The acute phase begins after an incubation period usually of 1-4 week after exposure to T. cruzi. The lesions known as “chagomas”, including the typical, but non-specific, Romaña’s sign, are a result of mucosal or cutaneous edema at the site of inoculation. Most patients present asymptomatic or show systemic infection symptoms (fever, hepatosplenomegaly, diaphoresis, myalgia), accompanied by equally non-specific laboratory alterations, especially leukocytosis, with absolute lymphocytosis. A minority of patients present a clinical picture of myocarditis, with signs and symptoms similar to myocarditis of other causes: dyspnea, fatigue and other commemorative symptoms of heart failure. In these cases, the ECG may show sinus tachycardia, ventricular ectopic beats, low voltage of the QRS complexes, branch block, diffuse disturbances of ventricular repolarization, first-degree or more advanced AV block. Chest x-ray may show increased cardiothoracic ratio in more severe cases, which may be associated with increased heart chambers and/or pericardial effusion. The echocardiogram often shows pericardial effusion, segmental changes in parietal mobility and insufficiency of mitral and tricuspid valves and, less frequently, cavitary dilation and decreased global systolic performance. These abnormalities usually resolve in the majority of patients over the first year of follow-up.12^,13

Serological tests are usually negative in the first weeks of infection. The diagnosis is made by detection of circulating parasites or their genetic material (PCR) through a variety of methods, such as blood culture, direct visualization of the parasite in peripheral blood, xenodiagnosis, or the presence of parasite nests in amastigote form revealed by biopsy with histopathology of affected organs, or of cutaneous “chagomas”.

Endomyocardial biopsy is rarely used for the diagnosis, but may be necessary, especially in cases of suspected reactivation of Chagas disease after heart transplantation, in which a clear distinction from implant rejection is critical in patient management.14

The clinical course of the acute phase in Chagas disease is often benign and the signs and symptoms typically resolve spontaneously over 1 to 3 months. It is estimated that fatal evolution occurs in < 5% of patients in the acute phase, when contaminated through classic vectorial transmission (via the bite of a triatomine bug), predominantly in patients with refractory heart failure. However, in acute cases resulting from contamination by the oral route (for example after ingestion of T cruzi contaminated sugar cane juice or açai), the acute disease is usually more severe and the rates of mortality are higher. This is probably due to inoculation of high parasite load and ease of penetration through the gastrointestinal mucosa, which is highly permeable to the parasite, in these cases.

The treatment of clinical manifestations of myocarditis and heart failure is similar to the one recommended for cases of myocarditis of other etiologies, including intensive measures of circulatory support in more severe cases.

However, more specifically, although there is no conclusive evidence concerning possible clinically relevant benefits that could be achieved, antiparasitic treatment with benznidazole or nifurtimox is indicated for all cases of acute Chagas disease, independently of the infection route or the reactivation mechanism, since it may decrease the severity of the symptoms, reduce disease time and the duration of detectable parasitemia. The occurrence of parasitological cure, besides clinical cure, is estimated in 60 to 85% of cases.15^,16

The indeterminate form of the disease is classically defined as the clinical situation of an individual with parasitological and/or serological evidence of chronic T. cruzi infection, but without symptoms or physical signs of the disease, with normal ECG and chest X-ray and without digestive tract (esophagus and colon) impairment seen in radiological exams.

However, more accurate complementary exams (i.e. echocardiography, nuclear angiocardiography, hemodynamic study and autonomic cardiac assessment) may demonstrate - usually subtle and of no prognostic relevance - cardiac alterations in this group of patients classified as indeterminate by classical criteria.17^-19 In spite of these minor abnormalities verified in many patients, those classified as indeterminate by the classical criteria, while maintaining their normal ECG status, present excellent prognosis and mortality rates comparable to those of the control group of the same age without T. cruzi.4^,20^,21

There are no formal guidelines in relation to the conduction of exams for early detection of left ventricular dysfunction in patients with the indeterminate form of Chagas disease. On the other hand, there are no identifiable factors in this phase that can distinguish the individuals who will develop the clinical cardiopathy from those who will remain asymptomatic during their whole lives, just keeping the serological positivity.

It is suggested that the ECG be repeated every 1 or 2 years and a simple chest X-ray every 3 or 5 years. Though it is more controversial, one may also suggest that the transthoracic echocardiography can be performed initially, and later on at regular intervals as well, every 3 to 5 years.8^,17^,22^,23

The absence of symptoms is most marked in individuals who are in incipient stages of the chronic disease, when (discrete) myocardial injury can be detected only due to alterations in complementary exams, such as ECG conduction disturbance, changes in LV segmental parietal mobility on ECG or Holter arrhythmias. In these individuals, however, sudden death may occur due to arrhythmic events, as evidenced by studies demonstrating a worsen prognosis in individuals with alterations in the ECG, even when asymptomatic.24 In a 10-year cohort study of 885 seropositive individuals, it was shown that T. cruzi infected individuals with normal ECG had a survival of 97.4%, comparable to the survival of seronegative individuals. On the other hand, survival of those with abnormal ECG was 61.3%, with a nine-fold increased risk in this group.25 It is estimated that 2 to 5% of patients without any apparent cardiopathy will develop new ECG alterations and evidence of cardiopathy each year.19^,26^,27

The clinical picture of CCC with ventricular dysfunction is described in a quite uniform manner in several literature reports, following the pioneering remarks of Chagas and Villela.28 In the early stages of manifestation, the most frequent symptoms are fatigue and dyspnea on exertion, but the registry of more intense symptoms of pulmonary congestion, such as paroxysmal nocturnal dyspnea and decubitus with orthopnea, are uncommon. In the disease evolution, there occur the symptoms of systemic venous congestion (jugular swelling, hepatomegaly, lower limb edema and ascites) and the evolution can still progress to anasarca, adynamia, or cardiac cachexia, similar to what happens in other cardiopathies with advanced ventricular dysfunction. At clinical examination, there are also signs of cardiomegaly resulting from deviation of the ictus cordis, there may be a muffled S1 heart sound in the mitral area, fixed doubling of second heart sound due to RBBB, third heart sound and atrioventricular valve regurgitation murmur, which may occur secondarily to the dilation of ventricular chambers. Signs of low systemic output may occur in advanced cases, such as filiform pulse, slow and oliguria peripheral perfusion. These signs are common to other clinical syndromes of heart failure. In contrast to these similarities, in heart failure of Chagasic etiology, pulmonary congestion is commonly mild in the advanced stages of the disease, compared to the more exuberant systemic congestion, and the pulmonary semiology may be more affected by the signs of pleural effusion than by crepitations, as well as by lower systemic pressure levels in this group of patients. The progression to acute pulmonary edema in these cases is even more rare.20 These particularities in the clinical presentation may relate to the most frequent concomitance of biventricular dysfunction, with right ventricular heart failure, sometimes earlier and more pronounced than the left one in T. cruzi infected patients.

Pulmonary and systemic embolisms are common manifestations in patients with CCC, as a result of murine thrombosis from cardiac chambers and systemic vein thrombosis, and are a major cause of embolic stroke and other morbidities. Thromboembolic accidents are often the first manifestation of the disease and may occur in stages without ventricular dysfunction (Stage B2). However, as in several other cardiopathies, cavity dilation and HF syndrome are known associated risk factors. Nevertheless, it is chronic regional ventricular dyskinesia, mainly apical, such as the classic aneurysm of the glove finger, which has shown a special propensity for the formation of mural thrombi and the consequent embolic events, particularly the systemic ones. As predicted, atrial fibrillation, even when present in the minority of this population, as a relatively late and secondary manifestation to the ventricular dysfunction, also constitutes an additional thrombogenic factor. Pulmonary embolization, which can originate from peripheral venous thrombi and right cardiac cavities, is much less frequently clinically diagnosed, but its incidence is certainly underestimated, compared to its prevalence in necropsy material.29

There is a clear lack of data to provide an estimate of the actual incidence of clinical thromboembolism in CCC, but series of autopsies and clinical studies indicate high rates of intracardiac thrombi and thromboembolic events in this population. In a revision of 1345 necropsies of patients with chronic chagasic cardiomiopathy, thromboemboli and/or intracardiac thrombi were observed in 44% of cases.29 Thrombi were equally frequent in right and left cardiac cavities. Systemic circulation thromboembolism was more frequent, but more associated with fatal events.

A transthoracic and transesophageal echocardiography study showed that thromboembolism frequently originated from the heart in 75 T. cruzi chronically infected patients without symptoms of heart failure, or with mild symptoms. Left ventricular mural thrombi were found in 23% of patients and were associated with previous history of stroke. Apical aneurysm was identified in 47% of patients and was significantly related to mural thrombosis and the occurrence of stroke. Left and right atrial appendage thrombosis was present in 4 and 1 patient, respectively. During the 24-month follow-up period, 1 non-fatal stroke event and 13 deaths were observed, 7 of which were sudden, 5 due to HF progression and 1 death by stroke.30^,31

Systematic revision of 8 observational studies, involving 4158 patients, addressed the association between CCC and the risk of stroke.32 The results indicate that chronically T. cruzi infected patients, when compared to the non-infected, had an excess risk of stroke of about 70% (RR = 1.70; HF 95%: 1.06 to 2.71). When the analysis was limited to 3 studies, with more strict criteria for stroke, an even higher excess risk was found (RR = 6.02; HF 95%: 1.86 to 19.49).

The characteristics of stroke patients with CCC were explored in a study of 94 patients with acute ischemic stroke, compared with the characteristics of a control group of patients without CCC. T. cruzi infected individuals showed higher rates of cardioembolic stroke (56% versus 9%), left ventricular dilatation (23% versus 5%), LV mural thrombosis (12% versus 2%), apical aneurysm (37% versus 1%) and atrial fibrillation (14% versus 5%).23

The I Latin American Guideline for the Diagnosis and Treatment of Chagas Cardiopathy adopted recommendations for estimation and prevention of cardioembolic stroke risk through the use of oral antithrombotic agents,1 based on a prospective cohorte study of 1,043 patient with CCC. The total incidence reported in this event was 3.0%, or 0.56%/year. In the final risk model for cardioembolic stroke prediction, a score was calculated in which the presence of LV systolic dysfunction added two points, and apical aneurysm, alteration of the ventricular repolarization at the ECG and age > 48 years added one point for each alteration. Considering the risk-benefit ratio, warfarin would be indicated for patients with 4-5 points (in this subgroup, there is incidence of 4.4% of stroke versus 2.0% of major bleeding per year). In the 3 point score group, stroke and major bleeding rates with OAC are equivalent, and acetylsalicylic acid (ASA) or warfarin can be indicated.32 In 2 point patients with low incidence of stroke (1.2% per year), ASA was recommended, or no prophylaxis. Patients with 0-1 point, with incidence close to zero, would not require prophylaxis.33

CCC is essentially an arrhythmogenic cardiomyopathy, with pathophysiological peculiarities in this context, which makes it uniquely distinct from other cardiopathies. Virtually, all types of atrial and ventricular arrhythmia may occur, including sinus node dysfunction, intermittent or complete AV block and complex ventricular arrhythmias. The arrhythmias may course asymptomatic or present with non-specific malaise or sudden, fleeting and spontaneously resolved onset palpitation, at rest or by exertion. Symptoms of low cardiac output due to Stokes-Adams syndrome are less common, but more ominous, including presyncope, lipotimia, or even syncope, which can occasionally be preceded by palpitations. These episodes can either correspond to (sustained or nonsustained) ventricular tachycardia, with or without hemodynamic instability, or to bradyarrhythmias due to atrioventricular block.34

In some cases, the standard 12-lead ECG with rhythm strip shows premature and ectopic ventricular depolarization, and even ventricular tachycardia outbreaks, in addition to atrial fibrillation or complete atrioventricularblock (AVB). Tachycardic ventricular arrhythmias and AV conduction disorders leading to periods of bradycardia can often alternate, frequently coexisting during the same Holter recording.

At clinical examination, it is possible to detect fixed doubling of the second heart sound (at a pulmonary focus), irregular heart rhythm, or even bradycardia, often associated with typical signs of a-waves, periodically incremented in the jugular venous pulse and reinforcement of the first heart sound, in cannon waves, when there is a temporal correlation between atrial and ventricular systoles, which is suggestive of complete atrioventricular block.

The presence and density of arrhythmia correlate with the degree of ventricular dysfunction in many cases, but can also occur in patients with preserved global left ventricular function, constituting the "isolated arrhythmogenic form” of the disease. This characteristic, which distinguishes CCC from coronary artery disease in patients with ventricular dysfunction, as well as from other cardiomyopathies, and makes T. cruzi infected patients especially susceptible to early sudden death, derives from its pathophysiological peculiarities and peculiar pathogenesis.

In fact, the mechanism of severe ventricular arrhythmia in CCC is mainly associated with the presence of regional fibrosis (especially in the posterolateral regions of the LV) and macroreentry circuits formation.35 Recent studies using cardiac magnetic ressonance have reinforced the idea that the presence of regional fibrosis is a major factor for the arrhythmic mechanism in this disease.36^-38

Another relevant pathophysiological factor that potentially triggers severe ventricular arrhythmia and sudden death in CCC patients is regional myocardial extensive and early sympathetic denervation. In a study with patients with CCC and normal or slightly reduced LV function, the presence of sustained ventricular tachycardia has been associated with more extensive areas of viable denervated myocardium, detected through I-MIBG myocardial scintigraphy^123.39

It is estimated that sudden death is the leading cause of mortality throughout the various phases of CCC, corresponding to 55 - 65% of deaths.22 Sudden death is often triggered by physical effort and may be caused both by severe tachyarrhythmias, such as ventricular tachycardia and fibrillation (probably in 80-90% of cases), and (less frequently) by asystole or complete AV block.40 The detection of nonsustained and especially of sustained ventricular tachycardias increases the chance of sudden death, but it occurs mainly in patients with advanced ventricular dysfunction.

Complaints of precordialgia in patients with CCC are quite common. This pain has characteristics that are often atypical for myocardial ischemia, described as stabbing, fleeting, or, conversely, long lasting (hours or even days), poorly located, usually not related to efforts, sometimes caused by emotional stress and with recurrent patterns throughout the day. However, sometimes, the episodes may be more acute, with typically ischemic characteristics, making the diagnosis even more difficult. Evidence of microcirculatory disorders as causes of anginal manifestations in this group accumulates in the literature.11^,41^,42

Given the low parasitaemia in the chronic phase of the disease, the serological tests must be able to detect antibodies against T. cruzi antigens. The most commonly used tests are: immunoenzymatic assay (ELISA), indirect immunofluorescence (IFI), and indirect hemagglutination (HAI). When all 3 are performed, agreement (90-98%) was observed among them. Since ELISA and IFI have similar characteristics in terms of their accuracy, with higher sensivity, but slightly reduced specificity than the HAI assay, a positive result in two of these three tests is recommended for the diagnosis. However, when the first test is negative, in the current practice based on the high sensitivity of all serological tests, there is no need for a second test.43

The main complementary diagnostic tests used in Chagas' heart disease are briefly described below, with emphasis on those focusing on the characterization and gradation of ventricular dysfunction.

Electrocardiography and Holter Monitoring: The most prevalent electrocardiographic alterations in patients with CCC are right bundle branch disorders and the left anterior hemiblock, and may reach 50% in patients of this group.44 These changes in the cardiac conduction system may be evolutionary, such as AV conduction delays. Sinus node dysfunction can also cause bradycardia. However, atrial arrhythmias tend to occur in the evolution of the heart disease with advanced ventricular dysfunction. It is crucial to observe that, although ventricular ectopic beats can be seen in normal individuals during the recording of a standard ECG, when it is verified in patients with CCC, the meaning of this alteration is completely different and it usually indicates that the ventricular arrhythmia is an integral part of the syndrome and it constitutes an element of strong prognostic value. Symptoms suggestive of arrhythmic syndromes also make ECG Holter monitoring mandatory, since it allows the assessment of episodes of both tachyarrhythmias and bradycardia for risk stratification in these patients.34 Recording for at least 24 hours allows determining the density of ventricular ectopy, detecting episodes of nonsustainable or sustained ventricular tachycardia, as well as determining the duration of sinus pauses and of asystolia of different origins. It is worth recalling that, to compose the Rassi score, used for predicting mortality in patients with CCC, the 24-hour Holter test is essential to assess the prognostic criteria, with independent value, of nonsustained ventricular tachycardia, as it will be seen later.

Other frequent changes in ECG at rest are: diffuse T wave and ST segment abnormalities, pathological Q waves, prolongation of the QT interval and increased QT dispersion. The evaluation of fibrosis using the QRS score applied to the standard ECG correlates with the NYHA functional class and with the extent of myocardial fibrosis detected in Late Gadolinium Enhanced (LGE) cardiac magnetic resonance imaging.45

Chest X-ray: Advanced stages of the disease are marked by (often massive) cardiomegaly, which may include signs of not only increased LV, but also of increased RV and both atrial dilation; however, pulmonary congestion differs from other cardiopathies because it is often discrete, in contrast to the degree of cardiothoracic ratio. Cardiomegaly is also an important prognostic factor in these patients, stratified by the Rassi score, as discussed next.

Echocardiography: Echocardiography is a non-invasive imaging method that allows the geometrical and functional diagnosis of both ventricles, which is essentially important in CCC. Alterations in segmental mobility in the inferior and inferolateral regions of the LV are quite common in patients with the cardiac chronic form.46^-50 Even though the detection of LV apical aneurysm (a very common alteration in patients with CCC)51 may be subject to operational limitations, it is one of the typical changes found in the disease, and may be filled with thrombus (Figure 3, A). Although early changes in LV regional mobility can be detected on ECG, both through conventional techniques17^,52^,53 and analysis of myocardial deformation, in some patients classified as having the indeterminate form, or even with CCC and function preserved by other methods,18^,53^,54 the prognostic value of these alterations is not well established yet.50^,55 Two-dimensional echocardiographic evaluation of the right ventricle can be performed through an acquisition protocol with images dedicated to this investigation.56 New methods for the assessment of right ventricular systolic function, such as the analysis of myocardial deformation (Figura 3, B), have already shown to correlate quite often to the quantification of RV ejection fraction by other methods, such as magnetic resonance imaging, in groups of patients with Chagas’ disease. Although three-dimensional echocardiography presents the benefit of volumetric quantification of cavities and, as a result, of ventricular ejection fractions, its role in patients with CCC has not been adequately established yet.

Figure 3

Nuclear Medicine: Radioisotope ventriculography (RIV), also known as radionuclear angiocardiography, can be used as an alternative method to echocardiography to measure the LV ejection fraction (EF), and presents the advantage of being a quantitative method free from geometric inferences, thereby granting it the role of a real gold standard method in this context. When simultaneous measurement of the right and left ventricular ejection fraction (RVEF, LVEF) is required, RIV has been used successfully and may detect earlier and more severe RV dysfunction, including in patients with the digestive form of Chagas' disease.57

Myocardial perfusion scintigraphy may be required for non-invasive investigation of Chagas disease patients with precordialgia. Negative findings for myocardial ischemia virtually excludes the presence of significant coronary artery disease, indicating a high negative predictive value. However, reversible perfusion defects have been detected in 30 to 50% of patients, in the absence of atherosclerotic epicardial CAD.58^-60 These perfusion changes have been attributed to coronary microcirculation in CCC and it has been postulated that such ischemic changes can contribute to regional myocardial damage in the chronic phase of the cardiomyopathy.37^,60 Fixed perfusion defects, on the other hand, are also frequently observed in patients with CCC and, in general, represent areas of fibrosis caused by the typical pathophysiology of Chagas disease.42

Iodine-123-labeled meta-iodobenzylguanidine myocardial scintigraphy (¹²³I-MIBG) allows the non-invasive evaluation of the neuronal integrity of the cardiac sympathetic nervous system at a myocardial level. The use of this imaging technique allowed the identification of regional myocardial denervation in early stages of chronic Chagas' disease in patients with no apparent impairment of left ventricular function, involving mainly the basal parts of the posterolateral and inferior walls and of the apical region.61 The results of these studies suggest that the extension/severity of the regional myocardial denervation and the intensity of the global derangement of sympathetic innervation (detected in planar images) correlate with the severity of LV systolic dysfunction.61 A more recent study has shown that cardiac patients with preserved or slightly reduced systolic function and sustained ventricular tachycardia (SVT) presented greater extension of sympathetic denervation evaluated by 123I-MIBG myocardial scintigraphy, compared to individuals without SVT, which reinforces the idea that autonomic cardiac denervation can play an important role in the arrhythmogenesis of this myocardial disease - Figure 4.39

Figure 4

Magnetic Resonance Imaging: MRI is a methodology which allows analysis of morpho-functional parameters of the heart with high degree of two-dimensional detailing, and can be quite elucidative, especially in cases where the quality of the echocardiographic images is poor, or when there are ventricular cavities with advanced geometric changes, making it difficult to perform echocardiographic measurement with the usual techniques. It is a method with great capacity for quantitative analysis of ventricular volumes and accurate calculation of LV ejection fraction.62^-66 It can also be quite useful for specific analysis of the right ventricular cavity, according to recent studies.67 More recent studies call attention for the potential of MRI for detecting the regions of myocardial fibrosis in patients with CCC and for being a potentially valuable non-invasive risk prediction tool to assess sudden death risk in these patients, even in those with preserved left ventricular ejection fraction.38 The fibrosis pattern is varied, with the presence of focal or diffusely distributed fibrosis, and even with transmural impairment, simulating a fibrosis area usually seen in myocardial infarction due to obstructive coronary disease (Figure 5).

Figure 5

Electrophysiological study (EPS): The general indications for EPS apply for patients with CCC. The EPS is required for the evaluation of the sinus node function and AV conduction when the origin of symptoms, particularly syncope, remains uncertain after noninvasive evaluation. In most patients with preserved left ventricular function who have nonsustained ventricular tachycardia or without spontaneous arrhythmia, the EPS does not provide any relevant additional prognostic information. The use of EPS has been proposed in survivors of sudden cardiac death and those with SVT for prognostic evaluation and indication of drug therapy and implantation of antiarrhythmic devices, but the data on the efficacy of this approach are still limited.68^-70

Cardiac catheterization: CCC can mimic several clinical aspects of ischemic heart disease. In fact, patients with CCC may show precordial pain, electrocardiographic changes of the ST-T segment and pathological Q waves, in addition to changes in LV segmental parietal mobility. Thus, the requirement of coronary angiography is not uncommon to rule out the presence of coronary artery disease in patients with risk factors for this condition. As stressed above, in the vast majority of patients referred for cardiac catheterization, with CCC, the subepicardial coronary arteries are essentially normal or have non-significant hemodynamically obstructive lesions.42^,59

The role of antiparasitic agents in the treatment of T. cruzi infection is considerably limited in the chronic phase of Chagas heart disease, since much reversal of established tissue damage should not be expected at these advanced stages of the disease.76

The BENEFIT study,77 released in 2015, was the only large-scale clinical trial carried out on Chagas Disease. The study randomized 2854 patients, who received benznidazole or placebo, with the essential objective of evaluating the efficacy of this treatment. Trypanosomicidal treatment was only effective to detect negative parasitological test results evaluated by the polymerase chain reaction technique (PCR) (66% in the treatment group versus 34% in the control group) even though a negative result did not correlate with clinical benefit over the 5-year follow-up.76 After the initial publication of the results, supplementary analysis comparing the outcomes verified in Brazil with the ones obtained in the other four countries allows the hypothesis that, probably due to the predominance of T. cruzi lineage II, which is more sensitive to the trypanocidal action of benznidazole, in Brazil, the effect of the etiological treatment may turn out to have a clinical benefit for chronically infected Brazilian patients without very advanced heart disease.77 Thus, the treatment could be offered, on an individual basis, for patients with this profile, in order to reach a potential decision following sharing models with the responsible physician.

The optimal approach for the management of severe ventricular arrhythmias and resuscitated sudden cardiac death secondary to CCC is still uncertain due to absolute lack of data. The first therapeutic measure in patients with CCC under risk of malignant ventricular arrhythmia is the optimization of drug therapy for those who also have heart failure, preferably with the concomitant use of beta-blockers and amiodarone.

The ICD implantation is useful in the secondary prevention of sudden cardiac death, in survivors of sudden arrhythmic death or with sustained ventricular tachycardia, especially when accompanied by hemodynamic instability. For those who are not candidates for the implantation of this device, the use of amiodarone is recommended. In fact, there is acceptable evidence of potential benefit for this antiarrhythmic drug in patients with ventricular arrhythmias of Chagas disease etiology. The concomitant use of amiodarone and beta-blockers is also recommended routinely to reduce the number of therapies, even when appropriate, due to ICD implantation in patients with CCC.

A multicenter randomized trial (CHAGASICS) is underway to assess the ICD benefit versus amiodarone, for primary prevention of sudden death in patients with CCC and high Rassi score.82 Amiodarone can be used, ideally associated with a beta-blocker, for patients with CCC, Rassi risk score of ≥10 points and nonsustained ventricular tachycardia detected on Holter monitoring.