Trypanosoma cruzi (Chagas, ), the etiologic agent of Chagas disease (CD), is a flagellate protozoan of the Trypanosomatidae family that is widely distributed in nature, occurring from the southern United States to southern Argentina and Chile. It is capable of infecting almost all tissues of hundreds of species of mammals to which it is transmitted by hematophagous hemiptera of the subfamily Triatominae, known as kissing bugs (barbeiros, vinchuca, bicudos and many other regional terms in Latin America) (Galvão et al., ; Noireau et al., ; Jansen et al., ). T. cruzi is a taxon that presents expressive intraspecific diversity whose interpretation has been challenging scientists since Carlos Chagas. Currently, seven genotypes or discrete typing units (DTUs) are recognized in the species, namely, TcI-TcVI, in addition to a genotype described as Tcbat. The latter, initially associated with Chiroptera (Marcili et al., ), has already been detected in humans (Guhl et al., ; Ramírez et al., a).
You will get efficient and thoughtful service from Accu-Tell.
Trypanosomiasis by T. cruzi is primarily a wild enzooty. Its transmission in nature takes place within a complex trophic network, in which each animal species plays a different role, in space and time, regarding its maintenance and infective competence (Jansen et al., ; Jansen et al., ), resulting in distinct enzootic scenarios. In other words, each region is unique and has a specific transmission network, which needs to be understood and known so that one can recognize areas of epidemiological risk and correctly guide health agents and local residents (Roque et al., ).
Dogs represent the first domestic T. cruzi hosts studied by Carlos Chagas and are among the first experimental models used by him (Jansen et al., ). Among the different species of mammals, Canis lupus familiaris (domestic dog) has a significant role in the epidemiology of T. cruzi because it can act as a bioindicator. This is because this taxon is demonstrably capable of acting as a competent sentinel, signalizing the transmission of the parasite among wild mammals (Travi, ). Unlike other countries, domestic dogs in Brazil rarely present high parasitemia; that is, they display low infective potential (Xavier et al., ; Brandão et al., ). A proposal from our group that has received increasing attention from the Brazilian Health Authorities is the longitudinal serological survey of domestic and peridomestic mammalian species to determine the prevalence and/or incidence of T. cruzi infection (Jansen et al., ).
Serological diagnosis is the method of choice to assess the spread of the wild transmission cycle of T. cruzi and, consequently, define areas where the risk of human disease occurs. However, serological tests such as the indirect immunofluorescence reaction test (IFAT) and enzyme immunoassay (ELISA) often use a complex mixture of parasitic antigens, which can be related to false negative or false positive results (Santos et al., ). This fact is due to the possibility of cross-reactions of T. cruzi antigens with those other trypanosomatids, such as Leishmania sp. or other Trypanosoma species because they share common epitopes. With the development of recombinant DNA technology, several bacterial and eukaryotic gene expression systems allow the production of parasite antigens in large quantities, with a high degree of purity and standardized quality (Silveira et al., ; Foti et al., ). Thus, studies have shown that to improve diagnostic accuracy, the selection of more specific antigenic fragments for T. cruzi would be effective (Silveira et al., ; Camussone et al., ; Marchi et al., ; Santos et al., ; Santos et al., ; Peverengo et al., ; Santos et al., ; Daltro et al., ; Del-Rei et al., ; Leony et al., ; Silva et al., ). Chimeric proteins consist of antigenic sequences of the parasite, presenting a series of epitopes, which increases the diagnostic sensitivity (Santos et al., ). The IBMP proteins (IBMP-8.1, IBMP-8.2, IBMP-8.3 and IBMP-8.4) were evaluated their potential for diagnosing T. cruzi. The T. cruzi proteins whose antigenic regions (IBMP muti-epitope antigens) were used to construct the chimeric antigens are described in the [Supplementary Material of the study by Santos et al. ()].
Silva et al. () showed that the lateral flow immunochromatographic rapid test (LFRT) for the diagnosis of T. cruzi in humans, using the chimeric proteins IBMP-8.1 and IBMP-8.4, is extremely efficient, and our idea was to expand the use of this test for the diagnosis of this parasite in domestic dogs and others species of wild canids in its original format. The use of two chimeric proteins increases the diagnostic potential of the rapid test as it increases the amount of available T. cruzi epitopes. This is because the two chimeric proteins used have different epitopes due to their different compositions. The rapid test results are quick, easy to perform and interpret, do not require laboratory infrastructure, and use a small amount of biological material (serum, plasma or whole blood). Additionally, there is no need for equipment and specific knowledge to carry it out. The use of the test will allow rapid preventive actions to be taken in places with or without notification of Chagas disease (Sánchez-Camargo et al., ). Consequently, it will bring benefits mainly to locations where access to a more complex laboratory test is limited.
The recombinant antigens were formatted in a rapid immunochromatographic test using either Staphylococcus aureus protein A or Streptococcus pyogenes protein G as the gold-labeled reagents for the visualization of the precipitin band formed between the immunoglobulin (Ig) G-specific antibodies and the recombinant antigen immobilized on the nitrocellulose stripe used in the test. Despite the fact that most rapid immunochromatographic tests are formatted with protein A, the rationale to test both protein A and protein G was based on the fact that these microbial molecules bind with different affinities and specificities to immunoglobulins of various species, including dogs (Frohman et al., ; Nilsson et al., ; Bjorck and Kronvall, ; Costa et al., ).
Here we need to consider that: 1) establishing a diagnosis based on only one serological test is not secure enough. Two tests are always required, one being confirmatory; 2) the screening of dogs (not to mentioning wild animals) in the field is extremely advantageous, due to the speed of the response of the test, which already allows the immediate adoption of some mitigation measures. Biological material is a difficult to be maintained in field conditions, and bad storage conditions may even result in its loss. Also, the subsequent communication of results to the examined community constitutes a challenge. There are still many places in Brazil without internet access or mail making the communication of the results a difficult task.Insufficient access to diagnosis is yet a major issue to diagnose T. cruzi infection. Therefore, there is a need for more practical diagnostics and/or diagnostic algorithms that better suit the demands and field conditions and are indispensable to prevent the reemergence of CD in vulnerable regions of Brazil with similar epidemiological characteristics. Such diagnostics should be made available not only to easily detect T. cruzi infection, including different DTUs and mixed infection but also to avoid crossed reaction by other trypanosomatid infections (Gállego et al., ). In this study, it was evaluated the performance of recombinant chimeric antigens (IBMP-8.1 and IBMP-8.4) for the detection of anti-T. cruzi IgG antibodies in dog sera using Chagas-LFRT.
All procedures performed with wild and domestic animals received authorization from IBAMA (wild canids) and followed protocols approved by FIOCRUZ’s Animal Use Ethics Committee (P-99; P-03; P/06; L-07; L-050/; LW-81/12).
Serum samples from species of the order Carnivora were submitted to the Protein A affinity test to validate the use of the Chagas/Bio-Manguinhos Lateral Flow Immunochromatographic Rapid Test (Chagas-LFRT) for domestic and wild canids. The species tested were Canis lupus familiaris (n = 4), Canis lupus (n = 2), Chrysocyon brachyurus (n = 1), Lycalopex vetulus (n = 2), Lycaon pictus (n = 1), Speothos venaticus (n = 1) and Cerdocyon thous (n = 3). These samples were submitted to the Bio-Manguinhos Canine Visceral Leishmaniasis DPP LVC® Rapid Test kit, which also uses protein A, and the intensity in the control band marking was used to classify the strength of the interaction between the antibodies of the dog species and this protein, regardless of the marking of diagnostic bands. The affinity of the interaction with Protein A of these samples was classified as strong (++++), medium (+++), low (++) and very low (+).
We used 253 serum samples from Canis lupus familiaris and nine serum samples from six species of wild canid to detect T. cruzi infection: European wolf, Canis lupus (n = 2), Cerdocyon thous (n = 2), Lycalopex vetulus (n = 2), Lycaon pictus (n = 1), Speothos venaticus (n = 1) and Chrysocyon brachyurus (n = 1). C. l. familiaris samples were collected in distinct scientific expeditions to different regions of Brazil by the Laboratory of Trypanosomatids Biology (LABTRIP) and by the Laboratory of Wild Mammals Reservoir Biology and Parasitology, both from the Oswaldo Cruz Institute – FIOCRUZ/Rio de Janeiro between and . Wild canid samples were obtained from the Zoo Park Foundation of São Paulo and from studies conducted in the southeastern Goiás state municipality of Cumari (Brandão et al., ) (Table 1 and Figure 1). This material is kept at -20°C in the LABTRIP serum bank. Sera were aliquoted and each aliquot thawed only once.
TABLE 1FIGURE 1Serum samples with previous serological diagnosis (IFAT and ELISA) were selected to compose the panel of positive (n = 133) and negative (n = 129) samples. The serologically positive samples were those with positive IFAT (titer ≥ 1/40) and ELISA (optical density ≥ 0.200). Within the panel of positive samples, samples with positive blood cultures (n = 20), including T. cruzi isolates characterized by different DTUs TcI (n = 8), TcII (n = 1), TcI/TcII (n = 2), TcIII (n = 2), TcIV (n = 1), and TcIII/TcV (n = 6), were included to validate the test based on the recognition of the different genotypes of the parasites deposited in the Coleção de Trypanosoma de Mamíferos Silvestres, Domésticos e Vetores, and COLTRYP/Fiocruz (Oswaldo Cruz Foundation, Rio de Janeiro, RJ, Brazil). To compose the panel of serum samples with different parasitic infections (n = 19), we used samples diagnosed as Leishmania infantum (n = 10), Trypanosoma caninum (n = 3), Trypanosoma rangeli (n = 1), Crithidia mellificae (n = 1), Anaplasma platys (n = 1), Dirofilaria immitis (n = 1), Ehrlichia sp. (n = 1) and a sample with mixed infection by Ehrlichia sp./Dirofilaria immitis (n = 1) (Table 1 and Figure 1). All samples of dogs serum used in the cross-reaction panel were subjected to immunofluorescence antibody test (IFAT) and ELISA to diagnose T. cruzi and Leishmania infantum infection and those positive for T. cruzi were excluded from the analysis.
The Chagas rapid test works with lateral flow immunochromatographic labeling, employing a combination of a protein conjugated with colloidal gold particles and T. cruzi antigens (recombinant proteins IBMP-8.1 and IBMP-8.4) bound to a solid phase (nitrocellulose membrane) (Silva et al., ). The sample to be analyzed is applied to a specific area of the plastic holder (cassette), followed by the addition of a running buffer. The buffer provides lateral flow of released components, promoting the binding of antibodies to antigens. Four different formulations of running buffer with and without blocking were tested, with modifications to the buffer formulation: high, medium or low blocking of nonspecific reactions in the tests and TR DPP LVC® canine visceral leishmaniasis from Bio-Manguinhos kit buffer.
To reveal the reaction, a protein A system conjugated to colloidal gold was used. The use of this type of conjugation basically privileges reactions with IgG because it makes use of the extraordinary affinity of protein A with the Fc region (crystallizable fragment) of IgG. Antibodies present in the sample bind to specific proteins conjugated to colloidal gold. In case a sample is positive, the “immunoconjugate” complex migrates on the nitrocellulose membrane, being captured by the fixed antigens in the test area and producing two purple/pink lines. In the absence of specific anti-T. cruzi antibodies, the purple/pink lines do not appear in the test area. In all cases, the sample continues to migrate across the membrane, producing a purple/pink line in the control area, which demonstrates proper functioning of the reagents (Figure 2A).
FIGURE 2The test was performed by applying 5 µL of sample (serum, plasma or whole blood) to the sample well followed by the addition of three drops of running buffer using a dropper bottle with a measuring nozzle. This mixture migrates by capillarity, eluting the conjugate (protein A conjugated to colloidal gold) to the test area where there are the antigens obtained through collaboration with the Proteomics and Protein Engineering Laboratory of the Carlos Chagas Institute - ICC/Fiocruz/PR, which provided two recombinant chimeras named IBMP-8.1 (21 KDa) and IBMP-8.4 (40 KDa). These chimeric proteins were expressed in Escherichia coli and purified by chromatography (Santos et al., ; Santos et al., ). Excess nonreaction-related components migrate to the other end of the tape where they are retained. The reading was carried out between 10 and 15 minutes, counted with the addition of running buffer, and incubated at room temperature. All inputs used in standardization are from the Chagas Rapid Test/Bio-Manguinhos (https://www.bio.fiocruz.br/index.php/br/produtos/reativos/testes-rapidos/teste-rapido-chagas) for Chagas disease produced by Bio-Manguinhos (Silva et al., ).
Visual reading of the recognition densities of the test and control bands was performed by two observers, as well as with a rapid lateral flow test reader to define the sensitivity and specificity parameters of the test to equate the inherent subjectivity of the human eye when reporting results in samples with low titers of antibodies. This type of device was developed in collaboration with the Carlos Chagas Institute-ICC/PR, Paraná Molecular Biology Institute-IBMP/PR and the Paraná Technological University-UTFPR/PR. Visual reading was standardized as follows: N1= strong negative; N2=weak negative; P1= positive of weak intensity; P2= positive of medium intensity and P3= positive of strong intensity (Figure 2B). The automated reader expresses the results quantitatively (ranging from 0 to 100) in optical density (OD), performing the reading for both diagnostic bands and the control band. These values were used to define the cutoff by the ROC curve.
A statistical summary of the data, a box-plot graph and a cluster analysis were performed to determine the Chagas rapid test cutoff point for classification of data collected by the automated reader of the IBMP-8.1 and IBMP-8.4 chimeric antigens by the matrix of confusion, which is intended to evaluate the performance of a binary classifier. The test performance (sensitivity and specificity) was evaluated format in single or duplex using the IBMP-8.1 and IBMP-8.4 proteins, separately and together to the serological diagnosis of T. cruzi infection wild and domestic Canids. The cut off, sensitivity, specificity and accuracy values were established by determining the largest area under the ROC curve (receiver operating characteristic) (Swets, ). The ROC curve is a statistical and graphical method for determining the best cutoff point for a diagnostic test. The highest point on the curve, corresponding to the upper left angle of the graph, represents 100% sensitivity and 0% false positives (d = 0); in this case, the ideal value for a diagnostic test, called the gold standard. Samples were separated by IFAT serological titers to generate the ROC curve and calculate sensitivity, specificity and cutoff values. Cohen’s kappa (K) analysis was used to determine the strength of agreement between the reference standard tests (IFAT and ELISA) and the Chagas LFRT, which was interpreted as weak agreement (< 0), slight agreement (0.01-0.20), fair agreement (0.21-0.40), moderate agreement (0.41-0.60), substantial agreement (0.61-0.80), and near perfect agreement (0.81-0.99) (Cohen, ). All data analyses were performed using RStudio software version 1.2. and R version 3.6.3, and a p-value under 5% (p < 0:05) was considered significant.
In the Protein A affinity test of samples from different species, all canids (wild and domestic), a strong interaction was demonstrated for the species Canis lupus familiaris, Canis lupus, Chrysocyon brachyurus, Lycalopex vetulus, Lycaon pictus, Speothos venaticus and Cerdocyon thous. This result allowed the choice of Canis lupus familiaris for the standardization of the Chagas rapid test for T. cruzi, in addition to this animal being considered an important sentinel of infection in different areas of study. Among the four formulations of running buffer with different types of blocking tested, the buffer selected was the original of the commercial kit of Chagas Rapid Test (with blocking) because it presented greater intensity in the labeling of the chimeric proteins IBMP-8.1 and IBMP-8.4. The selected buffer showed a high intensity recognition profile, visually classified as P3 for the two chimeric proteins and for the test control band.
Figure 3 shows the mean optical density values (OD 1.61) in serum samples from dogs with negative serological diagnosis. The mean of the optical density values of the samples from dogs infected by T. cruzi was 45.36 against the chimeric antigens IBMP-8.1 and IBMP-8.4. Among the positive samples considering the cutoff point OD ≥ 4.8, the IBMP-8.1 antigen was positive in 68% and the IBMP-8.4 antigen in 60% of the samples tested, and an agreement of 45% was observed between the IBMP-8.1 and 8.4 antigens in the positive samples against T. cruzi infection. This result highlights the importance of combining two antigens in a single test to increase the test sensitivity in the diagnosis of T. cruzi in dogs. Among the negative samples, only one (OD 5.1) reacted against the IBMP-8.1 antigen and means (OD 1.49), and four were false positive against the IBMP-8.4 antigen (OD ≥ 4.8 and ≤ 10.9) and the mean of the negative samples (OD 1.18).
FIGURE 3The optical densities obtained by the negative and positive samples were directly proportional to the serological titer by IFAT. The automated reading of the Chagas rapid test showed a pattern in which the optical density increased as the IFAT serological titer increased, mainly demonstrated by the means (Figure 4).
FIGURE 4Visual analysis showed high accordance with the band intensities of the IBMP-8.1 and IBMP-8.4 chimeric antigens according to the automated reader device. Negative titers showed total absence of color in 93.8%, classified as N1 and N2. Among the panel of negative samples by IFAT (titer ≤ 1/20), the observer rated was 78 (N1) and 13 (N2), 2 (N2/P1) and 4 (P1) false positive samples. In the panel of positive samples (titer ≥ 1/40), the observer classified as follows: 13 (P1), 15 (P2), 15 (P3) and 10 samples were classified as false negatives: 3 (N1) and 7 (N2). The intermediate titers (1/40 to 1/320) had a medium intensity in the test zone lines, mainly diagnosed as P1 (10), P2 (9) and P3 (1), as the higher titers (1/640 to 1/) showed the strongest intensities in the marking of chimeric proteins, with the most prevalent visual diagnosis being positive with strong intensity P3 (14), P2 (6) and P1 (3). The visual classification did not present a significant difference in the classification based on the optical densities of the automated reader in relation to the cut off 4.8 for positive and negative samples; that is, the visual classification can replace the reader’s classification by Kappa analysis (Figure 2B and Table 2).
TABLE 2Of the samples with positive blood cultures, 90% (18/20) had a positive result through visual and automated reading in the Chagas-LFRT. Samples from wild and domestic canids infected with the DTUs TcI, TcII, TcIII, TcIII/TcV and TcIV were considered positive by the test. Only two samples from domestic dogs with positive blood cultures (DTU TcI), with titers of 1/40 and 1/80, had a negative result in Chagas-LFRT (Figure 2B).
Regarding the samples infected with other trypanosomatids and other parasites from domestic canids, all samples from dogs infected by L. infantum, T. rangeli, T. caninum and D. immitis presented negative results in the Chagas-LFRT. However, positive samples for the parasites Anaplasma platys, Crithidia mellificae, Ehrlichia sp. and a sample with a mixed infection of Ehrlichia sp. and D. immitis showed weak positive results for T. cruzi in the rapid test (P1). The test was able to evaluate 78.9% (15/19) of the samples infected by other dog parasites as negative (Table 1).
When we only use one single of the chimeric proteins the sensitivity and specificity was lower (BMP-8.1: sensitivity 92.5% and specificity 85.6% or IBMP-8.4: sensitivity 75% and specificity 83.3%), already the duplex test format (with two chimeric proteins, IBMP 8.1 and IBMP8.4). The chimeric proteins IBMP-8.1 and IBMP-8.4 using Chagas-LFRT had a sensitivity of 94% and a specificity of 91%, increasing the diagnostic power of the test when used the two chimeric proteins to the serological diagnosis of T. cruzi infection wild and domestic Canids. Five cutoff points were tested and evaluated (2.3, 3.7, 4.8, 5.9 and 8.7), and the cutoff point was considered an optical density of 4.8 and an area under the curve (AUC) value of 92.6% (Figure 5A and Supplementary Figure S1). The ROC curve presented an AUC of 97.3% for the cutoff point by IFAT of >1/20 and 96.8% for samples >1/40 in IFAT (Figure 5B). In the Kappa test, the coefficient was 0.84 (elevated), accuracy was 0.90, predictive positive values was 0.96 and negative values was 0.87 between the IFAT and the automatic reader of the Chagas rapid test (Table 3 and Supplementary Figure S1).
FIGURE 5TABLE 3One of the main bottlenecks in monitoring parasitosis in wild and even domestic animals is the small number of simple and feasible diagnostic methods to be applied in the field. In this sense, the test that we are proposing fills an important gap both in the feasibility and in the spectrum of animal taxa in which trypanosomatid infection can be diagnosed. Establishing access to diagnostics for wild animals species is of utmost epidemiological and epizootiological importance as the health of all animals (including humans) is profoundly interdependent and this awareness is always increasing as clearly stated in the One Health perspective (Jansen et al., ).
At present, most Chagas disease occurs in a new epidemiological profile in which Trypanosoma cruzi transmission occurs independently of the domiciliation of triatomines. A new approach to control measures is necessary, as the measures previously applied, in this case, the use of insecticides to control species of intradomiciliary triatomines, does not fit the current epidemiological reality. Targeted control of reemerging transmission can be achieved by improved understanding of T. cruzi in canine populations. Castillo-Neyra et al. () suggest that dogs may be useful sentinels to detect reinitiation of transmission following insecticide treatment. A correlation between dogs with positive serology to T. cruzi and proximity to an infected triatomine (distance ≤50 m) was observed, suggesting that dogs may be useful sentinels to detect reinitiation of transmission following insecticide treatment (Castillo-Neyra et al., ).
The use of sentinel dogs has proven to be appropriate for any area that wants to assess the risk of human disease regardless of biome or cultural habits. Diagnosis using rapid tests is absolutely suitable for field conditions. It will enable large-scale diagnosis of areas of transmission of T. cruzi and, consequently, of risk for human disease. This work is perfectly aligned with the National Agenda of Priorities in Health Research, which textually recommends the development of new models and approaches for the surveillance of adverse health events and emerging diseases and the development of new technologies for the epidemiological surveillance of health problems, including specific forms of monitoring and scenario studies. Integration of canine T. cruzi blood sampling into existing interventions for zoonotic disease control (e.g., rabies vaccination programs) can be an effective method of increasing surveillance and improving the understanding of disease distribution (Castillo-Neyra et al., ).
The classical bottleneck has always been to preserve biological samples in field conditions that are often rather precarious besides transporting samples to the central level, as well as dispose of personnel sufficiently trained to carry out conventional methods of serological diagnosis on a large scale. Fast and easy tests to be performed have been increasingly sought after in regard to diagnosis. In this study, we performed an evaluation of the performance of the rapid lateral flow immunochromatographic test (Chagas-LFRT) for the serological diagnosis of T. cruzi infection in Canis lupus familiaris using two chimeric proteins (IBMP-8.1 and IBMP-8.4) as antigens. In this way, a quick decision-making process that will shape the epidemiological surveillance actions associated with the transmission of the parasite is made possible. Point-of-care tests are applicable in different sectors and bring short-term results in healthcare facilities and remote locations to complex diagnostic methods (Silva et al., ). Here, the use of a rapid test for dogs was validated, which has already been tested and approved for the diagnosis of T. cruzi infection in humans (Silva et al., ), optimizing its production and expanding its application in the field. Our results show that the test was sensitive to detect seroconversion, enhancing its application as a first measure to monitor the presence of the parasite, which indicates active transmission in the area, attested by seroconversion events, despite the low rate of positive parasitological tests in dogs from Brazil (Xavier et al., ; Brandão et al., ).
Dogs are animals that live together and accompany humans since they are still in the hunter-gatherer stage. In an evolutionary trade-off, dogs guaranteed safety, and humans, in turn, guaranteed dog food and shelter. In several cultures, dogs were used for different purposes, including prewarming the feet, hunting, guarding, adorning, sports and even a source of protein. Additionally, dogs are very common domestic animals and are easily traceable, and annual rabies vaccination campaigns as well as leishmaniasis control programs have resulted in the training of health service personnel in handling and collecting blood from these animals (Xavier et al., ; Daltro et al., ).
Dogs participate in the transmission cycle of T. cruzi in different ecoepidemiological scenarios, in addition to being a common animal in all areas, regardless of environmental and climatic conditions, which are not limiting for its occurrence. In Brazil, the pattern of T. cruzi infection in dogs is more related to the maintenance of the parasite in low/absence parasitemia, which results in low importance for the transmission of the parasite (Xavier et al., ). The plasticity of T. cruzi in nature ensures that it is transmitted in multiple environmental characteristics that vary in space and time, which is supported by the richness and diversity of wild mammals and vectors. In these multifactorial scenarios, the dog can be used as a sentinel for T. cruzi presence. Thus, the diagnosis of T. cruzi in domestic dogs is extremely important, as the dog’s role in the transmission of T. cruzi has been described as a dead-end host and sentinel of infection in the wild environment (Roque and Jansen, ). Areas where the dog’s infection is ≥ 30% signal the presence of a wild transmission cycle occurring close to the dog’s circulation areas, indicating the risk of Chagas disease cases (Xavier et al., ). As this test proved to be highly viable for the diagnosis of the parasite in dogs in our study, its use for detecting T. cruzi infection in dogs will allow area monitoring in real time, signaling areas with transmission of the parasite in the wild environment and perhaps also undetected human cases. Thus, accurate situational diagnosis is still in the field.
As T. cruzi is a parasite with high genetic variability, there is a significant difference in the performance of commercial serological tests related to the antigens used (Dopico et al., ). Therefore, the serological diagnosis of T. cruzi infection is quite complex due to the lack of reference methods. Furthermore, methods using parasite DNA are viable only in the initial phase of infection and are not capable of attesting to parasite viability (Daltro et al., ; Leony et al., ; Silva et al., ). Due to this situation, the World Health Organization (WHO) recommends the use of two tests to complete the diagnosis of T. cruzi infection (WHO Expert Committee, ). The diagnosis of the parasite is even more complicated in regard to wild and domestic canids due to the lack of validated tests. The antigenic matrices currently used can lead to low specificity and a high rate of cross-reaction with other trypanosomatids and other dog parasites (Leony et al., ).
Costa et al. () evaluated both IgG binding proteins in rapid immunochromatographic test kits for the diagnosis of canine visceral leishmaniasis. The tests were assembled with either Leishmania infantum recombinant antigens K39 or K26 and with either gold-labeled Staphylococcus aureus protein A or Streptococcus pyogenes protein G. The test using recombinant antigens K39 or K26 produced results that were only slightly superior to protein A than to protein G (no significant differences were observed), possibly because this minor difference in sensitivity between these 2 tests is due to the slightly higher affinity that protein A has over protein G for dog IgG (Miele and Krakowka, ; Peng et al., ). Additionally, our test that relies on gold-labelled Staphylococcus aureus protein A conjugate that may be easily adapted to other wild species thus, optimizing large-scale surveys. This is especially important in investigation of disease outbreaks. Therefore, our test is opening the possibility of the diagnosis not only of T. cruzi in dogs but also of other trypanosomatids species infections in other species that domestic dogs.
Chimeric antigens provide the diagnosis with a greater number of parasite epitopes, thus reducing false-negative results when compared to the use of nonrecombinant antigens (Dopico et al., ). The capacity of the rapid test to detect all T. cruzi DTUs that circulate in Brazil, also suggested by Santos et al. () and Dopico et al. (), is an important characteristic. In fact, in Brazil, there is a predominance of TcI and TcII DTUs, which are also the most frequent DTUs in wild carnivores and dogs (Rocha et al., ; Jansen et al., ; Brandão et al., ), but other DTUs may also be observed at minor rates. As an example, DTUs TcIII and TcIV have already been described in infected dogs in Brazil (Brandão et al., ). The chimeric antigens IBMP-8.1 and IBMP-8.4 in the rapid test platform are presented as tools capable of overcoming limitations imposed on other antigens and are also successful in other platforms investigated in previous studies, such as indirect ELISA and liquid microarray (Santos et al., ; Santos et al., ; Santos et al., ; Del-Rei et al., ; Leony et al., ; Silva et al., ).
Leony et al. () used three reference strains (Colombian, Y and Berenice), known to have TcI and TcII DTUs, for experimental infection of domestic dogs with T. cruzi. This study was successful in diagnosing the parasite using the chimeric proteins through ELISA. Here we emphasize that, in addition to expanding the recognition panel of DTUs different from those analyzed by Leony et al. (), our study showed the possibility of using a rapid test platform with chimeric proteins to detect T. cruzi in naturally infected domestic and wild dogs.
The breadth of the serological profile of the samples used in our study, which included samples with recent and older infections and with different serological titers, indicates that the cutoff point found in our study for dog samples was 4.8, indicating the high diagnostic accuracy of the chimeric proteins IBMP-8.1 and IBMP-8.4 for dog and wild carnivore infections. A sample of serum from a domestic dog from Guarapari (Espírito Santo) was included in two panels: panel of negative samples in the serological diagnosis (IFAT and ELISA) with titer of 1/20 in the RIFI, below the cut-off point and panel of positive samples in the parasitological examination. This type of phenomenon is not uncommon and suggests an initial infection, when the immune response is not yet established.
Leony et al. () found high sensitivity for four chimeric proteins, including two tested here (IBMP-8.1, IBMP-8.2, IBMP-8.3 and IBMP-8.4), in the diagnosis of T. cruzi infection in domestic dogs using ELISA. However, Leony et al. () found cross-reactivity to IBMP-8.1 for at least one of all tested parasites (anaplasmosis, babesiosis, dirofilariosis and ehrlichiosis), with the exception of other trypanosomatids. IBMP-8.1, in Chagas-LFRT, also cross-reacted with other parasites, with positive results for anaplasmosis, ehrlichiosis and Crithidia mellificae. However, unlike the study by Leony et al. (), this antigen did not react with dirofilariosis in the rapid test Chagas. The chimeric protein IBMP-8.4 did not cross-react on the ELISA platform, but in the rapid test, it reacted with Ehrlichia sp. This can be explained by the difference number of epitopes provided by IBMP-8.4 compared to IBMP-8.1 (Del-Rei et al., ). Herein, we propose the use of Chagas-LFRT for the serological diagnosis of T. cruzi infection in dogs as a screening test in the field to monitor the area and that a second confirmatory test (IFAT and/or ELISA) has to be used for individual diagnosis.
We did not observe cross-reactions with the main trypanosomatids described infecting dogs (L. infantum, T. caninum and T. rangeli), and these results show that the rapid test for T. cruzi also proves to be a useful and viable tool in places where there is spatial and temporal overlapping of transmission by other trypanosomatids. Daltro et al. () evaluated the cross-reactivity of chimeric proteins in the diagnosis of T. cruzi through ELISA in human samples with American Tegumentary Leishmaniasis and Visceral Leishmaniasis, finding practically negligible reactivity. Daltro et al. () state that the use of chimeric parasitic antigens can reduce the cost of diagnosis, since there would be no need to repeat tests on the same sample, suggesting their use where Leishmania sp. and T. cruzi are coendemic.
There are hundreds of wild mammal species in which T. cruzi is capable of infecting and sampling, and diagnosing infection in free-living wild fauna is still a huge challenge. One of the biggest obstacles is the lack of diagnostic improvement. As a result of the anthropocentric view of health, only products for serological diagnosis of humans and animals of economic interest are commercially found. Thus, the potential to perform the diagnosis through Protein A or G linked to colloidal gold in a rapid test will multiply the application, since a specific antibody is not necessary to reveal the antigen-antibody reactions as in conventional serological tests. It will bring benefits mainly for the diagnosis of wild mammal fauna.
Contact us to discuss your requirements of Dengue Combo Rapid Test Cassette. Our experienced sales team can help you identify the options that best suit your needs.
We propose here the spatiotemporal monitoring of infection in domestic dogs as an environmental diagnostic tool through Chagas-LFRT to be used as a first measure in the identification of areas where there is a potential risk of transmission of T. cruzi and the presence of the vector. The sole use of Chagas-LFRT can be used for an environmental diagnosis, while a second confirmatory test has to be employed for the individual diagnosis. Domestic dogs are animals that are easy to handle, in addition to having spatial and temporal traceability. The collection of blood from these animals does not require large infrastructure and cost, especially with the use of the rapid test for T. cruzi, a point-of-care technology that is fast and easy to handle. From this, Chagas-LFRT proved to be sensitive for use as a first environmental diagnostic tool for the presence of T. cruzi for early monitoring of the risk of new human cases.
Our main contribution was to validate and expand the use of the rapid Chagas test, which was developed in Bio-Manguinhos for the diagnosis of Chagas disease in humans (Silva et al., ), for field diagnosis of T. cruzi infection in domestic dogs and to evaluate its potential of application to wild canid species. Our motivation was to implement in the field work routine, a quick test that is easy to perform (point-of-care exam), which does not require technical training and is not dependent on a complex laboratory infrastructure for its execution, as in the conventional diagnostic tests available (ELISA and IFAT). One of the great advantages is to have one same type of test that can be used under the same conditions for the diagnosis of T. cruzi infection in humans, domestic and wild dogs.
The surveillance system as a whole will benefit because it will have in its hands a specific and reliable rapid test for “in loco” diagnosis of infected dogs and proximity or presence of infected triatomines, that is, the signalling of an expressive enzootic cycle of T. cruzi transmission near human dwellings. Chagas-LFRT will allow rapid diagnosis in a safer, reliable and low-cost manner without the need for more complex laboratory tests. The detection of “hot areas” of enzootic transmission detected by the test will streamline the decision-making process by allowing quick mapping of target areas for implementation of prevention measures.
The datasets presented in this article are readily available. Requests to access the datasets should be directed to Samanta Xavier,c2FtYW50YUBpb2MuZmlvY3J1ei5icg==.
The animal study was reviewed and approved by IBAMA (wild canids) and followed protocols approved by FIOCRUZ’s Animal Use Ethics Committee (P-99; P-03; P/06; L-07; L-050/; LW-81/12). Written informed consent was obtained from the owners for the participation of their animals in this study.
Conceptualization, SX, AJ, ES, and NR. Formal analysis, SX, ER, NR, AJ, and ES. Funding acquisition, SX. Investigation, SX, AJ, and AR. Methodology, SX, ER, GS, MS, JB, AB, RD, NR, ES, and AJ. Project administration, SX, NR, AJ, and ES. Resources, AB, RD, ES, and NR. Supervision, SX, ES, NR, and AJ. Visualization, SX, AJ, AR, NR, and ES. Writing original draft, ER, SX, JB, AJ, NR, and AR. Writing—review and editing, ER, SX, AJ, and AR. All authors have read and agreed to the published version of the manuscript.
SCCX has received financial support from CNPq (MCTIC/CNPq No. 28/ - Universal, process number /-2) and from Faperj ARC_ - Auxílio ao Pesquisador Recém-contratado (E-26/010./).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
The new TR Chagas development project, led by Bio-Manguinhos, is the result of the collaboration of several Fiocruz units, such as the team of Dr. Nilson Zanchin from the Carlos Chagas Institute (ICC/Fiocruz-PR), who developed the proteins and improved the processes of expression and purification, and the Ageu Magalhães Institute (IAM-Fiocruz-PE), Gonçalo Moniz Institute (IGM-Fiocruz-BA), Institute of Molecular Biology of Paraná (IBMP) and Oswaldo Cruz Institute (IOC-Fiocruz- RJ). We thank Dr. Julio Israel Fernandes and the staff of the Veterinary Hospital of Universidade Federal Rural do Rio de Janeiro (UFRRJ) for serum samples of dogs infected by Anaplasma platys, Dirofilaria immitis and Erlichia sp. We want to thank LABTRIP, Fundação Parque Zoológico de São Paulo and Programa de Conservação Mamíferos do Cerrado (PCMC), Brazil; Cristiane Varella for the COLTRYP isolate characterization and to the Scientific Initiation Scholars of the PIBITI Program of the Oswaldo Cruz Institute, Fiocruz: Matheus Santos Luquez, Victoria Brigido Lamim and Ana Carolina Ferreira de Carvalho for the contributions during this study.
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10./fcimb../full#supplementary-material
Supplementary Figure S1 | Receiver operating characteristic (ROC) curve, area under the curve (AUC) and statistical summary of cutoff point values (2.3, 3.7, 4.8, 5.9 and 8.7).
Chagas disease is caused by infection with the kinetoplastid parasite Trypanosoma cruzi (T. cruzi), generally transmitted by triatomine vectors that are endemic to the Americas. In recent decades, population movements from endemic to non-endemic regions and the chance of vector-independent transmission have globalized the disease impact to non-endemic countries like Spain [1].
According to WHO reports, there are 6–7 million people infected with the parasite worldwide and 25 million live in vector transmission risk areas [2]. Despite this gruelling data there are only two chemotherapies available, benznidazole and nifurtimox, and its indication is still a matter of debate [3,4]. Although highly efficacious against the acute stage, their efficacy diminishes the longer a person has been infected. Furthermore, the treatment is useless in the later stages of chronic cardiac complications, so as it can be concluded from the outcome of the BENEFIT clinical study that investigated the therapeutic capacity of benznidazole for treating chagasic patients with advanced heart damage [5]. Despite development of more efficacious and less toxic therapies is an urgent need of Chagas disease, nowadays early accessing to presently available drugs is yet a major issue. It is estimated that current chemotherapies barely reach 1% of the infected people [2,6], which is directly related to the stigma still surrounding this infection and the fact that, very usually, goes undiagnosed until it is too late.
A timely and accurate diagnosis remains a major obstacle to start treatment. Diagnosis is hampered by the disease own characteristics, depicted by its two distinct phases. Firstly, the short acute phase is mostly asymptomatic and as a consequence it is largely undiagnosed despite parasites can be detected in blood [7]. The long indeterminate stage that follows may span decades, throughout which parasite hides away and is not easily detected despite yielding a chronic infection [7]. Then, around 30–40% of those chronically infected will develop dreadful cardiac, digestive or cardiodigestive symptoms [7]. These cause massive tissue inflammation, leading to organ dysfunction and ultimately death if untreated. Thus, the earlier the diagnosis is achieved the sooner the treatment can be administered, and the more efficacious it would be.
At the chronic indeterminate stage, with scarce or no parasites circulating, diagnosis is performed serologically [8]. The most frequently used serological tests are enzyme linked immunosorbent assays (ELISAs), either based on whole parasite lysate or recombinant proteins as antigens; indirect immunofluorescence detection (IIF); and indirect hemagglutination assay (IHA). However, resources and personnel required for the performance of those tests are limited or directly non-existent in distant regions. In order to circumvent this limitation, rapid diagnostic tests (RDTs) were developed for Chagas disease detection [9]. These are easy-to-use diagnostic tools that usually provide a result read-out within an hour. They do not require electrical equipment or trained personnel for its working and are storable and functional at room temperature [9,10]. All of them fundamental features towards their implementation in remote regions, where access to/by health providers is difficult and electrical supply can be unstable or inexistent.
Amongst the American countries where Chagas disease is endemic, Bolivia registers the highest seroprevalence rate (6.1%; [11]). In Bolivia, Chagas disease conventional diagnosis is routinely made in laboratories located in larger urban sites where required facilities are available. However, most patients live in rural or peri-urban areas where there is not access to equipped labs, nor is available the trained personnel to perform and read the tests. Currently, upon guidelines established by the Chagas National Program, serodiagnosis of the disease is made first by rapid screening with the RDT Chagas Stat-Pak (CST; Chembio Inc., Medford, USA). Positive detections are then confirmed by conventional serological tests: either by ELISA based on parasite lysate antigen or by IHA [12]. Only upon confirmation of the infection by conventional serology the treatment can be started. If results from the RDT and conventional serology disagree, a third serological assay, an ELISA based on recombinant antigen, is used to tip the scales [12]. Conventional tests provide high sensitivity and specificity, but they are time-consuming, involve multiple steps, need trained personnel to perform them, and require at least two visits of the patient to the health centre to get diagnosed. Besides, conventional Chagas diagnosis demands resources (medical capacity, laboratory equipment, blood banks, cold storage) that are scarce in rural areas where the disease has a larger prevalence. There, diagnosis by RDTs can make the difference (if a diagnostic confirmation by a conventional serology was not needed to initiate therapy).
In the present study we hypothesize that implementing the synchronous use of two RDTs based on distinct antigenic sets would suffice to deliver a definitive diagnosis in a short time, accelerating patients´ access to treatment, thus resulting in a better adherence to it. With that objective, we evaluated in parallel the performance of two RDTs: CST (Chembio Inc., Medford, USA; [13]) and Chagas Detect Plus (CDP; InBios Inc., Seattle, USA; [14]). They were compared with routinely used serological tests (IHA, lysate-antigen ELISA, and recombinant-antigen ELISA).
Enrollment for this prospective study was offered to all patients ≥1 and <60 years old that attended the Chuquisaca Departmental Reference Laboratory in Sucre, either under medical prescription or by personal initiative to obtain a Chagas disease diagnosis. Patients that attended the Platform for Integral Attention to the Chagas Patient in Sucre were also offered enrollment. In any case patients came from Sucre urban area which has a population of 312,024 people (152,238 men and 159,785 women). The department of Chuquisaca, of which Sucre is the capital, is populated with 413,823 people (203,607 men and 210,216 women). None of the patients enrolled had received anti-Chagas treatment before. Upon signature of the informed consent, patients voluntarily agreed to participate in the study selflessly. In total, 342 samples were collected (223 women and 119 men). Sampling was made between March and May . Whole blood for the RDTs was obtained by fingertip puncture. At the same visit a larger blood volume was also obtained by venous extraction. Extracted blood was placed in 5 ml collection tubes without anti-coagulants and let to coagulate for 10 min at room temperature. Tubes were centrifuged 10 min at 3,000 rpm to segregate the sera which was then placed into appropriately labelled tubes. Serum samples were either used immediately or stored frozen until needed.
The study was a double-blind with three different observers involved: one for the CST, other for the CDP, and a third one for the conventional serological tests (IHA, ELISA based on lysate antigen, and ELISA based on recombinant antigen). None of the observers knew the results obtained by the others at any time. All tests (two RDTs and the three conventional serological assays) were assessed over the entire population enrolled (age range ≥1 and <60 years old; n = 342). Remarkably, RDTs used whole blood as sample whereas conventional tests relied on serum samples.
Study was reviewed and approved by the Ethical Committees of “Fundación Ciencia y Estudios Aplicados para el Desarrollo en Salud y Medio Ambiente” (CEADES) and the Department of Health Service of Chuquisaca (SEDES Chuquisaca). Similarly, the protocol was approved by the University Andina Simon Bolivar (Sucre, Chuquisaca, Bolivia), and it was sent to both the Departmental and National Chagas Programs for its acknowledgment and approval. All patients included in the study signed an informed consent. In those cases where patients were under 18 years old (legal age), inform consent was signed by the mother, father or assigned tutor.
Tests used in the study are those routinely performed by the National Chagas Program and by the Platform for Integral Attention to the Chagas Patient. ELISA tests were from Wiener Lab (Rosario, Argentina) and included the parasite lysate-based ELISA (Wiener v2.0) and the recombinant antigen-based ELISA (Wiener v3.0). Tests were performed following the manufacturer´s instructions.
Regarding IHA test, Chagas Polychaco kit from Lemos Laboratorios (Buenos Aires, Argentina) was used. Briefly, assay functionality is based on the ability of anti-T. cruzi immunoglobulins to induce hemagglutination of parasite pre-sensitized sheep red blood cells. Samples were tested in a 2-fold dilution range from 1/16 up to 1/. Sera 1/8 dilutions were incubated with non-sensitized sheep red cells as a control for cross-reactivity with heterologous antigens. Samples that induced a hemagglutination mesh over sensitized red cells at dilutions ≥1/16 were considered positive, as far as they did not cross-react with non-sensitized red blood cells at 1/8. If cross-reactivity occurred, protocol was repeated upon treating the samples with 2-mercaptoethanol following the manufacturer´s instructions.
All individuals confirmed to be positive were offered benznidazole treatment by the medical officer in the Platform for Integral Attention to the Chagas Patient in Sucre or at the Departmental Health Service if they fulfilled treatment criteria.
Selection of RDTs for the study was based on their ability to allow diagnosis of whole blood samples, generate results within an hour, not require instrumentation for results readout, have long shelf life at room temperature, and be based on distinct sets of parasite antigens. CST (Chembio Inc.) is in use as standard tool for Chagas disease screening by the Chagas National Program since . It is based on the detection of IgG specific for the T. cruzi antigens B13, 1F8 and H49/JL7 (the latter allowing the detection of IgG antibodies characteristic of the disease chronic stage; [15]). Despite it was not commercially available at Bolivia at the study time, CDP (InBios Inc.) was chosen due to its amenability to be used with whole blood samples, the fact that barely 5 μl of it were required for a viable reading, and because it has a result turnaround of about 25 min [14]. It is based on a multi-epitope antigen (ITC8.2; [16]) which contain differs in origin from that of CST.
RDTs were made immediately upon blood collection with the second blood drop attaching to manufacturers´ instructions.
The study variables considered were: (i) sensitivity, defined as the percentage of positive patients correctly diagnosed in comparison with the standard; (ii) specificity, defined as the percentage of negative patients correctly diagnosed in comparison with the standard; (iii) positive predictive value, the proportion of positive patients that were really positive; (iv) negative predictive value, the proportion of negative patients that were really so; and (v) diagnostic efficiency (i.e. accuracy), the proportion of valid results amongst the whole of them, defined therefore as the proportion of individuals correctly classified. Along the study, the standards to compare with were respectively the ELISA based on recombinant antigen (Wiener ELISA v3.0), and the coordinated results from all three conventional serological tests used.
Data from each participant in the study were entered in Microsoft Excel and analyzed with the freely available software EPIDAT (version 3.1; Sergas, Santiago de Compostela, Spain; available at http://www.sergas.es/Saude-publica/EPIDAT). The percentage of concordance between methodologies was determined with the Kappa coefficient (κ) that compares the agreement between actual and expected results, following the criteria scale of Landis and Koch [17]. In the present study, for a RDT to be considered reliable, its κ must be >0.8. A κ value >0.8 and ≤0.9 indicated a very good concordance, whereas [0.9> κ ≤1.0] was considered excellent. In all cases, the confidence interval (CI) was set at 95% and its values are provided in parenthesis where corresponding. The Z-test was used to determine the statistical level of significance at each case (p <0.05). Youden´s index was calculated to determine the diagnostic efficacy (accuracy) of the RDTs combination in comparison with the conventional serological tests.
Screening studies to validate CST and CDP have reported varied results. From the 100% agreement between ELISA and CST on almost 6,000 serum samples from Central American blood donors and Chagas patients [15], to CST 87% sensitivity and 93.2% specificity reported by Sanchez-Camargo et al. when testing sera from nine geographically apart locations [10]. In that very same study, CDP outperformed CST as it registered 93.4% sensitivity and 95.2% specificity [10]. However, CDP had averaged 84.8% sensitivity and 97.3% specificity on sera from four distinct countries (Argentina, Ecuador, Mexico and Venezuela) in another work [18]. To date, the only study that has compared CST and CDP on the same samples set, respectively determined 87.5% and 90.7% sensitivity in Bolivian sera, whilst CST and CDP sensitivity on sera from Peru was much lower (respectively, 26.6–33% and 54.3–55.2%, depending of the observer; [19]). As suggested elsewhere, it could be that in regions with lower prevalence rates, with expected lower generalized antibody titers, implementation of RDTs is not feasible at all due to very low sensitivity [20]. In both sites specificity was >98%, thus pointing at false positive detections as a cause of discordance too, which could be due to cross-reactivity with other infections, like leishmaniasis.
CST and CDP are functional as well with whole blood as sample [10,13,14], which represents a major advantage over sera as it reduces the required laboratory procedures. Besides, blood volume required for the tests is so little (10 and 5 μl respectively) that sampling can be made by digital puncture, which largely eases up the process and simplifies the screening of children. CST box includes as many Microsafe tubes (10 μL) for finger-stick whole blood acquisition as test devices, whereas CDP kit does not include them [13,14].
The first field evaluation of CST with whole blood samples (n = 1,913; all under 18 years old) from Bolivia (Sucre region) reported 93.4% sensitivity and 99% specificity (98.6% concordance between serum and whole blood results; [21]). With that sensitivity authors concluded that CST was not ideal as diagnostic tool, but claimed it was improved to 96.7% if test reading time was made at 20 min instead of the 15 min instructed by the manufacturer. We have not found such differences in our study. Considering the results of the conventional serological assays Roddy et al. found a seroprevalence rate of 6.3% (121/) [21]. In our case, testing a much smaller collection of patients, seroprevalence amongst those <15 years old was 11.1% (5/45). Another CST study on n = 874 blood samples from Latin American migrants in Switzerland (>16 years old; most of the infected patients were from Bolivia) determined a 95.2% sensitivity and 99.9% specificity (99.7% concordance between serum and blood samples; [22]). In agreement with that work, we have also found an excellent sensitivity (100%), a very high specificity (99.3%), and the same level of concordance between serum and blood usage with CST.
InBios CDP field evaluation with blood was also made with Bolivian samples, this time from Santa Cruz province (n = 108 adults), and from the Chaco, including pregnant women (n = 277) and children up to 17 years old (n = 200) [20]. Overall, compared to conventional serology CDP sensitivity was 96.2% and its specificity was 98.8% (95.3% serum and blood concordance; κ = 0.905). In our study, CDP also rendered excellent sensitivity and specificity (respectively, 100% and 99.3%), and the concordance between serum and blood was 99.7% (341/342; κ = 0.99). The higher seroprevalence detected by us in women (64.6%; 144/223) compared to Shah and co-workers congenital study branch (48.4%; [20]) could be due to the fact that women enrolled in their sub-study were younger in average (15–42 years old). However, within the inclusive age range between 15 and 42 years of age in our study (56.5% of women; 126/223) seroprevalence retrieved by us is still quite high (67.5%; 87/126). Thus, such difference with the previous work might be related to the locations studied and an overall higher seroprevalence in Sucre (Chuquisaca) compared to Camiri (Cordillera province, Chaco region; [20]).
Definitely, geographic variation, directly related to distinct seroprevalence rates and circulating parasite strains, is of major relevance towards RDTs implementation in the field. Therefore, in full agreement with others, RDTs must be field validated in each particular site before its use for screening is recommended [19,20]. Seroprevalence detected in this work was quite high (60.8%), but accordingly to the Bolivian government Chagas Management Departmental Program (with data from ), Chagas disease seroprevalence in the department of Chuquisaca was estimated to be 5.6% amongst children between 1–4 years old, 5.4% in the age range from 5 to 14 years old, and 56.6% in those ≥15 years old [23]. This is above the country´s mean (6.1%; [11]), illustrating a heterogeneous distribution of Chagas disease in the country. Although the main target population for these diagnostics are indeterminate patients, it would be interesting to address whether their performance varies in relation to the indeterminate or symptomatic stage of chronic Chagas' disease.
In the case of Bolivia, results from this study and others with Bolivian blood samples, depict CST and CDP as fast, reliable and accurate tools [19,20,22], thus bringing the chance to incorporate them in the diagnostic protocols (CST is already at the frontline for Chagas disease serological screening in Bolivia, both in blood bank screening and clinical diagnosis) [12]. However, despite RDT permeation in the health system, conventional serological diagnosis is yet necessary for infection confirmation [12,21]. This requirement for a definitive conventional diagnosis imposes a treatment start delay that can hinder patients´ adhesion to it. In the Department of Chuquisaca, most endemic municipalities present a high prevalence of T. cruzi infection and have problems to access lab services due to bad communications and lack of resources. Labs in the region are small, ill equipped, and have very limited resources. Nonetheless they are obliged by law to include Chagas disease diagnosis for the population that demands the service [24]. Taking into account that CST and CDP have shown to perform excellently in the criteria scale of Landis and Koch [17], and that they are based on distinct antigens sets, the use of the duo to substitute currently used conventional serology would be justified. Since same results were retrieved with both RDTs, it would have been very interesting to have considered other features for comparison between them (i.e. strength of signal, ease of use). However, due to the blind characteristics of the study, with a distinct observer in charge of each RDT and another one in charge of conventional serological tests, no further comparison criteria were included. Accordingly to a recent article by Angheben et al., a blind comparison of RDTs and conventional serological assays (gold-standard) should be made in a consecutive series of patients from a relevant clinical population for optimal assessment of their accuracy [25]. In view of the obtained results, the use of CST and CDP in distant regions stands as an alternative to conventional methodologies. It must be noted though that RDTs have been previously shown to lack appropriate sensitivity in regions where seroprevalence is lower than in Bolivia [19]. Since seroprevalence rates also vary between distinct regions of the country, it is inevitable to think that RDTs would perform worse than what we report here if the study had been carried out in other regions. Thus, either region-specific pilot studies or a larger study at national level including samples from geographically heterogeneous origin should be launched before fully adopting this approach.
Other issue could be RDTs´ price, which is in principle disadvantageous to that of conventional assays. In Bolivia, RDT costs range from 4$ to 7$ (US dollars) per single determination, whilst the average cost of a conventional test single readout is 1$. Nonetheless, if cost of logistics of the conventional techniques is accounted for, then their real cost is to be much higher. Despite it did not occur in this study it cannot be discarded that divergent results between both RDTs are observed. Then, a third test would still be needed for confirmation of the inconclusive outcome. Although this would increase costs, notably it should only be performed on a little percentage of expected inconclusive results and it would reasonably yet be cost-effective if conventional serology logistical costs are considered.
The retrieved results of this study validate the use of the CST and CDP diagnostic duo to reduce response time to patients and the logistic resources currently needed to elaborate a confirmed Chagas disease diagnosis. Therefore, we encourage considering the implementation of both RDTs with whole blood as sample to provide fast, reliable and definite diagnosis of Chagas disease in distant settings.