INTRODUCTION
The agent of toxocariasis is the ascarid canine nematode Toxocara canis, although Toxocara cati commonly found in cats probably causes a similar infection (Nichols, 1956; Kennedy et al. 1987). While the domestic dog represents the definitive host of T. canis, aberrant infection in humans is acquired through ingestion of the embryonated eggs, typically via contaminated soil or vegetables. The clinical manifestations are dependent on the organ involved. Although visceral larva migrans and ocular larva migrans are the two clinical syndromes typically described (Despommier, 2003), other manifestations include neurological larva migrans, and covert toxocariasis which reflects a generalized response of the immune system to constant challenge by the parasite antigens (Pawlowski, 2001). First described in 1950 by Wilder (Wilder, 1950) then in 1952 by Beaver et al. (1952), and initially regarded as an uncommon paediatric illness, toxocariasis is now believed to be the most prevalent helminthic infestation in industrialized countries (Magnaval et al. 2001). Through the use of newly-developed and reliable immunodiagnostic techniques, several studies from various countries around the world have recently investigated the prevalence of human infection and past exposure to the parasite (Herrmann et al. 1985; Marmor et al. 1987; Buijs et al. 1994; Montalvo et al. 1994; Holland et al. 1995; Rai et al. 1996; Jimenez et al. 1997; Sadjjadi et al. 2000; Hermanowska-Szpakowicz et al. 2001; Park et al. 2002).
The understanding of the epidemiology of Toxocara in Lebanon has been very limited. This pilot study addresses unanswered questions about the seroprevalence of toxocariasis in this country and its association with various socioeconomic and environmental factors.
MATERIALS AND METHODS
Study subjects
The study was conducted between 1 May 2004 and 30 June 2004 at the American University of Beirut Medical Center (AUBMC), a major teaching hospital in the country that offers outpatient as well as inpatient healthcare to a wide spectrum of the Lebanese population. Individuals [ges ]14 years of age attending the outpatient clinics were included.
Data collection
After obtaining informed consent, a short questionnaire was administered to each subject through a personal interview. Information was collected on demographics, education, life-style, and living conditions. Subjects were also screened for a variety of systemic symptoms.
Specimen analysis
Blood samples were collected into sterile plain collection tubes and centrifuged. The supernatant sera were stored at −70 °C until analysed. A commercially available enzyme-linked immunosorbent assay (ELISA) kit (Bordier Affinity Products SA, Crissier, Switzerland) was used to detect IgG antibodies directed against the excretory-secretory antigens (ES) produced by the second-stage Toxocara larvae. A Western blot (WB) assay (LDBio Diagnostics, Lyon, France) was used as a confirmatory test on samples that tested positive by ELISA. Specific instructions provided by the manufacturers were followed to perform the testing. The presence of 2 or more bands in the range of 24–35 kDa was indicative of a positive Western blot assay.
Statistical analysis
Data were entered into a database using SPSS 11.0 (SPSS Inc, Chicago, IL). Sample size calculation was not performed as this was intended to be a pilot study. The Mann-Whitney test was used to describe the age distribution in quartiles among seropositive and seronegative subjects. Bivariate analysis to study the association between various sociodemographic variables and the presence of anti-Toxocara antibodies was performed using the χ2 test. Variables that were significant on bivariate analysis were entered into a multivariate backward logistic regression model using Stata 7.0 (Stata Corporation, College Station, TX). The level of significance for all tests was set at 95% (two-sided P=0·05).
Ethical considerations
All data collected for the purpose of this study were kept confidential and questionnaires were delivered anonymously. All patients read and signed an informed consent form in Arabic prior to enrolment. Individuals under 18 years of age gave their verbal consent and the written consent form was signed by the accompanying adult parent. The study was reviewed and approved by the Institutional Review Board at AUBMC.
RESULTS
A total of 150 study subjects gave their consent to participate in the study. The ELISA test was performed first on all available sera. Eighteen samples tested positive (12%) and another 14 gave borderline results (9%). The remaining 118 samples were negative. The WB assay was next used to confirm the 32 positive and borderline results. All 18 samples positive by ELISA were also positive by WB. Of the 14 sera with borderline results, 10 were confirmed to be positive by WB. Therefore, the total number of positive specimens was 28, giving a seroprevalence rate for toxocariasis of 19%. Three of these 28 specimens reacted only to low molecular weight bands (24–35 kDa).
Table 1 shows the general characteristics of the study subjects categorized according to Toxocara serostatus. Variables associated with seropositivity included male gender (57 vs. 32%), less than high school education (32 vs. 12%), low family income as a child (25 vs. 8%), and 5 or more persons residing in the household as a child (96 vs. 80%). There were no significant differences among the two groups with regard to age (comparison of means as well as quartiles through the Mann-Whitney test), frequent contact with soil (farming, gardening, etc.), pet ownership, number of rooms in the house, sanitary facilities (sewer vs. cesspool) or source of drinking water (bottled vs. pipeline or underground water).
The variables that most strongly predicted a seropositive status were the number of persons in the household and low family income during childhood (odds ratios=6·6 and 3·7, respectively). However, only low level of education and male gender remained significant (odds ratios=3·1 and 2·8, respectively) with multivariate analysis.
Study subjects were also questioned about a variety of systemic symptoms including fever, headache, lethargy, abdominal pain, sleep disorders, behavioural problems, allergic symptoms, anorexia, and skin rashes. None were significantly different among seropositive compared to seronegative individuals (data not shown).
DISCUSSION
Four major forms of human toxocariasis are presently recognized: systemic, compartmentalized, covert and asymptomatic. Over the past 4 years, several cases of subacute and chronic myelitis of unclear etiology were encountered at AUBMC. Patients were ultimately found to have positive Toxocara serology in both serum and cerebrospinal fluid, and most exhibited clinical and radiological improvement with antihelminthic therapy. The diagnosis of this extremely rare entity of Toxocara myelitis aroused interest in the hitherto undetermined distribution of toxocariasis in Lebanon. This pilot study is the first to tackle this problematic parasitic infection and disclose its association with various epidemiological variables.
The diagnosis of toxocariasis relies on the detection of antibodies directed against the larval ES antigens. These immunogenic glycopeptides are released from the epicuticle of migrating larvae as it is shed following the binding of specific antibodies (Page et al. 1992). The most commonly utilized immunodiagnostic technique is ELISA (de Savigny et al. 1979) with a test sensitivity that ranges between 73% and 90% (Glickman et al. 1978; Jacquier et al. 1991). However, there has been concern about cross-reactivity with other helminths, such as Fasciola hepatica and Ascaris suum, giving the ELISA test a specificity of around 93% (Glickman et al. 1978; Speiser and Gottstein, 1984; Romasanta et al. 2003). The low-cost easy-to-use ELISA method has therefore been recommended as a first-line test, but confirmatory testing by WB should be performed, as the latter offers comparable sensitivity and increased specificity when considering lower molecular weight bands ( Magnaval et al. 1991; Courtade et al. 1995). The WB assay confirmed all of our positive and 10 of the 14 borderline ELISA results. As our knowledge about the time-course of the Toxocara antibody titres following infection is limited (how soon they start to rise and how long they remain elevated), caution should be employed when interpreting positive results through population screening. For the most part, these represent past exposure to the parasite rather than recent infection (Magnaval et al. 2001).
Toxocara has a worldwide distribution, both in developing and in industrialized countries. In the United States, Herrmann et al. (1985) found seropositivity rates between 5 and 7% among children in different geographical areas of the country. Much higher rates approaching 30% were found among African-American children of lower socioeconomic status. Some countries in the Middle East have also published seroprevalence rates for toxocariasis, up to 23% in Egypt (Safar et al. 1995), 19·5% in Jordan (Abo-Shehada et al. 1992) and 8·5% in Israel in mentally retarded children (Huminer et al. 1992). In our study, we calculated a seroprevalence rate of 19% among individuals 14 years of age and older, making Lebanon a country of relatively high endemicity for Toxocara. Including younger children in our study population would have probably yielded an even higher rate.
Several epidemiological studies have attempted to identify risk factors for acquiring toxocariasis with inconsistent results. For instance, some studies found a significant association between Toxocara seropositivity and variables that include dog ownership (Marmor et al. 1987; Holland et al. 1995), geophagia (Marmor et al. 1987; Holland et al. 1995), and rural residence (Herrmann et al. 1985; Holland et al. 1995; Hermanowska-Szpakowicz et al. 2001), while others could not substantiate such a correlation (Glickman et al. 1979; Jimenez et al. 1997; Sadjjadi et al. 2000). Other factors like male gender have exhibited more consistent associations with positive exposure to Toxocara ( Holland et al. 1995; Rai et al. 1996; Hermanowska-Szpakowicz et al. 2001). Several indicators of low socioeconomic status have also been linked to a diagnosis of toxocariasis. A survey in the United States over a 3-year period found that a positive serological test for Toxocara was more common with increased number of persons in the household, decreased income, lower education, and fewer rooms in the house (Herrmann et al. 1985). Two other studies reported higher rates of toxocariasis among children with low parental education (Worley et al. 1984; Sadjjadi et al. 2000). In our study, only below high school education and male gender remained significant with multivariate analysis.
In this first report from Lebanon, we found that toxocariasis is relatively common in this country with a seroprevalence rate of 19%. Males and individuals with less than a high school education constitute the most at-risk population. This preliminary pilot study provides valuable information and sets the ground for a broader population-based epidemiological survey in the future to validate these results. An environmental study is also being planned where stray dog faeces and soil samples from public parks and gardens will be examined.
