Human infection with the protozoan parasite Toxoplasma gondii occurs worldwide [¹,²]. Major routes of T. gondii infections include ingesting food or water that is contaminated with oocysts shed by cats or by eating undercooked or raw meat containing tissue cysts [^2-4]. The clinical spectrum of T. gondii infection varies from an asymptomatic state to severe illness. The parasite can affect the host's lymph nodes, eyes, central nervous system, liver, and heart [³,⁵,⁶]. In liver, the parasite has been associated with a number of pathological changes including hepatomegaly, granuloma, hepatitis, and necrosis [^7-14]. In addition, an epidemiological study has reported an association of T. gondii infection with liver cirrhosis [¹⁵]. However, epidemiological studies on the association of infection with T. gondii and liver disease are scarce, and have not been performed in Mexico. Therefore, we performed a case-control study in Northern Mexico to determine the seroprevalence of T. gondii infection and anti-T. gondii IgG levels in adult patients with liver disease attending the Department of Gastroenterology in a secondary-care public hospital in Durango City. Furthermore, we investigated socio-demographic, clinical, and behavioral characteristics associated with T. gondii seropositivity in these patients.

Anti-T. gondii IgG antibodies were found in 10 (13.3%) of 75 patients and in 16 (10.7%) of 150 controls (P = 0.71). Of the 10 anti-T. gondii IgG positive patients, 6 (8.0%) had IgG levels higher than 150 IU/ml, and 4 (5.3%) between 8 to 99 IU/ml. In comparison, of the 16 anti-T. gondii IgG positive controls, 9 (6.0%) had IgG levels higher than 150 IU/ml, 2 (1.3%) between 100 to 150 IU/ml, and 5 (3.3%) between 8 to 99 IU/ml. Anti-T. gondii IgG levels were comparable among patients and controls (P = 0.60). Anti-T. gondii IgM antibodies were found in 2 patients and in 5 controls (2.7% vs 3.3%, respectively; P = 0.57). The socio-demographic characteristics among seropositive and seronegative patients were not significantly different (Table 1). Seropositivity to T. gondii was significantly higher in patients with an occupation of truck driver than those with other occupations (3/3: 100% vs 8/65: 12.3%; P = 0.003). The type, duration and clinical response to treatment of liver disease did not show any association with the seroprevalence and levels of anti-T. gondii IgG (Table 2). The frequency of T. gondii seropositivity was higher in patients with reflex impairment (27.8%) than patients without this impairment (8.8%) (P = 0.05). Patients with a history of abdominal hernia repair had a significantly higher seroprevalence of T. gondii infection than those without this history (3/5: 60% vs 7/70: 10%, respectively; P = 0.01). In contrast, no statistically significant differences were observed among T. gondii positive and T. gondii negative patients in the frequencies of other clinical characteristics including concomitant diseases, frequent headaches, presence or history of lymphadenopathy, blood transfusion, or transplant, and impairments in memory, hearing or vision (Table 3).

Bivariate analysis showed a number of behavioral characteristics with a P value equal or less than 0.25 including cats at home, raising animals, traveling abroad, consumption of sheep, chicken, turkey, pigeon, rabbit, deer, squirrel, quail, skunk and armadillo meats, consumption of raw milk, ham, unwashed raw fruits, and untreated water, soil contact and soil floors at home. Multivariate analysis of these behavioral characteristics showed that consumption of sheep meat (OR = 8.69; 95% CI: 1.02-73.71; P = 0.04), rabbit meat (OR = 4.61; 95% CI: 1.06-19.98; P = 0.04), venison (OR = 40.46; 95% CI: 2.25-725.75; P < 0.01), and quail meat (OR = 38.50; 95% CI: 1.70-871.99; P < 0.01) were significantly associated with T. gondii infection in patients (Table 4). Other behavioral characteristics did not show an association with T. gondii infection. Raw data of patients and controls are included in additional files [additional file 1-cases and additional file 2-controls, respectively].

In this seroprevalence case-control study, we found a comparable frequency of anti-T. gondii IgG and IgM antibodies in liver disease patients and controls. Similarly, levels of anti-T. gondii IgG antibodies were comparable among these groups indicating that T. gondii infection is not likely to substantially contribute to the etiology of liver disease in our patient population. We are not aware of previous reports about the association of T. gondii infection in liver disease patients in Mexico, and reports in other countries are scarce. Our results conflict with those reported in a Turkish study where researchers found an association of T. gondii infection with liver cirrhosis [¹⁵]. Most of our patients suffered from liver cirrhosis but we did not find any association between seropositivity to T. gondii and cirrhosis. Similarly, the comparable seroprevalence of T. gondii infection in patients and controls differs from those reported in an Egyptian study where researchers found a 65.5% seroprevalence of T. gondii antibodies in patients with acute and chronic hepatic diseases against a 27% seroprevalence found in controls [¹⁷]. Certainly, differences in the characteristics of the studies might explain the differences in the seroprevalences including the use of different laboratory methods and matching procedures, difference in ages of participants and proportions of controls and patients.

None of the socio-demographic characteristics and diagnosis of liver disease associated with T. gondii seropositivity in our patients. Concerning behavioral characteristics, it was noteworthy that there was an association between T. gondii seropositivity and sheep meat consumption (OR = 8.69; 95% CI: 1.02-73.71; P = 0.04). Infections with T. gondii have been reported in sheep [¹⁸]. In addition, viable T. gondii has been found in lambs destined for meat consumption in the USA [¹⁹], and ovine meat consumed in France [²⁰]. It will therefore be of interest to examine the seroprevalence of T. gondii infection in sheep in Durango. In a recent study in the USA, elevated risk of recent T. gondii infection was associated with eating rare lamb [²¹]. Even frozen lamb meat has been associated with acute T. gondii infection in Brazil [²²]. The association of T. gondii infection and consumption of sheep meat in our patients was unexpected since lamb meat consumption was negatively associated with T. gondii infection in a previous study in psychiatric patients in Durango [²³]. We are not aware of any previous report about a positive association of T. gondii infection and consumption of sheep meat in Mexico. Remarkably, consumption of rabbit meat was also associated with T. gondii infection in patients (OR = 4.61; 95% CI: 1.06-19.98; P = 0.04). Infections with T. gondii in rabbits have been reported in several countries [^24-26]. Antibodies against T. gondii were found in 77 (26.9%) of 286 domestic rabbits from three rabbit farms in Mexico [²⁵]. However, the seroprevalence of T. gondii infection in rabbits in Durango is unknown. To the best of our knowledge there is not any previous report about the association of T. gondii infection and consumption of rabbit meat. In the present study, we also found an association between T. gondii seropositivity and consumption of quail meat (P < 0.01). Experimental infections with T. gondii in bobwhite [²⁷] and Japanese quail [²⁸] have been reported. We are not aware of any previous epidemiological report about the association of T. gondii infection and consumption of quail meat. More expectedly, we observed an association between T. gondii seropositivity and consumption of venison (P = < 0.01). Infections with T. gondii have been reported in deer [²⁹,³⁰]. Consumption of undercooked or uncooked venison has been linked to ocular toxoplasmosis in deer hunters [³¹]. Interestingly, toxoplasmosis with liver involvement has been reported in deer hunters who had eaten undercooked venison [³²].

The frequency of T. gondii seropositivity was higher in patients with reflex impairment (27.8%) than patients without this impairment (8.8%), and this difference showed a borderline significance (P = 0.05). In a previous study in patients with visual impairment in Durango, we found that patients with reflex impairment had a significantly higher frequency of T. gondii infection than those with normal reflexes [³³]. Reflex impairment might contribute to reducing the quality of life in T. gondii infected patients.

Seropositivity to T. gondii was comparable among liver disease patients and controls. Further studies with larger sample sizes are needed to elucidate the association of T. gondii with liver disease. Consumption of venison, and rabbit, sheep, and quail meats may warrant further investigation.

The authors declare that they have no competing interests.

CAE conceived and designed the study protocol, participated in the coordination and management of the study, applied the questionnaires, performed the laboratory tests and data analysis, and wrote the manuscript. JLTB and MFMS obtained clinical data, applied the questionnaires and performed the data analysis. SEM performed the statistical analysis. OL performed the data analysis, and wrote the manuscript. All authors read and approved the final manuscript.

This study was supported by Fondos Mixtos Durango-Consejo Nacional de Ciencia y Tecnología, Mexico. Grant No. 66718.