Chart 1.

Methodological classification of selected articles

Author	Year	Country	Type of study	N. of patients	Mean age (years) (range)	Selection criteria	Group of comparison	Results
Guerrero et al.13	1991	Venezuela	Cohort	269	> 18	Yes	Yes	The higher prevalence of complex ventricular arrhythmias in chronic Chagasic patients, compared to those with primary dilated cardiomyopathy, exposes Chagasic patients to a worse prognosis.
Marin-Neto et al.14	1998	Brazil	Prospective	45	With digestive tract abnormalities = 39.4 ± 10.7	Yes	Yes	Dysautonomia is evident in chagasic patients with the digestive form, but not with the indeterminate form. There is no causal association with early myocardial damage, apparent only in patients with impaired right ventricular function, which appears to be the mechanism for the predominance of systemic over pulmonary congestion when heart failure occurs.
					Without abnormalities = 42.4 ± 11.9
Barros et al.15	2001	Brazil	Prospective	40	Not informed	Yes	Yes	TDI allowed the identification of higher isovolumic contraction of the septal wall in chagasic patients with the indeterminate form.
Barros et al.16	2002	Brazil	Prospective	30	39.9 ± 9.9	Yes	Yes	TDI allows early identification of the RV dysfunction (increased isovolumic contraction of the septal and lateral walls).
Barros et al.17	2003	Brazil	Prospective cross-sectional	77	Control: 35.8 ± 10.6	Yes	Yes	Tissue Doppler imaging (TDI) identifies early contractility abnormalities.
					Chagasic com normal ECG: 40.7 ± 8.5
					Chagasic with abnormal ECG: 43.9 ± 11.1
Arias et al.18	2003	Brazil	Prospective cross-sectional	60	> 18	Yes	Yes	Troponin levels are elevated in different clinical presentations of Chagas' disease and detect early inflammation.
Nunes et al.19	2004	Brazil	Prospective cross-sectional	74	47.5 ± 12.9 (22-73)	Yes	No	RV dysfunction was significantly associated with LV involvement and pulmonary hypertension, which was more correlated with RV overload than to contractility.
Viotti et al.20	2004	Argentina	Cohort	849	> 18	Yes	No	Echocardiography determines the prognosis of chronic Chagasic patients without congestive heart failure (CHF).
Talvani et al.21	2004	Brazil	Prospective cross-sectional	81	43.5 ± 11.1	Yes	Yes	High levels of brain natriuretic peptide (BNP) in Chagasic patients detect heart failure (HF) or severe HF manifestations, generating arrhythmias.
Rocha et al.22	2004	Brazil	Cohort	60	47 ± 13	Yes	No	Left atrial volume (LAV) was an important predictor of death in Chagas disease.
Freitas et al.23	2005	Brazil	Cohort	1220	13 ± 72 (45)	Yes	No	Chagasic etiology in HF patients was the main prognostic factor for mortality.
Melo et al.24	2005	Brazil	Transversal	25	62.7 ± 7.7 chronic 42.2 ± 11.7 asymptomatic	Yes	Yes (Chagas asymptomatic x symptomatic)	BNP increases with worsening functional class and cardiac area.
Pazin-Filho et al.25	2006	Brazil	Prospective cross-sectional	59	37-76 > 55	Yes	No	In time, slight alterations in the left ventricular ejection fraction (LVEF) lead to the worsening of global LV systolic function.
Rassi Jr et al.26	2006	Brazil	Cohort	424	47 ± 11	Yes	Yes	Development of death risk assessment in chagasic cardiomyopathy based on six factors
Chaves et al.27	2006	Colombia	Prospective cross-sectional	430	Md = 40.5 (IQR = 36-45)	Yes	Yes	Comparing controls, stage I and stage II using tissue Doppler, it is possible to identify, in stage II, the greater relaxation time, lower velocity of A wave, higher E/A ratio, lower pulmonary A velocity, greater pulmonary/mitral A velocity ratio, greater annulus A wave velocity, and the RV shows increased tricuspid A wave velocity and isovolumic contraction time.
Benchimol Barbosa28	2007	Brazil	Cohort	50	34-74	Yes	No	In the male gender, pathological Q wave, ventricular extrasystoles, ventricular tachycardia and abnormal echocardiogram (diastolic and systolic dysfunction and apical aneurysm) are markers of poorer prognosis.
Barbosa et al.29	2007	Brazil	Prospective cross-sectional	59	48 ± 11	Yes	No	The pro-BNP levels remained a strong correlation with LVEF and duration indices of diastolic dysfunction of left atrial volume, as well as with the more severe forms of the disease, allowing their differentiation from the moderate forms.
Nunes et al.30	2008	Brazil	Cohort	158	48 ± 12	Yes	No	Evaluation of systolic and diastolic function using the Tei index (myocardial performance index) and CF showed to be useful prognostic tool in Chagasic cardiomyopathy.
Villas-Boas et al.31	2008	Brazil	Prospective cross-sectional	38	51.3 ± 1.6 for chagasic and 53.3 ± 2.3 for control patients	Yes	Yes	The concentration of BNP is increased and correlates with heart failure (direct), functional class (direct) natremia (reverse), blood pressure (reverse), right atrial pressure, R/L ventricular dysfunction, TNF-alpha (demonstrating inflammatory characteristic of the disease).
Nunes et al.32	2009	Brazil	Cohort	192	48 ± 12	Yes	No	LAV provides powerful information to predict mortality regardless of clinical and echocardiographic data.
Del Castillo et al.33	2009	Brazil	Prospective cross-sectional	40	55 ± 10	Yes	Yes	Based on the principle of speckle tracking and employing the assessment of intramyocardial velocity gradients, they identified the reduction in the percentage of deformation (X-strain) and the rate of deformation (strain rate) of the infero-lateral left ventricular wall in Chagas patients unidentifiable by clinical data or conventional echocardiography.
Terzi et al.34	2010	Brazil	Prospective cross-sectional	62	58	Yes	No	Chagasic patients with normal LVEF and normal electrocardiogram had a significant frequency of contractile alterations related with arrhythmias and reduced LVEF.
Pereira Nunes et al.35	2010	Brazil	Cohort	287	Chagasic: 48.3 ± 12	Yes	Yes	Identification of Chagasic etiology resulted in poor prognosis in patients with HF, regardless of clinical and echocardiography data.
					Idiopathic: 49.6 ± 15.9
Nunes et al.36	2010	Brazil	Cohort	65	48.6 ± 9.1	Yes	No	RV dysfunction is an important determinant, regardless of exercise capacity, and the systolic valvular velocity of this ventricle was associated with the peak volume of O2 consumed, regardless of age, gender and echocardiographic parameters.
Garcia-Alvarez et al.37	2010	Spain	Cohort	54	20-58 (mean 37)	Yes	Yes (4 groups)	The identification of diastolic dysfunction can be achieved by increasing BNP, which correlates with functional class (NYHA), TNF-alpha, PS, endothelin, ANP and reduction in the percentage of longitudinal and radial deformation to assess contractility.
Duarte et al.38	2011	Brazil	Cohort	56	56 ± 10	Yes	No	Although high prevalence of intraventricular dyssynchrony and moderate interventricular dyssynchrony was demonstrated among Chagasic patients, especially in patients without a pacemaker, dyssynchrony did not show to be a risk factor for poor prognosis.
Valerio et al.39	2011	Spain	Descriptive	100	38.2 ± 10.2	Yes	No	The fact that Spain has become a center of migration of Latin Americans in Europe has aroused the concern of Chagas disease becoming an important cause of cardiomyopathy in health services.
Nunes et al.40	2012	Brazil	Prospective endpoint	232	48 ± 12	Yes	Yes (survivors and deaths)	Using tissue Doppler imaging, they identified risk factors for death: functional class III and IV increased RV Tei, increased left atrial volume index and interaction of LV ejection fraction and E / E' index and protectors: E / E ' ratio and reduction in LV ejection fraction.
Vasconcelos and Junqueira41	2012	Brazil	Prospective cross-sectional	15 controls and 13 chagasic	controls = 40-46	Yes	Yes	The sympathetic and parasympathetic depressions with preserved balance were associated with heart rate variability and increased left ventricular systolic diameter in Chagasic patients with cardiomyopathy, indicating that these depressions may precede and be independently more severe than ventricular dysfunction, with no causal association between this depression and dysfunction.
					chagasic = 35-49
Del Castillo et al.42	2012	Brazil	Prospective cross-sectional	20	54.8 ± 13.5	Yes	Yes	Employing the speckle tracking technique, it was possible to prove reduction of twisting and torsional deformation of myocardial fibers in Chagas cardiomyopathy, thus differentiating it from ventricular hypertrophy
Melo et al.43	2010	Brazil	Cross-sectional	150	G1:51.02 ± 1.256	Yes	Yes	It was demonstrated that both groups had increased inflammatory activity, but there were no indications of greater prothrombotic status among chagasic patients.
					G2: 54.88 ± 0.8787

Legend: ANP: Atrial Natriuretic Peptide, BNP: brain natriuretic peptide; DTI: Doppler tissue imaging (Image by tissue Doppler), CF: cardiac function, LVEF, left ventricular ejection fraction, HF: Heart failure; NYHA: New York Heart Association; O2: oxygen; Tei: myocardial performance index, TNF: Tumor Necrosis Factor; LAV: left atrial volume, RV: right ventricle, LV: left ventricle.