Travel-acquired infections in Canada : CanTravNet 2011 — 2012

Background: Important gaps remain in our knowledge of the infectious diseases people acquire while travelling and the impact of pathogens imported by Canadian travellers. Objective: To provide a surveillance update of illness in a cohort of returned Canadian travellers and new immigrants. Methods: Data on returning Canadian travellers and new immigrants presenting to a CanTravNet site between September 2011 and September 2012 were extracted and analyzed by destination, presenting symptoms, common and emerging infectious diseases and disease severity. Results: During the study period, 2283 travellers and immigrants presented to a CanTravNet site, 88% (N=2004) of whom were assigned a travel-related diagnosis. Top three destinations for non-immigrant travellers were India (N=132), Mexico (N=103) and Cuba (N=89). Fifty-one cases of malaria were imported by ill returned travellers during the study period, 60% (N=30) of which were Plasmodium falciparum infections. Individuals travelling to visit friends and relatives accounted for 83% of enteric fever cases (15/18) and 41% of malaria cases (21/51). The requirement for inpatient management was over-represented among those with malaria compared to those without malaria (25% versus 2.8%; p<0.0001) and those travelling to visit friends and relatives versus those travelling for other reasons (12.1% versus 2.4%; p<0.0001). Nine new cases of HIV were diagnosed among the cohort, as well as one case of acute hepatitis B. Emerging infections among travellers included hepatitis E virus (N=6), chikungunya fever (N=4) and cutaneous leishmaniasis (N=16). Common chief complaints included gastrointestinal (N=804), dermatologic (N=440) and fever (N=287). Common specific causes of chief complaint of fever in the cohort were malaria (N=47/51 total cases), dengue fever (14/18 total cases), enteric fever (14/17 total cases) and influenza and influenza-like illness (15/21 total cases). Animal bites were the tenth most common diagnosis among tourist travellers. https://doi.org/10.14745/ccdr.v40i16a01


Introduction
In 2012, Canadians spent $36.5 billion on international tourism, up from $35.9 billion the year before (1). Top tourist destinations for Canadians continue to include tropical and economically developing countries such as Mexico, Cuba, the Dominican Republic and Jamaica (2). In 2012, of the 11,363,100 Canadians who stayed one or more nights at a destination other than the United States, almost a third travelled to one of those four countries.
International tropical travel puts travellers at risk of enteric and vector-borne infectious diseases (3)(4)(5), many of which are preventable through specific interventions such as chemoprophylaxis, immunization, insect repellents, personal protective measures and avoidance (4). A large-scale analysis of illness in returned Canadian travellers and new immigrants over a two-year period has recently been published by members of CanTravNet (6), provides Canadian practitioners with an epidemiologic roadmap of travel-acquired infections, which can be used to inform decision-making in both the pre-travel and post-travel setting.
This surveillance report provides an update to the two-years' worth of CanTravNet data published previously (6) and highlights the breadth of illnesses encountered by Canadians visiting >130 countries over a one-year period and presenting for care at a CanTravNet site.

Data sources
Six Canadian sites from four provinces (British Columbia, Alberta, Ontario and Quebec) also belonging to the GeoSentinel Global Surveillance Network have grouped together to form the core sites of CanTravNet (6). The six sites in Canada are large referral-based outpatient centres that primarily service the Greater Vancouver/Victoria, Calgary, Toronto, Ottawa and Montreal areas, which account for 47% of the Canadian population (or, a catchment of ~15.5 million people). They are staffed by specialists in travel and tropical medicine and immediate referral from the affiliated emergency departments is common.
Data were collected using the GeoSentinel data platform. This network is comprised of 56 specialized travel/tropical medicine clinics on six continents, which contribute denominalized clinician-and questionnairebased travel surveillance data on all ill travellers examined to a centralized Structured Query Language database (7). (For additional details see http://www.istm.org/geosentinel ). Collected data include patient demographics, details of recent travel, five-year travel history, purpose of travel and presence of absence of a pre-travel encounter with a healthcare provider. Final diagnoses are made by attending physicians and assigned a diagnostic code selected from a standardized list of >500 diagnostic entities, including etiologic (e.g. Giardia) and syndromic (e.g. cough) diagnoses. Syndromic codes are entered where an etiologic code cannot be assigned due to use of empiric therapy, self-limited disease, or inability to justify a more extensive workup as part of routine clinical practice. All CanTravNet sites contribute microbiologically confirmed data, where available, based on the best national reference diagnostic tests (including molecular diagnostics) available at the time. 'Probable' diagnoses are restricted to patients with pathognomonic physical findings (e.g. tick eschar), clinical response to highly specific therapy, or classical presentation and exposure history with laboratory exclusion of other possible etiologies (6). Further details regarding CanTravNet can be found at http://www.istm.org/cantravnet.

Definitions and classifications
Reason for most recent travel. Six possible travel purpose designations are available: immigration (including refugee), tourism, business, missionary/volunteer research/aid work, visiting friends and relatives, and "other" (students, military personnel and medical tourists). A visiting friends and relatives (VFR) traveller is defined as an immigrant who is ethnically and/or racially distinct from the majority population in their current country of residence and who returns to his homeland to "visit friends and relatives". VFR travellers also include children of foreign-born parents (e.g. second generation immigrants) who return to their parent's homeland to visit friends and relatives. A VFR traveller designation is typically applied to individuals travelling from a high-income country of current residence to a low-income country of origin (8). "Medical tourists" are defined as those for whom the primary purpose of travel is to seek health care, and as a consequence of travel, acquire an infectious complication secondary to the medical care received or become ill with an infectious or non-infectious disease while abroad.
Countries of exposure and travel were assigned 14 regional classifications: North America, Central America, the Caribbean, South America, Western Europe, Eastern Europe, the Middle East, North Africa, Sub-Saharan Africa, South Central Asia, Southeast Asia, Northeast Asia, Australia/New Zealand and Oceania.

Inclusion criteria
Demographic, clinical and travel-related data on Canadian citizens and new immigrants to Canada encountered after completion of their international travel or residence abroad and seen in any of five CanTravNet sites from September 2011 to September 2012 were extracted and analyzed. (The Calgary site was new to GeoSentinel as of 2012 and did not contribute cases during the study period.) Only patients with probable or confirmed final diagnoses (specific etiology or syndrome as described previously (6)) were included.

Descriptive analysis
Extracted data were managed in a Microsoft Access database and analyzed using standard parametric and nonparametric techniques. Travellers were described by purpose of travel, demographics, diagnoses, country of exposure and region of travel. Top syndromic and etiologic diagnoses were described for each purpose of travel. Top chief complaints were described by represented causative diagnoses and top source countries. Comparisons between categorical variables (e.g. purpose of travel) were made using Yates' corrected chi-square analysis, while continuous variables (e.g. age) were analyzed for significant differences using the Student's t-test and in the case of non-normally distributed parameters, the Mann-Whitney Rank Sum test. For a particular variable (e.g. purpose of travel or diagnosis), the reference population was all other travellers in the cohort without that variable (e.g. malaria versus non-malaria). Differences between groups of continuous variables were compared using One Way ANOVA or Kruskal Wallis One Way ANOVA on ranks. All statistical computations were performed using SigmaStat 2.03 software (SPSS Inc., Chicago, IL). Level of significance was set at p<0.05.
Non-immigration travellers in the cohort (i.e. all travellers in the cohort except those travelling for the purpose of immigration) (N=1511/2004) for whom exposure country was known (N=1349) visited 133 different countries, the most frequently visited of which included: India (N=132, 9.8%), Mexico (N=103, 7.6%), Cuba (N=89, 6.6%), Dominican Republic (N=71, 5.3%) and Thailand (N=49, 3.6%). Includes students (n = 42), military personnel (n = 6) and medical tourists (n = 2). † Among those born outside of Canada, people who travelled for the purpose of visiting friends and relatives were defined as immigrants who were ethnically and/or racially distinct from the majority population in their current country of residence and who returned to their homeland to visit friends and relatives. This group also included children of foreign-born parents (e.g. second generation immigrants) who returned to their parents' homeland to visit friends and relatives.

Diagnoses
A total of 2402 travel-related diagnoses were issued to 2004 ill returned travellers. Of these diagnoses, 2078 were confirmed and 324 were probable. The most frequently issued travel-related diagnoses in persons travelling for the purpose of immigration were: latent tuberculosis, chronic hepatitis B, active tuberculosis, chronic hepatitis C and strongyloidiasis ( Table 2). The most frequently issued travel-related diagnoses among non-immigration travellers were: post-infectious irritable bowel syndrome, acute diarrhea, chronic diarrhea, skin and soft tissue infections and arthropod bites ( Table 2). Table 3 lists the top travel-related diagnoses and countries of exposure for travellers presenting with a chief complaint of fever, gastrointestinal symptoms and skin disease. Table 4 describes cases of malaria among the 2004 ill returned travellers.    Malaria was the top specific diagnosis for those travelling for the purpose of visiting friends and relatives and was the fifth most common diagnosis among business travellers and missionaries, researchers, volunteers and aid workers ( In addition to malaria, enteric fever was also over-represented among those travelling for the purpose of visiting friends and relatives (p<0.0001) compared to other types of travellers. Cases of enteric fever due to Salmonella enterica serotype Typhi (N=2) and Paratyphi (N=2), as well as hepatitis E virus (N=1) and hepatitis A virus (N=1) were all represented among children who were visiting friends and relatives. The proportion of ill returned travellers who had been visiting friends and relatives who required inpatient management of their travel-acquired illness (12.1%) was more than five times that of ill returned travellers who had not been visiting friends and relatives (2.4%) (p<0.0001). VFR travellers also had the lowest proportionate uptake of pre-travel consultation among all ill returned non-immigrant travellers (p<0.0001) ( Table 1). In addition, those who were visiting friends and relatives travelled for longer periods of time compared to travellers who had not visited friends and relatives (median 36 versus 18 days; p<0.001).
Other emerging, life-threatening and notifiable diseases were represented among the cohort of ill returned travellers. There were nine cases of newly diagnosed or acute (febrile) HIV infection, three of which occurred in non-immigration travellers and a single case of acute hepatitis B in a tourist traveller. Hepatitis E is an emerging infection among travellers and there were six cases diagnosed in this cohort, with 4 (67%) occurring in males, three (50%) in VFR travellers and two (33%) requiring inpatient management. The age range of hepatitis E cases was 4 -63 years. Five of six cases of hepatitis E virus were acquired in South Central or Southeast Asia. Three cases of brucellosis were diagnosed in immigrant and VFR travellers. Trip duration was known for one case and was 177 days, consistent with the long duration of exposure typically associated with travel-related brucellosis.
Cutaneous larva migrans appears to be an emerging disease among ill returned travellers, with 27 of 43 (62.8%) cases occurring in travellers returning from the Caribbean. Jamaica was the top country of exposure for cutaneous larva migrans, with 18 (42%) cases acquired on this island. Ninety-three percent of cutaneous larva migrans cases (N=40) occurred in tourist travellers on short-stay trips (median trip duration 8.5 days). Of 16 cases of cutaneous leishmaniasis, 69% were imported from Central or South America and in particular, 50% were acquired in Costa Rica.

Discussion
Analysis of surveillance data on ill returned travellers presenting to a CanTravNet site between September 2011 and September 2012 has revealed the spectrum of travel-acquired illness encountered at CanTravNet sites. These data provide an update to the largest surveillance report on illness in Canadian travellers (2009 -2011) (6).

Potentially life-threatening illnesses can occur in ill returned travellers
Malaria remains the top specific cause of fever in ill returned travellers (9,10) and in this study, was caused by the potentially life-threatening P. falciparum in 60% of cases. Ill returned travellers with malaria were eight-times as likely to require inpatient management compared to those with alternate diagnoses. Malaria is a preventable infection, yet, of patients with malaria who clearly travelled to a risk area, only 20% sought pre-travel advice and only 6% received malaria chemoprophylaxis.
It has been previously demonstrated that travellers are more likely to be exposed to blood and body fluids while travelling than at home (11,12). We noted nine cases of acute or newly diagnosed HIV in our cohort, one third of which occurred in non-immigration travellers, as well as one case of acute hepatitis B in a tourist traveller. Hepatitis B, in particular, remains a risk to international travellers, despite being almost completely vaccinepreventable (13)(14)(15). Our data reiterate the importance of pre-travel strategies which aim to reduce the likelihood of blood and body fluid exposure while travelling (16), in addition to strategies which mitigate the risk of food-and water-borne and vector-borne diseases.
Animal bites were the tenth most common diagnosis among tourist travellers and were acquired in countries such as Thailand, Indonesia, Mexico, Costa Rica and Chile, which may not have readily available human rabies immune globulin (HRIG) or vaccine that meets minimum potency standards set forth by the World Health Organization (WHO) (17). As rabies is virtually always fatal, ensuring timely access to full post-exposure prophylaxis is essential, yet it is rarely available in rural developing world settings (17). Prevention of animal bites altogether eliminates the risk of rabies and obviates the need for access to post-exposure prophylaxis and, along with rabies pre-immunization, should therefore serve as a target for pre-travel intervention in the consultation setting.
Travellers visiting friends and relatives are at risk VFR travellers are known to acquire specific travel-related illnesses more frequently than others, likely because these travellers tend to stay in local homes, travel for longer durations and may fail to recognize the health risks inherent to travel to their country of origin (3,4,7,8,18). Although VFR travellers comprised only 10% of this cohort, they accounted for 83% of cases of enteric fever and 41% of cases of malaria. In addition, the proportion of VFR travellers requiring inpatient management of their travel-acquired illness was five-fold higher than ill returned travellers who had not visited friends and relatives, which may simply reflect that those visiting friends and relatives are more likely to seek care for more serious illness, rather than benign etiologies, as opposed to being more likely to acquire more serious illness. Those who visited friends and relatives travelled for a longer average period of time, yet, they were least likely of all travellers to have sought pre-travel advice. The data underscores the unique characteristics of these of travellers and the urgent need to identify strategies to minimize their travelacquired morbidity.

Emerging infections among travellers are difficult to prevent
Hepatitis E is a water-borne virus with large epidemics reported from Central America, Sub-Saharan Africa, the Middle East and Asia (19,20). In our cohort, 83% of cases were acquired from the Indian sub-continent or Southeast Asia and again, those visiting friends and relatives were the most well represented type of traveller, suggesting that, as is the case for hepatitis A and enteric fever (21), more prolonged, rural and in-home travel may be a risk factor for hepatitis E acquisition. Hepatitis E infection is a particular danger for pregnant women (20), especially in the third trimester, with associated maternal mortality as high as 25% (22). Pregnant travellers should therefore be advised of this risk when travelling to endemic countries (19) and food-and water-precautions should be emphasized.
Chikungunya fever has recently emerged in the Americas (23) and it is anticipated that Canadian physicians will increasingly encounter this disease (24). Prevention of chikungunya rests on mosquito avoidance measures, principally repellents, which are often deemed unappealing by travellers (25)(26)(27)(28). Cutaneous larva migrans, while not life-threatening, causes considerable morbidity, lost productivity and costly medical encounters due to the severity of pruritus and lack of ready access to the only effective medications, albendazole and ivermectin, which are not currently licensed in Canada. These drugs can only be acquired in Canada through the Special Access Programme of Health Canada, which necessitates a paper or electronic application for approval and has a processing time of at least one week for each request http://www.hc-sc.gc.ca/dhp-mps/acces/drugsdrogues/sapg3_pasg3-eng.php. Prevention of cutaneous larva migrans rests on avoidance of barefoot or bare skin exposure to sand, which is difficult to achieve in beach destinations where the causative organism is prevalent. Cutaneous leishmaniasis, an emerging vector-borne disease among travellers (29), is challenging to prevent as it requires absolute bite avoidance in destinations with typically high ambient temperatures and humidity, where the use of long clothing and repellents might be deemed inconvenient (25)(26)(27)(28). Insecticide treated bed nets and sleeping several feet above the ground may provide some protection.

Limitations
Analysis of CanTravNet data has several limitations. This report focuses only on those ill returned travellers who presented to a CanTravNet centre, thus, conclusions may lack external validity. It must be noted that many of the top illnesses in those travelling for the purpose of immigration, including latent tuberculosis, hepatitis B and hepatitis C, would have been diagnosed through screening of at-risk individuals from endemic areas and cannot be definitively linked to travel. Travellers with illnesses with very short or very long incubation periods may have sought care in different settings and these diagnoses are difficult to definitively link to travel. Similarly, ill travellers returning from destinations perceived to be low-risk may be under-represented in the CanTravNet database. The data cannot estimate incidence rates or destination-specific numerical risks for particular diseases (7,30). Variation among sites regarding screening protocols for new immigrants and refugees may have led to over-or under-contributions of particular diagnoses from individual sites. Fifty-three percent of cases were contributed by Montreal sites, which may have introduced bias given the inter-Provincial variation in travel patterns and preferences.

Conclusions
This surveillance report aims to better inform pre-and post-travel management and to illuminate changing patterns of imported diseases. Malaria remains the top specific diagnosis among travellers visiting friends and relatives and although still mostly acquired in Sub-Saharan Africa, India was the top single source country of imported malaria in this cohort. In addition to malaria, other preventable travel-acquired illnesses such as enteric fever, influenza, hepatitis B and animal bites were common and reinforce that improved translation of knowledge into action on the part of the traveller should be prioritized. In addition, barriers to the uptake of pre-travel consultation by particular risk groups, such as VFR travellers and barriers to the use of preventive interventions, such as insect repellent and malaria chemoprophylaxis, should be assessed in the travelling Canadian population.