Galectin-3 Associated with Severe Forms and Long-term Mortality in Patients with Chagas Disease

Abstract

Background

The histopathological characteristics of Chagas disease (ChD) are: presence of myocarditis, destruction of heart fibers, and myocardial fibrosis. Galectin-3 (Gal-3) is a biomarker involved in the mechanism of fibrosis and inflammation that may be useful for risk stratification of individuals with ChD.

Objectives

We sought to evaluate whether high Gal-3 levels are associated with severe forms of Chagas cardiomyopathy (CC) and whether they are predictive of mortality.

Methods

We studied anti-T. cruzi positive blood donors (BD): Non-CC-BD (187 BD without CC with normal electrocardiogram [ECG] and left ventricular ejection fraction [LVEF]); CC-Non-Dys-BD (46 BD with CC with abnormal ECG but normal LVEF); and 153 matched serum-negative controls. This cohort was composed of 97 patients with severe CC (CC-Dys). We used Kruskall-Wallis and Spearman’s correlation to test hypothesis of associations, assuming a two-tailed p<0.05 as significant.

Results

The Gal-3 level was 12.3 ng/mL for Non-CC-BD, 12.0 ng/mL for CC-Non-Dys-BD, 13.8 ng/mL for controls, and 15.4 ng/mL for CC-Dys. LVEF<50 was associated with higher Gal-3 levels (p=0.0001). In our linear regression adjusted model, we found association between Gal-3 levels and echocardiogram parameters in T. cruzi-seropositive subjects. In CC-Dys patients, we found a significant association of higher Gal-3 levels (≥15.3 ng/mL) and subsequent death or heart transplantation in a 5-year follow-up (Hazard ratio – HR 3.11; 95%CI 1.21–8.04; p=0.019).

Conclusions

In ChD patients, higher Gal-3 levels were significantly associated with severe forms of the disease and more long-term mortality, which means it may be a useful means to identify high-risk patients. (Arq Bras Cardiol. 2021; 116(2):248-256)

Chagas Disease; Chagas, Cardiomyopathy; Mortality; Galectin-3; Biomarkers; Electrocardiography/methods; Heart Failure

Resumo

Fundamento

As características histopatológicas da doença de Chagas (DCC) são: presença de miocardite, destruição das fibras cardíacas e fibrose miocárdica. A Galectina-3 (Gal-3) é um biomarcador envolvido no mecanismo de fibrose e inflamação que pode ser útil para a estratificação de indivíduos com DCC por risco.

Objetivos

Nosso objetivo foi avaliar se níveis elevados de Gal-3 estão associados a formas graves de cardiomiopatia chagásica (CC) e são preditivos de mortalidade.

Métodos

Estudamos doadores de sangue (DS) positivos para anti-T. cruzi: não-CC-DS (187 DS sem CC com eletrocardiograma [ECG] e fração de ejeção do ventrículo esquerdo [FEVE] normais); CC-Não-Dis-DS (46 DS com CC e apresentando ECG anormal, mas FEVE normal); e 153 controles negativos correspondentes. Esta amostra foi composta por 97 pacientes com CC grave (CC-Dis). Usamos as correlações de Kruskall-Wallis e Spearman para testar a hipótese de associações, assumindo um p bicaudal <0,05 como significativo.

Resultados

O nível de Gal-3 foi de 12,3 ng/mL para não-CC-DS, 12,0 ng/mL para CC-Não-Dis-DS, 13,8 ng/mL para controles e 15,4 ng/mL para CC-Dis. FEVE <50 foi associada a níveis mais elevados de Gal-3 (p=0,0001). Em nosso modelo de regressão linear ajustado, encontramos associação entre os níveis de Gal-3 e os parâmetros do ecocardiograma em indivíduos positivos para T. cruzi. Nos pacientes CC-Dis, encontramos uma associação significativa de níveis mais elevados de Gal-3 (≥15,3 ng/mL) e morte ou transplante cardíaco em acompanhamento de cinco anos (Hazard ratio – HR 3,11; IC95% 1,21– 8,04; p=0,019).

Conclusões

Em pacientes com CC, níveis mais elevados de Gal-3 estiveram significativamente associados a formas graves da doença e maior taxa de mortalidade em longo prazo, o que significa que pode ser um meio efetivo para identificar pacientes de alto risco. (Arq Bras Cardiol. 2021; 116(2):248-256)

Doença de Chagas; Cardiomiopatia Chagásica; Mortalidade; Galectina 3; Biomarcadores; Eletrocardiografia/métodos; Insuficiência Cardíaca

Introduction

Chagas cardiomyopathy (CC), one of the leading causes of heart disease and death in Latin America, has a poor prognosis compared to noninflammatory cardiomyopathies.¹

The natural history of Chagas disease (ChD) involves an acute phase, followed by the chronic phase. It is still unknown, however, which patients are more likely to progress to severe forms. Direct parasite injury, inflammation triggered by the immune system, and autonomic dysfunction are role-players in the pathogenesis of CC. When the cardiac tissue is injured, replacement fibrosis appears to be a cause of structural disorganization, geometry, and functional heart impairment.²

Galectin-3 (Gal-3) is secreted by activated macrophages and is involved in the fibrogenesis of heart failure (HF). This biomarker has recently been linked to development of HF and mortality. In an experimental model of ChD, Gal-3 promoted cell infiltration in the heart and fibrosis.³,⁴

The lack of a good marker of active infection or incipient CC makes the development of new treatments in this population a challenge. The use of biomarkers that can accurately predict clinical outcomes in CC would have the potential to guide therapy, by identifying patients at higher risk and who would need an earlier, more intensive, and personalized strategy.⁵

The aim of our study was to evaluate whether high Gal-3 levels are associated with severe forms of CC and whether it is a predictive factor for subsequent mortality or the need for heart transplantation.

Methods

Study Design

Samples were collected during the Retrovirus Epidemiology Donor Study-II (REDS-II),⁶ a retrospective cohort study in which T. cruzi-seropositive blood donors (BD) were identified by screening blood banks in 1996-2002 (233 from the city of Sao Paulo), in addition to 153 serum-negative control BDs (matched by year of donation, age and gender). This cohort of BD was composed of 97 previously diagnosed cases of CC from the Heart Institute (INCOR) of the Medical School of Universidade de São Paulo. From July 2008 to October 2010, recruited individuals (BD and CC patients) filled in health questionnaires and went through medical evaluations, including electrocardiogram (ECG), echocardiogram (ECHO), and phlebotomy with processing and cryopreservation of samples for subsequent batched blinded analyses of cardiac markers, polymerase chain reaction (PCR) for detection of T. cruzi, and other biomarkers (see below).

First, we performed a cross-sectional study in which T. cruzi-seropositive participants filled in a questionnaire and received a medical evaluation (laboratory, ECG, and ECHO parameters), resulting in groups stratified by CC status. Then, a longitudinal cohort study was carried out, in which 97 T. cruzi-seropositive patients suffering from chronic and more severe CC forms were followed up in an outpatient service from INCOR, in which time-to-event data was available.

All blood samples were collected in EDTA and serum tubes, processed for parasite detection, or submitted to spinning and divided in aliquots. All specimens were frozen in Brazil at -20°C until shipped to the REDS-II Central Laboratory (Blood Systems Research Institute, San Francisco, CA) on dry ice and maintained at -70°C.

Four groups were created: a control serum-negative group and three T. cruzi seropositive groups: BD without cardiomyopathy, presenting with normal ECG and left ventricular ejection fraction (LVEF) (Non-CC-BD); BD with CC, presenting with ECG abnormalities but normal systolic function (CC-Non-Dys-BD); and participants with CC and left ventricular dysfunction (CC-Dys). The dysfunction was defined as LVEF <50% upon ECHO.

Data regarding time-to-event analysis were available only for CC-Dys subjects. From July to September 2015, we proceeded with CC-Dys patients’ chart analysis and phone calls to monitor events and get respective dates.

The local ethics committees approved the study protocol, and all participants signed an informed consent form.

PCR Procedures

From each participant, we collected 20 mL of EDTA-anti-coagulated blood that were immediately mixed with an equal volume of 6 M guanidine hydrochloride -0.2M EDTA solution. The target-capture real-time PCR assay used in this study was developed based on the PCR method described by Virreira et al.,⁷ targeting the kinetoplast T. cruzi mitochondrial minicircle DNA. The DNA extraction was improved with a target-capture step using magnetic beads coated with a T. cruzi-specific 20-mer capture oligonucleotide.

Cardiac Measurements

Resting 12-lead ECGs were recorded (General Electric MAC 1200 electrocardiograph; GE Healthcare, Waukesha, WI).⁸,⁹

Echocardiograms were performed using a Sequoia^TM 512 ultrasound machine (Acuson, Mountain View, CA), per guidelines by the American Society of Echocardiography.¹⁰,¹¹ Studies were recorded in digital format, and all measurements were made on digital loops using a Digisonics offline analysis station (version 3.2 software; Digisonics, Houston, TX) at the Cardiovascular Branch, Echocardiography Laboratory, of the National Heart, Lung, and Blood Institute (Bethesda, MD).¹²

Soluble Biomarkers

Biomarkers associated with T. cruzi infection have been previously described.¹² Blinded plasma samples were tested using Milliplex kits (Millipore) with antibody-coated beads for detection of tumor necrosis factor alpha (TNF-α), interleukin (IL) 6 (IL-6), IL-8, IL-10, and interferon gamma (IFNγ). Standard curves and samples were tested in duplicate. Results were acquired on a Labscan 200 analyzer (Luminex) using the Bio-Plex manager software v6.1 (Bio-Rad). IFNγ was predominantly below the threshold of detection (57%).

Concentrations of N-terminal pro B-type natriuretic peptides (NT-proBNPs) and troponin were measured using U.S. Food and Drug Administration-cleared assays on the VITROS System (Ortho Clinical Diagnostics, Raritan, NJ).

Plasma Gal-3 levels were determined by an enzyme-linked fluorescence assay and measured on a BioMerieux Vidas 30 system (BioMerieux, Marcy l’Etoile, Lyon, France), following the manufacturer’s recommendations.

Statistical Analyses

Normality was tested with the Shapiro-Wilk test. Continuous non-normally distributed variables were: age, Gal-3 levels, ejection fraction, and other cardiac and inflammatory biomarkers, being expressed as median and interquartile range. Differences between groups as to variables were compared using the Kruskal-Wallis test, while chi-squared or Fisher’s test and logistic regression were used to assess variable type and distribution. Use of ranks in one-criterion variance analysis and post-hoc analysis was made with Dunn’s test to evaluate median-value differences between the groups with Bonferroni adjustment for multiple comparisons. For analysis of correlations, the Spearman’s correlation was used, reporting p values.

Receiver operating characteristic (ROC) curves were performed for Gal-3 and NT-proBNP to optimize the definition of the cutoff points that would best discriminates the event at follow-up, and an area under the curve (AUC) were identified. Both curves were compared with DeLong and chi-squared tests.

For a time-to-event analysis, the CC-Dys group was divided in two profiles regarding Gal-3 and NT-proBNP cutoff values: low Gal-3 or NT-proBNP (≤ cutoff) and high Gal-3 or NT-proBNP (≥ cutoff). Analysis of incidence of cumulative events across Gal-3 and NT-proBNP strata and the additive value of Gal-3 relative to NT-proBNP was made by a Kaplan-Meier-like method followed by log-rank test.

Bivariate and multivariate Cox proportional-hazards regression models were constructed to evaluate the association of Gal-3 and NT-proBNP values (below versus above or equal to the cutoff value) with incident events. Models were adjusted for sex, age, serum creatinine, New York Heart Association (NYHA) classification and LVEF; 95% confidence intervals (CI) were used to depict the association of each marker and the events in the final fitted Cox proportional-hazards model. Sensitivity was analyzed using Gal-3 and NT-proBNP as continuous variables. A two-tailed p<0.05 was considered as significant.

All graphs and statistical analyses were made in the software Stata (version 13.0, Stata Corp., College Station, TX).

Results

Of the original 570 participants in REDS-II, 483 had samples available for Gal-3 testing; 153 were anti-T. cruzi serum-negative BD, and the remainder were sorted into three groups: 187 Non-CC-BD presenting with normal ECG and LVEF; 46 CC-non-Dys; and 97 CC-Dys (Figure 1).

Figure 1
– Flowchart of inclusion.

Patient Clinical and Biomarker Characteristics

Demographic and clinical characteristics are described in Table 1. Gal-3 levels were greater in patients from the CC-Dys group, than other clinical groups. Higher Gal-3 levels were also seen in controls when compared with Non-CC-BD. We did not observe significant differences in Gal-3 levels between Non-CC-BD and CC-non-Dys-BD, or between CC-non-Dys-BD and controls. Inflammatory markers (TNF-α, IL-6, IL-8, IL-10), as well as biomarkers associated with, cardiac dysfunction or damage (NT-proBNP and troponin) were elevated in CC-Dys patients compared to other groups (Table 1).

Table 1
– Clinical, laboratory, and echocardiographic findings

	Non-CC-BD	CC-Non-Dys-BD	CC-Dys	Controls
	n=187 (38.6%)	n=46 (9.5%)	n=97 (20.2%)	n=153 (31.6%)	p-value
Age (y)	49 [41-58]	50 [44-59]	48.5 [43-54]	48 [42-55]	0.20
Male, n (%)	110 (58.8)	29 (63)	59 (60.2)	89 (58.2)	0.39
Ejection fraction (%)	63 [60-65]	60 [55-65]	30 [20-40]	64 [60-65]	<0.001*
Cardiac biomarkers
Galectin-3 (ng/mL)	12.3 [10-15.4]	12.0 [9.5-14.9]	15.4 [11.8-19.8]	13.8 [11.2-16.2]	<0.001†
NT-proBNP (pg/mL)	40.6 [23.6-66.8]	59 [35.0-109.1]	748 [379.20-2223.41]	27.5 [19.3-48.1]	<0.001‡
Troponin	0.012 [0.01-0.012]	0.012 [0.012-0.015]	0.021 [0.12-0.03]	0.012 [0.012-0.012]	<0.001§
Inflammatory markers
TNF-α	2.94 [1.64-4.59]	3.02 [1.25-4.69]	3.65 [2.57-5.52]	2.84 [1.62-3.91]	0.002#
IL-6	0.69 [0.32-1.63]	0.77 [0.32-1.8]	1.60 [0.64-3.13]	1.14 [0.32-1.7]	<0.001 **
IL-8	1.61 [0.95-2.79]	1.53 [0.99-2.5]	2.23 [1.38-3.2]	1.44 [0.95-2.54]	0.003††
IL-10	1.28 [0.32-4]	2.02 [0.32-4.11]	4.37 [1.62-8.06]	1.22 [0.32-3.35]	<0.001‡‡
IFN-γ	0.32 [0.32-0.64]	0.32 [0.32-0.84]	0.32 [0.32-1.07]	0.32 [0.32-0.39]	0.06
Parasite load
Parasite estimate per 20mL	0.05 [0-2.5]	0.68 [0.03-5.47]	1.77 [0.16-5]	-	<0.001§§

• p-values were reported for Kruskal-Wallis and Dunn post-hoc hypothesis testing. Median [interquartile range] reported for all biomarkers tested. Non-CC-BD, blood donors without Chagas cardiomyopathy; CC-Non-Dys-BD, blood donors with Chagas cardiomyopathy; CC-Dys, Chagas cardiomyopathy patients with cardiac dysfunction; NT-proBNP, N-terminal pro B-type natriuretic peptide; TNF, tumor necrosis factor; IL, interleukin; IFN, interferon. *Statistically significant difference in ejection fraction levels between Non-CC BD/CC-Dys (p<0.001); CC-Non-Dys-BD/CC-Dys (p<0.001); CC-Non-Dys-BD/controls (p=0.042); CC-Dys/controls (p<0.001). †Statistically significant difference in Gal-3 levels between Non-CC-BD/CC-Dys (p<0.001); Non-CC-BD/controls (p=0.010); CC-Non-Dys-BD/CC-Dys (p<0.001); CC-Dys/controls (p=0.028). ‡Statistically significant difference in NT-proBNP levels between Non-CC-BD/CC-Dys (p<0.001); Non-CC-BD/controls (p=0.004); CC-Non-Dys-BD/CC-Dys (p<0.001); CC-Non-Dys-BD/controls (p<0.001); CC-Dys/controls (p<0.001). §Statistically significant difference in troponin levels between Non-CC-BD/CC-Dys (p=0.024); Non-CC-BD/CC-controls (p<0.001); CC-Non-Dys-BD/CC-Dys (p<0.001); CC-Non-Dys-BD/controls (<0.001); CC-Dys/controls (p<0.001). #Statistically significant difference in TNF-α levels between Non-CC-BD/CC-Dys (p=0.019); CC-Dys/controls (p<0.001). **Statistically significant difference in IL-6 levels between Non-CC-BD/CC-Dys (p<0.001); CC-Non-Dys-BD/CC-Dys (p=0.032); CC-Dys/controls (p=0.004). ††Statistically significant difference in IL-8 levels between Non-CC-BD/CC-Dys (p=0.016); CC-Non-Dys-BD/CC-Dys (p=0.039); CC-Dys/controls (p=0.001). ‡‡Statistically significant difference in IL-10 levels between Non-CC-BD/CC-Dys (p<0.001); CC-Non-Dys-BD/CC-Dys (p=0.001); CC-Dys/controls (p<0.001). §§Statistically significant difference in parasite estimate per 20mL between Non-CC-BD/CC-Dys (p=0.011); Non-CC-BD/controls (p<0.001).

Positive T. cruzi PCR indicated a statistically significant difference between Non-CC-BD and CC groups: CC-Non-Dys-BD (p=0.010) and CC-Dys. By contrast, no difference in parasitemia was observed when comparing CC-Non-Dys-BD and CC-Dys (Table 1). However, we did not finda

significant difference in CC-Dys patients between T. cruzi PCR and event occurrence. No significant association between T. cruzi PCR and Gal-3 was found.

Among the T. cruzi infected subjects, Spearman’s correlation was applied to assess the relationship between Gal-3 and cardiac biomarkers, inflammatory mediators, and parasite load. There was a weak correlation for TNF- (r_s=0.25, p<0.001) and IL-8 (r_s=0.22, p<0.001). By contrast, no association between Gal-3 and Troponin, NT-proBNP, IL-6, IL-10, IFN-, or with parasite load was found.

Echocardiography was carried out, and we used Spearman’s correlation, to verify any relationship between Gal-3 levels and echocardiographic parameters among T. cruzi infected patients. No moderate or strong statistically significant correlation was found between Gal-3 levels and left ventricular end-diastolic diameter (LVEDD) (rs=0.09, p=0.07), left ventricular end-systolic dimension (LVESD) (rs=0.11, p=0.03), LVEF (r_s=-0.16, p=0.001), left atrial diameter (r_s=0.11, p=0.02), left atrial volume indexed to body surface area (r_s=0.09, p=0.18), right atrial area (r_s=0.032, p=0.53) and septal E/e’ ratio (r_s=0.135, p=0.009).

Survival and Risk Analysis

Time-to-event data were available for 97 patients, with mean follow-up of 51.2 ± 10.8 months and median of 58 months (range: 8 to 60 months). Events were observed in 28 patients (29%), and were due to three (10.8%) heart transplantations and 25 (89.2%) deaths by all causes. Among event-experienced patients, median concentrations of Gal-3 and NT-pro-BNP were significantly higher, while the ejection fraction was significantly lower. Age, sex, NYHA class >I, ejection fraction on echocardiography, and laboratory data of event-experienced patients are compared in Table 2.

Table 2
– Demographic and laboratory information of CC-Dys patients according to death or heart transplantation

	No events	Events	p-value
	(n=69)	(n=28)
Age (y)	49 [42-54]	47.5 [44.5-52]	0.96
Male, n (%)	42 (61%)	16 (57%)	0.73
Creatinine (mg/dL)	1.01 [0.85-1.14]	1.14 [0.87-1.23]	0.06
Ejection fraction (%)	35 [25-40]	20 [20-30]	0.001
NYHA >1	29	20	0.009
Galectin-3 (ng/mL)	14.4 [10.9-19.1]	18.5 [14.7-23.4]	0.005
Low	39	7
High	30	21
NT-proBNP (pg/mL)	542 [281-1337]	2643 [1047-4771]	<0.001
Low	50	7
High	19	21
Parasite estimate*	1.77 [0.19-4.2]	1.25 [0.16-12.61]	0.08

• ^thth

Gal-3 cutoff point (<15.3 ng/mL) by ROC curve was used to divide CC-Dys subjects into two strata (low and high levels), as was the NT-proBNP cutoff point (<1278 pg/mL). ROC identified the potential to reach an event. Although the AUC of NT-proBNP was larger than the Gal-3 AUC, there were no differences (p=0.22).

After adjusting for sex, age, renal function, NYHA functional class >I, and LVEF, we found a significant association of higher levels of Gal-3 with subsequent events in a 5-year follow-up (Table 3 and Figure 2). Complementarily, the risk of events also increased as the levels of NT-proBNP climbed (Table 3). Similar results were seen when Gal-3 and NT-proBNP (by 100-unit increase) were analyzed as continuous variables (Appendix).

Table 3
– Association between galectin-3 and death or heart transplantation in the CC-Dys subgroup using ROC and Cox regression models, both crude and adjusted for age, sex, serum creatinine level, NYHA functional class >I and LVEF, and using Galectin-3 and NT-proBNP as categorical variables

				Crude Model		Adjusted Model
	AUC	Cutoff	Cutoff Level	p	HR (95%CI)	p	HR (95%CI)
Galectin-3 (ng/mL)	0.71	15.3	Low
			High	0.007	3.27 (1.39-7.71)	0.04	2.63 (1.00-6.90)
NT-proBNP (pg/mL)	0.80	1278	Low
			High	<0.001	5.69 (2.41-13.42)	0.02	3.44 (1.21-9.78)

• AUC: area under the curve; CI: confidence interval; HR: hazard ratio; LVEF: left ventricular ejection fraction; NT-proBNP: N-terminal pro B-type natriuretic peptide; NYHA: New York Heart Association; ROC: receiver operating characteristic

Figure 2
– Time-to-event curves between Gal-3 levels. Time-to-event curves between stratified Gal-3 levels results in CC-Dys. Cutoff level: Gal-3 <15.3 ng/mL.

Among patients presenting higher Gal-3 levels, we found differences in events when dichotomized with both NT-proBNP strata: patients with additionally high NT-proBNP levels were more likely to experience any event than patients with low NT-proBNP. Moreover, patients in the lower Gal-3 strata, when dichotomized with NT-proBNP levels, were more likely to present any event when NT-proBNP was higher (Figure 3).

Figure 3
– Event frequency at 5-year as a function of Gal-3 and NT-proBNP concentrations among CC patients. The percentage of patients who experienced any event is shown for each group. Low Gal-3: values <15.3 ng/mL; high Gal-3: values ≥15.3 ng/mL; low NT-proBNP: values <1,278 pg/mL; high NT-proBNP: values ≥1,278 pg/mL.

Patients in the higher strata of both Gal-3 and NT-proBNP levels had 11 to 16-fold increased risk of event compared with those with the lowest biomarker levels (unadjusted HR, 16.22; 95%CI: 3.71–70.83; p<0.001; adjusted HR, 11.39; 95%CI: 1.97–65.76; p=0.007). Subjects with low Gal-3 and low NT-proBNP had the lowest event rates.

Discussion

Increased Gal-3 was significantly associated with severe forms of ChD and predictive of subsequent morbidity/mortality.

Gal-3 is an emerging biomarker and modulates several physiological processes that contribute to HF, inflammation, and fibrosis.¹³-¹⁵ Inflammation is a prerequisite for tissue healing and scar formation,¹⁶ and Gal-3 has been shown to play a major role as a mediator in parasitic, viral,¹⁴,¹⁷ and bacterial infection.¹⁸ In ChD, experimental studies have shown that Gal-3 has upregulated expression following T. cruzi infection in dendritic cells, B cells,¹⁹,²⁰ and CD68+ macrophages. Significantly, CD68+ macrophages represent 50% of the mononuclear cell infiltrate in hearts with CC.²¹

T. cruzi and several immune-mediated mechanisms have a direct involvement in CC. Previous studies have reported that Gal-3 binds to 45KD, enhancing its adhesion to the extracellular matrix and even its entry into cells. Other studies have shown the importance of Gal-3 in the early process of T. cruzi infection, as it allows parasites to accumulate in the extracellular matrix before invading host cells.²²,²³

An experimental model of acute T. cruzi infection showed that induction of myocarditis was associated with the upregulation of Col I, Gal-3, IFN-, and IL-13.²¹ Gal-3 was primarily detected in interstitial cells and was higher in fibrotic areas. In myocardial areas of fibrosis, the intensity of myocarditis and significant matrix extracellular remodeling was correlated with the presence of Col I and Gal-3. In addition, myofibroblasts can induce fibrosis, which results in myocardial stiffness and cardiac dysfunction. Importantly, myofibroblasts are also a significant source of proinflammatory cytokines, including TNF-α and IL-1, which have a known deleterious effect on the myocardium. However, we did not find any significant association between Gal-3 and inflammatory markers.

Sabino et al.²⁴ compared the detection of T. cruzi DNA with known clinical and laboratory markers of CC severity and observed that the presence of parasitemia was associated with markers of disease progression, such as QRS and QT interval duration, lower LVEF, and elevated troponin and NT-proBNP levels. It was also observed that detection of T. cruzi DNA was significantly higher in patients with cardiomyopathy as compared to Non-CC-BD group; however, T. cruzi PCR did not correlate with Gal-3. Moreover, there was no significant difference in the detection of T. cruzi DNA between CC patients with and without dysfunction, nor between CC patients who did or did not experience events. Thus, in our study, parasitism was a marker of typical ECG changes in cardiomyopathy, but not of disease severity or clinical prognosis.

De Boer et al.²⁵ suggested that Gal-3 likely represents a unique phenotype at high risk for the development and progression of HF or other cardiovascular diseases. Chronic elevations in Gal-3 induce active fibrogenesis and may provoke pathological cardiac remodeling. De Boer et al.²⁵ also hypothesized that patients with this phenotype of Gal-3 overexpression are more likely to have a “fibrogenic” pathway for cardiac remodeling. In our study, high levels of Gal-3 were associated with the most severe form of cardiomyopathy, but without a strong association with echocardiographic parameters. So, Gal-3 levels defined a population with more severe disease, characterized by left systolic and diastolic ventricular dysfunction, higher left and right diastolic diameter, and elevated of NT-proBNP and troponin levels.

Echeverria et al.²⁶ examined the diagnostic value of a panel of biomarkers to distinguish the severity of CC and found no associations between sST2 and Gal-3 levels. However, the sample size was small and did not include patients with stage A (positive T. cruzi, but normal ECG and echocardiography), which could have allowed the examination of the role played by the biomarkers in asymptomatic patients. They also do not provide any prognostic information of Gal-3.

We found higher Gal-3 levels in the control group compared to the Non-CC-BD group. However, the values were lower compared to the CC-Dys group. It is known, since Carlos Chagas’ pioneering studies,²⁷ that up to 60% of infected patients have no evidence suggesting cardiovascular or gastrointestinal involvement. These individuals are thought to have the so-called indeterminate form, defined as Non-CC-BD in our study. As a result, survival in this group of patients appears to be comparable to the general population. Our results showed low Gal-3 levels in this group, which supports this concept.

Galectin-3 phenotype is an important factor in the onset and progression of HF. It is known that HF patients with low Gal-3 levels have slow progression and better outcomes than patients with HF and high Gal-3 levels.²⁵,²⁸ Gal-3 was shown to predict the development of all-cause mortality and HF in the general population²⁸ and can be used to define and identify patients with HF at very low risk for 30-day and 180-day mortality, and HF rehospitalizations after an episode of acute HF.²⁹ A meta-analysis by Chen et al.³⁰ reported the value of serum Gal-3 as a predictor factor of all-cause mortality and cardiovascular mortality in HF patients.³⁰

Our study’s most striking finding was the relationship between Gal-3 and the risk of events among patients with CC. Because both Gal-3 and NT-proBNP were independent predictors of adverse events, we also showed that the increase of both markers was associated with the highest rates of death or heart transplantation in patients with CC.

Study Limitations

This was a single-center study with a relatively small sample. In addition, we had only a single-time point measure of Gal-3 and NT-proBNP and, therefore, did not assess dynamic changes in these biomarkers over time. Another limitation was the use of non-parametric tests to analyze associations between continuous variables, resulting in loss of efficiency.

Conclusions

High plasma Gal-3 levels were significantly associated with cardiac dysfunction and CC severity. Our findings suggest that a biomarker-based approach for risk stratification in ChD patients might help physicians identify patients who are more likely to have worse outcomes and potentially guide the development of treatment strategies for this high-risk group.

Acknowledgment

We thank Biomerieux for donating Galectin-3 kits.

Referências

• ¹
  Nadruz W, Gioli-Pereira L, Bernardez-Pereira S, Marcondes-Braga FG, Fernandes-Silva MM, Silvestre OM, et al. Temporal trends in the contribution of Chagas cardiomyopathy to mortality among patients with heart failure. Heart. 2018;104(18):1522-8.
• ²
  Cunha-Neto E, Coelho V, Guilherme L, Fiorelli A, Stolf N, Kalil J. Autoimmunity in Chagas’ disease. Identification of cardiac myosin-B13 Trypanosoma cruzi protein crossreactive T cell clones in heart lesions of a chronic Chagas’ cardiomyopathy patient. J Clin Invest. 1996;98(8):1709-12.
• ³
  Filipe MD, Meijers WC, Rogier van der Velde A, de Boer RA. Galectin-3 and heart failure: prognosis, prediction & clinical utility. Clin Chim Acta. 2015;443:48-56.
• ⁴
  Pineda MA, Cuervo H, Fresno M, Soto M, Bonay P. Lack of galectin-3 prevents cardiac fibrosis and effective immune responses in a murine model of Trypanosoma cruzi infection. J Infect Dis. 2015;212(7):1160-71.
• ⁵
  Pecoul B, Batista C, Stobbaerts E, Ribeiro I, Vilasanjuan R, Gascon J, et al. The BENEFIT Trial: Where Do We Go from Here? PLoS Negl Trop Dis. 2016;10(2):e0004343.
• ⁶
  Sabino EC, Ribeiro AL, Salemi VM, Di Lorenzo Oliveira C, Antunes AP, Menezes MM, et al. National Heart, Lung, and Blood Institute Retrovirus Epidemiology Donor Study-II (REDS-II), International Component. Ten-year incidence of Chagas cardiomyopathy among asymptomatic Trypanosoma cruzi-seropositive former blood donors. Circulation. 2013;127:1105-15.
• ⁷
  Virreira M, Torrico F, Truyens C, Alonso-Vega C, Solano M, Carlier Y, et al. Comparison of polymerase chain reaction methods for reliable and easy detection of congenital Trypanosoma cruzi infection. Am J Trop Med Hyg. 2003;68(5):574-82.
• ⁸
  Prineas RJ CR, Blackburn H. The Minnesota Code Manual of Electrocardiographic Findings. Littleton, MA: John Wright-PSG; 1982.
• ⁹
  Ribeiro AL, Sabino EC, Marcolino MS, Salemi VM, Ianni BM, Fernandes F, et al. Electrocardiographic abnormalities in Trypanosoma cruzi seropositive and seronegative former blood donors. PLoS Negl Trop Dis. 2013;7(2):e2078.
• ¹⁰
  Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr. 2005;18(12):1440-63.
• ¹¹
  Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, Smiseth OA, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. Eur J Echocardiogr. 2009;10(2):165-93.
• ¹²
  Keating SM, Deng X, Fernandes F, Cunha-Neto E, Ribeiro AL, Adesina B, et al. Inflammatory and cardiac biomarkers are differentially expressed in clinical stages of Chagas disease. Int J Cardiol. 2015;199:451-9.
• ¹³
  Liu FT, Rabinovich GA. Galectins: regulators of acute and chronic inflammation. Ann N Y Acad Sci. 2010;1183:158-82.
• ¹⁴
  Jaquenod De Giusti C, Ure AE, Rivadeneyra L, Schattner M, Gomez RM. Macrophages and galectin 3 play critical roles in CVB3-induced murine acute myocarditis and chronic fibrosis. J Mol Cell Cardiol. 2015;85:58-70.
• ¹⁵
  Ma F, Li Y, Jia L, Han Y, Cheng J, Li H, et al. Macrophage-stimulated cardiac fibroblast production of IL-6 is essential for TGF/Smad activation and cardiac fibrosis induced by angiotensin II. PLoS One. 2012;7(5):e35144.
• ¹⁶
  de Boer RA, Daniels LB, Maisel AS, Januzzi JL Jr. State of the art: Newer biomarkers in heart failure. Eur J Heart Fail. 2015;17(6):559-69.
• ¹⁷
  Sato S, Ouellet M, St-Pierre C, Tremblay MJ. Glycans, galectins, and HIV-1 infection. Ann N Y Acad Sci. 2012;1253:133-48.
• ¹⁸
  Mishra BB, Li Q, Steichen AL, Binstock BJ, Metzger DW, Teale JM, et al. Galectin-3 functions as an alarmin: pathogenic role for sepsis development in murine respiratory tularemia. PLoS One. 2013;8(3):e59616.
• ¹⁹
  Acosta-Rodríguez EV, Montes CL, Motrán CC, Zuniga EI, Liu FT, Rabinovich GA, et al. Galectin-3 mediates IL-4-induced survival and differentiation of B cells: functional cross-talk and implications during Trypanosoma cruzi infection. J Immunol. 2004;172(1):493-502.
• ²⁰
  Vray B, Camby I, Vercruysse V, Mijatovic T, Bovin NV, Ricciardi-Castagnoli P, et al. Up-regulation of galectin-3 and its ligands by Trypanosoma cruzi infection with modulation of adhesion and migration of murine dendritic cells. Glycobiology. 2004;14(7):647-57.
• ²¹
  Ferrer MF, Pascuale CA, Gomez RM, Leguizamón MS. DTU I isolates of Trypanosoma cruzi induce upregulation of Galectin-3 in murine myocarditis and fibrosis. Parasitology. 2014;141(6):849-58.
• ²²
  Nde PN, Lima MF, Johnson CA, Pratap S, Villalta F. Regulation and use of the extracellular matrix by Trypanosoma cruzi during early infection. Front Immunol. 2012;3:337.
• ²³
  Reignault LC, Barrias ES, Soares Medeiros LC, de Souza W, de Carvalho TM. Structures containing galectin-3 are recruited to the parasitophorous vacuole containing Trypanosoma cruzi in mouse peritoneal macrophages. Parasitol Res. 2014;113(6):2323-33.
• ²⁴
  Sabino EC, Ribeiro AL, Lee TH, Oliveira CL, Carneiro-Proietti AB, Antunes AP, et al. Detection of Trypanosoma cruzi DNA in blood by PCR is associated with Chagas cardiomyopathy and disease severity. Eur J Heart Fail. 2015;17(4):416-23.
• ²⁵
  de Boer RA, van der Velde AR, Mueller C, van Veldhuisen DJ, Anker SD, Peacock WF, et al. Galectin-3: a modifiable risk factor in heart failure. Cardiovasc Drugs Ther. 2014;28(3):237-46.
• ²⁶
  Echeverría LE, Rojas LZ, Calvo LS, Roa ZM, Rueda-Ochoa OL, Morillo CA, et al. Profiles of cardiovascular biomarkers according to severity stages of Chagas cardiomyopathy. Int J Cardiol. 2017;227:577-82.
• ²⁷
  Lewinsohn R. Carlos Chagas (1879-1934): the discovery of Trypanosoma cruzi and of American trypanosomiasis (foot-notes to the history of Chagas’s disease). Trans R Soc Trop Med Hyg. 1979;73(5):513-23.
• ²⁸
  de Boer RA, van Veldhuisen DJ, Gansevoort RT, Muller Kobold AC, van Gilst WH, Hillege HL, et al. The fibrosis marker galectin-3 and outcome in the general population. J Intern Med. 2012;272(1):55-64.
• ²⁹
  Meijers WC, de Boer RA, van Veldhuisen DJ, Jaarsma T, Hillege HL, Maisel AS, et al., Biomarkers and low risk in heart failure. Data from COACH and TRIUMPH. Eur J Heart Fail. 2015;17(12):1271-82.
• ³⁰
  Chen A, Hou W, Zhang Y, Chen Y, He B. Prognostic value of serum galectin-3 in patients with heart failure: a meta-analysis. Int J Cardiol. 2015;182:168-70.

• Study Association
  This study is not associated with any thesis or dissertation work.
• *Supplemental Materials
  For additional information, please click here.

• Sources of Funding.There were no external funding sources for this study.

Publication Dates

• Publication in this collection
  01 Mar 2021
• Date of issue
  Feb 2021

History

• Received
  02 July 2019
• Reviewed
  29 Nov 2019
• Accepted
  27 Dec 2019