## Abstract

Background Chagas disease affects mainly poor populations in Latin America. This review assesses the evidence on the independent risk of cardiovascular events associated with positive Chagas serology.

Methods We searched for studies using the following outcomes: death, stroke, new onset heart failure, heart failure hospitalization or evidence of left ventricular dysfunction. Studies comparing patients with positive serology for Chagas with a control group with a follow-up longer than 1 year were selected. The Medline, Lilacs and Embase databases were searched on 21 January 2011 without restrictions.

Results From 5236 potentially relevant studies, 25 fulfilled the inclusion criteria.

Fourteen included patients with heart failure, six with severe symptoms and nine with mild symptoms or asymptomatic patients with low ejection fraction. In four studies of patients in functional class III or IV and in three studies of patients with mild symptoms, a higher risk of death was reported among those with positive serology for Chagas.

Of the 11 studies of patients without symptoms or low ejection fraction, 3 showed a higher risk of mortality related to Chagas exposure. Two of these were based on the same cohort of people aged >60 years.

Overall, 8 out of the 14 heart failure studies and 2 out of the 11 heart damage studies adjusted for confounding factors.

Conclusion Positive serology for Chagas is associated with a higher risk of death for patients with heart failure. However, there is little evidence to link positive serology for Chagas with cardiovascular events in asymptomatic subjects.

## Introduction

Chagas disease is a public health problem that mostly affects poor rural populations in Latin America.1 However, a gradual increase in positive serology for Chagas is now being observed in non-endemic urban areas as a consequence of high rural to urban migration.2 Chagas has also been raised as a concern in non-endemic countries as a result of international migration.3 Chagas disease is caused by the parasite Trypanosoma cruzi. It is most commonly transmitted to humans through the faeces of haematophagous insects of the sub-family Triatomine, including Triatoma infestans and Rhodnius prolixus. Chagas can also be transmitted congenitally, through transfusions, and occasionally via oral contamination, although these account for a small minority of cases.4 The disease has two clinical phases: (i) an acute phase that is usually self-limited and asymptomatic in 95% of patients4 and (ii) a chronic phase, with a lengthy latency period of 10–30 years, where a percentage of infected individuals develop signs or symptoms consistent with cardiomyopathy.1,4,5 However, it is important to note that appropriate cohort studies specifically designed to determine the percentage of seropositive patients who will develop cardiomyopathy have not been published. Although widely cited, the estimate that 20–30% of seropositive individuals will go on to develop cardiomyopathy is therefore questionable.57

Chronic Chagas cardiomyopathy has been described as the most common cause of non-ischaemic cardiomyopathy worldwide.8 However, a consensus does not exist regarding the actual risk of cardiomyopathy and other cardiovascular events associated with positive Chagas serology independent of the confounding effect of poverty. Socio-economic status is associated with multiple confounding factors that can also increase the risk of developing cardiovascular events. Indeed, there is evidence of an association between poverty and heart failure.911 The associations between socio-economic status and the risk factors of heart failure and cardiovascular events, such as heart disease,12,13 poor access to risk factors treatment,1417 tobacco consumption18 or alcohol intake,19,20 are also well documented. This systematic review aims to assess the evidence base on the independent risk of cardiovascular events associated with positive Chagas serology.

### Methods

We searched the literature for studies comparing patients with positive serology for Chagas with a control group. To be included, a study needed to have a follow-up period >1 year and must have included at least one of the following end points: death, strokes, hospitalizations due to heart failure or development of myocardial injury evidenced by impaired ventricular function assessed by diagnostic imaging techniques. The damage measured by electrocardiographic examinations was not considered because no specific electrocardiographic alteration demonstrated consistently prognostic value.21

Studies that met the above-mentioned criteria were considered eligible, regardless of their design, whether published and the language in which they were written. Studies that assessed the prognosis of positive serology for Chagas disease in patients treated with devices (re-synchronization therapy or implantable cardioverter-defibrillator), invasive surgical procedures (cardiomyoplasty or ventricular remodelling surgery) or cardiac transplantation were excluded.

We searched for studies across three databases without language restrictions: Medline (1946–2011), Lilacs (1986–2011) and EMBASE (1947–2011). A strategy that showed high sensitivity (90%) to detect prognostic studies was used: [incidence (MeSH) or mortality (MeSH) or follow-up studies (MeSH) or prognos*(Text Word) or predict*(Text Word) or course*(Text Word)].22 As a complement, to further maximize the sensitivity, we also added words such as ‘cohort’ or ‘follow’. Therefore, the search strategy for Medline was: {[(Chagas Disease (MeSH) OR Chagas*(tiab) OR cruzi*(tiab) OR Cardiovascular Trypanosom*(tiab)] AND [incidence (MeSH) or mortality (MeSH) or follow-up studies (MeSH) or prognos*(Text Word) or predict*(Text Word) or course*(Text Word)] or [Cohort Studies (MeSH)] OR cohort*(all) OR follow*(all)]}. Similar criteria were used for the other two databases, including Spanish and Portuguese words for the search at Lilacs: (MH Enfermedad de Chagas OR Chagas$OR cruzi$ OR Trypanosom$) AND (MH Incidencia OR MH Mortalidad OR MH Estudios de Seguimiento OR prognos$ OR pronost$OR predic$ OR course$OR curso$ OR MH Estudios de Cohortes OR cohort$OR follow OR seguim$). We also consulted renowned Latin American experts to obtain unpublished information. All the titles and abstracts were assessed by two independent investigators (B.L., J.K.), who eliminated those studies that did not meet the inclusion criteria. Disagreements were resolved by consensus, and in the case of persistent dissent, the full text of the article was examined.

The full text of studies meeting the inclusion criteria was analysed by two reviewers who extracted the following points:

• General description of the study: Years of field work and publication, country of origin, whether the included participants suffered from heart damage and how it was defined, whether they were hospitalized and whether the sample was drawn from urban or rural areas.

• Patient data: Number of Chagasic and non-Chagasic subjects, along with description of these groups (age, sex and blood pressure).

• End points: How the risk of experiencing each end point was reported, comparing Chagasic and non-Chagasic groups (odds ratios, hazard ratios, relative risk, β coefficients, confidence intervals or standard errors, number of events and P value), and whether an adjustment by statistical methods was performed (and if so, which variables were used).

• Study quality23,24: Sampling method (probabilistic or using consecutive patients), along with whether it included a sample size calculation, a clear description of the inclusion and exclusion criteria, explanation of the follow-up method and the number of individuals lost in follow-up, adequate measurement of exposure to Chagas (two positive serologies by different methods as it is recommended today)25 and the events (how were the events defined and whether they were properly measured), confounding control through statistical methods and statistical analysis (missing data management and loss to follow-up, time-to-event analysis).

## Results

### Search

On 21 January, 2011, a search of Medline, Lilacs and Embase was carried out, finding a total of 5236 articles. After removing duplicated articles, 3617 articles were left for the screening of titles and abstracts. In this step, 3546 articles were deemed irrelevant, leaving 71 articles for a complete reading, although one of them could not be found.26 After reading, another 45 articles were eliminated, yielding a final set of 25 articles (Figure 1).

Figure 1

Details of study selection process

Figure 1

Details of study selection process

### Description of studies

Table 1 summarizes the studies included in this review. The studies were classified according to the clinical condition of subjects included in it. In 14 studies, the patients had heart failure or impaired ventricular function assessed by diagnostic imaging techniques. Six of these studies27–32 assessed patients with advanced heart failure (functional class III or IV), and nine studies included patients with mild symptoms or those who were asymptomatic but with an impaired ejection fraction.30,3340 Heringer et al.’s30 publication included both types of patients; thus, it is considered in both sets of conditions. Eleven studies included patients without heart failure or impaired ventricular function assessed by diagnostic imaging techniques.6,4150

Table 1

Descriptive characteristics of included studies

Corresponding author Year of publication Country Patients or individuals included and heart damage definition Year study began Year study ended
Studies that included patients with heart failure in functional class III or IV
De Campos 2006 Brazil Patients with FC III or IV, hospitalized for having descompensated heart failure 1998 2000
Cardoso 2010 Brazil Patients with FC IV, with LVEF <45. Stevenson profile C (Cold and wet) 2006 2007
Freitas 2005 Brazil Patients with FC III o IV as a result of systolic dysfunction, <75, refractory to treatment or heart transplant candidates 1991 2000
De Olivera 2005 Brazil Patients with FC IV as a result of systolic dysfunction 1993 1995
Heringera 2006 Brazil–Germany Patients with ICC FC III or IV, with LVEF <50% 2001 2006
Silva 2008 Brazil Patients with FC III or IV, hospitalized for having descompensated heart failure Not specified
Studies that included patients with heart failure in functional class I or II or asymptomatic with low LVEF measured
Nunes 2006 Brazil Patients with myocardiopathy in whom myocardic biopsy was performed, with LVEF <40% 1981 1998
Bestetti 1997 Brazil Chronic Chagasic cardiopathy (definition not specified). Patients with negative serology with a dilated left ventricle by echocardiography or radiography 1990 1993
Braga 2008 Brazil Heart failure diagnosis and LVEF <45% 2003 2004
Heringera 2006 Brazil–Germany Patients with FC I or II and LVEF <50% 2001 2006
Rassi 2005 Brazil HF symptoms with recent onset (<6 weeks) and LVEF <40% Not specified
Traina 2010 USA Patients with non-ischemic heart failure with LVEF <40 with history of having lived at least 12 month in a Latin-American country Not specified
Conceisao Souza 2009 Brazil HF symptoms with recent onset (less than a month and LVEF <45%) 2008 2009
Cruz 2010 Brazil Patients followed by a heart failure clinics Not specified
Pereira Nunes 2010 Brazil Patients with diagnosis of dilated cardiomyopathy, defined by LVEF <55% and D41S ≥ 31 1999 2008
Studies that included asymptomatic individuals without normal LVEF or with no LVEF measured
Pimenta 1999 Brazil Asymptomatic patients with ECG with right bundle block. Positive serology diagnosed in workplace or blood donor screening Not specified
Arribada 1987 Chile Individuals with abnormal electrocardiogram 1981 1984
Lauria Pires 2000 Brazil Street cleaners, urban area 1988 1998
Lima Costa A 2010 Brazil Individuals >60 years of age included in Bambui City cohort 1997 2007
Lima Costa Bb 2010 Brazil Individuals >65 years of age included In Bambui City cohort 1997 2007
Mota 1990 Brazil Rural population >5 years of age 1974 1983
Maguire 1987 Brazil Rural population >5 years of age 1974 1980
Puigbo 1969 Venezuela Rural population 1961 1965
Coura 1985 Brazil Not specified 1974 1984
Pereira 1985 Brazil Urban population 1976 1980
Espinozac 1985 Brazil Not specified 1973 1983
Corresponding author Year of publication Country Patients or individuals included and heart damage definition Year study began Year study ended
Studies that included patients with heart failure in functional class III or IV
De Campos 2006 Brazil Patients with FC III or IV, hospitalized for having descompensated heart failure 1998 2000
Cardoso 2010 Brazil Patients with FC IV, with LVEF <45. Stevenson profile C (Cold and wet) 2006 2007
Freitas 2005 Brazil Patients with FC III o IV as a result of systolic dysfunction, <75, refractory to treatment or heart transplant candidates 1991 2000
De Olivera 2005 Brazil Patients with FC IV as a result of systolic dysfunction 1993 1995
Heringera 2006 Brazil–Germany Patients with ICC FC III or IV, with LVEF <50% 2001 2006
Silva 2008 Brazil Patients with FC III or IV, hospitalized for having descompensated heart failure Not specified
Studies that included patients with heart failure in functional class I or II or asymptomatic with low LVEF measured
Nunes 2006 Brazil Patients with myocardiopathy in whom myocardic biopsy was performed, with LVEF <40% 1981 1998
Bestetti 1997 Brazil Chronic Chagasic cardiopathy (definition not specified). Patients with negative serology with a dilated left ventricle by echocardiography or radiography 1990 1993
Braga 2008 Brazil Heart failure diagnosis and LVEF <45% 2003 2004
Heringera 2006 Brazil–Germany Patients with FC I or II and LVEF <50% 2001 2006
Rassi 2005 Brazil HF symptoms with recent onset (<6 weeks) and LVEF <40% Not specified
Traina 2010 USA Patients with non-ischemic heart failure with LVEF <40 with history of having lived at least 12 month in a Latin-American country Not specified
Conceisao Souza 2009 Brazil HF symptoms with recent onset (less than a month and LVEF <45%) 2008 2009
Cruz 2010 Brazil Patients followed by a heart failure clinics Not specified
Pereira Nunes 2010 Brazil Patients with diagnosis of dilated cardiomyopathy, defined by LVEF <55% and D41S ≥ 31 1999 2008
Studies that included asymptomatic individuals without normal LVEF or with no LVEF measured
Pimenta 1999 Brazil Asymptomatic patients with ECG with right bundle block. Positive serology diagnosed in workplace or blood donor screening Not specified
Arribada 1987 Chile Individuals with abnormal electrocardiogram 1981 1984
Lauria Pires 2000 Brazil Street cleaners, urban area 1988 1998
Lima Costa A 2010 Brazil Individuals >60 years of age included in Bambui City cohort 1997 2007
Lima Costa Bb 2010 Brazil Individuals >65 years of age included In Bambui City cohort 1997 2007
Mota 1990 Brazil Rural population >5 years of age 1974 1983
Maguire 1987 Brazil Rural population >5 years of age 1974 1980
Puigbo 1969 Venezuela Rural population 1961 1965
Coura 1985 Brazil Not specified 1974 1984
Pereira 1985 Brazil Urban population 1976 1980
Espinozac 1985 Brazil Not specified 1973 1983

aAre the same study.

bBoth are included because they report different end point.

cIt also included heart failure or low LVEF. It only included the groups that have a comparative group with negative serology and includes all the patients in whom a myocardial biopsy was performed during the study years.

HF, heart failure; LVEF, Left ventricle ejection fraction; FC, functional class; LVDD/BS, left ventricular diastolic diameter/body surface.

Only 1 of the 14 studies that evaluated patients with heart failure was published before year 2000,34 whereas studies involving individuals without heart failure were mostly published before that date.6,4143,4650 Most of the studies were conducted in Brazil, except for three studies that were conducted in Chile,41 Venezuela49 and the United States.40 Most of the studies were published in journals, whereas three studies were only presented as abstracts in conferences/meetings.37,38,40 We did not find any unpublished sources.

Most of the studies identified were self-described as cohort studies. The majority of these were prospective,6,27,31,3340,42,4446 two were referred to as retrospective30,43 and in three of the studies, it was not specified whether the follow-up was prospective.28,29,32 Four studies were self-described as repeated surveys and not as a formal cohort.41,47,48,50

### Study results

Table 2 describes the number of patients included with and without Chagas and their demographic characteristics along with the main results of each study. Most of the studies describe the results in terms of death, with only four studies using other end points such as hospital admissions,37,40,51 development of heart failure symptoms or ventricular dysfunction6,44 and death due to stroke.44

Table 2

Characteristics of patients included and main results of each study

Corresponding author Sub-groups N patients included Age % Male End point N or % of events reported Effect (crude or adjusted)
S + S − S + S − S + S − S + S − Point estimation used and valuea CI 95% P Adjusting variables
Studies that included patients with heart failure in functional class III or IV
De Campos  102 392 57.5 ± 14.2 69.8% Death 75 NR OR 3.34 (1.9–5.89) <0.001 Sex, age, treatment, LVEF, diabetes, alcoholism, histroy of myocardial infarction, number of hospitalizations because of heart failure
Cardoso  33 67 52.9 ±14.5 59.8 ±14.9 54.5% 75.6% Death 66.7% 37.3% HR 2.75 (1.35–5.63) NR Hazard ratio for aetiology and age are reported. LVEF is significantly asociated but it is not reported. Rest of the variables included in the models are unknown
Freitas Idiopatic MD 242 454 43 ±10.8 44 ±11.2 79% 78% Death 110 156 HR 2.66 (1.10–6.46) 0.03 Sex, age, aetiology, haemodynamic variables by catheterisation, diameters and LVEF by echocardiography, rhythm and other variables by Holter
Hypertensive MD 170 48.7±8.6 45% 45 9.14 (2.20–38.03) 0.0024
Ischaemic MD 212 51.7 ±8.2 68% 64 4.91 (1.60–15.08) 0.0054
Other aetiologies 142 41 ±11.7 70% 40 6.12 (1.75–21.47) 0.0047
De Olivera  56 70 51.1 ±14.2 73% Death 50 51 HR 1.63 (1.10–1.43) 0.023 None. In multivariate analysis, aetiology is not significantly associated with death, but HR is not reported
Heringer  14 25 50.5 ±2.7 45.1 ±2.3 43% 64% Death NR NR NR NR NR
Silva  122 232 47.5 ±12 53 ±14.9 73.8% 69.4% Death 69.2% 47.9% NR NR NR P =0.001 In multivariate analysis, S+ is related to death (variables included with univariate significative association: age, blood pressure, LVEF, sodium and TNF levels)
Studies that included patients with heart failure in functional class I or II or asymptomatic with low LVEF measured
Nunes  26 19 42 ±2 43 ±3 89% 84% Death 3.8% 10.5% NR NR NR 0.131 NR
Bestetti  75 50 55 ±14 59 ±14 56% 70% Death 21 (28%) 3 (6%) OR 6.09 (1.70–21.73) 0.002 No
Braga  89 102 54 ±12.6 52.7 ±13.4 53.8% 59.4% Death 16 (21.6%) 10 (10.6%) OR 1.67 (0.67–4.41) NR Educational level achieved
Heringer  18 25 50.9 ±1.9 39% 52% Death NR NR NR NR NR
Rassi  57 45 61.5 ±13.5 61.8% Death NR NR RR 10.62 (3.46–32.6) <0.001 Age, history of diabetes, functional class, physical examination, laboratory, echocardiogram, electrocardiogram and radiography
Traina  22 102 57.5 55 50% 59% HF hospitalizations 22 50 TASA (100 PY) S+: 70.3 ±15 / S–: 27.2 ±3.9  0.0223 NR
Conceisao Souza (2)  100 62 NR NR Deathb 6% 4.8% NR NR NR P =ns NR
Cruz Idiopathic MD 21 55 50.7 ±11.8 73% Death 38.0% 20.0% 0.0% 16.0% NR NR NR NR NR
Hypertensive MD 22
Ischaemic MD 20
Pereira Nunes  224 63 48.3 ±12 49.6 ±15.9 61% 60% Death NR NR HR 2.48 (1.28; 4.78) 0.007 Sex, age, antecedents, treatment, echocardiography
Asymptomatic patients without demonstrated impaired ejection fraction
Pimenta  55 29 47.5 ±11.2 NR Death 20 NR NR NR 0.895 NA
Arribada  298 402 Range 11–90 36% 40% Death 23 (7.7%) 9 (2.2%) NR NR NR P <0.001 NA
Lauria Pires With treatment 45 41 90% between 31 and 60 years NR 100% Death NR NR NR NR NR
Without treatment 46
Lima Costa A  524 874 69.2 ±6.9 68.6 ±7.0 32% 44% Death due to stroke 25 20 HR 2.35 (1.25; 4.44) NR Sex, age, education, number of risk factors and CRP
Lima Costa B  563 915 69.1 ±7.2 39% Death NR NR HR 1.56 (1.32; 1.85) NR Sex, age, education, number of risk factors
Mota  488 509 ≥10 years 45% 45% Death 32 28 Rate (1000 PY) S+: 8.9 / S–:7.8 NR NR Age
Maguire  212 116 ≥10 years NR Death NR NR RR 1.2 (0.6;2.4) 0.27 Age
Puigbo  359 453 ≥5 years NR Death 17 NR NR NR NR NR
Coura  235 216 Range 10–80 years NR Death 54 (23%) 23 (10.6%) NR NR NR NR NR
Pereira  192 188 NR NR Death 22 (11.2%) 6 (3.2%) NR NR NR NR NR
Espinoza  18 22 37 ±13 38 ±13 50% 55% Deathc NR NR NR NR NR
Corresponding author Sub-groups N patients included Age % Male End point N or % of events reported Effect (crude or adjusted)
S + S − S + S − S + S − S + S − Point estimation used and valuea CI 95% P Adjusting variables
Studies that included patients with heart failure in functional class III or IV
De Campos  102 392 57.5 ± 14.2 69.8% Death 75 NR OR 3.34 (1.9–5.89) <0.001 Sex, age, treatment, LVEF, diabetes, alcoholism, histroy of myocardial infarction, number of hospitalizations because of heart failure
Cardoso  33 67 52.9 ±14.5 59.8 ±14.9 54.5% 75.6% Death 66.7% 37.3% HR 2.75 (1.35–5.63) NR Hazard ratio for aetiology and age are reported. LVEF is significantly asociated but it is not reported. Rest of the variables included in the models are unknown
Freitas Idiopatic MD 242 454 43 ±10.8 44 ±11.2 79% 78% Death 110 156 HR 2.66 (1.10–6.46) 0.03 Sex, age, aetiology, haemodynamic variables by catheterisation, diameters and LVEF by echocardiography, rhythm and other variables by Holter
Hypertensive MD 170 48.7±8.6 45% 45 9.14 (2.20–38.03) 0.0024
Ischaemic MD 212 51.7 ±8.2 68% 64 4.91 (1.60–15.08) 0.0054
Other aetiologies 142 41 ±11.7 70% 40 6.12 (1.75–21.47) 0.0047
De Olivera  56 70 51.1 ±14.2 73% Death 50 51 HR 1.63 (1.10–1.43) 0.023 None. In multivariate analysis, aetiology is not significantly associated with death, but HR is not reported
Heringer  14 25 50.5 ±2.7 45.1 ±2.3 43% 64% Death NR NR NR NR NR
Silva  122 232 47.5 ±12 53 ±14.9 73.8% 69.4% Death 69.2% 47.9% NR NR NR P =0.001 In multivariate analysis, S+ is related to death (variables included with univariate significative association: age, blood pressure, LVEF, sodium and TNF levels)
Studies that included patients with heart failure in functional class I or II or asymptomatic with low LVEF measured
Nunes  26 19 42 ±2 43 ±3 89% 84% Death 3.8% 10.5% NR NR NR 0.131 NR
Bestetti  75 50 55 ±14 59 ±14 56% 70% Death 21 (28%) 3 (6%) OR 6.09 (1.70–21.73) 0.002 No
Braga  89 102 54 ±12.6 52.7 ±13.4 53.8% 59.4% Death 16 (21.6%) 10 (10.6%) OR 1.67 (0.67–4.41) NR Educational level achieved
Heringer  18 25 50.9 ±1.9 39% 52% Death NR NR NR NR NR
Rassi  57 45 61.5 ±13.5 61.8% Death NR NR RR 10.62 (3.46–32.6) <0.001 Age, history of diabetes, functional class, physical examination, laboratory, echocardiogram, electrocardiogram and radiography
Traina  22 102 57.5 55 50% 59% HF hospitalizations 22 50 TASA (100 PY) S+: 70.3 ±15 / S–: 27.2 ±3.9  0.0223 NR
Conceisao Souza (2)  100 62 NR NR Deathb 6% 4.8% NR NR NR P =ns NR
Cruz Idiopathic MD 21 55 50.7 ±11.8 73% Death 38.0% 20.0% 0.0% 16.0% NR NR NR NR NR
Hypertensive MD 22
Ischaemic MD 20
Pereira Nunes  224 63 48.3 ±12 49.6 ±15.9 61% 60% Death NR NR HR 2.48 (1.28; 4.78) 0.007 Sex, age, antecedents, treatment, echocardiography
Asymptomatic patients without demonstrated impaired ejection fraction
Pimenta  55 29 47.5 ±11.2 NR Death 20 NR NR NR 0.895 NA
Arribada  298 402 Range 11–90 36% 40% Death 23 (7.7%) 9 (2.2%) NR NR NR P <0.001 NA
Lauria Pires With treatment 45 41 90% between 31 and 60 years NR 100% Death NR NR NR NR NR
Without treatment 46
Lima Costa A  524 874 69.2 ±6.9 68.6 ±7.0 32% 44% Death due to stroke 25 20 HR 2.35 (1.25; 4.44) NR Sex, age, education, number of risk factors and CRP
Lima Costa B  563 915 69.1 ±7.2 39% Death NR NR HR 1.56 (1.32; 1.85) NR Sex, age, education, number of risk factors
Mota  488 509 ≥10 years 45% 45% Death 32 28 Rate (1000 PY) S+: 8.9 / S–:7.8 NR NR Age
Maguire  212 116 ≥10 years NR Death NR NR RR 1.2 (0.6;2.4) 0.27 Age
Puigbo  359 453 ≥5 years NR Death 17 NR NR NR NR NR
Coura  235 216 Range 10–80 years NR Death 54 (23%) 23 (10.6%) NR NR NR NR NR
Pereira  192 188 NR NR Death 22 (11.2%) 6 (3.2%) NR NR NR NR NR
Espinoza  18 22 37 ±13 38 ±13 50% 55% Deathc NR NR NR NR NR

aTaking negative serology as a reference.

bHospital admissions: S+ 19%, S– 9.7%. P = 0.01.

cAnother end point: development of heart failure symptoms: Non- S+ or S–. Ventricular dysfunction: S+ 1, S– 0

NR, not reported; OR, Odds ratio; HR, hazard ratio; CRP, C-reactive protein; LVEF, left ventricular ejection fraction.

#### Studies in patients with advanced heart failure

In four out of six studies that included patients with functional class III or IV heart failure, a higher risk of death in patients with positive serology was reported, adjusted only by clinical variables.27,28,31,32 One of the remaining studies did not report whether the risk was increased,30 whereas the other one reported a high risk in the univariable analysis, but not in the multivariable analysis.29 None of these studies adjusted for socio-economic status or poverty.

#### Studies in patients with heart failure with mild or no symptoms and with impaired ejection fraction

In three out of nine studies with patients with impaired ventricular function or heart failure with mild symptoms, a higher risk of death was observed in the population exposed to Chagas.34,36,39 In three other studies, no differences were found33,35,37 (one of them, only after adjusting for education level).35 On the other hand, two studies did not report the results of a statistical test and present non-conclusive results based on small samples.30,38 Only one study assessed the risk of hospitalization in this sub-group, detecting an increased risk for subjects with positive serology.40 Only two studies of this group of patients performed an adjustment through a multivariable analysis, one analysis per educational level,35 and the other through multiple clinical, electrocardiographic, radiological and echocardiographic variables.36

#### Studies in asymptomatic patients without demonstrated impaired ejection fraction

Studies that evaluated subjects without heart failure or impaired ventricular function, as assessed by diagnostic imaging techniques, also showed mixed results. Among these studies, only three showed a higher risk of cardiovascular events among patients with positive serology for Chagas. Lima Costa’s studies,44,45 both of which focus on patients aged ≥60 years, show an increase in total and stroke mortality among patients with a positive serology for Chagas in a follow-up of 10 years, after adjusting by factors such as sex, age, education and number of risk factors. Arribada et al.’s41 study showed a greater crude mortality rate in patients with altered electrocardiography (7.7% for patients with Chagas and 2.2% for control subjects) after a 4-year period. In the other studies of this sub-group, statistical tests were not reported, except for studies by Maguire et al.46 and Pimenta et al.,42 where patients with positive serology were not found to experience a higher risk of mortality.

### Quality assessment

Only one study that examined advanced heart failure32 and two studies that included patients with mild symptoms or with impaired ventricular dysfunction35,39 reported data on patient attrition. On the other hand, in most of the studies with asymptomatic subjects, these data were described.6,41,43,44,46,48,50 The loss to follow-up rate in this sub-group reached values of up to 44.8%, and in some studies, the loss to follow-up was different in populations with and without positive serology for Chagas6,41,43,50 (Table 3).

Table 3

Follow-up and quality assessment by study

Corresponding author Follow-up Inclusion of individuals/patients Follow-up method reported Variables properly measured Bias and control of confounding factors Statistical analysis
Time (months) Lost in follow-up (%) Inclusion criterion Exclusion criterion Methods of selection of participants reported Method of selection Chagas exposure Events Multivariate analysis Exposed and non-exposed comparables Explained sample size Method described Sub-groups and interaction Time-to- event analysis Management of loss in follow-up described
S + S − S + S − Consecutive patients Census
Studies that included patients with heart failure in functional class III or IV
De Campos Atleast 24 NR ✓ ✓ ✓ ✓ ✕ ✓ ✕ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✕
Cardoso 25 NR ✓ ✓ ✓ ✓ ✕ ✕ ✓ ✓ ✓ ✓ ✕ ✓ ✕ ✓ ✕
Freitas 120 8.7% Between 3.5% and 10.6%a ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓
De Olivera 15.3 ±13.8 NR ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✓ ✓ ✕ ✕ ✓ ✕ ✓ ✕
Heringer 31.4 (range 13 to 54) NR ✕ ✕ ✕ ✓ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Silva 12 NR ✓ ✕ ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕
Studies that included patients with heart failure in functional class I or II or asymptomatic with low LVEF measured
Nunes 60 NR ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✓ ✕
Bestetti NR NR ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Braga 12 12% total ✓ ✕ ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕
Heringer 31.4 NR ✕ ✕ ✕ ✓ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Rassi 24 NR ✓ ✓ ✓ ✓ ✕ ✓ ✕ ✓ ✓ ✕ ✕ ✓ ✕ ✓ ✕
Traina 20.3 22.8 NR ✓ ✕ ✓ ✕ ✕ ✕ ✕ ✓ ✕ ✓ ✕ ✕ ✕ ✕ ✕
Conceisao Souza (2) 12 NR ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✓ ✓ ✕ ✓ ✕
Cruz 13.4 NR  ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Pereira Nunes 39.5 0.7% ✓ ✓ ✓ ✕ ✕ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓
Asymptomatic patients without demonstrated impaired ejection fraction
Pimenta 121 94.9 NR ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✓ ✕
Arribada 48 25.8% 44.0% ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Lauria Pires 120  15.6%y 15.5%b 0% ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Lima Costa A 120 6% total ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕
Lima Costa B 104.6 NR ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕
Mota 86.4 94.8 NR ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✓ ✕
Maguire 84 0.5% 4.3% ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✓ ✕
Puigbo 48 NR ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Coura 120 9.6% 16.9% ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✓
Pereira 72 33.6% 31.4% ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Espinoza 72 113 0% 9.09% ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Corresponding author Follow-up Inclusion of individuals/patients Follow-up method reported Variables properly measured Bias and control of confounding factors Statistical analysis
Time (months) Lost in follow-up (%) Inclusion criterion Exclusion criterion Methods of selection of participants reported Method of selection Chagas exposure Events Multivariate analysis Exposed and non-exposed comparables Explained sample size Method described Sub-groups and interaction Time-to- event analysis Management of loss in follow-up described
S + S − S + S − Consecutive patients Census
Studies that included patients with heart failure in functional class III or IV
De Campos Atleast 24 NR ✓ ✓ ✓ ✓ ✕ ✓ ✕ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✕
Cardoso 25 NR ✓ ✓ ✓ ✓ ✕ ✕ ✓ ✓ ✓ ✓ ✕ ✓ ✕ ✓ ✕
Freitas 120 8.7% Between 3.5% and 10.6%a ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓
De Olivera 15.3 ±13.8 NR ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✓ ✓ ✕ ✕ ✓ ✕ ✓ ✕
Heringer 31.4 (range 13 to 54) NR ✕ ✕ ✕ ✓ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Silva 12 NR ✓ ✕ ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕
Studies that included patients with heart failure in functional class I or II or asymptomatic with low LVEF measured
Nunes 60 NR ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✓ ✕
Bestetti NR NR ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Braga 12 12% total ✓ ✕ ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕
Heringer 31.4 NR ✕ ✕ ✕ ✓ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Rassi 24 NR ✓ ✓ ✓ ✓ ✕ ✓ ✕ ✓ ✓ ✕ ✕ ✓ ✕ ✓ ✕
Traina 20.3 22.8 NR ✓ ✕ ✓ ✕ ✕ ✕ ✕ ✓ ✕ ✓ ✕ ✕ ✕ ✕ ✕
Conceisao Souza (2) 12 NR ✓ ✓ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✓ ✓ ✕ ✓ ✕
Cruz 13.4 NR  ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Pereira Nunes 39.5 0.7% ✓ ✓ ✓ ✕ ✕ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓
Asymptomatic patients without demonstrated impaired ejection fraction
Pimenta 121 94.9 NR ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✓ ✕
Arribada 48 25.8% 44.0% ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Lauria Pires 120  15.6%y 15.5%b 0% ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕
Lima Costa A 120 6% total ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕
Lima Costa B 104.6 NR ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✕
Mota 86.4 94.8 NR ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✓ ✕
Maguire 84 0.5% 4.3% ✓ ✓ ✓ ✕ ✓ ✓ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✓ ✕
Puigbo 48 NR ✕ ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Coura 120 9.6% 16.9% ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✓
Pereira 72 33.6% 31.4% ✕ ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✕ ✕ ✕ ✕ ✕ ✕ ✕
Espinoza 72 113 0% 9.09% ✕ ✕ ✕ ✕ ✕ ✓ ✓ ✓ ✕ ✕ ✕ ✓ ✕ ✕ ✕

aIdiopathic miocardiopathy 4.8%, hypertensive 10.6%, ischaemic 6.6% and other aetiologies 3.5%.

bPatients with positive serology with and without treatment.

Of the 14 studies that included patients with heart failure, 9 explicitly described their patient inclusion and exclusion criteria, and a total of 8 described their method of selecting patients. None of the studies used a probabilistic sampling strategy, although six studies described sampling consecutive patients. Moreover, only 5 of 11 studies with patients without heart failure or impaired ventricular function described inclusion and exclusion criteria and the process used for their selection. Four studies were based on community censuses, and one study was based on consecutive patients (Table 3).

A majority of studies (16 out of 25) used two different techniques to measure Chagas exposure, whereas two used one technique, and seven did not state how they carried out their diagnosis. In most of the studies, the events were correctly assessed, and only one study reported them wrongly because they included deaths together with subjects lost to follow-up43 (Table 3).

Time-to-event analysis was used in a majority of studies (14 out of 25), with articles presenting survival curves or tests such as Cox regression. Only a small number of studies presented discussion of how patient attrition was managed in the analysis or whether missing data were imputed,2,3 and only two studies described a sample size calculation (Table 3).

Statistical adjustment through a multivariable analysis was carried out in five of six studies of patients with advanced heart failure and in three of nine studies of patients with heart failure in earlier stages. The adjustment variables were mainly clinical, except one study that included the educational level of the subjects.35 Among patients without heart failure, only the studies of Lima Costa et al.44,45 carried out a multivariable analysis, and these adjusted for sex, age, educational level and the presence of other risk factors (Table 3).

## Discussion

The analysis of studies that compare groups with positive and negative serology for Chagas shows that seropositive patients experience an elevated risk of death, mainly when considering patients with advanced stage heart failure. However, the evidence is not conclusive when considering patients with early stage heart failure, and even less in subjects without evident heart disease.45 In addition, the risk of experiencing events such as stroke and/or development or hospitalization due to heart failure was evaluated among seropositive patients in only very few studies.6,37,40,44

This systematic review reveals that few studies of subjects without heart failure included appropriate confounding factors through multivariable analysis. This is particularly problematic, as Chagas disease is strongly related to poverty. To estimate the real risk associated with the infection, the assessment of the effect of multiple confounding factors is needed because those may increase the risk of cardiomyopathy in vulnerable groups. Only the studies by Lima Costa et al.44,45 that included elderly subjects, adjusted for socio-economic status through level of education.

The association between presenting positive serology for Chagas and the risk of cardiovascular events seems to be more conclusive in patients with heart failure in functional class III or IV. Indeed, a positive serology for Chagas was associated with an increase in mortality within this sub-group in most of the studies included in this review. A number of factors associated with Chagas disease may explain this effect, including a higher frequency of ventricular arrhythmias, as well as the risk associated with right ventricular dysfunction, extensive myocardial damage, dysfunction of the autonomous nervous system and social deprivation.31,32

These results suggest the need for well-designed observational studies of subjects without heart failure. Such studies should include a long-term follow-up and a sample size large enough to enable detection of not only the risk of death but also the risk of developing other cardiovascular events in subjects with positive serology compared with those not exposed to Chagas disease. In this context, the Prospective Urban Rural Epidemiology study52 is carrying out a 10-year follow-up of urban and rural populations with and without positive serology for Chagas; this will enable researchers not only to assess the risk of the different end points checked for this review but also to make the necessary adjustment for confounding factors associated with poverty.

A lack of evidence of the risk for cardiovascular events in subjects without heart failure has implications both for the population with positive serology and from a public health perspective. It may diminish the unfair stigmatization of seropositive patients as ‘Chagasic’. It is currently a common practice in many endemic countries to ask for serology tests as pre-employment medical examinations, an obvious discrimination of seropositive patients.53 Our findings increase new challenges for public health, including the need to determine the independent risk of cardiovascular events associated with chronic Chagas infection, to establish, in a rigorous manner, the true burden of this disease. Such knowledge is needed to develop evidence-based public health policies.

Future research should seek to identify the poverty-related factors that increase the risk of cardiovascular events among seropositive patients. Identification of those factors may be crucial to the development of strategies to prevent disease progression.

The results of this review should be considered cautiously, taking into account the quality of the studies, particularly that of studies that included subjects without heart failure that were mostly performed with limited methodology. Many studies in this literature used an insufficient sample size and lacked a sample size calculation. Many studies also experienced high rates of attrition and failed to adequately discuss it. Perhaps most importantly, almost all of the studies in this review failed to include confounding factors, which would have allowed the establishment of the independent effect of the positive serology in the risk of cardiovascular events.

## Conclusion

This review demonstrates a higher risk of death in patients with advanced heart failure and positive serology for Chagas. However, the results should be considered cautiously, taking into account the quality of the studies. The evidence of a higher mortality in subjects without cardiomyopathy is weak, and in relation to other events such as stroke or heart failure development or hospitalizations, the evidence is null.

Well-designed observational studies of subjects with positive serology for Chagas are needed to determine the risk of cardiovascular events associated with this disease, after adjusting for the multiple confounding factors associated with poverty.

## Acknowledgements

Dr Sergio Sosa Estani, Dr Daniel Ferrante and Dr Agustin Ciapponi.

Funding support: None declared.

Disclosures: None declared.

KEY MESSAGES

• This is the first systematic review to assess the evidence linking positive serology for Chagas with cardiovascular events.

• Positive serology for Chagas is associated with a higher risk of death for patients with heart failure. However, there is little evidence to link positive serology for Chagas with cardiovascular events in asymptomatic subjects.

• Little research exists that addresses the confounding factors related to both poverty and Chagas seropositivity.

• Further epidemiological studies are needed that include an appropriate methodology to determine the true independent risk associated with positive serology for Chagas. These studies should include not only death but also other cardiovascular end points such as development of heart failure symptoms, ventricular dysfunction and stroke.

## References

1
Estimación cuantitativa de la enfermedad de Chagas en las Américas

Montevideo, Uruguay: Organización Panamericana de la Salud. Washington, DC: Organización Panamericana de la Salud, 2006
2
Reporte sobre la enfermedad de Chagas
,
2007
Organización Mundial de la Salud
3
Schmunis
GA
Epidemiology of Chagas disease in non-endemic countries: the role of international migration
Mem Inst Oswaldo Cruz
,
2007
, vol.
102

Suppl 1
(pg.
75
-
85
)
4
Elizari
MV
Chagasic myocardiopathy: historical perspective [Spanish]
Medicina (B Aires)
,
1999
, vol.
59

Suppl 2
(pg.
25
-
40
)
5
Dias
E
Laranja
FS
Miranda
A
Nobrega
G
Chagas’ disease; a clinical, epidemiologic, and pathologic study
Circulation
,
1956
, vol.
14
(pg.
1035
-
60
)
6
Espinosa
R
Carrasco
HA
Belandria
F
, et al.  .
Life expectancy analysis in patients with Chagas' disease: prognosis after one decade (1973–1983)
Int J Cardiol
,
1985
, vol.
8
(pg.
45
-
56
)
7
Rosenbaum
MB
Cerisola
JA
Epidemiology of Chagas' disease in the Argentine Republic [Spanish]
Hospital (Rio J).
,
1961
, vol.
60
(pg.
55
-
100
)
8
Rassi
A
Jr
Rassi
A
Little
WC
Chagas' heart disease
Clin Cardiol
,
2000
, vol.
23
(pg.
883
-
89
)
9
He
J
Ogden
LG
Bazzano
LA
Vupputuri
S
Loria
C
Whelton
PK
Risk factors for congestive heart failure in US men and women: NHANES I epidemiologic follow-up study
Arch Intern Med
,
2001
, vol.
161
(pg.
996
-
1002
)
10
McAlister
FA
Murphy
NF
Simpson
CR
, et al.  .
Influence of socioeconomic deprivation on the primary care burden and treatment of patients with a diagnosis of heart failure in general practice in Scotland: population based study
BMJ
,
2004
, vol.
328
pg.
1110

11
Schaufelberger
M
Rosengren
A
Heart failure in different occupational classes in Sweden
Eur Heart J
,
2007
, vol.
28
(pg.
212
-
18
)
12
Loucks
EB
Lynch
JW
Pilote
L
, et al.  .
Life-course socioeconomic position and incidence of coronary heart disease: the Framingham Offspring Study
Am J Epidemiol
,
2009
, vol.
169
(pg.
829
-
36
)
13
Salomaa
V
Niemela
M
Miettinen
H
, et al.  .
Relationship of socioeconomic status to the incidence and prehospital, 28-day, and 1-year mortality rates of acute coronary events in the FINMONICA myocardial infarction register study
Circulation
,
2000
, vol.
101
(pg.
1913
-
18
)
14
Kotchen
JM
Shakoor-Abdullah
B
Walker
WE
Chelius
TH
Hoffmann
RG
Kotchen
TA
Am J Public Health
,
1998
, vol.
88
(pg.
1696
-
99
)
15
Moreira
GC
Cipullo
JP
Martin
JF
, et al.  .
Evaluation of the awareness, control and cost-effectiveness of hypertension treatment in a Brazilian city: populational study
J Hypertens
,
2009
, vol.
27
(pg.
1900
-
07
)
16
Thomsen
RW
Johnsen
SP
Olesen
AV
, et al.  .
Socioeconomic gradient in use of statins among Danish patients: population-based cross-sectional study
Br J Clin Pharmacol
,
2005
, vol.
60
(pg.
534
-
42
)
17
Brown
AF
Gross
AG
Gutierrez
PR
Jiang
L
Shapiro
MF
Mangione
CM
Income-related differences in the use of evidence-based therapies in older persons with diabetes mellitus in for-profit managed care
J Am Geriatr Soc
,
2003
, vol.
51
(pg.
665
-
70
)
18
Ciapponi
A
Systematic review of the link between tobacco and poverty
,
2011
Geneva
World Health Organization Institute for Clinical Effectiveness and Health Policy
19
Droomers
M
Schrijvers
CT
Stronks
K
van de Mheen
D
Mackenbach
JP
Educational differences in excessive alcohol consumption: the role of psychosocial and material stressors
Prev Med
,
1999
, vol.
29
(pg.
1
-
10
)
20
van Oers
JA
Bongers
IM
van de Goor
LA
Garretsen
HF
Alcohol consumption, alcohol-related problems, problem drinking, and socioeconomic status
Alcohol Alcohol
,
1999
, vol.
34
(pg.
78
-
88
)
21
Rassi
A
Jr
Rassi
A
Rassi
SG
Predictors of mortality in chronic Chagas disease: a systematic review of observational studies
Circulation
,
2007
, vol.
115
(pg.
1101
-
08
)
22
Wilczynski
NL
Haynes
RB
Developing optimal search strategies for detecting clinically sound prognostic studies in MEDLINE: an analytic survey
BMC Med
,
2004
, vol.
2
pg.
23

23
Altman
DG
Systematic reviews of evaluations of prognostic variables
BMJ
,
2001
, vol.
323
(pg.
224
-
28
)
24
Sanderson
S
Tatt
ID
Higgins
JP
Tools for assessing quality and susceptibility to bias in observational studies in epidemiology: a systematic review and annotated bibliography
Int J Epidemiol
,
2007
, vol.
36
(pg.
666
-
76
)
25
World Health Organization

Control of Chagas Disease, Second report of the WHO Expert Committee. Geneva: WHO, 2002
26
Pereira
JB
da Cunha
RV
Willcox
HP
Coura
JR
Development of chronic human chagas cardiopathy in the hinterland of the Paraiba State, Brazil, in a 4.5 year period [Portuguese]
Rev Soc Bras Med Trop
,
1990
, vol.
23
(pg.
141
-
47
)
27
de Campos Lopes
CB
AT
Araujo
F
Pereira Barreto
AC
Mansur
AJ
Socioeconomic factors in the prognosis of heart failure in a Brazilian cohort
Int J Cardiol
,
2006
, vol.
113
(pg.
181
-
87
)
28
Cardoso
J
Novaes
M
Ochiai
M
, et al.  .
Chagas cardiomyopathy: prognosis in clinical and hemodynamic profile C
Arq Bras Cardiol
,
2010
, vol.
95
(pg.
518
-
23
)
29
Oliveira
MT
Jr
Canesin
MF
Munhoz
RT
, et al.  .
Major clinical characteristics of patients surviving 24 months or more after hospitalization due to decompensated heart failure [Portuguese]
Arq Bras Cardiol
,
2005
, vol.
84
(pg.
161
-
66
)
30
Heringer-Walther
S
Moreira Mda
C
Wessel
N
, et al.  .
Does the C-type natriuretic peptide have prognostic value in chagas disease and other dilated cardiomyopathies?
J Cardiovasc Pharmacol
,
2006
, vol.
48
(pg.
293
-
98
)
31
Silva
CP
Del Carlo
CH
Oliveira Junior
MT
, et al.  .
Por que os portadores de cardiomiopatia chagásica têm pior evolução que os não-chagásicos?
Arq Bras Cardiol
,
2008
, vol.
91
(pg.
389
-
94
)
32
Freitas
HFG
Chizzola
PR
Paes
AT
Lima
ACP
Mansur
AJ
Risk stratification in a Brazilian hospital-based cohort of 1220 outpatients with heart failure: role of Chagas' heart disease
Int J Cardiol
,
2005
, vol.
102
(pg.
239
-
47
)
33
Nunes
VL
Ramires
FJA
Pimentel
WdS
, et al.  .
O
papel do acúmulo de colágeno no interstício miocárdico na sobrevida dos pacientes com cardiomiopatia dilatada idiopática e chagásica [Portuguese]
Arq Bras Cardiol
,
2006
, vol.
87
(pg.
757
-
62
)
34
Bestetti
RB
Muccillo
G
Clinical course of Chagas' heart disease: a comparison with dilated cardiomyopathy
Int J Cardiol
,
1997
, vol.
60
(pg.
187
-
93
)
35
Braga
JC
Reis
F
Aras
R
Jr
, et al.  .
Is Chagas cardiomyopathy an independent risk factor for patients with heart failure?
Int J Cardiol
,
2008
, vol.
126
(pg.
276
-
78
)
36
Rassi
S
Barretto
AC
Porto
CC
, et al.  .
Survival and prognostic factors in systolic heart failure with recent symptom onset [Portuguese]
Arq Bras Cardiol
,
2005
, vol.
84
(pg.
309
-
13
)
37
Souza
GECS
Pessoa-De-Melo
LMJV
Valim
LR
, et al.  .
Prognosis of Chagas' cardiomyopathy versus non-Chagas' cardiomyopathy in beta-blockers era: 12 month follow-up
Eur J Heart Fail (Supplement)
,
2009
, vol.
8
(pg.
S126
-
27
)
38
Cruz
FDC
Issa
VS
Ayub
SAM
Bacal
FB
Bocchi
EB
Components of life quality evaluation in heart failure clinic
Eur J Heart Fail (Supplement).
,
2010
, vol.
9
pg.
S173

39
Pereira Nunes Mdo
C
Barbosa
MM
Ribeiro
AL
Amorim Fenelon
LM
Rocha
MO
Predictors of mortality in patients with dilated cardiomyopathy: relevance of chagas disease as an etiological factor
Rev Esp Cardiol
,
2010
, vol.
63
(pg.
788
-
97
)
40
Traina
M
Hernandez
S
Garcia
EP
, et al.  .
Journal of Cardiac Failure Conference: 14th Annual Scientific Meeting Heart Failure Society of America San Diego

CA United States Conference Start, 2010;16(Suppl 8)
41
A
Apt
W
Ugarte
JM
Evolución de la cardiopatía chagásica durante un período de cuatro años en un grupo de pacientes chilenos
Bol Oficina Sanit Panam
,
1987
, vol.
102
(pg.
49
-
73
)
42
Pimenta
J
Valente
N
Miranda
M
Long-term follow up of asymptomatic chagasic individuals with intraventricular conduction disturbances, correlating with non-chagasic patients
Rev Soc Bras Med Trop
,
1999
, vol.
32
(pg.
621
-
31
)
43
Lauria-Pires
L
Braga
MS
Vexenat
AC
, et al.  .
Progressive chronic Chagas heart disease ten years after treatment with anti-Trypanosoma cruzi nitroderivatives
Am J Trop Med Hyg
,
2000
, vol.
63
(pg.
111
-
18
)
44
Lima-Costa
MF
Matos
DL
Ribeiro
AL
Chagas disease predicts 10-year stroke mortality in community-dwelling elderly. The Bambui cohort study of aging
Stroke
,
2010
, vol.
41
(pg.
2477
-
82
)
45
Lima-Costa
MF
Peixoto
SV
Ribeiro
AL
Chagas disease and mortality in old age as an emerging issue: 10 year follow-up of the Bambui population-based cohort study (Brazil)
Int J Cardiol
,
2010
, vol.
145
(pg.
362
-
63
)
46
Maguire
JH
Hoff
R
Sherlock
I
, et al.  .
Cardiac morbidity and mortality due to Chagas' disease: prospective electrocardiographic study of a Brazilian community
Circulation
,
1987
, vol.
75
(pg.
1140
-
45
)
47
Mota
EA
Guimaraes
AC
Santana
OO
, et al.  .
A nine year prospective study of Chagas' disease in a defined rural population in northeast Brazil
Am J Trop Med Hyg
,
1990
, vol.
42
(pg.
429
-
40
)
48
Pereira
JB
Willcox
HP
Coura
JR
Morbidity in Chagas' disease. III. Longitudinal study of 6 years, in Virgem da Lapa, MG, Brazil [Portuguese]
Mem Inst Oswaldo Cruz
,
1985
, vol.
80
(pg.
63
-
71
)
49
Puigbo
JJ
Nava Rhode
JR
Carcia Barrios
H
Gil Yepez
C
A 4-year follow-up study of a rural community with endemic Chagas' disease
Bull World Health Organ
,
1968
, vol.
39
(pg.
341
-
48
)
50
Coura
JR
de Abreu
LL
Pereira
JB
Willcox
HP
Morbidity in Chagas' disease. IV. Longitudinal study of 10 years in Pains and Iguatama, Minas Gerais, Brazil [Portuguese]
Mem Inst Oswaldo Cruz
,
1985
, vol.
80
(pg.
73
-
80
)
51
Conceicao-Souza
GE
Vieira
JL
Moreira
LFP
, et al.  .
Evaluation of ventricular remodeling as prognostic predictor in advanced heart failure: comparison of chagasic and non-chagasic aetiology
Eur J Heart Fail (Supplement)
,
2010
, vol.
9
pg.
S15

52
Teo
K
Chow
CK
Vaz
M
Rangarajan
S
Yusuf
S
The prospective urban rural epidemiology (PURE) study: examining the impact of societal influences on chronic noncommunicable diseases in low-, middle-, and high-income countries
Am Heart J
,
2009
, vol.
158
(pg.
1
-
7
e1
53
Ianni
BM
Arteaga
E
Frimm
CC
Pereira Barretto
AC