Seroepidemiology of Toxocara canis infection among mountain
aboriginal schoolchildren living in contaminated districts
in eastern Taiwan
Chia-Kwung Fan1, Chien-Ching Hung2,3, Wen-Yuan Du4, Chien-Wei Liao1and Kua-Eyre Su2
1 Department of Parasitology, College of Medicine, Taipei Medical University, Taiwan 2 Department of Parasitology, National Taiwan University College of Medicine, Taiwan
3 Division of Infectious Diseases, Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan 4 School of Medicine, Fu-Jen Catholic University, Taipei, Taiwan
Summary We conducted a seroepidemiological study of Toxocara canis infection among mountain aboriginal schoolchildren aged 7–12 years living in contaminated districts in eastern Taiwan. To detect sera IgG (‡1:64) we used a T. canis larval excretory-secretory antigen-based enzyme-linked immunosorbent assay. A short questionnaire elicited information on the practices of raising dogs, playing with soil, eating raw vegetables, or whether the child normally washed his/her hands before eating. The overall seroprevalence was quite high, reaching 76.6% (252/329). Neither age nor gender seemed to be important factors related to a positive serology. Aboriginal schoolchildren who raised dogs (OR ¼ 1.83, 95% CI: 1.04–3.19, P ¼ 0.03), or played with soil (OR ¼ 2.52, 95% CI: 1.49–4.25, P < 0.001) seemed to be more susceptible to T. canis infection than those who did not. Children who habitually washed their hands before eating (OR ¼ 0.57, 95% CI: 0.33–0.97, P ¼ 0.04) had a lower chance of acquiring T. canis infection than those who did not.
keywords aboriginal schoolchildren, enzyme-linked immunosorbent assay, eastern Taiwan, Toxocara canis
Introduction
Toxocariasis is an infection predominantly caused by migration of the roundworm Toxocara canis larvae to organs and tissues. The major clinical consequences of prolonged migration of T. canis larvae in humans are visceral larva migrans (VLM) and ocular toxocariasis (OT) (Glickman 1993). Humans can be infected either by ingesting parasite eggs, by contact with larvae contamin-ating the teats of bitches that have recently given birth or the muzzle of puppies, or by means of the paratenic hosts of the parasite. Young children up to the age of 12 years are the main population supposedly susceptible to T. canis infection due to dirt pica, poor hygiene, or frequent contact with dogs (Glickman 1993). In the last 10 years, child toxocariasis cases associated with endomyocarditis, gen-eralized lymphadenopathy, endophthalmitis, asthma, hepatosplenomegaly, and meningoencephalitis have been widely reported (Szczepanski et al. 1996; De Cock et al. 1998; Kincekova et al. 1999; Chan et al. 2001; Vidal et al. 2003). It is noteworthy that there is now considerable interest in the role of T. canis in epilepsy, particularly partial epilepsy (Nicoletti et al. 2002).
As the larvae migrate in tissues and do not develop into adult worms, making a definitive diagnosis through tissue biopsy and a stool examination is unlikely; diagnosis of toxocariasis mainly relies on a T. canis larval excretory-secretory (TcES) antigen-based enzyme-linked immuno-sorbent assay (ELISA), which reportedly shows 78% sensitivity and 92% specificity (Glickman et al. 1978). The seroprevalence of toxocariasis among children in different countries has been reported to be within a range of 4.0–86% using the TcES-ELISA (Holland et al. 1995; Jimenez et al. 1997; Luo et al. 1999; Ajayi et al. 2000; Alonso et al. 2000; Sadjjadi et al. 2000; Alderete et al. 2003). In general, reports concerning the seroprevalence of T. canis infection in aboriginal children living in moun-tainous areas are rather rare (Hakim et al. 1992).
Taiwanese aboriginal adults hunt wild animals with dogs (Fan et al. 1998a). It is unknown whether T. canis eggs shed from infected dogs contaminate the environment, thus causing T. canis infection in aboriginal populations. Our previous study revealed that the prevalence of T. canis embryonated eggs in stools deposited in the soil was fairly high (18%, 3/17) from a survey in Bunun aboriginal districts in eastern Taiwan (Fan et al. 2000). In this study,
we investigated the seroprevalence of and risk factors for T. canis infection among schoolchildren through screening for sera anti-Toxocara IgG antibodies by means of a TcES-based ELISA and a questionnaire interview in those aboriginal districts of Bunun that are contaminated.
Materials and methods
Study population and subject selection
In 1993, there were 10 aboriginal tribes, with the Bunun (approximately 41 000 people) being the fourth largest aboriginal population in Taiwan. Most Bunun populations live at elevations of 100–600 m in eastern Taiwan (Xiun 1996). Bunun aboriginal schoolchildren living in moun-tainous areas at elevations of 300–400 m in Hei-dong and Yen-ping districts in Taitung County of eastern Taiwan, respectively, were chosen as the study population (Figure 1). All 12 primary schools located in five villages in these two districts were included in the present study and the total number of schoolchildren in these primary schools was approximately 618.
In total, 329 serum samples were obtained by veni-puncture, 164 serum samples from boys and 165 from girls, all randomly collected from apparently healthy schoolchildren. The mean ages were similar in both genders and ranged between 7 and 12 years for all schoolchildren. All serum specimens were kept at)30 C until laboratory examination. Most aboriginal adults in both districts were labourers, or farmers (Ministry of the Interior, ROC 2000).
Questionnaire interview to assess risk factors Informed consent was obtained from the parents or guardians of schoolchildren, giving the participation rate of 53.2% (329/618). Each participating child completed a short self-administered questionnaire, the results of which were reviewed by three trained public health nurses. The questionnaire requested various personal details including age, sex, and residential district, and whether the children had been raising dogs, playing with soil, eating raw vegetables, and whether they washed their hands before eating. This study was approved by the Taipei Medical University ethical committee.
Egg culture
Adult T. canis were collected from the intestines of necropsied stray dogs (Fan et al. 1998b). Eggs harvested from the anterior third of the uterus were cultured according to the method of Bowman et al. (1987) with
slight modifications. Briefly, eggs were stirred in 1% sodium hypochlorite solution and incubated for 5 min at room temperature; thereafter, the mixture was centrifuged for 5 min at 2000 g. The pellet was washed twice with distilled water and once with 2% formalin. The eggs were resuspended in 2% formalin and placed in a 250-ml Erlenmeyer flask, to which 2% formalin was added to bring the liquid level to approximately 1 cm deep. The flask was covered with parafilm, and incubated at room temperature for 8–9 weeks with gentle weekly agitation. They were then stored at 4 C until use.
Preparation of larval excretory-secretory antigens Toxocara canis larvae were hatched according to De Savigny (1975). All solutions were sterile and experi-mental procedures were carried out under aseptic condi-tions. Briefly, T. canis embryonated eggs were washed with PBS, resuspended in 1% sodium hypochlorite and incuba-ted in an atmosphere containing 5% CO2at 37 C for 30 min. After several washings with PBS containing antibiotics (100 IU/ml penicillin, 250 lg/ml streptomycin, and 25 lg/ml nystatin; Biochrom KG, Berlin, Germany), the pellet was resuspended in RPMI-1640 medium (JRH Biosciences, Lenexa, KS, USA) containing the same con-centration of antibiotics. Motile larvae were collected from the bottom of the jar containing a modified Baermann apparatus (Yen et al. 1987) made up of two layers of cotton cloth in a steel sieve that had been kept in an atmosphere containing 5% CO2at 37 C for 12 h. They were transferred to new RPMI-1640 medium containing antibiotics in 50-ml tissue culture flasks (BD Biosciences, Franklin Lakes, NJ, USA) with a larval concentration of 1 · 104/ml. Supernatants containing ES antigens derived from the T. canis larvae (TcES) were collected weekly, pooled, and centrifuged to precipitate all debris. The resulting supernatant was sterilized by membrane filtration (pore size 0.2 lm) and dialysed (molecular weight cutoff, 6000–8000 kDa) against PBS for 12 h at 4 C, or until the phenol red disappeared. Protein content of the dialysate was estimated by the method of Bradford (1976), it was then concentrated by lyophilization (Labconco, Kansas City, MO, USA) and stored at)70 C until use. Enzyme-linked immunosorbent assay
Determination of the serum IgG specific for TcES was performed according to the method of Jimenez et al. (1997) and our previous study with modifications (Fan et al. 2003). Optimal dilutions of the antigens and antibodies (Abs) were pre-determined by check board titration. Briefly, wells of microtitre plates (Nunc, Roskilde, Denmark) were coated
with TcES antigen (protein concentration: 10 lg/ml) diluted in carbonate buffer at pH 9.6 and 2% skimmed milk in PBS was used as the blocking solution. Optimally diluted test sera (1:64), which were further verified by immunoblotting analysis, were used as the primary antibodies. Positive control sera were from patients with toxocariasis with proven clinical and laboratory diagnosis (kindly provided by Dr B. Gottstein, Institute of Parasitology, Berne University, Switzerland). Horseradish peroxidase-conjugated goat anti-human IgG (heavy and light chains) (Amersham,
Piscataway, NJ, USA), diluted 1:1000 in PBS, were used as the secondary antibodies, and 2,2-azinobis[3-ethylbenzo-thiazoline-6-sulfonic acid] diammonium salt (ABTS) solu-tion (Calbiochem, Darmstadt, Germany) was used as the substrate and the reaction was stopped by 1% sodium dodecyl sulphate (SDS). Duplicate tests were run on each test serum. The absorbance at 405 nm was determined in individual wells with an automated spectrophotometer (EIA
reader model EL312e Bio-Tec, Virginia Beach, VA, USA). Positive and negative control sera were included in each plate. Tested serum whose mean OD value was equal to or higher than the mean OD value minus 2 standard deviations of the positive control serum was considered to be positive.
Immunoblotting verification
The serum titre set at 1:64 selected by an ELISA titration study was further verified by immunoblotting (IB) analysis using negative and positive control sera. Briefly, TcES Ag preparations (9 lg per slab) were separated by 12.5% SDS-PAGE and transferred to a nitrocellulose membrane (Amersham) in a semi-blotter (Hoffer, Fremont, CA, USA). Strips were then incubated with sera diluted at 1:64. A Western Lightning kit (PerkinElmer Life Sciences, Bos-ton, MA, USA) was employed to detect the immunoreac-tions, and positive reactions were ascertained by the presence of low-molecular-weight bands of either 24, 28, 30, or 35 kDa which specifically correlate to T. canis infection (Magnaval et al. 1991). Some ELISA-positive and -negative tested sera were also subjected to IB confirmation.
Statistical analysis
In the present study, subjects were categorized into six age groups (7-, 8-, 9-, 10-, 11- and 12-year-old groups). Statistical analysis was performed using the SAS software system (SAS Institute, Cary, NC, USA). The increasing trend of age-specific seropositive rates was tested for statistical significance using the chi-square test for trends. Multivariate adjusted odds ratios (ORs) with their 95% CIs were estimated by means of multiple logistic regression analysis. The statistical significance of differences in age-, gender- and risk factor-adjusted seropositive rates among comparison groups was examined by testing the statistical significance of the regression coefficients.
Results
The optimal titre for detection of T. canis infection as determined by a titration study using TcES-ELISA was 1:64. Further verification by IB revealed that only positive but not negative control serum at titre of 1:64 could recognize both of the specific low-molecular-weight bands at 30 and 35 kDa which are specifically related to toxocariasis (Figure 2). A subsequent ELISA study revealed that mean OD values ± SD calculated from positive control sera from each plate were 0.440 ± 0.047, thus the OD cutoff was 0.346 (mean)2SDs of positive control sera OD values). Randomly selected ELISA-positive tested sera
119 °E 120 °E 10 km Taitung County 23 °27’ N Tropic of cancer Pacific Ocean N
Figure 1 Map of Taiwan showing selected study areas: (¯) Bunun villages; ( ) mountain areas.
could recognize both specific low-molecular-weight bands at 30 and 35 kDa, whereas ELISA-negative tested serum did not react with either of these bands (Figure 2).
Of 329 serum samples studied, 252 (76.6%; 252/329) were positive for Toxocara IgG antibody for aboriginal schoolchildren as determined by TcES-ELISA. The overall seroprevalence was 78.0% (128/164) in boys and 75.2% (124/165) in girls (Table 1). Seroprevalence was highest (83.9%, 47/56) in 8-year-olds, followed in sequence by 81.8% (45/55) in 7-year-olds, 77.6% (45/58) in 10-year-olds, 73.7% (42/57) in 12-year-10-year-olds, 73.3% (33/45) in 9-year-olds and 69.0% (40/58) in the 11-year-old age group (Table 1). Seropositive values for the IgG antibody against TcES in aboriginal schoolchildren were 79.4% (197/248) in those who had histories of raising dogs, 82.5% (179/217) in children who played with soil, 73.0% (100/137) in those who ate raw vegetables, and 72.3% (133/184) for those who washed their hands before eating (Table 1). In the multiple logistic regression analysis, gender, age and risk factors were included in the regression model. As shown in Table 2, gender did not seem to have a significant association with seropositivity for T. canis antibody (OR ¼ 0.85, 95% CI ¼ 0.71–1.96, P ¼ 0.53). Age was also not an important factor related to T. canis infection due to there being no significant differences in seroprevalence among the 7-, 8-, 9-, 10-, 11- and 12-year-old age groups (P > 0.05). Moreover, the multivariate-adjusted ORs were 1.83 and 2.52 for those
who had histories of raising dogs or playing with soil against those without (OR ¼ 1.83, 95% CI ¼ 1.04–3.19, P ¼ 0.03; OR ¼ 2.52, 95% CI ¼ 1.49–4.25, P < 0.001). 180 73 35 30 kDa 1 2 3 4 5 6
Figure 2 Immunoblotting analysis of clinically proven positive control sera and randomly selected ELISA-positive sera showing reactive low-molecular-weight bands at 30 and 35 KDa specific-ally related to toxocariasis. Lane 1: positive control serum; lane 2: negative control serum; lane 3: ELISA-positive tested serum 1 (mean ± SD: 1.369 ± 0.053 OD); lane 4: ELISA-positive tested serum 2 (0.656 ± 0.062 OD); lane 5: ELISA-negative tested serum 1 (0.313 ± 0.088 OD); lane 6: ELISA-negative tested serum 2 (0.269 ± 0.022 OD).
Table 1 Demographic characteristics of the seroprevalence of Toxocara canis IgG antibody among mountain aboriginal schoolchildren in eastern Taiwan
Variable Group No. tested No. positive Percentage Gender Boys 164 128 78.0 Girls 165 124 75.2 Age (years) 7 55 45 81.8 8 56 47 83.9 9 45 33 73.3 10 58 45 77.6 11 58 40 69.0 12 57 42 73.7 Risk factors Raising dogs No 81 55 67.9 Yes 248 197 79.4 Playing in the soil No 112 73 65.2 Yes 217 179 82.5 Eating raw vegetables No 192 152 79.2 Yes 137 100 73.0 Washing hands before eating No 145 119 82.1 Yes 184 133 72.3 Total 329 252 76.6
Table 2 Multivariate-adjusted ORs for various risk factors associated with seropositivity of Toxocara canis antibodies among aboriginal children in Taiwan
Variable Group
Multivariate-adjusted
ORs (95% CI) P-value Gender Female 1.00* (referent)
Male 0.85 (0.71–1.96) 0.53 Age (years) 7 1.00 (referent)
8 1.16 (0.43–3.12) 0.77 9 0.61 (0.24–1.58) 0.31 10 0.77 (0.31–1.93) 0.58 11 0.49 (0.20–1.19) 0.11 12 0.62 (0.25–1.54) 0.30 Risk factors
Raising dogs No 1.00à (referent)
Yes 1.83 (1.04–3.19) 0.03 Playing in the soil No 1.00à (referent) Yes 2.52 (1.49–4.25) <0.001 Eating raw vegetables No 1.00à (referent) Yes 0.71 (0.43–1.19) 0.19 Washing hands before eating No 1.00à (referent) Yes 0.57 (0.33–0.97) 0.01 * Adjusted variables included age and risk factors.
Adjusted variables included gender and risk factors. àAdjusted variables included gender and age.
Schoolchildren who had the habit of washing their hands before eating were less likely to acquire T. canis infection than those who did not (OR ¼ 0.57, 95% CI ¼ 0.33– 0.97, P < 0.001). We found no significant difference in seroprevalence between schoolchildren with and without a history of eating raw vegetables (OR ¼ 0.71, 95% CI ¼ 0.43–1.19, P ¼ 0.19).
Discussion
Serological tests are of considerable importance in the detection of infection by T. canis, as the clinical symp-toms of toxocariasis are of limited value in the differential diagnosis (Schantz 1989). The use of TcES-ELISA to detect antibodies against T. canis does not require the pre-absorption of sera with embryonated Ascaris suum egg antigen because TcES-ELISA is now widely
recognized as having good specificity in the diagnosis of toxocariasis (Glickman & Schantz 1981; Hakim et al. 1992; Gueglio et al. 1994; Luo et al. 1999). In the present study, a fixed ELISA cutoff titre of 1:64 in the screening of sera for the anti-Toxocara IgG antibody in Taiwanese aboriginal schoolchildren by TcES-ELISA appeared to be valid as further verified by immunoblotting analysis that not only positive control sera (1:64) could react with the specific low-molecular-weight bands at 30 and 35 kDa which were correlated with toxocariasis (Mag-naval et al. 1991), but also ELISA-positive tested sera exhibited similar results, while negative control sera as well as ELISA-negative tested sera showed no reactive bands.
Our study revealed that the overall seroprevalence among healthy aboriginal schoolchildren living in egg-contaminated districts was quite high (76.6%, 252/329). Comparison of seroprevalence data between the present result and other studies is difficult due to the use of different cutoff titres and the difficulty in evaluating relationships between titre level, infection, and clinical findings (Alderete et al. 2003). However, lower seroprevalences in healthy schoolchildren ranging from 4.0% to 49.1% have been observed in Argentina, Brazil, China, Iran, Ireland, Nigeria, Spain and the US (Holland et al. 1995; Jimenez et al. 1997; Luo et al. 1999; Ajayi et al. 2000; Alonso et al. 2000; Sadjjadi et al. 2000; Alderete et al. 2003). A higher seroprevalence (86%) was observed in St Lucia (Thompson et al. 1986).
Gender did not seem to be an important factor related to T. canis infection among healthy Taiwanese aboriginal children due to a lack of a significant association between gender and frequency of Toxocara seropositivity in the present study. Similar findings have been reported in China, Iran, Nigeria and Spain (Jimenez et al. 1997; Luo
et al. 1999; Ajayi et al. 2000; Sadjjadi et al. 2000); while some reports indicated that boys had a greater opportunity to acquire T. canis infection, as they had more contact with dogs (Holland et al. 1995; Alonso et al. 2000). Sero-prevalence of T. canis infection in aboriginal school-children did not increase with age, perhaps due to the age range being too small to find a discrepancy of age vs. seroprevalence. Nevertheless, the children older than 7 years might have acquired T. canis infection earlier in their childhood, as antibody titres might persist for a long period of time (Schantz 1989). Similar findings of a lack of an age-related increase in seroprevalence were also repor-ted in children in Argentina, Iran, Nigeria and Spain (Jimenez et al. 1997; Ajayi et al. 2000; Alonso et al. 2000; Sadjjadi et al. 2000).
On the other hand, a significant association was observed between ownership of dogs and Toxocara infec-tion in aboriginal schoolchildren. They probably acquired the infection through inadvertently ingesting eggs con-taminating a dog’s body. A recent study indicated that dogs infected with T. canis might infect people by direct contact because of the high density of embryonating and embry-onated eggs in their fur (Wolfe & Wright 2003). Never-theless, high seropositive rates of 79.4% in dog owners and 67.9% in non-owners of dogs in the present study suggest that these two groups are equally at risk of being infected. The results are in line with those of Woodruff et al. (1978) who observed that 50% of patients with clinical toxoca-riasis had never owned a dog or had close contact with pets. However, controversy exists regarding the import-ance of contact with dogs as a risk factor for human toxocariasis. Some authors reported a higher frequency of infection for individuals who maintained contact with dogs (Matsumura & Endo 1983); while others found no association of ownership and professional contact with dogs with the frequency of Toxocara infection (Woodruff et al. 1978). Nevertheless, the presence of dogs seems to be important for the determination of human infection, as the eradication of dogs in Iceland during the 1940s led to the disappearance of human Toxocara infection in that coun-try at the beginning of the 1980s (Woodruff et al. 1982).
The present results further suggest that toxocaral infec-tion in Taiwanese aboriginal schoolchildren in the study area was also acquired by the ingestion of soil containing infective eggs, confirming our earlier report that contam-ination of soils with T. canis eggs was relatively high (18%) in the same Bunun districts (Fan et al. 2000). This confirms that children who do not wash their hands before eating are more likely to get infected. This high incidence of T. canis infection among Bunun schoolchildren may well exist in other aboriginal districts in Taiwan due to their similar lifestyles (Xiun 1996).
It was noteworthy that OT was reportedly more com-mon in children over 6 years (Schantz 1989); in addition, decreased cognitive development or increased behavioural disorders were also observed in asymptomatic children (Marmor et al. 1987; Nelson et al. 1996). Recently, it was proposed that the cause of partial epilepsy was highly related to T. canis infection (Nicoletti et al. 2002). However, our laboratory has diagnosed (using patients’ vitreal fluid and sera as determined by TcES-ELISA and Western blotting) several cases in Taiwanese children referred from different hospitals in Taipei City, who were suspected of being infected with OT (Fan & Su KE, unpublished data). There is thus an urgent need to address whether Toxocara infection really is a hazard to the health of Taiwanese schoolchildren. In fact, VLM and OT are practically unknown to local physicians in aboriginal districts; whether these asymptomatic aboriginal school-children with positive serology are susceptible to develop-ment of vision impairdevelop-ments or other Toxocara-associated clinical features, such as partial epilepsy, should be further evaluated.
Acknowledgements
We thank H.S. Lee and D.C. Kao for their valuable technical assistance and collection of human sera. Dr D. Chamberlin kindly undertook a critical revision of this paper.
References
Ajayi OO, Duhlinska DD, Agwale SM & Njoku M (2000) Frequency of human toxocariasis in Jos, Plateau State, Nigeria. Memorias do Instituto Oswaldo Cruz 95, 147–149.
Alderete JMS, Jacob CMA, Pastorino AC et al. (2003) Prevalence of Toxocara infection in schoolchildren from the Butanta region, Sao Paulo, Brazil. Memorias do Instituto Oswaldo Cruz 98, 593–597.
Alonso JM, Bojanich MV, Chamorro M & Gorodner JO (2000) Toxocara seroprevalence in children from a subtropical city in Argentina. Revista do Instituto de Medicina Tropical de Sao Paulo 42, 235–237.
Bowman DD, Mika-Grieve M & Grieve RB (1987) Circulating excretory-secretory antigen levels and specific antibody responses in mice infected with Toxocara canis. American Journal of Tropical Medicine and Hygiene 36, 75–82. Bradford MM (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytic Biochemistry 72, 248–254.
Chan PW, Anuar AK, Fong MY, Debruyne JA & Ibrahim J (2001) Toxocara seroprevalence and childhood asthma among Malay-sian children. Pediatrics International 43, 350–353.
De Cock C, Lemaitre J & Deuvaert FE (1998) Loeffler
endomyocarditis: a clinical presentation as right ventricular tu-mor. Journal of Heart Valve Diseases 7, 668–671.
De Savigny DH (1975) In vitro maintenance of Toxocara canis larvae and a simple method for the production of Toxocara ES antigen for use in serodiagnostic tests for visceral larva migrans. Journal of Parasitology 61, 781–782.
Fan CK, Su KE, Chung WC et al. (1998a) Seroprevalence of Toxoplasma gondii antibodies among Atayal aboriginal people and their hunting dogs in northeastern Taiwan. Japanese Journal of Medical Sciences and Biology 51, 35–42.
Fan CK, Su KE, Chung WC et al. (1998b) A parasitological survey among stray dogs in Taipei City with hematological findings. Journal of Chinese Society of Veterinary Science 24, 288–294. (in Chinese, with English abstract)
Fan CK, Tsai YJ, Su KE & Lin CC (2000) Epidemiological study of zoonotic parasite infection among dogs in Bunun aboriginal tribes (Da-Wu and Dan-Zen districts) in Taitung County. Chinese Journal of Parasitology 13, 47–52. (in Chinese with English abstract)
Fan CK, Lin YH, Du WY & Su KE (2003) Infectivity and pathogenicity of 14-month-cultured embryonated eggs of Toxocara canis in mice. Veterinary Parasitology 113, 145–155.
Glickman LT (1993) The epidemiology of human toxocariasis. In: Toxocara and Toxocariasis: Clinical, Epidemiological and Molecular Perspectives (eds JW Lewis & RM Maizels) Birbeck & Sons, London, pp. 3–10.
Glickman LT & Schantz PM (1981) Epidemiology and patho-genesis of zoonotic toxocariasis. Epidemiological Review 3, 230–250.
Glickman LT, Schantz PM, Dombroske R & Cypess R (1978) Evaluation of serodiagnostic tests for visceral larva migrans. American Journal of Tropical Medicine and Hygiene 27, 492–498.
Gueglio B, de Gentile L, Nguyen JM & Achard J (1994) Epide-miologic approach to human toxocariasis in western France. Parasitology Research 80, 531–536.
Hakim SL, Mak JW, Lam PLW, Nazma S & Normaznah Y (1992) Seroprevalence of Toxocara canis antibodies among Orang Asli (aborigines) in Peninsular Malaysia. Southeast Asian Journal of Tropical Medicine and Public Health 23, 493–496.
Holland CV, O’Lorcain P, Taylor MRH & Kelly A (1995) Seroepidemiology of toxocariasis in schoolchildren. Parasitology 110, 535–545.
Jimenez JF, Valladares B, Fernandez-Palacios JM, de Armas D & del Castillo A (1997) A serologic study of human toxocariasis in the Canary Islands (Spain): environmental influences. American Journal of Tropical Medicine and Hygiene 56, 113–115. Kincekova J, Reiterova K & Dubinsky P (1999) Larval
toxocar-iasis and its clinical manifestation in childhood in the Slovak Republic. Journal of Helminthology 73, 323–328.
Luo ZJ, Wang GX, Yang CI, Luo CH, Cheng SW & Liao L (1999) Detection of circulating antigens and antibodies in Toxocara canis infection among children in Chengdu, China. Journal of Parasitology 85, 252–256.
Magnaval JF, Fabre R, Maurieres P, Charlet JP & de Larrard B (1991) Application of the western-blotting procedure for the immunodiagnosis of human toxocariasis. Parasitology Research 77, 697–702.
Marmor M, Glickman L & Shofer F et al. (1987) Toxocara canis infection of children: epidemiologic and neuropsychologic findings. American Journal of Public Health 77, 554–559. Matsumura K & Endo R (1983) Seroepidemiological study of
toxocaral infection in man by enzyme-linked immunosorbent assay. Journal of Hygiene 90, 61–65.
Nelson S, Greene T & Ernhart CB (1996) Toxocara canis infection in preschool age children: risk factors and the cognitive devel-opment of preschool children. Neurotoxicology and Teratology 18, 167–174.
Nicoletti A, Bartoloni A, Reggio A et al. (2002) Epilepsy, cysti-cercosis, and toxocariasis: a population-based case-control study in rural Bolivia. Neurology 58, 1256–1261. Sadjjadi SM, Khosravi M, Mehrabani D & Orya A (2000)
Seroprevalence of Toxocara infection in schoolchildren in Shiraz, southern Iran. Journal of Tropical Pediatrics 46, 327–330.
Schantz PM (1989) Toxocara larva migrans now. American Journal of Tropical Medicine and Hygiene 41, 21–34. Szczepanski T, Sonta-Jakimczyk D, Janik-Moszant A & Olejnik I
(1996) Generalized lymphadenopathy as initial presentation of
toxocariasis in a seven-year-old boy. Pediatric Infectious Disease Journal 15, 717–718.
Thompson DE, Bundy DAP, Cooper ES & Schantz PM (1986) Epidemiological characteristics of Toxocara canis zoonotic infection of children in a Caribbean community. Bulletin of the World Health Organization 64, 283–290.
Vidal JE, Sztajnbok J & Seguro AC (2003) Eosinophilic men-gioencephalitis due to Toxocara canis: case report and review of the literature. American Journal of Tropical Medicine and Hygiene 69, 341–343.
Wolfe A & Wright IP (2003) Human toxocariasis and direct contact with dogs. Veterinary Record 152, 419–422. Woodruff AW, de Savigny DH & Jocobs DE (1978) Study of
toxocaral infection in dog breeders. British Medical Journal 2, 1747–1748.
Woodruff AW, de Savigny DH & Hendy-Ibbs PM (1982) Toxocaral and toxoplasmal antibodies in cat breeders and in Icelanders exposed to cats but not to dogs. British Medical Journal (Clinical Research Education) 284, 309–310.
Xiun HL (1996) Development of aborigines in Taiwan. In: Taiwan Aborigines (ed. DT Sun) Shan-Hei, Taipei, pp. 1–10. (in Chinese)
Yen CM, Chen ER & Hsieh HC (1987) Angiostrongylus canto-nensis infection in immunosuppressed mice. Japanese Journal of Parasitology 36, 397–404.
Authors
Chia-Kwung Fan (corresponding author) and Chien-Wei Liao, No. 250, Wu-Hsing Street, Department of Parasitology, College of Medicine, Taipei Medical University, Taipei 110, Taiwan. Tel.: +886 2 22434856; Fax: +886 2 22434856;
E-mail: tedfan@tmu.edu.tw, toxoplasma0422@yahoo.com.tw
Chien-Ching Hung and Kua-Eyre Su, No. 1, Sec. 1, Jen-Ai Rd, Department of Parasitology, National Taiwan University College of Medicine, Taipei 100, Taiwan. E-mail: hcc0401@ha.mc.ntu.edu.tw; kesu@ha.mc.ntu.edu.tw
Wen-Yuan Du, 510 Chung-Cheng Rd, Hsien-Chuang, School of Medicine, Fu-Jen Catholic University, Taipei 242, Taiwan. E-mail: lins2202@mails.fju.edu.tw
