One health approach on serosurvey of anti-Leptospira spp. in homeless persons and their dogs in South Brazil

Although leptospirosis has been described as a worldwide bacterial zoonosis primarily affecting vulnerable populations, to date no study has focused on concomitant serosurvey of homeless persons and their dogs. The aim of the present study was, to use a One Health approach to serologically assess homeless persons and their dogs in 3 major cities of south Brazil (São Paulo, Curitiba, and Foz do Iguaçu). Environmental information was obtained with an epidemiological questionnaire given to all participants. A total of 200 human and 75 dog samples were tested for anti-Leptospira spp. antibodies to thirty different serovars using the microscopic agglutination test. None of the homeless persons were positive while 5 of the 75 (6.7%) dogs were positive. Among homeless population, 89% (177 of 200) were male, 61% (122 of 200) self-declared Non-white, and 67% (134 of 200) were educated up to the 8th school grade. Lower exposure of homeless persons to Leptospira spp. in the present study when compared to other vulnerable populations (slum and low-income residents) may be result of less direct exposure as they are able to rapidly change locations in response to flooding events. In addition, these results may reflect the effectiveness of a specific healthcare service provided to people living in the streets in the 3 cities. While dogs may be used as environmental sentinels for leptospirosis, the low seropositivity results found in this report may indicate low transmission risk to homeless owners in direct daily contact with their dogs.


Introduction
Leptospirosis, caused by a bacterium of genus Leptospira, is considered one of the neglected tropical diseases of public health importance worldwide [1]. Transmission may occur through contact with contaminated urine of sewer rats (Rattus novergicus) that serve as the main reservoir in urban settings, or through exposure to soil or water contaminated with Leptospira spp. [2]. This disease affects vulnerable populations in tropical countries due to socio-environmental issues including inadequate infrastructure and sanitation, water and soil contamination, garbage accumulation, and rodent proliferation; all of which are common scenarios in Brazilian slums [3,4].
Homeless populations worldwide live without adequate or permanent housing and have little access to public and private resources. These populations are susceptible to multimorbidity of infectious and mental illness, substance abuse, stigmatization, and interpersonal violence, and social exclusion [5]. In Brazil, the homeless population is estimated at 140,559 persons, primarily living in poor conditions of highly populated urban cities, along with 13,151 slums distributed in 734 cities of the 27 Brazilian states [6].
Dogs may share the same ecological bioaccumulation environment as owners, and may act as sentinels or reservoirs for several zoonotic diseases [7]. Among these, leptospirosis may be life-threatening to both humans and dogs as it can lead to multiple organ involvement and death in severe cases [8]. Although the World Health Organization (WHO) has defined neighborhood dogs as semi-dependent on multiple families for living maintenance [9]. The role of stray dogs and dogs owned by homeless persons in zoonotic diseases remains to be fully established, particularly in urban settings. Accordingly, the aim of this study was to serologically assess homeless persons and their dogs in São Paulo, Curitiba, and Foz do Iguaçu, 3 major cities of south Brazil.

Study area and sample collection
The study herein was conducted in 3 major cities of south Brazil. São Paulo ( Homeless persons were contacted through local official health services and voluntarily participated by signed consent for themselves and their dogs. An epidemiological questionnaire was given to participants to assess environmental information while dogs were clinically examined. Municipality nurses collected blood from the people and certified veterinarians collected blood from the dogs. Serum samples were obtained by centrifugation, stored at − 20 • C, and tested by microscopic agglutination test (MAT) to detect anti-Leptospira spp. antibodies, according to the Brazilian Ministry of Health [10] protocols.

Statistical analysis
The data were tabulated, descriptive analysis was performed with the Epi Info version 7 statistical software (CDC, Atlanta, GA, USA), frequency distributions with 95% confidence intervals were calculated, and results were organized for presentation.

Results
Of the 200 humans samples, there were 119 (59.5%) homeless persons sampled in São Paulo, 59 (29.5%) in Curitiba, and 22 (11.0%) in Foz do Iguaçu. No human samples were positive for the presence of anti-Leptospira spp. antibodies. While answers to the epidemiological questionnaire were obtained and gathered (Table 1), no associated risk factor was statistically significant due to the absence of seropositivity.

Discussion
To the authors knowledge, this study is the first epidemiological serosurvey to date of Leptospira spp. in homeless persons worldwide. Previous studies were case reports of Leptospira spp. infection in homeless or outdoor persons in London [12], Tokyo [13], Florida [14], Lisbon [15], and Marseille [16].
Although negative results for human leptospirosis in homeless persons of 3 major Brazilian cities was a pleasant surprise, human leptospirosis has sustained high endemic levels in Brazil, mostly in urban areas, with an average annual report of 3810 cases (1.9 cases/100,000), according to a recent 16-year (2000-2015) retrospective survey by the Brazilian Ministry of Health [17]. São Paulo State accounted for 11,884 cases, corresponding to 21% of the leptospirosis cases nationwide [17].
As a vulnerable and outdoor population, homeless people have greater environmental exposure to infectious diseases. Our research group has recently shown a 55% seropositive rate for COVID-19 among homeless persons in São Paulo, the highest prevalence worldwide at the time [21]. However, homeless people may be less directly exposed to leptospirosis, as they are able to rapidly change locations in response to flooding events. In comparison, residents of slums or low-income flooding areas who normally refuse to evacuate may be exposed to overflowing open sewers, which are 3-times more likely to contain pathogenic Leptospira spp. and have 6-times more pathogen load [22] compared to areas with closed sewers. Thus, as floods in tropical countries may directly increase Leptospira spp. infection and outbreaks [23], the lack of fixed housing may allow homeless persons to avoid infection by migrating within the city itself [24], and therefore decrease their exposure to the consequences of floods. Despite the study herein has not surveyed environmental Leptospira spp., a recent systemic review has shown that leptospiral organisms may grow into a biofilm in both nutrient-free and complex microbiota environments, remaining virulent for months, particularly in soils and sediments [25]. Thus, populations environmentally exposed to Leptospira spp., such as homeless persons and their animals, are likely highly exposed in endemic areas such as the three surveyed cities of São Paulo, Curitiba and Foz do Iguaçu.
Despite living outdoors, the homeless persons in this study had low (absent) environmental infection of Leptospira spp., similar to lower seropositivity for Toxoplasma gondii of other vulnerable Brazilian populations [26]. In that study, 36% of homeless persons were seropositive for T. gondii, compared to 57% of people in a riverside community and 80% of indigenous individuals [26]. While homeless persons in this study were less exposed to bacterial Leptospira spp. infection probably due to their ability to quickly relocation during flooding, their lower exposure to protozoan T. gondii infection was mostly attributed to habits of eating mainly processed food and lacking fresh vegetables and raw or undercooked meat in their daily diet. When considering vector-borne diseases, a previous study has indicated that homeless populations of Houston, Texas, USA may be at risk for Chagas disease, due to outdoors exposure to vectors including triatomines and blood-borne pathogen risk behaviors, such as drug use [27]. Further studies should be conducted to establish the occurrence of zoonotic pathogens among homeless persons, particularly those of environmental exposure.
The One Health approach herein has shown seropositive dogs with no concurrent human exposure, indicating that each pathogen may have different animal and environmental risks for causing human infection, and all should be surveyed together. In contrast, homeless persons are reportedly more exposed to diseases transmitted from human-to-human, such as sexually transmitted infections from unprotected sex and needlesharing behavior (syphilis, HIV, and hepatitis [28], and COVID-19 [21]. Despite the lack of positive results for leptospirosis in homeless persons in this study, lack of adequate water and sanitation should be considered intrinsic human rights [29]. In addition, housing itself has been a social determinant of health [30] and health equity [31]. Negative human leptospirosis in this study may also be attributed to the Brazilian Unified Health System in the 3 cities operating the Street Clinic (healthcare for people living in the streets), which attempts to improve access to health services, a lack of which may substantially increase homeless persons vulnerability [32].
Besides the human leptospirosis cases, all 3 cities also had confirmed dog cases of leptospirosis [33]. However, the 7% frequency of positive dogs we found was lower than the 11% of street and shelter dogs in São Paulo [34], and 14% of dogs from a slum area in Curitiba, Parana State [34]. In a downtown urban area of Parana State, 2% of owners and 21% of dogs were serologically positive [35]. Although the frequency we found for dogs may have been lower due to a relatively high vaccination rate (47% of dogs), results were corroborated by absence of seropositive homeless persons. In addition to positivity, one dog presented high titers (400 and 1600) to two serovars within serogroup Icterohaemorragiae, the same as previous studies mentioned [35,36].
As limitations in the present study, human and dog leptospirosis were solely surveyed by serological methods, as association of serology and molecular tests may improve sensitivity and specificity of leptospirosis [37]. Nonetheless, a comparative study of MAT and PCR in acute leptospirosis found similar detection results [38].
As limitations in the present study, human and dog leptospirosis was surveyed only by serology with no molecular testing that might have increased the rate of detection. In addition, samples were obtained through contact with local health services, which may have missed new, non-registered and homeless persons refusing health assistance. The high percentage of vaccinated dogs may have indicated a biased sampling of well-treated dogs and a well-assisted homeless population. Finally, despite being a multicentric study, relatively low sampling per city may impair extrapolation of these results to other homeless populations in Brazil or worldwide. As a complex disease, leptospirosis has reportedly affected mostly vulnerable and low-income populations with several socio-environmental issues [39], which includes homeless populations. Thus, further studies should be conducted to confirm our results, including more sampling of homeless persons, other major cities in different regions, and different locations within each city. In summary, lower exposure of homeless persons to Leptospira spp. in the present study when compared to other vulnerable populations (slum and low-income residents) may be the result of less direct exposure, as they are able to rapidly change locations in response to flooding events. In addition, such results may reflect the effectiveness of a specific healthcare service (Street Clinic) provided to people living in the streets in the 3 cities. While dogs may be used as environmental sentinels for leptospirosis, low seropositivity results in this study could indicate low transmission risk of direct daily contact with homeless owners.

Funding
No funding.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.