†Guidelines for the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes: The Task Force for the Diagnosis and Treatment of Non-ST-Segment Elevation Acute Coronary Syndromes of the European Society of Cardiology

Preamble

Guidelines and Expert Consensus Documents summarize and evaluate all currently available evidence on a particular issue with the aim to assist physicians in selecting the best management strategies for a typical patient, suffering from a given condition, taking into account the impact on outcome, as well as the risk–benefit ratio of particular diagnostic or therapeutic means. Guidelines are no substitutes for textbooks. The legal implications of medical guidelines have been discussed previously.

A great number of Guidelines and Expert Consensus Documents have been issued in recent years by the European Society of Cardiology (ESC) as well as by other societies and organizations. Because of the impact on clinical practice, quality criteria for development of guidelines have been established in order to make all decisions transparent to the user. The recommendations for formulating and issuing ESC Guidelines and Expert Consensus Documents can be found on the ESC website (http://www.escardio.org/knowledge/guidelines/rules).

In brief, experts in the field are selected and undertake a comprehensive review of the published evidence for management and/or prevention of a given condition. A critical evaluation of diagnostic and therapeutic procedures is performed including assessment of the risk–benefit ratio. Estimates of expected health outcomes for larger societies are included, where data exist. The level of evidence and the strength of recommendation of particular treatment options are weighed and graded according to pre-defined scales, as outlined in Tables 1 and 2.

The experts of the writing panels have provided disclosure statements of all relationships they may have which might be perceived as real or potential sources of conflicts of interest. These disclosure forms are kept on file at the European Heart House, headquarters of the ESC. Any changes in conflict of interest that arise during the writing period must be notified to the ESC. The Task Force report was entirely supported financially by the ESC and was developed without any involvement of the industry.

The ESC Committee for Practice Guidelines (CPG) supervises and coordinates the preparation of new Guidelines and Expert Consensus Documents produced by Task Forces, expert groups, or consensus panels. The Committee is also responsible for the endorsement process of these Guidelines and Expert Consensus Documents or statements. Once the document has been finalized and approved by all the experts involved in the Task Force, it is submitted to outside specialists for review. The document is revised, and finally approved by the CPG and subsequently published.

After publication, dissemination of the message is of paramount importance. Pocket-sized versions and personal digital assistant (PDA)-downloadable versions are useful at the point of care. Some surveys have shown that the intended end-users are sometimes not aware of the existence of guidelines, or simply do not translate them into practice, so this is why implementation programmes for new guidelines form an important component of the dissemination of knowledge. Meetings are organized by the ESC and directed towards its member National Societies and key opinion leaders in Europe. Implementation meetings can also be undertaken at national levels, once the guidelines have been endorsed by the ESC member societies and translated into the national language. Implementation programmes are needed because it has been shown that the outcome of disease may be favourably influenced by the thorough application of clinical recommendations.

Thus, the task of writing Guidelines or Expert Consensus documents covers not only the integration of the most recent research, but also the creation of educational tools and implementation programmes for the recommendations. The loop between clinical research, writing of guidelines, and implementing them into clinical practice can then only be completed if surveys and registries are performed to verify that real-life daily practice is in keeping with what is recommended in the guidelines. Such surveys and registries also make it possible to evaluate the impact of implementation of the guidelines on patient outcomes. Guidelines and recommendations should help physicians to make decisions in their daily practice; however, the ultimate judgement regarding the care of an individual patient must be made by the physician in charge of his/her care.

1. Introduction and definitions

Cardiovascular diseases are presently the leading causes of death in industrialized countries and expected to become so in emerging countries by 2020.¹ Among these, coronary artery disease (CAD) is the most prevalent manifestation and is associated with high mortality and morbidity. The clinical presentations of ischaemic heart disease include silent ischaemia, stable angina pectoris, unstable angina, myocardial infarction (MI), heart failure, and sudden death. Patients with chest pain represent a very large proportion of all acute medical hospitalizations in Europe. Distinguishing those with acute coronary syndrome (ACS) within the very large proportion with suspected cardiac pain represents a diagnostic challenge, especially in those without clear symptoms or electrocardiographic features. In spite of modern treatment, the rates of death, MI, and re-admission of patients with ACS remain high.

It is well established that ACS in their different clinical presentations share a widely common pathophysiological substrate. Pathological, angioscopic, and biological observations have demonstrated that atherosclerotic plaque rupture or erosion, with differing degrees of superimposed thrombosis and distal embolization, resulting in myocardial underperfusion, represent the basic pathophysiological mechanisms in most ACS.

As this is a life-threatening state of atherothrombotic disease, criteria for risk stratification have been developed to allow the clinician to make timely decisions on pharmacological management as well as on coronary revascularization strategies, tailored to the individual patient. The leading symptom that initiates the diagnostic and therapeutic cascade is chest pain, but the classification of patients is based on the electrocardiogram (ECG). Two categories of patients may be encountered:

• Patients with typical acute chest pain and persistent (>20 min) ST-segment elevation. This is termed ST-elevation ACS (STE-ACS) and generally reflects an acute total coronary occlusion. Most of these patients will ultimately develop an ST-elevation MI (STEMI). The therapeutic objective is to achieve rapid, complete, and sustained reperfusion by primary angioplasty or fibrinolytic therapy.²

• Patients with acute chest pain but without persistent ST-segment elevation. They have rather persistent or transient ST-segment depression or T-wave inversion, flat T-waves, pseudo-normalization of T-waves, or no ECG changes at presentation. The initial strategy in these patients is to alleviate ischaemia and symptoms, to monitor the patient with serial ECG, and to repeat measurements of markers of myocardial necrosis. At presentation, the working diagnosis of non-STE-ACS (NSTE-ACS), based on the measurement of troponins, will be further qualified into non-ST elevation MI (NSTEMI) or unstable angina. (Figure 1) In a certain number of patients, CAD will subsequently be excluded as the cause of symptoms. The therapeutic management is guided by the final diagnosis.

The management of patients with STEMI is addressed in the ESC Guidelines for management of ST-elevation acute MI.² The present document deals with the management of patients with suspected NSTE-ACS, replacing the document published in 2000 and last updated in 2002.³ It includes all scientific evidence fully published as peer-reviewed papers in a journal, before 30 April 2007.

The class A level of recommendations of this document is based primarily on randomized, double-blind studies of adequate size using contemporary adjunctive treatment and endpoints that are not subject to observer bias, such as death and MI. These studies were considered to represent the greatest weight of evidence. Studies that were randomized, but not double blind, and/or studies using less robust endpoints (such as refractory ischaemia or need for revascularization) were considered to confer a lower weight of evidence. If only smaller studies were available, meta-analyses were used. However, even the largest controlled trials do not cover all aspects seen in real life. Therefore, some recommendations are derived from subset analyses of larger trials, in the absence of sufficiently powered independent studies. Furthermore, in this rapidly moving field, new studies will constantly challenge the current recommendations.

Costs of health care become an increasing issue in many countries. Although this should not play a role in decision-making, cost consciousness is necessary today. Therefore, we provide the numbers needed to treat (NNT) to prevent an event for the most important treatment options. The NNT seems to be the most transparent approach to compare studies of different size and different endpoints. For example, an NNT of 50 to prevent one death is to be interpreted differently from an NNT of 50 to avoid one rehospitalization.⁴

2. Epidemiology and natural history

The diagnosis of NSTE-ACS is more difficult to establish than STEMI and therefore its prevalence is harder to estimate. In addition, in recent years, a new definition of MI has been introduced to take into account the use of more sensitive and more specific biomarkers of cell death.⁵ In this context, the prevalence of NSTE-ACS, relative to STEMI, has been determined from multiple surveys and registries.^6–15 Overall, data suggest that the annual incidence of NSTE-ACS is higher than that of STEMI. The ratio between NSTE-ACS and STEMI has changed over time, as the rate of NSTE-ACS increased relative to STEMI, without any clear explanation for the reasons behind this evolution.¹⁶ This change in the pattern of NSTE-ACS could actually be linked to changes in the management of disease and greater efforts in prevention of CAD over the last 20 years.^17–20 Overall, from these registries and surveys, it has been shown that the annual incidence of hospital admissions for NSTE-ACS is in the range of 3 per 1000 inhabitants. To date, there are no clear estimates for Europe as a whole, because of the absence of a common centre for centralized health statistics. However, the incidence of the disease is greatly variable among European countries, with a strong west-to-east gradient, the higher incidences and death rates occurring in Central-Eastern Europe.

Overall, the prognosis of NSTE-ACS can be derived from surveys carried out around the world that have included more than 100 000 patients. Data consistently show that the mortality rate at 1 and 6 months is higher in survey populations than in randomized clinical trials. Hospital mortality is higher in patients with STEMI than among those with NSTE-ACS (7 vs. 5%, respectively), but, at 6 months, the mortality rates are very similar in both conditions (12 vs. 13%, respectively).^21,22 Long-term follow-up of those who survive to reach hospital showed that death rates were higher among those with NSTE-ACS than with STE-ACS, with a two-fold difference at 4 years.²³ This difference in mid- and long-term evolution may be due to different patient profiles, since NSTE-ACS patients tend to be older, with more co-morbidities, especially diabetes and renal failure. The difference could also be due to the greater extent of coronary artery and vascular disease or persistent triggering factors such as inflammation.^24,25

The implications for therapy are as follows.

• NSTE-ACS is more frequent than STEMI.

• In contrast to STEMI, where most events occur before or shortly after presentation, in NSTE-ACS these events continue over days and weeks.

• Mortality of STEMI and NSTE-ACS after 6 months is comparable.

This implies that treatment strategies for NSTE-ACS need to address the requirements of the acute phase as well as longer-term treatment.

3. Pathophysiology

Atherosclerosis is a chronic, multifocal immunoinflammatory, fibroproliferative disease of medium-sized and large arteries mainly driven by lipid accumulation.²⁶ CAD involves two distinct processes: a fixed and barely reversible process that causes gradual luminal narrowing slowly over decades (atherosclerosis) and a dynamic and potentially reversible process that punctuates the slow progression in a sudden and unpredictable way, causing rapid complete or partial coronary occlusion (thrombosis or vasospasm, or both). Thus, symptomatic coronary lesions contain a variable mix of chronic atherosclerosis and acute thrombosis. Since the exact nature of the mix is unknown in the individual patient, the term atherothrombosis is frequently used. Generally, atherosclerosis predominates in lesions responsible for chronic stable angina, whereas thrombosis constitutes the critical component of culprit lesions responsible for the ACS.^27,28

ACS represent a life-threatening manifestation of atherosclerosis usually precipitated by acute thrombosis, induced by a ruptured or eroded atherosclerotic plaque, with or without concomitant vasoconstriction, causing a sudden and critical reduction in blood flow. In the complex process of plaque disruption, inflammation was revealed as a key pathophysiological element. In rare cases, ACS may have a non-atherosclerotic aetiology such as arteritis, trauma, dissection, thrombo-embolism, congenital anomalies, cocaine abuse, and complications of cardiac catheterization. Some key pathophysiological elements will be described in more detail because they are important to understand the therapeutic strategies.

3.1 The vulnerable plaque

Atherosclerosis is not a continuous, linear process but rather a disease with alternate phases of stability and instability. Sudden and unpredictable changes in symptoms appear to be related to plaque disruption. The plaques prone to instability and rupture have a large lipid core, a low density of smooth muscle cells, a high concentration of inflammatory cells, and a thin fibrous cap covering the lipid core compared with stable plaques.²⁹ Plaque vulnerability may also depend on circumferential wall stress, on the location and size of the plaque, and on the impact of flow on the luminal plaque surface. In addition to plaque rupture, plaque erosion is another underlying mechanism in ACS. When erosion occurs, the thrombus adheres to the surface of the plaque, whereas in the case of plaque rupture, the thrombus involves the deeper layers down to the lipid core. This may contribute to the growth and rapid progression of the plaque, if the thrombus is not accommodated by positive remodelling.

The fibrous cap usually has a high concentration of type I collagen and can support high tensile stress without breaking. However, it is a dynamic structure in continuous equilibrium between growth factor-modulated collagen synthesis and its degradation by proteases arising from activated macrophages. The apoptosis of smooth muscle cells can also weaken cap tissue and favour plaque rupture. Macrophage infiltration has been consistently demonstrated in pathological studies; the proportion of macrophages is six to nine times greater in ruptured than in stable plaques and is characterized by the presence of activated T-lymphocytes at the site of plaque rupture that can release various cytokines that activate macrophages and promote smooth muscle cell proliferation.³⁰ These cells may produce proteases that digest the extracellular matrix. In vitro, macrophages induce the breakdown of collagen obtained from human fibrous caps, and protease inhibitors can block this process.

3.2 Coronary thrombosis

The central role of thrombosis in the development of ACS has been widely demonstrated by means of autopsy data^31,32 and angiographic and angioscopic detection of thrombi at the site of the culprit lesion.³³ In addition, the detection of markers of thrombin generation and platelet activation³⁴ and evidence of improved outcomes with antithrombotic treatments have contributed to our understanding of the role of thrombosis in ACS.

Coronary thrombosis in ACS usually develops at the site of a vulnerable plaque. The lipid-rich core exposed after plaque rupture is highly thrombogenic and has a high concentration of tissue factor.³⁵ Thrombosis is induced at the site of plaque rupture or erosion and may lead to rapid changes in the severity of stenosis that may cause subtotal or total vessel occlusion. The thrombus is fibrin-rich and completely occlusive in STEMI, whereas it is platelet-rich and partially or intermittently occlusive in NSTE-ACS.

Spontaneous thrombolysis may explain transient episodes of thrombotic vessel occlusion/subocclusion and the associated transient ischaemia. A platelet-rich thrombus at the site of plaque rupture may fragment into small particles, which embolize downstream and may occlude arterioles and capillaries. These platelet emboli may cause small areas of necrosis in the myocardium supplied by the culprit vessel, thus leading to the release of markers of myocardial necrosis.^31,32

3.3 The vulnerable patient

There is increasing experimental and clinical evidence to suggest the diffuse nature of unstable plaques in patients with ACS. Multiple sites of plaque rupture with or without intracoronary thrombosis, along with elevated levels of various systemic markers of inflammation and thrombosis as well as coagulation system activation, have been documented in patients with ACS.^36–38 Hypercholesterolaemia, tobacco smoking, and increased fibrinogen levels have been reported to contribute to instability in these patients, leading to thrombotic complications.

The concept of widespread instability has important therapeutic implications, because beyond the focal revascularization strategies, such individuals should have systemic therapies aimed at stabilizing the high-risk profile that may cause recurrent ischaemic events.

3.4 Endothelial vasodilatory dysfunction

Minor changes in coronary tone may greatly affect myocardial blood supply and thus cause insufficient flow at rest or during exercise. Vasospasm most frequently occurs at the site of atherosclerotic plaques in which local vasoconstricting substances, such as serotonin, thromboxane A₂, and thrombin, are released locally by platelets and intracoronary thrombi. It has been shown that the endothelium is a multifunctional organ, the integrity of which is essential for normal tone modulation. Endothelial dysfunction is linked to prognosis and is unmasked by vasoconstriction induced by acetylcholine and methacholine.^39,40 The prototype of dynamic coronary obstruction as a cause of ACS is Prinzmetal's variant angina, in which coronary vasospasm is the main determinant of an abrupt reduction in flow. This usually occurs at sites of critical or subcritical stenoses.⁴¹

3.5 Accelerated atherosclerosis

Severe endothelial injury appears to be the critical initiating event that causes smooth muscle cell proliferation in accelerated atherosclerosis. This is followed by intense platelet activation and thrombus formation leading to rapidly progressive coronary narrowing. An angiographic study of patients on the waiting list for percutaneous coronary revascularization has shown that rapid progression of pre-existing atherosclerotic stenoses is relatively common, and the risk arising from complex stenoses is greater than that associated with smooth lesions.⁴²

3.6 Secondary mechanisms

A number of extra-cardiac mechanisms can cause a critical increase in myocardial oxygen consumption to above the supply threshold, and thus elicit an ACS episode with or without a pre-existing coronary stenosis. The mechanisms related to an increase in myocardial oxygen consumption are fever, tachycardia, thyrotoxicosis, a hyperadrenergic state, sudden emotional stress, and increased left ventricular (LV) afterload (hypertension, aortic stenosis), whereas those related to reduced myocardial oxygen delivery are anaemia, methaemoglobinaemia, and hypoxaemia. Triggers such as emotional upset, vigorous physical exercise, lack of sleep, or overeating have been shown to precipitate the onset of ACS.⁴³

3.7 Myocardial injury

Pathological studies in patients with NSTE-ACS show a broad spectrum of findings in the myocardium supplied by the culprit vessel. The myocardium may be normal or there may be varying degrees of necrosis. In some patients, focal areas of cell necrosis in the myocardium supplied by the culprit artery have been shown, which have been attributed to repeated episodes of thrombus embolization.^31,32 Focal myocardial necrosis was shown to be surrounded by areas of inflammation.⁴⁴ In clinical practice, this minor damage may be detected only by cardiac troponin T (cTnT) or troponin I (cTnI) elevations and are classified as MI according to the ESC/AHA/ACC Consensus Document.⁵ This concept is of clinical importance, because it has major practical implications with respect to short-term prognosis and the choice of the therapeutic regimen.

4. Diagnosis and risk assessment

Diagnosis and risk stratification are closely linked in ACS. During the process of establishing the diagnosis of ACS and excluding differential diagnoses, the risk is repeatedly assessed and serves as a guide for the therapeutic management. Patients with NSTE-ACS are at high risk for MI, recurrence of MI, or death. Risk must not be understood in a binary way, but rather as a continuum from patients with very high risk to patients with low risk.

4.1 Clinical presentation and history

The clinical presentation of NSTE-ACS encompasses a wide variety of symptoms. Traditionally, several clinical presentations have been distinguished:

• Prolonged (>20 min) anginal pain at rest,

• New onset (de novo) severe angina [Class III of the Classification of the Canadian Cardiovascular Society⁴⁵ (CCS)]

• Recent destabilization of previously stable angina with at least CCS III angina characteristics (crescendo angina), or

• Post-MI angina.

Prolonged pain is observed in 80% of patients, whereas de novo or accelerated angina is observed in only 20%.⁴⁶ It is important to note that a reliable distinction between ACS with or without ST-elevation cannot be based on symptoms.

The typical clinical presentation of NSTE-ACS is retro-sternal pressure or heaviness (‘angina’) radiating to the left arm, neck, or jaw, which may be intermittent (usually lasting several minutes) or persistent. These complaints may be accompanied by other symptoms such as diaphoresis, nausea, abdominal pain, dyspnoea, and syncope. However, atypical presentations of NSTE-ACS are not uncommon.⁴⁷ These include epigastric pain, recent-onset indigestion, stabbing chest pain, chest pain with some pleuritic features, or increasing dyspnoea. Atypical complaints are often observed in younger (25–40 years) and older (>75 years) patients, in women, and in patients with diabetes, chronic renal failure, or dementia.^47,48 Absence of chest pain leads to under-recognition of the disease and under-treatment.⁴⁹ The diagnostic and therapeutic challenges arise especially when the ECG is normal or nearly normal, or conversely when the ECG is abnormal at baseline due to underlying conditions such as intraventricular conduction defects or LV hypertrophy.¹³

There are certain features regarding the symptoms that may support the diagnosis of CAD and guide the management. The exacerbation of symptoms by physical exertion or their relief at rest or after nitrates support a diagnosis of ischaemia. Symptoms at rest carry a worse prognosis than symptoms elicited only during physical exertion. In patients with intermittent symptoms, an increasing number of episodes preceding the index event may also have an impact on outcome. The presence of tachycardia, hypotension, or heart failure upon presentation indicates a poor prognosis and needs rapid diagnosis and management. It is important to identify clinical circumstances that may exacerbate or precipitate NSTE-ACS, such as anaemia, infection, inflammation, fever, and metabolic or endocrine (in particular thyroid) disorders.

A classification of unstable angina was introduced by Braunwald⁵⁰ and was based on the severity of pain, the circumstances under which it occurs, and precipitating factors associated with its onset, and was later validated as a prognostic tool.⁵¹ However, its usefulness in the clinical setting is limited to the finding that patients with pain at rest during the last 48 h are at increased risk, particularly if troponins are elevated.⁵²

When faced with a symptomatic patient, there are several clinical findings that increase the probability of a diagnosis of CAD and therefore NSTE-ACS. These include older age, male gender, and known atherosclerosis in non-coronary territories, such as peripheral or carotid artery disease. The presence of risk factors, in particular diabetes mellitus and renal insufficiency as well as prior manifestation of CAD, i.e. previous MI, percutaneous coronary intervention (PCI), or coronary bypass graft surgery (CABG), also raises the likelihood of NSTE-ACS. However, all of these factors are not specific, so that their diagnostic value should not be overestimated.

4.2 Diagnostic tools

4.2.1 Physical examination

The physical examination is frequently normal. Signs of heart failure or haemodynamic instability must prompt the physician to expedite the diagnosis and treatment of patients. An important goal of the physical examination is to exclude non-cardiac causes of chest pain and non-ischaemic cardiac disorders (e.g. pulmonary embolism, aortic dissection, pericarditis, valvular heart disease), or potentially extra-cardiac causes, such as acute pulmonary diseases (e.g. pneumothorax, pneumonia, pleural effusion). In this regard, differences in blood pressure between the upper and lower limbs, an irregular pulse, heart murmurs, a friction rub, pain on palpation, and abdominal masses are physical findings that may suggest a diagnosis other than NSTE-ACS. Other physical findings such as palor, increased sweating, or tremor may orientate towards precipitating conditions, such as anaemia and thyrotoxicosis.

4.2.2 Electrocardiogram

The resting 12-lead ECG is the first-line diagnostic tool in the assessment of patients with suspected NSTE-ACS. It should be obtained within 10 min after first medical contact upon arrival of the patient in the emergency room and immediately interpreted by a qualified physician.⁵³ The finding of persistent (>20 min) ST-elevation suggests STEMI which requires different treatment.² In the absence of ST-elevation, additional recordings should be obtained when the patient is symptomatic and compared with recordings obtained in an asymptomatic state. Comparison with a previous ECG, if available, is valuable, particularly in patients with co-existing cardiac disorders such as LV hypertrophy or a previous MI. ECG recordings should be repeated at least at 6 and 24 h, and in the case of recurrence of chest pain/symptoms. A pre-discharge ECG is advisable.

ST-segment shifts and T-wave changes are the ECG indicators of unstable CAD.^21,54 The number of leads showing ST-depression and the magnitude of ST-depression are indicative of the extent and severity of ischaemia and correlate with prognosis.⁵⁵ ST-segment depression ≥ 0.5 mm (0.05 mV) in two or more contiguous leads, in the appropriate clinical context, is suggestive of NSTE-ACS and linked to prognosis.⁵⁶ Minor (0.5 mm) ST-depression may be difficult to measure in clinical practice. More relevant is ST-depression of ≥ 1 mm (0.1 mV) which is associated with an 11% rate of death and MI at 1 year.⁵⁴ ST-depression of ≥ 2 mm carries about a six-fold increased mortality risk.⁵⁷ ST-depression combined with transient ST-elevation also identifies a high-risk subgroup.⁵⁸

Patients with ST-depression have a higher risk for subsequent cardiac events compared with those with isolated T-wave inversion (>1 mm) in leads with predominant R-waves, who in turn have a higher risk than those with a normal ECG on admission. Some studies have cast doubt on the prognostic value of isolated T-wave inversion. However, deep symmetrical inversion of the T-waves in the anterior chest leads is often related to a significant stenosis of the proximal left anterior descending coronary artery or main stem.⁵⁹

It should be appreciated that a completely normal ECG does not exclude the possibility of NSTE-ACS. In several studies, around 5% of patients with normal ECG who were discharged from the emergency department were ultimately found to have either an acute MI or an unstable angina.^60,61 Particularly, ischaemia in the territory of the circumflex artery frequently escapes the common 12-lead ECG, but may be detected in lead V₄R and V₃R as well as in leads V₇–V₉. Transient episodes of bundle branch block occasionally occur during ischaemic attacks.

Continuous ST-segment monitoring

The standard ECG at rest does not adequately reflect the dynamic nature of coronary thrombosis and myocardial ischaemia. Almost two-thirds of all ischaemic episodes in the phase of instability are clinically silent and, hence, not likely to be detected by conventional ECG. On-line continuous computer-assisted 12-lead ST-segment monitoring is a valuable diagnostic tool. Several studies revealed that 15–30% of patients with NSTE-ACS have transient episodes of ST-segment changes, predominantly ST-segment depression. These patients have an increased risk of subsequent cardiac events. ST-monitoring adds independent prognostic information to the ECG at rest, troponins, and other clinical parameters.^62–65

Exercise or other stress testing

In patients who continue to have typical ischaemic chest pain, no stress test should be performed. However, a stress test has a predictive value and is therefore useful before discharge in patients with non-diagnostic ECG provided there is no pain, no signs of heart failure, and normal biomarkers (repeat testing). Early exercise testing has a high negative predictive value. Parameters reflecting cardiac performance provide at least as much prognostic information as those reflecting ischaemia, while the combination of these parameters gives the best prognostic information.⁶⁶

4.2.3 Biochemical markers

Several biomarkers have been investigated in recent years to be used for diagnostic and risk stratification. These reflect different pathophysiological aspects of NSTE-ACS, such as minor myocardial cell injury, inflammation, platelet activation, or neurohormonal activation. For the long-term prognosis, indicators of LV and renal dysfunction or diabetes also play an important role.

Markers of myocardial injury

cTnT or cTnI are the preferred markers of myocardial injury, because they are more specific and more sensitive than the traditional cardiac enzymes such as creatinine kinase (CK) or its isoenzyme MB (CK-MB). In this setting, myoglobin is not specific and sensitive enough to allow the detection of myocardial cell injury and therefore not recommended for routine diagnosis and risk stratification.⁶⁷

It is believed that the elevation of cardiac troponins reflects irreversible myocardial cellular necrosis typically resulting from distal embolization of platelet-rich thrombi from the site of a ruptured plaque. Accordingly, troponins may be seen as a surrogate marker of active thrombus formation. In the setting of myocardial ischaemia (chest pain, ST-segment changes) troponin elevation has to be labelled as MI according to the ESC/ACC/AHA Consensus Document⁵ currently under revision.⁶⁸

Troponins are the best biomarker to predict short-term (30 days) outcome with respect to MI and death.^69–72 The prognostic value of troponin measurements has also been confirmed for the long term (1 year and beyond). The increased risk associated with elevated troponin levels is independent of and additive to other risk factors such as ECG changes at rest or on continuous monitoring, or markers of inflammatory activity.^52,71 Furthermore, the identification of patients with elevated troponins levels is also useful for selecting appropriate treatment in patients with NSTE-ACS.^73–75

In patients with MI, an initial rise in troponins in peripheral blood occurs after 3–4 h. Troponin levels may persist elevated for up to 2 weeks caused by proteolysis of the contractile apparatus. In NSTE-ACS, minor elevation of troponins may be measurable only over 48–72 h (Figure 2). The high sensitivity of troponin tests allows the detection of myocardial damage undetected by CK-MB in up to one-third of patients presenting with NSTE-ACS. Minor or moderate elevations of troponins appear to carry the highest early risk in patients with NSTE-ACS.⁷²

A single negative test for troponins on arrival of the patient in hospital is not sufficient for ruling out an elevation, as in many patients an increase in troponins can be detected only in the subsequent hours. In order to demonstrate or to exclude myocardial damage, repeated blood sampling and measurements are required 6–12 h after admission and after any further episodes of severe chest pain.⁷⁶ A second sample in the absence of any other suspicious findings may be omitted only if the patient's last episode of chest pain was more than 12 h prior to the initial determination of troponins.

It is important to stress that other life-threatening conditions presenting with chest pain, such as dissecting aortic aneurysm or pulmonary embolism, may also result in elevated troponins and should always be considered as a differential diagnosis. Elevation of cardiac troponins also occurs in the setting of non-coronary-related myocardial injury (Table 3). This reflects the sensitivity of the marker for myocardial cell injury and should not be labelled as false-positive test results. True ‘false-positive’ results have been documented in the setting of skeletal myopathies or chronic renal failure. Elevation of troponins is frequently found when serum creatinine level is > 2.5 mg/dL (221 µmol/L) in the absence of proven ACS, and is also associated with adverse prognosis.^77,78 Troponin elevations that cannot be explained are rare.

There is no fundamental difference between troponin T and troponin I. Differences between study results are predominantly explained by varying inclusion criteria, differences in sampling patterns, and the use of assays with different diagnostic cut-offs. The diagnostic cut-off for MI using cardiac troponins should be based on the 99th percentile of levels among healthy controls as recommended by the Consensus committee. Acceptable imprecision (coefficient of variation) at the 99th percentile for each assay should be ≤ 10%.⁵ Each individual laboratory should regularly assess the range of reference values in their specific setting.

The diagnosis of NSTE-ACS should never be made only on the basis of cardiac biomarkers, whose elevation should be interpreted in the context of other clinical findings.

Markers of inflammatory activity

Of the numerous inflammatory markers that have been investigated over the past decade, C-reactive protein measured by high-sensitive (hsCRP) assays is the most widely studied and linked to higher rates of adverse events. The exact source of elevated hsCRP levels among patients with NSTE-ACS remains unclear. Given that myocardial damage is also a major inflammatory stimulus, an acute inflammatory process induced by myocardial damage is superimposed on a chronic inflammatory condition, both of which might influence long-term outcome in NSTE-ACS.

There is robust evidence that even among patients with troponin-negative NSTE-ACS, elevated levels of hsCRP are predictive of long-term mortality (>6 months).^37,71,79,80 The FRISC study confirmed that mortality is associated with elevated hsCRP levels at the time of the index event and continues to increase over 4 years.³⁶ This was also observed in large cohorts of patients submitted to planned PCI.⁸¹ However, hsCRP has no role for the diagnosis of ACS.

Markers of neurohumoral activation

Neurohumoral activation of the heart can be monitored by measurements of systemic levels of natriuretic peptides secreted from the heart. Natriuretic peptides, such as brain-type [B-type natriuretic peptide (BNP)] or its N-terminal prohormone fragment (NT-proBNP), are highly sensitive and fairly specific markers for the detection of LV dysfunction. There are robust retrospective data in NSTE-ACS showing that patients with elevated BNP or NT-proBNP levels have a three- to five-fold increased mortality rate when compared with those with lower levels.^82,83 The level is strongly associated with the risk of death even when adjusted for age, Killip class, and LV ejection fraction (EF).⁷¹ Values taken a few days after onset of symptoms seem to have superior predictive value when compared with measurements on admission.^84,85 Natriuretic peptides are useful markers in the emergency room in evaluating chest pain or dyspnoea and were shown to be helpful in differentiating between cardiac and non-cardiac causes of dyspnoea. However, they are markers of long-term prognosis, but have limited value for initial risk stratification and hence for selecting the initial therapeutic strategy in NSTE-ACS.⁸⁶

Markers of renal function

Impaired renal function is a strong independent predictor for long-term mortality in ACS patients.^71,87,88 Serum creatinine concentration is a less reliable indicator of renal function than creatinine clearance (CrCl) or glomerular filtration rate (GFR), because it is affected by a multitude of factors, including age, weight, muscle mass, race, and various medications.⁸⁹ Several formulae have been devised to improve the accuracy of serum creatinine level as a surrogate of GFR, including the Cockcroft–Gault⁹⁰ and the abbreviated Modification of Diet in Renal Disease (MDRD) equations.⁹¹ Long-term mortality is influenced by the degree of renal function, as it increases exponentially with decreasing GFR/CrCl. When compared with patients with normal renal function, the odds ratio (OR) for death at 1 year was 1.76 for mild renal dysfunction, 2.72 for moderate renal dysfunction, and 6.18 for severe renal dysfunction.⁸⁸ (see section 7.4 Chronic kidney disease).

Cystatin C is considered to be a surrogate marker of renal function superior to CrCl or GFR estimation.^93,94 Cystatin C is a cysteine proteinase inhibitor produced by all nucleated cells at a constant rate and excreted into the bloodstream. Because of its low molecular weight (13 kDa), it is freely filtered at the glomerulus and is almost completely reabsorbed and catabolized, but not secreted, by tubular cells. Cystatin C levels have been shown to be good markers of prognosis,⁹⁵ although not widely available yet.

Novel biomarkers

A considerable number of patients can still not be identified as being at high risk by today's routine biomarkers. Accordingly, a great number of novel biomarkers have been investigated in recent years to explore their usefulness as diagnostic tools and for risk stratification in addition to established markers. Several novel biomarkers have been studied. These include markers of oxidative stress (myeloperoxidase),^96,97 markers of thrombosis and inflammation (e.g. soluble CD40 ligand),^98,99 or markers involved more upstream in the inflammation cascade, i.e. markers specific of vascular inflammation. All have shown their incremental value over troponins in retrospective analyses, but have not been tested prospectively and are not yet available for routine use.

Multimarker approach

Since NSTE-ACS is a complex event, several markers reflecting the respective pathophysiological pathways may be advantageous for risk stratification. It is useful to distinguish between markers for the acute risk of MI and for long-term mortality. The combined use of markers for myocardial necrosis, inflammation, myocardial and renal dysfunction, and neurohumoral activation may significantly add to our ability to identify correctly patients who are at high risk for future cardiovascular events. Several studies demonstrated that a multimarker approach improves risk stratification.^71,79,98

Currently, it is recommended to use troponins (cTnT or cTnI) for the acute risk stratification on arrival of the patient in the hospital. At the same time or during the subsequent days, CrCl and BNP or NT-pro-BNP allow estimation of any renal or myocardial dysfunction with its inherent impacts on treatment and long-term outcome. Currently, only hsCRP is available on a routine basis for the detection of the underlying inflammatory activity responsible for long-term mortality.

Point-of-care (bedside) biomarker testing

The diagnosis of NSTE-ACS and the assignment to a risk group should be undertaken as rapidly as possible (see section 8 Management strategies). Point-of-care testing for biochemical markers is advantageous to establish diagnosis. These tests can be performed either directly at the bedside or at ‘near patient’ locations such as the emergency department, chest pain evaluation centre, or intensive care unit.^76,100,101 Point-of-care tests for troponins should be implemented when a central laboratory cannot consistently provide test results within 60 min.¹⁰² No special skill or prolonged training is required to read the result of these assays. Accordingly, these tests can be performed by various members of the health-care team after adequate training.¹⁰³ However, reading of these mostly qualitative tests is performed visually and therefore is observer-dependent. Some companies provide optical reading devices for the emergency room setting.¹⁰⁴ The tests are usually reliable when positive. However, in the presence of a remaining suspicion of unstable CAD, negative tests should be repeated at a later time point and verified by a central laboratory.

4.2.4 Echocardiography and non-invasive myocardial imaging

LV systolic function is an important prognostic variable in patients with ischaemic heart disease and can be easily and accurately assessed by echocardiography. In experienced hands, transient localized hypokinesia or akinesia in segments of the left ventricle wall may be detected during ischaemia, with normal wall motion on resolution of ischaemia. Furthermore, differential diagnoses such as aortic stenosis, aortic dissection, pulmonary embolism, or hypertrophic cardiomyopathy may be identified.¹⁰⁵ Therefore, echocardiography should routinely be used in emergency units.

Stress echocardiography is helpful in stabilized patients to obtain objective evidence of ischaemia and has the same indications as other exercise modalities.¹⁰⁶ Similarly, stress scintigraphy^107,108 or magnetic resonance imaging (MRI)¹⁰⁹ may be used, if available. MRI is useful to assess myocardial viability. Rest myocardial scintigraphy was shown to be helpful for initial triage of patients presenting with chest pain without ECG changes or evidence of ongoing MI.¹¹⁰

4.2.5 Imaging of the coronary anatomy

Imaging modalities provide unique information on the presence and the severity of CAD. The gold standard is still conventional invasive coronary angiography.

Patients with multiple vessel disease as well as those with left main stenosis are at the highest risk of serious cardiac events.¹¹¹ Angiographic assessment of the characteristics and location of the culprit lesion as well as other lesions is essential if revascularization is being considered. Complex, long, heavily calcified lesions, angulations and extreme tortuosity of the vessel are indicators of risk. Highest risk is associated with the occurrence of filling defects indicating intracoronary thrombus formation.

At the current state of development, cardiac computed tomography (CT) cannot be recommended as the coronary imaging modality in NSTE-ACS, because of suboptimal diagnostic accuracy. Fast technical evolution may result in improved diagnostic accuracy in the near future and lead to reconsideration of the use of this tool in the decision-making process.¹¹² Furthermore, because of the high likelihood of PCI, time is lost and the patient is exposed to unnecessary radiation and contrast medium utilization if CT is used as the first diagnostic option.

MRI is not established as an imaging tool for coronary arteries. It may only be useful in the course of hospitalization in quantifying myocardial injury or excluding myocarditis.¹⁰⁹ CT or MRI may, however, be indicated for evaluation of differential diagnoses, such as pulmonary embolism or aortic dissection.

4.3 Differential diagnoses

There are several cardiac and non-cardiac conditions that may mimic NSTE-ACS (Table 4).

Underlying chronic heart conditions such as hypertrophic cardiomyopathy and valvular heart disease (i.e. aortic stenosis, aortic regurgitation) may be associated with typical symptoms of NSTE-ACS, elevated cardiac biomarkers, and ECG changes.¹¹³ Since some patients with these underlying conditions also have CAD, the diagnostic process can be difficult.

Myocarditis, pericarditis, or myopericarditis of different aetiologies may be associated with chest pain that resembles the typical angina of NSTE-ACS and be associated with a rise in cardiac biomarker levels, ECG changes, and wall motion abnormalities. A flu-like, febrile condition with symptoms attributed to the upper respiratory tract often precedes or accompanies these conditions. However, infections, especially of the upper respiratory tract, also often precede or accompany NSTE-ACS.¹¹⁴ The definitive diagnosis of myocarditis or myopericarditis may frequently only be established during the course of hospitalization.

Non-cardiac, life-threatening conditions may mimic NSTE-ACS and must be diagnosed. Among these, pulmonary embolism may be associated with dyspnoea, chest pain, ECG changes, as well as elevated levels of cardiac biomarkers similar to those of NSTE-ACS.¹¹⁵ Chest X-ray, CT, or MRI angiography of the pulmonary arteries, pulmonary perfusion scans, and blood levels of d-dimer are the recommended diagnostic tests. Aortic dissection is another condition to be considered as an important differential diagnosis. NSTE-ACS may be a complication of aortic dissection when the dissection involves the coronary arteries. In a patient with undiagnosed aortic dissection, the current therapies for NSTE-ACS may exacerbate the patient's condition and result in detrimental outcomes. Stroke may be accompanied by ECG changes, wall motion abnormalities, and a rise in cardiac biomarker levels.¹¹⁶ Conversely, atypical symptoms such as headache and vertigo may in rare cases be the sole presentation of myocardial ischaemia.

4.4 Risk scores

Several risk stratification scores have been developed and validated in large patient populations. In clinical practice, only simple risk scores are useful.

The GRACE risk scores^8,117,118 are based upon a large unselected population of an international registry of the full spectrum of ACS patients. The risk factors were derived with independent predictive power for in-hospital deaths¹¹⁸ and post-discharge deaths at 6 months.⁸ Easy to assess clinical/ECG/laboratory variables such as age, heart rate, systolic blood pressure, serum creatinine level, Killip class at admission, presence of ST-depression, and elevated cardiac biomarkers as well as cardiac arrest are included in the calculation. The models were validated in GRACE and GUSTO-2B, as well as externally in a Mayo Clinic population, a Canadian ACS Registry, and a Portuguese Registry. The GRACE models have very good discriminative power. Their complexity, however, requires special tools (graphs, tables, or computer programs) to estimate risk at the bedside. Computer or PDA software of a simplified nomogram are freely available at http://www.outcomes.org/grace. According to the GRACE risk score, three risk categories have been developed (Table 5). On the basis of direct comparisons,¹¹⁹ the GRACE risk score is recommended as the preferred classification to apply on admission and at discharge in daily clinical routine practice.

The TIMI risk score¹²⁰ is derived from the TIMI-11B trial population and was validated in the TIMI-11B and ESSENCE patients, as well as externally such as in a Mayo Clinic population, in the TIMI-3 and Portuguese Registries. The TIMI risk score was applied in analysing treatment efficacy in various risk groups. It is less accurate in predicting events, but its simplicity makes it useful and widely accepted. The FRISC score is based on similar variables and was derived from the 1-year outcome of the FRISC-2 trial.¹²¹ This is the only risk score that has repeatedly been shown to identify patients with a long-term benefit of an early invasive treatment strategy in a randomized trial.¹²² The PURSUIT risk score is based upon the PURSUIT trial population and was validated externally in the Canadian ACS Registry, a Mayo Clinic population, and in a Portuguese Registry.¹²³ It allows separate risk stratification of patients with unstable angina and NSTE-MI. It is a complex model with high discriminative power, but poor calibration in the Canadian ACS Registry.

Recommendations for diagnosis and risk stratification

• Diagnosis and short-term risk stratification of NSTE-ACS should be based on a combination of clinical history, symptoms, ECG, biomarkers, and risk score results (I-B).

• The evaluation of the individual risk is a dynamic process that is to be updated as the clinical situation evolves.

  • A 12-lead ECG should be obtained within 10 min after first medical contact and immediately read by an experienced physician (I-C). Additional leads (V₃R and V₄R, V₇–V₉) should be recorded. ECG should be repeated in the case of recurrence of symptoms, and at 6 and 24 h and before hospital discharge (I-C).

  • Blood must be drawn promptly for troponin (cTnT or cTnI) measurement. The result should be available within 60 min (I-C). The test should be repeated after 6–12 h if the initial test is negative (I-A).

  • Established risk scores (such as GRACE) should be implemented for initial and subsequent risk assessment (I-B).

  • An echocardiogram is recommended to rule in/out differential diagnoses (I-C).

  • In patients without recurrence of pain, normal ECG findings, and negative troponins tests, a non-invasive stress test for inducible ischaemia is recommended before discharge (I-A).

• The following predictors of long-term death or MI should be considered in risk stratification (I-B).

  • Clinical indicators: age, heart rate, blood pressure, Killip class, diabetes, previous MI/CAD;

  • ECG markers: ST-segment depression;

  • Laboratory markers: troponins, GFR/CrCl/cystatin C, BNP/NT-proBNP, hsCRP;

  • Imaging findings: low EF, main stem lesion, three-vessel disease;

  • Risk score result.

5. Treatment

The treatment options described in this section are based on evidence from numerous clinical trials or meta-analyses.

Four categories of acute treatment are discussed: anti-ischaemic agents, anticoagulants, antiplatelet agents, and coronary revascularization. Generally, the therapeutic approach is based on whether the patient is to be only medically treated, or in addition referred to angiography and revascularization. Many of the treatment options were evaluated more than two decades ago or tested only in specific subsets of patients. The recommendations take these circumstances into account.

5.1 Anti-ischaemic agents

These drugs decrease myocardial oxygen consumption (decreasing heart rate, lowering blood pressure, or depressing LV contractility) and/or induce vasodilatation.

5.1.1 Beta-blockers

Evidence for the beneficial effects of beta-blockers in unstable angina is based on limited randomized trial data, along with pathophysiological considerations and extrapolation from experience in stable angina and STEMI. Beta-blockers competitively inhibit the myocardial effects of circulating catecholamines. In NSTE-ACS, the primary benefits of beta-blockers are related to their effects on beta-1 receptors that result in a decrease in myocardial oxygen consumption.

Two double-blind, randomized trials have compared beta-blockers with placebo in unstable angina.^124,125 A meta-analysis suggested that beta-blocker treatment was associated with a 13% relative reduction in risk of progression to STEMI.¹²⁶ Although no significant effect on mortality in NSTE-ACS has been demonstrated in these relatively small trials, the results may be extrapolated from larger randomized trials of beta-blockers in patients with unselected MI.¹²⁷

Beta-blockers are recommended in NSTE-ACS in the absence of contraindications and are usually well tolerated. In most cases, oral treatment is sufficient. The target heart rate for a good treatment effect should be between 50 and 60 b.p.m. Patients with significantly impaired atrioventricular conduction and a history of asthma or of acute LV dysfunction should not receive beta-blockers.

5.1.2 Nitrates

The use of nitrates in unstable angina is largely based on pathophysiological considerations and clinical experience. The therapeutic benefits of nitrates and similar drug classes such as sydnonimines are related to their effects on the peripheral and coronary circulation. The major therapeutic benefit is probably related to the venodilator effects that lead to a decrease in myocardial pre-load and LV end-diastolic volume, resulting in a decrease in myocardial oxygen consumption. In addition, nitrates dilate normal as well as atherosclerotic coronary arteries and increase coronary collateral flow.

Studies of nitrates in unstable angina have been small and observational.^128–130 There are no randomized placebo-controlled trials to confirm the benefits of this class of drugs either in relieving symptoms or in reducing major adverse cardiac events. Only very scarce data exist about the best route for administrating nitrates (intravenous, oral, sublingual, or topical) and about the optimal dose and duration of therapy.^131,132

In patients with NSTE-ACS who require hospital admission, intravenous nitrates may be considered in the absence of contraindications. The dose should be titrated upwards until symptoms (angina and/or dyspnoea) are relieved unless side effects (notably headache or hypotension) occur. A limitation of continuous nitrate therapy is the phenomenon of tolerance, which is related to both the dose administered and the duration of treatment. When symptoms are controlled, intravenous nitrates may be replaced by non-parenteral alternatives with appropriate nitrate-free intervals. An alternative is to use nitrate-like drugs such as sydnonimines or potassium channel activators. Nitric oxide donor therapy (nitrates and sydnonimines) is contraindicated in patients taking phosphodiesterase-5 inhibitors (sildenafil, vardenafil, tadalafil) because of the risk of profound vasodilatation and blood pressure drop in the case of concomitant administration.

5.1.3 Calcium channel blockers

Calcium channel blockers are vasodilating drugs. In addition, some have significant direct effects on atrioventricular conduction and heart rate. There are three subclasses of calcium blockers, which are chemically distinct and have different pharmacological effects: the dihydropyridines (such as nifedipine), the benzothiazepines (such as diltiazem), and the phenylalkylamines (such as verapamil). The agents in each subclass vary in the degree to which they cause vasodilatation, decrease myocardial contractility, and delay atrioventricular (A-V) conduction. A-V block may be induced by non-dihydropyridines. Nifedipine and amlodipine produce the most marked peripheral arterial vasodilatation, whereas diltiazem has the least vasodilatory effect. All subclasses cause similar coronary vasodilatation.

There are only small randomized trials testing calcium channel blockers in NSTE-ACS. Generally, they show efficacy in relieving symptoms that appear equivalent to beta-blockers.^133,134 The HINT study, the largest randomized trial, tested nifedipine and metoprolol in a 2 × 2 factorial design.¹²⁵ Although no statistically significant differences were observed, there was a trend towards an increased risk of MI or recurrent angina with nifedipine (compared with placebo), whereas treatment with metoprolol, or with a combination of both drugs, was associated with a reduction in these events.

The beneficial effect after discharge is somewhat controversial.^135,136 A meta-analysis of the effects of calcium channel blockers on death or non-fatal MI in unstable angina suggests that this class of drugs does not prevent the development of acute MI or reduce mortality.¹³⁷ In particular, several analyses of pooled data from observational studies suggest that short-acting nifedipine might be associated with a dose-dependent detrimental effect on mortality in patients with CAD.^138,139 On the other hand, there is evidence for a protective role of diltiazem in NSTEMI in one trial.¹⁴⁰

Calcium channel blockers, particularly dihydropyridines, are the drugs of choice in vasospastic angina.

5.1.4 New drugs

New antianginal drugs with different modes of action have been investigated in recent years. Ivabradine selectively inhibits the primary pacemaker current in the sinus node and may be used in patients with beta-blocker contraindications.¹⁴¹ Trimetazidine exerts metabolic effects without haemodynamic changes.¹⁴² Ranolazine exerts antianginal effects by inhibiting the late sodium current.¹⁴³ It was not effective in reducing major cardiovascular events in MERLIN-TIMI 36.⁹² Nicorandil has nitrate-like properties. Nicorandil significantly reduced the rate of occurrence of the primary composite endpoint (coronary death, non-fatal MI, or unplanned hospital admission for cardiac pain) in chronic stable angina patients in the IONA study,¹⁴⁴ but has never been tested in the setting of NSTE-ACS.

Recommendations for anti-ischaemic drugs

• Beta-blockers are recommended in the absence of contraindications, particularly in patients with hypertension or tachycardia (I-B).

• Intravenous or oral nitrates are effective for symptom relief in the acute management of anginal episodes (I-C).

• Calcium channel blockers provide symptom relief in patients already receiving nitrates and beta-blockers; they are useful in patients with contraindications to beta-blockade and in the subgroup of patients with vasospastic angina (I-B).

• Nifedipine, or other dihydropyridines, should not be used unless combined with beta-blockers (III-B).

5.2 Anticoagulants

Anticoagulants are used in the treatment of NSTE-ACS to inhibit thrombin generation and/or activity, thereby reducing thrombus-related events. There is clear evidence that anticoagulation is effective in addition to platelet inhibition and that the combination of the two is more effective than either treatment alone.^145,146 With all anticoagulants, there is an increased risk of bleeding. The risk factors for bleeding are well defined (see section 6.1 Bleeding complications). Several anticoagulants, which act at different levels of the coagulation cascade, have been investigated in NSTE-ACS:

• unfractionated heparin (UFH) as intravenous infusion;

• low molecular weight heparin (LMWH) as subcutaneous injection;

• fondaparinux as subcutaneous injection;

• direct thrombin inhibitors (DTIs) as intravenous infusion;

• vitamin K antagonists (VKAs) as oral medication.

5.2.1 Unfractionated heparin

Pharmacology

UFH is a heterogeneous mixture of polysaccharide molecules, with a molecular weight ranging from 2000 to 30 000 (mostly 15–18 000) Da. One-third of the molecules found within a standard UFH preparation contain the pentasaccharide sequence, which binds to antithrombin and accelerates the rate at which antithrombin inhibits factor-Xa. Inhibition of factor-IIa requires heparin to bind to both thrombin and antithrombin to bridge them. This can only be achieved if the chains containing the pentasaccharide sequence comprise at least 18 saccharide units to provide sufficient length to bridge to factor-IIa. UFH is poorly absorbed by the subcutaneous route, so that intravenous infusion is the preferred route of administration. The therapeutic window is narrow, requiring frequent monitoring of the activated partial thromboplastin time (aPTT), with an optimal target level of 50–75 s, corresponding to 1.5–2.5 times the upper limit of normal. At higher aPTT values, the risk of bleeding complications is increased, without further antithrombotic benefits. At aPTT values lower than 50 s, the antithrombotic effect is limited and the number of ischaemic events not reduced. A weight-adjusted dose of UFH is recommended, at an initial bolus of 60–70 IU/kg with a maximum of 5000 IU, followed by an infusion of 12–15 IU/kg/h, to a maximum of 1000 IU/h. This regimen is that currently recommended as being the most likely to achieve the target aPTT values.^145,146

The maintenance of a well-controlled anticoagulation is difficult by UFH infusion in NSTE-ACS patients, especially as the clinical condition usually improves within the first 24 h, when the patients often are mobilized and even ambulatory. The anticoagulant effect of UFH is rapidly lost within a few hours after interruption. During the first 24 h after termination of treatment, there is a risk of reactivation of the coagulation process and thereby a transiently increased risk of recurrence, despite simultaneous acetylsalicylic acid (aspirin) treatment.¹⁴⁷

Treatment effects

A pooled analysis of six trials testing short-term UFH vs. placebo or untreated controls showed a significant risk reduction for death and MI of 33% (OR 0.67, 95% CI 0.45–0.99, P = 0.045).¹⁴⁸ The risk reduction for MI accounted for practically all of the beneficial effect. When the data from FRISC, which compared LMWH with placebo, are added to this pooled analysis, the risk reduction is even greater (Figure 3). In trials comparing the combination of UFH plus aspirin vs. aspirin alone in NSTE-ACS, a trend towards a benefit was observed in favour of the UFH–aspirin combination, but at the cost of an increase in the risk of bleeding. Recurrence of events after interruption of UFH explains why this benefit is not maintained over time, unless the patient is revascularized before the interruption of UFH^148–150 (Figure 3).

5.2.2 Low molecular weight heparin

Pharmacology

LMWH represents a class of heparin-derived compounds with molecular weights ranging from 2000 to 10 000 Da. LMWHs have pharmacological advantages over UFH. They link to antithrombin through the pentasaccharide sequence. This is the basis of anti-Xa activity. The anti-IIa activity is lower than that with UFH and depends on the molecular weight of the molecule, with greater anti-IIa activity with increasing molecular weight. The advantages of LMWH are an almost complete absorption after subcutaneous administration, less protein binding, less platelet activation, and, thereby, a more predictable dose–effect relationship.^145,146 Furthermore, there is a lower risk of heparin-induced thrombocytopenia (HIT) with LMWH when compared with UFH due to less interaction with platelet factor 4 (PF4) (see section 6.2 Thrombocytopenia). LMWHs are eliminated at least partially by the renal route. They are contraindicated in the case of renal failure with CrCl < 30 mL/min (in certain countries such as the USA, dose adaptation in the case of renal failure is recommended; see section 7.4 Chronic kidney disease).

The LMWH dosages used in NSTE-ACS are body weight adjusted and are of the same magnitude as those used in the treatment of venous thrombo-embolism (VTE), which are higher than the doses used in prophylaxis of deep vein thrombosis (DVT). LMWHs are commonly administered subcutaneously every 12 h in NSTE-ACS to avoid the risk of inadequate anti-Xa level during treatment.^{149,151–155} An initial intravenous bolus in high-risk patients has also been advocated.¹⁵¹ From VTE studies, the therapeutic range of anti-Xa activity has been considered to be 0.6–1.0 IU/mL, without any clear relationship between anti-Xa activity and clinical outcome. However, the bleeding risk increases above 1.0 IU/mL of anti-Xa activity.^145,146 In TIMI-11A, where the enoxaparin dose was 1.5 mg/kg twice daily, patients with major haemorrhage had anti-Xa activity in the range of 1.8–2.0 IU/mL. Excess bleeding led to a reduction in the dosage.¹⁵⁶ With the current doses used in clinical practice, monitoring of anti-Xa activity is not necessary, except in special populations of patients, such as those with renal failure and obesity.

The treatment can be prolonged without immobilization of the patient.^153,157 This allows treatment to be continued until a decision on early coronary angiography is made.¹⁵³ As no superior protection against recurrence of ischaemic events and more bleeding occurred with prolonged treatment, discontinuation of LMWH is recommended at hospital discharge.¹⁵⁷ The risk of bleeding with LMWH is dose-related and is increased with higher age, female gender, lower body weight, reduced renal function, and interventional procedures.¹⁴⁶

Treatment effects

The efficacy of LMWH in aspirin-treated patients with NSTE-ACS has been evaluated vs. placebo in FRISC¹⁵⁷ with dalteparin 120 U/kg twice daily, and in another smaller study.¹⁵⁸ The results showed a substantial risk reduction of death and MI associated with a modest increase in bleeding risk. Several trials have assessed the respective efficacy and safety of various LMWHs in comparison with UFH. Dalteparin and nadroparin were shown to be equally efficacious and safe as UFH in aspirin-treated patients.^155,159 Dalteparin was shown to have greater efficacy in troponin-positive than in troponin-negative patients.¹⁶⁰ Enoxaparin has been compared with UFH in several trials. In ESSENCE and TIMI-11B, an invasive strategy was not encouraged and as a result the revascularization rate was low in both trials when compared with contemporary practice.^151,152

A pooled analysis of both trials showed a significant risk reduction of death and MI at the end of the study period, achieved at the cost of a significant increase in minor (but not major) bleeding complications. In INTERACT and ACUTE-2, a regimen of enoxaparin plus eptifibatide or tirofiban was compared with a regimen of UFH plus eptifibatide or tirofiban in aspirin-treated patients.^161–163 These two trials were insufficiently powered to draw definitive conclusions as to whether enoxaparin regimen had a better efficacy/safety profile than the UFH regimen. In A to Z, enoxaparin plus tirofiban was shown to be non-inferior to a regimen of UFH plus tirofiban.¹⁶¹ SYNERGY was the largest trial to test enoxaparin against UFH in a contemporary approach with a high rate of invasive procedures, use of PCI/revascularization, stent implantation, and active antiplatelet therapy with aspirin, clopidogrel, and a high rate of glycoprotein IIb/IIIa (GP IIb/IIIa) receptor inhibitors. It included 10 027 high-risk patients undergoing early invasive evaluation/revascularization, of whom 76% received anticoagulants pre-randomization. No significant difference was observed in terms of death and MI at 30 days (14.0 vs. 14.5%, OR 0.96; 95% CI 0.86–1.06, P = NS) for enoxaparin vs. UFH.¹⁶⁴ More bleeding occurred with enoxaparin, with a statistically significant increase in TIMI major bleeding (9.1 vs. 7.6%, P = 0.008) but non-significant excess in GUSTO severe bleeding (2.7 vs. 2.2%, P = 0.08) and transfusions (17.0 vs. 16.0%, P = 0.16).

The meta-analysis of these six trials, totalling 21 946 patients, showed no significant difference between the two compounds for death at 30 days (3.0 vs. 3.0%, OR 1.00, 95% CI 0.85–1.17, P = NS).¹⁶⁵ A significant reduction in the combined endpoint of death or MI at 30 days was observed in favour of enoxaparin vs. UFH (10.1 vs. 11.0%, OR 0.91, 95% CI 0.83–0.99). Post hoc subgroup analysis showed a significant reduction in death or MI at 30 days in enoxaparin-treated patients, who did not receive UFH prior to randomization, vs. the UFH group (8.0 vs. 9.4%, respectively, OR 0.81, 95% CI 0.70–0.94). No significant differences in blood transfusions (7.2 vs. 7.5%, OR 1.01, 95% CI 0.89–1.14) or in major bleeding (4.7 vs. 4.5%, OR 1.04, 95% CI 0.83–1.30) were observed at 7 days after randomization in the overall population, as well as in the population of patients who received no anticoagulant therapy prior to randomization (Figure 4).

Enoxaparin has been used in combination with aspirin and thienopyridines in many recent trials or observational studies apparently without safety concerns, particularly bleeding complications.^164,167 However, as the bleeding risk specifically incurred by this combination of different antithrombotic agents was not the main objective of these studies, it is difficult to extract specific data on safety. However, it is known from registry data that combination of antithrombotic drugs is a predictor of increased risk of bleeding¹⁶⁸ (see section 6.1 Bleeding complications). Enoxaparin and other LMWHs have also been tested in combination with GP IIb/IIIa inhibitors in several trials or observational studies, in most without dose adaptation. No complications of excess bleeding were reported, except in SYNERGY where high rates of pre-randomization administration of anticoagulants and crossover from one anticoagulant to the other may have played a role in the excess bleeding observed with enoxaparin compared with heparin.^{161–165,169–171} In INTERACT, eptifibatide and enoxaparin in aspirin-treated patients demonstrated better efficacy than UFH plus eptifibatide, but the sample size was too small to draw any definitive conclusion.¹⁶³

Registry data have shown that the use of enoxaparin in an unselected cohort of patients with NSTE-ACS may lead to a significant risk reduction for death or MI when compared with UFH.¹⁷²

5.2.3 Factor-Xa inhibitors

Pharmacology

The only selective factor-Xa inhibitor available for clinical use is fondaparinux. This is a synthetic pentasaccharide modelled after the antithrombin-binding sequence of UFH. It exerts a selective antithrombin-mediated inhibition of factor-Xa, a dose-dependent inhibition of thrombin generation without inhibition of the thrombin molecule per se. It has a 100% bioavailability after subcutaneous injection with an elimination half-life of 17 h and can therefore be given once daily. It is eliminated mainly by the renal route. It is contraindicated if CrCl is lower than 30 mL/min. It is insensitive to inactivation by platelet-released heparin neutralization proteins. Because it does not induce the formation of heparin–PF4 complexes, HIT is unlikely to occur with fondaparinux. No case of HIT has been reported with this drug, even after extensive use in the setting of prevention and treatment of VTE. Therefore, monitoring of platelet count is not necessary. In ACS, a 2.5 mg fixed dose is recommended. No dose adjustment and no monitoring of anti-Xa activity is required. Fondaparinux has no significant influence on the usual variables that monitor anticoagulant activity, such as aPTT, activated clotting time (ACT), prothrombin, and thrombin times.

Clinical studies have shown advantageous effects compared with UFH and LMWH in the settings of prevention of DVT in orthopaedic and general surgery, in acutely ill medical patients, and in the treatment of VTE. In two small, phase II studies, fondaparinux also showed promising results as a substitute for enoxaparin or UFH in NSTE-ACS¹⁷³ and PCI.¹⁷⁴

Treatment effects

In a double-blind dose-ranging study of fondaparinux vs. enoxaparin in the setting of NSTE-ACS involving 1147 patients, the 2.5 mg dose of fondaparinux was shown to have the best efficacy/safety profile when compared with 4, 8, and 12 mg doses of fondaparinux and with enoxaparin 1 mg/kg twice daily, and was then chosen as the dose for subsequent phase III trials.¹⁷³ Fondaparinux was also tested in a phase II trial in the setting of PCI, at doses of 2.5 or 5 mg, vs. the standard dose of UFH, and was shown to have comparable efficacy and safety to UFH.¹⁷⁴ In this study, catheter thrombi were reported in both UFH and fondaparinux groups, but with a higher frequency with fondaparinux. There was no impact on the rate of clinical events, including peri-procedural MI, but the study was underpowered to detect significant differences in the rate of events.¹⁷⁴

In the OASIS-5 study,^175,176 20 078 patients with NSTE-ACS were randomized to receive 2.5 mg subcutaneous fondaparinux once daily vs. subcutaneous enoxaparin 1 mg/kg twice daily for 8 days maximum (average 5.2 vs. 5.4 days). The primary efficacy outcome of death, MI, or refractory ischaemia at 9 days was 5.7 vs. 5.8% for enoxaparin vs. fondaparinux, respectively (HR 1.01, 95% CI 0.90–1.13). The upper limit of the CI was well below the pre-specified boundary of 1.185 for non-inferiority (P = 0.007). At the same time point, major bleeds were halved with fondaparinux, 2.2% compared with 4.1% with enoxaparin (HR 0.52, 95% CI 0.44–0.61, P < 0.001), and the composite outcome of death, MI, refractory ischaemia, or major bleeding also favoured fondaparinux, 7.3 vs. 9.0% for fondaparinux vs. enoxaparin (HR 0.81, 95% CI 0.73–0.89, P < 0.001). Major bleeding was an independent predictor of long-term mortality, which was significantly reduced with fondaparinux at 30 days (2.9 vs. 3.5%, HR 0.83, 95% CI 0.71–0.97, P = 0.02) and at 6 months (5.8 vs. 6.5%, HR 0.89, 95% CI 0.80–1.00, P = 0.05). The composite outcome of death, MI, or stroke was significantly lower with fondaparinux at 6 months (11.3 vs. 12.5%, HR 0.89, 95% CI 0.82–0.97, P = 0.007). In summary, at 30 days, 167 patients needed to be treated to avoid one death or one death/MI, and 53 patients to avoid one major bleed. In the population submitted to PCI, a significantly lower rate of major bleeding complication (including access site complications) was observed in the fondaparinux group vs. enoxaparin at 9 days, 2.3 vs. 5.1%, respectively (HR 0.45, 95% CI 0.34–0.59, P < 0.001).¹⁷⁶ Catheter thrombus formation during PCI was observed in both groups, though at a significantly higher rate with fondaparinux than with enoxaparin (see section 5.2.6 Anticoagulants during percutaneous coronary intervention). On the basis of OASIS-5, if fondaparinux is chosen as anticoagulant therapy, it should be maintained for up to 5 days or until hospital discharge, and cannot be used as sole anticoagulant during PCI procedures (see section 5.2.6 Anticoagulants during percutaneous coronary intervention).

5.2.4 Direct thrombin inhibitors

Pharmacology

DTIs bind directly to thrombin (factor-IIa) and thereby inhibit the thrombin-induced conversion of fibrinogen to fibrin. They inactivate fibrin-bound, as well as fluid-phase thrombin. As they do not bind to plasma proteins, the anticoagulant effect is more predictable. Unlike heparin, DTIs do not interact with PF4. The model compound for this treatment principle is hirudin extracted from the medical leech (Hirudo medicinalis). Currently, several DTIs (hirudin, argatroban, bivalirudin) are available. Hirudin and its derivatives are eliminated by the renal route. Hirudin and bivalirudin prolong aPTT and ACT. Coagulation tests correlate well with plasma concentrations. Therefore, these two tests can be used to monitor the anticoagulant activity of these compounds.

Treatment effects

Intravenous direct thrombin inhibition with hirudin has been compared with aPTT-monitored UFH infusion in several large-scale randomized trials. In a meta-analysis including all these trials, there was a significantly lower event rate with hirudin vs. UFH infusion^149,177,178 (Figure 5). However, these differences were not sustained during long-term follow-up. Furthermore, the use of hirudin as the primary treatment of NSTE-ACS has been associated with a higher bleeding rate (1.7 vs. 1.3%, OR 1.28, 95% CI 1.06–1.55)¹⁷⁷ and therefore its use in this indication has not been approved. Bivalirudin compared with UFH in the setting of PCI significantly reduced the rate of major adverse cardiac events (death, MI, or repeat revascularization) (6.2 vs. 7.9%; OR 0.78% CI 0.62–0.99, P = 0.039) and of bleeding (3.5 vs. 9.3%; OR 0.34 95% CI 0.26–0.45, P < 0.001 for bivalirudin vs. UFH) at 7 days in one trial.¹⁷⁹ More recently, bivalirudin plus provisional GP IIb/IIIa inhibitors was shown to be non-inferior to UFH plus GP IIb/IIIa inhibitors in the protection against ischaemic events during PCI procedures, but with a significantly lower rate of major bleeding complications (2.4 vs. 4.1%, P < 0.001 for bivalirudin vs. UFH plus GP IIb/IIIa inhibitors).^180,181 Bivalirudin is currently recommended as an alternative anticoagulant for urgent and elective PCI.¹⁸² Hirudin, bivalirudin, and argatroban have been used for treatment of HIT complicated by thrombotic events.^183–185

The ACUITY trial was a randomized, open-label trial in 13 819 patients with moderate to high risk NSTE-ACS planned for an invasive strategy.^186,187 Three primary 30-day endpoints were pre-specified, some of them including not so robust components: composite ischaemia (death from any cause, MI, or unplanned revascularization for ischaemia), major bleeding (non-CABG-related), and net clinical outcome (composite ischaemia or major bleeding). Major bleeding definition included intracranial or intraocular bleeding, haemorrhage at the access site requiring intervention, haematoma > 5 cm in diameter, haemoglobin level drop > 4 g/dL without or > 3 g/dL with overt bleeding, reoperation for bleeding, or transfusion. Some elements of the definition such as haematoma > 5 cm in diameter were never used in any other definition of major bleeding and may have left room for potential biased assessment in a non-blinded trial.

Patients were randomized to one of three unblinded treatment groups: standard combination treatment with either UFH or LMWH with GP IIb/IIIa inhibitor (control arm) (n = 4603) or bivalirudin with GP IIb/IIIa inhibitor (n = 4604) or bivalirudin alone (n = 4612). A pre-specified non-inferiority hypothesis with a boundary of 1.25 was generated to compare the relative efficacy of the two test arms vs. the control arm as to the composite ischaemia endpoint. In the two arms with GP IIb/IIIa inhibitors, patients were randomized to receive GP IIb/IIIa inhibitors either upstream of the catherization laboratory, or after angiography if PCI were to be undertaken. The randomization was stratified for pre-treatment with clopidogrel, which was administered prior to PCI in 62.3% of patients. Coronary angiography was performed in 98.9%, PCI in 56.3%, and CABG in 11.1%, whereas 32.6% had no revascularization procedure. There was no significant difference between standard UFH/LMWH plus GP IIb/IIIa inhibitors, and the combination of bivalirudin and GP IIb/IIIa inhibitors for the composite ischaemia endpoint at 30 days (7.3 vs. 7.7%, respectively, RR 1.07, 95% CI 0.92–1.23, P = 0.39), or for major bleeding (5.7 vs. 5.3%, RR 0.93, 95% CI 0.78–1.10, P = 0.38). Bivalirudin alone was shown to be non-inferior to the standard UFH/LMWH combined with GP IIb/IIIa inhibitors as to the composite ischaemia endpoint (7.8 vs. 7.3%, RR 1.08, 95% CI 0.93–1.24, P = 0.32 for bivalirudin alone vs. UFH/LMWH plus GP IIb/IIIa inhibitors), but with a significantly lower rate of major bleeding (3.0 vs. 5.7%, RR 0.53, 95% CI 0.43–0.65, P < 0.001 for bivalirudin alone vs. UFH/LMWH plus GP IIb/IIIa inhibitors). Therefore, the rate of 30-day net clinical outcome was significantly lower (10.1 vs. 11.7%, RR 0.86 95% CI 0.77–0.94, P = 0.015) with bivalirudin alone vs. UFH/LMWH plus GP IIb/IIIa inhibitors. It has to be stressed that in ACUITY, about 41% of patients had no troponin elevation. Therefore, in the two groups where UFH/LMWH or bivalirudin were combined with GP IIb/IIIa inhibitors, the risk of bleeding may have been increased, without any beneficial effect on the ischaemic side in troponin-negative patients. This may have favoured the group bivalirudin alone. The treatment effects of bivalirudin monotherapy as regards net clinical outcome were consistent among most pre-specified subgroups, including patients who had positive tests for biomarkers, those undergoing PCI, those who were randomly assigned to immediate or deferred treatment with GP IIb/IIIa inhibitors, and those who did or did not undergo early angiography. However, in patients not pre-treated with clopidogrel prior to PCI, a significant excess of composite ischaemia endpoints was observed (9.1 vs. 7.1%, RR 1.29, 95% CI 1.03–1.63) for bivalirudin alone vs. UFH/LMWH plus GP IIb/IIIa inhibitors, with borderline significant interaction (P = 0.054) between pre-treatment with clopidogrel and the effect of bivalirudin alone. The lack of efficacy in patients not pre-treated with clopidogrel suggests that a potent antiplatelet therapy remains necessary for PCI. This is consistent with the results of ISAR-REACT 2 that show that GP IIb/IIIa inhibitors in addition to aspirin plus clopidogrel are beneficial in troponin-positive NSTE-ACS patients.¹⁸⁸ In retrospect, the margins of non-inferiority as regards the composite ischaemia endpoint might have been too liberal and did not meet the approach recommended by regulatory authorities.^189–197

The risk reduction for bleeding with bivalirudin alone compared with UFH/LMWH plus GP IIb/IIIa inhibitors was significant in all subgroups, except in patients submitted to CABG surgery. The reduction in bleeding with bivalirudin alone did not translate at 30-day follow-up into a risk reduction of death or ischaemic events as observed in OASIS-5.

5.2.5 Vitamin K antagonists

The VKAs produce their anticoagulant effect by interfering with the hepatic metabolism of vitamin K that results in the production of partially carboxylated and decarboxylated proteins with reduced coagulant activity. Their therapeutic effects are not observed until after 3–5 days of treatment. This treatment is therefore not useful in the acute phase of NSTE-ACS. In order to maintain anticoagulation efficacy without an excessive risk of bleeding, laboratory monitoring of the prothrombin time aiming for international normalized ratio (INR) levels 2.0–3.0 is required in the setting of MI.¹⁹⁸ The level of anticoagulation requires continuous monitoring as it is influenced by several food–drug and drug–drug interactions. VKA treatment, especially VKAs in combination with aspirin, was shown to be more effective than aspirin alone in the long-term prevention of death, re-MI, and stroke,¹⁹⁹ but at the cost of a higher risk of major bleeding.²⁰⁰ Better efficacy and safety of VKAs plus aspirin over aspirin alone can only be obtained if good compliance is achieved.²⁰¹ In the current era of combining aspirin with clopidogrel in NSTE-ACS, VKAs are mostly used in the presence of other indications for anticoagulation, such as atrial fibrillation, or after implantation of a mechanical heart valve.

The optimal antithrombotic therapy post-PCI in patients on ongoing anticoagulation therapy, for example because of atrial fibrillation, remains to be defined. Administration of aspirin and clopidogrel to patients already on VKA increases the risk of bleeding, whereas withholding antiplatelet therapy increases the risk of thrombotic events, particularly if a stent has been implanted. Discontinuation of VKAs in turn increases the risk of thrombo-embolic events. There are no data from randomized trials to clarify the optimum treatment in these patients and the feasibility of such studies is questionable. Hence, treatment decisions continue to be made on an individualized basis and should include information on key factors, including bleeding and thrombo-embolic risks. On the basis of experiences from clinical practice, it seems that antiplatelet and VKA combinations lead to only modest increases in bleeding risk in elderly patients, provided tight control of INR can be obtained^200,202 (see section 5.3.4 Resistance to antiplatelet agents/drug interactions for recommendations for association of a VKA and dual antiplatelet therapy). In patients with active VKA treatment presenting with ACS, initiation of the anticoagulants recommended during the acute phase (UFH, LMWH, fondaparinux, or bivalirudin) should be withheld as long as the INR is not known, and not started before the INR is < 2.0. Reversal of anticoagulation with vitamin K supplements is not recommended unless necessary for bleeding complications.

5.2.6 Anticoagulants during percutaneous coronary intervention procedures in non-ST-segment elevation acute coronary syndromes

The use of platelet inhibition with aspirin and systemic anticoagulation with UFH has been the standard of care for PCI from the beginning.¹⁸² The current recommendation, based on empiric evidence, is to give UFH as an intravenous bolus of 100 IU/kg or 50–60 IU/kg if GP IIb/IIIa inhibitors are given.¹⁸² The efficacy of UFH is monitored by ACT. However, the relationship between ACT and the rate of clinical events, and the real utility of ACT monitoring remain controversial.

Direct thrombin inhibition with bivalirudin and provisional GP IIb/IIIa inhibitor infusion has been shown to be at least as effective as and associated with a lower risk of bleeding than UFH/LMWH plus GP IIb/IIIa inhibitors in the setting of planned PCI.¹⁸¹ In addition, bivalirudin during PCI procedures was tested in comparison with UFH/LMWH or bivalirudin plus GP IIb/IIIa inhibitors in the ACUITY trial. As already mentioned, a significant risk reduction for bleeding was observed with bivalirudin alone when compared with UFH/LMWH or bivalirudin with GP IIb/IIIa inhibitors, but with a significantly higher rate of ischaemic events in patients not pre-treated with clopidogrel.¹⁸⁷

As most of the data and evidence in the setting of PCI have been collected so far with UFH and bivalirudin, including in NSTE-ACS, either drug can be recommended as first choice if the patient is directed immediately to the catherization laboratory because of life-threatening condition or refractory ischaemia and/or angina (see section 8 Management strategies).

LMWH have been used in the setting of PCI, but most of the data have been obtained with enoxaparin.^164,203 Until recently, because of the lack of the clinical studies, it was recommended to add UFH in enoxaparin-treated patients taken to the catherization laboratory for PCI.¹⁵² More recent data have shown that no additional UFH is needed if PCI is carried out within 6–8 h following the last subcutaneous dose of enoxaparin. After 6–8 h, an additional 0.3 mg/kg intravenous bolus of enoxaparin is recommended.²⁰⁴

Enoxaparin (1 mg/kg twice daily) was compared with UFH as antithrombotic agent in a PCI setting in 4687 NSTE-ACS in the SYNERGY trial. There was no difference in outcome during or after PCI, whatever the drug used in the catherization laboratory (UFH or enoxaparin). However, there was a strong trend towards an excess of bleeding (non-CABG-related TIMI major bleeds) with enoxaparin, when compared with UFH, possibly augmented by post-randomization crossover antithrombotic therapy.¹⁶⁴ A recent trial (STEEPLE) involving 3258 patients undergoing elective PCI suggests that lower doses of enoxaparin may be favourable with respect to bleeding.²⁰⁵

Enoxaparin and fondaparinux were used in the setting of PCI in 6239 patients in OASIS-5.¹⁷⁶ There was a significantly higher risk of vascular access site complications (8.1 vs. 3.3%, RR 0.41, 95% CI 0.33–0.51, P < 0.001) with enoxaparin than with fondaparinux. Catheter thrombus was significantly more common with fondaparinux (0.4 vs. 0.9%, RR 2.25, 95% CI 1.64–7.84, P = 0.001 for enoxaparin vs. fondaparinux). Per protocol in OASIS-5, patients taken to the catheterization laboratory more than 6 h after the last enoxaparin injection received an additional dose of UFH. In the fondaparinux group, no additional UFH was given. This may have generated a protective effect against catheter thrombus formation in the enoxaparin group. The excess of catheter thrombus had no impact on the rate of PCI-related coronary complications (8.6 vs. 9.5%, RR 1.11 95% CI 0.94–1.29, P = 0.21) for enoxaparin vs. fondaparinux and no significant impact on clinical events at 9 days. Peri-procedural complications, in terms of death, MI, stroke, and major bleeding, were significantly more frequent with enoxaparin compared with fondaparinux (20.6 vs. 16.6%, RR 0.81, 95% CI 0.73–0.90, P = 0.001) at 9 days as well as at 30 days (11.7 vs. 9.5%, RR 0.81, 95% CI 0.70–0.93, P = 0.004 for enoxaparin vs. fondaparinux).

Fondaparinux has also been used in the setting of PCI in ASPIRE and OASIS-6 in addition to OASIS-5.^166,174,176 Catheter thrombus was also observed in the setting of primary PCI in STEMI in the OASIS-6 study.¹⁶⁶ Catheter thrombus formation was reduced in OASIS-5 and eliminated in OASIS-6 by the administration of UFH upstream of or during PCI. Adding UFH to fondaparinux did not increase bleeding risk in OASIS-5, but a larger data set of patients is needed to confirm the safety of this association.²⁰⁶ Until new data are available, a standard dose of UFH (50–100 IU/kg bolus)¹⁸² is needed in addition to fondaparinux at the time of PCI, if fondaparinux was initiated prior to the procedure.

Recommendations for anticoagulation

• Anticoagulation is recommended for all patients in addition to antiplatelet therapy (I-A).

• Anticoagulation should be selected according to the risk of both ischaemic and bleeding events (I-B) (see also section 6.1 Bleeding complications, section 7.4 Chronic kidney disease, and section 7.5 Anaemia).

• Several anticoagulants are available, namely UFH, LMWH, fondaparinux, and bivalirudin. The choice depends on the initial strategy (see section 8 Management strategies: urgent invasive, early invasive, or conservative strategies) (I-B).

• In an urgent invasive strategy, UFH (I-C), enoxaparin (IIa-B), or bivalirudin (I-B) should be immediately started.

• In an non-urgent situation, as long as a decision between an early invasive or conservative strategy is pending (see section 8 Management strategies):

  • Fondaparinux is recommended on the basis of the most favourable efficacy/safety profile (I-A) (see sections 5.2.3 Factor-Xa inhibitors and 6.1 Bleeding complications).

  • Enoxaparin with a less favourable efficacy/safety profile than fondaparinux should be used only if the bleeding risk is low (IIa-B).

  • As the efficacy/safety profile of LMWH (other than enoxaparin) or UFH relative to fondaparinux is unknown, these anticoagulants cannot be recommended over fondaparinux (IIa-B).

  • At PCI procedures, the initial anticoagulant should also be maintained during the procedure regardless of whether this treatment is UFH (I-C), enoxaparin (IIa-B), or bivalirudin (I-B), whereas addititional UFH in standard dose (50–100 IU/kg bolus) is necessary in the case of fondaparinux (IIa-C).

  • Anticoagulation can be stopped within 24 h after an invasive procedure (IIa-C). In a conservative strategy, fondaparinux, enoxaparin, or other LMWH may be maintained up to hospital discharge (I-B).

5.3 Antiplatelet agents

Platelet activation plays a key pathophysiological role in NSTE-ACS. The time course of events following presentation with NSTE-ACS necessitates immediate treatment with antiplatelet therapy, once the diagnosis is made. Platelet activation needs to be seen not only in the context of an acute plaque rupture event, but also as a contributor to subsequent atherothrombotic events in patients with up-regulated inflammation in the arterial wall and systemic circulation. Thus, antiplatelet therapy is necessary for the acute event, and subsequent maintenance therapy. Three related but complementary strategies provide effective antiplatelet therapy: cyclooxygenase (COX)-1 inhibition (aspirin), inhibition of adenosine diphosphate (ADP)-mediated platelet aggregation with thienopyridines (ticlopidine and clopidogrel), and GP IIb/IIIa inhibition (tirofiban, eptifibatide, abciximab).

5.3.1 Acetylsalicylic acid (aspirin)

Aspirin irreversibly inhibits COX-1 in platelets, thereby limiting the formation of thromboxane A₂, thus inhibiting platelet aggregation. Three trials have consistently shown that aspirin decreases death or MI in patients with unstable angina.^147,207,208 In the meta-analysis of the Antithrombotic Trialists Collaboration, a 46% reduction in the rate of vascular events was evidenced.²⁰⁹ This meta-analysis suggested that 75–150 mg aspirin was as effective as higher doses. No robust relationship between dose and efficacy has been demonstrated. Initial doses of chewed, non-enteric aspirin from 160 to 325 mg are recommended to minimize delay before COX-1 inhibition occurs.²⁰⁹ In another meta-analysis, including four studies, the reduction in the rate of vascular events was 53% (Figure 6). Intravenous aspirin is an alternative mode of application, but has never been validated in trials.

The most common side effect of aspirin is gastrointestinal intolerance, reported in 5–40% of aspirin-treated patients. Gastrointestinal bleeding appears to increase with higher doses. In the CAPRIE study, the rate of gastrointestinal bleeding leading to aspirin discontinuation was 0.93%.²¹⁰ Hypersensitivity (‘allergy’) to aspirin is rare, but its prevalence depends on the clinical manifestation. Desensitization may be an option in selected patients.²¹¹ Most frequent is aspirin-exacerbated respiratory tract disease. Aspirin-induced rash or skin manifestations occur in 0.2–0.7% of the general population. More serious reactions such as anaphylactic shock are extremely rare.^212,213

In the CURE trial, aspirin was given in combination with clopidogrel at doses ranging from 75 to 325 mg.¹⁶⁷ The incidence of major bleeding increased as a function of the aspirin dose, both in patients treated with aspirin alone and in patients treated with a combination of aspirin and clopidogrel. The risk of bleeding was lowest with doses up to 100 mg of aspirin, and there was no evidence of improved efficacy with higher doses of aspirin.²¹⁴

5.3.2 Thienopyridines

Both ticlopidine and clopidogrel are ADP receptor antagonists, which block the ADP-induced pathway of platelet activation by specific inhibition of the P2Y₁₂ ADP receptor. Ticlopidine has been investigated in the setting of NSTE-ACS in only one study, in which a significant 46% risk reduction of death and MI at 6 months has been documented.²¹⁵ However, the use of ticlopidine has declined because of potentially serious side effects, in particular gastrointestinal, and risk of neutropenia or thrombocytopenia, and slower onset of action. As a result, ticlopidine was replaced by clopidogrel over time.

In the CURE trial, clopidogrel was administered for 9–12 months in addition to aspirin (75–325 mg) vs. aspirin alone in 12 562 patients suffering from NSTE-ACS. Patients received placebo or a loading dose of 300 mg clopidogrel followed by 75 mg daily, in addition to conventional therapy. A significant risk reduction for death from cardiovascular causes, non-fatal MI, or stroke was observed in the treatment arm (9.3 vs. 11.4%, RR 0.80, 95% CI 0.72–0.90, P < 0.001). The risk reduction was significant for MI, and there was a trend towards reduction of death and stroke. The risk reduction was consistent across all risk groups (low, moderate, or high) and among all subsets of patients (elderly, ST-deviation, with or without elevated cardiac biomarkers, diabetic patients). The benefit was obtained early, with a significant 34% risk reduction of cardiovascular death, MI, stroke, or severe ischaemia at 24 h in the clopidogrel group (1.4 vs. 2.1%, OR 0.66, 95% CI 0.51–0.86, P < 0.01) and throughout the 12 months of the study period.^167,216

In non-acute settings, two other mega-trials tested clopidogrel against aspirin²¹⁰ and clopidogrel plus aspirin against placebo plus aspirin.²¹⁷ In CAPRIE, clopidogrel 75 mg once daily was compared with 325 mg of aspirin once daily in a population of 19 185 patients suffering from documented atherosclerotic disease, manifested as either recent ischaemic stroke, recent MI, or symptomatic peripheral arterial disease. The mean follow-up was 23 months. A significant relative risk reduction of 8.7% in favour of clopidogrel was observed vs. aspirin (95% CI 0.3–16.5, P = 0.043). There was no significant difference in the rate of major bleeding, particularly intracranial or gastrointestinal haemorrhage.²¹⁰

The CHARISMA study included 15 603 patients, of whom 12 153 had documented cardiovascular disease, including 10.4% with prior MI. In this trial, clopidogrel was added to low-dose aspirin (75–160 mg daily). There was no difference with respect to the primary endpoint (cardiovascular death, MI, or stroke).²¹⁷ In the pre-specified subgroups of patients with documented cardiovascular disease, the composite endpoint was significantly reduced with dual antiplatelet therapy, when compared with aspirin (6.9 vs. 7.9%, RR 0.88, 95% CI 0.77–0.99, P = 0.046). There was a non-significant increase in the risk of GUSTO major bleeding complications (1.7 vs. 1.3%, RR 1.25, 95% CI 0.97–1.61, P = 0.09). Although clopidogrel cannot be recommended for long-term treatment for the full spectrum of patients investigated in the CHARISMA trial, subsidiary analyses suggest that among those with documented cardiovascular disease, benefits may outweigh risks.

Limited data are available about the combination of aspirin, clopidogrel, and GP IIb/IIIa inhibitors in the setting of NSTE-ACS. In the CURE study, the need for GP IIb/IIIa inhibitors was considered as a surrogate marker of anti-ischaemic efficacy. Overall, 5.9% of patients received GP IIb/IIIa inhibitors in conjunction with clopidogrel plus aspirin, when compared with 7.2% in the placebo group (RR 0.82, 95% CI 0.72–0.93, P = 0.003). No specific data have been reported about the safety and efficacy of this triple association. Triple antiplatelet therapy has been shown to be superior in preventing ischaemic events without compromising safety in post hoc analyses of PCI trials and in the ISAR-REACT-2 trial^188,218,219 (see section 5.3.3 Glycoprotein IIb/IIIa receptor inhibitors).

Newer P2Y₁₂ inhibitors with more potent receptor affinity and more rapid onset of action are currently under evaluation (e.g. prasugrel, cangrelor, AZD6140).

Risk of bleeding

In CURE, an increase in the rate of major bleeding was observed in clopidogrel-treated patients (3.7 vs. 2.7%, RR 1.38, 95% CI 1.13–1.67, P = 0.001), but with a non-significant increase in life-threatening and fatal bleeds.¹⁶⁷ Bleeding rates were higher in patients who underwent CABG, but this reached only borderline significance in 912 patients submitted to surgery less than 5 days after cessation of clopidogrel treatment (9.6 vs. 6.3%, RR 1.53, 95% CI 0.97–2.40, P = 0.06). For those treated more than 5 days after interruption of clopidogrel, there was no significant increase in bleeding.²²⁰ There is no known antidote to clopidogrel or other ADP receptor antagonists (see section 6.1 Bleeding complications).

However, in the entire cohort, the benefit of clopidogrel treatment, including among patients submitted to revascularization by both PCI and CABG, outweighed the risk of bleeding, since overall, treating 1000 patients resulted in 21 fewer cardiovascular deaths, MI, or stroke, at the cost of an excess of seven patients requiring transfusion, and a trend for four patients to experience life-threatening bleeds.²²⁰

Overall, the benefit of clopidogrel treatment outweighs the risk in all patients with NSTE-ACS, including those submitted to CABG. The excess bleeding risk in patients submitted to surgery may be attenuated or eliminated by stopping clopidogrel for 5 days before surgery. However, it has not been investigated whether this results in increased complication rates during washout.

Dose and timing of clopidogrel

A number of studies have employed higher loading doses of clopidogrel (usually 600 mg), and these have demonstrated more rapid inhibition of platelet aggregation than that achieved with 300 mg. However, no large-scale outcome clinical trials have tested higher doses of clopidogrel in the setting of NSTE-ACS. Nevertheless, experience in other clinical settings suggests that faster platelet inhibition with higher loading doses (≥600 mg) may be more effective in reducing clinical endpoints.^221–225 Definitive evidence of risk vs. benefit will be provided by ongoing large-scale clinical trials.

Pre-treatment of unselected patients with clopidogrel before angiography results in better outcome of PCI.^218,219,226 The approach of postponing clopidogrel administration until coronary anatomy is known in patients submitted to very early invasive angiography is not based on evidence. The potential advantage of this approach is to avoid increased bleeding risk in patients requiring immediate surgery. However, this situation is rare, and frequently surgery can be deferred for a few days. Therefore, postponing clopidogrel to after angiography cannot be recommended, because the highest rates of events are observed in the early phase of NSTE-ACS. In patients who cannot be given clopidogrel before PCI, GP IIb/IIIa inhibitors should be administered.

Recommendations for oral antiplatelet drugs (Table 6)

• Aspirin is recommended for all patients presenting with NSTE-ACS without contraindication at an initial loading dose of 160–325 mg (non-enteric) (I-A), and at a maintenance dose of 75–100 mg long-term (I-A).

• For all patients, an immediate 300 mg loading dose of clopidogrel is recommended, followed by 75 mg clopidogrel daily (I-A). Clopidogrel should be maintained for 12 months unless there is an excessive risk of bleeding (I-A).

• For all patients with contraindication to aspirin, clopidogrel should be given instead (I-B).

• In patients considered for an invasive procedure/PCI, a loading dose of 600 mg of clopidogrel may be used to achieve more rapid inhibition of platelet function (IIa-B).

• In patients pre-treated with clopidogrel who need to undergo CABG, surgery should be postponed for 5 days for clopidogrel withdrawal if clinically feasible (IIa-C).

5.3.3 Glycoprotein IIb/IIIa receptor inhibitors

Three GP IIb/IIIa inhibitors have been approved for clinical use, namely abciximab, eptifibatide, and tirofiban. They block the final common pathway of platelet activation by binding to fibrinogen and, under high shear conditions, to von Willebrand factor, and thus inhibiting the bridging between activated platelets. Abciximab is a monoclonal antibody fragment, eptifibatide is a cyclic peptide, and tirofiban a peptido-mimetic inhibitor. Clinical studies with oral GP IIb/IIIa inhibitors were stopped because of an excess of ischaemic events and/or excess bleeding.^227,228

The results obtained with the use GP IIb/IIIa inhibitors differed according to whether their use was associated with a conservative or an invasive strategy.

Glycoprotein IIb/IIIa inhibitors in a conservative strategy

All three GP IIb/IIIa inhibitors were tested in trials where an invasive strategy was not encouraged. A meta-analysis including 31 402 NSTE-ACS patients treated in clinical trials using GP IIb/IIIa inhibitors showed a 9% significant risk reduction for death and MI at 30 days with GP IIb/IIIa inhibitors (11.8 vs. 10.8%, OR 0.91, 95% CI 0.84–0.98, P = 0.015).²²⁹ This risk reduction was consistent across multiple subgroups and was evident particularly in high-risk patients, (diabetic patients, ST-segment depression, and troponin-positive patients) and in patients submitted to PCI during initial hospitalization. GP IIb/IIIa inhibitors had no effects in troponin-negative patients and in women. However, most of them were actually troponin negative,²²⁹ and women with troponin release derived the same benefit as men. The use of GP IIb/IIIa inhibitors was associated with an increase in major bleeding complications, but intracranial bleeding was not significantly increased²²⁹ (Figure 7).

Outcome as a function of the utilization of GP IIb/IIIa inhibitors in patients initially medically managed, and submitted to PCI, was explored in a further meta-analysis involving 29 570 patients.²³⁰ A 9% risk reduction overall was confirmed, but the benefit was non-significant in purely medically managed patients receiving GP IIb/IIIa inhibitors vs. placebo, with a rate of death and MI at 30 days of 9.3 vs. 9.7% (OR 0.95, 95% CI 0.86–1.04, P = 0.27). The only significant beneficial effect was observed when GP IIb/IIIa inhibitors were maintained during PCI (10.5 vs. 13.6%, OR 0.74, 95% CI 0.57–0.96, P = 0.02). These data confirm previous reports showing a risk reduction for ischaemic events in patients pre-treated with GP IIb/IIIa inhibitors before PCI.^231,232 In diabetic patients, a meta-analysis showed a highly significant risk reduction for death at 30 days with the use of GP IIb/IIIa inhibitors,²³³ particularly pronounced when submitted to PCI. This confirmed previous reports on this issue.^229,231,233

Abciximab

Abciximab was tested in the GUSTO-4-ACS trial.²³⁴ In this trial, an invasive strategy and revascularization were discouraged during the acute phase. Seven thousand patients on aspirin and UFH were randomized to one of three drug regimens: placebo, abciximab bolus plus 24 h infusion, or abciximab bolus plus 48 h infusion. No significant benefit was demonstrated for the two groups treated with abciximab, and an increased bleeding risk was observed. Thrombocytopenia (defined as platelet count < 50 000 µL⁻¹) was observed in 1.5% of patients receiving abciximab vs. 1% in the placebo group. Thus, abciximab cannot be recommended for patients with NSTE-ACS, unless in the context of PCI (discussed subsequently).

Eptifibatide

In the PURSUIT trial,²³⁵ 10 948 patients were enrolled and randomized into three arms. In addition to conventional therapy including aspirin and UFH, patients were randomized to placebo or two different regimens of eptifibatide infusion, after the same initial bolus. The lower dose of eptifibatide was dropped because of lack of efficacy. The comparison involved the high-dose eptifibatide regimen vs. placebo. A significant reduction of the 30-day composite endpoint of death or non-fatal MI was observed (14.2 vs. 15.7%, eptifibatide vs. placebo, P = 0.04). The benefit was maintained at 6 months. This benefit was obtained at the cost of an excess risk of TIMI major bleeding (10.6 vs. 9.1%, P = 0.02), but no excess of intracranial bleeding. Thrombocytopenia (defined as platelet count of < 100 000 mm⁻³ or nadir < 50% of baseline value) occurred at a similar rate in both arms (6.8% eptifibatide vs. 6.7% placebo). The rate of profound thrombocytopenia (defined as platelet count < 20 000 mm⁻³) was low in both treatment arms (0.2 vs. < 0.1%, eptifibatide vs. placebo, P = NS).

Tirofiban

Tirofiban has been tested in two separate randomized trials.^236,237 In the PRISM trial, 3231 patients presenting with NSTE-ACS were randomized to receive either tirofiban or UFH for 48 h. A significant reduction in the composite endpoint of death, MI, or refractory ischaemia was observed at 48 h and maintained at 30 days, but not thereafter (3.8 vs. 5.6%, RR 0.67, 95% CI 0.48–0.92, P = 0.01 at 48 h). The rate of thrombocytopenia (defined as platelet count < 90 000 mm⁻³) was significantly more frequent with tirofiban than with UFH (1.1 vs. 0.4%, P = 0.04).

In the PRISM-PLUS trial, 1915 patients at higher risk than in the PRISM trial were randomized to three different arms: tirofiban alone, tirofiban plus UFH, or UFH alone. The tirofiban-alone arm was stopped soon after the start of the trial, because of an excess of adverse events. A significant reduction of the risk of death, MI, and refractory ischaemia was obtained at 7 days (12.9 vs. 17.9%, RR 0.68, 95% CI 0.53–0.88, P = 0.004) and maintained at 30 days and 6 months in the tirofiban plus UFH group, when compared with UFH alone. Major bleeds (according to the TIMI criteria) were not statistically more frequent in the tirofiban group, despite a trend towards an increase (1.4 vs. 0.8%, P = 0.23).

Glycoprotein IIb/IIIa inhibitors in an invasive strategy

Consistent results have been obtained in three different meta-analyses exploring the impact of the use of GP IIb/IIIa inhibitors in the setting of PCI. Two meta-analyses showed that a significant risk reduction for death and MI at 30 days could be achieved when GP IIb/IIIa inhibitors were administered before taking patients to the catherization laboratory and maintained during PCI.^230,231 Kong et al.²³⁸ reported a significant risk reduction in 30-day mortality among a total of 20 186 patients (0.9 vs. 1.3%, OR 0.73, 95% CI 0.55–0.96, P = 0.024). Importantly, thienopyridines and stents were not routinely used in these trials.

Abciximab

Abciximab has been tested in three trials as an adjunct to PCI in the setting of ACS.^239–241 Altogether, 7290 patients were included in these three trials, which convincingly showed a significant reduction in the combination of death and MI or need for urgent revascularization at 30 days. Pooled data from these three trials showed a significant late mortality benefit (HR 0.71, 95% CI 0.57–0.89, P = 0.003).²⁴²

In CAPTURE, abciximab has also been tested in patients with NSTE-ACS with planned PCI pre-treated with abciximab for 24 h and maintained for 12 h. In this trial without routine use of stents and clopidogrel, abciximab significantly reduced the rate of death, MI, and need for urgent intervention for recurrent ischaemia when compared with placebo at 30 days (11.3 vs. 15.9%, P = 0.012).²⁴³ The benefit was restricted to patients with elevated TnT levels.⁷⁴

More recently, in ISAR-REACT-2, 2022 high-risk NSTE-ACS patients were randomized following pre-treatment with aspirin and 600 mg of clopidogrel to either abciximab or placebo.¹⁸⁸ There was a similar proportion of diabetic patients in each group (average 26.5%); 52% of patients had elevated troponins and 24.1% had previous MI. The 30-day composite endpoint of death, MI, or urgent target vessel revascularization (TVR) occurred significantly less frequently in abciximab-treated patients vs. placebo (8.9 vs. 11.9%, RR 0.75, 95% CI 0.58–0.97, P = 0.03). Most of the risk reduction generated by abciximab resulted from a reduction in the occurrence of death and MI. The effect was more pronounced in certain pre-specified subgroups, particularly troponin-positive patients (13.1 vs. 18.3%, RR 0.71, 95% CI 0.54–0.95, P = 0.02). The duration of pre-treatment with clopidogrel had no influence on outcome, and there was no detectable effect in troponin-negative patients or among diabetic patients. However, the number of diabetic patients included in this trial may have been too low to provide robust statistical power to detect any effect.

Abciximab was tested in a head-to-head comparison vs. tirofiban in the TARGET trial, in which two-thirds of the patients had recent or ongoing NSTE-ACS. Abciximab was shown to be superior to tirofiban in standard doses in reducing the risk of death, MI, and urgent revascularization at 30 days, but the difference was not significant at 6 months and 1 year.^244,245

Eptifibatide

Eptifibatide has been tested in patients undergoing PCI including 38% with unstable angina (IMPACT-2) and exhibited no significant benefit when compared with placebo.²⁴⁶ Subsequently, eptifibatide was tested in the ESPRIT trial, in which the dose was increased to a double bolus of 180 µg/kg followed by an infusion of 2.0 µg/kg/min for 18–24 h vs. placebo.²⁴⁷ In this trial, a significant reduction in the risk of death, MI, urgent TVR, and bail-out use of GP IIb/IIIa inhibitors was demonstrated at 48 h, maintained at 30 days and at 6 months (6.6 vs. 10.5%, RR 0.63, 95% CI 0.47–0.84, P = 0.0015 at 48 h for eptifibatide vs. placebo). The secondary composite endpoint of death, MI, or urgent TVR was also significantly reduced at the same time point (6.0 vs. 9.3%, RR 0.65, 95% CI 0.47–0.87, P = 0.0045). However, in this study, a smaller proportion (46%) of patients with recent or ongoing NSTE-ACS was included when compared with TARGET.

Tirofiban

Tirofiban was tested in the RESTORE trial, involving 2139 patients suffering from recent NSTE-ACS. In this trial, a significant 38% relative risk reduction in the primary composite endpoint of death, MI, repeat revascularization, or recurrent ischaemia at 48 h was observed at 7 days but not at 30 days.²⁴⁸ Tirofiban was used at the same dose in the TARGET and RESTORE trials. In retrospect, the dose may have been too low.

Further trials explored higher doses of tirofiban in various clinical settings. In a small trial of 202 patients, high-dose bolus (25 µg/kg) and infusion (0.15 µg/kg/min for 24–48 h) was shown to reduce the incidence of ischaemic thrombotic complications during high-risk PCI vs. placebo.²⁴⁹ TENACITY, a large-scale study testing high-dose tirofiban against abciximab, was stopped for financial reasons after inclusion of only 383 patients.

Use of glycoprotein IIb/IIIa inhibitors prior to revascularization

Glycoprotein IIb/IIIa inhibitors and percutaneous coronary intervention

Two GP IIb/IIIa inhibitors (tirofiban and eptifibatide) have shown efficacy in reducing ischaemic events in NSTE-ACS, particularly in high-risk subgroups, such as troponin-positive or diabetic patients, and in patients submitted to revascularization.^73,235 Therefore, they may be used as first-line treatment in addition to other antithrombotic agents, before invasive evaluation of the patient is undertaken. This so-called ‘upstream’ use of GP IIb/IIIa inhibitors prior to revascularization has been shown in meta-analyses to further reduce the risk of death and MI at 30 days, if GP IIb/IIIa inhibitors are prescribed upstream of and maintained during the PCI procedure.^230,231 However, this question will be further explored in upcoming trials (EARLY-ACS).²⁵⁰

In ACUITY-TIMING, deferred selective vs. routine upstream administration of GP IIb/IIIa inhibitors was tested in a 2 × 2 factorial design. GP IIb/IIIa inhibitors were used in 55.7% of patients for 13.1 h in the deferred selective and in 98.3% of patients for 18.3 h in the routine upstream strategy.²⁵¹ Deferred selective vs. routine upstream strategy resulted in a reduced 30-day major bleeding rate (4.9 vs. 6.1%, RR 0.80, 95% CI 0.67–0.95) but the rate of ischaemic events did not meet the criteria for non-inferiority, with a trend towards a higher rate (7.9 vs. 7.1%, RR 1.12, 95% CI 0.97–1.29; P = 0.13 deferred selective vs. routine upstream strategy). TIMI major bleeding rate was not significantly different in both groups (1.6 vs. 1.9%, P = 0.20) for deferred selective vs. routine upstream, whereas TIMI minor bleeding rate was significantly lower (5.4 vs. 7.1%, P < 0.001) for deferred selective vs. routine upstream. Ischaemic composite endpoint was achieved significantly more frequently in patients submitted to PCI with deferred selective vs. routine upstream GP IIb/IIIa inhibitors (9.5 vs. 8.0%, RR = 1.19, 95% CI 1.00–1.42, P = 0.05). Considering these results, it can be concluded that more frequent and more prolonged use of GP IIb/IIIa inhibitors with upstream treatment leads to an excess risk of major bleeding, but with potentially greater protection against ischaemic events in patients submitted to PCI.

In routine practice, as shown by several registries,^252,253 patients are often taken to the catherization laboratory without prior infusion of GP IIb/IIIa inhibitors. In this setting, if there is a need for immediate PCI, the administration of GP IIb/IIIa inhibitors in the catherization laboratory is recommended on the basis of ISAR-REACT-2 as an alternative strategy, which, however, has no proven superiority when compared with upstream administration.

Glycoprotein IIb/IIIa inhibitors and coronary artery bypass graft

Inhibition of platelet aggregation may result in bleeding complications, either spontaneously or at the time of cardiac surgery. However, surgery in patients receiving GP IIb/IIIa inhibitors has been shown to be safe when appropriate measures are taken to ensure adequate haemostasis. GP IIb/IIIa inhibitors should be discontinued at the time of cardiac surgery. Eptifibatide and tirofiban have a short half-life, so platelet function should recover by the end of CABG. Abciximab has a longer effective half-life, and earlier discontinuation may be needed. If excessive bleeding occurs, fresh platelet transfusions may be administered (see section 6.1 Bleeding complications). Fibrinogen supplementation with fresh–frozen plasma or cryoprecipitate either alone or in combination with platelet transfusion can also be considered for restoring haemostatic potential and managing major haemorrhagic complications associated with the administration of small-molecule, GP IIb/IIIa inhibitors.²⁵⁴

Adjunctive therapy

All trials carried out with GP IIb/IIIa inhibitors used UFH. Nowadays, the use of LMWH, particularly enoxaparin, is more widespread. Several trials in the field of NSTE-ACS, as well as observational studies in PCI, have shown that LMWH, predominantly enoxaparin, can be safely used with GP IIb/IIIa inhibitors without compromising efficacy.^{162–172,255,256} In OASIS-5, GP IIb/IIIa inhibitors were used with aspirin, clopidogrel, and fondaparinux in 1308 patients or enoxaparin in 1273 patients. Overall, bleeding complications were lower with fondaparinux than with enoxaparin (see section 5.2 Anticoagulants).

Thienopyridines were not used in earlier trials testing GP IIb/IIIa inhibitors. Therefore, the efficacy and safety of a triple association of aspirin, clopidogrel, and GP IIb/IIIa inhibitors was not clearly defined. Recently, the ISAR-REACT-2 study has confirmed that better outcome can be obtained with abciximab added to pre-treatment with aspirin and a 600 mg loading dose of clopidogrel, when compared with a dual regimen with only aspirin plus clopidogrel in the setting of high-risk PCI patients with NSTE-ACS.¹⁸⁸ Other trials testing the same hypothesis are ongoing (e.g. Early-ACS).²⁵⁰

Bivalirudin and UFH/LMWH were shown to have equivalent safety and efficacy when used with triple antiplatelet therapy, including GP IIb/IIIa inhibitors in the ACUITY trial. However, bivalirudin alone was associated with a lower bleeding risk when compared with any combination with GP IIb/IIIa inhibitors.²⁵⁷

Recommendations for glycoprotein IIb/IIIa inhibitors (Table 6)

• In patients at intermediate to high risk, particularly patients with elevated troponins, ST-depression, or diabetes, either eptifibatide or tirofiban for initial early treatment is recommended in addition to oral antiplatelet agents (IIa-A).

• The choice of combination of antiplatelet agents and anticoagulants should be made in relation to risk of ischaemic and bleeding events (I-B).

• Patients who receive initial treatment with eptifibatide or tirofiban prior to angiography should be maintained on the same drug during and after PCI (IIa-B).

• In high-risk patients not pre-treated with GP IIb/IIIa inhibitors and proceeding to PCI, abciximab is recommended immediately following angiography (I-A). The use of eptifibatide or tirofiban in this setting is less well established (IIa-B).

• GP IIb/IIIa inhibitors must be combined with an anticoagulant (I-A).

• Bivalirudin may be used as an alternative to GP IIb/IIIa inhibitors plus UFH/LMWH (IIa-B).

• When anatomy is known and PCI planned to be performed within 24 h with GP IIb/IIIa inhibitors, most secure evidence is for abciximab (IIa-B).

5.3.4 Resistance to antiplatelet agents/drug interactions

Resistance to antiplatelet agents describes partial or total failure of an antiplatelet agent to achieve the expected inhibition of platelet function and would therefore be better named low or hypo-responsiveness. The term refers to the variability in the magnitude of platelet aggregation inhibition measured ex vivo achieved in a population of treated patients. Resistance to antiplatelet agents is often confused with recurrence of events despite antiplatelet treatment. This does not necessarily imply that antiplatelet resistance is the causal phenomenon, since atherothrombosis is multifactorial, and the recurrence of events may be due to a mechanism other than resistance to treatment. Resistance to antiplatelet agents may be assessed by various tests of platelet function. The magnitude of true resistance to antiplatelet agents remains poorly defined. No simple test has been reliably validated to assess the level of platelet function inhibition for any antiplatelet agent used in atherothrombosis.^258,259

In addition, drug interactions may result in partial or total inhibition of the activity of a given compound. As the prescription list for patients suffering from NSTE-ACS at the acute phase and for the long term involves several different pharmacological classes, caution has to be exerted before deciding on a polypharmacy prescription. Only few interactions have been reported for the drugs used in this setting. The most serious interactions have recently been suspected with non-steroidal anti-inflammatory drugs (NSAIDs) and aspirin (discussed subsequently).

Aspirin resistance/drug interactions

Resistance to aspirin describes a range of phenomena, including inability to protect individuals from thrombotic complications, inability to cause prolongation of the bleeding time, inability to reduce thromboxane A₂ production, and failure to produce an anticipated effect on one or more in vitro tests of platelet function, including in particular aggregometry, shear stress-induced activation, and expression of platelet surface receptors.²⁵⁹ A proportion of patients treated for any clinical manifestation of atherothrombosis, CAD, cerebrovascular disease, or peripheral artery disease may develop treatment failure over time, even with increasing doses.^260–266

However, few studies have shown that aspirin resistance could lead to treatment failure. One HOPE substudy showed that different degrees of thromboxane A₂ inhibition were associated with a significant difference in rate of events. However, these differences may reflect different levels of treatment compliance.^267,268

There are at least three potential mechanisms that may lead to aspirin resistance, namely transient expression of COX-2 in newly formed platelets,²⁶⁹ extra-platelet sources of thromboxane A₂, ²⁷⁰ and interference with NSAIDs. Concomitant administration of NSAIDs, such as ibuprofen, may interfere with the inactivation of COX-1 due to a competitive action on the docking site of aspirin in the COX channel.²⁷¹ This possible interaction is not observed with selective COX-2 inhibitors or other anti-inflammatory drugs such as diclofenac. Nevertheless, in some reports, it has been shown that more events occur in patients treated with this association.^258,272,273 Recently, a retrospective analysis of a large cohort of patients discharged from hospital after MI showed that the use of selective COX-2 inhibitors and non-selective NSAIDs in the post-MI period led to a higher risk of death, with any of these agents.²⁷⁴ There was also an increased risk of hospitalization for MI with the use of selective COX-2 inhibitors and NSAIDs. This and other studies indicate that anti-inflammatory drugs should be avoided in the post-MI period, whatever the mechanism for the increased risk of death and MI may be.

Clopidogrel resistance/drug interactions

Clopidogrel is an inactive compound, which needs oxidation by hepatic cytochrome P450 to generate an active metabolite. CYP3A4 and CYP3A5 are the P450 isoforms responsible for the oxidation of clopidogrel, which, through a multistage metabolic degradation, leads to the active form of the drug. The standard dose of clopidogrel achieves, through antagonism of the P2Y₁₂ platelet ADP receptor, approximately 30–50% inhibition of ADP-induced platelet aggregation.²⁷⁵

Clopidogrel resistance is an inappropriate terminology which refers actually to the variability in the clopidogrel-induced inhibition of platelet aggregation. Light transmission aggregometry is the most frequently used test to measure platelet inhibition. However, there is no unanimity on how to define clopidogrel resistance with regard to the cut-off value. With these restrictions, clopidogrel resistance was shown to occur in 4–30% of patients.^258,275 The mechanisms of clopidogrel resistance are currently under investigation. Despite small studies that have shown a higher rate of events associated with low inhibition of platelet aggregation, there is little tangible evidence that clopidogrel resistance results in treatment failure.^276–278 There are still efforts to overcome this problem by raising and/or tailoring the clopidogrel dose. Newer ADP receptor antagonists (e.g. prasugrel, cangrelor, and AZD6140) are currently under clinical investigation.

In some instances, reduced bioavailability through drug interactions has been elicited, particularly with some statins which are metabolized by CYP3A4 and CYP3A5. They have been shown in in vitro studies to limit by 90% the degradation of clopidogrel in its active metabolite form.^258,275 However, in clinical practice, this has not been translated into any demonstrable negative effect.²⁷⁹ Indeed, in the GRACE registry, the association of clopidogrel and statins is suggestive of an additive beneficial effect on outcome.²⁸⁰

In vitro, clopidogrel metabolites can inhibit enzymatic activity of cytochrome P4502C9 and lead to increased plasma levels of NSAIDs, which are metabolized by this cytochrome. This could lead to an increased risk of gastrointestinal bleeding in the case of concomitant administration of clopidogrel and NSAIDs (particularly naproxen).²⁸¹

Lastly, the association of clopidogrel with a VKA is not recommended, since it may potentially increase the risk of bleeding. This combination can, however, be necessary in the context of mechanical heart valve or in the case of high risk of thrombo-embolic events, where VKAs cannot be interrupted and clopidogrel is mandatory. In these cases, the lowest efficacious INR and shortest duration for the prescription should be targeted. Strict control of the INR is necessary.

Recommendations for resistance to antiplatelet treatment/drug interactions

• Routine assessment of platelet aggregation inhibition in patients submitted to either aspirin or clopidogrel therapy, or both, is not recommended (IIb-C).

• NSAIDs (selective COX-2 inhibitors and non-selective NSAIDs) should not be administered in combination with either aspirin or clopidogrel (III-C).

• Clopidogrel can be administered with all statins (I-B).

• The triple association of aspirin, clopidogrel, and a VKA should only be given if a compelling indication exists, in which case, the lowest efficacious INR and shortest duration for the triple association should be targeted (IIa-C).

5.3.5 Withdrawal of antiplatelet agents

Some recent reports have shown that, in patients with CAD, withdrawal of antiplatelet agents, whatever the reason, may lead to an increased rate of recurrence of events.²⁸² In a recently published multicentre prospective cohort of 1521 patients with recent MI, 184 patients discontinued all three recommended medications (aspirin, beta-blockers, and statins), 56 two medications, and 272 only one medication during a 12-month follow-up period. Patients who discontinued all three medications had the lowest 12-month survival rate (88.5 vs. 97.7%; log-rank P < 0.001) compared with patients who maintained at least one medication. In multivariate analysis, medication discontinuation was independently associated with higher mortality rates (HR 3.81; 95% CI 1.88–7.72). Results were consistent when evaluating aspirin, beta-blockers, and statins separately.²⁸³

Interruption of dual antiplatelet therapy soon after stent implantation increases the risk of acute stent thrombosis, which carries a particularly adverse prognosis, with mortality varying from 15 to 45% at 1 month.^284–286 In addition, interruption of antiplatelet agents long after implantation of drug-eluting stents (DES) may expose the patient to late stent thrombosis.^{285,287–289} Similarly, interruption of dual antiplatelet treatment soon after the acute phase of NSTE-ACS may expose the patients to a high risk of recurrence of events even in non-stented patients, though only few data are available to support this notion. However, interruption of dual antiplatelet therapy in the case of a necessary surgical procedure more than 1 month after ACS in patients without DES may be reasonable.

If interruption of dual antiplatelet therapy becomes mandatory, such as need for urgent surgery or major bleeding that cannot be controlled by local treatment, no proven efficacy alternative therapy can be proposed as a substitute. Different alternatives to dual antiplatelet treatment have been proposed depending on the clinical setting, the type of stent and date of implantation, and type of surgery. None of them was formerly proven efficacious, and all are based on experts' consensus opinion. LMWH have been advocated without tangible proof of efficacy.^290,291

Recommendations for withdrawal of antiplatelet treatment

• Temporary interruption of dual antiplatelet therapy (aspirin and clopidogrel) within the first 12 months after the initial episode is discouraged (I-C).

• Temporary interruption for major or life-threatening bleeding or for surgical procedures where even minor bleeding may result in severe consequences (brain or spinal surgery) is mandatory (IIa-C).

• Prolonged or permanent withdrawal of aspirin, clopidogrel, or both is discouraged unless clinically indicated. Consideration should be given to the risk of recurrence of ischaemic events which depends (among other factors) on initial risk, on the presence and type of stent implanted, and on the time window between proposed withdrawal and the index event and/or revascularization (I-C).

5.4 Coronary revascularization

Revascularization for NSTE-ACS is performed to relieve angina and ongoing myocardial ischaemia and to prevent progression to MI or death. The indications for myocardial revascularization and the preferred approach (PCI or CABG) depend on the extent and severity of the lesions as identified by coronary angiography, the patient's condition, and co-morbidity.

5.4.1 Coronary angiography

Invasive coronary angiography remains pivotal in determining suitability for percutaneous and/or surgical revascularization. It is recommended to perform angiograms after intracoronary administration of vasodilators (nitrates) in order to attenuate vasoconstriction and offset the dynamic component that is frequently present in ACS.²⁹² In haemodynamically compromised patients (pulmonary oedema, hypotension, severe life-threatening arrhythmias), it may be advisable to perform the examination after placement of an intra-aortic balloon pump, to limit the number of coronary injections and to omit LV angiography.

Data from TIMI-3B²⁹³ and FRISC-2²⁹⁴ show that 30–38% of the patients with unstable coronary syndromes have single-vessel disease and 44–59% have multivessel disease (>50% diameter stenosis). The incidence of left main narrowing varies from 4 to 8%. Although stenosis severity is usually well determined by angiography, sometimes intravascular ultrasound may be useful.²⁹⁵ Coronary angiography in conjunction with ECG findings and wall motion abnormalities frequently allows identification of the culprit stenosis that often shows eccentricity, irregular borders, ulceration, haziness, and filling defects suggestive of the presence of intracoronary thrombus.²⁹⁶ However, sometimes it may be difficult to determine the culprit lesion because the above-mentioned features are either present in more than one vessel or absent. Diffuse atherosclerotic infiltration without significant narrowing is found in 14–19% of cases.²⁵³ A number of new investigational invasive diagnostic tools are currently being tested for their ability to identify the presence of one or more vulnerable segments, monitor changes that occur either spontaneously or under medical treatment, and relate those markers of plaque vulnerability to patient outcome.^297,298 Focal accumulation of specific plaque components such as lipid necrotic core and weakening of the fibrous cap are associated with instability.^299,300 At this point in time, it remains undetermined whether non-culprit coronary segments showing vulnerable features will merit mechanical intervention.³⁰¹

5.4.2 Invasive vs. conservative strategy

Choice of strategy

Coronary angiography should be planned as soon as possible (urgent invasive strategy) in patients with severe ongoing angina, profound or dynamic ECG changes, major arrhythmias, or haemodynamic instability upon admission or thereafter. These patients represent 2–15% of the patients admitted with NSTE-ACS.^302–304 In patients with intermediate to high-risk features, but without the aforementioned life-threatening features, early coronary angiography (within 72 h) followed by revascularization when possible and indicated, or initial medical stabilization and selective performance of coronary angiography based on the clinical course have been tested as alternative strategies. In low-risk patients, a non-invasive assessment of inducible ischaemia should be performed prior to discharge. If this is positive, coronary angiography should be performed³⁰⁵ (see section 8 Management strategies).

A meta-analysis of seven randomized trials (including early studies prior to the widespread use of stents and multidrug adjunctive therapy) comparing routine angiography (n = 4608) followed by revascularization with a more conservative strategy (invasive care only in patients with recurrent or inducible ischaemia, n = 4604) showed reduced rate of death and MI at the end of follow-up (12.2 vs. 14.4%, OR 0.82, 95% CI 0.72–0.93, P = 0.001) for routine invasive vs. selective invasive.³⁰⁶ At the same time point, there was a non-significant trend towards fewer deaths (5.5 vs. 6.0%, OR 0.92, 95% CI 0.77–1.09), whereas a significant reduction in MI alone was observed (7.3 vs. 9.4%, OR 0.72, 95% CI 0.65–0.88, P < 0.001) for routine invasive vs. selective invasive. These results were obtained despite an early hazard observed during initial hospitalization in the routine invasive group, where a significantly higher risk of death, and death and MI was noted (1.8 vs. 1.1%, OR 1.6, 95% CI 1.14–2.25, P = 0.007 for death; 5.2 vs. 3.8%, OR 1.36, 95% CI 1.12–1.66, P = 0.002 for death and MI) for routine invasive vs. selective invasive. The beneficial effect was actually achieved from hospital discharge to the end of follow-up, where a significant risk reduction in death and death and MI was observed (3.8 vs. 4.9%, OR 0.76, 95% CI 0.62–0.94, P = 0.01 for death; 7.4 vs. 11.0%, OR 0.64, 95% CI 0.55–0.75, P < 0.001 for death and MI) routine invasive vs. selective invasive. Over a mean follow-up of 17 months, recurrent angina was reduced by 33% and rehospitalization by 34% in the routine invasive group. In another meta-analysis including six contemporary trials, the OR was 0.84, 95% CI 0.73–0.97 for early invasive vs. conservative strategy (Figure 8). The benefit of the routine invasive strategy was present in patients with elevated troponins at baseline, but not in troponin-negative patients (from the analysis of the three most recently performed trials with available troponin data).^122,307,308 A more recent meta-analysis including seven trials with 8375 patients available for analysis showed after a mean follow-up of 2 years a significant risk reduction for all-cause mortality (4.9 vs. 6.5%, RR 0.75, 95% CI 0.63–0.90, P = 0.001) for early invasive vs. conservative, without excess of death at 1 month (RR = 0.82, 95% CI 0.50–1.34, P = 0.43). At 2 years of follow-up, the incidence of non-fatal MI was 7.6 vs. 9.1% (RR = 0.83, 95% CI 0.72–0.96, P = 0.012), without excess at 1 month (RR = 0.93, 95% CI 0.73–1.19, P = 0.57.³⁰⁹ Long-term mortality reduction has been confirmed in the follow-up of RITA-3 at 5 years³¹⁰ and FRISC-2 at 2 and 5 years.^122,308 Many of the trials analysed in the meta-analysis by Mehta et al.³⁰⁶ were not contemporary. In four of the trials, namely TIMI-3B, VANQWISH, MATE, and FRISC-2, the use of stents and GP IIb/IIIa inhibitors was low or non-existent.^293,311,312 More recently, a review of the most contemporary trials by the Cochrane collaboration confirmed the initial observations reported by Mehta et al. This meta-analysis confirmed the existence of a trend towards an early excess of mortality with an invasive strategy (RR 1.59, 95% CI 0.96–2.54), but with a significant long-term benefit in terms of death (RR 0.75, 95% CI 0.62–0.92) or MI (RR 0.75, 95% CI 0.62–0.91) with invasive vs. conservative at 2–5-year follow-up.³¹³ The recently published ICTUS trial was not included in this meta-analysis, although its results challenge the paradigm of superior outcome with routine invasive strategy.³¹⁴ In this trial, 1200 patients were randomized to an early invasive strategy vs. a more conservative (selective) approach. There was no difference in the incidence of the primary composite endpoint of death, MI, or rehospitalization for angina within 1 year (22.7 vs. 21.2%, RR 1.07, 95% CI 0.87–1.33, P = 0.33) with early vs. selective invasive strategy. These results were maintained at 3-year follow-up³¹⁵ In keeping with previous studies, routine intervention was associated with a significant early hazard. MI was significantly more frequent in the early invasive group (15.0 vs. 10.0%, RR 1.5, 95% CI 1.10–2.04, P = 0.005). The majority (67%) of MI [defined as ≥ 1–3 times the upper limits of normal (ULN) CK-MB] was indeed associated with revascularization procedures. The discrepancy between this and previous trials could be attributed in part to the small difference in revascularization rates between the two study groups and the high overall rate of revascularization before discharge (76% in the routine invasive and 40% in the selective group). In addition, the criterion for diagnosis of MI (any CK-MB elevation above ULN as opposed to more than three times ULN) differs between studies. Furthermore, the selection of patients may have been biased, as some studies included all consecutive patients admitted while others did not enter severely unstable patients.

In all randomized trials, a large proportion of patients in the conservative arm eventually underwent revascularization (‘crossover’) such that the true benefit of revascularization is underestimated.³¹⁶ When comparing the relative mortality benefit between routine and selective revascularization strategies with the actual difference in the revascularization rates between arms, a linear relationship emerges: the greater the difference in the rate of revascularization, the greater the benefit on mortality (Figure 9).

Timing of invasiveness

With the exception of indications for emergency angiography and revascularization, controversy remains as to the optimal timing between hospital admission, initiation of medical therapy, and invasive evaluation. In 410 consecutive, high-risk patients with either ST-segment depression (65%) or elevated cTnT (67%) enrolled in the ISAR-COOL trial, deferral of intervention did not improve outcome.³¹⁷ On the contrary, patients randomized to immediate PCI (on average 2.4 h after admission) had a lower incidence of death or MI at 30 days than patients randomized to deferred PCI (86 h after admission and medical therapy) (5.9 vs. 11.6%, RR 1.96, 95% CI 1.01–3.82, P = 0.04). Likewise, no early hazard was observed in TACTICS-TIMI-18 (mean delay for PCI was 22 h) with upstream treatment with GP IIb/IIIa inhibitors.⁷³

At variance with these findings, early routine invasive care in the ICTUS trial within 48 h of randomization in 56% and during initial hospitalization in 76% of cases was associated with an excess of MI (15.0 vs. 10.0%, RR 1.5, 95% CI 1.10–2.04, P = 0.005). Expedited catheterization was also associated with worse outcome in FRISC-2 as well as in the GRACE and CRUSADE registry.^318–320

Accordingly, currently available evidence does not mandate a systematic approach of immediate angiography in NSTE-ACS patients stabilized with a contemporary pharmacological approach. Likewise, routine practice of immediate transfer of stabilized patients admitted in hospitals without onsite catheterization facilities is not mandatory, but should be organized within 72 h.

5.4.3 Percutaneous coronary intervention

Outcome after PCI in NSTE-ACS has been markedly improved with the use of intracoronary stenting and contemporary antithrombotic and antiplatelet therapy. The risk of bleeding complications should be balanced against the severity of ischaemia and the patient's risk profile. The choice of access site depends on operator expertise and local preference. Non-pharmacological strategies to reduce access site bleeding complications include the use of closure devices and the radial approach. The femoral approach is preferred in haemodynamically compromised patients to permit the use of intra-aortic balloon counterpulsation. As for all patients undergoing PCI, stent implantation in this setting helps to reduce the threat of abrupt closure and restenosis. The safety and efficacy of DES has not been prospectively tested in this specific population, although patients with NSTE-ACS represent up to 50% of patients included in most PCI trials. The approved DES appear to be equally effective in reducing restenosis in this setting as shown from subgroup analyses of randomized trials and real-world data.³²¹ Although the incidence of (sub)acute stent thrombosis is higher in NSTE-ACS patients, when compared with stable patients undergoing PCI, the use of DES does not seem to portend a higher risk of (sub)acute stent thrombosis in this specific setting.²⁸⁷ In view of the potentially severe consequences of acute or subacute stent thrombosis, it is advisable to use a bare metal stent (BMS) in patients scheduled to undergo extra-cardiac interventions or surgery that will require interruption of clopidogrel within the first year after stent implantation.^322,323 This strategy should also be considered in patients requiring long-term VKA treatment. In addition, doubts have been cast on the risk of stent thrombosis and long-term safety of DES in terms of risk of death and MI, particularly when used off-label in complex situations.³²⁴ Recent data suggest that dual antiplatelet therapy should be maintained for 1 year in the case of DES implantation irrespective of the active drug (sirolimus or paclitaxel).^325,326 As long as the concerns have not been addressed and the situation has not been completely clarified, the choice between the use of BMS or DES should be based on an individual assessment of benefit vs. potential risk.^325,326

The main issue with PCI for NSTE-ACS remains the relatively high incidence of peri-procedural MI, up to 10% in the ICTUS trial.³¹⁴ The use of antiplatelet therapy has significantly reduced the incidence of peri-procedural MI.³²⁷ However, embolization of debris or plaque fragments cannot be entirely prevented by state-of-the-art antithrombotic and antiplatelet adjunctive therapy.³²⁸ A wide variety of filter and/or distal protection devices have been tested but failed to improve clinical outcome, with the exception of the subset of saphenous vein graft interventions.³²⁹

Currently, there are no outcome data supporting routine PCI in non-significant culprit or non-culprit coronary obstructions, as perceived by angiography, even with the use of DES (‘plaque sealing’).³⁰¹

5.4.4 Coronary artery bypass graft

The proportion of patients with NSTE-ACS undergoing bypass surgery during initial hospitalization is about 10%.³¹⁴ It is important to consider the risk of bleeding complications in patients who undergo bypass surgery, although initially treated with aggressive antiplatelet treatment.^330,331 Overall, pre-treatment with a triple or even dual antiplatelet regimen should be considered as only a relative contraindication to early bypass surgery but does require specific surgical measures to minimize bleeding and platelet transfusions (see sections 5.3.3 Glycoprotein IIb/IIIa receptor inhibitors and 6.2 Thrombocytopenia).

5.4.5 Respective indications for percutaneous coronary intervention or coronary artery bypass graft

With the exception of an urgent procedure, the choice of revascularization technique in NSTE-ACS is the same as for elective revascularization procedures. From the randomized controlled trials comparing multivessel-stented PCI with bypass surgery, there was no interaction between the presence of NSTE-ACS, treatment strategy, and outcome.^331,332 In patients with multivessel disease, all significant stenoses can be treated at once. A staged procedure may be considered, with immediate PCI of the culprit lesion and subsequent reassessment of the need for treatment of other lesions.

Recommendations for invasive evaluation and revascularization (see also section 8 Management strategies).

• Urgent coronary angiography is recommended in patients with refractory or recurrent angina associated with dynamic ST-deviation, heart failure, life-threatening arrhythmias, or haemodynamic instability (I-C).

• Early (<72 h) coronary angiography followed by revascularization (PCI or CABG) in patients with intermediate to high-risk features is recommended (I-A).

• Routine invasive evaluation of patients without intermediate to high-risk features is not recommended (III-C), but non-invasive assessment of inducible ischaemia is advised (I-C).

• PCI of non-significant lesions is not recommended (III-C).

• After critical evaluation of the risk–benefit ratio, and depending on known co-morbidities and potential need for non-cardiac surgery in the short/medium term (e.g. planned intervention or other conditions) requiring temporary withdrawal of dual antiplatelet therapy, consideration should be given to the type of stent to be implanted (BMS or DES) (I-C).

5.5 Long-term management

Patients with NSTE-ACS after the initial phase carry a high risk of recurrence of ischaemic events. Therefore, active secondary prevention is an essential element of long-term management. Several measures and therapies have been proven to be effective in reducing the risk of recurrence of events after NSTE-ACS either in clinical randomized trials or in observational studies and registries. However, several registries have shown that these lifestyle measures and drug therapies are underused. The role of the physician is to make sure that NSTE-ACS patients receive the appropriate therapy and lifestyle counselling in order to improve long-term outcome. It is beyond the scope of this document to review in detail all the measures and treatments that should be implemented for secondary prevention, but emphasis will be placed on those of paramount importance. Detailed recommendations on secondary prevention have been extensively described in other guidelines.^333–335

5.5.1 Lifestyle

Several lifestyle interventions, described extensively in other reports, have been proven effective in reducing long-term risk of recurrence of events in patients with CAD, including NSTE-ACS.^333–336

Smoking cessation is difficult to achieve in the long-term. Smoking resumption is frequent. Active counselling, in addition to adjunctive drug interventions, such as nicotine replacement and bupropione, is necessary.^333–335

Regular physical activity must be encouraged. Thirty minutes of moderate intensity aerobic activity, if possible every day, or at least five times per week, is recommended. A medically supervised programme for high-risk patients may be necessary.^333–335

Healthy diet based on low salt intake with reduced intake of saturated fats is essential. Encourage regular intake of fruit and vegetables. Moderate alcohol consumption may be beneficial.³³⁷

5.5.2 Weight reduction

Weight reduction in obese and overweight patients has to be encouraged. Return to physical activity can help to facilitate weight loss. Significant weight reduction is difficult to achieve, and, to date, no pharmacotherapy can firmly be recommended, although some specific drugs interacting with the endocannabinoid system have been shown to lead to sustained weight loss with minimal side effects.^338,339 Weight reduction has a favourable impact on lipid profile and glycaemic control. The theoretical goal is to achieve a body mass index (BMI) < 25 kg/m² or a waist circumference < 102 cm in men and < 88 cm in women. Whereas these are the long-term goals, an initial weight loss of 10% from baseline is a first step. Further weight reduction can be attempted if the initial 10% weight loss is successfully achieved and maintained.

5.5.3 Blood pressure control

The goal is to achieve blood pressure < 140/90 mmHg in non-diabetic patients and < 130/80 mmHg in patients with diabetes or chronic renal dysfunction. Lifestyle interventions are an important means of achieving blood pressure control, particularly physical activity, in addition to weight loss and pharmacotherapy.^333–335

5.5.4 Management of diabetes

Glycaemic balance abnormalities (impaired fasting glucose level, impaired glucose tolerance, or abnormal fasting glucose level) should be actively searched for in every patient with proven NSTE-ACS. In patients with established diabetes, the aim is to achieve HbA1c levels ≤ 6.5%. Counselling with endocrinologists is advisable. Lifestyle measures, in addition to weight loss in obese patients, and adapted pharmacotherapy are of great importance. In patients with impaired fasting glucose level or impaired glucose tolerance, no specific treatment is so far recommended except for lifestyle changes.³⁴⁰

5.5.5 Interventions on lipid profile

Interventions on low-density lipoprotein (LDLc) and high-density lipoprotein (HDLc) cholesterol as well as triglycerides are an important component of long-term management of NSTE-ACS. Most of the evidence has been obtained in the field of LDLc reduction, which is best achieved with statins or with a combination of statins and other lipid-lowering agents. Other interventions to correct low HDLc or high triglycerides might be necessary in some patients, although the impact of these measures on long-term outcome is less well established.

Statins

Long-term statin therapy improves outcome for all forms of CAD, after NSTE-ACS or in patients with chronic manifestations of CAD.^341–344 This beneficial effect was shown in all subgroups, including men and women, the elderly, smokers, diabetic patients, hypertensive patients, or patients with chronic kidney disease (CKD). The recent guidelines recommend combining dietary interventions with pharmacotherapy by statins, or a combination of statins with other lipid-lowering agents, to reduce LDLc to < 100 mg/dL (<2.6 mmol/L). However, two aspects of LDLc reduction have to be discussed specifically, namely early prescription of statins at the acute phase of NSTE-ACS and the impact of aggressive lipid-lowering therapy, to achieve LDLc levels < 70 mg/dL (<1.81 mmol/L).^333–335

The rationale behind the prompt initiation of statin therapy after NSTE-ACS includes the possibility of plaque stabilization, anti-inflammatory effects, and restoration of endothelial function. Moreover, an NSTE-ACS may serve as an impetus for the initiation and maintenance of long-term therapy, whereas in the chronic phase, treatment may be received and pursued less vigilantly.

So far little or no benefit of statins initiated soon after the acute phase has been reported in trials, registries, meta-analyses, and post hoc analyses of NSTE-ACS studies.^14,345–349 More recent randomized trials specifically addressing this issue have shown that early aggressive lipid-lowering therapy led to a swift and important drop in LDLc, but apparently without major impact on short-term outcome.^350–352 A more recent meta-analysis including 13 trials and 17 963 patients revealed that early initiation of statin therapy was safe and had a positive impact on outcome, with beneficial effects on the rate of death and cardiovascular events over 2 years of follow-up (HR 0.81, 95% CI, 0.77–0.87, P < 0.001). Survival benefit was apparent only after 4 months, achieving statistical significance by 12 months.³⁵³

The potential benefit of aggressive when compared with moderate lipid-lowering therapy among a wide spectrum of NSTE-ACS patients was assessed in PROVE-IT.³⁵⁴ This trial enrolled STE-ACS and NSTE-ACS patients with a total cholesterol level < 240 mg/dL (6.2 mmol/L). Treatment with pravastatin 40 mg or atorvastatin 80 mg was initiated within 10 days of admission, and follow-up was continued over 18–36 months. By the end of the study, the levels of LDLc were reduced by 21% in the pravastatin arm [to a median level of 95 mg/dL (2.46 mmol/L)], when compared with a 49% reduction in the atorvastatin arm [to a median level of 62 mg/dL (1.6 mmol/L)], with most of the treatment effect achieved within 30 days. The primary composite endpoint (death, MI, unstable angina requiring rehospitalization, revascularization, or stroke) was reduced by 16% in the intensive compared with the non-intensive therapy arm. The difference in outcomes was already apparent within 30 days of randomization. Patients who achieved levels of LDLc < 70 mg/dL (1.81 mmolL) had lower event rates than those with higher levels. A similar difference was observed between those who achieved hsCRP levels < 2 mg/L after statin therapy when compared with those with levels > 2 mg/L. Thus, intensive lipid-lowering therapy, associated with reductions in the levels of LDLc or hsCRP to values of < 70 mg/dL (1.81 mmol/L) or < 2 mg/L, respectively, leads to improved outcome after ACS.

Other lipid-lowering agents

Limited data are available about the benefit of fibrates, nicotinic acid, and ezetimibe in the setting of NSTE-ACS. A combination of statin and ezetimibe has shown considerable capacity to reduce LDLc and is being tested vs. conventional statin therapy in a large clinical trial in ACS patients (IMPROVE-IT). Low HDLc has been shown to be a risk factor for CAD and death from CAD. Epidemiological studies suggest also that raising HDLc levels may prevent the development of CAD. Each increase in baseline HDLc of 1 mg/dL (0.03 mmol/L) is associated with a 6% decrease in the risk of death from CAD or MI.^355,356 Nicotinic acid has been shown to be able to raise HDLc levels significantly. Evidence from old or small-size studies suggests that raising HDLc may lead to a significant risk reduction for coronary events.³⁵⁷ A large clinical trial exploring this therapeutic possibility is ongoing. It will assess the potential of an association of statin plus nicotinic acid (niacin) to reduce the rate of cardiovascular events when compared with statin alone in a population of patients with established atherosclerotic disease and atherogenic lipid profile (AIM-HIGH study). Other therapeutic approaches aiming at raising HDLc have failed.³⁵⁸

In addition, chronic aerobic exercise was shown to increase HDLc and should be advocated whenever possible³⁵⁹ (see section 5.6 Rehabilitation and return to physical activity).

Recommendations for lipid-lowering therapy

• Statins are recommended for all NSTE-ACS patients (in the absence of contraindications), irrespective of cholesterol levels, initiated early (within 1–4 days) after admission, with the aim of achieving LDLc levels < 100 mg/dL (<2.6 mmol/L) (I-B).

• Intensive lipid-lowering therapy with target LDLc levels < 70 mg/dL (<1.81 mmol/L) initiated within 10 days after admission is advisable (IIa-B).

5.5.6 Antiplatelet agents and anticoagulants

See sections 5.2 Anticoagulants and 5.3 Antiplatelet agents.

5.5.7 Beta-blockers

Beta-blocker therapy should be initiated in all patients and maintained indefinitely in the case of reduced LV function, with or without symptoms of heart failure, unless formal contraindications exist. In other patients, beta-blockers may be useful, but evidence of their long-term benefit is not established. Meta-analysis and registry data have shown that long-term treatment with beta-blockers in patients suffering from NSTE-ACS may lead to a significant risk reduction for death.³⁶⁰

Recommendations for use of beta-blockers

• Beta-blockers should be given to all patients with reduced LV function (I-A).

5.5.8 Angiotensin-converting enzyme inhibitors

Angiotensin-converting enzyme inhibitors (ACE inhibitors) are beneficial in reducing remodelling and improving survival in patients with reduced LV systolic function (with or without clinical heart failure) after MI.^361–363 Therefore, their initial use in the context of ACS was limited to patients with reduced LV systolic function. Subsequently, several trials suggested an anti-atherogenic effect of ACE inhibitors in patients with risk factors for atherosclerosis or established atherosclerotic disease, irrespective of LV function and beyond their effect on blood pressure.^268,364,365 Meta-analyses of major trials carried out with the main objective of demonstrating the anti-atherogenic effect of ACE inhibitors have shown a 14% risk reduction for death at 4 years with ACE inhibitors in this setting.^366–368 To date, only ramipril and perindopril have shown efficacy. Prescription of ACE inhibitors in this indication should be restricted to agents and doses of proven efficacy.³⁶⁹ The applicability of these findings, although documented in the context of stable CAD, has been extended to all patients with NSTE-ACS. For patients with reduced LV systolic function, the use of an oral ACE inhibitor should be initiated within the first day after admission, in the absence of contraindications. For other patients, treatment should be initiated during the course of hospitalization.

Recommendations for the use of ACE inhibitors

• ACE inhibitors are indicated long-term in all patients with LVEF ≤ 40% and in patients with diabetes, hypertension, or CKD, unless contraindicated (I-A).

• ACE inhibitors should be considered for all other patients to prevent recurrence of ischaemic events (IIa-B). Agents and doses of proven efficacy are recommended (IIa-C).

5.5.9 Angiotensin-2 receptor blockers

Recent trials have clearly documented that angiotensin-2 receptor blockers (ARBs) may be used in acute MI patients with reduced LV systolic function.^370,371 They can be used instead of ACE inhibitors or in combination with ACE inhibitors. In contrast to ACE inhibitors, there are no firm data regarding their use as anti-atherogenic agents. For patients with reduced LV systolic function, their use should be initiated within the first day after admission, in the absence of contraindications.

Recommendations for the use of angiotensin receptor blockers

• ARBs should be considered in patients who are intolerant to ACE inhibitors and/or who have heart failure or MI with LVEF < 40% (I-B).

5.5.10 Aldosterone receptor antagonists

Spironolactone has been shown to be beneficial in the treatment of patients with LV systolic dysfunction and chronic, severe heart failure (NYHA classes III and IV).³⁷² During chronic use of spironolactone, a minority of patients develop gynaecomastia related to binding of the drug to progesterone receptors. Eplerenone is a new aldosterone receptor antagonist with 1000-fold less affinity for the progesterone receptor than spironolactone. Eplerenone has been evaluated in a randomized, placebo-controlled trial of patients after MI (with or without ST-elevation) and LV systolic dysfunction with either symptomatic heart failure or diabetes mellitus.³⁷³ The acute use of oral eplerenone, in addition to optimal medical and invasive therapy, was associated with improved outcome (morbidity and mortality). Aldosterone receptor antagonists should not be used in the case of severe renal failure [serum creatinine > 2.5 mg/dL (221 µmol/L) for men and > 2.0 mg/dL (177 µmol/L) for women], hyperkalaemia, or inability to perform serial examinations to monitor potassium levels.

Recommendations for aldosterone receptor antagonists

• Aldosterone blockade should be considered in patients after MI who are already treated with ACE inhibitors and beta-blockers and who have an LVEF < 40% and either diabetes or heart failure, without significant renal dysfunction or hyperkalaemia (I-B).

5.6 Rehabilitation and return to physical activity

After NSTE-ACS, an assessment of functional capacity and the ability to carry out daily activities or work is needed. This ability is influenced, among other factors, by cardiac function, extent of CAD, the presence and magnitude of residual ischaemia, and the propensity for cardiac arrhythmias. Every patient after NSTE-ACS should undergo an ECG-guided exercise test (or another non-invasive test if exercise cannot be performed or if the ECG is difficult to interpret) within 4–7 weeks after discharge.^374–376 As a rule of thumb, physical activity that includes leisure, professional, and sexual activities should be resumed at 50% of the maximal exercise capacity, expressed as metabolic equivalents (METS), and gradually increased over time. A patient with preserved LV systolic function (EF > 0.40) and without inducible ischaemia or arrhythmias in a stress test can return to work. If the work is in an office environment, an 8 h day can be resumed. If the work is manual, the workload should not exceed 50% of the maximal exercise capacity assessed in the stress test. The workday should not exceed 4 h in the first month, with monthly 2 h increments thereafter. A patient with moderate LV systolic dysfunction (EF between 0.30 and 0.40) or with mild ischaemia in a stress test can resume office work, but should be limited to static manual work. A patient with severe LV systolic dysfunction (EF < 0.30) or significant ischaemia in a stress test can perform office work, provided exercise capacity is > 5 METS without symptoms. Otherwise, the patient should refrain from working. As for other physical activities, including sexual activity, the non-invasive test can also guide the physician's advice. In general, a patient with an exercise capacity > 5 METS can engage in routine sexual activity. The physician should inform the patient about the timing of resumption of physical and sexual activity, taking into account the above-mentioned cardiac parameters as well as other factors such as the status of the arterial puncture site in a patient after cardiac catheterization. In all cases, close collaboration between the cardiologist and the primary physician is necessary.

Recommendations for rehabilitation and return to physical activity

• After NSTE-ACS, assessment of functional capacity is recommended (I-C).

• Every patient after NSTE-ACS should undergo an ECG-guided exercise test (if technically feasible) or an equivalent non-invasive test for ischaemia, within 4–7 weeks after discharge (IIa-C).

• On the basis of cardiovascular status and the results of functional physical capacity assessment, patients should be informed about the timing of resumption and the recommended level of physical activity, including leisure, work, and sexual activities (I-C).

6. Complications and their management

6.1 Bleeding complications

Bleeding complications are the most frequent non-ischaemic complications observed in the management of NSTE-ACS. Several definitions, including clinical aspects of bleeding (location and impact on haemodynamics) and/or need for blood transfusion as well as magnitude of haemoglobin drop, are used to grade bleeding severity (Table 7).³⁷⁷ Bleeding is graded as severe, life threatening, major, or minor. However, the same term may represent a different bleeding severity depending on the definition used. This implies that different bleeding complication rates may be observed within the same study population depending on whether different definitions are used to grade severity of bleeding. This also implies that the frequency of bleeding may be difficult to compare across studies.

Taking into account all these limitations about bleeding definition, it is estimated that the frequency of major bleeding ranges from 2 to 8% across the spectrum of NSTE-ACS and depends greatly on the type of treatment used, particularly the type and dose of antithrombotic and antiplatelet therapies, invasive procedures, and other factors (Table 8).^377,378 In randomized trials, the reported frequency ranges from < 2% in OASIS-2, PRISM, and PURSUIT to > 8% in SYNERGY.^164,178,236 Figures from registries are generally higher than those in clinical trials. In the CRUSADE registry, blood transfusion used as a surrogate marker of major bleeding was used in more than 15% of patients,²⁵² possibly reflecting a higher rate of invasive strategy used in the USA. In the GRACE registry, data from 24 045 patients revealed that the overall incidence of major bleeding was 3.9% in patients with STE-ACS and 4.7% in patients with NSTE-ACS, and 2.3% in patients with unstable angina.³⁷⁹

6.1.1 Predictors of bleeding risk

The independent predictors of major bleeding in the GRACE registry were advanced age (OR 1.22 per 10-year increase, P = 0.0002), female sex (OR 1.36, P = 0.0116), history of bleeding (OR 2.18, P = 0.014), use of PCI (OR 1.63, P = 0.0005), history of renal insufficiency (OR 1.53, P = 0.0062), and use of GP IIb/IIIa inhibitors (OR 1.86, P = 0.0001), among others (Table 8).³⁷⁹ Excessive doses of drugs, especially in women, elderly patients, or those with renal failure, also increase the risk of bleeding.¹⁶⁸ Renal dysfunction plays a critical role. Bleeding risk exponentially increases with decreasing CrCl.^176,382 A steep increase in bleeding risk is already observed for CrCl levels below 60 mL/min. Better definition of the appropriate dose of antithrombotic agents to be given according to the level of renal dysfunction is needed.

In addition, the same baseline characteristics, namely age, gender, and renal dysfunction, influence the risk of both death and bleeding. In the GRACE registry, the increase in the risk of bleeding with declining renal function parallels the increase in the risk of death as shown in Figure 10. This implies that caution has to be exerted with high-risk patients when deciding on aggressive invasive and/or anticoagulant/antiplatelet treatments. Particular attention has to be paid to the selection of the doses of anticoagulants in CKD patients.

According to recent reports, baseline haemoglobin/haematocrit has also been shown to be an independent predictor of bleeding complications, both procedure-related and non-procedure-related bleeds.³⁸³

6.1.2 Impact of bleeding on prognosis

Bleeding has a strong impact on prognosis. Major bleeding in the GRACE registry was associated with an increased risk of hospital death (OR 1.64, 95% CI 1.18–2.28, P < 0.001).³⁷⁹ According to several reports including a large meta-analysis of registries and trials, including more than 30 000 patients, major bleeding was shown to be associated with a four-fold increase in the risk of death, a five-fold increase in risk of recurrent MI, and a three-fold increase in risk of stroke at 30 days.^{377–379,384} Pooled data from four multicentre randomized clinical trials of patients with ACS, totalling 26 452 patients, documented a stepwise increase in the risk of death at 30 days and 6 months, according to the severity of bleeding. At 1 month, the hazard ratios for death were 1.6, 2.7, and 10.6 for mild, moderate, and severe bleeding, respectively (GUSTO definition), and at 6 months, the hazard ratios were 1.4, 2.1, and 7.5, respectively.³⁸⁴ The same impact on prognosis has been shown to exist for both procedure-related and non-procedure-related bleeding, as well as in other settings, such as PCI.³⁸⁴ In the OASIS-5 trial, at 30 days, the risk of ischaemic events was strongly influenced by the occurrence of major bleeding. The rate of death was 12.9 vs. 2.8%, the risk of MI 13.9 vs. 3.6%, and the risk of stroke 3.6 vs. 0.8% for patients who suffered a major bleeding vs. no bleeding, respectively. The same is also true for minor bleeding, although the increase in risk is of lesser magnitude.¹⁷⁶ Beyond 30 days, the risk is lower, but still present, as modern treatment of NSTE-ACS includes the use of dual antiplatelet therapy for 12 months, which has been shown to result in a higher long-term risk of bleeding.^378,385

Several factors contribute to the worse outcome associated with bleeding. Renal failure, as well as haemodynamic consequences of the bleed, and potential deleterious effects of transfusions may account for the higher risk. In addition, bleeding triggers a pro-thrombotic and pro-inflammatory state. The main component of the risk is probably the need to discontinue antiplatelet and antithrombotic drugs, which can lead to an increased risk of events through a rebound phenomenon. However, as the risk factors for both bleeding and ischaemic events are mostly the same, the higher risk patients are exposed to both risks, and also submitted to the most aggressive drug and procedure strategies. Hence, occurrence of bleeding may simply be a precipitating factor for worse outcome in a frail population.

6.1.3 Management of bleeding complications

Prevention of bleeding has become a target as important as the prevention of ischaemic events. In the OASIS-5 trial, the difference in mortality between the two groups was almost entirely associated with the reduction in bleeding in the fondaparinux group. Therefore, risk assessment in patients with NSTE-ACS needs to address the risk of both thrombotic and bleeding complications. Prevention of bleeding encompasses the choice of safer drugs, appropriate dosage (taking into account age, gender, and CrCl), reduced duration of antithrombotic treatment, use of combination of antithrombotic and antiplatelet agents according to proven indications, as well as the choice of radial over femoral approach if angiography or PCI is being considered. In addition, if an invasive procedure is planned, unnecessary delays should be avoided, as this prolongs the duration during which the patient is at risk for bleeding.

Minor bleeding, unless persistent, does not require the interruption of active treatments. Major bleeding such as gastrointestinal, retro-peritoneal bleeding, intracranial haemorrhage, or severe blood loss requires the interruption and neutralization of both antiplatelet and antithrombotic treatment if bleeding cannot be controlled by appropriate interventions. It might not be necessary to interrupt antithrombotic/antiplatelet agents if complete control of the haemorrhage can be obtained with local treatment.³⁸⁶ In clinical practice, the risk of interrupting antithrombotic and antiplatelet agents must be weighed against the risk of a thrombotic event, particularly if the patient has been submitted to revascularization and stent implantation. The risk of acute thrombotic events after interruption of antithrombotic/antiplatelet agents is maximum after 4–5 days, but persists for up to 30 days.³⁷⁸

UFH can be inhibited by an equimolar concentration of protamine sulfate, which neutralizes the factor-IIa activity. However, protamine sulfate has less or no impact on neutralization of factor-Xa activity achieved with LMWH or fondaparinux. In this situation, recombinant factor-VII has been recommended.³⁸⁷ However, there is no firm evidence that this can control bleeding, and recent data show that the use of recombinant factor-VIIa is associated with an increased risk of thrombotic complications.³⁸⁷

Antiplatelet activity is also difficult to reverse. Aspirin and clopidogrel are irreversible platelet inhibitors. Their action is slowly reversed by the continuous generation of new platelets (around 10–20% per day), so antiplatelet effects persist for 5–10 days after cessation of treatment. No compound was found to reverse significantly the pharmacological activity of clopidogrel. If prompt correction of prolonged bleeding time is required, platelet transfusion is the only possibility to reverse the effects of clopidogrel/aspirin. The recommended minimum dose in adults is 0.5–0.7 × 10¹¹ platelets per 7 kg of bodyweight. This is not based on firm evidence, but only on expert consensus.³⁸⁸

GP IIb/IIIa inhibitors have different pharmacological properties that are important to consider when evaluating the modalities for reversal. Because little free abciximab circulates in plasma, an infusion of platelets replenishes the number of viable GP IIb/IIIa receptors, and thus allows a return to normal haemostasis. However, although platelet administration may be beneficial in patients experiencing major bleeding with abciximab, no recommendations exist regarding the quantity required to reverse its antiplatelet effect. The situation is different with tirofiban or eptifibatide. As these drugs undergo significant renal elimination, the baseline platelet function of patients with normal renal function can be expected to return to normal within 4–8 h after infusion interruption. If immediate reversal of platelet inhibition is needed, platelet infusion alone may be insufficient because of the large amount of freely circulating molecules. Fibrinogen-containing plasma supplementation may help to restore platelet aggregation.^254,389

Antiplatelet and/or antithrombotic agents cannot be re-introduced until strict control of the haemorrhage has been obtained for at least 24 h. In the case of peptic ulcer, re-introduction of antiplatelet therapy—whatever the drug or combination of drugs used—needs to be associated with proton pump inhibitors.

6.1.4 Impact of blood transfusion

Blood transfusion can be required to control anaemia and haemodynamic compromise. However, there is ongoing controversy about its real efficacy and safety in the context of NSTE-ACS. Blood transfusion has been shown to improve prognosis in elderly patients suffering from acute MI with haematocrit levels < 30% and may possibly be useful for haematocrit levels in the range of 30–33%.³⁹⁰ The utility of blood transfusion for higher values of haematocrit is not proven. In another report, blood transfusion improved 1-month outcome in STEMI patients, if given for baseline haemoglobin < 12 g/dL.³⁹¹ However, in the same report, transfusion was associated with an increased risk of death, MI, and refractory ischaemia in NSTE-ACS. Similarly, blood transfusion was associated with poorer outcome, even after adjustment for baseline characteristics and in-hospital procedures, in a meta-analysis involving more than 24 000 ACS patients.³⁸³ In a more recent meta-analysis, a 20% increase in mortality was reported in patients who received transfusions.³⁹²

Some small size randomized trials have tested the efficacy of transfusion in critically ill patients, vascular surgery, or patients with recent trauma, and showed that blood transfusion may have no effect on mortality, or even be associated with lower survival.^393–396 In two clinical trials, a restrictive blood transfusion strategy led to better results than a liberal one in terms of mortality and organ failure at 30 days in critically ill patients suffering from acute (including cardiac) conditions and treated in intensive care units. However, no significant difference in 30-day outcome was observed in cardiac patients. In these trials, blood transfusion was given for haemoglobin levels < 7 g/dL, aiming at haemoglobin levels between 7 and 9 g/dL in the restrictive and between 10 and 12 g/dL in the liberal strategy.^394,395 Nonetheless, despite several studies, the correct haematocrit or haemoglobin levels to achieve after blood transfusion in patients with anaemia (with or without cardiovascular disease) have not yet been adequately defined.

It is not clearly understood why transfusion may be associated with adverse outcome. Alterations in erythrocyte, nitric oxide biology in stored blood, and high haemoglobin oxygen affinity due to a low rate of 2,3-diphosphoglyceric acid, leading to decreased oxygen delivery to tissues, have been put forward, as well as increases in inflammatory mediators.^397–399

All in all, the information about the efficacy of and the indications for blood transfusion needs to be critically considered. In mild to moderate anaemia (haematocrit > 25% or haemoglobin levels > 8 g/dL), blood transfusion may be associated with increased risk of death at 30 days and should be avoided if anaemia is haemodynamically well tolerated. Below these haematocrit/haemoglobin levels, blood transfusion should be given.³⁸³

Recommendations for bleeding complications

• Assessment of bleeding risk is an important component of the decision-making process. Bleeding risk is increased with higher or excessive doses of antithrombotic agents, length of treatment, combinations of several antithrombotic drugs, switch between different anticoagulant drugs, as well as with older age, reduced renal function, low body weight, female gender, baseline haemoglobin, and invasive procedures (I-B).

• Bleeding risk should be taken into account when deciding on a treatment strategy. Drugs, combination of drugs, and non-pharmacological procedures (vascular access) known to carry a reduced risk of bleeding should be preferred in patients at high risk of bleeding (I-B).

• Minor bleeding should preferably be managed without interruption of active treatments (I-C).

• Major bleeding requires interruption and/or neutralization of both anticoagulant and antiplatelet therapy, unless bleeding can be adequately controlled by specific haemostatic intervention (I-C).

• Blood transfusion may have deleterious effects on outcome and should therefore be considered individually, but withheld in haemodynamically stable patients with haematocrit > 25% or haemoglobin level > 8 g/L (I-C).

6.2 Thrombocytopenia

Thrombocytopenia is defined as a decrease in platelet count to < 100 000 µL⁻¹ or a drop of > 50% from baseline platelet count. Thrombocytopenia is considered to be moderate if the platelet count is between 20 000 and 50 000 µL⁻¹ and severe if it is less than 10 000 µL⁻¹.

6.2.1 Heparin-induced thrombocytopenia

Thrombocytopenia can occur during UFH or LMWH treatment, but has a different significance and potential for complications depending on whether or not it is immune-mediated.

Mild and transient decline in platelet count occurring 1–4 days after initiation of therapy is common and observed in up to 15% of UFH-treated patients. It is not immune-mediated and rarely leads to a severe reduction in platelet levels. It resolves spontaneously, despite continuation of UFH. Pseudo-thrombocytopenia is a laboratory artefact due to platelet clumping in EDTA-containing tubes and can be avoided by the use of citrate instead of EDTA for blood sampling.

The immune-mediated form of HIT is a serious complication that often leads to severe thrombo-embolic events. It is not dose-dependent, usually causes a severe drop in platelet levels (at least 50%), and typically appears 5–14 days after the start of UFH treatment,⁴⁰⁰ but much earlier in patients with recent (within 3 months) UFH exposure.⁴⁰¹ Delayed-onset HIT occurring several days or weeks after the cessation of UFH treatment has also been described.⁴⁰² It is beyond the scope of this document to discuss the mechanisms and causes of HIT. When HIT is suspected, laboratory confirmation can be obtained with various tests, but treatment of HIT must be undertaken as soon as the diagnosis is suspected, without waiting for laboratory confirmation.

HIT must be suspected when there is a drop of > 50% in platelet count or a decrease in platelet count to < 100 000 µL⁻¹. Immediate interruption of UFH or LMWH is mandatory, as soon as HIT is suspected. Alternative antithrombotic therapy must be introduced, even in the absence of thrombotic complications. Heparinoids such as danaparoid sodium (Orgaran) may be used, although in vitro cross-reactions with UFH or LMWH have been observed, but apparently without causing thrombosis. The alternative is to use DTIs, such as argatroban, or hirudin or derivatives, which do not carry any risk of thrombocytopenia, and make it possible to have sustained and controllable antithrombotic activity, easily monitored by aPTT.⁴⁰³ Fondaparinux (pentasaccharide) also has the potential to be used in this type of situation, since it has a potent antithrombotic effect, without any cross-reaction with platelets,⁴⁰⁴ but is not approved in this indication.

6.2.2 Glycoprotein IIb/inhibitor-induced thrombocytopenia

Thrombocytopenia has been reported to occur at rates ranging from 0.5 to 5.6% in clinical trials of parenteral GP IIb/IIIa inhibitors, a rate comparable with that observed with UFH alone.^{244,405–407} Abciximab more than doubles the incidence of severe thrombocytopenia in comparison with placebo. The risk is lower with eptifibatide (0.2% severe thrombocytopenia in PURSUIT)²³⁵ or tirofiban. In the TARGET study, thrombocytopenia developed in 2.4% of the patients treated with abciximab and 0.5% of those treated with tirofiban (P < 0.001).⁴⁰⁸ Again, it is beyond the scope of this document to discuss the mechanisms and causes of GP IIb/IIIa inhibitor-induced thrombocytopenia.

Severe and profound thrombocytopenia due to GP IIb/IIIa inhibitors may remain asymptomatic with only minor bleeding at the access site and minor oozing. Major bleeds are rare, but may be life threatening. It is recommended that all patients treated with GP IIb/IIIa inhibitors undergo platelet counts within 8 h of onset of drug infusion or in the case of bleeding. In the case of acute profound thrombocytopenia (<10 000 µL⁻¹), discontinuation of GP IIb/IIIa inhibitors as well as UFH or LMWH is recommended. Platelet transfusions are indicated in the case of bleeding. Fibrinogen supplementation with fresh–frozen plasma or cryoprecipitate either alone or in combination with platelet transfusion has also been advocated.²⁵⁴

After drug discontinuation, tirofiban-induced thrombocytopenia resolves after a mean of 2.1 days (range 1–6 days), whereas abciximab-induced thrombocytopenia resolves after a mean of 4.5 days (range 1–24 days). Thrombocytopenia induced by GP IIb/IIIa inhibitors is associated with worse outcome, including increased 30-day rates of bleeding, recurrent ischaemia, urgent revascularization, and death.⁴⁰⁸

Recommendations for thrombocyopenia

• Significant thrombocytopenia (<100 000 µL⁻¹ or > 50% drop in platelet count) occurring during treatment with GP IIb/IIIa inhibitors and/or heparin (LMWH or UFH) requires the immediate interruption of these drugs (I-C).

• Severe thrombocytopenia (<10 000 µL⁻¹) induced by GP IIb/IIIa inhibitors requires platelet transfusion with or without fibrinogen supplementation with fresh–frozen plasma or cryoprecipitate in the case of bleeding (I-C).

• Interruption of heparin (UFH or LMWH) is warranted in the case of documented or suspected HIT. In the case of thrombotic complications, anticoagulation can be achieved with a DTI (I-C).

• Prevention of HIT can be achieved with the use of anticoagulants devoid of risk of HIT, such as fondaparinux or bivalirudin, or by brief prescription of heparin (UFH or LMWH) in cases where these compounds are chosen as anticoagulant (I-B).

7. Special populations and conditions

Some special populations deserve additional considerations for the management of NSTE-ACS. The following groups of patients are at substantial risk of adverse cardiac events or merit alternative therapeutic strategies. Although discussed separately, there is great overlap between the subgroups, i.e. many elderly patients are women and/or have renal dysfunction, diabetes, or anaemia. In this section, some considerations for these populations will be provided. Comprehensive reviews can be found elsewhere.^{391,409–412}

7.1 The elderly

There is a substantial increase of elderly patients with CAD worldwide. Although there is no common definition of what represents elderly, either age > 65 or > 75 years are the two most common definitions. Although such dichotomous cut-offs are in general helpful, it should be recognized that the risk for mortality increases in a continuous curvilinear manner with each decade after age 50. Thus, the risk of cardiac adverse events such as death, stroke, MI, and heart failure is substantial among patients over the age of 75 with CAD. In the USA, persons over the age of 75 represent only 6% of the population, but account for 37% of all admissions for acute MI and 60% of all MI-related deaths.⁴¹³

In Europe, the rate of patients > 75 years old in registries of NSTE-ACS varies from 27 to 34.1%.^414,415 In patients > 75 years old, the death rate is at least twice as high as in patients aged < 75.⁴¹⁵ Despite the high proportion of elderly patients in registries, the elderly (>75 years) represent less than 10% of all patients in recent trials.⁴¹⁶ Furthermore, it has recently been shown that the elderly enrolled into NSTE-ACS trials also had substantially fewer co-morbid illnesses, particularly renal and heart failure, compared with the general population of elderly patients at the same institutions.⁴¹⁷ Thus, the applicability of findings from the trials that enrolled predominantly younger patients to an older and generally sicker population is questionable. On the basis of these observations, the risk–benefit ratio with any therapeutic strategy should be determined in the elderly, with special consideration for estimated life expectancy, patient wishes, and co-morbidities before invasive strategies and therapy that increase bleeding risk and/or risk of renal failure are applied.

7.1.1 Early diagnostic evaluation in the elderly

The clinical presentation of NSTE-ACS in the elderly can sometimes be difficult. The elderly are more likely to have mild symptoms and frequently have atypical symptoms or no chest pain. Common symptoms in the elderly are shortness of breath (49%), diaphoresis (26%), nausea–vomiting (24%), and syncope (19%).^48,49 The ECG among elderly patients with MI is also more likely to be non-diagnostic, without either ST-elevation or ST-depression in as many as 43%. Presentation with overt heart failure is also common, with up to 41% having symptoms of heart failure on admission to hospital.⁴¹⁸ Thus, among elderly patients presenting with non-specific symptoms, a high level of suspicion for NSTE-ACS should be entertained, even among patients with non-specific ECG findings.

7.1.2 Therapeutic considerations

The bleeding risk linked to LMWH is higher in elderly patients.^379,419 Although there has been a suggestion of a greater therapeutic effect with LMWH compared with UFH, this has not held up after multivariable modelling has adjusted for important baseline characteristics in the elderly vs. younger patients.⁴¹⁹ In OASIS-5, patients over 65 years of age suffered a higher rate of bleeding complications than younger patients, but with a significantly lower risk of bleeding with fondaparinux than with enoxaparin.¹⁷⁶ The meta-analysis of all trials with GP IIb/IIIa inhibitors showed that the therapeutic benefit appears to be less for older patients (OR 0.86 in < 60 years vs. 0.96 in > 70 years, P-value for interaction 0.10), while major bleeding also appeared to be about 60% higher.²²⁹ However, the CURE trial documented a more consistent benefit: about 2% absolute reduction of death, MI, and stroke among elderly patients (>65 years) receiving clopidogrel and aspirin vs. aspirin alone.¹⁶⁷ Attention to the risk–benefit ratio of these therapies needs to be individualized in the elderly patient depending on the treatment and co-morbid illness profiles. Attention has to be paid to the dosage of antithrombotic agents, since in the CRUSADE registry, it was shown that excessive dosing was frequently observed in elderly patients and led to a significantly higher rate of bleeding.¹⁶⁸

Elderly patients are substantially less likely to receive an invasive strategy after NSTE-ACS, and adjusted observational analysis has failed to show any early survival benefit compared with younger patients.²⁵² However, a subgroup analysis of one of the largest randomized trials of invasive vs. conservative strategies, using current interventional strategies (stents and GP IIb/IIIa inhibitors), showed a substantial treatment effect in favour of an invasive strategy⁴²⁰ (Figure 11). Among patients > 75 years old, there was a 56% relative reduction in death and non-fatal MI. This was offset by a three-fold higher risk of in-hospital major bleeding. Although the FRISC-2 trial did not enrol patients over the age of 75, the greatest reduction in death and non-fatal MI was observed among patients over the age of 65 during a 5-year follow-up [24.4 vs. 31.5% (OR 0.77, CI 0.64–0.93) invasive vs. non-invasive strategy].¹²² Taken together, these findings suggest that the invasive strategy is associated with an overall better long-term outcome. However, the risk–benefit ratio needs to be carefully evaluated in elderly patients considered for routine invasive care. It becomes increasingly important with advancing age when evaluating elderly patients to select strategies and/or drugs to minimize the risk of bleeding and adverse outcome. CrCl should always be calculated in elderly patients in order to adapt dosage of drugs with exclusive or substantial renal elimination (see section 7.4 Chronic kidney disease).

Recommendations for the elderly

• Elderly patients (>75 years old) often have atypical symptoms. Active screening for NSTE-ACS should be initiated at lower levels of suspicion than among younger (<75 years old) patients (I‐C).

• Treatment decisions in the elderly should be tailored according to estimated life expectancy, patient wishes, and co-morbidities to minimize risk and improve morbidity and mortality outcomes in this frail but high-risk population (I-C).

• Elderly patients should be considered for routine early invasive strategy, after careful evaluation of their inherent raised risk of procedure-related complications, especially during CABG (I-B).

7.2 Gender

In general, women have their first cardiovascular event an average of 10 years later in life compared with men. Thus, in NSTE-ACS, they tend to be older and therefore have more co-morbid illnesses, including renal impairment and heart failure. In registries in Europe, the average age of women with NSTE-ACS was 6 years higher than in men (71 vs. 65 years). On average, 45% of females and 20.5% of males were aged > 75. Diabetes was more frequent in females than in males (26 vs. 22%). However, other risk factors were equally distributed in men and women.⁴²¹ In a registry of 201 114 patients with a first MI, multivariable analysis showed that younger women had a 25% higher 30-day mortality compared with men. However, gender was not an independent predictor of 1-year survival. Interactions between age and gender observed in short-term case fatality can be explained by increased pre-hospital mortality in men.⁴²² However, among older women and men, the mortality rates were similar after adjustments for co-morbid illnesses. In the analysis from the GUSTO-2B trial, women with NSTE-ACS had a significantly higher mortality rate at 30 days than did men and similar rates of re-infarctions. In a subgroup with unstable angina, female sex was associated with an independent protective effect.⁴²³

Females in general are less likely to receive evidence-based therapies including diagnostic procedures in NSTE-ACS.⁴²⁴ In European registries, women were under-treated when compared with men, especially in terms of PCI (24.4% for men vs. 22.9% for women), prescription of clopidogrel (49% for men vs. 39% for women), and prescription of GP IIb/IIIa inhibitors (24.8% for men vs. vs 23.8% for women). Referral for revascularization, percutaneous or surgical, was significantly lower for women.^{253,331,421–426} For most treatments, there has not been any gender differential treatment effect with new therapeutic agents. However, with GP IIb/IIIa inhibitors and early revascularization (by either PCI or CABG), several trials have reported more adverse events in women, especially in those at lower risk. Registry data do not suggest that gender is an independent adverse risk of outcome. Therefore, it is recommended that women be evaluated and treated similarly to men, with special attention to co-morbid risk factors in NSTE-ACS.

7.2.1 Glycoprotein IIb/IIIa inhibitors in women

A meta-analysis of major clinical trials with GP IIb/IIIa inhibitors in NSTE-ACS showed a lack of treatment effect in women, with a significant interaction between sex and allocated treatment, with a treatment benefit in men. There was also a significant treatment interaction in favour of troponin-positive patients.²²⁹ In a pooled analysis of trials with abciximab, no gender difference in the protection from major adverse outcomes was demonstrated. Women had higher rates of bleeding.^427,428 It was suggested that women were more likely to have non-obstructive CAD, in which therapeutic advantage of agents that act on the atherothrombotic process may be minimal.⁴²⁹ It is recommended that GP IIb/IIIa inhibitors in NSTE-ACS should be used primarily among women with troponin elevation and a high likelihood of CAD.

7.2.2 Revascularization and early invasive strategy among women

In the contemporary practice of PCI with stents and GP IIb/IIIa inhibitors, a meta-analysis of randomized trials of an invasive approach (with revascularization using PCI or CABG) has shown a 23% risk reduction in 2-year mortality (RR 0.77, 95% CI 0.60–0.99). However, when the outcomes of men and women were examined, the treatment benefit was restricted to men (RR 0.68, 95% CI 0.57–0.81), whereas in women there appeared to be no benefit at 6 months to 1 year of follow-up (RR 1.07, 95% CI 0.82–1.41).⁴³⁰ Both the RITA-3 and the FRISC-2 trials showed a higher rate of death and non-fatal MI among women.^431,432 In the TACTICS-TIMI-18 trial, no gender difference was observed with an invasive strategy. Revascularization compared with a conservative strategy improved the prognosis in women (OR 0.72; 95% CI 0.47–1.11), in the same range as in men (OR, 0.64; 95% CI, 0.47–0.88; P = 0.60 for sex interaction). The benefit of invasive therapy was enhanced in women with elevated troponin T levels (OR, 0.47; 95% CI 0.26–0.83).⁴³³ Better long-term outcomes have been demonstrated in unselected women with an early invasive strategy when compared with men.⁴³⁴ However, in the 5-year follow-up of the FRISC-2 trial, an invasive strategy was not shown to improve outcome in women [21.9 vs. 19.6% rate of death or MI invasive vs. conservative strategy (RR 1.12, 95% CI 0.83–1.50)], whereas a significant improvement was observed in men [19.0 vs. 26.8% rate of death or MI for invasive vs. conservative strategy (RR 0.70, 95% CI 0.59–0.86)], with significant interaction between women and men (P = 0.01).¹²² Lastly, in a more recent meta-analysis by the Cochrane collaboration, women were shown to derive significantly better long-term benefit than men in terms of death or MI (RR 0.73 95% CI 0.59–0.91) for invasive vs. conservative strategy, but with early hazard.³¹³ These conflicting results suggest that further randomized trials are required among women to establish whether a routine invasive strategy is of benefit. In the meantime, it is recommended that a routine early invasive strategy should primarily be considered among women who have high-risk criteria such as ongoing ischaemia and troponin elevation, taking into account the existing co-morbid illnesses.

Recommendations for women

• Women should be evaluated and treated in the same way as men, with special attention to co-morbidities (I-B).

7.3 Diabetes mellitus

The presence of diabetes mellitus is an independent predictor of higher mortality among patients with NSTE-ACS, and is associated with a two-fold higher risk of death when compared with non-diabetic patients,⁴³⁵ placing diabetic patients in a high-risk category. Diabetic patients have more co-morbid illnesses, including impaired renal function, heart failure, stroke, and general vascular disease.⁴³⁶ Overall, ∼20–30% of all patients with NSTE-ACS have diabetes and the vast majority has type 2 with insulin resistance. Data from recent registries carried out in the USA and Europe have shown that the rate of diabetes mellitus is increasing among patients with NSTE-ACS and ranges between 29 and 35% in Europe. Diabetes is more frequently observed in women than in men (41.6 vs. 30.7%). Diabetic patients are more often hypertensive (81 vs. 66% in non-diabetic patients) and obese (BMI > 30 is more frequent in diabetic patients than among non-diabetic patients, 28.5 vs. 18.6%), and suffer more frequently from renal failure (7.2 vs. 2.4% among non-diabetic patients).^437,438 When an established diagnosis of diabetes, impaired glucose tolerance, or impaired fasting glycaemia is considered, two-thirds of all patients suffering from either acute or chronic CAD have glucose regulation abnormalities.⁴³⁹ Patients with impaired glucose tolerance or impaired fasting glycaemia also have a worse prognosis than patients with no glucose regulation abnormalities, but a slightly better prognosis than patients with confirmed diabetes.

As patients with diabetes are at a higher risk for adverse events, a more comprehensive approach to primary and secondary prevention is recommended. Tight glycaemic control using intravenous insulin and glucose was shown in DIGAMI to reduce the 1-year mortality by 30% in STEMI patients.⁴⁴⁰ This early benefit was extended up to 39 months.⁴⁴¹ These observations were not confirmed in DIGAMI-2 which, however, showed that glucose level is a strong, independent predictor of long-term mortality following MI in patients with type 2 diabetes, with a 20% increase in long-term mortality for an increase of 3 mmol/L in plasma glucose.⁴⁴² Current knowledge indicates that insulin infusion is needed in diabetic patients with high blood glucose levels at admission in order to reach normoglycaemia as soon as possible. Moderate or minor elevation of blood glucose levels at admission may be handled with oral glucose-lowering agents. In the follow-up, strict glucose control is beneficial. Appropriate diet, lifestyle changes, oral agents, and insulin may be needed to achieve this goal. More detailed information on this issue can be found in the specific guidelines describing the management of diabetes and cardiovascular disease.³⁴⁰

In the case of angiography and/or angioplasty, the use of contrast medium raises the risk of contrast-induced nephropathy (CIN). Metformin should be interrupted ideally 24 h before the examination or at the latest on the day of the procedure. The risk of lactic acidosis is very low, but is increased in the case of renal failure. Metformin can be re-introduced 48 h after the use of contrast medium, if renal failure has not developed.

Similarly, an invasive treatment and potent antithrombotic therapeutic strategies are recommended. Both FRISC-2 and the TACTICS-TIMI-18 trials showed a 22–27% reduction in death and non-fatal MI in diabetic patients randomized to the early invasive strategy when compared with conservative strategy. Thus, the early invasive strategy is recommended for diabetic patients with NSTE-ACS. As many diabetic patients have multivessel CAD, CABG is frequently recommended based on the BARI trial.⁴⁴³ We await current trials that are examining the relative merits of the most appropriate invasive strategy for diabetic patients using either DES or CABG.

In the BARI trial (which was not specifically directed at NSTE-ACS), there was a survival advantage for patients with multivessel CAD who were randomized to CABG rather than PCI.^410,444 It has to be stressed that in the BARI trial, old technology was used, and it was more a comparison of PCI (without stents) and surgery. The use of modern technology in the PCI arm may produce different results. However, this was not observed in the most recent trial of CABG vs. PCI in the diabetic population with medical refractory unstable angina where the 3-year survival was not statistically different between CABG (72%) and PCI (81%).⁴⁴⁵ This trial differs from previous trials of PCI vs. CABG as it enrolled only medical refractory patients. Contemporary therapies with intracoronary stents and GP IIb/IIIa inhibitors were used extensively, but it was not powered enough to show statistical differences. Registry data show that an early invasive approach with a contemporary pharmacological environment and extensive use of stents significantly reduces in-hospital mortality when compared with a conservative approach.²⁵² The medical management with GP IIb/IIIa inhibitors in diabetic patients has also been examined in a meta-analysis.²³³ In 6458 diabetic patients enrolled in the six large NSTE-ACS trials of GP IIb/IIIa inhibitors, there was a 26% reduction in 30-day mortality (6.2% vs. 4.6%, OR 0.74; 95% CI 0.59–0.92; P = 0.007) as shown in Figure 12. Hence, diabetic patients with NSTE-ACS should receive intravenous GP IIb/IIIa inhibitors as part of the initial medical management that should be continued through the completion of PCI. It is worth mentioning that more recent data obtained in the setting of PCI do not confirm the meta-analysis data, since in elective PCI as well as in high-risk NSTE-ACS patients, abciximab was not shown to lead to higher benefit in diabetic patients.^188,446

Despite the evidence, it would appear that diabetic patients still remain undertreated when compared with their non-diabetic counterparts. In the European registries, revascularization (any form), thienopyridines, and GP IIb/IIIa inhibitors were prescribed less frequently among diabetic patients than among non-diabetic patients, with a clear impact on in-hospital and long-term mortality (5.9 vs. 3.2% mortality at 1 month and 15.2 vs. 7.6% at 1 year). In addition, registry data show that the presence of diabetes does not influence the choice of revascularization strategy.⁴⁴⁷ For a comprehensive review of the management of diabetes in cardiovascular disease, readers are referred to existing guidelines on this topic.³⁴⁰

Recommendations for diabetes

• Tight glycaemic control to achieve normoglycaemia as soon as possible is recommended in all diabetic patients with NSTE-ACS in the acute phase (I-C).

• Insulin infusion may be needed to achieve normoglycaemia in selected NSTE-ACS patients with high blood glucose levels at admission (IIa-C).

• An early invasive strategy is recommended for diabetic patients with NSTE-ACS (I-A).

• Diabetic patients with NSTE-ACS should receive intravenous GP IIb/IIIa inhibitors as part of the initial medical management which should be continued through the completion of PCI (IIa-B).

7.4 Chronic kidney disease

CKD is classified into five different stages (Table 9).⁴⁴⁸ Kidney function is best assessed with GFR according to the MDRD equation, which includes ethnicity and sex in its calculation. It should be assessed in all patients with or at increased risk of CAD.⁴⁴⁹ However, in daily clinical practice, CrCl is used instead of GFR. Cystatin C was shown to be a good surrogate marker of renal dysfunction.^93,94

7.4.1 Chronic kidney disease as a marker of risk of coronary artery disease

According to a large US registry, renal dysfunction has been shown to be quite common among the general population and is associated with a higher risk of cardiovascular and all-cause mortality, which increases exponentially with progressive decrease of GFR, with a swift increase in events for GFR < 60 mL/min/1.73 m².⁴⁵⁰ The risk of death from any cause, including cardiovascular disease, rises from an adjusted HR 1.2–5.1 from the least altered to the most severe levels of renal dysfunction, taking GFR > 60 mL/min/1.73 m² as a reference. The adjusted HRs for occurrence of any cardiovascular disease were 1.4 and 3.4, respectively.⁴⁵⁰ Similar observations were made in other reports,⁴⁵¹ some of which confirm that the prevalence of CAD is high at every stage of renal disease, including stage I, and results in a high rate of complications and in a two-fold increase in early mortality when compared with patients without renal dysfunction.^{412,451–453} The high prevalence of CAD in CKD patients is due to a high incidence of traditional and also non-traditional risk factors such a intense pro-inflammatory state, hyperhomocysteinaemia, and pro-thrombotic state.⁴⁵⁴ Diabetes, which accounts for about 50% of all cases of end-stage renal dysfunction, is an aggravating factor.⁴⁵⁵

Renal dysfunction is frequently observed in NSTE-ACS as in other forms of CAD. It is associated with worse prognosis in patients with overt clinical manifestations of atherosclerosis, including NSTE-ACS, STE-ACS, and PCI, as well as in diabetic patients.^11,456–461 In addition, renal dysfunction is a potent independent predictor of bleeding risk in patients with ACS; the more severe the dysfunction, the higher the bleeding risk (Figure 13) (see section 6.1 Bleeding complications).

The presence of renal dysfunction complicates the management of patients suffering from NSTE-ACS. In case of severe renal failure (CrCl < 30 mL/min), many drugs with exclusive or substantial renal elimination need to be down-titrated or might even be contraindicated, particularly LMWH, fondaparinux, bivalirudin and GP IIb/IIIa inhibitors. In this situation, UFH does not protect against bleeding complications, since in the GRACE registry, a gradual increase in the risk of bleeding with declining renal function was observed with UFH, similar to that observed with LMWH.³⁸² Since a much lower risk of bleeding complications was observed in OASIS-5 with fondaparinux compared to enoxaparin, even in patients with severe renal failure, fondaparinux has the potential to be used in this situation. Recent data confirm with eptifibatide that dose reduction may reduce bleeding risk⁴⁶² (Table 10).

7.4.2 Contrast-induced nephropathy

Baseline renal dysfunction may increase the risk of CIN in the case of angiography/angioplasty.⁴⁶³ Risk of CIN is particularly high in patients with older age, diabetes, dehydration, high volume contrast medium injection, and the use of high-osmolar, as opposed to non-ionic, low osmolar contrast medium. Hydration prior to and following angiography and/or angioplasty is the strategy that has been shown to have the greatest impact in reducing the risk of CIN.^464–468 Patients who need to undergo coronary angiography and/or angioplasty must receive special care in order to reduce or avoid CIN. Current protocols recommend hydration with 250–500 mL of sodium chloride 0.9% before and after the procedure, being cautious in those patients with a history of heart failure. The amount of contrast medium should be limited to a maximum of 50 mL for a diagnostic procedure. Assessment of creatinine level up to day 3 after contrast injection is necessary to detect CIN.

In patients with serious renal dysfunction, immediate angiography and/or revascularization may be postponed, unless clinically indicated, in order to reduce the risk of acute renal failure post-intervention. If PCI is required, consideration should be given to deferring it by several days after angiography, if clinically possible. In the case of multivessel PCI, staged procedures may be considered. CABG is associated with a raised risk of renal dysfunction so the risk–benefit ratio should be considered carefully.

7.4.3 Management of chronic kidney disease in patients with coronary artery disease

ACE inhibitors and ARBs have both been shown to reduce microalbuminuria and progression towards end-stage renal dysfunction. ACE inhibitors must be administered under strict monitoring of serum creatinine, which may initially increase when ACE inhibitors are introduced, and thereafter return to baseline in most patients. Their use is contraindicated in patients with renal artery stenosis. ARBs can be used as an alternative to ACE inhibitors. Only ACE inhibitors and statins have been shown to reduce the risk of cardiovascular events in patients with renal dysfunction and should therefore be used as for any other patient suffering from NSTE-ACS. Data on the impact of revascularization on outcome in patients with CKD are sparse, since in most trials, renal dysfunction was an exclusion criterion so that patients with CKD are under-represented.⁴⁶⁹ Revascularization was shown in a large registry as well as in substudies of trials in the setting of NSTE-ACS to improve outcome of CKD patients, not only at the stage of end-stage renal failure but also at the stage of moderate renal dysfunction.^458,470,471 In the same registry, as in others, it was shown that patients with renal dysfunction are often treated suboptimally and do not receive guideline-recommended therapy.⁴⁵⁸

7.4.4 Biomarkers in chronic kidney disease

Troponin elevations are sometimes found in asymptomatic patients with renal dysfunction, particularly those under haemodialysis, without clear evidence of ongoing NSTE-ACS. These troponin elevations may render the diagnosis of NSTE-ACS difficult in this setting. However, the prognosis of patients with CKD is impaired in the case of troponin elevation independent of the anginal status.^{77,78,472–474}

Recommendations for patients with CKD

• CrCl and/or GFR should be calculated for every patient hospitalized for NSTE-ACS (I-B). Elderly people, women, and low body weight patients merit special attention as near normal serum creatinine levels may be associated with lower than expected CrCl and GFR levels (I-B).

• Patients with CKD should receive the same first-line treatment as any other patient, in the absence of contraindications (I-B).

• In patients with CrCl < 30 mL/min or GFR < 30 mL/min/1.73 m², a careful approach to the use of anticoagulants is recommended, since dose adjustment is necessary with some, while others are contraindicated (I-C).

• UFH infusion adjusted according to aPTT is recommended when CrCl < 30 mL/min or GFR < 30 mL/min/1.73 m² (I-C).

• GP IIb/IIIa inhibitors can be used in the case of renal failure. Dose adaptation is needed with eptifibatide and tirofiban. Careful evaluation of the bleeding risk is recommended for abciximab (I-B).

• Patients with CKD with CrCl < 60 mL/min are at high risk of further ischaemic events and therefore should be submitted to invasive evaluation and revascularization whenever possible (IIa-B).

• Appropriate measures are advised in order to reduce the risk of CIN (I-B).

7.5 Anaemia

Anaemia has been shown to be associated with worse prognosis, and particularly higher mortality in various conditions, including heart failure, renal failure, various types of surgery, and malignancy,^475–481 but also across the whole spectrum of CAD, including STEMI, NSTE-ACS, PCI, and CABG.^391,482,483

According to the World Health Organization criteria (haematocrit < 39% or haemoglobin levels < 13 g/dL in men and < 12 g/dL in women),⁴⁸⁴ anaemia may be present in 5–10% of patients with NSTE-ACS.³⁸³ Figures as high as 43% were observed among elderly patients with acute MI, but only 4.2% had haematocrit levels < 30%.³⁹⁰ In a more recent report, anaemia was observed in 30.6% of cases of ACS, but only 5.4% had haemoglobin < 10 g/dL.³⁹¹

Anaemia in NSTE-ACS is associated with worse prognosis. A recent meta-analysis involving nearly 40 000 patients with both NSTE-ACS and STEMI showed that outcome at 30 days is significantly influenced by admission haemoglobin levels. The probability of cardiovascular death, MI, or recurrent ischaemia increases as haemoglobin falls below 11 g/dL, with an OR of 1.45 per 1 g/dL decrement in haemoglobin, taking those with a haemoglobin level of 15–16 g/dL as the reference group. The rate of cardiovascular events also increases with haemoglobin levels higher than 16 g/dL.³⁹¹ The same reverse J-shaped relationship between mortality and haemoglobin levels was observed in a cohort of 5888 elderly patients followed up for 11 years in the Cardiovascular Health Study.⁴⁸⁵ Other reports in different contexts, such as STEMI, PCI, and CABG, have also identified anaemia as a marker of worse prognosis.^390,483,486 Anaemia is associated with more co-morbidities, such as older age, presence of diabetes, and renal failure^482,483 but also non-cardiovascular conditions (haemorrhagic diathesis or malignancy), which may partly account for the adverse prognosis. However, after adjustment on a broad array of baseline characteristics, a dose–response relationship across the spectrum of ACS was observed; the more profound the baseline haemoglobin, the worse the prognosis.^391,485 There seems to be a causal link between anaemia and risk of cardiovascular death. Indeed, anaemia increases heart rate and cardiac output, leading to development of LV hypertrophy, and an imbalance between oxygen demand and supply to the myocardium. These mechanisms, in addition to compromised oxygen supply to infarcted or ischaemic myocardium, can lead to increased infarct size, development of arrhythmias, and may also aggravate hypotension, and eventually worsen prognosis.

Baseline haemoglobin was also shown to be an independent predictor of the risk of bleeding; the lower the baseline haemoglobin, the higher the risk, for both procedure-related and non-procedure-related bleeding.^383,487 Therefore, as modern treatment of NSTE-ACS may lead to a worsening of anaemia, because of increased risk of bleeding, special attention has to be paid to baseline haemoglobin level when deciding upon the therapeutic approach^179,488,489 (see section 6.1 Bleeding complications).

Recommendations for anaemia

• Low baseline haemoglobin is an independent marker of the risk of ischaemic and bleeding events at 30 days. It should be taken into consideration in assessing initial risk (I-B).

• All necessary measures should be taken during the course of initial management to avoid worsening of anaemia by bleeding (I-B) (see section 6.1 Bleeding complications).

• Well-tolerated anaemia at baseline in patients with NSTE-ACS should not lead to systematic blood transfusion which should be considered only in the case of compromised haemodynamic status (I-C) (see section 6.1 Bleeding complications).

7.6 Normal coronary arteries

A sizeable proportion of patients with NSTE-ACS have normal coronary arteries or only minor abnormalities. The pathophysiology of NSTE-ACS is not homogeneous and possible mechanisms include: (i) a coronary artery spasm (Prinzmetal's angina); (ii) an intramural plaque complicated by acute thrombosis with subsequent recanalization; (iii) coronary emboli; and (iv) syndrome X.

In patients admitted with suspected NSTE-ACS, the demonstration of normal or nearly normal coronary arteries at angiography challenges the diagnosis. However, ST-segment changes and release of biomarkers in patients with typical chest pain and patent coronary arteries without significant stenotic lesions may be due to true necrosis rather than false-positive results. About 15% of patients with proven NSTE-ACS actually have normal or nearly normal coronary arteries. This tends to be more common in women. Important atherosclerotic burden may be present even in the absence of angiographically significant stenoses because it may occur in a diffuse manner and lead to arterial wall remodelling in which the wall thickens and expands outwards without encroaching on the lumen.⁴⁹⁰ The prognosis of these patients is similar to that of patients with NSTE-ACS and significant coronary atherosclerosis, and they therefore merit optimal antithrombotic therapy and secondary prevention with antiplatelet agents and statins.⁴¹

Prinzmetal's variant angina refers to an unusual syndrome of cardiac pain secondary to myocardial ischaemia that is not precipitated by physical exertion or emotional stress, and is associated with transient ST-segment elevation.⁴⁹¹ The original hypothesis that it is the result of a coronary vasospasm has been convincingly demonstrated by coronary arteriography. The vasospasm causes a transient, abrupt, marked decrease in the diameter of epicardial coronary arteries, thus leading to severe myocardial ischaemia. Vasospasm may occur at sites of severe focal stenoses, but often in patients with an apparently normal vessel at angiography. Patients with variant angina tend to be younger than those with conventional NSTE-ACS and are often heavy smokers. The symptom is often severe and may be accompanied by syncope. Attacks of Prinzmetal's angina tend to be clustered between midnight and eight o'clock in the morning.^491,492 Spasm of epicardial coronary artery, leading to transmural ischaemia, is the diagnostic hallmark of Prinzmetal's angina. The spasm may be spontaneous or provoked by an acetylcholine or ergonovine test or hyperventilation. The mainstay therapy for Prinzmetal's angina is the administration of calcium antagonists, shown to be effective in preventing coronary spasm, alone or in combination with nitrates. They should be prescribed at maximally tolerated doses on a long-term basis.⁴⁹²

In rare cases, NSTE-ACS with normal or near-normal coronary arteriogram is linked to coronary embolism, due to atrial fibrillation or atrial flutter. As atrial fibrillation is often clinically unrecognized, the frequency of this mechanism of NSTE-ACS may be underestimated.⁴⁹³

The term ‘syndrome X’ is used to describe patients with angina precipitated by exercise, ST-segment depression on stress test, and a non-obstructed coronary artery at angiography. The chest pain may increase in frequency or intensity, or occur at rest. The patients may present with typical features of unstable angina.^494,495 The prognosis is usually excellent. The real cause of the syndrome has not been established, but it is most frequently associated with impaired endothelial-dependent arterial vasodilatation, decreased nitric oxide production, and increased sensitivity to sympathetic stimulation. There is growing evidence that such patients often have an increased response to pain. Because the prognosis is excellent, the most important therapy is reassurance and symptom relief, for which nitrates, beta-blockers, and calcium antagonists have been found to be effective.

Apical ballooning, recently described, may present clinically as NSTE-ACS and is characterized by normal coronary arteries at angiography accompanied by apical and sometimes medioventricular akinesis unrelated to the distribution of a coronary artery. This is typically fully reversible within weeks. The exact mechanism of this syndrome is presently unknown.^496,497

8. Management strategies

NSTE-ACS encompasses a heterogeneous spectrum of patients with different levels of risk in terms of death, MI, or recurrence of MI. In the following paragraphs, a stepwise strategy is outlined which is based on the above detailed analysis of the available scientific data and which should be applicable to most patients admitted with suspected NSTE-ACS. It must be appreciated, however, that specific findings in individual patients may result in appropriate deviations from the proposed strategy. For every patient, the physician must make an individual decision taking into account the patient's history (co-morbid illnesses, age, etc.), his/her clinical condition, findings during the initial assessment on first contact, and the available pharmacological and non-pharmacological treatment options.

8.1 First step: initial evaluation

Chest pain or discomfort will be the symptom that leads to the patient seeking medical attention or hospitalization. A patient with suspected NSTE-ACS must be evaluated in a hospital and immediately seen by a qualified physician. Specialized chest pain units provide the best and most expeditious care.⁴⁹⁸

The initial step is to assign the patient without delay to a working diagnosis on which the treatment strategy will be based. The criteria are the following:

• quality of chest pain and a symptom-oriented physical examination;

• assessment of the likelihood of CAD (e.g. age, risk factors, previous MI, CABG, PCI);

• ECG (ST-deviation or other ECG abnormalities).

On the basis of these findings which should be available within 10 min after first medical contact, the patient can be assigned to one of the three major working diagnoses:

• STEMI requiring immediate reperfusion;

• NSTE-ACS;

• ACS (highly) unlikely.

The treatment of patients with STEMI is covered in the respective guidelines.² The assignment to the category ‘unlikely’ must be done with caution and only when another explanation is obvious (e.g. trauma). Additional ECG leads (V₃R and V₄R, V₇–V₉) should be recorded, especially in patients with persisting chest pain.

Blood is drawn on arrival of the patient in hospital and the result should be available within 60 min to be used in the second strategy step. Initial blood tests must at least include: troponin T or troponin I, CK (-MB), creatinine, haemoglobin, and leukocyte count.

The assignment to the category NSTE-ACS will result in the second strategy step.

8.2 Second step: diagnosis validation and risk assessment

8.2.1 Diagnosis validation

After the patient is assigned to the group NSTE-ACS, intravenous and oral treatments will be started according to Table 11.

The first-line treatment should be made up of at least nitrates, beta-blockers, aspirin, clopidogrel, and anticoagulation, the type depending on the management strategy, i.e. urgent invasive, early invasive, or conservative (see Third step).

The further management will be based on additional information/data:

• routine clinical chemistry, particularly troponins (on presentation and after 6–12 h) and other markers according to working diagnoses (e.g. d-dimers, BNP, NT-proBNP);

• repeat, preferably continuous, ST-segment monitoring (when available);

• Echocardiogram, MRI, CT, or nuclear imaging for differential diagnoses (e.g. aortic dissection, pulmonary embolism);

• responsiveness to antianginal treatment;

• risk score assessment;

• bleeding risk assessment.

During this step, other diagnoses may be confirmed or excluded, such as acute anaemia, pulmonary embolism, and aortic aneurysm (see Table 4; section 4.3 Differential diagnoses).

8.2.2 Risk assessment

The treatment of the individual patient is tailored according to the risk for subsequent events which should be assessed early at the initial presentation as well as repeatedly thereafter in the light of continuing or repetitive symptoms and additional information from clinical chemistry or imaging modalities.

Risk assessment is an important component of the decision-making process and is subject to constant re-evaluation. It encompasses assessment of both ischaemic and bleeding risk. The risk factors for bleeding and ischaemic events overlap considerably, with the result that patients at high risk of ischaemic events are also at high risk of bleeding complications. Therefore, the choice of the pharmacological environment (dual or triple antiplatelet therapy, anticoagulants) has become critical, as has the dosage of the drugs. In addition, in cases where invasive strategy is needed, the choice of the vascular approach is very important, since the radial approach has been shown to reduce the risk of bleeding when compared with the femoral approach. In this context, particular attention has to be paid to renal dysfunction, shown to be particularly frequent in elderly patients and among diabetics.

During this step, the decision has to be made whether the patient should go on to cardiac catheterization or not.

8.3 Third step: invasive strategy

Cardiac catheterization is advised to prevent early complications and/or to improve long-term outcome (Figure 14). Accordingly, the need for and timing of an invasive strategy has to be tailored according to the acuteness of risk into three categories: conservative, early invasive, or urgent invasive.

8.3.1 Conservative strategy

Patients that fulfil all the below criteria may be regarded as low risk and should not be submitted to early invasive evaluation:

• no recurrence of chest pain;

• no signs of heart failure;

• no abnormalities in the initial ECG or a second ECG (6–12 h);

• no elevation of troponins (arrival and at 6–12 h);

Low risk as assessed by a risk score (see section 4.4 Risk stratification) can support the decision-making process for a conservative strategy. The further management in these patients is according to the evaluation of stable CAD.⁴⁹⁹ Before discharge, a stress test for inducible ischaemia is useful for further decision-making.

Patients who cannot be excluded by the above criteria should go on to cardiac catheterization.

8.3.2 Urgent invasive strategy

Urgent invasive strategy should be undertaken for patients who are early in the process of developing major myocardial necrosis escaping the ECG (e.g. occlusion of the circumflex artery) or are estimated to be at high risk of rapid progression to vessel occlusion.

These patients are characterized by:

• refractory angina (e.g. evolving MI without ST-abnormalities);

• recurrent angina despite intense antianginal treatment associated with ST-depression (≥2 mm) or deep negative T-waves;

• clinical symptoms of heart failure or haemodynamic instability (‘shock’);

• life-threatening arrhythmias (ventricular fibrillation or ventricular tachycardia).

In addition to the medication shown in Table 11, a GP IIb/IIIa inhibitor (tirofiban, eptifibatide) should be added in symptomatic patients bridging the time to catheterization.

8.3.3 Early invasive strategy

Most patients initially respond to the antianginal treatment, but are at increased risk and need early angiography. The timing depends on the local circumstances, but it should be performed within 72 h.

The following features indicate patients who should undergo routine early angiography:

• elevated troponin levels;

• dynamic ST or T-wave changes (symptomatic or silent) (≥0.5 mm);

• diabetes mellitus;

• reduced renal function (GFR < 60 mL/min/1.73 m²);

• depressed LVEF < 40%;

• early post-MI angina;

• PCI within 6 months;

• previous CABG;

• intermediate to high risk according to a risk score (Table 5).

A GP IIb/IIIa inhibitor (tirofiban, eptifibatide) should be added to the standard treatment prior to catheterization in the case of elevated troponins, dynamic ST/T changes, or diabetes provided there is no overt excessive bleeding risk.

The decision about the timing of catheterization must continuously be re-evaluated and modified according to clinical evolution and occurrence of new clinical findings.

8.4 Fourth step: revascularization modalities

If the angiogram shows no critical coronary lesions, patients will be referred for medical therapy. The diagnosis of NSTE-ACS may be reconsidered and particular attention should be paid to other possible reasons for symptoms at presentation before the patient is discharged. However, the absence of critical coronary lesions does not rule out the diagnosis if clinical presentation was suggestive of ischaemic chest pain and if biomarkers were positive. In this situation, patients should receive treatment according to recommendations in NSTE-ACS.

Recommendations for the choice of a revascularization modality in NSTE-ACS are similar to those for elective revascularization procedures. In patients with single-vessel disease, PCI with stenting of the culprit lesion is the first choice. In patients with multivessel disease, the decision for PCI or CABG must be made individually. A sequential approach with treating the culprit lesion by PCI followed by elective CABG may be advantageous in some patients.

The anticoagulant should not be changed for PCI. In patients pre-treated with fondaparinux, UFH must be added before PCI. In patients pre-treated with tirofiban or eptifibatide, the infusion should be maintained throughout the intervention. Patients untreated with GP IIb/IIIA inhibitors should preferably receive abciximab before PCI. There is less evidence for the use of eptifibatide or tirofiban in this setting.

If CABG is planned, clopidogrel should be stopped and surgery deferred for 5 days, if the clinical condition and the angiographic findings permit this.

If angiography shows no options for revascularization, owing to the extent of the lesions and/or poor distal run-off, freedom from angina at rest should be achieved by intensified medical therapy, and secondary preventive measures should be instituted.

8.5 Fifth step: discharge and post-discharge management

Although in NSTE-ACS most adverse events occur in the early phase, the risk for MI or death remains elevated over several months. Patients treated with early revascularization are at low (∼2.5%) risk for developing life-threatening arrhythmias, with 80% occurring during the first 12 h after onset of symptoms.⁵⁰⁰ Accordingly, monitoring of the patients beyond 24–48 h is not warranted.

Discharge from hospital depends on clinical and angiographic findings. Patients with NSTE-ACS should be hospitalized for at least 24 h after successful stenting of the culprit lesion.

Intense risk factor modification is warranted in all patients following the diagnosis of NSTE-ACS (see section 5.5 Long-term management).

9. Performance measures

Despite the presence of consistent European and national guidelines, substantial variations exist in adherence to guidelines both within and between countries. There are large variations in the use of diagnostic procedures and in the application of risk stratification. Similarly, there are great variations in the use of medical and interventional treatments and variations in the selection of patients for specific therapeutic strategies.^{10,11,252,501–503} Such variations in the application of evidence-based strategies, within and between countries, are associated with differences in outcome. Registry studies have examined the relationship between the use of evidence-based treatment strategies and event rates, and they provide clear support that improving guideline adherence will improve patient outcomes.^{14,15,504–508} Thus, priority needs to be given to improving the uptake of evidence-based guidelines.

Publication of guidelines may have a very limited impact on quality of care unless national and local audits and performance measures are adopted. A systematic, multidisciplinary approach is required combining education, and the identification, and resolution of logistic problems. A well-structured treatment process associated with continuous monitoring of performance indicators can improve outcomes.^7,509–514

Quality is a relative concept that requires comparison either with the performances of others or with the previous standards. The standard of care in a unit is a consequence of a large number of individual decisions and actions, by several care providers, for each patient. Estimates of quality need to be based on indicators of care that are measurable and relevant for the health-care provider and the patient. To permit temporal comparisons within a centre, and across centres, performance indicators need to be well defined and standardized.⁵¹⁰ The measurements need to be performed in similar patient populations, and with risk adjustment. This requires measurement of individual patient's risk characteristics, an adequate sample size, and robust statistical comparisons. In general, estimates of quality of care in individual units need to assess process of care (e.g. rates of usage of medications and interventions with Class I-A recommendations) rather than infrequent outcome events (mortality or MI). Reliable estimates of rates of death or MI require large populations and prolonged observation periods.

Estimates of quality can be applied to individual units, hospitals, regions, or nations. However, for performance indicators to change the process of care, they have to be applied consistently, locally, and based on current rather than historical performance. Thus, this necessitates repeated or continuous measurements and feedback of performance to individual care centres.^{7,252,503,508,509,511–514}

For patients with ACS, the CARDS data set (available from http//:www.escardio.org) is an appropriate and standardized ESC- and EU-recommended database for quality development.⁵¹⁰ The CARDS data set, or similar national data sets with continuous monitoring of the standards of treatment of ACS patients, has been implemented in a number of European countries.^7,508 These data sets have revealed almost as large variation within a country⁵⁰⁸ as between countries.^11,508 Nevertheless, such continuous registry programmes have contributed to impressive improvements in the standards of care and in outcomes.^{14,15,504–508}

Currently, the most useful performance indicators in individual centres for monitoring and improving the standards of care in non-STEMI ACS include the following.

• Class I-recommended antiplatelet and anticoagulation: use of aspirin, clopidogrel, and UFH/LMWH (enoxaparin)/fondaparinux/bivalirudin; use of GP IIb/IIIa inhibitors before and/or during early PCI procedures.

• Class I-recommended interventional therapy: use of early invasive procedures in intermediate–high risk patients.

• Risk stratification: use of the above treatment measures in target populations in accordance with risk stratification (using risk scores), in the absence of contraindications.

• Class I-recommended secondary prevention therapy: statins, beta-blockade in patients with reduced LV function, ACE inhibition, smoking cessation, glycaemic control, lifestyle changes.

Regional, national, and international audit programmes that include thousands rather than hundreds of patients can measure outcome events (i.e. recurrent MI and mortality) and can be used to evaluate the impact of improved adherence to guidelines on clinical outcomes.^{14,15,504–508} However, such large-scale programmes also require action at a local level. In each hospital, continuous monitoring of performance indicators is strongly encouraged to enhance the quality of treatment and to minimize unwarranted variations in evidence-based care. Consistent application of therapies based on robust evidence (e.g. Class I recommendation) may have larger effects on real-life cardiovascular health than those seen in selected trial populations, especially with the combined implementation of several effective treatment modalities. Such programmes have been successfully implemented in several countries including Sweden (RIKS-HIA Registry), the UK (MINAP Registry), Germany, Italy, and Israel on a regional basis, or in intermittent programmes in many other countries. These performance measure programmes are also proposed and developed by the ESC through the continuous ACS Registry within the Euro Heart Survey Programme.

Recommendations for performance measures

• Development of regional and/or national programmes to measure performance indicators systematically and provide feedback to individual hospitals is strongly encouraged (I-C).

10. Abbreviations

• ACC (American College of Cardiology)

• ACE inhibitors (Angiotensin-converting enzyme inhibitors)

• ACS (Acute coronary syndrome)

• ACT (Activated clotting time)

• ADP (Adenosine diphosphate)

• AHA (American Heart Association)

• aPTT (Activated partial thromboplastin time)

• ARB (Angiotensin receptor blocker)

• A-V (Atrioventricular)

• BMS Bare metal stent

• BNP (Brain natriuretic peptide)

• CABG (Coronary bypass graft surgery)

• CAD (Coronary artery disease)

• CARDS (Cardiology Audit and Registration Data Standards)

• CCS (Canadian Cardiovascular Society)

• CI (Confidence interval)

• CIN (Contrast-induced nephropathy)

• CK (Creatinine kinase)

• CKD (Chronic kidney disease)

• CK-MB (Creatinine kinase myocardial band)

• COX (Cyclo-oxygenase)

• CPG (Committee for Practice Guidelines)

• CrCl (Creatinine clearance)

• CT (Computed tomography)

• cTnT or cTnI (Cardiac troponin T or cardiac troponin I)

• DES (Drug-eluting stent)

• dL (decilitre)

• DPG (Diphosphoglyceric)

• DTI (Direct thrombin inhibitor)

• DVT (Deep vein thrombosis)

• e.g. (for example)

• ECG (Electrocardiogram)

• EDTA (Ethylenediamine triacetic acid)

• EF (Ejection fraction)

• ESC (European Society of Cardiology)

• EU (European Union)

• Factor-Xa (Activated factor-X)

• GFR (Glomerular filtration rate)

• GPIIb/IIIa inhibitors (Glypoprotein IIb/IIIa inhibitors)

• Hct (Haematocrit)

• HDL (High-density lipoprotein)

• HIT (Heparin-induced thrombocytopenia)

• HR (Hazard ratio)

• hsCRP (High-sensitive C-reactive protein)

• i.e. (that is)

• INR (International normalized ratio)

• IU (International units)

• kg (kilogram)

• LBBB (Left-bundle branch block)

• LDL (Low-density lipoprotein)

• LMWH (Low molecular weight heparin)

• LV (Left ventricular)

• LVEF (Left ventricular ejection fraction)

• MB (Myocardial band)

• MDRD (Modification of Diet in Renal Disease)

• METS (Metabolic equivalents)

• mg (milligram)

• MI (Myocardial infarction)

• mL (millilitre)

• mm (millimetre)

• MPO (Myeloperoxidase)

• MRI (Magnetic resonance imaging)

• mV (millivolt)

• NNT (Numbers needed to treat)

• NSAID (Non-steroidal anti-inflammatory drug)

• NSTE-ACS (Non-ST-elevation acute coronary syndromes)

• NSTEMI (Non-ST elevation myocardial infarction)

• NT-proBNP (N-terminal pro-hormone brain natriuretic peptide)

• OR (Odds ratio)

• PCI (Percutaneous coronary intervention)

• PDA (Personal digital assistant)

• PF4 (Platelet factor 4)

• RR (Risk ratio)

• STE-ACS (ST-elevation-acute coronary syndrome)

• STEMI (ST-elevation myocardial infarction)

• t-PA (Tissue plasminogen activator)

• TVR (Target vessel revascularization)

• UFH (Unfractionated heparin)

• ULN (Upper limits of normal)

• VKA (Vitamin K antagonist)

• VF (Ventricular fibrillation)

• VT (Ventricular tachycardia)

• VTE (Venous thrombo-embolism)

11. Trial acronyms

• ACUITY (Acute Catheterization and Urgent Intervention Triage strategy)

• ACUTE-2 (Antithrombotic Combination Using Tirofiban and Enoxaparin)

• ASPIRE (Arixtra Study in Percutaneous Coronary Interventions)

• BARI (Bypass Angioplasty Revascularisation Investigation)

• CAPRIE (Clopidogrel vs. Aspirin in Patients at Risk of Ischaemic Events)

• CAPTURE (Chimeric 7E3 Antiplatelet Therapy in Unstable Angina Refractory to Standard Treatment)

• CARDS (Collaborative Atorvastatin Diabetes Study)

• CHARISMA (Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management and Avoidance)

• CRUSADE (Can Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the ACC/AHA Guidelines)

• CURE (Clopidogrel in Unstable Angina to Prevent Recurrent Events)

• DIGAMI (Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction)

• EARLY-ACS (Early Glycoprotein IIb/IIIa Inhibition in patients with Non-ST-segment Elevation Acute Coronary Syndromes)

• ESPRIT (Enhanced Suppression of the Platelet IIb/IIIa Receptor with Integrilin Therapy)

• ESSENCE (Efficacy and Safety of Subcutaneous Enoxaparin in Non-Q-Wave Coronary Events)

• FRISC (Fast Revascularisation during InStability in Coronary artery disease)

• FRISC-2 (Fragmin and Fast Revascularisation during InStability in Coronary artery disease II)

• GRACE (Global Registry of Acute Coronary Events)

• GUSTO (Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries)

• GUSTO-2 (Global Use of Strategies To open Occluded coronary arteries II)

• GUSTO-4 (Global Utilization of Strategies To open Occluded coronary arteries IV)

• GUSTO-4-ACS (Global Utilization of Strategies To open Occluded coronary arteries trial IV in Acute Coronary Syndromes)

• HINT (Holland Interuniversity Nifedipine/metopropol Trial)

• HOPE (Heart Outcomes Prevention Evaluation study)

• ICTUS (Invasive versus Conservative Treatment in Unstable Coronary Syndrome)

• IMPACT-2 (Integrilin to Minimize Platelet Aggregation and Coronary Thrombosis II)

• INTERACT (Integrilin and Enoxoparin Randomized Assessment on Acute Coronary Syndrome Treatment)

• IONA (Impact of Nicorandil in Angina)

• ISAR (Intracoronary Stenting and Antithrombotic Regimen)

• ISAR-COOL (Intracoronary Stenting With Antithrombotic Regimen Cooling-Off)

• ISAR-REACT-2 (Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment 2)

• MATE (Medicine vs. Angiography in Thrombolytic Exclusion)

• MINAP (National Audit of Myocardial Infarction Project)

• OASIS (Organization to Assess Strategies for Ischaemic Syndromes pilot study)

• OASIS-5 (Organization to Assess Strategies in Acute Ischemic Syndromes 5)

• OASIS-6 (Organization for the Assessment of Strategies for Ischemic Syndromes 6)

• PRISM (Platelet Receptor Inhibition in Ischaemic Syndrome Management)

• PRISM-PLUS (Platelet Receptor Inhibition in Ischaemic Syndrome Management in Patients Limited by Unstable Signs and Symptoms)

• PROVE-IT (Pravastatin or Atorvastatin Evaluation and Infection Therapy trial)

• PURSUIT (Platelet Glycoprotein IIb/IIIa in Unstable Angina; Receptor Suppression Using Integrilin Therapy)

• RESTORE (Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis)

• RITA-3 (Randomized Intervention Trial of unstable Angina 3)

• STEEPLE (The Safety and Efficacy of Enoxaparin In PCI Patients In International Randomized Evaluation)

• SYNERGY (Superior Yield of the New strategy of Enoxaparin, Revascularization and Glycoprotein IIb/IIIa inhibitors)

• TACTICS–TIMI-18 (Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Strategy)

• TARGET (Tirofiban and Reopro Give Similar Efficacy Outcomes)

• TENACITY (Tirofiban Novel Dosing vs. Abciximab with Evaluation of Clopidogrel and Inhibition of Thrombin Study)

• TIMI (Thrombolysis in Myocardial Infarction)

• TIMI-11A (Thrombolysis in Myocardial Infarction Phase 11A)

• TIMI-11B (Thrombolysis in Myocardial Infarction Phase 11B)

• TIMI-3B (Thrombolysis in Myocardial Ischaemia Phase III)

• VANQWISH (Veterans Affairs Non-Q-Wave Infarction Strategies in Hospital)

Acknowledgements

We are indebted to Fiona Ecarnot for her invaluable support and editorial assistance during the preparation of the manuscript.

†

References

Author notes