The persistence of inflammatory processes in the myocardium in varying degrees of chronic Chagas heart disease has been poorly investigated. We hypothesized that edema could occur in patients with chronic chagasic cardiomyopathy and corresponds to the persistence of inflammatory processes in the myocardium. Eighty-two Chagas disease (CD) seropositive patients (64.6% females; age = 58.9 ± 9.9) without ischemic heart disease or conditions that cause myocardial fibrosis and dilation were considered. Late gadolinium enhancement (LGE) and T2-weighted magnetic resonance imaging of edema were obtained and represented using a 17-segment model. Patients were divided into three clinical groups according to the left ventricular (LV) ejection fraction (EF) as G1 (EF > 60%;
Chagasic heart disease, caused by
After infection with
Recently, cardiac magnetic resonance (CMR) has been considered a sensitive technique to detect myocardial damage in Chagas heart disease [
This study was performed with 82 patients in different clinical phases of Chagas heart disease. Patients were divided into groups according to the left ventricular (LV) ejection fraction (EF) as G1 (EF > 60%;
The exclusion criteria considered refusal of the patient or their relatives to participate in the study diagnosis of coronary disease; dilated cardiomyopathy of ischemic origin and/or significant valvular disease; contraindications to magnetic resonance; presence of tattoos, metallic pigments, nonremovable metal prostheses, metallic sutures and/or heart devices not compatible with magnetic resonance; pregnancy; fever; severe claustrophobia; severe psychiatric disorders; creatinine <30 mL/minute in a 24-hour depuration test; and contraindication to gadolinium as a contrast medium.
Patients were recruited from the cardiology, electrophysiology, emergency, and hospitalization services of seven cardiovascular reference hospitals of the city of Bogotá DC, Colombia.
CMR was performed in all selected patients using a Philips Ingenia 1.5 T scanner. The exploration protocol comprised an initial morphological evaluation considering bright and black-blood gradient-echo sequences in the axial and coronal planes. The functional evaluation considered maps of T1 and T2 sequences of cine resonance with the technique of steady-state free precession (SSFP). Additionally, a quantitative functional evaluation was carried out. Structural measures (diastolic and systolic diameter of the left ventricle, diastolic thickness of the walls, anterior systolic left atrial diameter, volume of the left atrium in two dimensions, mean lateral diameter of the right ventricle, and presence of pericardial effusion) and functional measures (ejection fraction of the left ventricle, estimation of pulmonary arterial pressure, and presence and degrees of mitral and tricuspid regurgitation) were obtained from each patient. Tissue characterization evaluated the presence of myocardial edema by means of T2 sequences with fat suppression and the presence of LGE with T1 IR acquisitions in 2D and 3D. Additionally, images of LGE were obtained specifying the location and degree of transmurality. The intravenous contrast material used was Gadobutrol (0.1 mmol/kg), a gadolinium-based MRI contrast agent.
The interpretation of the resonances was carried out by specialists in radiology and diagnostic images of the participating institutions. A second reading of the study was made by an expert in advanced cardiac imaging with experience in cardiovascular magnetic resonance analysis; the specialists were blinded with respect to the previous analysis, in order to estimate the reproducibility of the imaging variables obtained in the study.
Comparisons of normally distributed continuous variables between groups were performed by the one-way analysis of variance (ANOVA) with the Bonferroni post hoc test for multiple comparisons. The nonparametric test for discrete variables and nonnormal continuous variables was the Kruskal–Wallis test by ranks. Normality was determined by the Shapiro–Wilk test. The Fisher exact test was used for proportion comparisons between the groups. A confidence level of 95% was used to considered test results statistically significant.
Multiple correspondence analysis (MCA), a multivariate data analysis, was employed to construct relationships among categorical variables associated with the presence and location of LGE in the different LVEF groups. Chi-squared tests were performed to verify the association between categorical variables. All statistical analyses were performed using the R language [
Table
Clinical characteristics of Chagas heart disease patients classified by groups.
Total |
Group 1FE > 60% |
Group 260% < FE > 35% |
Group 3FE < 35% |
| |
---|---|---|---|---|---|
Age (years) | 58.9 (SD 9.9) | 57.6 (SD 11.0) | 58.8 (SD 9.1) | 63.0 (SD 7.7) | 0.13 |
BMI | 26.2 (SD 4.3) | 26.7 (SD 3.9) | 26.0 (SD 4.3) | 25.6 (SD 5.7) | 0.38 |
Socioeconomic | 2.0 (SD 1.0) | 2.25 (SD 1.0) | 1.42 (SD 0.99) | 1.4 (SD 0.7) | 0.02 |
Female (%) | 53 (64.6) | 30 (81.0) | 18 (54.5) | 5 (41.7) | 0.013 |
Blood pressure (%) | 39 (47.6) | 18 (50.0) | 12 (40.0) | 9 (75.0) | 0.12 |
Type 2 diabetes (%) | 7 (8.9) | 4 (11.1) | 2 (6.7) | 1 (8.3) | 0.82 |
DLP (%) | 12 (15.4) | 8 (22.2) | 2 (6.7) | 2 (16.7) | 0.22 |
Smoking (%) | 3 (3.8) | 0 (0) | 2 (6.7) | 1 (8.33) | 0.25 |
NYHA | 1.5 (SD 0.5) | 1.4 (SD 0.5) | 1.4 (SD 0.6) | 1.7 (SD 0.4) | 0.19 |
|
|||||
|
|||||
LGE (%) | 48 (58.5) | 16 (43.2) | 21 (63.6) | 11 (91.7) | 0.01❖ |
FEVD | 54.2 (SD 9.8) | 58.6 (SD 5.5) | 55.0 (SD 6.8) | 38.2 (SD 11.6) | 0.01❖ |
LV telediastolic volume | 153.2 (SD 60.5) | 123.0 (SD 32.1) | 149.0 (SD 39.7) | 256 (SD 67.2) | 0.01❖ |
LV telediastolic volume index | 87.1 (SD 34.7) | 70.9 (SD 16.9) | 82.5 (SD 24.7) | 150.6 (SD 29.8) | 0.01❖ |
LV telesystolic volume | 78.5 (SD 60.1) | 43.5 (SD 14.1) | 74.5 (SD 33.0) | 197.0 (SD 57.0) | 0.01❖ |
LV telesystolic volume index | 44.2 (SD 35.8) | 24.3 (SD 7.6) | 41.0 (SD 20.1) | 114.0 (SD 26.5) | 0.01❖ |
LV mass | 105.4 (SD 40.6) | 87.7 (SD 19.7) | 98.5 (SD 25.5) | 163.0 (SD 54.7) | 0.01❖ |
LV mass index | 57.9 (SD 21.3) | 49.2 (SD 13.4) | 53.0 (SD 15.3) | 89.7 (SD 20.0) | 0.01❖ |
VD telediastolic volume | 128.1 (SD 42.4) | 122.0 (SD 35.3) | 132.0 (SD 41.9) | 136.0 (SD 62.1) | 0.48 |
VD telediastolic volume index | 72.1 (SD 21.3) | 69.5 (SD 18.1) | 73.1 (SD 24.1) | 77.4 (SD 23.0) | 0.51 |
VD telesystolic volume | 58.9 (SD 29.2) | 49.8 (SD 16.1) | 60.2 (SD 27.0) | 83.9 (SD 48.4) | 0.01❖ |
VD telesystolic volume index | 33.9 (SD 15.1) | 28.5 (SD 8.3) | 34.8 (SD 16.0) | 48.2 (SD 19.5) | 0.01❖ |
|
|||||
|
|||||
Transmural (%) | 15 (31.2) | 3 (18.7) | 8 (38.1) | 4 (41.6) | 0.41 |
Subendocardic (%) | 27 (56.2) | 7 (43.7) | 10 (47.6) | 10 (90.9) | 0.03❖ |
Subepicardic (%) | 26 (54.2) | 7 (43.7) | 14 (66.7) | 5 (45.4) | 0.31 |
Midwall (%) | 41 (85.4) | 12 (75.0) | 20 (95.2) | 9 (81.8) | 0.21 |
Basal segments |
35 (72.9) | 10 (62.5) | 18 (85.7) | 7 (63.6) | 0.21 |
Apical segments |
21 (43.7) | 6 (37.5) | 7 (33.3) | 8 (72.7) | 0.08 |
Midcavity segments |
18 (37.5) | 4 (25.0) | 7 (33.3) | 7 (63.7) | 0.11 |
|
|||||
|
|||||
Creatinine | 0.9 (SD 0.2) | 0.84 (SD 0.2) | 0.89 (SD 0.2) | 1.1 (SD 0.3) | 0.00❖ |
BUN | 17.6 (SD 7.9) | 16.2 (SD 5.6) | 18.5 (SD 8.3) | 19.4 (SD 11.6) | 0.34 |
Continuous variables are expressed as mean (SD, standard deviation). Categorical variables are expressed as the number of patients (%). BMI: body mass index; NYHA FC: New York Heart Association functional class. Post hoc comparisons between groups were not statistically significant otherwise indicated ⦿: significant difference between G2 and G1;
LGE was observed in 48 (58.5%) patients; 16 (43.2%) of G1, 21 (63.6%) of G2, and 11 (91.7%) of G3 (
Results of the (a) LGE and (b) edema distribution in a 17-segment heart model. LAD: left anterior descending; RCA: right coronary artery; LCX: left circumflex. Modified from
(a) LVEF of different LGE distributions and (b) scatterplot of LVEF and RVEF between groups (
Our confirmation of myocardial fibrosis in Chagas-seropositive patients, in different stages of disease severity, coincides with previous studies that reported myocardial fibrosis in patients within the indeterminate form [
We found that patients with LGE had a trend toward worse NYHA functional class compared to patients without LGE, although LGE was found in 25 (52.1%) patients with NYHA functional class I and 22 (45.8%) patients with NYHA functional class II, and it was found in a patient with NYHA functional class III. LVEF and CMR-derived LV volumes showed high negative correlation with echocardiographic measurements for LV telediastolic volume (
Although usually associated with reduced LVEF, we found that RVEF was significantly lower in G3 patients. This finding is in accordance with recent studies which suggest that RV systolic dysfunction in Chagas disease depends mainly on LV increased afterload [
Edema was observed in 8 (9.8%) patients, 2 (5.4%) of G1, 4 (12.1%) of G2, and 2 (16.7%) of G3. It was observed mainly at the basal inferolateral segment in 7 (87.5%) cases. Edema distribution is represented in a 17-segment model in Figure
Clinical characteristics of Chagas heart disease in patients with cardiac edema.
Edema( |
Nonedema( |
| |
---|---|---|---|
Age | 53.8 (SD 12.9) | 59.4 (SD 9.5) | 0.12 |
BMI | 27.9 (SD 4.3) | 26.1 (SD 4.3) | 0.25 |
Socioeconomic | 1.75 (SD 0.9) | 2.07 (SD 1.0) | 0.39 |
Female (%) | 6 (75.0) | 47 (63.5) | 0.79 |
Blood pressure (%) | 3 (37.5) | 36 (51.4) | 0.71 |
Type 2 diabetes (%) | 1 (12.5) | 11 (15.7) | 0.99 |
DLP (%) | 10 (16.9) | 6 (24.0) | 0.29 |
Smoking (%) | 1 (12.5) | 2 (2.9) | 0.71 |
NYHA | 1.8 (SD 0.5) | 1.4 (SD 0.5) | 0.12 |
|
|||
|
|||
FEVI | 45.6 (SD 15.4) | 54.6 (SD 14.6) | 0.10 |
LGE (%) | 8 (100) | 40 (54.1) | 0.03 |
FEVD | 51.1 (SD 5.9) | 54.5 (SD 10.2) | 0.36 |
LV telediastolic volume | 178.0 (SD 78.3) | 151.0 (SD 58.3) | 0.23 |
LV telediastolic volume index | 104.0 (SD 51.6) | 85.3 (SD 32.3) | 0.15 |
LV telesystolic volume | 118.0 (SD 79.4) | 74.2 (SD 56.7) | 0.04 |
LV telesystolic volume index | 69.6 (SD 50.1) | 41.4 (SD 31.6) | 0.03 |
VD telediastolic volume | 142.0 (SD 23.7) | 127.0 (SD 43.9) | 0.34 |
VD telediastolic volume index | 81.0 (SD 10.4) | 71.1 (SD 22.0) | 0.21 |
VD telesystolic volume | 66 (SD 17.2) | 58.2 (SD 30.2) | 0.47 |
VD telesystolic volume index | 39.0 (SD 7.6) | 33.4 (SD 15.6) | 0.31 |
|
|||
|
|||
Transmural (%) | 3 (37.5) | 12 (16.2) | 0.32 |
Midwall (%) | 7 (87.5) | 24 (32.4) | 0.01 |
Subendocardic (%) | 2 (25.0) | 15 (20.3) | 0.99 |
Subepicardic (%) | 7 (87.5) | 12 (16.2) | 0.00 |
Basal (%) | 7 (87.5) | 27 (67.5) | 0.47 |
Septal (%) | 2 (25.0) | 8 (20.0) | 0.99 |
Mesial (%) | 1 (12.5) | 16 (40.0) | 0.28 |
Apical (%) | 4 (50.0) | 17 (42.5) | 0.99 |
|
|||
|
|||
Creatinine | 0.9 (SD 0.2) | 0.9 (SD 0.2) | 0.93 |
BUN | 20.5 (SD 6.9) | 17.2 (SD 8.1) | 0.40 |
Continuous variables are expressed as mean (SD, standard deviation). Categorical variables are expressed as number of patients(%). ∗Significant difference between cardiac edema and nonedema patients.
The presence of edema in the chronic phase of the disease was previously related with severe impairment of myocardial status [
In Figure
Multiple correspondence analyses of the variables associated with the presence and location of LGE within LVEF groups. It shows the difference between the variables associated with G3 versus G2 and G1 groups.
Our findings indicate that the concept of indeterminate form of Chagas disease, although valid for the collective prognostic evaluation of this group of patients, should be updated in the analysis of individual cases, considering the findings obtained by the application of more accurate diagnostic methods, such as CMR [
The aim of this study was to characterize CMR findings caused by Chagas at different stages of disease severity in Chagas-seropositive patients and solve the hypothesis that edema could occur in patients with chronic chagasic cardiomyopathy and corresponds to the persistence of the inflammatory process in the myocardium. We found that edema can be found in patients with chagasic cardiomyopathy in the chronic stage. In later stages of cardiac dilation with low LVEF, the LGE pattern involves subendocardium and mid locations. Deteriorations of RV and LV are positively correlated without evidence of fibrosis in the RV.
The data used to support the findings of this study may be released upon application to the Fundación Universitaria de Ciencias de la Salud (FUCS), Bogotá, Colombia, who can be contacted at
The authors declare that there are no conflicts of interest regarding the publication of this paper.
Special thanks go to Tania Sánchez Vargas and the fellows: Dr. Lorena Gonzalez Russi, Dr. Juan Sebastián Salcedo, Dr. Camilo Calvache, and Dr. Leonor Mariño. The authors thank the financial support of the Departamento Administrativo de Ciencia, Tecnología e Innovación-COLCIENCIAS, Colombia (Convocatoria 744/2016).