In spite of about 2 cases of congenital toxoplasmosis per 1000 newborns occur in Mexico City , the epidemiology of T. gondii infection in general, and in pregnant women in particular, have been poorly studied in Mexico. A national survey in the general population showed that T. gondii infections exist in the whole country , and prevalences clearly vary among the regions [33, 34]. With respect to research on T. gondii infection in pregnant women of Mexico, only three studies have been reported. A study performed in a humid tropical State located in the south-east of Mexico showed that 60% of the pregnant women studied were positive for anti-T. gondii antibodies . Other study performed with high risk pregnant women from central Mexico showed that 34.9% had anti-T. gondii IgG antibodies , and a study of women with spontaneous abortions from the south of Mexico showed a 47% prevalence of anti-T. gondii antibodies . In this study, we found a 6.1% prevalence of latent T. gondii infection in pregnant women of Durango City, Mexico. This prevalence is much lower than those reported in the previous studies of pregnant women in other regions of Mexico [35–37]. Similarly, our prevalence found in Durango city is much lower than those reported in pregnant women from European countries [11–20], Vietnam , India [23, 24], Malaysia , Nepal , Sudan , New Zealand , Brazil , and Cuba , where prevalences vary from 9% to 74.5%. In contrast, our prevalence is higher than the one reported in a Korean study, where researchers found a prevalence of about 0.8% in the pregnant women studied . The lower prevalence found in our study compared with those reported elsewhere might be explained by differences in the characteristics of the pregnant women studied and the environment of Durango City. Concerning the characteristics of the pregnant women that may contribute to the low prevalence found could be: 1) the pregnant women studied belonged to a low socio-economic level and although they do eat meat, they certainly do it at a lower frequency and quantity than those with a better socioeconomic level because they can not afford to buy meat in a regular basis. Since eating contaminated meat is a well known route of T. gondii infection, the lower the frequency of meat consumption the lower the risk of infection; and 2) consumption of undercooked or raw meat is a rare practice both in the studied women and in people of the general population of Durango City. The presence of these two characteristics in our infected population points towards transmission that might have occurred in some cases, not by ingesting tissue cysts of the parasite in infected meat but by ingesting parasite oocysts in contaminated food or water. The poorest housing conditions found among the infected population support this statement. With respect to environmental characteristics, we may speculate that the low prevalence found could be also explained by a number of environment characteristics: 1) Durango City has a dry climate, and prevalence of T. gondii infection in dry climates has been reported lower than other climates [3, 19, 33]; 2) Most of the year Durango City usually has high temperatures during the day, and these high temperatures may contribute to reduce infectivity of T. gondii oocysts. This likely explanation could be supported by the observation in an experimental model that the higher the storing temperature of T. gondii oocysts the shorter the survival and infectivity time of the parasites ; and 3) Durango City also has a high altitude (1880 meters above sea level), and prevalences of T. gondii infection have been found lower in high altitudes than in low altitudes in some studies in humans [39, 40] and a study in animals .
In this study, living in a house with soil floors was associated with T. gondii infection (adjusted OR = 7.16; 95% CI: 1.39–36.84). Contaminated soil with parasite oocysts is a source of infection  and might contribute to explain the higher frequency of infection in women living in a house with floors made of soil than those living in a house with floors made of concrete or other materials. Theoretically, infected cats play a major role in contaminating soil; therefore persons living in a house with soil floors need to have a cat or have contact with it to become infected by this route. The fact that in this study the variable "contact with cats" was not associated with infection might be explained by a number of possible reasons: 1) contact with cats was a very common practice among the studied population, but parasite transmission could be more efficient in those houses with soil floors than in those with other kind of floors; 2) the frequency of infection in cats may be higher in those living in the poorest environment than those living in better housing conditions. The food consumed by the cats is certainly different in those living in the poorest housing condition than those living in a regular housing condition. More hunting of possibly infected animals may occur in the poorest environment than in a better one; 3) soil floors reflect extreme poverty and there are always further bad sanitary conditions in women with this status that may all facilitate transmission of T. gondii, and; 4) animals other than domestic cats may also contaminate soil floors. In this study we also found an association of latent T. gondii infection with a residence outside Durango State (adjusted OR = 4.25; 95% CI: 1.72–10.49). This finding suggests that those women could have been infected outside Durango State. In addition, logistic regression showed that turkey meat consumption was also associated with T. gondii infection (adjusted OR = 3.85; 95% CI: 1.30–11.44). Turkey meat consumption could be thus responsible for at least some cases of T. gondii infection among our infected population. This finding was unexpected since meats other than turkey meat have been implicated more frequently as a source of T. gondii infection than turkey meat. This finding deserves further investigation. The rest of the sociodemographic, clinical, and behavioural characteristics of the studied women did not show any association with T. gondii infection. Known factors associated with T. gondii infection in pregnant women such as drinking untreated water [13, 42], contact with garden soil , increasing parity , consumption of cured meat , and contact with cats , were not associated with T. gondii infection in our study.
We performed serum analysis by Microparticle Enzyme Immunoassays for anti-T. gondii IgG and IgM antibodies because these assays have been shown a high overall agreement as compared with the dye test . In addition, routine quality control measure were applied when analysing samples, therefore, we considered results of our assays were reliable.