Myocardial scars correlate with eletrocardiographic changes in chronic Trypanosoma cruzi infection for dogs treated with Benznidazole

Authors


Corresponding Author Ivo Santana Caldas, Núcleo de Pesquisas em Ciências Biológicas, Universidade Federal de Ouro Preto, Morro do Cruzeiro, campus universitário, Ouro Preto, 35400-000, Minas Gerais, Brazil. E-mail: ivocaldas@nupeb.ufop.br

Abstract

Objectives  The cardiac form of Chagas disease is evidenced by a progressive cardiac inflammation that leads to myocarditis, fibrosis and electrocardiographic (ECG) conduction abnormalities. Considering these characteristics, the aim of this study was to prospectively evaluate the early ECG changes in dogs that were experimentally inoculated with Benznidazole (Bz)-susceptibly (Berenice-78) and Bz-resistant (VL-10, and AAS) Trypanosoma cruzi strains and, later, evaluate the efficacy of Bz treatment for preventing these ECG alterations.

Methods  Electrocardiographic changes of treated and untreated animals were prospectively evaluated for up to 270 days after infection, at which point collagen (right atrium) quantification was performed.

Results  All infected dogs had a high intensity of heart fibrosis (4616.00 ± 1715.82 collagen/74931 μm2 in dogs infected with Berenice-78 strain, 5839.2 ± 1423.49 collagen/74931 μm2 in infected by AAS and 6294.40 ± 896.04 collagen/74931 μm2 in animals infected with VL-10 strain), while 78.57% of all infected dogs showed ECG alterations. Bz Therapy reduced or prevented fibrosis in Bz-susceptible Berenice-78 (2813.00 ± 607.13 collagen/74931 μm2) and Bz-resistant AAS strains (4024 ± 1272.44 collagen/74931 μm2), coincident with only 10% de ECG alterations at 270 days. However, in those animals infected with a Bz-resistant VL-10 strain, specific treatment did not alter collagen deposition (6749.5 ± 1596.35 collagen/74931 μm2) and there was first atrioventricular block and chamber overload at 120 and 270 days after infection, with 75% abnormal ECG exams.

Conclusions  These findings indicate that an effective antiparasitic treatment in the early stage of Chagas disease can lead to a significant reduction in the frequency and severity of the parasite-induced cardiac disease, even if parasites are not completely eliminated.

Abstract

Objectifs

La forme cardiaque de la maladie de Chagas se manifeste par une inflammation cardiaque progressive qui conduit à des myocardites, à la fibrose et à des anomalies de conductions électrocardiographiques (ECG). Compte tenu de ces caractéristiques, le but de cette étude était d’évaluer prospectivement les premiers changements ECG chez des chiens qui ont été inoculés expérimentalement avec des souches de T. cruzi sensibles au Benznidazole (Bz) (Berenice-78) et résistantes au Bz (VL-10 et AAS) et, plus tard, d’évaluer l'efficacité du traitement au Bz pour prévenir ces altérations de l’ECG.

Méthodes

les changements dans l’ECG des d'animaux traités et non traités ont été évalués de façon prospective sur un maximum de 270 jours après l'infection, date à laquelle la quantification du collagène (oreillette droite) a été réalisée.

Résultats

Tous les chiens infectés avaient une forte intensité de fibrose cardiaque (4616,00 ± 1715,82 collagène/74931 μm2 chez les chiens infectés par la souche Berenice-78, 5839,2±1423,49 collagène/74931 μm2 chez ceux infectées par la souche AAS et 6294,40±896,04 collagène/74931 μm2 chez ceux infectés avec la souche VL-10), alors que 78,57% de tous les chiens infectés ont montré des altérations ECG. La thérapie au Bz réduit ou empêche la fibrose dans les infections avec les souches Berenice-78 sensibles au Bz (2813,00±607,13 collagène/74931 μm2) et avec les souches AAS résistantes au Bz (4024±1272,44 collagène/74931 μm2), coïncidant avec seulement 10% d'altérations de l’ECG à 270 jours. Cependant, chez les animaux infectés par une souche VL-10 résistante au Bz, un traitement spécifique ne modifie pas le dépôt de collagène (6749,5±1596,35 collagène/74931 μm2) et il y avait un premier bloc auriculo-ventriculaire et une surcharge de la chambre à 120 et 270 jours après l'infection, avec 75% d'examens ECG anormaux.

Conclusions

Ces résultats indiquent qu'un traitement efficace contre les parasites au stade précoce de la maladie de Chagas peut conduire à une réduction significative de la fréquence et de la gravité de la maladie cardiaque induite par le parasite, même si les parasites ne sont pas totalement éliminés.

Abstract

Objetivos

La afección cardiaca de la enfermedad de Chagas se presenta como una inflamación cardiaca progresiva que conlleva a una miocarditis, fibrosis y anomalías en el electrocardiograma (ECG). Teniendo en cuenta estas características, el objetivo de este estudio era evaluar de forma prospectiva los cambios tempranos en el ECG de perros a los que se les había inoculado, de manera experimental, con cepas de T. cruzi susceptibles al Benznidazol (Bz) (Berenice-78) y resistentes al Bz (VL-10 y AAS), y posteriormente evaluar la eficacia del tratamiento con Bz para prevenir las alteraciones en el ECG.

Métodos

Se evaluaron, de forma prospectiva, los cambios en el ECG de los animales tratados y sin tratar, hasta 270 días después de la infección, punto en el cual se hizo una cuantificación del colágeno (en el atrio derecho).

Resultados

Todos los perros infectados tenían una alta densidad de fibrosis en el corazón (4616.00±1715.82 colágeno/74931 μm2 en perros infectados con la cepa Berenice-78, 5839.2±1423.49 colágeno /74931 μm2 en perros infectados con AAS y 6294.40±896.04 colágeno /74931 μm2 en animales infectados con la cepa VL-10), mientras que un 78.57% de todos los perros infectados mostraban alteraciones en el ECG. La terapia con Bz redujo o previno la fibrosis en cepas susceptibles a Bz, Berenice-78, (2813.00±607.13 colágeno/74931 μm2) y en cepas AAS, resistentes a Bz (4024±1272.44 colágeno /74931 μm2), coincidiendo con solo un 10% de las alteraciones en el ECG a día 270. Sin embargo, en aquellos animales infectados con una cepa VL-10 Bz-resistente, el tratamiento específico no alteró la deposición de colágeno (6749.5±1596.35 colágeno/74931 μm2) y primero hubo un bloqueo atrioventricular y sobrecarga de la cámara en los días 120 y 270 después de la infección, siendo anormales un 75% de los exámenes de ECG.

Conclusiones

Estos hallazgos indican que un tratamiento antiparasitario efectivo, en una etapa temprana de la enfermedad de Chagas, puede conllevar a una reducción significativa en la frecuencia y la severidad de la enfermedad cardiaca inducida por el parásito, inclusive sin una eliminación completa de los parásitos.

Introduction

Chagas disease, induced by Trypanosoma cruzi, is responsible for substantial morbidity, mortality and economic adversity in Latin America. Chagas disease is also an emerging health threat in regions with large South and Central American immigrant populations (Schmunis 2007). Chemotherapy, together with vector control, remains one of the most important elements in its control since currently there are no vaccines to prevent the infection (Urbina 2009; Dujardin et al. 2010).

Cardiomyopathy is the most serious manifestation of Chagas disease and one of the primary causes of heart failure and other clinical manifestations, including cardiac arrhythmias, heart blocks and arterial or venous thromboembolism (Coura 2007). Clinical presentation varies widely according to the degree of myocardial damage and the signs of the cardiomyopathy evolution can be monitored by electrocardiogram (ECG) and echocardiography (Viotti et al. 2004; Biolo et al. 2010). Several studies showed that the prevalence of electrocardiographic abnormalities increases with age with a peak at the fourth to fifth decade of life, while the differential risk of ECG alterations comparing seropositive with seronegative people decreases with age (Maguire et al. 1983). More recently, a high prevalence rate of ECG abnormalities associated with T. cruzi-positive serology among children was shown (Andrade et al. 1998; Salazar-Schettino et al. 2009). The authors suggest the existence of endemic areas with a particular accelerated disease progression, emphasising the importance of longitudinal cardiologic follow-up (Salazar-Schettino et al. 2009) and a health chemotherapy programme that focuses on the treatment of young seropositive children (Andrade et al. 1998).

In clinical trials during the acute phase of the infection, 40–76% rates of parasitological cure have been attained (Andrade et al. 1992; Bahia-Oliveira et al. 2000; Cançado 2002). Long-term studies have also shown significant success for chemotherapy with Benznidazole (Bz) among children and adolescents who have short-term chronic infections (Sosa & Segura 1999; Silveira et al. 2000; Andrade et al. 2004). In different regions of Bolivia, however, the Bz-treatment of children and adolescents induced a cure rate of 0–5.4% (Yun et al. 2009). According to authors, the variability in apparent treatment effectiveness may reflect differences in the parasite population’s lineages in different geographic regions

The aim of this study was to prospectively evaluate the early ECG changes in dogs that were experimentally inoculated with Benznidazole (Bz)-susceptibly (Berenice-78) and Bz-resistant (VL-10, and AAS) T. cruzi strains and, later, evaluate the efficacy of Bz treatment for preventing these ECG changes.

Material and methods

T. cruzi stocks

For the infection, two T. cruzi II stocks (AAS and VL-10) previously classified as drug resistant (Caldas et al. 2008) and one T. cruzi II stock, Berenice-78, classified as drug susceptible (Veloso et al. 2001) were used. The Berenice-78 strain was used as a reference drug-susceptible strain. All T. cruzi stocks used in this study was isolated from human hosts, in the State of Minas Gerais, Brazil (Lana & Chiari 1986; Schlemper et al. 1986; Filardi & Brener 1987).

Drug

Benznidazole (N-benzil-2-nitro-1-imidazoleacetamide) was synthesised by Produtos Roche Q.F.S.A., Rio de Janeiro, Brazil.

Animal infection and Benznidazole treatment

Thirty five mongrel dogs (all 4 months old) obtained from the Animal Facilities at the Ouro Preto Federal University (UFOP), Minas Gerais State, Brazil were used in this study. All procedures and experimental protocols were conducted in accordance with the guidelines issued by the Brazilian College of Animal Experimentation (COBEA) and approved by the Ethics Committee in Animal Research at UFOP. The animals were fed with commercial dog food and water ad libitum. Prior to study, the animals were de-wormed and vaccinated against several infectious diseases. A total of 10 dogs were inoculated with 2.0 × 103 bloodstream trypomastigotes per kg of bodyweight with each T.cruzi strain described previously. Five age-matched uninfected dogs were used as negative controls.

The animals infected with each T. cruzi strain were divided in two experimental groups: (i) five dogs were treated with Bz, at 7.0 mg/kg, suspended in 4% gum arabic, twice a day [b.i.d.; i.e. every 12 h (q12 h)] for 60 days; and (ii) five dogs were maintained as non-treated. An additional five animals were maintained as an uninfected and untreated control group. These treatment schemes were previously described by Guedes et al. (2002). Oral treatment was initiated after parasite detection that occurs around 20 post-infection days.

Assessment of treatment efficacy

The efficacy of treatment was assessed by blood culture and PCR assays performed on the first and sixth month post-treatment. Blood culture assay was performed in treated and control animals as described by Chiari et al. (1989). Blood cultures were examined monthly up to 120 days for parasitic detection.

For the PCR assay, 10 ml of blood from each animal were collected, immediately mixed with an equal volume of 6 m guanidine hydrochloride/0.2 m EDTA solution and maintained at room temperature for 2 weeks. Subsequently, these samples were boiled for 15 min before DNA extraction from 200 μl aliquots taken from each sample (Guedes et al. 2002). DNA amplification was performed in a total volume of 20 μl that contained 0.1% Triton ×−100, 10 mm Tris-HCl (pH 9.0), 75 mm KCl, 5 mm MgCl2, 0.2 mm dATP, 0.2 mm dTTP, 0.2 mm dGTP. 0.2 mm dCTP (Sigma-Aldrich), 1 μl of Taq DNA polymerase (Invitrogen, USA), 20 pmol of S35 (5′ AAATAATGTACGGG(T/G)GAGATGCATGA3′) and S36 (5′ GGGTTCGATTGGGGTTGGTGT3′) primers and 2 μl of DNA from each sample (Ávila et al., 1991). The reaction mixture was subjected to 35 cycles of amplification in an automatic thermocycler (Biocycler). The temperature profile was as follows: denaturation at 95 °C for 1 min (with a longer initial time of 5 min at 95 °C in the 1st amplification cycle), and primer annealing at 65 °C for 1 min with an extension for 1 min at 72 °C, after which there was a final incubation at 72 °C for 10 min to extend the annealed primers. The PCR products were visualised by 6% polyacrylamide gel electrophoresis followed by silver staining. All DNA extraction steps and reaction mixtures used for PCR were monitored and compared to positive and negative controls. The PCR analysis was considered negative after three failed DNA extractions of a given sample.

Electrocardiography

All animals were submitted to electrocardiographic (ECG) exams before infection and on the 4th and 9th months of post-infection. Qualitative ECG variables were verified, blinded and independently, by two cardiologists who were experts in Chagas disease. Animals were injected intraperitoneally with the anaesthetic thiopental sodium (0.03 g per ml of 0.8% saline solution) and placed on an electrically insulated surface where the anesthetised dogs were positioned with the limbs oriented perpendicularly to their body. Electrical tracings were recorded with standard ECG measurements and cardiac vectors were recorded for the limb (aVR, aVL, aVF) and precordial (V10, CV5RL, CV6LL and CV6LU) leads. Traces were made at 50 mm/s with a voltage of 1 mV standardised to 1 cm.

Fibrosis quantification in heart tissue

From the middle of the right atrial wall from uninfected and infected dogs, a fragment of approximately 1.0 cm × 1.0 cm × 0.2 cm was taken for histopathological analyses after euthanasia at 270 days post-infection. Tissue fragments were fixed in 10% buffered formalin solution, dehydrated, cleared and embedded in paraffin. Blocks were cut into 4 mm-thick sections and stained by Masson’s trichromic for fibrosis quantitative evaluation. Thirty fields from each stained section were randomly chosen at a 40x magnification, giving a total of 224 × 106 μm2 analysed-myocardium area. Images were obtained through a Leica DM 5000 B microchamber (Leica Application Suite, UK, version 2.4.0 R1) and processed by software Leica Qwin (V3) image analyzer. The fibrosis area was quantified using the image segmentation function. All pixels with blue hues in the Masson’s trichromic section were selected to build a binary image, subsequently calculating the total area occupied by connective tissue in uninfected and T. cruzi infected dogs (Diniz et al. 2010).

Statistical analysis

The results of quantitative ECG and histopathological data parameters were compared by variance analysis followed by Tukey’s multiple comparison test. In all cases, differences were considered to be significant when < 0.05. PR interval prolongation and intensity of the fibrosis in the right atrium tissue were correlated using the Spearman rank test. Statistical calculations were made using GraphPad Prism software.

Results

First, a quantitative analysis of the right atrium’s fibrotic area of all animals infected with T. cruzi strains was assessed and it was observed that all infected dogs presented a higher intensity of heart fibrosis when compared to a healthy one (P < 0.05) (Figure 1). Then, it was prospectively evaluated for the early appearance of electrocardiographic (ECG) changes during the experimental infection of dogs inoculated with T. cruzi strains. ECG alteration measurements (P wave duration, PR and QRS interval) were evaluated at the baseline and at 120 and 270 post-infection days (Figure 2). These results showed significant differences between the infection and its baseline values and the values for the uninfected dogs with proportional time of life for the most of the quantitative ECG variable assessed. This was particularly true for values related with P wave duration and PR variables. Interestingly, in animals infected with VL-10 and AAS strains, the increase in quantitative ECG variables was detected earlier than in those infected with Berenice-78. In contrast, despite having been detected over a prolonged QRS interval in infected animals in relationship to baseline values, a difference was evidenced only in animals infected with VL-10 T. cruzi strain.

Figure 1.

 Quantitative analysis of fibrotic area in right atrium of dogs inoculated with 2000 trypomastigotes per kg of bodyweight of VL-10, AAS and Berenice-78 (Be-78) Trypanosoma cruzi strains (analyses were done at 270 days of infection). Masson trichrome staining was used and differences considered significant at < 0.05, &indicating significance in relationship to uninfected (NIC) animals (magnification, x 40).

Figure 2.

 Comparative analysis of P wave, PR and QRS intervals duration in dogs inoculated with 2000 trypomastigotes per kg of bodyweight of VL-10, AAS and Berenice-78 (Be-78) Trypanosoma cruzi strains. ECG measurements were available at the baseline and at 120 and 270 post-infection days. T1, baseline of uninfected animals; T2, time correspondent to 120 days; T3, time correspondent to 270 days; *indicates significant difference in relationship to baseline values among animals infected with each T. cruzi strain; #indicates difference in relationship to the group of uninfected animals in the corresponding time. Differences were considered significant at < 0.05.

The frequency of a variety of qualitative Chagas disease-related ECG in infected dogs was also assessed, as shown in Table 1. Conduction abnormalities were detected in 20% (3 of 15) of the infected dogs at 120 post-infection days. Three incidental new cases of A-V conduction abnormalities were diagnosed among the infected animals at 270 days of infection. At this time, considering all ECG alterations shown in Table 1, 73.33% (11 of 15) of the infected dogs presented at least one qualitative ECG abnormality.

Table 1. Electrocardiographic abnormalities among VL-10, AAS and Berenice-78 Trypanosoma cruzi infected dogs
 ECG AlterationsInfected dogs (n = 15)
VL-10 strainAAS strainBerenice-78 strain
Baseline120 dai270 daiBaseline120 dai270daiBaseline120 dai270 dai
  1. ASDB, anterosuperior divisional block of the left bundle branch; CRBBB, complete right bundle branch block; dai, days after infection; ECG, electrocardiographic; First A-VB, first degree atrioventricular block; First S-A, first degree sinoatrial block; IRBBB, incomplete right bundle branch block.

  2. Rhythm alteration – premature ventricular contraction, presence of ectopic P wave, ventricular extrasystole and brady-tachy syndrome.

AV conduction defectsIRBBB0/50/50/50/50/50/50/51/50/5
CRBBB0/50/50/50/50/50/50/50/51/5
First A-V B0/51/51/50/50/50/50/51/50/5
First S-A B0/50/50/50/50/51/50/50/50/5
ASDB0/50/50/50/50/50/50/50/51/5
Rhythm alteration0/50/50/50/50/53/50/50/52/5
Chambers overload0/53/53/50/51/51/50/50/53/5
 Abnormal ECG/dog0/5 (0%)3/5 (60%)3/5 (60%)0/5 (0%)1/5 (20%)4/5 (80%)0/5 (0%)2/5 (40%)4/5 (80%)

To evaluate the efficacy of the Bz treatment for preventing heart scars and ECG changes, dogs were infected with T. cruzi strains and treated for 60 days. Daily examination of fresh blood in all dogs treated with Bz showed a suppression of blood parasites, regardless of the degree of susceptibility to Bz. Meanwhile, the untreated dogs maintained particular profiles according to each strain of parasite: Berenice-78 (parasitemia peak: on the 15th day / average of trypomastigotes: 4.966/0.1 ml of blood); then, VL-10 (parasitemia peak: on the 33rd day / average of trypomastigotes: 5.555/0.1 ml of blood) and finally, AAS with no patent parasitemia. In particular, with the AAS strain, the infection was confirmed by haemoculture assay on the 20th post-infection day. After treatment, the blood culture presented negative results in 100%, 90% and 10% of the animals infected with Berenice-78, AAS and VL-10 strains, respectively, but T. cruzi kDNA was detected in the blood of all animas infected with VL-10 and AAS strains, while in those infected with the Berenice-78 strain, the PCR assay was persistently negative.

In dogs infected with drug-susceptible (Berenice-78) and drug-resistant (AAS and VL-10) T. cruzi strains, the Bz treatment efficacy in preventing chronic heart scars was determined through quantitative heart fibrosis analysis of the uninfected and infected dogs treated and untreated with Bz (Figure 3). Heart scars were prevented in those animals that were infected with Berenice-78, since the fibrosis intensity in the right atrium tissue was similar between the treated (2813.00 ± 607.13 collagen/74931 μm2) and uninfected dogs (2283.60 ± 343.20 collagen/74931 μm2), and lower in relationship to those infected and untreated (4616.00 ± 1715.82 collagen/74931 μm2). The treatment also induced a reduction in the cardiac fibrotic area intensity in animals infected with the AAS drug-resistant strain. In these treated animals, for the collagen values of 74931 μm2, the heart fibrosis was significantly lower (4024 ± 1272.44) in comparison to infected dogs (5839.2 ± 1423.49). However, the heart fibrotic area detected in the treated animals infected with AAS strain was significantly higher than in the uninfected dogs (2283.6 ± 343.20 of collagen/74931 μm2). Differently, the quantification of the heart fibrotic area (in 74931 μm2) in the drug-resistant VL-10 revealed very similar values between treated (6749.5 ± 1596.35) and untreated (6294.40 ± 896.04) animals.

Figure 3.

 Quantitative analysis of the right atrium fibrotic area. The dogs were inoculated with 2000 trypomastigotes per kg of bodyweight of VL-10, AAS, and Berenice-78 (Be-78) Trypanosoma cruzi strains, and treated with Benznidazole (Bz) at 7 mg/kg bid (Q12) for 60 days. Comparison is made with the mean of all infected and untreated (INF) and uninfected (NIC) controls groups. Masson trichrome staining was performed at 180 post-treatment days or 270 days after infection. Differences were considered significant at < 0.05, #indicates a significant difference in relationship to uninfected animals; &indicates significant difference in relationship to infected animals (magnification, x 40).

In order to analyse the ECG, treated-dogs were divided into two groups based on the atrium heart fibrosis evaluation: (i) treated without reduction in the heart fibrosis area (VL-10 strain) and (ii) treated with reduction in the heart fibrosis area (Berenice-78 and AAS strains). There was good agreement between the efficacy of the heart fibrosis reducing treatment and the ECG evaluation, as demonstrated by the reduction of ECG abnormalities in group (ii) and by the higher incidence of ECG alterations in group (i) (Table 2). In accordance, specific treatment prevented the occurrence of the qualitative ECG alteration until 120 days of infection, but two new incident cases of the conduction abnormalities and three of the chamber overload were detected 270 days after infection in 75% (three of four) of the treated dogs included in group (i).

Table 2. Electrocardiographic abnormalities among Berenice-78, AAS and VL-10 Trypanosoma cruzi infected dogs after Benznidazole treatment
ECG AlterationsBenznidazole treated dogs (n = 14)
(i) Absence of heart fibrosis reduction(ii) Heart fibrosis reduction
Baseline120 dai270 daiBaseline120 dai270 dai
  1. dai, day after infection; ECG, electrocardiographic; First A-VB, first degree atrioventricular block.

First A-V B0/40/42/40/100/100/10
Chambers overload0/41/43/40/100/101/10
Abnormal ECG/dog0/4 (0%)1/4 (25%)3/4 (75%)0/10 (0%)0/10 (0%)1/10 (10%)

In addition, Bz treatment prevented alterations of the quantitative ECG parameters in animals infected with Berenice-78 and AAS strains, since these treated animals showed very similar values in all quantitative ECG-derived variables analysed until 270 days after infection (Figure 4). By contrast, among VL-10-infected animals, Bz was not efficient in preventing the increase in the P wave duration and the PR interval. All treated animals showed increase in these variables which were significantly larger than baselines values (P < 0.05) and similar to that of the infected animals (P > 0.05).

Figure 4.

 Comparative analysis of P wave, PR interval and QRS interval duration in dogs inoculated with 2000 trypomastigotes per kg of bodyweight of VL-10, AAS, and Berenice-78 (Be-78) Trypanosoma cruzi strains and treated with Benznidazole (Bz) at 7 mg/kg bid (Q12) for 60 days. ECG measurements were available at the baseline and at 120 and 270 post infection days. *indicates significant difference (P < 0.05) in relationship to baseline values among animals infected with each T. cruzi strain.

Our findings were that infecting dogs with Berenice-78, VL-10 and AAS strains results in both electrical abnormalities (prolonged P wave and PR interval) and fibrosis in the right atrium tissue, and that the specific treatment was effective in preventing heart fibrosis in Berenice-78 and AAS infected animals, but not in the VL-10 infected. Based on this, we investigated whether PR interval prolongation and intensity of the fibrosis in the right atrium tissue were correlated using the Spearman rank test. As seen in Figure 5, correlation was observed between the PR interval and right atrium fibrosis (r = 0.551 and P = 0.0007).

Figure 5.

 Correlation between PR interval prolongation and fibrosis intensity in the right atrium tissue obtained from dogs inoculated with 2000 trypomastigotes per kg of bodyweight of VL-10, AAS, and Berenice-78 (Be-78) Trypanosoma cruzi strains and treated with Benznidazole (Bz) at 7 mg/kg b.i.d. (Q12) for 60 days. The analysis was performed using the Spearman rank test.

Discussion

Chagas cardiomyopathy is the most important clinical form of Chagas disease and clinical presentation typically varies widely according to the degree of myocardial damage (Biolo et al. 2010). Considering that ECG monitoring is used as the primary tool for disease detection and progression (Sosa-Estani et al. 2009), in this study we evaluated prospectively the early appearance of ECG changes during experimental infection of dogs inoculated with Bz-resistant T. cruzi strains. Clinical aspects of the human Chagas heart disease have become an increasingly useful target for investigation and non-invasive techniques. In this respect, ECG and echocardiography have been successfully applied to the study of dogs in recent years (Lana et al. 1992; Melo et al. 2011; Santos et al. 2012), due to their clinical, immunological and pathological similarities with the human disease.

In the last decades, ECG analysis from ours and other groups have allowed us to verify the reference values for the duration and amplitude of ECG waves that take into consideration the differences among gender, size and genetic breeds of dogs (Laranja et al. 1949; de Lana et al. 1992; Hanton & Rabemampianina 2006). Therefore, it is acceptable that each laboratory establishes its own reference parameters according to the genetic breeds of dog used. In our particular case, using mongrel dogs, our ECG reference values for normal dogs are shown in Figure 2. These approaches permit prospective evaluations in a single animal and direct assessment of functional changes during the disease’s evolution.

Some of the most common ECG alterations described in human and in experimentally or naturally infected dogs with T.cruzi are: complete right bundle branch block (CRBBB); left anterior fascicular block; complex ventricular arrhythmias and sustained ventricular tachycardia; supraventricular tachyarrhithymias, second degree; complete A-V blocks; and sinus tachycardia (Pellegrino 1947, Laranja et al. 1949; Lana et al. 1992; Sosa-Estani et al. 2009). ECG measurements such as P-wave duration or dispersion and consequently, the duration of the PR interval are ECG indexes evaluated in human cardiology and veterinary medicine. It is assumed that the duration of the P-wave reflects the electrophysiological properties of the atrium muscle in the human and animal models. So the regional depolarisation disturbances caused by various heart illnesses, including Chagas disease, may lead to a variation of the duration of the P-wave at different ECG leads (Noszczyk-Nowak et al. 2011). The P wave dispersion is also evaluated in humans and dogs as a prognosis index: for the case of atrial fibrillation, for the development of some types of supra-ventricular arrhythmia and for dilated cardiomyopathy; which occurs in these individuals/dogs with enlarged atria due to mitral/tricuspid insufficiency (Dilaveris & Gialafus 2001, Noszczyk-Nowak et al. 2011). Particularly in veterinarian medicine, the P wave dispersion value is a factor independent from body weight and sex and it is a constant parameter in healthy dogs, with no supra-ventricular conduction disturbances and changes in the atria size, resulting in a low SD value for a big and diversified group of healthy dogs.

In our investigation, this was evidenced by progressive prolongation of wave P, PR and QRS intervals at the baseline and up till 270 days of infection. In some groups, CRBBB, ventricular arrhythmias and chamber overload were also observed, particularly at 270 days of infection. This data could be supported by the evolvement of the incomplete right bundle branch block, previously detected in dogs infected by Berenice-78, to the point of CRBBB during the clinical follow-up. Several experimental and clinical studies have previously reported ECG abnormalities in the murine model and human infection including increased PR (Postan et al. 1987; Garcia et al. 2005; Williams-Blangero et al. 2007) and QRS interval (Bustamante et al. 2002; Williams-Blangero et al. 2007; Eickhoff et al. 2010). Our results agree with recent clinical studies that showed a high frequency of early ECG abnormalities among seropositive children (Andrade et al. 1998; Salazar-Schettino et al. 2009). Experimentally, it has been described that some T. cruzi stocks induce more cardiac damage than others (Andrade 1990; Guedes et al. 2010) and the consequent fibrosis process in heart muscles present and detectable at the end of acute phase, might contribute, in part, to the ECG and other clinical changes described in infected animals.

An inflammatory process is the first step to the development of a progressive pathological process and consequent myocardium scars from collagen deposition. Myocarditis leads to the occurrence of fibrosis, formed by a progression of myocardial cells and their replacement by collagen (Andrade et al. 1997). The deposition of collagen fibers and fat tissue in the interstitial spaces hinder the passage of nutrients to the myocardium. Furthermore, areas occupied by fibrous tissue change the myocardial structure and play an important role in the new position of cardiac cells, providing them with flexibility and stiffness for adequate heart functioning (Weber et al. 1989).

This fibrosis could, in part, be responsible for the alterations in electrical conductance as well as the functionality of the organ (Andrade et al. 1997). Benznidazole (Bz) has been shown to be effective in eliminating parasites during the acute phase of infection, but also interferes in the course of immunological and pathological responses (Caldas et al. 2008). The benefit of the early Bz treatment was clearly demonstrated by the prevention or reduction of fibrosis in the myocardium muscle for the susceptible-Berenice-78 and resistant-AAS strains. Differently, Bz treatment did not prevent cardiac scars in the VL-10 infected dogs, confirming the drug-resistance of the VL-10 strain to be the highest, as previously shown by Caldas et al. (2008). The positive effect of the current specific treatment, despite its inability to eradicate the parasite, has been explained by the drug-induced reduction of the parasite load in infected tissue that consequently reduces the severity of the associated heart tissue lesions (Tarleton et al. 2007). It is suggested that apparent variability in effectiveness of treatment of dogs infected with drug-resistant T. cruzi stocks may reflect the existence of different levels of drug-resistance among these resistant parasite stocks.

Consistently, considering that Chagas disease induces heart disease by provoking extensive myocardial fibrosis that has a positive correlation with disease severity (Strauss et al. 2011), in our study, Bz-treated animals were divided according to the intensity of heart fibrosis: (i) animals without reduction of myocardial fibrosis after treatment (VL-10 strain) and (ii) animals that presented prevention (Berenice-78 strain) or reduction (AAS strain) of myocardial fibrosis after treatment. This data reinforced the importance of the relationship between the efficacy of Bz treatment, the reduction of heart fibrosis, and the ECG changes. These findings indicate that an effective antiparasitic treatment in the early stages of Chagas disease may lead to a significant reduction in the frequency and severity of the parasite-induced cardiac disease, even when parasites are not completely eliminated.

Finally, our results demonstrated that T. cruzi infection was able to induce progressive ECG changes in dogs and treatment with Bz, administrated during the early phase of the disease was efficient in preventing ECG changes, being these benefits related with T. cruzi stocks. The present study extends our previous observations (Guedes et al. 2002) regarding the use of the dog model to explore desired benefits concerning specific treatment in the clinical Chagas disease.

Acknowledgements

We recognise the support of Universidade Federal de Ouro Preto, Fundação de Amparo a Pesquisa do Estado de Minas Gerais, Conselho Nacional de Desenvolvimento Científico e Tecnológico. M. T. Bahia and A. Talvani are recipients of research fellowships from CNPq and I. S. Caldas is funded by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior.

Ancillary