MI is one of the leading causes of malpractice claims for emergency department physicians in the United States, and patients with missed MI have a higher mortality. The chest pain unit was developed as a strategy to provide systematic care, prevent missed MI diagnoses, and avoid unnecessary hospitalizations.

Most chest pain units are in or near the emergency department and are designed to monitor patients with either possible ACS or definite ACS but with low-risk features. Criteria usually include chest pain or related symptoms that may indicate myocardial ischemia but with a normal or nondiagnostic ECG and a normal initial set of cardiac enzymes. In most units, cTnT or cTnI is measured at serial intervals for 6 to 9 hours; as noted, hsTn assays increase first-draw sensitivity and provide utility to considerably shorter “rule-out” protocols.^21,22

Patients promptly receive ASA, an intravenous (IV) line, ECG monitoring, and a 12-lead ECG at specific intervals. If cardiac biomarkers remain negative and no ischemia is detected, a symptom-limited stress test may be performed for diagnostic and prognostic purposes or the patient may be discharged and sent home.¹ In low-risk patients who present at night or on the weekend, it is reasonable to perform the stress test as an outpatient, provided it can be done within 72 hours.

Coronary Computed Tomography Angiography

Coronary computed tomography angiography (CTA) is a potentially useful test among some patients undergoing an ACS/chest pain evaluation^23,24 (see Chap. 17), providing high-quality imaging of the coronary artery anatomy and high sensitivity for CAD (Fig. 39–5). The high sensitivity of CTA for coronary plaque has been found to impart a high negative predictive value for ACS among low- to intermediate-risk patients typically seen in chest pain observation units. Emerging techniques, such as identification of high-risk plaque, help to add to the evaluation of patients with coronary atherosclerosis detected on CTA.²⁵ Other advantages of CTA are that the procedure allows for visualization of other cardiac and pulmonary structures (potentially allowing for detection of unsuspected pathology, such as pericardial effusion) and delivers good sensitivity for detection of less common causes of ACS, including SCAD.

In contrast to the high sensitivity of CTA for plaque presence, specificity is a limitation with CTA because many lesions that appear obstructive by CTA are nonobstructive with coronary angiography; this is most often a result of “negative remodeling” of the coronary vessel, such that large plaque burden may not project into the lumen of the artery. Because of this, the positive predictive value is moderate for significant luminal obstruction.²⁶ Additional limitations of CTA include difficulty in plaque severity quantification in the context of heavy coronary calcification, the requirement for intravenous contrast administration, and exposure to ionizing radiation; newer algorithms for CTA performance now allow for substantial reduction in radiation exposure, however. Establishment of a CTA program requires institutional commitment along with specialized training for interpreting physicians.

Randomized comparative effectiveness studies suggest that appropriate use of CTA in patients at low-to-moderate risk for ACS may reduce need for hospital discharge and shorten time to disposition from the emergency department.²⁷ In such studies, CTA was highly sensitive to detect presence of coronary atherosclerosis and thus provided high sensitivity for ACS. Outcomes were largely the same in those evaluated with CTA versus standard methods (such as myocardial perfusion imaging); downstream testing (such as invasive coronary angiography) may also be increased, presumably as a result of visualization of coronary anatomy.

In the balance, CTA represents a potentially useful test when the probability of severe CAD is low to moderate, study quality may be optimized (heart rate can be reduced to 70 bpm or lower with a β-blocker, and nitrates may be provided), dye exposure is not contraindicated, and anatomic rather than physiologic information is preferred. How the use of CTA will be affected by more widespread use of hsTnT and hsTnI is ambiguous, but the combination appears promising.²⁸

Initial risk assessment of UA/NSTEMI dictates the appropriate intensity of initial therapy. At the low end of the risk scale, a patient may be discharged home with ASA and an early outpatient stress test. By contrast, high-risk patients may be hospitalized in a coronary care unit, treated with multiple drugs, and undergo coronary angiography urgently as a prelude to revascularization.

Clinical Features

The American College of Cardiology (ACC)/American Heart Association (AHA) 2014 Guideline Update for the Management of Patients with Unstable Angina and Non–ST-Segment Elevation Myocardial Infarction¹ and the 2015 ESC Guidelines for the Management of Acute Coronary Syndromes in Patients Presenting Without Persistent ST-Segment Elevation² recommend (Class I, Level of Evidence [LOE] B) using risk-stratification models, such as the Global Registry of Acute Coronary Events (GRACE) Risk Model Nomogram²⁹ or the Thrombolysis in Myocardial Infarction (TIMI) Risk Score³⁰ to assess prognosis. Web-based applications from GRACE (gracescore.org) and TIMI (timi.org) are available.

Features suggesting high risk include ongoing chest pain that lasts longer than 20 minutes, reversible ST-segment changes of at least 0.5 mm, elevated markers of myonecrosis, and signs of significant LV dysfunction. In addition to these features, the GRACE Risk Model Nomogram uses Killip class, vital signs, age, and creatinine in a stepwise scoring system and considers whether the patient had cardiac arrest at admission. Generally, lower risk patients are younger, have worsening angina without rest pain, have a normal or unchanged ECG without evidence of previous infarction, and have normal cardiac enzymes.

It is increasingly recognized that risk assessment should be both integrative and dynamic. Tools that account for multiple risk predictors provide clear enhancement over evaluation of single variables, as discussed later. Moreover, the risk assessment should be updated during hospitalization. For example, continuing angina with ST-segment changes despite medical therapy is an ominous sign that should prompt urgent coronary arteriography with probable revascularization. Episodes of silent ST-segment depression detected by ambulatory ECG recordings also predict an unfavorable course.

Electrocardiography

The ECG is too often overlooked as a powerful risk stratification tool in patients with UA/NSTEMI. Even minor ST-segment depression (> 0.5 mV) is associated with a markedly increased mortality. Among 9461 patients enrolled in the PURSUIT (Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy) study, mortality at 30 days was 5.1% in patients with ST-segment depression versus 2.1% among those without ST-segment depression.³¹ Also, patients with isolated T-wave inversion have a more favorable prognosis than those with ST-segment depression.³² It bears repeating that repeat ECG is recommended in those with an initially normal tracing, particularly during periods of repeat symptoms when an initially normal tracing was obtained at a time when chest discomfort was not present.

Biochemical Markers

Troponin measurements should be used in the risk stratification of patients with UA/NSTEMI to supplement the assessment from clinical and ECG data; consensus documents suggest that CK-MB not be used when troponin is available.⁶ The combination of troponin elevation and ST-segment depression identifies a group at particularly high risk.³³ The association of troponin elevation with high-risk coronary lesion morphology provides mechanistic insight into studies demonstrating that patients with even minor elevations in cTnT or cTnI derive substantial benefit from an aggressive approach with antithrombotic therapies and an early invasive management strategy.^33,34,35

Plasma levels of B-type natriuretic peptide (BNP) and the N-terminal fragment of its prohormone (NT-proBNP) may increase in response to cardiac ischemia in proportion to the size and severity of the ischemic insult.³⁶ In patients with UA/NSTEMI, higher levels of BNP or NT-proBNP measured at presentation or during hospitalization are associated with increased risk for subsequent death or heart failure. As such, measurement of BNP or NT-proBNP may provide additive information to troponin values; such measurement received a Class IIb indication in the 2014 guidelines. Because coronary ischemia in a patient with incident or prevalent heart failure is associated with substantially greater risk than in a patient without heart, failure and because coronary ischemia may lead to decompensation of heart failure itself, use of natriuretic peptide testing may be helpful to round out the diagnostic picture in such patients.

Although many other biomarkers have been proposed for measurement at various time points after ACS, with the exception of troponin and BNP or NT-proBNP,³⁷ few have demonstrated the level of robust evidence necessary for consideration of measurement in routine practice. With increased use of hsTn, the bar will be even higher for newer biomarkers to show clinical utility.

Noninvasive Testing for Ischemia

Stress testing (Table 39–4) is commonly used for risk assessment in patients with UA/NSTEMI who are managed with an initial conservative strategy (also called an ischemia-guided strategy; see below). Low-risk and some intermediate-risk patients who stabilize with medical therapy are candidates for stress testing for risk stratification. Stress tests should be symptom limited rather than submaximal, and a stress ECG without adjunctive imaging is appropriate unless baseline ECG abnormalities would preclude adequate interpretation. Those with high-risk findings, such as ST-segment depression at low exercise levels, or large, reversible perfusion defects should undergo coronary arteriography; those with negative or low-risk results can be treated medically. Low-risk patients who complete a stay in a chest pain unit without objective evidence of myocardial ischemia can safely undergo stress testing for diagnosis and prognostic purposes either immediately or, when possible, within 48 hours as an outpatient.¹

In patients who are unable to exercise, pharmacologic testing with dipyridamole, adenosine, regadenoson, or dobutamine can be used to provide the stress, and sestamibi imaging or echocardiography can be used as a method of assessment (see Chaps. 13, 15, and 18). Stress testing is not needed in patients whose clinical features already put them into a high-risk category; instead, they should undergo early coronary arteriography.

Coronary Angiography

The use of coronary CTA has been previously discussed (see also Chap. 17). Among patients with stable CAD, risk is proportional to the number of vessels with greater than 50% diameter stenosis and the presence and severity of LV dysfunction. However, the relative prognostic impact of the extent of CAD is probably less with ACS because the risk of short-term events is dominated by features of the culprit lesion, such as whether it induces ST-segment depression or troponin release.

Among patients with UA/NSTEMI who undergo arteriography, approximately 25% have one-vessel disease, 25% have two-vessel disease, and 25% have three-vessel disease. Ten percent have significant main stenosis, and the other 15% will have coronary luminal narrowing of less than 50% or normal-appearing vessels on arteriography³⁸ (see also Chap. 20). Patients with left main stenosis of at least 50% or three-vessel disease with LV dysfunction derive a survival benefit from coronary artery bypass graft (CABG) surgery compared with medical therapy. Importantly, patients with no significant lesions at angiography benefit from a reorientation of their management. Noncardiac causes of chest pain should be considered (including pulmonary embolism), as well as “syndrome X” and variant angina. If the coronary arteries are completely angiographically normal, antithrombotic and antiplatelet drugs can often be discontinued and the need for antianginal medication reassessed. Patients who are most likely to have no significant lesions at angiography tend to be women with no ECG changes. Nevertheless, the finding of no significant lesions at angiography is usually unanticipated. Importantly, symptomatic patients without significant obstructive CAD seen with angiography may have more severe atherosclerosis detected by IVUS, caused by eccentric coronary artery remodeling, which preserves the lumen size. Therefore, in selected patients, more invasive testing at the time of coronary arteriography, including use of IVUS or measurement of coronary flow reserve using adenosine as a vasodilating agent, may be indicated to help better establish the cause of the acute chest pain.

Risk Models and Risk Scores

Risk models and scores predict a probability for adverse outcomes based on combinations of clinical, ECG, and laboratory data available at presentation. These tools integrate multiple predictors and thus provide a more comprehensive risk assessment than focus on a single variable. Whereas the TIMI and PURSUIT scores were derived from clinical trial databases, the GRACE score was derived from a large international registry.^30,31,39 Although all three methods discriminate patients at high versus low risk for short- and intermediate-term adverse outcomes, the GRACE model provides better calibration between predicted and observed rates of death.^40,41 The TIMI score does not include any measurement of heart failure. The GRACE model was developed in a less-selected patient population and includes renal insufficiency as a variable, which are two potential advantages over the other methods. The major advantage of the TIMI score is that it is a simple integer sum that can be calculated at the bedside without a calculator (see Fig. 39–6); PURSUIT and GRACE use weighted averages of multiple risk factors and may require a computer to calculate. None of the models incorporates information from biomarkers, such as troponin or BNP/NT-proBNP.

In summary, the short-term outcome of UA/NSTEMI can be predicted by a variety of methods, and the TIMI and GRACE risk scores are the most systematic and widely tested approaches.

Prognosis

Prognosis in patients with UA or NSTEMI depends on the combination of the morbidity or mortality expected from the extent of coronary stenosis and LV dysfunction and the short-term risk associated with the culprit lesion and the unstable state of ACS. Risk is highest early after the onset of symptoms. Published reports concerning UA/NSTEMI are influenced by patient selection and treatment and can be misleading. The inclusion and exclusion criteria for clinical trials introduce bias by sometimes eliminating low- or high-risk patients. Comparisons of incidence and outcome of UA/NSTEMI since the adoption of routine troponin testing are problematic because the proportion of patients classified as having NSTEMI has increased, and the overall mortality of NSTEMI may have fallen as a consequence.⁴² Data from the GRACE study show a 6-month mortality of 6.2% in patients with NSTEMI and 3.6% in those with UA. Rehospitalization rates over the 6-month period were approximately 20%, and revascularization rates were approximately 15%.⁴³

Treatment Overview

Please see Table 39–5 for recommendations for acute treatments (ie, early hospital care) from the 2014 AHA/ACC Guideline for the Management of Patients with Non-ST-Elevation Acute Coronary Syndromes. The aims of therapy for UA/NSTEMI patients are to relieve ischemia, control symptoms, and prevent complications; such complications may include recurrent episodes of myocardial ischemia or myonecrosis, heart failure, and death.

β-Blockers, nitrates, and (to a lesser extent) calcium channel blockers (CCBs) reduce the risk of recurrent ischemia. Revascularization eliminates ischemia in many patients. The risk of progression to MI, or recurrent MI, is diminished by antiplatelet and antithrombotic drugs and by invasive treatment of the culprit lesion. Aggressive statin therapy plays an increasingly important role after ACS.

Hospitalized patients with NSTE-ACS should be treated with DAPT (aspirin and a P2Y₁₂ inhibitor), antithrombin therapy, a β-blocker, and a statin. In selected high-risk individuals, a glycoprotein (GP) IIb/IIIa inhibitor may be indicated. Furthermore, critical decisions are required regarding the angiographic strategy. One option, commonly termed the early invasive strategy, incorporates an angiographic approach in which patients undergo coronary angiography within 24 to 48 hours and revascularization is performed if suitable coronary anatomy is identified. The alternative approach, the early conservative strategy, is guided by myocardial ischemia, with angiography reserved for patients with recurrent ischemia at rest or findings on a predischarge noninvasive evaluation for ischemia that do not support low-risk status. This strategy is better described as selective invasive because it requires aggressive medical intervention and risk stratification. Regardless of the angiographic strategy used, an assessment of LV function should be strongly considered because it carries prognostic information, and it is imperative to treat patients who have impaired LV systolic function with both angiotensin-converting enzyme (ACE) inhibitors and β-blockers unless contraindicated. It is also important to consider a more aggressive approach to revascularization, including CABG surgery in appropriate candidates.

Use of Anti-Ischemic Drugs

β-Adrenergic Blockers

β-Blockers reduce heart rate, contractility, and blood pressure, thereby reducing myocardial oxygen demand (MVO₂). Slowing of the heart rate with β-blockers also increases the duration of diastole, which improves coronary and collateral blood flow. These effects are mediated by competitive antagonism of β₁-receptors in cardiomyocytes. In contrast, antagonism of β₂-receptors in the peripheral circulation and the lungs may cause vasoconstriction and bronchoconstriction.

It is recommended that β-blockers be initiated orally within the first 24 hours after it has been determined that the patient is not in heart failure or at significant risk for cardiogenic shock and does not otherwise have contraindications, such as significant sinus bradycardia (heart rate < 50 bpm), hypotension (systolic blood pressure < 90 mm Hg), or evidence of significant heart failure (rales > lung bases). If there are concerns about possible intolerance to β-blockers, initial selection should favor a short-acting β₁-specific drug, such as metoprolol or esmolol if a very short-acting agent is needed. Mild wheezing or a history of asthma or chronic obstructive pulmonary disease does not preclude use of a β-blocker, but initiating therapy with a low dose of a short-acting cardioselective agent is prudent. IV β-blockers should be reserved for patients with a clear indication for parenteral therapy and an absence of risk factors for cardiogenic shock or other contraindications.

Nitrates

NTG dilates vascular smooth muscle cells in an endothelium-independent manner and has venous, peripheral, and coronary vascular effects. Nitrates improve oxygen supply by dilating coronary and collateral blood vessels and reduce MVO₂ by dilating the venous bed and reducing preload.

Patients whose symptoms are not relieved with sublingual NTG may experience symptom relief from IV NTG, and such therapy is recommended in the absence of contraindications for patients with ongoing ischemia or heart failure. Nitrates are absolutely contraindicated with concomitant use of phosphodiesterase-5 inhibitors (eg, sildenafil, tadalafil, or vardenafil) because concomitant therapy may lead to severe hypotension.

Clinical trial evidence does not support benefit of nitrates on “hard” adverse outcomes among patients with suspected ACS beyond the first 48 hours of treatment.⁴⁴ Thus, nitrates should not be administered routinely or for extended periods in the absence of ongoing or recurrent chest pain or heart failure symptoms. However, nitrates represent excellent treatment options for patients with acute or chronic symptomatic ischemia as well as among patients in whom the ACS presentation is associated with hypertension or pulmonary congestion from heart failure.

Morphine

Morphine effectively relieves chest pain and symptoms of anxiety in patients with ACS. It has venodilating properties that may be particularly helpful in relieving symptoms related to pulmonary congestion when heart failure complicates ACS. Morphine may be reasonable for patients with persistent or recurrent symptoms after initiation NTG or other anti-ischemic agents, but reduction in discomfort after morphine administration should not be necessarily held as improvement in ischemia. Morphine should be administered in IV boluses of 1 to 5 mg, which are repeated as needed to relieve symptoms provided careful blood pressure monitoring is performed. Recent studies have suggested a delayed onset of action of P2Y₁₂ inhibitors when the loading dose is administered soon after IV morphine; this is presumably a result of a delay in drug absorption from the gut.

Calcium Channel Antagonists

These agents inhibit vascular smooth muscle cell function and have effects on sinus node and atrioventricular (AV) nodal function. All of these agents are coronary and peripheral vasodilators, but members of the drug class differ notably with regard to their effects on cardiac contractility and electrophysiologic effects. Verapamil and diltiazem have broad effects, slowing heart rate and AV nodal conduction, reducing contractility, and causing peripheral vasodilation. In contrast, the dihydropyridines (eg, amlodipine and nifedipine) have powerful vasodilatory effect but show minimal or no direct effects on sinus or AV nodal function.

Rate-limiting calcium antagonists should be considered second- or third-line agents to treat persistent or recurring ischemia in patients on full doses of β-blockers and nitrates or in whom these agents are contraindicated or poorly tolerated. Several randomized trials assessing the use of diltiazem or verapamil in patients with UA/NSTEMI suggest that these agents relieve or prevent symptoms and ischemia, but should be avoided in patients with heart failure or known LV dysfunction, because they have been shown to increase mortality in this setting.¹

Dihydropyridine CCBs do not play a primary role in the management of ACS but may be useful agents for blood pressure control when persistent hypertension is present after initiation of β-blockers and ACE inhibitors or angiotensin receptor blockers (ARBs). Only long-acting dihydropyridines, such as amlodipine, should be used for this purpose; short-acting preparations of nifedipine must be avoided in ACS because they cause adverse outcomes.

Antagonists of the Renin-Angiotensin-Aldosterone System

Although not specifically anti-ischemic agents, ACE inhibitors antagonize adverse neurohormonal pathways that contribute to adverse ventricular remodeling and heart failure after ACS. In patients with acute MI, particularly when complicated by LV systolic dysfunction, ACE inhibitors show powerful long-term mortality benefits.⁴⁵ Although data focusing specifically on populations with UA/NSTEMI are sparse, ACE inhibitors have also demonstrated risk reduction among high-risk subsets with chronic CAD, including those with normal LV function.⁴⁶ Accordingly, ACE inhibitors should be used in patients with UA/NSTEMI complicated by heart failure or LV dysfunction, as well as in those with hypertension (see Chap. 25) or diabetes (see Chap. 28). ARBs are effective alternatives to ACE inhibitors in patients with LV dysfunction or heart failure after acute MI, but should not routinely be used in combination with ACE inhibitors.⁴⁷ Because of the larger evidence base for ACE inhibitors, ARBs should predominantly be used in place of ACE inhibitors in patients with ACE inhibitor intolerance.

Aldosterone antagonists such as spironolactone or eplerenone should be considered for high-risk UA/NSTEMI patients with an LV ejection fraction ≤ 0.40 and either symptomatic heart failure or diabetes mellitus, provided they are receiving adequate doses of ACE inhibitors and do not have significant renal dysfunction or hyperkalemia.¹

An algorithm for clinical management from the 2014 Update of the AHA/ACC Guidelines for Management of UA/NSTEMI is shown in Fig. 39–6.¹ The decision between early invasive and ischemia-guided strategies has been moved to a very proximal time point in the management algorithm of NSTE-ACS.¹ Although initial treatments for these strategies are very similar, patients who undergo an early invasive approach are separated into immediate (< 2 hours), early (< 24 hours), and delayed (1-3 days) groups according to the urgency of coronary angiography. Patients needing immediate angiography are those with refractory angina, sustained ventricular tachyarrhythmias, worsening heart failure or shock, or new or worsening mitral regurgitation. Patients with a low-risk score can be initially managed with an ischemia-guided strategy covered below. Early, but not immediate, angiography is performed for patients at increased risk by the TIMI or GRACE risk score, whereas a delayed invasive approach can be considered for patients at intermediate risk.

In prior studies, two broad strategies were considered with regard to the early use of coronary angiography and revascularization after NSTE-ACS. In the early invasive strategy, patients are referred for coronary angiography, typically within 24 to 48 hours of presentation, with coronary revascularization (either PCI or CABG) performed based on anatomic findings identified during angiography.

In the early conservative strategy, which is more appropriately termed a selective invasive strategy, or now called an ischemia-guided strategy, angiography is deferred and reserved for patients with spontaneous recurrent ischemia, development of heart failure, or evidence of significant ischemia or LV dysfunction detected on a predischarge noninvasive evaluation. As discussed later, both strategies should incorporate aggressive antiplatelet and antithrombotic strategies, although there are some differences in the preferred options depending on the strategy selected. Proponents of the early invasive strategy argue that it allows earlier and more definitive risk stratification. It identifies the 10% to 15% of patients without obstructive coronary stenoses and the approximately 20% of patients with three-vessel or left main CAD. Proponents of the early conservative strategy argue that the high- and low-risk patients can also be identified with appropriate noninvasive imaging.

Several, now older, large, randomized controlled trials compared a routine early invasive strategy with selective invasive strategies in patients with NSTE-ACS and produced conflicting results.^{48,49,50,51,52,53} Comparisons among the trials were difficult because the studies included heterogeneous patient populations, and the comparisons could not account for the rapid advances in PCI technology and medical therapies over time. Nonetheless, meta-analysis of these studies showed that applying a routine early invasive approach was associated with a trend toward an early hazard (more periprocedural ischemic and bleeding events) without a reduction in long-term mortality. In contrast, when considering patients at high risk for adverse events, an early invasive approach reduced episodes of refractory angina and MI during early and late follow-up.

As an example, in the TACTICS (Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Strategy)–TIMI 18 trial, the benefit of the early invasive approach was confined to patients at intermediate or high risk, as defined using troponin elevation, ECG changes, or the TIMI risk score.⁵¹ In the meta-analysis by Hoenig et al,⁵⁴ five studies including nearly 8000 patients were assessed and revealed an approximately 25% reduction in early and late MIs and a halving of refractory angina events. Long-term mortality was similar between the strategies, although it is noted that a substantial proportion of patients initially assigned to a conservative approach ultimately undergo an invasive procedure.

Moreover, an early invasive strategy appears to be cost effective.⁵⁵ Thus, in high-risk patients, a routine invasive strategy is generally preferred if there are no contraindications to coronary angiography and if the patient is a good candidate for prolonged DAPT.¹ Among high-risk patients, the benefits of an invasive approach are well established; among low-risk patients, no benefit exists for a routine early invasive approach (see Fig. 39–7).

Timing of Invasive Therapy

Beyond favoring an early invasive strategy for many intermediate-to-high-risk ACS patients, there has been a steady move to have patients undergo coronary angiography earlier. The rationale for earlier intervention includes a desire to arrive at a definitive treatment plan and shorten hospital stays as much as possible, as well as the recognition that improvements in interventional technique largely driven by coronary stenting, radial artery access, and adjunctive antiplatelet and antithrombin therapy have increased the safety of catheter interventions.

However, in contrast to STEMI, acceleration of the timing of invasive therapy does not appear to markedly improve outcomes for most patients with UA/NSTEMI. In the Timing of Intervention in Acute Coronary Syndrome (TIMACS) trial,⁵⁶ 3031 patients with UA/NSTEMI were randomized to routine angiography performed early (within 24 hours of randomization) or delayed (≥ 36 hours from randomization). The median times to angiography in the two groups were 14 and 50 hours, respectively. At 6 months, the primary outcome of death, MI, or stroke occurred in 9.6% of patients in the early intervention arm and 11.3% of patients in the delayed intervention arm (hazard ratio [HR], 0.85; 95% confidence interval [CI], 0.68-1.06; P = .15). No differences in bleeding or other safety end points were found between the two arms. Refractory ischemia was significantly reduced at 30 and 180 days in the early arm, but other individual secondary end points were not different between the groups. In a prespecified subgroup in the highest tertile of GRACE risk score (> 140), the primary end point was reduced by 35% in the early intervention arm (P = .006); in contrast, no difference was observed among patients at intermediate or low risk (see Fig. 39–7).⁵⁶

These findings, considered together with other trials comparing timing of PCI after UA/NSTEMI,⁵⁷ suggest that for most low- and intermediate-risk patients with UA/NSTEMI in whom an invasive approach is planned, outcomes are similar when PCI is performed earlier versus later in the hospital stay; as such, there is no indication for urgent (ie, immediate) intervention, and it is appropriate to consider logistical issues when determining the timing of coronary angiography and PCI. On the other hand, because no hazard has been observed with earlier intervention, catheterization and PCI within 24 hours (or earlier) may prove efficient in many circumstances and would be generally preferred in the highest risk patients if logistically feasible in that health care system. A meta-analysis including four trials, over 4000 patients, and with extended follow-up yielded similar findings—a lower occurrence of recurrent ischemic events and shorter hospital stays among patients undergoing an earlier invasive approach.⁵⁸

Noninvasive Test Selection in Patients Managed Conservatively

A detailed discussion of noninvasive stress testing in patients with CAD is presented in the ACC/AHA Guidelines for Exercise Testing, ACC/AHA Guidelines for the Clinical Use of Cardiac Radionuclide Imaging, and ACC/AHA Guidelines for the Clinical Application of Echocardiography. Likewise, Chaps. 13, 14, 15, 16, 17, 18, 19 cover all contemporary modalities for noninvasive cardiac testing. Coronary angiography is usually indicated in patients with UA/NSTEMI who either have recurrent symptoms of ischemia despite adequate medical therapy or who are at high risk as categorized by clinical signs (congestive heart failure, malignant ventricular arrhythmias) or noninvasive test findings (significant LV dysfunction: ejection fraction < 0.35, large anterior or multiple perfusion defects) (see Table 39–4 for low-, intermediate-, and high-risk predictors from noninvasive testing).

Coronary revascularization (PCI or CABG) is performed to improve prognosis, relieve symptoms, prevent ischemic complications, and improve functional capacity. The decision to proceed from diagnostic angiography to revascularization is influenced not only by the coronary anatomy but also by a number of additional factors, including anticipated life expectancy, LV function, comorbidity, functional capacity, severity of symptoms, quantity of viable myocardium at risk, and the individual patient's preferences. Careful clinical judgment is required when deciding which lesions identified at coronary angiography merit revascularization.

Data from both retrospective observations and randomized clinical trials indicate that PCI success rates are now very high in most patients with UA/NSTEMI. Procedural safety appears to be enhanced by the use of a radial approach to vascular access, particularly among high-risk ACS patients. Patients with multivessel CAD can now often undergo complete revascularization with the use of multiple DESs. Although it is not clear that PCI of nonculprit stenoses at the time of culprit lesion intervention improves outcomes, increasing procedural safety and improving revascularization durability may be affording more single-setting complete revascularization procedures. Data to guide selection between PCI and CABG in patients with ACS and multivessel CAD continue to give some preference for surgical revascularization, especially among patients with more extensive disease and diabetes. In general, the indications for CABG in UA/NSTEMI are similar to those for stable angina.¹ High-risk patients with diabetes mellitus or LV dysfunction and multivessel CAD and those with two-vessel disease with severe proximal involvement of the left anterior descending artery or severe three-vessel or left main disease should be considered for CABG. However, some of these patients may be candidates for multivessel PCI.⁵⁹ Moreover, although the Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery (SYNTAX) study (Chap. 42) did not focus specifically on patients with ACS, the demonstration of exaggerated benefits of CABG over multivessel PCI among individuals with more diffuse and complex CAD likely applies to the ACS population as well.⁶⁰ Thus, selection between PCI and CABG in patients with multivessel CAD without diabetes is based on patient preference, lesion morphology, and the likelihood of a complete revascularization result with PCI.⁵⁹

Antiplatelet and antithrombotic agents (see also Chap. 41) are cornerstone therapies to help passivate the active vascular disease process and prevent thrombotic complications, including death and recurrent myocardial ischemia. Currently, a combination of ASA, a P2Y₁₂ antagonist (clopidogrel, prasugrel, or ticagrelor), antithrombin therapy (unfractionated heparin [UFH], low-molecular-weight heparin [LMWH], fondaparinux, or bivalirudin), and in certain instances, a platelet GP IIb/IIIa receptor antagonist represents the leading effective options. The intensity and duration of treatment are tailored to individual risk, though nearly all patients should receive DAPT and an antithrombin initially.¹

Aspirin

ASA irreversibly acetylates and inhibits cyclooxygenase-1 within platelets, preventing the formation of thromboxane A₂, which is a potent platelet activator and vasoconstrictor. In patients with UA/NSTEMI, ASA reduces the probability of death or MI by 25% to 30%.⁶¹

ASA should be initiated in patients with UA/NSTEMI at a dose of 160 or 325 mg, with the first dose chewed to rapidly establish a high blood level.¹ Thereafter, daily doses of 75 to 325 mg are prescribed and continued indefinitely. Data generally support a lower dose of maintenance aspirin to minimize bleeding, especially when combined with a P2Y₁₂ inhibitor. Debate remains regarding the optimal ASA dose after coronary artery stenting; although many interventional cardiologists prefer a 325-mg dose for 1 month after stent implantation. One arm of the Clopidogrel Optimal Loading Dose Usage to Reduce Recurrent Events-Organization to Assess Strategies in Ischemic Syndromes (CURRENT)/Organization for the Assessment of Strategies for Ischemic Syndromes (OASIS) 7 trial compared high-dose (300-325 mg) versus low-dose (75-100 mg) ASA in approximately 25,000 patients with ACS (70% with UA/NSTEMI; 70% with PCI). No difference in early efficacy or safety was seen between ASA doses, including in the subgroup undergoing PCI.⁶² In the Study of Platelet Inhibition and Patient Outcomes (PLATO) trial, post hoc analysis of aspirin dose on outcome was performed to understand the striking difference in efficacy of ticagrelor between countries outside North America versus inside North America. A strong association between ticagrelor superiority (over clopidogrel) was found when patients received low-dose aspirin. For this reason, the NSTEMI guidelines specifically recommend low-dose aspirin be used in combination with ticagrelor when DAPT is prescribed.⁶³ The 2015 ESC NSTEMI guidelines recommend low-dose aspirin for maintenance therapy regardless.

Oral P2Y₁₂ Inhibitors

Clopidogrel

Clopidogrel is a thienopyridine derivative that blocks binding of adenosine diphosphate (ADP) to the P2Y₁₂ receptor on the platelet surface, inhibiting ADP–mediated platelet activation and aggregation by approximately 50% to 60%. Because ASA and P2Y₁₂ antagonists inhibit platelet function via different pathways, combined use of the two drugs represents an attractive antiplatelet strategy. The Clopidogrel in Unstable Angina to Prevent Recurrent Ischemic Events (CURE) trial randomized 12,562 patients with UA/NSTEMI to ASA alone or to the combination of ASA and clopidogrel for 3 to 12 months (average, 9 months).⁶⁴ The composite end point of CV death, MI, or stroke occurred in 11.5% of patients assigned to placebo versus 9.3% assigned to clopidogrel (relative risk, 0.80; P < .001). Importantly, this benefit was similar whether patients were managed with medical therapy or with coronary revascularization.⁶⁵ Major bleeding was increased in the clopidogrel arm (3.7% vs 2.7%; P = .003). Bleeding was notably increased in patients who underwent CABG surgery within the first 5 days of stopping clopidogrel.⁶⁴

In a subgroup analysis of CURE consisting of the 2658 patients who underwent PCI after UA/NSTEMI, CV death or MI was reduced by 31% in the clopidogrel plus ASA group versus the ASA alone group (P = .002).⁶⁶ In the Clopidogrel for the Reduction of Events During Observation (CREDO) trial,⁶⁷ 2116 patients scheduled to undergo PCI were randomized to clopidogrel or placebo in addition to ASA. A 27% reduction in the composite outcome of CV death, MI, or stroke was reported in patients treated with combination clopidogrel and ASA for 1 year versus those treated for 1 month (8.5% vs 11.5%; P = .02). As with CURE, there was an increased risk of major bleeding in the combination-therapy group (8.8% vs 6.7%; P = .07), but this excess was largely restricted to patients who underwent CABG. Thus, elective CABG (and other major surgery) should be delayed for at least 5 days after the last dose of clopidogrel. In some practices, therefore, physicians have preferred to withhold clopidogrel (as well as other thienopyridines) until the coronary anatomy is known, to be certain that CABG will not be delayed if needed.

It is important to recognize that the PCI-CURE and CREDO trials were performed before newer DESs were available. DESs inhibit vascular smooth muscle cell proliferation and prevent in-stent restenosis; however, they also delay regrowth of the protective vascular endothelium, which results in a longer time period during which patients are at risk for stent thrombosis.⁶⁸ Because stent thrombosis has been observed up to and even beyond 1 year following placement of a first-generation DES, current guideline recommend at least 1 year of DAPT after placement of a DES.⁶⁹ Patients who, for any reason, are poor candidates for long-term clopidogrel, including those who are planned for major surgery in the near future (in the next few months after the PCI procedure), should not receive a first-generation DES. Some data suggest lower rates of major adverse CV events in those treated with prolonged DAPT compared to treatment for 1 year.⁷⁰ Several meta-analyses have been performed comparing different durations of DAPT following DES. Most include approximately 30,000 patients and report lower bleeding rates and lower overall mortality with shorter-duration DAPT (6-12 months), but with lower rates of stent thrombosis and MI with prolonged DAPT (> 12 months).^71,72,73,74 The 2014 AHA/ACC NSTEMI guidelines recommend DAPT for up to 12 months irrespective of treatment strategy.

In addition to the ASA dose comparison described earlier, the CURRENT/OASIS 7 trial compared standard-dose (300-mg load; 75 mg/d) clopidogrel versus “double-dose” clopidogrel (600-mg load; 150 mg/d × 7 days; then 75 mg/d). In the overall study population, no difference was observed in the primary end point of CV death, MI, or stroke at 30 days; however, a significant interaction was observed with PCI, such that in the subgroup undergoing PCI, the higher dose strategy resulted in a 15% reduction in the primary end point and a 29% reduction in stent thrombosis. In contrast, among patients not undergoing PCI, there was a nonsignificant trend toward worse outcomes in the higher dose clopidogrel arm. Bleeding rates were increased significantly in the high-dose clopidogrel arm, both in the overall population as well as the PCI subgroup.⁶² The OASIS 7 results are consistent with multiple studies evaluating platelet function outcomes in supporting use of a 600-mg clopidogrel dose before PCI among ACS patients.

Clopidogrel Resistance

All thienopyridines are prodrugs that require more than one step of metabolism before achieving their antiplatelet effects. Clopidogrel has complex pharmacokinetic and pharmacodynamic properties that result in widely variable levels of the circulating active metabolite. The in vitro response to clopidogrel, as assessed by inhibition of ADP-mediated platelet aggregation, is heterogeneous with a normal (bell-shaped) distribution of response among individuals. After absorption, clopidogrel requires a two-step oxidation by the hepatic cytochrome P450 (CYP) system to generate an active metabolite. Multiple CYP enzymes are involved (CYP3A4, CYP3A5, CYP2C9, CYP1A2, CYP2B6, and CYP2C19), several of which have important genetic variants and are affected by concomitant medications. The clinical relevance of these findings is likely muted by the multiple other factors affecting outcome and the benefit of concomitant medications such as proton pump inhibitors. For example, initial studies suggested an in vitro interference of antiplatelet effects of clopidogrel by omeprazole,⁷⁵ yet clinical trials have not generally detected a conclusive signal to suggest concomitant omeprazole use in clopidogrel-treated patients should be avoided, especially because a lower rate of gastrointestinal bleeding has been noted.⁷⁶ Use of other proton pump inhibitors without such in vitro interaction may be considered⁷⁷ if concerns remain.

Although response variability to clopidogrel has emerged as a potentially important issue in recent years, the clinical implications remain controversial for several reasons. First, many studies have defined “clopidogrel resistance” on the basis of a single measurement of platelet reactivity performed on clopidogrel; however, residual platelet aggregation to ADP is determined only in part by response to clopidogrel because patient-related factors such as diabetes, smoking, and proximity to the ACS event also contribute to platelet aggregation. Second, the limitations of currently available platelet function tests are noteworthy because the available tests correlate poorly with each other and capture neither the full functionality of the platelet nor the response to clopidogrel.

Prasugrel

Prasugrel is a thienopyridine that, similar to clopidogrel, requires conversion to an active metabolite to bind to the P2Y₁₂ receptor. Prasugrel requires only a one-step hepatic metabolism with CYP3A, CYP2B6, CYP2C9, and CYP2C19. This more favorable pharmacokinetic profile translates into better pharmacodynamics with faster onset of action, more potent and irreversible platelet inhibition, and lower interindividual variability as compared to even high doses of clopidogrel. In the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel–Thrombolysis in Myocardial Infarction (TRITON-TIMI) 38 trial,⁷⁸ which enrolled 13,608 patients with STEMI or high-risk UA/NSTEMI who were scheduled to undergo PCI, prasugrel (60-mg load; then 10 mg/d) was compared with clopidogrel (300-mg load; 75 mg/d). Both drugs were initiated at the time of PCI with no pretreatment administered in advance of PCI. The primary efficacy end point—the same as was used in the CURE trial—was a composite of CV death, MI, and stroke and was reduced from 12.1% in the clopidogrel arm to 9.9% in the prasugrel arm over a median duration of 14.5 months of follow-up (HR, 0.81; 95% CI, 0.73-0.90; P < .001) (Table 39–6). This benefit was mediated by a 24% reduction in nonfatal MI without a significant effect seen on mortality or stroke rate. Two-thirds of the MI events were procedural MIs detected based on serial cardiac enzyme measurement rather than clinical recognition of symptoms and signs.

Importantly, definite or probable stent thrombosis (per the Academic Research Consortium definition) was reduced from 2.4% with clopidogrel to 1.1% with prasugrel (P < .001).⁷⁸ Interestingly, most of the stent thromboses were observed within the first 30 days after randomization, regardless of whether DES or bare-metal stents were used, a finding that highlights the relatively high frequency of early stent thrombosis after ACS, in distinction to much lower rates after elective PCI. Interpreting the magnitude of benefit on stent thrombosis and procedural MI is challenging, given that clopidogrel was not administered before catheterization and the dose used was only 300 mg.

Bleeding rates were significantly increased in the prasugrel arm of the trial, including an increase from 1.8% to 2.4% in TIMI major bleeding (HR, 1.32; 95% CI, 1.03-1.68; P = .03) as well as notable increases in life-threatening and fatal bleeding. Patients older than age 75 years or with body weight below 60 kg were at particular risk for bleeding, such that the increase in bleeding negated any efficacy advantage. Those with prior stroke or transient ischemic attack had a prohibitive increase in bleeding, and a net harm was identified; in such patients, prasugrel is contraindicated. Among individuals who underwent CABG after at least one dose of study drug, major bleeding rates were 13.4% in the prasugrel arm and 3.2% in the clopidogrel arm (P < .0001).⁷⁸ Thus, although some institutions have become comfortable performing CABG after clopidogrel administration, such a strategy would not be prudent after using prasugrel given the much higher rates of surgical bleeding anticipated.

Ticagrelor

Ticagrelor is a nonthienopyridine, direct-acting, and reversible oral antagonist of the P2Y₁₂ receptor. This agent does not require conversion to an active metabolite and provides more rapid onset of action and a more potent and predictable antiplatelet response than clopidogrel. Ticagrelor achieves rapid antiplatelet activity after oral loading, with therapeutic activity observed within 30 minutes and near full activity at 2 hours.⁷⁹

In PLATO,⁸⁰ ticagrelor (180-mg loading dose; 90 mg twice daily) was compared with clopidogrel (300- to 600-mg loading dose; 75 mg/d) in 18,624 patients with STEMI or UA/NSTEMI, including patients managed with and without PCI. At the end of the 12-month follow-up period, the primary end point (CV death, MI, and stroke) was reduced from 11.7% in the clopidogrel arm to 9.8% in the ticagrelor arm (HR, 0.84; 95% CI, 0.77-0.92; P < .001) (see Table 39–6). In addition to significant reductions in MI alone, there was also a significant 21% relative risk reduction in vascular mortality and a 22% reduction in total mortality (5.9% vs 4.5%; P < .001). Similar to the CURRENT/OASIS 7 and TRITON-TIMI 38 trials, stent thrombosis was reduced significantly with the more potent oral antiplatelet regimen. Also consistent with prior studies, an increase in non-CABG major bleeding was observed in the ticagrelor arm (4.5% vs 3.8%; P = .03); however, bleeding rates after CABG were lower with ticagrelor, likely because of the shorter half-life and more rapid reversibility of the drug.⁸⁰

In contrast to TRITON-TIMI 38, clopidogrel pretreatment was permitted in PLATO; however, only 27% of patients received 600 mg of clopidogrel within the first 24 hours after their ACS event, again raising questions of what the comparative benefit of this novel compound would be versus a more aggressive clopidogrel strategy.

Several unique side effects have been observed with ticagrelor, which may be adenosine mediated, and contribute to a slightly higher rate of discontinuation of ticagrelor than clopidogrel. These include dyspnea, which tends to occur early after starting the drug in 10% to 15% of treated patients but is not associated with evidence of heart failure and usually lasts less than a week. Ventricular pauses also occurred more commonly in the ticagrelor group early after treatment initiation, but these also tended to decrease in frequency over time, were rarely symptomatic, and were not associated with clinically significant bradycardia. It should be noted that although reversible, this drug still has residual antiplatelet effects for up to 5 days among individuals on chronic therapy.⁷⁹

Clopidogrel, prasugrel, and ticagrelor each receive a Class I recommendation in the current AHA/ACC NSTEMI guidelines, with a preference for ticagrelor over clopidogrel for patients managed with an early invasive or ischemia-guided strategy (Class IIa).¹

Glycoprotein IIb/IIIa Inhibitors

When the platelet is activated, the GP IIb/IIIa receptors on the platelet surface increase in number and demonstrate improved binding affinity for fibrinogen. The binding of fibrinogen to receptors on different platelets results in aggregation. The platelet GP IIb/IIIa receptor antagonists act by occupying the receptors sites, thus opposing fibrinogen and von Willebrand factor binding. The occupancy of 80% or more of the receptor sites and inhibition of platelet aggregation to ADP by 80% or more results in potent antiplatelet effects. The various GP IIb/IIIa antagonists, however, possess significantly different pharmacokinetic and pharmacodynamic properties (see Chap. 41). The role of GP IIb/IIIa inhibitors in the contemporary treatment of patients with ACS continues to undergo a reappraisal, and overall usage has decreased significantly in recent years. The main reasons for the decreased utilization stems from lack of benefit evidenced from pretreatment versus in-lab provisional use and the introduction of rapidly acting, potent oral antiplatelet agents. For these reasons, the current AHA/ACC NSTEMI guidelines give GP IIb/IIIa inhibitors a Class IIb recommendation when used with DAPT in an early invasive approach.

Adjunctive Glycoprotein IIb/IIIa Inhibitor Use During Percutaneous Coronary Intervention

The efficacy of GP IIb/IIIa antagonists during PCI has been documented in numerous trials, with particular benefit seen for patients with NSTEMI. In many of the earlier trials, however, high-dose clopidogrel loading had not occurred. Subsequently, the Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment (ISAR-REACT)-2 study compared abciximab versus placebo immediately before PCI in 2022 patients with UA/NSTEMI who were treated with ASA and 600 mg of clopidogrel. Abciximab reduced the primary end point of death or MI by 25% (P = .03). This benefit was restricted to patients with elevated troponin values.⁸¹

Comparison of Upstream Versus Delayed Glycoprotein IIb/IIIa Inhibitor Use

Two recent trials have compared routine upstream versus delayed selective GP IIb/IIIa inhibitor use in high-risk patients with UA/NSTEMI. In the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) Timing trial, 9207 patients assigned to receive a GP IIb/IIIa inhibitor in the primary ACUITY randomization were randomized subsequently in a 2 × 2 factorial design to routine upstream GP IIb/IIIa initiation or deferred selective therapy started in the catheterization laboratory.⁸² In the upstream arm, 98.3% of patients received GP IIb/III inhibitors versus 55.7% in the delayed arm. At 30 days, the composite ischemia outcome was not significantly different between the two arms, but rates of major bleeding were significantly higher in the upstream GP IIb/IIIa inhibitor arm. A major limitation of this study was that in the early upstream arm, GP IIb/IIIa inhibitors were only started an average of 4 hours before PCI.⁸²

The Early Glycoprotein IIb/IIIa Inhibition in Non–ST-Segment Elevation Acute Coronary Syndrome (EARLY ACS) trial also compared a strategy of early, routine GP IIb/IIIa inhibition (with eptifibatide) versus a strategy of delayed provisional use among 9492 patients with NSTE-ACS who were planned for invasive therapy.⁸³ Early ACS required an infusion of 12 hours or more before PCI in the upstream eptifibatide arm, and patients received an average of 21 hours of therapy before PCI compared with 4 hours in ACUITY Timing. Provisional GP IIb/IIIa inhibitor at the time of PCI was used in only 26% of patients. Compared with the delayed, provisional strategy, upstream eptifibatide did not result in a significant reduction in the primary end point of death, MI, urgent revascularization for ischemia, or a thrombotic complication during PCI at 4 days (9.3% vs 10.0%; odds ratio, 0.92; P = .23). Moreover, no significant difference was observed in the secondary end point of death or MI at 30 days. No benefit was observed in key subgroups, including those who were troponin positive and those not loaded with clopidogrel. In contrast, rates of TIMI major (2.6% vs 1.8%; P = .015) and minor (3.6% vs 1.7%, P <.001) bleeding were significantly increased in the early upstream eptifibatide arm.⁸³

Taken together, these findings suggest that an appropriate strategy is to defer GP IIb/IIIa inhibitor use until the decision to perform PCI has been made and to use them selectively among high-risk patients defined clinically or angiographically. Upstream use should be restricted to patients with ongoing clinical instability and perhaps those with dynamic ST-segment changes, particularly when oral ADP receptor antagonist loading is not performed. Upstream use is no longer recommended simply on the basis of troponin elevation.

Anticoagulants available for parenteral use include UFH, LMWH, a synthetic pentasaccharide, and direct thrombin inhibitors. All receive a Class I recommendation in the treatment guidelines (see also Chap. 41).

UFH is a heterogenous mixture of polysaccharides of various molecular weights, which accelerate the action of circulating antithrombin, a proteolytic enzyme that inactivates factor IIa (thrombin), factor IXa, and factor Xa. UFH prevents thrombus generation but is not active against clot-bound thrombin. LMWHs are relatively more potent inhibitors of factor Xa than of thrombin. Relative to UFH, LMWH has more specific binding, a longer half-life, and more predictable anticoagulation, permitting once- or twice-daily subcutaneous administration. Use of LMWHs does not usually require laboratory monitoring, except during pregnancy and with changing renal function. Direct thrombin inhibitors (DTIs) act by binding directly to the anion binding and catalytic sites of thrombin to produce potent, predictable anticoagulation. Unlike UFH and LMWH, DTIs are active against clot-bound thrombin (see Chap. 41).⁸⁴

Unfractionated Heparin

Several trials have compared UFH with placebo and form the basis of the class I recommendation for use of UFH with ASA for patients with UA/NSTEMI.¹ In a meta-analysis of six trials with end point assessment varying from 2 to 12 weeks, death or MI was reduced by 33% in UFH-treated patients (P = .06).⁸⁵

Pharmacokinetic limitations of UFH translate into poor bioavailability and marked variability in anticoagulant response. UFH requires monitoring with the activated partial thromboplastin time (aPTT). A weight-adjusted regimen is recommended, with an initial bolus of 60 U/kg (maximum, 5000 U) and an initial infusion of 12 U · kg¹ · 1 h¹ (maximum, 1000 U/h). Guideline committees recommend dosage adjustments of the nomograms to correspond to a therapeutic range equivalent to heparin levels of 0.3 to 0.7 U/mL by antifactor Xa determinations, which correlates with aPTT values between 50 and 75 seconds.

Serial platelet counts are necessary to monitor for heparin-induced thrombocytopenia for patients receiving more than 96 hours of therapy or being readministered heparin following a recent prolonged heparin course. Thrombocytopenia from heparin takes two forms. One form involves a mild decrease in platelet counts (rarely < 100,000/μL) that occurs early (1-4 days) after initiation of therapy, reverses quickly after discontinuation of heparin, and is of little clinical consequence because it is not antibody mediated. The more severe form of thrombocytopenia is an immune-mediated thrombocytopenia that typically occurs more than 4 days from therapy, although it may occur earlier in patients who have received heparin within the previous several months. Heparin immune thrombocytopenia is caused by a heparin-dependent, antiplatelet antibody that can activate normal platelets. This syndrome can be associated with thrombosis that can produce severe morbidity and mortality. It should be treated with a pentasaccharide or direct thrombin inhibitor initially, especially if thrombosis is present, and then with an oral anticoagulant after hospital discharge.

Low-Molecular-Weight Heparin

Several large, randomized trials have directly compared LMWH with UFH (see Chap. 40). Important differences have emerged depending on the approach to revascularization used in the studies.

Among the earlier trials, those studying enoxaparin versus UFH for medical therapy showed a consistent reduction in ischemic events. The Efficacy and Safety of Subcutaneous Enoxaparin in Non–Q-Wave Coronary Events (ESSENCE)⁸⁶ and TIMI 11B⁸⁷ trials found approximately a 15% to 20% reduction in death or MI at 4- to 6-week follow-up. Both trials reported a small excess in bleeding in the enoxaparin arms.

In contrast to a conservative-approach trial, the Superior Yield of the New Strategy of Enoxaparin, Revascularization, and Glycoprotein IIb/IIIa Inhibitors (SYNERGY) trial (n = 10,027) was an open-label superiority trial comparing enoxaparin with UFH among patients with a planned early-invasive strategy.⁸⁸ GP IIb/IIIa inhibitors were used in more than 50% of patients. The primary end point of death or nonfatal MI occurred in 14.0% of patients randomized to enoxaparin versus in 14.5% of patients randomized to UFH (HR, 0.96; 95% CI, 0.86-1.06). Rates of TIMI major bleeding were increased among patients receiving enoxaparin (9.1% vs 7.6%).

Fondaparinux

Fondaparinux is a synthetic pentasaccharide that can bind antithrombin and inactivate factor Xa. It has no effect on thrombin and, thus, is a pure factor Xa inhibitor. It offers predictable pharmacokinetics and a half-life of 15 hours, allowing once-daily subcutaneous dosing without a need for routine laboratory monitoring. In the OASIS 5 trial,⁸⁹ which enrolled more than 20,000 patients with UA/NSTEMI, 2.5 mg/d of fondaparinux versus 1 mg/kg twice a day of enoxaparin yielded similar rates of the primary end point of death, MI, or refractory ischemia at 9 days (5.8% vs 5.7%). However, major bleeding events were reduced by 48% with fondaparinux, and mortality trended lower in the fondaparinux group at 30 days (2.9% vs 3.5%; P = .02) and at 180 days (5.8% vs 6.5%; P = .05); all but three of the 64 excess deaths with enoxaparin were associated with major or minor bleeding.⁸⁹

Several caveats to the OASIS 5 trial merit mention. First, patients were allowed entry with a creatinine up to 3.0 mg/dL. Major bleeding risk was particularly high among patients treated with weight-adjusted enoxaparin who had a creatinine clearance below 30 mL/min (9.9% rate of major bleeding). Second, UFH was administered after randomization in a larger proportion of enoxaparin-treated patients, a practice that is thought to increase bleeding risks. Third, the long half-life of fondaparinux may create logistical problems in centers that perform early cardiac catheterization. Finally, among patients undergoing cardiac catheterization, an excess in catheter-related thrombotic complications was observed, a finding that has also been observed in other trials using fondaparinux.

Although fondaparinux receives a Class I recommendation in the latest AHA/ACC NSTEMI guideline, other anticoagulants represent better first-line options when an invasive approach is planned, given the long half-life of the agent, lack of reversibility, and the need to administer additional UFH at the time of PCI to prevent catheter-associated thrombus in patients who have received fondaparinux. In contrast, fondaparinux represents a more attractive alternative in conservatively managed patients and is a preferred agent among conservatively managed patients at increased risk for bleeding.

Direct Thrombin Inhibitors

Several DTIs are commercially available (argatroban, lepirudin, and bivalirudin), although bivalirudin has been the most studied and widely used. Bivalirudin is a semisynthetic direct-acting antithrombin with a 25-minute half-life. In the Randomized Evaluation in PCI Linking Angiomax to Reduced Clinical Events (REPLACE) II trial,⁹⁰ bivalirudin alone was noninferior to the combination of UFH plus GP IIb/IIIa inhibitor with regard to death, MI, or urgent revascularization at 30 days (odds ratio, 1.09; 95% CI, 0.90-1.32). Major bleeding was significantly lower in the bivalirudin arm (2.4% vs 4.1%; P < .001).

The ACUITY trial was a complex trial performed in 13,819 patients with UA/NSTEMI, all of whom underwent cardiac catheterization within 72 hours.⁹¹ Patients were randomized to one of three arms: UFH or enoxaparin plus routine GP IIb/IIIa inhibition, bivalirudin plus routine GP IIb/IIIa inhibition, or bivalirudin alone (with provisional GP IIb/IIIa use only). Results from this trial largely support the findings from REPLACE II, demonstrating significantly lower bleeding rates without a significant increase in the primary ischemic end point (death, MI, or unplanned revascularization for ischemia) in patients treated with bivalirudin alone compared with UFH (or enoxaparin) plus a GP IIb/IIIa inhibitor.⁹¹ Importantly, however, in the GP IIb/IIIa inhibitor arms, no safety or efficacy benefits of bivalirudin were noted over UFH or enoxaparin; thus, bivalirudin is not indicated when GP IIb/IIIa inhibitor use is planned. Because the median duration of pretreatment prior to PCI in ACUITY was only 4 hours, this study does not provide support for use of this agent outside the catheterization laboratory. Moreover, there was a treatment interaction based on whether patients are pretreated with clopidogrel before PCI. Among patients who were pretreated with a thienopyridine, ischemic outcomes were similar between the bivalirudin alone and heparin plus GP IIb/IIIa inhibitor arms. In contrast, among patients not pretreated with a thienopyridine, ischemic outcomes were worse in the bivalirudin arm. In aggregate, these studies support bivalirudin as a safe alternative to UFH or enoxaparin during PCI, provided patients have been pretreated with a P2Y₁₂ inhibitor or a postprocedure infusion of bivalirudin until the P2Y₁₂ effect is present. A meta-analysis including 13 trials with 24,605 ACS patients randomized to bivalirudin versus UFH with or without GP IIb/IIIa inhibitors confirmed a marked reduction in major bleeding with bivalirudin (odds ratio, 0.52; 95% CI, 0.45-0.60; P < .001) versus UFH with GP IIb/IIIa inhibitors. When GP IIb/IIIa inhibitors were provisionally used with UFH, the bleeding advantage of bivalirudin was attenuated (odds ratio, 0.66; 95% CI, 0.33-1.32). The rate of subacute (< 24 hours) stent thrombosis was several-fold higher with bivalirudin, although the rate of stent thrombosis at 30 days was not different between the bivalirudin and heparin groups.⁹²

Long-Term Anticoagulation

Several studies have evaluated the combination of warfarin and ASA after ACS. In initial trials, low fixed-dose warfarin was not superior to ASA monotherapy, and the combination was associated with excess bleeding risk. More recently, several studies have shown that the combination of ASA and warfarin may be effective in preventing ischemic events after ACS when the international normalized ratio is maintained at a higher level consistently and is carefully monitored.^93,94 However, these findings are of questionable significance in light of the results of more recent antiplatelet trials, which demonstrated similar or greater benefit with a simpler regimens of ASA and ADP receptor antagonists. Warfarin should be prescribed for UA/NSTEMI patients who also have other well-proven indications for warfarin, such as atrial fibrillation and mechanical prosthetic heart valves. In such patients, combination with low-dose ASA is reasonable if the bleeding risk is not markedly increased.

One issue of particular concern is the extended-duration combination of ASA, clopidogrel, and warfarin. Evidence suggests that “triple therapy” is associated with substantially increased risks for bleeding. In one recent retrospective study of more than 40,000 patients discharged from Danish hospitals with ACS, admission for bleeding occurred in over 12% of patients in a 15-month period, a rate more than four-fold higher than among patients treated with ASA alone.⁹⁵ Bleeding rates were also markedly increased in patients treated with only clopidogrel and warfarin. It may be expected that risks will be even higher with combinations that include the newer and more potent antiplatelet agents, such as prasugrel and ticagrelor. As such, it is recommended that attempts be made to minimize the duration of triple therapy. For patients who require triple therapy, the ASA dose should be reduced to 81 mg, the international normalized ratio should be maintained at the lower end of the therapeutic range, and gastrointestinal prophylaxis (eg, pantoprazole) should be considered. When the indication for warfarin is atrial fibrillation, a careful reevaluation of the risks of stroke (using a tool such as the CHADS₂ score) and bleeding (HAS-BLED score) should be performed; in general, the threshold at which warfarin is initiated or continued should be higher among patients receiving ASA and clopidogrel. It should be noted that, although it is not as effective as warfarin for stroke prevention, ASA combined with clopidogrel is superior to ASA alone.⁹⁶ In patients with atrial fibrillation receiving bare-metal stents, it may be reasonable to treat with ASA and clopidogrel for 1 month if the CHADS₂ score is low, at which time the clopidogrel or aspirin can be stopped and warfarin reinitiated.⁹⁷ Until data are available, prasugrel should be avoided with warfarin except in special circumstances, such as stent thrombosis with clopidogrel.

Several large ongoing trials with over 10,000 patients are assessing whether aspirin can be discontinued among patients with an indication for DAPT and anticoagulant therapy. In some trials, novel oral anticoagulants are replacing warfarin, whereas in others, more potent P2Y₁₂ inhibitors are replacing clopidogrel. These trials follow the intriguing results from the What Is the Optimal Antiplatelet and Anticoagulant Therapy in Patients With Oral Anticoagulation and Coronary Stenting (WOEST) study.⁹⁸ In WOEST, 573 patients with a long-term indication for oral anticoagulant therapy (eg, chronic atrial fibrillation) and undergoing PCI with stent placement were randomized to triple therapy (DAPT plus warfarin) versus double therapy (clopidogrel with warfarin). At 1-year follow-up, bleeding events were substantially less frequent in the double therapy group (odds ratio, 0.36; 95% CI, 025-0.50; P < .0001). More impressive, perhaps, is that the 1-year occurrence of ischemic events was also lower in the group without aspirin.

Timing and Intensity of Lipid Lowering

A series of large randomized controlled trials has evaluated the role of statin therapy in patients with ACS. In the Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) study, 3086 patients were randomized to treatment with 80 mg/d of atorvastatin or placebo 24 to 96 hours after an ACS.⁹⁹ After 16 weeks, the primary end point of death, MI, resuscitated cardiac arrest, or recurrent myocardial ischemia was reduced from 17.4% in the placebo group to 14.8% in the atorvastatin group (P = .048).

The Zocor (Z) phase of the A to Z trial compared an early intensive statin regimen (40 mg followed by 80 mg of simvastatin) with a delayed and less intensive regimen (placebo for 4 months followed by 20 mg of simvastatin) for up to 24 months in 4497 patients with ACS.¹⁰⁰ The primary end point of CV death, MI, readmission for ACS, or stroke occurred in 14.4% of patients in the early intensive statin arm versus in 16.7% of patients in the delayed, less intensive arm (HR, 0.89; 95% CI, 0.76-1.04). CV death was reduced by 25% in the early intensive statin arm (P = .05). During the first 4 months, no difference was evident between treatment groups, but from 4 to 24 months, the primary end point was reduced in the intensive statin group.

In the Pravastatin or Atorvastatin Evaluation and Infection Therapy (PROVE IT)-TIMI 22 trial, an intensive statin regimen of 80 mg of atorvastatin was compared with a standard statin regimen of 40 mg/d of pravastatin among 4162 patients with ACS.¹⁰¹ The atorvastatin arm achieved an average low-density lipoprotein (LDL) cholesterol of 62 mg/dL versus an average of 95 mg/dL in the pravastatin arm. The primary end point of death, MI, UA, or revascularization after 30 days was reduced by 16% in the atorvastatin arm (P < .0001).

Thus, findings are generally consistent with a “lower is better” approach to LDL cholesterol management after ACS. Current clinical practice guidelines eschew targets for LDL concentrations, recommending immediate use of high-dose statin therapy (eg, atorvastatin 80 mg) for patients with ACS.^1,102

The risk of progression to MI or the development of recurrent MI or death is highest during the first few months after the index ACS event. At 1 to 3 months after the acute phase, most patients resume a clinical course similar to that in patients with chronic stable CAD (see Chap. 45).

The broad goals during the hospital discharge phase are two-fold: (1) to prepare the patient for normal activities to the extent possible and (2) to use the acute event as an opportunity to implement long-term care, particularly lifestyle and risk factor modification.

Preparing patients for resumption of normal activities after discharge should include education regarding risk factors for coronary atherosclerosis, importance of medication therapy, and the importance of healthy behaviors including proper diet and exercise habits. Patients should be educated not only about their medications but also which medications should be avoided after discharge (eg, nonsteroidal anti-inflammatory drugs). Discussion regarding referral to CV rehabilitation program is appropriate at this time. Such programs provide an excellent means by which patients may achieve the goals discussed.

There is no better time to implement long-term improvements in care for CV disease than soon after hospital discharge following an ACS. Goals for continued medical therapy after discharge are to ensure that (1) all potential therapies that have been demonstrated to improve long-term outcomes are administered to eligible patients, including ASA (with or without DAPT), β-blockers, statins, and ACE inhibitors (especially for patients with LV ejection fraction < 40%); (2) ischemic symptoms are controlled, using as-needed nitrates, β-blockers, calcium antagonists and, occasionally, ranolazine; and (3) major risk factors such as hypertension, smoking, hyperlipidemia, and diabetes mellitus (see Chaps. 25, 26, 27, 28, 29 and 45) are optimally managed. Monitoring for depression after hospital discharge is recommended and may be achieved in a cardiac rehabilitation environment.

Several patient groups with UA/NSTEMI deserve special commentary.

The Elderly

Older patients with NSTE-ACS should generally be managed similarly to younger patients, but patient-centered decision making is necessary in older patients, and it is important to individualize pharmacotherapy to older patients, given differences in renal function and more medical comorbidities.¹⁰³

Cardiogenic Shock

Although cardiogenic shock is much less common in patients with NSTE-ACS compared to those with STEMI, should cardiogenic shock complicate UA/NSTEMI, management should be similar to those with STEMI, including revascularization, whenever possible.¹⁰⁴

Spontaneous Coronary Artery Dissection

SCAD is an increasingly recognized cause of ACS, particularly in women age 30 to 50 years.¹⁰⁵ The pathophysiology of SCAD is unknown; many cases occur in situations such as uncontrolled hypertension, in the puerperium, in those with fibromuscular dysplasia, or as a complication of disorders of collagen integrity, such as Marfan syndrome. The diagnosis of SCAD should be entertained when younger (usually female) patients present with symptoms and signs of acute coronary ischemia in the absence of other risk factors. Diagnostic studies for those with SCAD should be similar to those without the diagnosis; the presence of a coronary dissection plane may be challenging to recognize at the time of coronary angiography (Fig. 39–7), so a high level of suspicion for presence of the diagnosis should be maintained during such procedures.

Standard treatment for SCAD is not established. For many, conservative management may be quite effective, particularly if flow is present in the coronary artery at the time of diagnostic coronary angiography; spontaneous healing of the dissection may be seen at the time of follow-up angiography.¹⁰⁶ Should revascularization be indicated (ie, occlusion of a coronary vessel is present), both PCI and CABG have been reported as effective choices, although risk for further dissection is present in the context of a fragile vessel, so caution is advised. Given the high prevalence of unrecognized fibromuscular dysplasia in affected patients,¹⁰⁷ some recommend screening for presence of intracranial aneurysm in those affected by SCAD.

Stress Cardiomyopathy

Widely recognized as an ACS “mimic,” stress cardiomyopathy (also known as “apical ballooning” syndrome or takotsubo cardiomyopathy as a result of resemblance of the dysfunctional LV to a Japanese octopus pot trap) was first reported in Japan in 1990. Following more widespread recognition of this entity, incidence and prevalence of stress cardiomyopathy have grown; stress cardiomyopathy is found in 1.7% to 2.2% of patients presenting with ACS, and the vast majority (> 90%) are women.¹⁰⁸

Stress cardiomyopathy most often follows an acute emotional stress, such as the death of a loved one, and may appear indistinguishable from a high-risk ACS, with typical angina, rise and fall of troponin, and onset of heart failure symptoms. Characteristic ECG changes of stress cardiomyopathy include development of symmetric T-wave inversion in most leads; however, such changes are not specific enough to stress cardiomyopathy to avoid diagnostic coronary angiography. At angiography, nonobstructed coronary arteries are the rule, along with characteristic myocardial dysfunction of the LV apex, with compensatory basal hyperkinesis (Fig. 39–8).

Management of patients with stress cardiomyopathy is typically supportive¹⁰⁹; patients most often recover rapidly after presentation, though shock at presentation is not uncommon. Use of vasodilating inotropic agents or intra-aortic balloon counterpulsation in those with shock may actually worsen hemodynamics as such treatment may precipitate LV outflow tract obstruction in the context of basal hyperkinesis. β-Blockers and ACE inhibitors or ARBs are often used during convalescent periods, but their value in stress cardiomyopathy is unknown.

Variant Angina

In 1959, Prinzmetal described a syndrome characterized by angina at rest with transient marked ST-segment elevation.¹¹⁰ The attacks were cyclic in nature, often occurring at rest and in the early morning hours. Ventricular arrhythmias and AV block sometimes occurred at the peak of an attack, and both MI and sudden death were reported as potential consequences. With the use of coronary angiography, it soon became evident that the syndrome was caused by coronary spasm, usually focal and often at the site of a coronary artery stenosis. Coronary spasm might occur in more than one artery in some patients, and the site of spasm may fluctuate from one coronary artery to another.

Variant angina is more common in heavy cigarette smokers, but their age, gender, and risk factor profiles are otherwise similar to those of other coronary patients. Of patients with variant angina, 25% have a history of migraine headaches, and approximately 25% have symptoms of Raynaud phenomenon. Syncope that occurs during rest angina, likely caused by ischemia-induced ventricular arrhythmia or AV block, is a clue to the diagnosis.

Pathophysiologically, it is thought that coronary spasm is a result of abnormalities in endothelial function and nitric oxide activity at sites of coronary spasm.¹¹¹ Severe spasm rapidly induces transmural ischemia, resulting in ST-segment elevation. The risks of serious ventricular arrhythmias and conduction abnormalities (including compete heart block) increase in proportion to the severity and extent of ischemia. Prolonged spasm may lead to MI.

Patients with variant angina can be diagnosed most easily by an ECG recorded during an episode of rest angina, particularly if ST-segment elevation that occurs during an attack disappears promptly after the administration of NTG. Ambulatory ECG monitoring or an event monitor can sometimes be useful to confirm the diagnosis. Exercise testing provokes angina with ST-segment elevation in approximately one-third of patients with variant angina during an active phase of the disease. All patients with variant angina should undergo coronary angiography unless an absolute contraindication is present.

Patients with variant angina are often difficult to treat because attacks are unpredictable and often occur without an obvious precipitating factor. NTG relieves variant angina attacks within minutes and should be used promptly; long-acting nitrates are initially effective in preventing variant angina attacks. CCBs are very effective in preventing attacks of variant angina; higher CCB doses are frequently required. For example, long-acting nifedipine (90 mg/d), diltiazem (360 mg/d), verapamil (480 mg/d), and amlodipine (20 mg/d) are commonly used.¹¹² Statin therapy is also indicated, given the beneficial effects of statins on endothelial function and the common presence of atherosclerosis underlying focal spasm.

Although intuitively logical, the use of revascularization to manage spasm should be reserved for absolutely refractory cases, where documented spasm continues despite optimal medical therapy; results of revascularization procedures in such circumstances are mixed as a result of recurrence of spasm in other coronary territories.

We would like to thank Dr. James A. de Lemos, Dr. Robert A. O’Rourke, and Dr. Robert A. Harrington who substantially contributed to the previous version of this chapter in the 13th edition.