Angina is a clinical syndrome characterized by pain or discomfort in any of the following body regions: chest, epigastrium, mandible, shoulder, dorsum, or upper limbs. It is typically triggered or aggravated by physical activity or emotional stress and attenuated with the use of nitroglycerin and its derivatives. Angina usually occurs in patients with CAD showing stenosis of at least one epicardial artery. However, it can also occur in patients with valvular heart disease, hypertrophic cardiomyopathy, and uncontrolled hypertension. Patients with normal coronary arteries and myocardial ischemia associated with spasms or endothelial dysfunction may also present with angina. In this respect, changes in the microcirculation, such as those that occur during left ventricular (LV) hypertrophy and syndrome X, can also lead to coronary insufficiency. This observation has attracted attention for years, particularly in women with angina but without obstructions on coronary angiography, and is known as syndrome X. In fact, even without arterial obstruction, both epicardial arteries and those involved in the microcirculation have inadequate flow, which is sufficient to cause myocardial ischemia. This can be diagnosed on exercise electrocardiography (ECG) or myocardial perfusion scintigraphy (MPS). This syndrome is interpreted as a disease of the coronary microcirculation and is caused by endothelial dysfunction or changes in the vascular tone, resulting in decreased oxygen supply at the cellular level. Recently, it has been shown that subjects with this syndrome have atherosclerotic plaques in various segments where coronary obstruction is not present, with a frequency of up to 55%-60% on coronary intravascular ultrasound (IVUS)12. It is known that changes in vascular reactivity occur in this condition13^,14, leading to myocardial ischemia, particularly in women, as reported in the WISE study15. In addition, this same study revealed that, even in the absence of obstructions, changes in coronary reactivity were a risk factor for future coronary events in these female patients, although the possibility of atherosclerosis in epicardial vessels in the absence of obstruction was not discarded. Another similar condition is the occurrence of low coronary flow16. Anatomical abnormalities, e.g., anomalous origins of coronary arteries, can also lead to CAD. Furthermore, a decrease in coronary flow may also occur in the presence of changes in vascular tone known as coronary artery spasms. These spams can alter the degree of obstruction of the vessel lumen and even lead to (or promote) coronary artery occlusion, and consequently to a clinical manifestation of MI, especially because these changes occur in patients with moderate or significant atherosclerotic lesions. When ECG changes occur, this condition is designated Prinzmetal's angina. However, in a few patients, we observe angiographically normal coronary arteries. In addition, many chest pain conditions, or symptoms manifested in other common regions, are identified in other diagnoses, including complications of the esophagus, stomach, lung, mediastinum, pleura, and chest wall. After the diagnosis of heart disease is discarded, the recommendations for the management of these patients are beyond the scope of this guideline.

a) Clinical history: Clinical examination is one of the most important steps in the evaluation of patients with chest pain, because it helps the doctor estimate the probability of significant CAD with a high degree of accuracy9; angiographically, CAD is defined by the presence of stenosis ≥ 70% of the diameter of at least one segment of a major epicardial artery, or stenosis ≥ 50% of the diameter of the left main disease (LMD). Although lesions with a lower degree of stenosis can cause angina, they have a lower prognostic significance8. A clinical history with a detailed assessment of symptoms enables physicians to adequately characterize chest pain. Some symptoms should be carefully investigated to determine the likelihood of occurrence of angina:

Many symptoms are reported by patients when describing angina, including suffocation, burning, distress, and heaviness. They frequently complain of discomfort but not precordial pain. Angina is rarely reported as stabbing pains and generally has no association with breathing or the decubitus position. Typically, the anginal episode lasts a few minutes. It is generally precipitated by exercise or emotional stress, with frequent improvement or relief at rest. The use of nitroglycerin compounds, such as sublingual nitrate, can relieve angina in approximately 1 min. A sudden, brief, or continuous discomfort, which can last several hours, is rarely angina. Angina usually affects the retrosternal region and radiates to the neck, mandible, epigastrium, and upper limbs. Pain located in the chondrosternal joints rarely has cardiac origin.

Several classification systems have been proposed and the most commonly used divides chest pain into three groups: typical, atypical, and noncardiac17 (Chart 1). It can also be classified according to its severity (Chart 2). Angina can also be stable or unstable. It is important to diagnose unstable angina because of its close association with acute coronary events. Unstable angina can be divided into three groups according to certain clinical characteristics: at rest, recently emerged, and in development (Chart 3).

b) Physical examination: Physical examination is usually normal in patients with stable angina18. However, during the anginal episode, it can provide important clues as to the presence or absence of CAD. When a physical examination is performed during a pain episode, findings such as the third heart sound (S3), fourth heart sound (S4) or gallop, mitral regurgitation, paradoxical splitting of the second heart sound (S2), and bibasilar lung rales are suggestive symptoms and predictors of CAD19^,20. The occurrence of atherosclerosis in other body regions, including decreased pulse in the lower limbs, arterial hardening, and abdominal aneurysm, increase the likelihood of CAD. Moreover, other findings, including high blood pressure, retinal exudates, and xanthomas, suggest the presence of risk factors for CAD. Muffled heart sounds and facial flushing may indicate diseases of the pericardium and/or pleura adjacent to the heart. Even if the physical examination does not indicate direct or indirect signs of CAD, a complete clinical examination should be carefully performed, especially that of the cardiovascular system, because it can help elucidate other associated conditions, including valvular diseases and hypertrophic cardiomyopathy. Chest wall palpation often reveals the pain location in the patients with musculoskeletal syndromes. However, pain can also occur in patients with typical angina.

In all patients, particularly those with typical angina, associated diseases that can precipitate "functional" angina in the absence of significant anatomical coronary obstruction (e.g., myocardial ischemia) should be considered. These diseases generally cause myocardial ischemia by increasing the oxygen consumption of the myocardium or decreasing the oxygen supply to it (Chart 4)18^-21. The increased oxygen consumption can be produced during hyperthermia, hyperthyroidism, and cocaine use. Hyperthermia, particularly if accompanied by decreased blood volume due to sweating or loss of other body fluids, can precipitate angina, even in the absence of significant CAD. Hyperthyroidism, and its associated tachycardia and high metabolic rate, can increase oxygen consumption and decrease oxygen supply. It is important to emphasize that older patients may not present clinical signs typical of thyrotoxicosis; therefore, this possibility should always be considered, particularly among older people. Sympathomimetic toxicity, of which cocaine use is the most common, not only increases oxygen consumption but also decreases its supply by simultaneously provoking coronary spasms, which may lead to MI in young patients. Over the long term, cocaine use can lead to angina by causing CAD prematurely22. Angina may occur in patients with uncontrolled hypertension owing to increased LV wall stress, decreased coronary flow reserve (ability to increase coronary flow during physiological and pharmacological stress), and increased LV end-diastolic pressure, and the latter can decrease subendocardial myocardial perfusion. These same mechanisms contribute to anginal symptoms in patients with aortic valve stenosis and hypertrophic cardiomyopathy. Moreover, both ventricular and supraventricular sustained tachycardia can increase oxygen consumption. Paroxysmal tachycardia is among the conditions that most often contribute to angina and is usually difficult to diagnose. Conditions that decrease the oxygen supply should also be considered during differential diagnosis or the diagnosis of aggravating diseases in anginal patients. Anemia decreases the oxygen-carrying capacity of the blood and increases cardiac overload. Increased cardiac output usually occurs in anemic patients with hemoglobin levels < 9 g/dL, and ST-T changes (depression or inversion) can occur when hemoglobin levels reach < 7 g/dL. Hypoxemia resulting from pulmonary diseases such as pneumonia, asthma, chronic obstructive pulmonary disease, pulmonary hypertension, interstitial fibrosis, or obstructive sleep apnea can also precipitate angina. Obstructive sleep apnea should be seriously considered in patients with significant nocturnal symptoms.

Conditions associated with increased blood viscosity may increase coronary resistance and consequently decrease coronary artery blood flow, which can precipitate angina in patients with significant coronary artery stenoses. Increased blood viscosity is observed in medical conditions, including polycythemia, leukemia, thrombocytosis, and hypergammaglobulinemia. In addition, other diagnoses should be considered during the assessment of the patient's medical history because these associated conditions may be the cause of the symptoms experienced by the patient. To differentiate these diagnoses, it is necessary to evaluate all conditions (Chart 5) and to identify which of these conditions may differ from angina.

ECG has limited importance in chronic CAD because repolarization changes do not necessarily indicate CAD and can be associated with other causes, including LV hypertrophy, electrolyte abnormalities, left bundle-branch block (LBBB), and T-wave changes, among others. Therefore, considering this limitation, a normal ECG does not exclude the possibility of coronary obstruction. However, ECG has diagnostic importance: (1) presence of Qr or QS waves plus negative T wave suggest the diagnosis of previous MI; (2) changes in the ventricular repolarization are suggestive of subepicardial ischemia (negative, sharp, and symmetric T waves) in a specific myocardial region: anteroseptal (V1, V2, V3, V4), anterolateral (V4, V5, V6, DI, and aVL), high lateral (DI and aVL), extensive anterior (V1-V6 in DI, and aVL), inferior (D2, D3, and aVF), and dorsal (V7 and V8 with reciprocal image in V1, V2, and V3); (3) changes in ventricular repolarization are suggestive of subendocardial ischemia (positive, sharp, and symmetrical T waves) in a specific region (anteroseptal, anterolateral, high lateral, extensive anterior, inferior, and dorsal); (4) changes in the ventricular repolarization are suggestive of subendocardial injury (depression of the J point and ST-segment, with superior concavity of this segment in leads exploring the injury) in a specific region (anteroseptal, anterolateral, high lateral, extensive anterior, inferior, and dorsal).

Therefore, ECG is indicated for patients with suspected cardiac cause of chest pain (Grade of recommendation I, Level of evidence B) or during an episode of chest pain (Grade of recommendation I, Level of evidence B).

Chest radiography is usually the first imaging modality performed in patients with chest pain, with the main purpose of performing a differential diagnosis of angina; it can be diagnosed in patients with pneumothorax, pneumomediastinum, rib fractures, and acute infections. Other conditions that cause acute chest pain with noncardiogenic etiology, including aortic aneurysm, aortic dissections, and pulmonary embolism, can be diagnosed with CR. However, the sensitivity of the method for the diagnosis of these conditions is very low. Therefore, CR is indicated for patients with CAD and signs or symptoms of congestive heart failure (Grade of recommendation I, Level of evidence B), patients with signs and symptoms of lung disease (Grade of recommendation IIa, Level of Evidence B), and for other conditions (Grade of recommendation IIb, Level of evidence B).

Exercise treadmill test (ETT) is a noninvasive method used most frequently in stable angina, aiming to confirm diagnosis, determine prognosis, and define the therapeutics. For ETT interpretation, the clinical responses related to symptoms and functional capacity, and ECG and hemodynamic profiles, need to be considered. The most predictive variables related to the diagnosis of coronary obstruction are ST-segment depression ≥ 1 mm (measured at 0.80 s from the J point), with a horizontal or downsloping pattern, and the presence of anginal pain. For the diagnosis of myocardial ischemia, the test results should be correlated with the pretest probability of CAD. In revascularized patients, particularly those who present ECG with baseline changes in the ST segment, hemodynamic and clinical profiles, and the functional capacity, should be taken into consideration during the test. Imaging tests (scintigraphy, echocardiography, or stress cardiac magnetic resonance imaging) are recommended for these patients for the detection of the presence and location of residual ischemia. Recently, novel analysis criteria, including QT dispersion, have been used in revascularized group for the diagnosis of residual ischemia after revascularization, with some improvment in the sensitivity and specifivity of ETT26.

Echocardiography is an important test for confirming the diagnosis and evaluating the prognosis in patients with chronic CAD27^,28. Echocardiography can provide valuable diagnostic aid in elucidating reversible or irreversible segmental wall motion abnormalities in patients with clinical CAD, particularly when the clinical history and ECG are inconclusive. Considering that echocardiography allows a real-time assessment of the LV wall motion, tests using physical or pharmacological stress, whether inotropic or vasodilator, allow the evaluation of the extent and severity of transient changes in LV wall motion. Echocardiography with microbubble-based ultrasound contrast represents a breakthrough in the diagnosis of these patients. These microbubbles are approximately 3 µm in diameter and behave as red blood cells in the bloodstream, and can map the entire tissue perfusion using ultrasound. The microbubbles fill the LV cavity, allowing the accurate evaluation of segmental wall motion abnormalities and, after coronary sinus filling, the evaluation of the intramyocardial blood flow, i.e., myocardial perfusion29^-31.

a) Role of transthoracic echocardiography in the diagnosis of coronary atherosclerosis and its complications: Transthoracic echocardiography is an excellent screening method in patients with CAD during acute events, because segmental wall motion abnormalities occur seconds after coronary occlusion and are reliable markers of previous MI. Although these abnormalities can indicate previous ischemia or infarction rather than acute infarction, they help eliminate other causes of chest pain, e.g., aortic dissection, pericarditis, and massive pulmonary embolism (Chart 6). In the stable patient, the anatomical information is important but not necessary in the routine of all cases. Echocardiography at rest provides a lot of information on the LV function, including the myocardial viability, with important therapeutic and prognostic implications after AMI. The wall motion score is obtained from the echocardiographic division of LV into 16 segments, and values between 1 and 4 are assigned to each segment according to the degree of motility. This finding is valuable in establishing the degree of LV dysfunction, especially because it is better correlated with the total mass involved in the process of ischemic injury compared with the actual ejection fraction (EF), which may be overestimated. A value of "1" indicates a normal contractile movement. Subsequently, hypokinesia, akinesia, and dyskinesia are assigned other values. The LV wall motion score index (LVWMSI) is calculated by the sum of each of 16 segments with motion dysfunction. A score between 1 and 1.6 indicates normal or slightly impaired ventricular function, a score between 1.61 and 2.0 indicates moderate stenosis, and a score > 2.0 corresponds to significant stenosis. The complete echocardiographic study, i.e., with color Doppler flow mapping, is critical for assessing complications, such as diastolic dysfunction, mitral regurgitation, interventricular communication, pericarditis, aneurysm, and pseudoaneurysm. It is the method of choice for the differential diagnosis of diseases, including aortic stenosis, hypertrophic cardiomyopathy, and mitral valve prolapse27^-30.

b) Use of stress echocardiography in chronic coronary atherosclerotic disease: Stress echocardiography is a noninvasive method used to evaluate patients with suspected or known obstructive CAD, make the diagnosis and assess the prognosis, assess the impact of revascularization therapies, detect myocardial viability, and aid in therapeutic decisions. Cardiovascular stress causes myocardial ischemia in regions supplied by an artery with a significant degree of stenosis, and this phenomenon is manifested by transient changes in segmental contraction. Two-dimensional echocardiography allows the evaluation of all myocardial segments of LV with high spatial and temporal resolution, making it an ideal strategy for the noninvasive evaluation of myocardial ischemia. The methods available for stress induction include physical exercise (treadmill or stationary bike), transesophageal atrial pacing, and the use of vasodilators (dipyridamole, adenosine), or adrenergic stimulants (dobutamine). Stress echocardiography shows good accuracy in the detection of myocardial ischemia induced in patients with intermediate or high pretest probability, and has greater sensitivity and specificity for the diagnosis of CAD than ETT Dobutamine stress echocardiography and exercise stress echocardiography have similar diagnostic accuracy (83% and 85%, respectively), whereas dipyridamole stress echocardiography seems to have a slightly lower diagnostic accuracy. This difference can be attributed to a lower sensitivity of dipyridamole in identifying patients with one-vessel diseases (38% for dipyridamole stress echocardiography, in contrast with 70% for exercise stress test and 61% for dobutamine stress echocardiography). The addition of atropine to dobutamine in stress echocardiography improves accuracy and decreases the rate of ineffective tests, particularly in patients using beta-blockers31^-34.

c) General Indications of stress echocardiography: The choice of the type of stress to which the patient will be subjected must be based on the aim of the test and associated clinical conditions, considering the specific contraindications of each method31^-33. Stress echocardiography is indicated to evaluate myocardial ischemia in symptomatic patients, in cases when ETT is not diagnostic, and to evaluate ischemia in patients with a clinical presentation not suggestive of coronary insufficiency and for those with positive or doubtful ETT. In patients with increased clinical suspicion of CAD, stress echocardiography is valuable in the identification of concurrent conditions that may not be efficiently diagnosed with ETT including changes in the ST segment and T wave at rest, LBBB, LV hypertrophy, ventricular pacemaker rhythms, and digitalis therapy. Similarly, stress echocardiography is not recommended for the initial evaluation of asymptomatic patients without CAD. Stress echocardiography may be important in the clinical management of the patient but is not indicated for the routine periodic evaluation of stable patients with unchanged clinical status.

d) Preoperative evaluation: According to the recommendations of ACC/AHA and the European Association of Cardiovascular Imaging (EACVI), dobutamine stress echocardiography has become an invaluable tool for risk stratification of preoperative patients with CAD35^,36 (Chart 7). In addition, it is recommended for the evaluation of patients with one or more clinical cardiovascular risk factors and limited ability to perform physical activities. Several studies have demonstrated the importance of this method in preoperative risk stratification in patients subjected to vascular surgery. A normal test shows a negative-predictive value (NPV) between 93% and 100% for cardiovascular events, and patients with negative test results may undergo surgery after other examinations. The detection of segmental contraction abnormalities has a positive-predictive value (PPV) between 7% and 30%, which is similar to the results found using thallium-201 perfusion scintigraphy37.

Nuclear cardiology evaluates the heart and focuses on aspects related to myocardial perfusion, cellular integrity, myocardial metabolism, myocardial contractility, and global or segmental ventricular function. The limited availability of equipment and radiotracers (e.g., thallium-201, technetium-99m, isonitrile, and tetrofosmin) can restrict the large-scale use of nuclear methods.

Myocardial perfusion studies are important in the diagnosis of ischemic heart disease because this method is noninvasive, virtually free of adverse reactions to the radiotracer, and is easily administered to patients. By using single-photon-emission CT (SPECT), it is possible to diagnose CAD with high sensitivity and specificity. Current ECG synchronization and regional quantification techniques associated with tomographic studies allow the analysis of concomitant data on myocardial perfusion, wall motion, and overall LV function, thereby significantly increasing the diagnostic power of the method. This is particularly important when the method has limitations, as in cases of obstructive disease isolated from the circumflex artery and in patients with multivessel disease, and can increase the sensitivity > 90% in these cases38^,39. It should be emphasized that the method specificity - usually between 80% and 90%40 - is affected by underestimations due to bias, because most patients with normal perfusion scintigraphy are not referred to the gold standard coronary angiography examination. Another important aspect about the exam, highlighted in recent publications, is related to false-positive scintigraphy results associated with moderate reversible defects detected in patients who do not present >70% obstructions in CA. In most of these occasions, intracoronary ultrasound examination indicates that even angiographically insignificant injuries cause major changes in the vasodilatory capacity of the coronary circulation and can potentially cause ischemia and infarction. Furthermore, special attention should be given to the detection of artifacts, especially when using tomographic techniques, to minimize problems with attenuations, movements41, and interference of the intestinal loops. Sensitivity and specificity values are equivalent in scintigraphic studies using tracers labeled with 99mTc or 201Tl. Considering use of lowest doses, best-quality images, and ease of handling, the examinations with tracers labeled with 99mTc are indicated as the first choice for investigations on ischemia42^-44. Labeling with 201Tl is less used because of its association with increased radiation and is indicated for investigations on ischemia associated with viable myocardium. New SPECT cameras have significantly decreased radiation and image acquisition time45. The indications for the use of scintigraphy are shown in Chart 846.

Coronary lesions are significant when there is obstruction of one or more epicardial arteries with at least 70% stenosis and/or that of LMD with at least 50% stenosis, and such obstructions are evaluated and measured using CA, which is a diagnostic test with low rates of complications47.

Some patients need to undergo invasive tests because it is the most accurate method for the diagnosis of obstructive coronary lesions. In addition, in some unusual cases, nonatherosclerotic causes for angina, including coronary spasm, coronary anomaly, Kawasaki disease, and primary coronary dissection, may exist.

However, in most cases, noninvasive tests are performed first, as already explained. Analysis of the symptoms as an initial diagnostic method may have an important role in special cases, such as in cases with angina, those in which noninvasive tests are contraindicated or their benefits are negligible, those with severe illness, those having physical disabilities that limit the use of noninvasive methods, and those without a suitable patient profile. Invasive tests are also reasonably indicated for patients at high CAD risk, those with noninvasive tests presenting conflicting results, those with poorly diagnosed conditions, or even those whose incapacity can affect the general population, e.g., aircraft pilots, firefighters, and professional athletes. Other groups require special considerations. Recent studies suggest that women with positive exercise stress test and thallium stress test results are less frequently indicated for additional noninvasive tests than men (4% vs. 20%, respectively) and invasive tests (34% vs. 45%, respectively)48^-50. The causes of these differences and how they affect diagnosis are uncertain51^,52. In addition, the assessment of chest pain in older patients may be difficult48^,49^,52^,53 because complaints of chest pain, fatigue, dyspnea, and comorbidities with angina-mimicking symptoms are common, which results in decreased evaluation of ischemic symptoms with advanced age. The increased frequency of abnormal resting ECG and the difficulty in performing physical activities also limit the results of noninvasive tests. Furthermore, the high prevalence of the disease in this population group decreases the importance of negative results of noninvasive tests. In contrast, the diagnostic CA poses little risk to older patients compared with younger patients. Under these conditions, some authors54 prefer invasive tests for this population group. Other patients need to undergo CA with ventriculography to determine prognosis, extent of CAD, and degree of LV dysfunction, because these are the main determinants for the long-term results55^-58. The simplest and most widely used method for assessing the extent of CAD classifies patients as those with one-vessel, two-vessel, and three-vessel disease or as those with LMD lesions59^-61. The survival time decreases with the involvement of additional vessels, stenosis of the left anterior descending artery (LAD), and LV dysfunction56^,62^,63 (Table 2).

Furthermore, patients who are candidates for revascularization with angioplasty or surgery are indicated for CA. Therefore, the assessment of the coronary vasculature is necessary to determine whether this procedure is indicated. This group includes patients with angina, those with clinical evidence of heart failure and those who have experienced cardiac arrest or serious ventricular arrhythmia. In such cases, the performance of CA as a first option is indicated. Most of these cases include chronic anginal symptoms, which are not included in the above categories. It is necessary to decide whether these patients should either undergo revascularization or simply initiate clinical treatment and undergo revascularization in situations of clinical failure. Therefore, it is important to perform CA along with left ventriculography, when revascularization is proposed, aiming to improve survival. CA is effective when the prognosis related to clinical treatment is poor and when the prognosis improves after revascularization with either angioplasty or surgery. Chart 9 shows the recommendations for coronary angiography in patients with CAD.

Cardiac computed tomography (CT) offers two main modes of examination, which employ different techniques and provide different information: the calcium score (CS) and coronary CTA.

In recent years, CMR has become a reliable diagnostic method for the evaluation of a variety of cardiovascular diseases. It allows the accurate and reproducible assessment of cardiac and vascular anatomy, ventricular function, myocardial perfusion, and tissue characterization, and all this information is available after a single examination42. Moreover, it uses neither ionizing radiation nor contrast medium, which have a greater potential for nephrotoxicity. Its diagnostic versatility and accuracy make it a highly attractive method for the evaluation of various heart diseases, but it is primarily indicated for ischemic heart disease in clinical practice.

Magnetic resonance imaging is a process based on capturing the energy released by protons (hydrogen nuclei) subjected to the action of a strong external magnetic field and stimulated by repeated radiofrequency (RF) waves. On the basis of the organization of RF pulses (pulse sequences), it is possible to obtain high-quality images with exquisite anatomical detail and contrast resolution, allowing the investigation of different molecular properties.

CMR is now considered to be the gold standard method for the quantification of ventricular volume, EF, and myocardial mass103. This is mainly owing to its ability to visualize the entire cardiac anatomy with high spatial and temporal resolution and to provide enhanced detail of the endocardial and epicardial borders of the left and right ventricles. Furthermore, CMR allows the application of the Simpson's formula without major technical obstacles, making this method extremely accurate for the assessment of global and segmental biventricular functions. With particular regard to ischemic heart disease, CMR can be used for the assessment of myocardial ischemia, myocardial fibrosis/infarction/viability, and for the noninvasive assessment of coronary arteries103.

A team composed of a cardiologist, hemodynamicist, and surgeon is recommended to individualize the decision about the best treatment for patients with LMD lesions or complex CAD. Level of Evidence C 266.

The calculation of the Syntax and STS scores is suitable for patients with LMD lesions or complex CAD. Level of Evidence B 46^,266

Most patients with stable angina can control symptoms using antianginal drugs, PCI, and/or CABG. However, a select group of patients with extensive coronary atherosclerosis and persistence of symptoms despite drug therapy is not eligible for percutaneous or surgical revascularization. In this group, classified as having refractory angina, TLRS and percutaneous myocardial laser revascularization (PMLR) have emerged as therapeutic options.

TLRS, described by Mirhoseini el al.271 in 1983, was introduced in Brazil by Galantier et al.272 in 1995 and was performed after lateral anterior thoracotomy without extracorporeal circulation. Most studies have used three types of high-energy lasers (Holmium YAG laser, CO₂ laser, and XeCl excimer laser) capable of creating transmural channels with approximately 1 mm in diameter, from the epicardium to the LV endocardium, distributed in the area of the ischemic myocardium. TLRS can be associated with myocardial revascularization and venous or arterial grafts. At first, the mechanism proposed was that these microchannels would be responsible for the direct perfusion of the ischemic myocardium. However, it was shown that microchannels were occluded by necrotic and scar tissue several weeks after the procedure273^,274. Other hypotheses intended to explain the mechanism of action involved the emergence of new blood vessels (angiogenesis) and/or denervation of sympathetic epicardial myocardial fibers, although an associated placebo effect cannot be discarded273^,274.

Several observational studies have been conducted to evaluate the efficacy and safety of TLRS in patients with refractory angina. Most observational studies have shown decreased angina, improvement in the functional capacity, and increased exercise tolerance. In contrast, the myocardial perfusion imaging results during patient recovery are conflicting. Of note, the surgical mortality varied between 3% and 20% and one-third of the patients had complications associated with the procedure274^,275.

Seven prospective, randomized, and open studies compared TLRS with optical coherence tomography (OCT) available at that time276^-281. Recently, a systematic review of the Cochrane database reviewed these studies, which involved 1,137 participants, including 559 patients who were randomized for TLRS282. This review indicated that none of the studies met all the established quality criteria primarily because they were open studies with great potential for bias in the evaluation of the primary outcome - angina relief.

No significant differences in survival after the 12-month follow-up period were observed between the two groups. Approximately 44% of the patients randomized for TLRS experienced a lowering of at least two angina classes [established by the Canadian Cardiovascular Society (CCS) and New York Heart Association (NYHA)] in contrast to only 15% of the group subjected to OMT (odds ratio (OR) = 4.63, 95% CI: 3.43-6.25). With regard to exercise tolerance, despite the overall improvement at 12 months, no significant differences were observed between the two treatments. Only two studies evaluated the quality of life of patients and both indicated only a slight improvement in the quality of life in the TLRS group when evaluated with the Seattle Angina Questionnaire. Most patients in both treatments experienced improved myocardial perfusion during recovery, but this improvement did not differ significantly between the TLRS and OMT groups in six of the seven studies evaluated.

Furthermore, there was no significant difference in mortality at 12 months (12.2% vs. 11.9% for the TLRS and OMT groups, respectively; OR = 1.12; 95% CI: 0.77-1.63). However, hospital mortality assessed at 30 days was significantly higher in the group subjected to surgery (6.8% vs. 0.8% in the TLRS and OMT groups, respectively; OR = 3.76; 95% CI: 1.63-8.66). A systematic review concluded that the clinical benefits of TLRS do not outweigh the risk of the procedure and that the procedure was associated with increased hospital mortality. Moreover, randomized double-blind studies with sham procedures are needed to avoid evaluation biases in the perception of angina in the clinical group.

Recently, NICE analyzed these randomized trials and highlighted that, although the patients subjected to TLRS achieved better subjective outcomes, including exercise tolerance, angina score, and quality of life, than patients subjected to OMT, these results contradict the increased perioperative morbidity and mortality, indicating an unfavorable risk/benefit ratio. These studies conclude that TLRS alone should not be used in patients with refractory angina283.

A single prospective, randomized, single-center, and double-blind study compared CABG alone with CABG associated with TLRS in patients with refractory angina. This study involved 263 patients and indicated no significant differences in the relief of angina and exercise tolerance between the two methods in the short term. However, the operative mortality was significantly lower in the group subjected to CABG and TLRS (1.5% vs. 7.6%; p < 0.05). After a 5-year follow-up period, although both groups had improved angina relief compared with baseline values, the CABG/TLRS group had a lower angina score (0.4 ± 0.7 vs. 0.7 ± 1.1; p = 0.05) and lower proportion of angina III/IV than the group subjected only to CABG [0% (0/68) vs. 10% (6/60), p = 0.009]. The survival after 5 years was similar between the two groups (76% vs. 80%; p = 0.90). The authors concluded that, in selected patients, CABG associated with TLRS might be superior to isolated CABG in relieving refractory angina284.

In 1997, Kim et al.285 demonstrated the formation of microchannels through the endocardial surface using a system of catheters introduced through the femoral artery puncture. PMLR emerged as an alternative to TLRS with the prospect of maintaining the supposed benefits observed in angina relief but with decreased rates of morbidity and mortality and without the need for thoracotomy. The mechanism of action would be the same proposed by PMLR (microchannels, microvascular angiogenesis, cardiac denervation, and placebo effect).

Moreover, some observational studies demonstrated the efficacy of PMLR in patients with refractory angina, including a lower angina class, increased exercise tolerance, and improved quality of life, but without the high rates of complications reported in open surgery286^,287.

After the encouraging results of these observational studies, other prospective and randomized studies compared PMLR with OCT in patients with refractory angina without therapeutic options. Recently, a meta-analysis evaluated the results of the seven largest randomized studies involving PMLR288. Of the 1213 patients evaluated, 651 were allocated to the PMLR group. All studies used Holmium YAG laser and three types of catheter (Eclipse System, Coaxial Cardiogenesis System, and Axcis PMLR) percutaneously, included patients with refractory angina (CCS classes III/IV), and evaluated the improvement in angina class, exercise tolerance, and quality of life during follow-up.

The authors noted that the patients subjected to PMLR showed lower angina class (OR = 2.13, 95% CI: 1.22-3.73), improvement in the quality of life by assessing the difference in the mean deviation (DMD); (DMD = 0.29, 95% CI: 0.05-0.52), perception of disease (DMD = 0.37, 95% CI: 0.14-0.61) and exercise tolerance (DMD = 0.29, 95% CI: 0.05-0.53) compared with patients on OCT. There was no impact on mortality after a 12-month follow-up period. It was concluded that the patients with refractory angina without therapeutic options might benefit from PMLR for decreased anginal symptoms and improved quality of life, without impacting mortality. However, the authors highlighted that not all studies were double blind and that adverse events related to the procedure would need to be assessed because of the heterogeneity of the trials.

Interestingly, in the two largest randomized and double-blind studies involving 439 patients and that used a percutaneous sham procedure in the group randomized for OCT (to avoid the placebo effect of the procedure), no statistically significant difference was observed in any of the clinical outcomes evaluated289^,290. This suggests that the benefits previously observed with PMLR and TLRS were more associated with a placebo effect of the procedure than with therapy itself (PMLR).

NICE also conducted an analysis of these seven studies involving PMLR283. Although they noted some improvement in secondary outcomes (exercise tolerance and anginal symptoms), this study highlighted that little could be concluded in view of the heterogeneity of the studies. Moreover, it emphasized that, although the perioperative mortality was low, the rates of nonfatal complications were high, including MI (7%), ventricular perforation (1%-4%), cardiac tamponade (≤3%), and stroke (2%). Therefore, PMLR may not be as effective in decreasing anginal symptoms and improving the functional capacity and is associated with unacceptable nonfatal complication rates.

Over the past 10 years, we have observed a great progress in the TLRS technique, surgical or percutaneous, with encouraging results from a few isolated clinical trials. However, more qualified, randomized, double-blind studies with sham procedures in the control group as well as recent evidence from meta-analyses and systematic reviews do not support the widespread use of this technique.

Class III, Level of Evidence A (meta-analysis and systematic review)

Class IIB, Level of Evidence B (one randomized study)

Class III, Level of Evidence A (meta-analysis and systematic review)

When the noninvasive evaluation of ischemia using imaging techniques is contraindicated, inconclusive, or unavailable, the evaluation of FFR during adenosine infusion is particularly beneficial for identifying hemodynamically or functionally significant stenoses that are inducers of ischemia, which can justify the indication for revascularization291^-293.

Previous studies with conventional stent demonstrated that medical treatment provides better results than immediate revascularization in patients with FFR > 0.80322^-324. Therefore, a patient with stenosis and FFR > 0.80 (with two measurements or during adenosine infusion) should not be revascularized. A recent study known as FFR versus Angiography for Multivessel Evaluation (FAME-2) confirmed that patients with stable CAD and stenoses with FFR ≤ 0.80 benefited from revascularization using PCI compared with OMT, and this benefit was solely the result of the decreased need for urgent revascularization. In contrast, patients without ischemia had excellent results when treated with OMT without revascularization325. FFR is already being used clinically to evaluate the efficacy of PCI.

In general, although FFR is not important in lesions angiographically > 90% (almost all these injuries have a FFR ≤ 0.80), FFR can help decide when revascularization is necessary in some unclear clinical conditions, including those of patients with multivessel diseases of heterogeneous presentation. In these patients, the measurement of FFR can change the revascularization strategy (PCI vs. CABG) beyond its scope, according to the functional assessment of stenoses in critical areas of the coronary arteries.

The use of IVUS has been widely investigated in stable CAD in distinct lesion subtypes. In contrast to FFR, IVUS is a diagnostic imaging tool and does not provide a functional assessment of stenosis severity. The cutoff limits of 3.5 or 4.0 mm2, previously accepted for stenosis in major epicardial arteries, and 6.0 mm2 for stenoses in LCAT326, proved to be unreliable and poorly correlated with FFR, and better results are achieved when absolute IVUS measurements are corrected for the reference vessel size. In contrast, considering that the indication for treatment is well established and more information is needed, IVUS is far superior to FFR because the former can perform an anatomical characterization of the lesion in terms of vessel size and plaque composition. It can also control the expansion and position of the stent arm.

Recently, optical coherence tomography (OCT) has been developed as a novel intracoronary image tool with higher resolution (10 mm) and can make a detailed evaluation of surface components, including thickness measurements of the fibrous lipid cap of the plaque327. The benefits of OCT in patients with stable CAD and with potential vulnerable plaques have not been well established, and certainly the treatment of severe nonfunctional lesions based solely on the presence of elements of instability is not recommended. On the other hand, its easy image acquisition allows the optimized expansion and placement of the stent and the assessment of its long-term endothelialization327.

The goal of revascularization in stable patients is to improve survival and relieve symptoms. The decision to revascularize a patient should be based on the presence of significant obstructive coronary artery stenosis, amount of associated ischemia, and the expected benefits for the prognosis and/or symptoms (Figure 3). Many clinical, anatomical, technical, and environmental factors should be considered and discussed before the benefit of revascularization is anticipated (Chart 3, Figure 3)86^,328^-331. The heterogeneity of these factors makes absolute recommendations unfeasible for the management of each clinical condition. Therefore, for a particular patient in a particular hospital, the clinical assessment should be consensual and not individual, preferably after discussion with the Heart Team of cardiologists (in difficult cases) and also considering the patient's preferences. However, this decision should be individualized, because to many patients, the preferred method is often quite clear. When technically feasible, with an acceptable level of risk and a good life expectancy, revascularization is indicated in patients with angina that is refractory to OMT and chronic stable CAD.

Although contemporary interventional treatments have decreased the risk of restenosis compared with previous techniques (up to 40%), previous meta-analyses did not demonstrate that the introduction of the conventional stent conferred advantage in terms of survival in comparison with balloon angioplasty291^,293^,322. Similarly, the use of covered stents did not have any advantage over conventional stents in terms of survival323.

With regard to two-vessel disease patients, classical studies that compared CABG with drug treatment indicated increased survival of patients with two-vessel diseases with involvement of the LAD, particularly those with both LV dysfunction and stenosis in the proximal LAD58^,298^,301^,332^-334. Previous studies that compared the results of CABG with those of PCI using first-generation stents in multivessel-disease patients demonstrated that both interventions yielded similar results regarding late survival and the rate of MI, and the main difference was associated with a larger number of subsequent revascularization procedures in patients subjected to PCI294^,304^,316^,335^-337.

To date, no study could demonstrate that PCI in patients with stable CAD improves the survival rates58^,301^,332^,333^,265^,294^,316^,335^-339. The three most recent studies comparing CABG with OMT are the largest and the most informative and involved contemporary and OMT. Similarly, these three studies indicated no survival advantage with PCI. The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial evaluated 2,287 patients and compared PCI and OMT with OMT alone in patients with stable CAD or ischemia and coronary lesions suitable for PCI294. The target population of the study comprised patients with chronic angina pectoris (CCS class I-III), stable post-MI, and asymptomatic patients with objective evidence of myocardial ischemia. All patients were diagnosed with CAD using angiography in at least one vessel with indication for PCI (class I or II) according to AHA/ACC. Of these, 30% had one-vessel disease and 39% had two-vessel disease. Patients with stenosis > 80% in one or more vessels encompassing a large myocardial area at risk could participate even in the absence of documented ischemia. The primary outcome of overall death or nonfatal MI did not differ between the two groups during a mean follow-up period of 4.6 years294. However, in patients treated invasively, free (angina) time was significantly better in a 3-year follow-up. However, a substudy indicated that patients with ischemia > 10% on stress myocardial perfusion scintigraphy exhibited a higher rate of death or AMI. Moreover, the group subjected to PCI and OMT experienced a significant decrease in ischemia (33% vs. 19%, p = 0.0004) and those with decreased ischemia had a lower unadjusted risk of death or MI, particularly if baseline ischemia was moderate to severe316.

The Bypass Angioplasty Revascularization Investigation in Type 2 Diabetes (BARI 2D) study, involving 2,368 patients, assessed whether PCI or CABG (the choice was left to the cardiologist), combined with OMT was better than OMT alone in patients with chronic stable CAD and diabetes mellitus type 2335. The target population was patients diagnosed with type 2 diabetes and CAD documented angiographically, whose revascularization was not required for immediate control of severe or unstable angina. Patients with stenosis > 70% showing symptoms of angina were eligible for randomization even without documented ischemia. In contrast, approximately 30% of the patients with a positive stress test were asymptomatic. With regard to the risk factors associated with atherosclerotic burden, 69% of the cases involved two-vessel disease and only 13.2% involved the stenosis of the proximal LAD. The primary outcome of overall mortality and the rate of MI or stroke did not differ between the two treatment strategies in a 5-year follow-up. Patients with more severe disease were selected for CABG instead of PCI and formed a high-risk group, which resulted in increased benefit of early revascularization (decreased MI compared with OMT)336.

In the FAME-2 study, 888 patients with stable CAD and functionally significant stenosis (FFR ≤ 0.80) were randomly assigned to PCI on the basis of FFR and OMT or OMT alone, and 34.9% of these patients had two-vessel diseases with at least one significant injury in the middle third or the proximal third of LAD324^,325. The target population of the study were patients who had at least one functionally significant stenosis and, on average, large areas of myocardial ischemia (mean FFR = 0.68), whereas low-risk patients with nonischemic FFR were not randomized but were monitored separately to maintain a record. The study was discontinued prematurely by the Health Safety Monitoring Board owing to a very significant decrease in the rate of hospital readmission and of urgent revascularization in the group with FFR ≤ 0.80 treated with PCI compared with those with FFR ≤ 0.80 treated with OMT. There was no difference in the rates of death or MI between the two strategies. In patients without ischemia (on record), the progress and outcomes were favorable for the group treated with OMT325.

In summary, a total of seven major randomized revascularization trials vs. medical therapy in stable chronic CAD were published in the last 10 years265^,294^,325^,335^,337^-339. In addition, for low-risk patients with stable CAD, after documentation of ischemia and careful clinical and angiographic screening, the initial OMT strategy is safe and should be the standard approach. Until an adequate duration of OMT is administered, cardiologists and surgeons should be more conservative when making decisions about revascularization, particularly in cases of high-risk comorbidities, unfavorable anatomy, or symptomatic patients without extensive induced ischemia. Previous studies have shown that, despite frequent revascularization procedures, most patients can be treated with OMT alone during these procedures.

However, when the initial OMT procedure fails and the patients remain symptomatic or when the risk of ischemia is relevant, various options need to be discussed, including strengthening of OMT and revascularization. The advantages, limitations, and opinions of the surgeon, clinician, and interventionist should be fully addressed in the discussion with the patient.

The well-known early revascularization complications include periprocedural MI, thrombosis, and late restenosis (nowadays, the latter is greatly decreased with the advent of the second-generation stents) after PCI, as well as perioperative MI, stroke, cognitive impairment, surgical-wound infection, and prolonged hospitalization and rehabilitation after myocardial revascularization. The potential advantages of an early revascularization strategy (PCI or CABG) include greater relief of symptoms and insignificant increase in mortality, decreased duration of drug therapy, decreased hospitalization, and lower chance of repeated revascularization in the first year, resulting in improved quality of life.

However, the advantage of CABG in relation to OMT with regard to the relief of symptoms is attenuated over time. OMT is safer in the short term and as is safe as CABG considering the mortality rate in low-risk patients in up to 5 years. However, OMT requires more and higher doses of medications, which can have a direct impact on treatment adherence, side effects, drug interactions, quality of life, and long-term costs to patients and the health care system.

The results of the studies that compared myocardial revascularization with OMT confirms that, except for better symptom relief and decreased frequency of urgent revascularization, there is no advantage of CABG in relation to isolated OMT to decrease mortality in patients with angiographically diagnosed stable CAD with a low-risk profile (e.g., two-vessel disease and preserved ventricular function), despite the possibility of OMT during follow-up.

Finally, although interventional and surgical techniques have improved over the past two decades, drug therapy also improved greatly over the same period. As a result, OMT can substantially improve the long-term results of patients treated conservatively, in addition to those subjected to revascularization, decreasing the impact of disease progression on survival and future events, even in revascularized patients. Therefore, for the treatment of chronic stable CAD, OMT is the foundation of treatment whereas myocardial revascularization is the coadjuvant and complementary treatment at some point in disease progression.

During the last two decades, approximately 20 clinical studies that compared PCI with CABG have indicated no difference in the overall survival rate between the two interventions, which is possibly associated with the low CAD risk in the populations evaluated304^,314^,340^-342. In contrast, several records matched by propensity scores have consistently demonstrated increased survival rates with CABG, compared with catheter intervention, followed by a decreased need for repeated intervention, although these results are still susceptible to confounding factors265^,343^,344.

The relative indications for PCI and CABG in patients with chronic stable CAD have been clearly defined by recent recommendations86^,328^-331. There has been a growing recognition of the importance of the Heart Team staff in defining the consensus about whether, when, and how revascularization should be performed. Figure 3 and Chart 3 show the algorithms recommended to simplify the decision-making process. These guidelines emphasize the importance of OMT for all patients and for both procedures.

In relation to patients with diabetes, even among those with two-vessel disease, the FREEDOM study recently demonstrated a significant reduction in the primary ischemic outcome in a 5-year follow-up for 1900 patients treated with CABG or PCI (16% of the cases involved two-vessel disease)267, corroborating the results of previous studies that suggested a significant mortality benefit of CABG compared with PCI in patients with diabetes with multivessel disease when both revascularization techniques are technically viable345, even at the cost of increased risk of nonfatal stroke. In this same study, all patients subjected to PCI received covered stents and glycoprotein inhibitors during the procedure, in addition to dual antiplatelet regimen for at least 1 year, and the control of risk factors was optimized in both groups.

Furthermore, the results of large records matched by propensity scores comparing the results of PCI with those of CABG, are consistent in favor of the use of CABG346^-348. In fact, in a recent study involving 7235 patients matched for numerous baseline characteristics, the overall survival rates in a 8-year follow-up were 78.0% for CABG and 71.2% for PCI (HR = 0.68; 95% CI = 0.64-0.74, p < 0.001]. For anatomical groups, RR was 0.53 (p < 0.001) in patients with three-vessel disease involving the proximal LAD, and 0.78 (p = 0.05) for patients with two-vessel disease and healthy LAD. A lower risk of death after CABG was observed in all subgroups stratified by different basal risk factors348.

More recently, Weintraub et al.349 reported the survival rates of 86,244 patients subjected to CABG (19.7% of the cases involved two-vessel disease) and 103,549 patients subjected to PCI (68.9% involving two-vessel disease) matched for propensity score with two-vessel CAD (53%) or three-vessel CAD. In a 4-year follow-up, increased mortality was observed in the PCI group compared with the CABG group. Despite statistical adjustment, this large record could not eliminate confounding factors, in addition to the fact that the patients in most severe conditions were probably selected to undergo PCI349.

Considering the above, the need for decision-making on an individual basis is important by assessing risk using scores that are available for both percutaneous and surgical intervention (Syntax score and EuroSCORE), by evaluating the benefits to each patient, and by assessing the cost-benefit ratio and patient preferences to prevent the indiscriminate and improper use of percutaneous interventions. To this end, in addition to common sense, responsibility, and ethics, adequacy criteria related to indication for revascularization are available and have recently been revised to serve as a guideline in the management of patients with chronic stable CAD. Of note, the effect of PCI on the quality of life of patients has not been addressed in the current guidelines, but it deserves attention in the context of chronic CAD and should be considered during the decision to intervene295^,350.

Some comments regarding thrombosis occurring in novel stents need to be made. The risk of thrombosis dramatically increases in patients who prematurely interrupt the double platelet inhibition therapy; the mortality rate associated with stent thrombosis varies between 20% and 45%351. Therefore, the ability of the patient to tolerate and respect the period of at least 30 days of treatment with dual platelet inhibition for conventional stents, and 12 months for covered stents, is an important factor when choosing PCI for treatment of patients with stable CAD.

In conclusion, PCI is an alternative to CABG in patients with two-vessel disease when there is appropriate indication for revascularization, as shown in Chart 4. The overall assessment of comorbidities, surgical risk, ischemia, and coronary anatomy is important when choosing the most appropriate method of revascularization for each patient.

Adequacy criteria are based on consensus among specialists regarding when the indication for a procedure is appropriate or questionable352^,353. Coronary revascularization is considered appropriate when the expected benefits in terms of survival and overall status (symptoms, functional status, and/or quality of life) exceed its expected negative consequences329. Similarly, indication for revascularization is considered inappropriate when the procedure probably will not improve patient outcome and/or increase patient survival299^,329.

However, this is an important and complex issue because the cost of imaging techniques and revascularization comes under increasing but appropriate scrutiny. In addition, the excessive and inappropriate number of indications for PCI in recent years, without considering the benefit to the patient, can cause complications to the patient. Therefore, stricter guidelines with appropriate indication criteria for intervention in the context of chronic stable CAD are required.

Recently, Chan et al.352 applied these criteria to patients subjected to PCI in the context of stable angina in the United States and found that 50% of the indications were considered appropriate, 38% were considered questionable, and 12% were inappropriate. Furthermore, Ko et al.353 observed patients with stable CAD in Ontario, Canada, and found that 68% of the indications for revascularized patients were appropriate, 18% were questionable, and 14% were inappropriate. Among the patients with questionable or inappropriate indications, 86% and 82% were subjected to PCI, respectively353.

Therefore, PCI is an alternative to CABG for two-vessel disease patients when there is appropriate indication for revascularization, as shown in Chart 4. An overall assessment of comorbidities, surgical risk, ischemia, and coronary anatomy is important when deciding on the most appropriate method of revascularization for each patient.

In conclusion, even in the era of contemporary stents, PCI can be used to decrease the incidence of angina but not the risk of MI in the long term and this risk may even increase in the short term. Furthermore, improved survival compared with optimized drug therapy has not been demonstrated in patients with stable angina. The indications are summarized below.

Grade of recommendation I, Level of evidence A.

To improve symptoms, CABG or PCI are beneficial to patients with one or more injured arteries (stenosis > 70%) that are amenable to treatment and to patients with refractory angina, despite optimized drug therapy.

When introduced in 1977, PCI was only considered to be suitable to treat patients with single-vessel disease and with noncomplex angiographic characteristics.354 Since then, there has been great progress in the development of novel devices and in the technical capacity to perform these procedures, with the consequent increase in the number of indications. At present, patients with complex artery diseases are treated routinely, particularly with the use of coronary stents.

With regard to patients with multivessel disease, several studies have compared the results of CABG with those of PCI355^-360. Many of these evaluations have been limited by the nonrandomized selection of patients, inclusion of less-complex cases, or insufficient statistical power.

The SYNTAX study is the best contemporary analysis of revascularization strategies in multivessel disease patients361. In this series, 1,800 patients with significant LMD disease or three-vessel disease were randomized for CABG (n = 897) or PCI (n = 903) with the use of the first-generation SF Taxus(r). The primary outcome was MACE comprising death from any cause, MI, stroke, and repeated revascularization. Moreover, this study was responsible for two important contributions to adequate patient revascularization: the angiographic Syntax score and the Heart Team concept. The Syntax score was based on the location, severity, and extent of stenosis, to determine the level of anatomical complexity of CAD. Subsequently, patients were categorized as low score if Syntax score ≤ 22, intermediate score if Syntax score was between 23 and 32, and high score if Syntax score was ≥ 33. With regard to the Heart Team, the revascularization strategy was decided collectively by the cardiac surgeon and the interventional cardiologist.

In the SYNTAX study, 1,096 patients had three-vessel disease. After a 5-year follow-up period, no significant difference in the MACE rate was observed among three-vessel disease patients with low Syntax score (0-22) subjected to PCI or MACE (33.3% vs. 26.8% respectively; p = 0.21)362. The patients with intermediate Syntax score (23-32) and randomized for PCI had a significantly higher MACE rate than those subjected to CABG (37.9% vs. 22.6% respectively; p = 0.0008). The performance of PCI became even more unfavorable for patients with high Syntax score (≥ 33), and an increase in MACE rate of 57% was observed in the patients subjected to this strategy (41.9% vs. 24.1%, respectively; p = 0.0005).

At present, CABG is the preferred strategy for three-vessel disease patients with more severe clinical and angiographic characteristics (increased age, low EF, renal dysfunction, peripheral vascular disease, diabetes mellitus, or Syntax score > 22). However, for lower-risk patients, PCI has a safety profile similar to that of CABG and can be considered the initial revascularization strategy for these patients294^,314^,315^,363. The choice of treatment on an individual basis is the most relevant factor when choosing the best therapeutic option for these patients. The adherence to the Heart Team concept, detailed clinical evaluation, evaluation of the experience of the operators and the surgical clinic, and respect for patient preferences after giving the necessary recommendations are essential for the achievement of favorable outcomes.

In the last decades, the presence of stenosis > 50% in LM has been indicative of CABG. Only high-surgical-risk patients with disease in LMD were eligible for PCI. Recently, however, PCI for treatment of significant stenoses in the unprotected LMD has become an alternative to surgical treatment, considering that several clinical trials have consistently demonstrated the feasibility, safety, and effectiveness of PCI with stenting in this context, particularly with DES. Together with the development of percutaneous treatment devices, technical approaches, and adjunct pharmacological treatment, PCI proved to be a less-invasive procedure and has been associated with lower rates of intraprocedural complications and decreased hospitalization364.

Obstructive disease of LMD can show different degrees of complexity, depending primarily on location, and may involve LMD bifurcation and the origin of the LAD and Cx arteries. Therefore, stenoses of LMD can be classified on the basis of the stenosis of the arterial ostium, body, or bifurcation. Several studies have demonstrated better clinical prognosis of PCI when the LMD bifurcation is unimpaired, and this factor is important when choosing the best strategy and revascularization technique. In general, compared with CABG, PCI in the unprotected LMD has shown high success rates and a similar safety profile in the long-term follow-up but a higher incidence of target lesion revascularization (TLR) at late follow-up (2%-38%), with most of the scientific evidence being derived from retrospective, single-center studies, multicenter studies, nonrandomized comparisons, and prespecified subanalyses of randomized trials. Even so, the percutaneous treatment of the unprotected LMD is already considered in situations of decreased anatomical complexity without significant stenosis of the bifurcation as an alternative approach to surgery364.

However, before referring a patient for myocardial revascularization, either percutaneous or surgical, it is important to confirm the presence of significant atherosclerotic obstruction, considering that spasm or artifacts generated by the positioning of the guiding catheter or angiographic projection are common and may lead to incorrect conclusions about the degree of LMD stenosis. Therefore, in addition to a detailed clinical evaluation, intracoronary ultrasound (IVUS) is recommended to confirm the presence of significant LMD stenosis. Such assessment is obviously more important in cases of moderate stenosis on coronary angiography and should include the measurement of the minimum diameters and areas as well as the load and distribution of the atherosclerotic plaque. A minimal luminal area (MLA) of > 6 mm2 indicates safety for nonrevascularization of LMD365. In contrast, MLA of 4.8 mm2 was correlated with FFR <0.80 and AML of 4.1 mm2 was correlated with FFR < 0.75, indicating functional impairment366. A subanalysis of the MAIN-COMPARE study evaluated the use of IVUS to guide PCI in the unprotected LMD. Overall, procedures in 756 patients were guided by IVUS whereas those in 219 patients were guided only by angiography. Using the propensity score, there was a lower (statistically nonsignificant, p = 0.06) mortality rate after a 3-year follow-up associated with the use of IVUS. However, when considering the group treated with DES, the mortality rate was significantly lower in those guided by IVUS (4.7% vs. 16.0%; p = 0.048)367.

The impact of PCI and DES in the unprotected LMD lesions has been shown in previous studies. The Left Main Coronary Artery Stenting (LE MANS) registry368 included 252 patients with LMD lesions treated with nonpharmacological stents (NDES) and DES (36.2%). At 30 days, the rate of MACE + stroke and the mortality rate were 4.8% and 1.5%, respectively. After a mean follow-up period of 3.8 years, the rate of MACE + stroke and the mortality rate were 25.4% and 13.9%, respectively. After a 5-year and 10-year follow-up period, the survival rates were 78.1% and 68.9%, respectively. Importantly, the rates of MACE were significantly lower with DES than with NDES (p = 0.04). In the multicenter observational study known as Revascularization for Unprotected Left Main Coronary Stenosis: Comparison of Percutaneous Coronary Angioplasty Versus Surgical Revascularization (MAIN-COMPARE)357, 2,240 patients with significant unprotected LMD lesions were treated with PCI or CABG (NDES = 318, DES = 784, CABG = 1,138). Patients were subjected to PCI when the anatomy was favorable to stenting and in cases of refusal or high risk during CABG. In patients subjected to PCI, the rate of complications during hospitalization and the mortality rate were 2.7% and 0.8%, respectively. At late follow-up (approximately 3 years), no difference was observed between PCI and CABG in 542 pairs regarding the survival rate (92.2% vs. 92.1%; p = 0.45) and mortality-free survival rate + AMI + stroke (90.8% vs. 90.7%; p = 0.61), respectively. However, patients subjected to PCI had increased target vessel revascularization (TVR) (p < 0.001). In the very late follow-up stage (mean of 5.2 years), after adjustment with propensity score, there was no significant difference in the mortality rate (p = 0.35) and death + AMI + stroke (p = 0.59) for PCI compared with CABG. However, TVR was higher in the groups subjected to PCI treated with both NDES and DES. Furthermore, a meta-analysis performed by Pandya et al.369 involving 44 studies and 10,342 patients treated with DES or NPS revealed that the mortality rate, AMI, and new TVR were 8.8%, 4.0%, and 8.0% among those treated with DES and 12.7%, 3.4%, and 16.4% among those treated with NDES, respectively. Considering only nine comparative studies (n = 5,081), the DES group had lower rates of adverse events, mortality rate (p = 0.01), MI (p = 0.03), and TVR (p < 0.001) than the NDES group after a 3-year follow-up period369. Therefore, when PCI is indicated for LMD, the latter should be performed preferably with DES.

The prospective, randomized, and multicenter SYNTAX study370 compared PCI and DES with CABG in complex multivessel disease patients with or without involvement of LMD, and a prespecified subanalysis involving 705 patients evaluated the impact of the two strategies for LMD treatment. The clinical results of a 5-year follow-up indicated similar values for mortality rate (12.8% vs. 14.6%; p = 0.53), cardiac death (8.6% vs. 7.2%; p = 0.46), MI (8.2% vs. 4.8%; p = 0.10), and the composite outcome of death + stroke + AMI (19.0% vs. 20.8%; p = 0.57) for the PCI group compared with the CABG group. In contrast, the rates of stroke were significantly lower in the PCI arm (1.5% vs. 4.3%; p = 0.03). However, a lower rate of new TVR was observed in the surgical arm (26.7% vs. 15.5%; p < 0.01). When the results were stratified by anatomical complexity, comparable or better results for PCI were observed in the subgroups with low (< 23) and intermediate Syntax score (23-32). In this context, the event rates for those treated with PCI compared with those treated with CABG in the low-score subgroup were as follows: total death (7% vs. 11.3%; p = 0.28), stroke (1.8% vs. 4.1%; p = 0.28), MI (6.2% vs. 3.1%; p = 0.32), total death + stroke + AMI (13.9% vs. 15.2%; p = 0.71), and TVR (23.0% vs. 20.3%; p = 0.65), and the results in the intermediate-score subgroup were as follows: overall death (8.9% vs. 19.3%; p = 0.04), stroke (1.0% vs. 3.6% p = 0.23), MI (6.0% vs. 4.6%; p = 0.71), total death + stroke + AMI (15.7% vs. 24.9%; p = 0.11), and TVR (22.2% vs. 16.6%; p = 0.40). In the high-score subgroup (> 32), the event rates were similar for overall death (20.9% vs. 14.1%; p = 0.11), stroke (1.6% vs. 4.9%; p = 0.13), MI (11.7% vs. 6.1%; p = 0.40), and total death + stroke + AMI (26.1% vs. 22.1%; p = 0.33). However, the rate of TVR was higher in the PCI group than with the CABG group (34.1% vs. 11.6%; p < 0.001). Overall, the SYNTAX study demonstrated that multivessel CAD patients subjected to PCI for treatment of lesions in the unprotected LMD is a worse clinical prognostic factor, particularly when the lesion is accompanied by lesions in two or three other vessels. Similarly, the associated presence of lesions in the right coronary artery (particularly in cases of total occlusion) was found to be a predictor of mortality in the late follow-up of patients subjected to PCI of LMD371^,372.

In addition, the PRE-COMBAT prospective study randomized 600 patients with unprotected LMD lesions subjected to PCI or CABG. After 1 year, the composite outcome of death, AMI, stroke, and ischemia-associated TVR was observed in 8.7% of those from the PCI group compared with 6.7% of those from the CABG group (p < 0.001 for noninferiority). However, this incidence was below the value considered in the calculation of sample size. Consequently, the study lacked statistical power to adequately respond to the tested hypothesis373.

The angiographic Syntax score proved to be a good discriminator for PCI but not for CABG because it did not include clinical factors considered important for the prognosis of CABG patients already covered in surgical risk scores, including EuroSCORE, Parsonnet, and the score of the Society of Thoracic Surgeons (STS). Accordingly, PCI should not be used in isolation when making clinical decisions. For this reason, the Syntax score II was recently introduced, considering eight variables strongly associated with mortality after a 4-year follow-up in the SYNTAX study. In addition to the original angiographic Syntax score, the other seven variables were age, creatinine levels, LV function, disease of the unprotected LMD, peripheral vascular disease, being of the female gender, and chronic obstructive pulmonary disease. The validation study by Farooq et al.374 demonstrated the greater predictive power of the Syntax II score than the purely angiographic Syntax score; in theory, the former score could facilitate clinical decision-making for patients with multivessel disease and unprotected LMD.

According to current evidence, the following factors should be considered when choosing the treatment strategy for unprotected LMD:

Therefore, the development of both DES and management techniques have supported PCI as a viable alternative to CABG for the treatment of unprotected LMD lesions. Randomized clinical trials involving patients eligible for both PCI and CABG helped elucidate the feasibility, safety, and efficacy of PCI in this setting. At present, it is reasonable to consider PCI for treatment of unprotected LMD lesions in clinically stable patients with the aim of improving survival by following these recommendations:

Gene therapy can be defined as a medical intervention in which the transfer of genetic material is used for the modification of somatic cells in vivo, allowing the in situ expression of the transferred gene with a consequent therapeutic effect375. The transfer of therapeutic genes requires the use of a vehicle known as a vector, which carries the gene of interest and guides it to the target cell, thereby facilitating the transfer of genetic material in somatic cells in vivo 376. Basically, there are two groups of gene transfer vectors: viral and nonviral. Among the viral vectors, the most commonly used for the optimization of gene transfer are modified retroviruses and adenoviruses377.

Until June 2013, 27 gene therapy protocols for CAD (24 protocols for chronic CAD and three for arterial restenosis) had been submitted to the Office of Biotechnology Activities at NIH in the United States, which is responsible for regulating all protocols involving gene manipulation in the country378.

The growth of knowledge regarding vascular growth and angiogenic cytokines, and the parallel development of more efficient vectors, allowed for the testing of the hypothesis that gene transfer of growth factors could mitigate the damage from myocardial ischemia by stimulating vascular growth, and this strategy became known as therapeutic angiogenesis379.

Around the late 1990s, Losordo et al.380, Symes et al.381, and Rosengart et al.382, among others, reported the initial results of gene transfer of vascular endothelial growth factor 165 (VEGF₁₆₅) via direct intramyocardial injection in patients with refractory angina. During follow-up, researchers documented a significant decrease in anginal episodes, increased number of normally perfused myocardial segments, and increased Rentrop score (number of collateral vessels) in all patients. No procedure-related adverse effects were observed.

More recently, the first multicenter trial known as Angiogenic Gene Therapy (AGENT)383 selected 79 patients with symptomatic CAD to receive one of five increasing doses of a viral vector encoding FGF4, or placebo. Although the results of therapeutic efficacy of ETT were not significantly different between the groups, the analysis of the subgroup with greater initial functional limitation indicated that gene therapy proved effective in increasing exercise tolerance. Subsequent studies such as AGENT-3 and AGENT-4384 involving > 500 patients in several countries did not replicate the initial results related to increased exercise tolerance after administration of FGF4 in patients with stable angina, and therefore these studies were discontinued. Similar neutral results were obtained in the Vascular Endothelial Growth Factor in Ischemia for Vascular Angiogenesis (VIVA) trial385. However, in an evaluation performed 120 days after treatment, the group treated with the highest dose of VEGF showed a significant decrease in angina (functional class improvement) and a favorable tendency of increased performance effort and of decreased anginal episode frequency.

Grade of recommendation: Till date, no recommendation is available for the clinical use of gene therapy, except within the context of experimental clinical research in accordance with current standards.

The transplantation of stem cells and/or progenitor cells may be a therapeutic option for inducing vascular growth (angiogenesis)386 and/or limiting the postischemic myocyte loss387 and can therefore decrease or even prevent the onset of heart failure secondary to chronic ischemic heart disease.

In the last decade, the first reports on cell therapy in patients with CAD began to be published. Assmus et al.388 transplanted, via intracoronary infusion, progenitor cells from bone marrow or peripheral blood in patients who had acute stroke after reperfusion. During the 4-month follow-up, these patients showed increased LVEF, improved regional wall motion in the infarct zone, decreased end-systolic volume, and increased coronary flow reserve in the artery supplying the infarcted region. No adverse events were observed.

The use of cells derived from adult bone marrow for treatment of severe ischemic heart disease associated with heart failure has been proposed in the study conducted by Perin et al.389 involving 14 patients. Patients were subjected to transendocardial injection guided by electromechanical mapping in viable but ischemic areas (hibernating myocardium). In a 4-month follow-up period, the authors observed functional class improvement, significant reduction of the perfusion defects according to SPECT, and an increase in EF from 20% to 29%.

Stamm et al.390 proposed the combined use of intramyocardial injections of bone marrow-derived stem cells with the potential to induce angiogenesis using CABG in six post-AMI patients. All patients survived after 3-9 months of follow-up. In addition, increased overall wall motion was observed in four of six patients and increased perfusion of the infarcted area was observed in five of six patients. Gowdak et al.391 adopted a similar strategy for the treatment of patients with severe and diffuse CAD who were refractory to clinical treatment and ineligible for complete surgical revascularization because of the disease extent. In 21 patients, stem cells and autologous hematopoietic progenitor cells were injected during revascularization surgery in myocardial areas previously identified as viable and ischemic. No adverse events related to the procedure were observed392. The analysis of myocardial perfusion in the injected and nonrevascularized segments indicated the reversal of ischemia in these segments and a contractile improvement. A large, randomized, double-blind, and placebo-controlled clinical trial is underway to test the role of cell therapy complementary to incomplete CABG in patients with stable angina393.

The RENEW study, currently underway, will test the efficacy and safety of the intramyocardial injection of autologous CD34⁺ cells in patients with angina refractory to OMT and in patients who are ineligible for revascularization procedures394. Moreover, the IMPACT-CABG395 study was recently initiated and will test the safety and efficacy of intramyocardial injection of autologous CD133⁺ cells in patients subjected to CABG procedure.

Human adult adipose tissue has high plasticity and contains two cell populations with distinct functionalities, which can contribute to neovascularization in ischemic tissues: endothelial cells and mesenchymal cells derived from adipose tissue396. Several clinical studies have been initiated to test the angiogenic potential of mesenchymal cells derived from adipose tissue in patients with chronic ischemic heart disease397, AMI, and cardiac failure398. The conclusion of these important clinical trials will help investigate the possibility of using this abundant cell source for the treatment of patients with heart disease.

Furthermore, one of the most recently investigated cell types for the treatment of patients with ischemic cardiomyopathy was obtained via the identification of resident cardiac stem cells with the potential for myocardial regeneration399. Numerous preclinical studies have demonstrated the efficacy of these cells for the treatment of post-AMI LV dysfunction400^,401. In the SCIPIO study402, resident cardiac stem cells were obtained from the right atrial appendage during the CABG procedure. After isolation, these cells were expanded and injected via intracoronary infusion approximately 4 months after surgery. The assessment of cardiac function using MRI showed a significant increase in LVEF in the treated group from 27.5% (baseline) to 35.1% and 41.2%, 4 and 12 months after cell infusion, respectively, as well as significant reduction in the infarcted area.

Grade of recommendation: Till date, there is no recommendation for the clinical use of cell therapy, except within the context of experimental clinical research in accordance with current standards.

In candidates for revascularization in the presence of ECVAD (extracranial carotid and vertebral artery disease), catheter-based angiography may be useful when noninvasive imaging studies are not sufficient for assessment. Level of Evidence C.

Stroke is a major cause of morbidity and mortality worldwide. Large differences between countries in incidence, prevalence and mortality have been attributed to changes in risk factors, uncontrolled hypertension, and other risk factors, resulting in more frequent and/or severe strokes in some countries. Stroke is the leading cause of morbidity and disability in the developed world. It is the second leading cause of dementia, the most common cause of epilepsy in the elderly, and a common cause of depression.

It is recommended that all patients with stroke be treated in a unit specialized in treating this type of patient. Class I, Level of Evidence A