Acute Toxoplasma gondii Infection among Family Members in the United States

We investigated 32 families of persons with acute toxoplasmosis in which >1 other family member was tested for Toxoplasma gondii infection; 18 (56%) families had >1 additional family member with acute infection. Family members of persons with acute toxoplasmosis should be screened for infection, especially pregnant women and immunocompromised persons.


Identification of Additional Family/Household Members
To identify all the members from the same households we searched in the PAMF-TSL electronic database the individuals with similar last names and collection dates and hand reviewed the hard copies of their PAMF-TSL records. Individuals with the above code were then grouped into study families. If the IC was a congenitally infected infant or a pregnant woman, an additional household member, beyond the mother-infant pair, had to be tested as well, in order for this group of individuals to qualify for a household.

Diagnostic Criteria
Acute T. gondii Infection within ≤6 Months from Sample Collection We used the following three very strict composite criteria for the diagnosis of acute toxoplasma infection as reported elsewhere (8)(9)(10)(11); A) IgG-Dye test titer ≥1:1024 AND IgM ELISA ≥5.0, AND acute pattern in the differential agglutination test; B) IgG-Dye test titer ≥1:1024 AND IgM ELISA ≥3.0 AND acute pattern in the differential agglutination test AND either IgA ELISA ≥5.0 or a low IgG avidity (< 10); C) IgG-Dye test titer ≤ 1: 512 AND IgM ELISA ≥5.0 AND acute pattern in the differential agglutination test AND either IgA ELISA ≥5.0 or low IgG avidity (< 10) (8,10). Patients meeting any of the above three composite criteria are likely to be infected within less than <6 months from the time of serum sampling. The above criteria were not applied if the IC was a congenitally infected infant or a patient with chorioretinitis. For congenitally infected infants there are special considerations: first, the diagnosis is based not only on serologic test results but also on the presence of clinical findings suggestive of congenital toxoplasmosis; second, IgM ISAGA is used instead of IgM ELISA (for infants < 6 months of age) and third, in 25-50% of infants with congenital toxoplasmosis the IgM ISAGA can be negative at birth (12). Also, for patients with chorioretinitis, if the ocular findings are typical of toxoplasmosis, a high IgG titer, even in the absence of other serologic markers, can suggest a recent postnatally acquired T. gondii infection. The above diagnostic criteria are routinely used in the daily clinical practice at PAMF-TSL to estimate of the most likely time of the T. gondii infection and their performance has been previously validated at the PAMF-TSL (8)(9)(10)(11).

Data Extraction
For all study family individuals we recorded the date of birth, the unique PAMF-TSL identifying number, the date of specimen collection, the serology and/or PCR results, characterization of the T. gondii infection status of those individuals according to the above described diagnostic criteria, any reported clinical manifestations and the risk factors for T. gondii infection. Clinical information was limited since it was based on answers to a short questionnaire regarding clinical signs, symptoms and risk factors (eg exposure to cat feces, ingestion of raw/undercooked meat, gardening, none of the above, other), routinely requested to assist in the more accurate interpretation of serologic test results.

Analyses
We calculated the prevalence of group 1 families (primary endpoint) and group 2 families (secondary endpoint) by dividing the number of families in group 1 and 2 respectively, by the total number of study-families over the 20-year study-period.
We used the Kruskal-Wallis non-parametric test to compare the IgG-Dye test titers and IgM-ELISA titers of the index cases across the three family groups. All analyses were done in STATA SE12 (StataCorp LP, College Station, TX, USA).

Identification of Eligible Families
Among 97,279 individuals serologically tested for T. gondii in the PAMF-TSL database over the 20-year study period, we identified 107 individuals who had in their record the specific code 90, indicating that at least one individual from their household was diagnosed with acute toxoplasma infection and at least one additional household member was serologically tested for T. gondii at PAMF-TSL. All samples were sent from diverse laboratories across the United States. Those 107 individuals were grouped in 32 families ( Figure 1).

Characteristics of the Study Families
The mean number of additional family members tested per household was 1.7 for families in the first group; 2.6 for families in the second group and 1.3 for families in the third group. In group 1 families, the ICs first tested were three congenitally infected infants, nine pregnant women, one patient with chorioretinitis, four patients with T. gondii lymphadenopathy, and one patient with fibromyalgia treated with corticosteroids who had a fatal outcome (Table 1).
In group 2 families, the ICs were two T. gondii-infected children, two congenitally infected infants, and one patient with chorioretinitis. In group 3 families, the ICs were five pregnant women, one T. gondii-infected child, and three congenitally infected infants.
The screening of additional household members of an IC led to the identification of a pregnant woman who had been infected during gestation and was tested because her husband had developed toxoplasmic lymphadenitis (IC 1, group 1) ( Table 1). In eight families, the diagnosis of acute toxoplasma infection during pregnancy was made in the mothers after their infants were diagnosed with congenital toxoplasmosis. We also documented one family in which the IC (IC-12, group 1) ( Table 1)

Serology Titers
In all study families, all additional household members were tested within 1-2 months from the time the family IC was first tested ( Table 2). The median IgG-Dye test and IgM-ELISA titers of the ICs in group 1 families were 3072 and 7.9, respectively ( Table 2). The corresponding median IgG-Dye test and IgM-ELISA titers for the ICs in group 2 families were 8000 and 7.7 and in group 3 families were 8000 and 4.8, respectively. The IgG-Dye test and IgM-ELISA titers of the ICs were not significantly different across the three family groups (p=0.27 and p=0.07, respectively).

Risk Factors
For the majority of the study families, the reporting of risk factors for acute T. gondii infection was incomplete, precluding a meaningful risk factor analysis between the family groups. In 11/18 families in group 1, in which risk factors were reported for at least one of their household members, exposure to cat feces was reported in 2 families; exposure to food likely contaminated with T. gondii was reported in 5 families (e.g., eating raw meat or food handled on surfaces where raw meat was cut and not washed); gardening was reported in 1 family and eating wild game meat was reported in 3 families (the meat was tested at PAMF-TSL and was positive for the presence of T. gondii DNA and all individuals who ate this meat were infected) ( Table 1). Some families reported more than one risk factor. Two study families reported that they had no known risk factors.

Prevalence of Group-One Families, among Families Tested at PAMF-TSL during 1991-2010 (Primary Endpoint)
The prevalence of group-one families, among the whole cohort of 32 eligible families was 56% (18/32) (Figure 1). In 14 of these group-one families, at least one additional household member had an acute T. gondii infection and for the remaining four families, at least one additional household member was found to be recently infected. Specifically, in the family of IC-10 (Table 2) -a pregnant woman with acute toxoplasma infection in the third trimester of her pregnancy-we documented that her 2-year old daughter (Daughter 2) also had a very recently acquired infection (very high IgG and IgA titer, low IgG avidity and acute pattern on the differential agglutination test; although her IgM was negative). In the family of IC-11 (Table  2),an infant with congenital toxoplasmosis, we documented that the child's father had serologic evidence of T. gondii infection acquired around the time of his wife's gestation; which was the time during which his wife was also infected (very high IgG titer, low avidity and an acute pattern in the differential agglutination test; although his IgM and IgA were negative). In the family of IC-16 (Table 2), a 70 year old immunocompromised woman with a fatal outcome from disseminated toxoplasmosis, we documented that her daughter also had a recent infection (high IgG titer, positive IgM and IgA and low avidity). In the family of IC-18 (Table 2), a man with toxoplasmic lymphadenopathy, we documented that this man's wife also had a recent infection (high IgG titer, positive IgM and an acute pattern on the differential agglutination test).