Demographics and travel history of imported and autochthonous cases of leishmaniosis in dogs in the United States and Canada, 2006 to 2019

Abstract Background Leishmania infantum infections are reported in foxhounds throughout the United States (US) and Canada, but only rarely in other dog breeds. A seroprevalence report from 2006 documented leishmaniosis in foxhounds (8.9%) tested in the US between 2000 and 2003. All other breeds were seronegative. Objective To reexamine demographics and travel history of L. infantum‐infected dogs in the US and Canada, we hypothesize detection of L. infantum in more foxhounds than nonfoxhounds and that infected nonfoxhounds will have traveled to endemic regions. Animals A total of 125 dogs positive for L. infantum by immunofluorescent antibody, PCR, or both. Methods Retrospective, descriptive study of L. infantum‐infected dogs between 4 January 2006 and 22 May 2019. Travel history and known lineage to foxhounds was collected from questionnaires. Results Leishmania infantum was detected in 125 (6.4%) of 1961 dogs tested between 4 January 2006 and 22 May 2019, of which 10 (8%) were foxhounds and 115 (92%) were nonfoxhound breeds. Travel history available for 69 (55%) dogs showed 60 (86.9%) dogs had traveled outside of the US or Canada. Nine (13%) dogs had not traveled outside of the US or Canada, 5 of which were nonfoxhounds. Conclusions and Clinical Importance The majority of L. infantum cases were detected in nonfoxhounds, many of which had traveled to L. infantum‐endemic countries, and several nonfoxhound breeds had no travel history. Leishmania surveillance should be considered for dogs that return from L. infantum‐endemic regions to monitor emergence of this zoonotic disease in the US and Canada.


| INTRODUCTION
Leishmania infantum causes leishmaniosis in dogs and visceral leishmaniasis in humans throughout the world. 1 Leishmania is transmitted by female phlebotomine sandflies, where endemicity depends on geographic location, climate, reservoir hosts, and competent vectors. [2][3][4] Leishmaniosis in dogs caused by L. infantum has been documented in foxhounds in the United States (US), with the first reported case in Oklahoma in 1980. [5][6][7][8] A serosurvey of L. infantum, primarily in foxhounds sampled between 2000 and 2003 in the US and Canada, reported an overall seroprevalence of 8.9%. 9 Infection with L. infantum has been documented by case reports in nonfoxhounds in the US, 8,10-13 but L. infantum was not definitively identified in nonfoxhounds in 2 US serosurveys. 9,14 A recent PCR-based survey conducted over 9 years on hunting hounds reported an average L. infantum prevalence of 20%, but specific breeds were not specified. 15 Given the lack of known, competent sandfly vectors in the US and evidence supporting vertical transmission in foxhounds, the latter is widely accepted as the only autochthonous method of L. infantum transmission within the US, primarily occurring in foxhounds. 16,17 Dogs could serve as epidemiological reservoirs for Leishmania emergence in the US and Canada as climate changes occur and sandfly ranges expand. The South American vector Lutzomyia anthrophora was predicted to migrate north through the US by 2020 based on environmental niche modeling. 18 It is conceivable that sandfly vectors transmitting L. infantum (also known as L. chagasi) could migrate northward as well. 19 A study showed that infected US foxhounds can serve as reservoirs for L. infantum via Lu. longipalpis, a known L. infantum vector native to Central and South America. 20 Lutzomyia shannoni, a sandfly native to Georgia, US acquired L. infantum after exposure to dogs with leishmaniosis, but it is unknown if Lu. shannoni can transmit L. infantum to another host. 21 Leishmania infantum screening is not required for dogs entering the US. 22 Because dogs are routinely imported into the US, consideration should be given to the risk they pose as reservoirs for leishmaniosis. 18,[23][24][25] To our knowledge, no surveys have evaluated the number of L. infantum-infected dogs imported into the US and Canada from Leishmania-endemic countries, only case reports are available. [26][27][28] Updated demographics on dogs infected with L. infantum in the US and Canada, which may not be confined to foxhounds, will inform regulatory agencies and veterinarians of a potentially increasing group of L. infantum reservoirs.
We aimed to describe demographic data and travel history for dogs with positive L. infantum serology, PCR, or both from a veterinary diagnostic laboratory. We hypothesized that L. infantum would be detected in more foxhounds than nonfoxhounds in the US and Canada, and that L. infantum-positive nonfoxhounds would have travel histories to endemic regions. to report demographics. Thirteen of the 125 positive dogs were tested more than once, but each dog was only counted as 1 dog in the analysis and considered either PCR negative, positive, or not tested, and IFAT negative, positive, or not tested. Dogs negative for L. infantum or dogs that were reported to be positive for Trypanosoma cruzi were excluded from analysis because of the potential for crossreactivity with L. infantum IFAT. For each dog positive for L. infantum by PCR or serology, attempts were made to contact the veterinarian on record by email. When email information was not provided, or when the referring veterinarian failed to respond to 2 emails, attempts were made to contact the clinic by phone. Dogs with veterinarians who did not reply to the questionnaire were excluded from analysis of travel history but not the study in general (Figure 1).

| Questionnaire and demographic data
Data were obtained and reviewed from the VBDDL database. Data included age, breed, and sex as reported by the referring veterinarian, location within the US or Canada, and date of sample submission for each dog. Region of sample origin was defined based on the US Census Bureau. 29 Additional data were collected using a questionnaire (Appendix S1) by email or phone contact with the primary veterinarian. Epidemiological information requested included known travel history outside of the US, reason for testing, clinical signs of leishmaniosis, whether the dog was simultaneously tested for T. cruzi, and familial ties to a foxhound bloodline. Veterinarians were interviewed about case history by phone or email for nonfoxhounds without travel history to obtain additional details about this unique subset of dogs.

| Immunofluorescent antibody test for Leishmania infantum
Serial, 2-fold dilutions of canine sera were made in phosphate-buffered saline (PBS) containing 0.05% Tween 20, 0.5% nonfat dry milk, and 1% normal goat serum (Fisher Scientific, Gibco cat# PCN5000) before adding 8 to 10 μL to slide wells prepared with cultured L. infantum  All qPCRs included a positive control consisting of either a previously characterized L. infantum-or L. guyanensis-infected sample, Leishmania kDNA plasmid DNA or sod plasmid DNA; and 2 negative controls including a no-template control consisting of filter-sterilized, moleculargrade water and an uninfected canine genomic DNA control.

| Retrospective testing to exclude potential Trypanosoma cruzi-infected dogs
Because antibodies against T. cruzi can cross-react with L. infantum, retrospective Leishmania PCR was performed on archived EDTA whole blood samples from L. infantum seropositive dogs that were only tested by IFAT and resided in the West South Central region of the US, where T. cruzi is prevalent. [31][32][33] Archived serum was not available for retrospective T. cruzi serology testing.

| Statistics
Descriptive statistics were used to calculate percentages, means, and ranges, when appropriate. Maps were created using ArcGIS (ArcMap  For 1 dog, the region of sample origin could not be determined.    foxhound (4/125; 3.2%), and unknown (7/125; 5.6%;   T. cruzi is known to be most common in the US. [31][32][33] Two dogs from this region were L. infantum IFAT positive; stored whole blood was available for PCR on 1 of these 2 dogs. This dog was positive for L. infantum by kDNA and sod PCR analysis, with amplicon sequence confirmation. Archived serum from these 2 dogs was not available for T. cruzi serology testing to evaluate for potential cross-reactivity.

| Demographics and travel history for L. infantum-positive dogs
More in-depth clinical information was obtained by phone or Leishmania infantum-exposed or infected dogs can and should then be treated and provided appropriate sandfly repellent products to limit the risk L. infantum transmission poses to humans and other dogs.  45 Overall, Leishmania PCR has been reported to be 100% specific, 45 indicating that a positive PCR with amplicon sequencing in this population likely reflects a true L. infantum-positive.
Specificity of L. infantum IFAT, however, is lower (approximately 91%-99%) and has the potential for cross-reactivity increasing the probability for false positives specifically in the population of IFAT positive and PCR negative (or PCR untested) dogs. 46 Antibodies against T. cruzi potentially can cross-react with Leishmania. 47 Only 1 dog that we were aware of underwent testing to identify T. cruzi exposure: a hound from the West South Central region (Texas). The dog was reported to be PCR negative for Leishmania by the primary veterinarian, with positive T. cruzi and L. infantum serology. This dog ultimately was excluded from analysis because of concern for cross-reactivity. The West South Central region (Texas, Louisiana, Oklahoma and Arkansas) was a focus of possible T. cruzi cross-reactivity because of the greater distribution of Chagas disease in the Southern US where its associated vector and reservoirs are found. 48 The remaining L. infantum-positive dogs from the West South Central region included 4 PCR and IFAT positive dogs, as well an IFAT positive dog where PCR testing was not performed. Although it is possible, this dog was only exposed to T. cruzi, additional evidence supported an infection with L. infantum, including clinical signs consistent with leishmaniosis and a history of travel to a L. infantum endemic country (Spain) before testing. Considering that over half of the L. infantum-positive dogs in our study were confirmed by PCR and sequence analyzes, and the remaining dogs tested did not reside in or travel to regions with a high prevalence of T. cruzi, we believe it is unlikely that cross-reactivity substantially influenced the outcome of our study.
In conclusion, we report that L. infantum was detected in more nonfoxhounds than foxhounds residing in the US and Canada, and that many of these dogs had traveled to, or been imported from, endemic regions. Our study also emphasizes the need for clinical veterinarians to obtain good travel histories, screen for L. infantum in dogs with travel histories to endemic regions and consider L. infantum in breeds other than foxhounds. We also identified 5 nonfoxhounds that had no travel history outside of the US and no known source of vertical transmission. This finding emphasizes the importance of L. infantum surveillance within the US and Canada so as to monitor the potential for endemic transmission. If vectorborne transmission is occurring on rare occasions in dogs in the US and Canada, it undoubtedly would have important public health implications for humans.

ACKNOWLEDGMENT
No funding was received for this study. This study was presented as an ePoster at the 2020 ACVIM Forum On Demand. The authors thank the technicians in the VBDDL, for sample handling and performing all Leishmania diagnostic assays, and participating veterinarians who provided additional information on Leishmania positive dogs.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.