Serological profiles and evaluation of parasitaemia by PCR and blood culture in individuals chronically infected by Trypanosoma cruzi treated with benzonidazole


Corresponding Author: Sandra C. B. Costa, Universidade Estadual de Campinas, Faculdade de Ciências Médicas, PO Box 6111, 13083-970, Campinas, SP, Brazil. Tel.: +55 19 35219215; Fax: +55 19 32894107; E-mail:


Objective  To evaluate the serological and parasitological status of patients with chronic Chagas disease (CD) after chemotherapy with benzonidazole.

Methods  Retrospective study of patients treated with benzonidazole (5 mg/kg/day for 60 days) between 1980 and 2010. Twenty-nine patients who had CD confirmed by two reagent immunological tests and/or one positive xenodiagnosis before treatment were included. Conventional serology (ELISA and IIF) and parasitological tests (haemoculture and N-PCR) were performed.

Results  At the time of treatment, the mean age of patients was 36 ± 7.24 years (20–39 years) and the time post-treatment varied from 1 to 29 years. After chemotherapy, all individuals had reagent ELISA and 93.1% had positive results for the IIF test. T. cruzi DNA was detected by N-PCR in 48.3%. Negative results were observed in 41.4% and inconclusive ones in 10.3%. Haemoculture was negative for all individuals.

Conclusions  Our results suggest that N-PCR may be useful in the early identification of therapeutic failure of CD. Although it is difficult to determine parasitological cure in negative N-PCR cases, we can infer that this condition represents a declination of parasitaemia as a favourable consequence of aetiological treatment.


Objectif:  Evaluer le statut sérologique et parasitologique des patients atteints de la maladie de Chagas chronique après la chimiothérapie au benzonidazole.

Méthodes:  Etude rétrospective de patients traités au benzonidazole (5 mg/kg/jour pendant 60 jours) entre 1980 et 2010. Vingt-neuf patients qui ont eu la maladie de Chagas confirmée par deux tests immunologiques et/ou un xénodiagnostic positif avant le traitement ont été inclus. La sérologie (ELISA et IIF) conventionnelle et des tests parasitologiques (hémoculture et N-PCR) ont été réalisés.

Résultats:  Au moment du traitement, l’âge moyen des patients était de 36 ± 7.24 ans [20–39 ans] et la durée post-traitement variait de 1 à 29 ans. Après la chimiothérapie, tous les individus avaient une ELISA réactive et 93.1% avaient des résultats positifs pour le test IIF. L’ADN de T. cruzi a été détecté par N-PCR chez 48.3% d’eux. Des résultats négatifs ont été observés chez 41.4% et des résultats non concluants chez 10.3%. L’hémoculture était négative pour tous les individus.

Conclusions:  Nos résultats suggèrent que la N-PCR peut être utile dans l’identification précoce de l’échec thérapeutique de la maladie de Chagas. Bien qu’il soit difficile de déterminer la guérison parasitologique pour les cas N-PCR négatifs, nous pouvons en déduire que cette condition représente une déclinaison de la parasitémie comme une conséquence favorable du traitement étiologique.


Objetivo:  Evaluar el estatus serológico y parasitológico de pacientes con enfermedad de Chagas crónica, después de recibir quimioterapia con benzonidazol.

Métodos:  Estudio retrospectivo de pacientes tratados con benzonidazol (5 mg/kg/día durante 60 días) entre 1980 y 2010. Se incluyeron veintinueve pacientes que tenían enfermedad de Chagas confirmada mediante dos pruebas inmunológicas reactivas y/o un xenodiagnóstico positivo antes del tratamiento. Se les realizó una serología convencional (ELISA e IIF) y pruebas parasitológicas (hemocultivo y PCR anidada).

Resultados:  En el momento de recibir el tratamiento, la edad media de los pacientes era de 36 ± 7.24 años (20–39 años) y el tiempo post-tratamiento varió entre 1 y 29 años. Tras la quimioterapia, todos los individuos tenían una ELISA reactiva y un 93.1% tenía un resultado positivo para la prueba IIF. Se detectó ADN de T. cruzi mediante una PCR anidada en un 48.3%. Se observaron resultados negativos en un 41.4% y resultados inconclusos en un 10.3%. Los hemocultivos eran negativos para todos los individuos.

Conclusiones:  Nuestros resultados sugieren que la PCR anidada puede ser útil en la identificación temprana del fallo terapéutico de la enfermedad de Chagas. Aunque es difícil determinar la cura parasitológica en casos con resultados negativos de la PCR anidada, podemos inferir que esta condición representa un declive de la parasitemia, como una consecuencia favorable del tratamiento etiológico.


Chagas disease (CD) is a parasitological infection caused by Trypanosoma cruzi protozoa that remains an important public health problem even 101 years after its description (Chagas 1909). An estimated 8–9 million people around the world are affected, especially in Latin America, where the vectorial transmission by triatomine insects causes 40 000 new cases annually (PAHO 2009). Other important routes of infection are blood transfusion, organ transplantation, congenital transmission and oral contamination (Moncayo & Ortiz Yanine 2006). The acute phase is asymptomatic in most cases or can present mild and non-specific clinical symptoms. In about one-third of all cases, the chronic form is characterized by lesions on heart, oesophagus and colon tissues, with severe disorders of nerve conduction of these organs (Urbina 1999). In the chronic stage, intermittent parasitaemia is common and antibodies (IgG) against T. cruzi are frequently detected by conventional immunological tests. Xenodiagnosis and haemoculture have limited sensitivity at this stage because of the lack of parasites in blood and are not useful in diagnosis (Portela-Lindoso & Shikanai-Yasuda 2003).

Two mechanisms explain the pathogenesis of CD: the hypothesis of autoimmunity and the parasite persistence hypothesis. The autoimmunity hypothesis prevailed for a long time based on the apparent absence of parasites in damaged tissues of chronic patients. This observation suggests that T. cruzi infection induces inflammatory responses to host tissues and does not depend on the parasite persistence. However, recent studies using molecular assay-based methods have reported the presence of parasites in damaged tissues of cardiac and digestive chronic patients (Vago et al. 1996; Marcon et al. 2011), which reinforces the parasite persistence hypothesis. CD lacks effective treatment in all its phases and no immunoprophylaxis is available. The drugs available are nifurtimox (Lampit®) and benzonidazole (Rochagan®). Both drugs must be administered orally; nifurtimox (8–10 mg/kg/day) for at least 30 days and benzonidazole (5–7 mg/kg/day) for at east 60 days, to eradicate the parasites.

In the acute stage, these drugs are strongly recommended with an incidence rate of cure about 70%. In the chronic phase, however, there is a disagreement about therapy utility because of allergic reactions, low cure rates and absence of criteria of cure (Marin-Neto et al. 2009). Side effects could limit the use of these drugs in some cases, mainly nifurtimox, which is not available in some countries such as Brazil because of its toxicity. Late chronic patients could be treated to prevent or reduce the evolution of lesions, with a parasitological cure estimated in 10–20% according to different analyses. In addition to the lower efficacy of aetiological treatment, the lack of safe methods for cure evaluation compounds chemotherapy invalidation (Coura & Castro 2002).

In the acute stage, conventional serological and parasitological tests become negative after 1 year of chemotherapy and could be useful to indicate success of treatment. In the chronic stage, the persistence of positive results of immunological assays after drug administration is frequent and negative results, when they occur, can take years to appear, requiring very long follow-up periods (Flores-Chavez et al. 2006). Even decreased titres of T. cruzi antibodies do not indicate a process of cure, because they may be part of the late evolution of CD (Prata 2001).

Xenodiagnosis and haemoculture, because of their low sensitivity, could present false-negative results (Britto et al. 1999). A recent alternative approach to monitor post-treatment of T. cruzi infection are polymerase chain reaction (PCR)-based strategies, because of their high sensitivity to parasite DNA in blood samples of chronic patients, in comparison with other parasitological tests, even in cases of doubtful or negative serology (Gomes et al. 1999; Marcon et al. 2002; Batista et al. 2010). To evaluate PCR as a laboratory tool for the establishment of parasite clearance, this retrospective study aimed to compare the serological and parasitological results of individuals who were treated with benzonidazole in chronic phase of CD, using three diagnostic methods: conventional serology, haemoculture and nested-PCR (N-PCR).



This retrospective study to evaluate the presence of parasites in blood of individuals treated with benzonidazole for T. cruzi infection at Clinical Hospital of the University of Campinas, from 1980 to 2010 was approved by the Institutional Ethical Committee in Human Research. A total of 135 clinical records were consulted and we identified 68 cases submitted to specific treatment. From this group, 14 persons did not conclude the recommended treatment period of 60 days (5–7 mg/kg/day) and were excluded from the study. Considering the remaining population, 29 individuals were recruited and gave written consent to participate. All these participants had CD confirmed by two of three positive immunological tests (Machado-Guerreiro, IIF or ELISA) and/or one positive xenodiagnosis before treatment (except in two cases where they came from another institution with positive diagnostic but without the test data). To certify the sensitivity of laboratorial tests, 10 individuals chronically infected (with two positive serologies), who did not receive treatment, were included in the control group.

Serological assays

All patients included in this study were tested by two serological assays (ELISA and IIF) after specific chemotherapy. Serological assays were performed at the Laboratory of Immunology in the Department of Clinical Pathology and the results were obtained through electronic data system of the Clinical Hospital of University of Campinas. The tests used were commercial kits for ELISA (BIOZIMA®) and IIF (Biolab®).


Haemoculture was performed according to methods described by Chiari et al. (1989) and Luz (1999), with modifications. We collected 12 ml of peripheral blood of each patient in three EDTA tubes. Immediately after, the material was centrifuged at 1900 g, during 10 min (4 °C), to discard plasma. The blood cells sediment was transferred to another tube and washed in LIT (liver infusion tryptose) medium, supplemented with FBS. The samples were centrifuged under the same conditions and supernatant was discarded. The blood cells were resuspended in 5 ml of LIT and incubated at 28 °C. The period of culture was 90–120 days, with monthly microscopic observation.

DNA isolation of blood samples

Four millilitres of peripheral blood were collected from each patient in EDTA tubes to prevent coagulation. Genomic DNA was isolated according to methodology previously described by Sambrook and Russel (2001), with modifications. After centrifugation at 800 g during 10 min (4 °C), plasma was immediately separated from blood cells. Red blood cells were treated with a lysis buffer (0.0114 m NH4Cl and 0.01 m NH4HCO3) and leucocytes were washed with TKM1 (10 mm Tris–HCl pH 7.6, 10 mm KCl, 10 mm MgCl2, 20 mm EDTA) and with TKM2 (10 mm Tris–HCl pH 7.6, 10 mm KCl, 0.4 m NaCl, 10 mm MgCl2, 2 mm EDTA) and sodium dodecyl sulphate 10%. The samples were incubated at 55 °C for 30 min and NaCl (5 m) was added to the supernatant after centrifugation. DNA was extracted with phenol/chloroform and precipitated with 3 m ethanol/sodium acetate. After overnight storage at −20 °C, DNA was washed with 70% ethanol and solubilized in 30 ml of deionized and sterile water.

N-PCR assays

One microlitre of extracted DNA was amplified at least in duplicate through N-PCR based on a procedure described by Marcon et al. (2002). DNA amplification was carried out in a 20 ml reaction mixture containing 10 mm Tris (pH 8.3), 50 mm KCl, 1.5 mm MgCl2, 0.2 mm of each dNTP, 0.1 mm of each primer, and 1 U of TaqDNA polymerase. The primers used in first amplification set were TCZ1 (forward 5′–CGAGCTCTTGCCCACACGGGTGCT-3′) and TCZ2 (reverse 5′-CCTCCAAGCAGCGGATAGTTCAGG-3′). In the second round, 1 μl of amplified DNA was re-amplified using primers TCZ3 (forward 5′-TGCTGCA(G/C)TCGGCTGATCGTT-TTCGA-3′) and TCZ4 (reverse 5′-CA(A/G)G(C/G)TTGTTTGGTGTCCAGTGTGTGA-3′). Annealing temperatures were 65 °C and 55 °C for the first and second rounds, respectively. After the second thermal cycling, 10 μl of each sample was electrophoresed in a 2% agarose gel stained with ethidium bromide and the 149-bp amplicon was visualized by ultraviolet transillumination. In addition, the beta-globin human gene was amplified for each sample to check DNA quality and the absence of inhibitors. To minimize the risk of contamination, first and second runs were set up in distinct laboratory areas and a reaction mixture with no DNA was processed in parallel with the samples in each reaction set.


We studied a group of 29 individuals who received anti-T. cruzi specific therapy, composed of 51.7% (15/29) women and 48.3% (14/29) men. All patients were born in endemic areas for CD in Brazil, except in one case of possible congenital infection. The mean age at the time of treatment was 36 ± 7.24 years old (8–56 years). The time post-treatment varied from 1 to 29 years and most of analysed subjects were followed up for 5 years (62.1%), therefore 27.6% had 5–10 years post-treatment and 10.3% more than 10 years (Table 1). All patients completed treatment without serious reactions, despite side effects in 20.7%. The symptoms reported were as follows: dermatitis (two cases), gastrointestinal disorders (two cases), paraesthesia (one case) and one case of generalized indisposition.

Table 1.   Epidemiological, clinical and laboratorial data of Chagas disease patients before and after treatment with benzonidazole
PatientSexAgeLaboratorial tests before treatmentTime post-treatment (years)Laboratorial tests after treatment
  1. MG, Machado-Guerreiro test; ELISA, enzyme-linked immunosorbent assay; IIF, indirect immunofluorescence; XN, xenodiagnosis; HC, haemoculture; N-PCR, nested-polymerase chain reaction; R, reagent; P, positive; N, negative; I, inconclusive; –, not informed.


Only 12 patients (41.4%) presented available data of xenodiagnosis before chemotherapy, with seven cases of positive results. Of these cases, four subjects repeated xenodiagnosis after treatment, with three negative and one persisting positive result (Table 1).

Before chemotherapy, 89.65% had available data for serology, three with IIF by Machado-Guerreiro test (titres 1:4, 1:8 and 1:32) and 88.5% with IIF by ELISA. The antibody titration by IIF varied from 1:40 to 1:1280. The titre of 1:40 was found in 7.7%, 1:80 in 23.1%, 1:160 in 34.6%, 1:320 in 19.2%, 1:640 in 7.7% and 1:1280 in 7.7%. After chemotherapy, 100.0% (29/29) of the cases were reagent to ELISA. According to IIF, two individuals (6.9%) showed inconclusive results with titres under the 1:40 cut-off. The rest of the group (27/29) had positive results with titres between 1:40 and 1:1280. IIF titres found were: 1:40 in 17.2%, 1:80 in 13.8%, 1:160 in 31.0%, 1:320 in 20.7%, 1:640 in 6.9% and 1:1280 in 3.5%, respectively. 61.5% of IIF titres dropped after treatment, 11.6% remained the same and 26.9% rose (Table 1).

All samples showed positive results for beta-globin human gene. The N-PCR detected T. cruzi DNA in 48.3% of the samples and parasite DNA was not amplified in 41.4%. Three cases (10.3%) were considered inconclusive, with discordant results (one positive result without concordance with repetitive reactions). None of the 29 collected samples were positive for haemoculture. The comparison of immunological and parasitological results after chemotherapy is summarized in Table 2.

Table 2.   Comparison of serological and parasitological results of experimental group after aetiological treatment for Chagas disease with benzonidazole
n = 29ELISA (%)IIF (%)HC (%)N-PCR (%)
  1. ELISA, enzyme-linked immunosorbent assay; IIF, indirect immunofluorescence; HC, haemoculture; N-PCR, nested-polymerase chain reaction.

Positive results29/29 (100)27/29 (93.1)0 (0)14/29 (48.3)
Negative results0 (0)0 (0)29/29 (100)12/29 (41.4)
Inconclusive results0 (0)2/29 (6.9)0 (0)3/29 (10.3)

In the control group (non-treated), the mean age was 60 ± 3.68 years old and all individuals reacted to ELISA and IIF tests, with titration varying from 1:40 to 1:1280. According to N-PCR assays, 70.0% (7/10) were positive, 20.0% (2/10) negative and one case (10.0%) was inconclusive. The haemoculture was positive in 40.0% (4/10) of the cases and the parasites were observed in a mean time of 63.5 (37–85) days after blood culture.


Our study compared two parasitological assays (N-PCR and haemoculture) with conventional immunological assays (ELISA and IIF) as tools for the investigation of chemotherapy effectiveness in a cohort of 29 individuals infected by T. cruzi who received benzonidazole treatment. N-PCR evidenced therapeutic failure in 48.3% of the cases, and therefore, parasite DNA was not found in 51.7% of the samples (negative and inconclusive results). Other reports based on PCR analysis demonstrated unsuccessful specific eradication of T. cruzi at a variable rate of 6.9–88.7% positive results for chronic patients (Britto et al. 1999; Lacunza et al. 2006; Fernandes et al. 2009; Murcia et al. 2010). This discrepancy is related to many factors, such as CD stage, PCR protocols and number of patients studied. With a similar cohort of 28 individuals, Lana et al. (2009) found 85.2% positive PCR and 7.1% haemoculture results. A possible explanation could be that these authors collected two samples of each patient in an interval of 1 year, increasing the possibility of reaching T. cruzi DNA in bloodstream, because parasitaemia in the chronic phase is intermittent. The 100.0% of negative blood culture indicates the low efficacy of this technique to monitor cure of CD, as demonstrated by Fernandes et al. (2009). In the control group, 70.0% of the individuals presented positive results by haemoculture, reinforcing the lack of sensibility in chronic patients even without specific treatment.

The immunological assays showed that all patients remained with positive or inconclusive results with no seronegativation; these findings corroborate other reports where post-treatment serology was superior to 90.0% (Solari et al. 2001; Flores-Chavez et al. 2006; Murcia et al. 2010). According to Brazilian Consensus in CD (Ministério da Saúde 2005), three consecutive and persistent drop off of immunological tests dilution can be suggestive of cure; however, in this study, two cases with decreased titres presented positive PCR. As the same manner as occurs with parasitaemia, the antibody titres oscillate during the period of infection and the serology titration does not necessarily correspond to the number of parasites circulating. This also can be verified in the cases where titres increased even after antitrypanocidal chemotherapy.

These results demonstrate the great difficulty of determining the success of specific treatment for CD, as from an immunological standpoint, all patients analysed could be considered non-cured (positive N-PCR in 48.3% of the cases). Considering the other cases, it is impossible to confirm that they are parasitologically cured, because the negative results of parasitological tests do not guarantee T. cruzi clearance. As suppressive activity of benzonidazole on parasitaemia is expected (Coura & Castro 2002), we could suggest that T. cruzi levels in blood was, at least, reduced by drug administration. The possibility of re-infection after treatment cannot be discarded, especially in the old cases evaluated. However, these individuals have been followed since chemotherapy at an ambulatory located in a non-endemic area, which minimizes the possibility of vector contact.

Non-severe side effects of chemotherapy were observed in 20.7% of the cases, as described before (Marin-Neto et al. 2009). Although undesirable, the adverse reactions are not necessarily an impediment to drug prescription when adequate follow-up is performed. Despite its limited usefulness in assessing cure in individuals treated for CD, aetiological treatment with benzonidazole in the chronic phase is recommended by several authors because there is evidence about its utility in preventing or minimizing lesions, improving the prognosis of these patients (Viotti et al. 1994; Gallerano & Sosa 2000; Fabbro et al. 2007).

In conclusion, our findings showed therapeutic failure in some cases, as evidenced by NPCR. However, we benzonidazole did decrease the number of parasites in the bloodstream, because parasitological assays were negative/inconclusive in most cases. Further studies including more patients, and systematic blood sampling and quantitative measurement of parasitaemia levels during and after chemotherapy could elucidate the true contribution of molecular assays as an efficient tool to evaluate T. cruzi clearance in CD treatment.


We thank the patients who agreed to participate of this work, laboratory colleagues and the Clinical Hospital staff. This work was supported by CAPES and FAEPEX.