Introduction

Toxocariasis, caused by intestinal roundworm of dogs (Toxocara canis) or cats (Toxocara catis), is a zoonotic infection with a worldwide distribution [1,2]. Humans can get infected by ingestion of embryonated eggs present on the soil, plants or soil-dwelling invertebrates contaminated by dog or cat feces, and less frequently by ingestion of encapsulated larvae from undercooked paratenic hosts such as chickens, cattle and sheep. Infection is followed by the migration of third-stage larvae through the tissues which is usually asymptomatic but may be associated with a variety of non-specific clinical features [2,3]. Presentation may be acute or sub-acute, with systemic, abdominal or respiratory manifestations, classically described as the syndrome of visceral larva migrans (VLM) sometimes associated with dermatological symptoms as well. Occasionally, the course is complicated by the involvement of the central nervous system (neurological toxocariasis) or the eyes (ocular toxocariasis). Like in other helminthic infections with larval migration, blood eosinophilia is common and/or numbers of eosinophils may be increased in other relevant tissues or fluids; other laboratory abnormalities such as elevated inflammatory parameters or liver enzymes disturbances are often mild and inconstant. Of note, a distinct clinical presentation called “covert” (in children) or “common” (in adults) toxocariasis has been described more recently, with more subtle and chronic symptoms such as cough, abdominal pain or pruritus, associated with mild eosinophilia and Toxocara spp. seropositivity [4–6]. It is however assumed that most cases of human toxocariasis go unrecognized.

A definitive diagnosis of human toxocariasis can only be made by histological examination of infected tissue, demonstrating Toxocara spp. larvae within eosinophilic granulomatous lesions. Since sampling of appropriate tissue is rarely justified on clinical grounds, diagnosis of human toxocariasis relies in the daily practice on a constellation of suggestive symptoms (if present), combined with laboratory abnormalities (blood eosinophilia) and the detection of circulating immunoglobulin G (IgG) antibodies to Toxocara excretory–secretory (TES) antigens [7]. In Western countries with low prevalence of helminthic infections, the association of clinical symptoms (although frequently poorly specific) with eosinophilia usually triggers clinicians aware of the disease to perform the specific serological investigations [1]. In the tropics however, where polyparasitism is highly prevalent, the low specificity of eosinophilia, the lack of diagnostic facilities and the cross-reactivity with other helminthic infections often preclude the etiologic diagnosis [2].

While larvae may remain viable for years in the tissues, human toxocariasis is usually a self-limiting disease, although its course can sometimes be invalidating and prolonged (up to several weeks). Treatment firstly aims at controlling the inflammatory reaction when needed, and despite the limited knowledge about its clinical benefit, anti-helminthic therapy is often associated. In such cases, preference usually goes to albendazole for its parasitological efficacy and good clinical tolerance [8].

The global burden of human toxocariasis is poorly quantified [9]. Serological surveys demonstrate that the infection is more frequent in tropical settings and in rural areas, with population-based seroprevalence ranging from 2.5% in urban Europe up to 85% in rural tropics [9–14]. Visceral larva migrans has been reported in individuals born in tropical countries having migrated to Europe [15]. However, although many susceptible European travelers are assumed to be at increased risk of exposure to Toxocara spp. during a stay in highly endemic developing countries, frequency and presentation of toxocariasis are largely unknown in this population. A recent large multicenter GeoSentinel study has reported 16 cases diagnosed as visceral larva migrans among 42,173 travelers evaluated from 2007 to 2011 (38 cases per 100,000 travelers) but there was no standard diagnostic approach in the 53 participating travel clinics and no clinical description [16]. In the present study, we aimed to assess the frequency, presentation and outcome of toxocariasis diagnosed among travelers and migrants presenting at the travel clinic of the Institute of Tropical Medicine of Antwerp, Belgium.

Materials and Methods

Study Setting and Population

For this retrospective study, a query was first undertaken in the database of the Central Laboratory of Clinical Biology of the Institute of Tropical Medicine, Antwerp (ITMA), to retrieve all results of Toxocara serology requested in patients having attended the travel clinic of the ITMA from 2000 to 2013. The ITMA is the national reference center for tropical medicine in Belgium, with on average about 6500 consultations a year for post-travel care. The medical records of all travelers and migrants found with a positive Toxocara serology during this period were then reviewed. Relevant clinical and laboratory data were extracted, de-identified and entered in a Microsoft Access 2010 database. Variables included: demographic data, month and year of first Toxocara positive serology, most recent travel destination, time of symptom onset after travel, duration of symptoms before consultation, clinical presentation, result of the chest X-rays if requested, absolute blood eosinophil count (and percentage of white blood cell count), Toxocara antibody optical density result, results of parasitological and serological tests prescribed by the physician targeting other helminths according to epidemiological relevance (Ascaris spp., Echinococcus granulosus, Fasciola spp., Filaria spp., Schistosoma spp., Taenia solium, Strongyloides stercoralis, Trichinella spp. and Anisakis simplex), administered treatment(s), clinical and laboratory evolution and outcome.

Next, we categorized all reviewed Toxocara-seropositive cases in four groups according to the clinical reasons for requesting Toxocara serology: 1) presence of symptoms compatible with VLM combined with eosinophilia; 2) asymptomatic eosinophilia; 3) symptoms compatible with VLM without eosinophilia; and 4) possible exposure/no clear reason (Fig. 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Flow chart and distribution of study participants (VLM: visceral larva migrans).

https://doi.org/10.1371/journal.pntd.0003559.g001

Results

Epidemiological, Clinical and Laboratory Features of Toxocariasis Cases

The clinical presentation of the 28 cases is detailed in Table 1. Cases were evenly distributed throughout the study period with no cluster phenomenon. All patients were adult travelers born or residing in Europe, except one Ethiopian child evaluated after adoption and one Lebanese adult living in the Democratic Republic of the Congo. Mean age was 46 years (range: 4–68 years) with a male/female ratio of 0.87. Regions of most recent travels (and presumed exposure) were North and sub-Saharan Africa in 10 patients, Southern and Southeast Asia in 9 and Southern Europe (including Turkey) in 6; for 3 patients (numbers 18, 19 and 26), the continent of acquisition could not be traced with certainty because of multiple travel destinations within a short timeframe. Duration of travel was less than 1 month in 12 (43%) patients and more than 3 months in 10 (36%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Epidemiological, clinical, laboratory and outcome features of patients diagnosed with toxocariasis (n = 28) at the Institute of Tropical Medicine, Antwerp, Belgium from 2000 to 2013.

https://doi.org/10.1371/journal.pntd.0003559.t001

For the 23 cases presenting with VLM, symptoms started during the stay abroad in 11 (48%) or developed within 3 weeks in average (range: 0–8 weeks) after return. For one patient (number 11), the dates of recent travel were not clearly reported. Clinical manifestations included abdominal complaints in 11 (48%) patients, respiratory symptoms and skin abnormalities in 10 (43%) each and fever in 9 (39%); in most cases, symptoms were combined or developed sequentially. Radiological pneumonia was found in 5 patients and one of them (number 22) had to be admitted elsewhere because of the severity of respiratory symptoms. Two patients (numbers 11 and 16) presented with progressive neurological features of transverse myelitis but no other symptoms. In both cases, the diagnosis was made by the demonstration of an increased eosinophil count and of anti-TES IgG in the cerebrospinal fluid, in the absence of another etiology (Table 1).

For the 21 patients without neurological complications, median duration of symptoms before first documented medical evaluation at ITMA or elsewhere was 3 weeks (range 5 days- 4 months), while both patients with neurological toxocariasis were diagnosed several months (2,5 and 7 respectively) after symptom onset.

At first presentation, the median blood eosinophil count was 1720/μL (range: 510–14160) in the 21 VLM cases without neurological complication and 2080/μL (range: 1100–2970) in the 5 asymptomatic cases. Of note, the median eosinophil count was significantly lower in the VLM patients presenting more than 4 weeks after symptom onset than in those consulting earlier (700/μL versus 2340/μL, p<0.001). The blood eosinophil count was within the reference ranges for both patients with neurological toxocariasis.

Discussion

Although toxocariasis is highly prevalent in most tropical areas, this condition has been hardly studied in returning travelers and migrants. We describe here a case series of 28 patients diagnosed with toxocariasis acquired from all over the world. Clinical presentation was extremely varied and resembled that of many other endemic helminth infections. Morbidity was important and complications sometimes serious.

This study has many limitations. It was indeed a retrospective single-center study conducted in a reference travel clinic, meaning that collection of data was not systematic and that findings may not be generalizable to all clinical settings. For instance, some cases with longer incubation period may have been directly attended in the primary care setting given the fact that the link between symptoms and travel had become less obvious. Also our observations are not transposable as such to the features observed in autochthonous toxocariasis sporadically seen in Belgium or elsewhere, mainly in children [22]. Other limitations could be related to the restrictive case definition that may have missed several true cases without eosinophilia (as sometimes observed in milder cases of “covert/common” toxocariasis) or with negative Toxocara serology (during the serological window period or because sensitivity does not reach 100%, in particular in low burden infection such as ocular toxocariasis [2]). In the same line, some true toxocariasis cases may have been disregarded just because serological tests against other helminths were also positive, either by cross-reactivity or as part of infection with multiple parasites. Conversely, false positive result was also possible if anti-Toxocara seropositivity was just reflecting remote exposure or cross-reaction while another helminthic infection was missed during the workup. The commercial test we used is widely considered as an adequate screening tool for clinical practice but indeed detects anti-TES IgG that may persist for years; IgM or IgE-based serological tests that could better discriminate recent infection are not routinely available. On the other side, for European travelers with little previous exposure to parasites, cross-reaction is probably not a major issue, since test threshold has been set at value providing good specificity. In addition, the consistent clinical and laboratory diagnostic approach by a stable group of expert physicians throughout the study probably reflected the best accuracy that can be obtained in routine care. Finally, in this series, infection was most likely acquired abroad since symptoms developed during or shortly after travel, but infection in Belgium before travel or after return cannot be fully excluded.

The observation of toxocariasis in travelers is not surprising, although poorly studied so far. In a 10-year retrospective study in Spain, Toxocara antibodies were detected in 31 (4.9%) of 634 Latin American migrants and VLM was diagnosed in 4 of them [15], but this was in a particular segment of the travel population. With 28 highly probable/definitive toxocariasis cases diagnosed in about 85,000 travelers during the 13-year study period (33/100,000 travelers), our findings are in line with the recent multicenter GeoSentinel study (38/100,000 travelers) although the actual frequency was probably somewhat underestimated given the very restrictive case definition. Because of the high prevalence of toxocariasis in tropical countries and the inherent risks related to visiting regions with substandard hygiene (exposure to locally prepared food, incidental contacts with animals [23], it is reasonable to include toxocariasis in the differential diagnosis of most travel-related illnesses. However, surprisingly, the rate of Toxocara seropositivity in the (suspected) travelers for whom the test was requested (5.5%) was quite similar to that found in suspected autochthonous cases in Denmark (5.5%) [13] or the Netherlands (5–10%) [12] at similar ages. This observation does not support the idea that exotic travel by itself represents a major risk factor for Toxocara seropositivity, but since the frequency of clinical toxocariasis was not reported in those studies, comparisons with our findings remain inconclusive. Finally, only one study conducted in the Netherlands has investigated the incidence rate of Toxocara infection in travelers by comparing pre- and post-travel serology and found 1.1 seroconversion per 1000 person-months [24]. We confirm here that, even if infrequent, toxocariasis does occur in travelers and has to be considered after any type or duration of travel and from any destination.

Several factors contribute to underdiagnosis of toxocariasis, even in settings with higher resources [25]. Observed symptoms were often little specific and mimicked many other parasitic infections occasionally seen in travelers [26,27]. They were sometimes mild and self-limiting, not always triggering a complete etiological workup. Eosinophilia was rather high within the first weeks after symptom onset but tended to normalize quite rapidly. Laboratory investigations often detected evidence of other helminthic infection, with almost no possibility to discriminate between cross reaction, dual infection or remote exposure. Finally, we observed almost always excellent clinical and laboratory responses to albendazole, which is liberally used as empiric anti-helminthic treatment in travel medicine [28]. Clinicians often tend therefore to consider the specific diagnosis of toxocariasis as somehow difficult and of secondary importance.

Morbidity of toxocariasis should however not be underestimated. In the present series, most patients experienced a protracted illness, also because diagnosis was often delayed. Admission was necessary for three patients (10%), including both neurological cases for whom the diagnosis was particularly challenging. The clinical features of transverse myelitis had indeed developed insidiously, with no concomitant systemic symptoms and no blood eosinophilia at diagnosis, but well laboratory findings indicating an eosinophilic meningitis and non-specific alterations at the Magnetic Resonance Imaging of the spine [29–32]. Both cases did not fully recover despite maximal anti-helminthic therapy and prolonged corticosteroids. For all other cases however, administration of a 5-day course of albendazole was, when tolerated, immediately beneficial, suggesting that the current recommended practice is adequate [8]. This was not the case for ivermectin treatment with which clinical failures were observed [33]. Of note, no clinical exacerbation was observed during anti-parasitic treatment in contrast with what often occurs in other helminthic infections with larval invasion such as acute schistosomiasis [26]. Finally, we did not observe any case of ocular toxocariasis, but such cases are often serologically negative (due to low parasite load) and usually diagnosed in specialized ophthalmologic clinics [34].

In conclusion, symptomatic and asymptomatic toxocariasis was sporadically diagnosed in international travelers attending our center and had clinical and laboratory features overlapping those of many other tropical infections. In the present series, morbidity was non negligible and occasionally severe. A standard 5-day course of albendazole provided substantial clinical benefit without evidence of clinical exacerbation. Research is needed to develop antigen-based tests that would better reflect the disease activity both for diagnostic and monitoring purposes in clinical care.

Supporting Information