Prevalence and risk factors for Toxoplasma gondii infection among pregnant and postpartum women attended at public healthcare facilities in the City of Niterói, State of Rio de Janeiro, Brazil

INTRODUCTION

Toxoplasmosis is a zoonosis caused by the obligate intracellular parasite Toxoplasma gondii, which infects warm-blooded animals, including human beings, and is found in almost every country, of various climates and social conditions, with variable levels in the percentage of positive tests ^1,2 . Congenital toxoplasmosis results mainly from primary maternal infection during pregnancy ³ , but can occasionally result from preconceptional infection or reactivation of latent infection in immunodeficient pregnant women, or even in immunocompetent women who were previously immune ⁴ . Transmission of T. gondii to the fetus can result in serious health problems including mental retardation, seizures, blindness and death. Some health problems may not become apparent until the second or third decade of life ^5–7 .

This infection is widespread around the world, in different countries and geographical areas within the same country and in different ethnic groups living in the same area ^8–14 . Thus, over the past three decades, antibodies anti-T. gondii have been detected in various adult human populations in proportions ranging from 9.1 to 97.4% of these individuals ^15,16 . In Brazil, studies on pregnant and/or postpartum women have indicated different prevalences in different states within the country ^17–24 , which adds to the importance of studying toxoplasmosis.

Risk factors for T. gondii infection are associated with hygiene, dietary and other cultural habits and climate ¹⁸ . According to infection routes and potential exposure, the most prevalent risk factors for toxoplasmosis are: deficiencies in environmental sanitation (untreated water), presence of domestic cats and rodents, consumption of potentially contaminated food (raw or undercooked meat and insufficiently washed vegetables) and contact with contaminated soil ²⁵ .

The aims of the present study were to determine the prevalence of immunoglobulin G (IgG) and immunoglobulin M (IgM) anti-T. gondii antibodies among pregnant and postpartum women attended within the public healthcare system in Niterói, Rio de Janeiro, and to detect possible exposure factors associated with T. gondii infection in this population.

METHODS

A cross-sectional study was carried out among 276 pregnant and 124 postpartum women who received prenatal care between September 2010 and September 2011, through the public healthcare system in Niterói, State of Rio de Janeiro, in order to determine the prevalence of IgG and IgM anti-T. gondiiantibodies. The study was conducted in the maternity outpatient clinic at the Hospital Universitário Antonio Pedro (the university hospital) of Universidade Federal Fluminense (HUAP/UFF), Policlínica Comunitária Sergio Arouca (PCSA; a community polyclinic) and five Family Doctor Program units (Viradouro, Preventório I and II, Palacio and Souza Soares). Among the volunteers, there were 300 (75%) from HUAP/UFF, 67 (16.8%) from PCSA and 33 (8.2%) from the Family Doctor Program units. The Hospital Universitário Antônio Pedro is a referral hospital for high-risk pregnancy, and receives patients living in Niterói as well as in other neighboring cities, such as São Gonçalo, Maricá and Itaboraí, among others. The Family Doctor Program units and the Policlínica Comunitária Sergio Arouca provide prenatal care for low-risk patients who live in specified regions of Niterói. Thus, each unit has its own area of activity and number of registered families. The people who use these services are characterized mostly by low educational level, financial difficulties and absence of adequate sanitary conditions.

All the 400 patients filled out a free and informed consent statement. Serum samples were collected, and then the patients answered an epidemiological questionnaire and received information about the disease. These samples were stored at -20°C until they were assayed. The questionnaires were applied at face-to-face interviews by the researchers, who provided guidance about the proposed research and read out the consent form, which the interviewees then signed. The questions were answered orally by the interviewees and recorded by the researchers, so that the study participants would find it easier to understand the subject matter. After this stage, guidance on T. gondii infection was provided with special emphasis on the ways of preventing infection. Because the participant’s name was written on the questionnaire and the serological results were subsequently sent to each participant, the risk of interviewing the same person twice was eliminated.

IgG and IgM anti-T. gondii antibodies were analyzed by means of the indirect fluorescent antibody test (IFAT) with a cutoff point of 1:16; and ELISA (BioKit®, Barcelona, Spain), in which readings at or above the cutoff value were considered positive. The tests were performed at the Toxoplasmosis Laboratory (LABTOXO) of the Oswaldo Cruz Institute (IOC/FIOCRUZ). Participants were selected by convenience sampling. Postpartum women between the first and tenth day’s postpartum and pregnant women at any gestational age were eligible. On the other hand, women taking immunosuppressive drugs and those who did not agree to sign the consent statement were excluded. In cases of positive IgM samples, IgG avidity for T. gondii was tested at the Evandro Chagas Clinical Research Institute, Fiocruz. The test was performed on the fully automated VIDAS system (VIDAS Toxo-IgG Avidity®, BioMérieux, Marcy-l’Etoile, France). The results were interpreted as recommended by the manufacturer, as follows: < 4 IU ml-1, negative; 4-8 IU ml-1, borderline; > 8 IU ml-1, positive. The results were given to the physician responsible for the patients’ prenatal care, for follow-up purposes and so that treatment could be started if necessary. A simple ratio was used to estimate seropositivity for toxoplasmosis among pregnant and postpartum women. Univariate relationships between a diagnosis of toxoplasmosis and the factors of interest were studied using the chi-square test. All the significant variables were included in a multivariate analysis, in which a multiple logistic regression model was used. The results from the logistic regression were expressed as odds ratios (OR) and the corresponding 95% confidence interval (95%CI). Adjusted odds ratios were calculated and their statistical significance was evaluated in accordance with the Wald test. A significance level of 0.05 was used to denote statistical significance. The data were analyzed using the Statistical Package for the Social Sciences 10.0 (SPSS) computer software.

The copositivity and conegativity rates and the concordance of the tests were determined as described by Teva, Fernandez and Silva ²⁶ . The Kappa index was used to measure the real agreement between the serological techniques.

RESULTS

The Kappa value for concordance between the two techniques for detection of IgG against T. gondii was 0.89, which reflects almost perfect agreement between the two techniques, as described by Teva, Fernandez and Silva ²⁶ . Furthermore, for detection of IgM against T. gondii, the Kappa was 0.2 which is considered to be a fair level of agreement.

In the data analysis, samples that were reactive in the IFAT and/or ELISA test were taken into consideration, thus totaling 234 (58.5%) samples that were reactive to IgG anti-T. gondii antibodies, and 166 (41.5%) samples that were susceptible to toxoplasmosis. Among the 234 positive samples, 161 (68.8%) were from pregnant and 73 (31.2%) from postpartum women. In the present study, we found no statistically significant difference (p=0.919) in relation to seroprevalence among pregnant and postpartum women. Seventeen (4.2%) samples were found to be reactive to IgM anti-T. gondii antibodies. One patient presented an intermediate result in the IgG avidity test, while the other patients showed high IgG avidity. The woman who had an intermediate result started treatment under supervision by the physician responsible for her case and remained under observation throughout her pregnancy. The newborn showed no signs or symptoms of congenital toxoplasmosis. This patient’s umbilical cord blood was analyzed using ELISA (BioKit®) and was found to be negative for IgM anti-T. gondii antibodies and positive for IgG anti-T. gondii antibodies. The relationship between serological profile, age, maternal education level and previous abortions is shown in Table 1 . There was a linear trend of increasing proportion of seropositivity for toxoplasmosis with increasing age among the subjects (p=0.002). Completed basic education was not significantly associated (p=0.171) with seropositivity in the bivariate analysis. There was no significant association between previous abortions and seropositivity for toxoplasmosis (p=0.212).

The relationship between serological profile and possible risk factors for toxoplasmosis is shown in Table 2. There were significant associations between seropositivity for toxoplasmosis and contact with cats (p=0.039) and between seropositivity and presence of rodents at home (p=0.008). No significant association was found between seropositivity for toxoplasmosis and the other variables. There was no evidence of association with consumption of water from the public water supply (p=0.075) at the significance level of 5%.

Multivariate analysis only confirmed the associations between seropositivity for toxoplasmosis and age group and between seropositivity and contact with cats. The odds of seropositivity among participants who reported contact with cats were 1.68 times higher than among those who had no contact with these animals. According to the OR of the age group variable, the chance that women in the range of 31 to 45 years old would be positive for toxoplasmosis was 3.97 times higher than for those who were outside this age group. The education variable significantly modified seropositivity for toxoplasmosis, such that education at least of the high school level was a protective factor against the infection ( Table 3 ). The presence of rodents at home did not show any significant association (p>0.05) after multivariate analysis.

DISCUSSION

This was the first epidemiological study carried out among pregnant and postpartum women in the City of Niterói, State of Rio de Janeiro, Brazil in which the prevalence of IgG anti-T. gondii antibodies found was 58.5%. In Brazil, this value was similar to results found by other authors in studies on pregnant women in southern Brazil: 59.8% and 54.8% in Rio Grande do Sul ^18,19 , 49.5% in Minas Gerais ¹⁷ and 55.1% in Paraná ²² . However, it was lower than what was observed in other states such as Mato Grosso ²¹ (70.7%), Mato Grosso do Sul ²⁷ (91.6%) and Goiás ²³ (67.7%) and in the indigenous Amazonas population ²⁸(73.5%). The comparison of the seropositivity of the pregnant women in Niterói with the prevalence of toxoplasmosis in other countries showed that the number of cases found in this study is higher than in Palestine ²⁹ (27.9%), China ³⁰ (12.3%), and Venezuela ³¹ (33%) and similar to the prevalence found in France ³² (47%), Albânia ³³ (48.6%), and Egypt ³⁴ (57.6%). The results observed in the present study show that in this population a significant number of susceptible women can acquire toxoplasmosis over the course of pregnancy. These susceptible patients present the risk that, through vertical transmission, the parasite may be passed on to the fetus, which could even lead to fetal death. This means that guidance on primary prevention measures and serological monitoring of pregnant women in the at-risk group (seronegative) are important measures for preventing congenital toxoplasmosis.

The prevalence of IgM anti-T. gondii antibodies was 4.2%, but only one patient presented an intermediate result in the IgG avidity test, while the other patients showed high IgG avidity. In this case, the physician’s decision was to begin treatment for congenital toxoplasmosis. However, it is important to take into consideration that a positive IgM result and/or low/intermediate avidity may influence the decision to start treatment. The treatment can be often unnecessary, as a result of an incorrect interpretation of serological results turning the treatment more dangerous than the parasite itself ^4,10 .

Although a large number of women aged 21 to 30 years were found to bepositive, the percentage of women who were seropositive for toxoplasmosis was higher in the age group from 31 to 45 years (66.9%). The odds of seropositivity among participants in the age group from 31 to 45 years were 3.97 times higher than among other participants, after multivariate analysis. This probably occurred because the women who were in the age group from 31 to 45 years had more chances of contact with the parasite throughout life. This finding reinforces the importance of routine serological screening for pregnant women. This increase in prevalence with age is consistent with the results from previous studies ^{18,19,21–23} .

Multivariate analysis indicated that pregnant and postpartum women with a low educational level (less than eight years of schooling) showed greater chance of being infected with T. gondii. Patients who had completed high school were clearly shown to be protected against anti-T. gondii seropositivity. Several other authors who have studied toxoplasmosis among pregnant women have also found similar results^{18,22,23,25,35} . These results show the importance of social and educational investments as preventive measures ³⁶ , not only against toxoplasmosis but also against other diseases. In relation to histories of abortions, there was no significant association between occurrences of previous abortions and seroprevalence for toxoplasmosis (p=0.212). Similar results have also been found in other studies ^21,37,38 .

Contact with domestic cats was regularly associated with Toxoplasma gondii transmission. In a study on the importance of farm animals for T. gondii infection in Brazil, it was concluded that cats were the key to controlling toxoplamosis ³⁹ . In the present study, a statistically significant association was found between seropositivity for toxoplasmosis and contact with cats (p=0.039). However, some studies on pregnant women have obtained different results, finding no significant association between seroprevalence and contact with domestic cats ^19,22,40 . In contrast, other authors have found an association with the presence of cats ^25,35 . It is known that the risk of infection is higher when there is contact with the feces of cats that are eliminating oocysts ^5,39 . Proper hygiene measures may decrease the risk of infection, such as correct disposal of cat feces, frequent cleaning of litter boxes and feeding cats with manufactured food or cooked meat ⁵ . It is important that pregnant women should avoid contact with cat feces and delegate these cleaning tasks to other people.

The presence of rodents at home was reported by 98 (41.8%) seropositive women. Although there was initially a significant association (p=0.008) with T. gondii infection, this variable ceased to have any significant association with seropositivity for toxoplasmosis after multivariate analysis (p>0.05). Multiple logistic regression makes a comparison between the variables that best explain the outcome. Thus, the presence of rodents at home was compared with other variables that were entered into the multivariable logistic regression model, such as age, educational level and contact with cats, and was found not to be statistically significantly associated with positivity for T. gondii infection in this population. Nonetheless, rodents, along with flies and cockroaches, have been epidemiologically implicated in mechanical transmission of T. gondii oocysts, and may facilitate contamination of food and water ²⁵ . Moreover, cats are commonly present in homes in order to control rodents in peridomestic areas. The cat/rodent combination may be related to transmission of T. gondii in humans and animals ⁴⁰ .

Consumption of raw or undercooked meat has been recognized as an important means of transmission^19,20,25,41 . However, in the present study, it was not associated with infection by T. gondii. Similarly, studies in Rolândia ²² , in the State of Paraná, and in Recife ⁴² , in the State of Pernambuco, Brazil, did not find any association between seropositivity for T. gondii and raw meat consumption. These findings may reflect differences in consumers’ eating habits, or different prevalences of infection in the meat-producing animals of these regions. There was no statistically significant association between seroprevalence and consumption of dried or cured meat (chorizo, ham, sausages or salami). However, some studies on pregnant women in the State of Rio Grande do Sul, Brazil, showed that women who had the habit of preparing and sometimes tasting sausage meat before it was cooked were at risk of becoming infected with cysts from bradyzoites present in the meat ^19,40 . Preparation or consumption of dried or cured meat is not a common practice in the State of Rio de Janeiro, which makes people less susceptible to this source of infection.

Contact with soil was not associated with T. gondii infection, although this has been widely cited in the literature as a risk factor for toxoplasmosis ^19,22,40,43 . Moreover, the water supply can also be a source of infection by T. gondii, acting as a disseminator for oocysts, and contamination of municipal water tanks by the feces of infected cats can lead to outbreaks or epidemics ^44,45 . However, in the present study, consumption of untreated water was not associated with infection by T. gondii. The urban origin of the pregnant and postpartum women studied and their low level of contact with water and soil, that could be contaminated with oocysts of the parasite, probably places this population at a lower risk of infection through these sources.

The habit of drinking raw (unpasteurized or untreated) milk may be a risk factor for infection with T. gondiiduring pregnancy. It was found that 33 (14.1%) of the seropositive women in this study had the habit of consuming untreated cow’s or/and goat’s milk. In Europe, 6 to 17% of pregnant women in different countries have been found to consume untreated milk or its derivatives ³⁸ . In the State of Goiás, Brazil, a study showed that 18.6% of the pregnant women consumed unpasteurized goats’ milk ²⁵ . In Rio Grande do Sul, 8.7% of the pregnant women reported consuming untreated milk ¹⁹ . Milk has been considered to be a potential vehicle for spreading toxoplasmosis in humans, since experimental work has shown that milk from infected animals contains tachyzoites and that this milk is capable of transmitting the infection to these animals’ offspring ^46,47 .

In relation to garbage disposal, there was no significant difference between seropositive and seronegative pregnant and postpartum women in this study. However, this variable is important because poor sanitary conditions favor transmission of diseases. In the case of toxoplasmosis, accumulation of waste in unsafe locations could favor transmission of oocysts by flies, cockroaches and rats that were attracted to these locations ²⁵ . Thus, lack of garbage collection from homes would be a risk factor for infection. In southern Brazil, some authors ⁴⁰ have stated that the garbage in the region that they studied played an important role as a source of infection. In the present study, the patients’ hometown (Niterói) has regular public garbage collection and, thus, these women were less exposed to contact with potential vectors of oocysts, like rodents, cockroaches and other insects.

The prevalence of IgG anti-T. gondii antibodies in the City of Niterói was high, 58.5%, and the risk factors for infection, after multivariate analysis, were: age over 30 years, contact with cats and education levels lower than university graduate level.

The results observed in the present study lead to the conclusion that there is considerable exposure to risk factors in this population and that a significant number of susceptible women can acquire Toxoplasma gondii infection over the course of pregnancy. These susceptible patients present the risk that, through vertical transmission, the parasite may be passed on to the fetus, which could even lead to fetal death. This means that guidance on primary prevention measures and serological monitoring of pregnant women in the at-risk group (seronegative) are important measures for preventing congenital toxoplasmosis and should be implemented within prenatal care.