Seroprevalence and risk factors associated with visceral leishmaniasis in dogs in Jaciara , State of Mato Grosso

Introduction: Visceral leishmaniasis is a neglected disease. Jaciara, State of Mato Grosso has the most cases. This study aimed to determine the risk factors and seroprevalence of canine visceral leishmaniasis in urban and rural areas. Methods: This cross-sectional study of domestic dogs used enzyme-linked immunosorbent assay and indirect immunofl uorescence test. Results: The prevalence was 54.7% among 345 analyzed samples. Short coat, age of 1-6 years, and living less than 100m from vegetation posed the highest infection risks. Conclusions: Certain dog behaviors and characteristics, and their correlation with environmental conditions, were relevant in the high prevalence of canine leishmaniasis in Jaciara.

Visceral leishmaniasis (VL) is a neglected disease.Approximately 3,500 cases are reported annually, and approximately 10% of these lead to death 1 .In Mato Grosso, canine visceral leishmaniasis (CVL) has been identifi ed in 41 municipalities with an average seroprevalence of 9% (1998-2005).The municipality of Jaciara has the highest recorded rate of seropositive dogs (40%) and one of the highest rates of human infection (5.8%) 2 .Lutzomyia longipalpis and Lutzomyia cruzi are vectors incriminated in the transmission of Leishmania chagasi (syn.Leishmania infantum) 3 .
From an epidemiological point of view, dogs are important in the urban environment because of the higher prevalence of the disease in this setting, and a large number of asymptomatic dogs harbor the parasite in their skin and thereby provide a source of infection for the vector 1 .Epidemiological studies indicate that the infection is associated with particular risk factors such as the length of an animal's coat, the areas in the house frequented by the dog 4 , and the socioeconomic and demographic indicators of the region 5 .The aim of this study was to determine the seroprevalence of CVL in urban and rural areas by identifying the risk factors involved in the transmission of this disease in the municipality of Jaciara, State of Mato Grosso.
A cross-sectional study was conducted from June 2010 to May 2011 in the municipality of Jaciara (16º 02' 30" S; 54º 59' 45" W), which is situated in the southern region of the State of Mato Grosso.The town is located in the Cerrado biome; its climate is predominantly semi-humid tropical; and its population is 25,666 inhabitants 6 .
Samples were collected in areas with a history of human cases and, on request, in the following neighborhoods in urban areas: Center, Jardim Aeroporto, Jardim Aurora, Jardim Esmeralda, Jardim Clementina, Jardim Leblon, Jardim Vitória, João de Barro, Nova Jaciara, Santa Luzia, São Lourenço, Santo Antônio, São Sebastião, Santa Rita and Vila Planalto (Figure 1); samples were also collected in the following rural areas: Brilhante, São Francisco, 25 de Novembro, Celma District, and seven farms on the BR 364 highway.
The sampling was estimated using a confi dence interval (CI) of 95% and an acceptable error of 5%, assuming a corrected fi nite population of 4,015 (Municipal Health Department: Personal Communication, 2010) and a prevalence of 40% 2 .The owners signed voluntary, informed consent forms to allow clinical examination of their animals and collection of samples from the animals, and to provide information on characteristics such as the origin of the dog, its breed (i.e., defi ned breed or mongrel), sex, castration status (i.e., yes or no), age (<1 year, ≥1-6 years or >6 years), length of coat, time living in the house, access to the streets and rural area (i.e., urban dogs), household status (i.e., indoor or outdoor dog), presence of other dogs and animals, vaccination against the disease, use of repellent collar on the dog, and environmental characteristics.In addition, the population's knowledge of the disease was investigated.The dogs were examined and classifi ed as symptomatic or asymptomatic 7 .Blood was collected by jugular venipuncture.After centrifugation, the serum samples were put into microtubes and stored at -20°C until the serological tests could be performed.These consisted of the immunoenzymatic assay (ELISA) and the indirect immunofluorescence test (IFAT), which were performed using commercial kits (Bio Manguinhos/FIOCRUZ, Rio de Janeiro, Brazil) in accordance with the manufacturer's recommendations.The IFAT is considered the gold standard 1 .
Statistical analyses were performed using the Epi Info software (Center for Disease Control and Prevention [CDC], Atlanta, GA, USA), the chi-square or Fisher's exact test, and logistic regression model.The level of significance was 0.05.The agreement between the techniques was investigated using the kappa index.
This study was conducted in accordance with the ethical principles approved by the Ethics Committee for Animal Research of Federal University of Mato Grosso [Comitê de Ética no Uso de Animais-Universidade Federal De Mato Grosso (CEUA-UFMT)] under protocol number 23108.024440/10-0.
When analyzing the chance that the dogs in each region would acquire the infection, the risk was three times higher in urban areas than in rural areas (OR = 3.23; 1.62 < OR < 6.53).
Table 1 presents the risk factors observed in this study.There were no statistically signifi cant differences between any of the areas studied with regard to sex, breed, being born in the home, having free access to the streets or to a rural environment (in the case of urban dogs), the presence of other dogs, the death of other animals with CVL that lived at the location, or the presence of a sick animal or person in the neighborhood (p > 0.05).
Control measures such as the use of repellent collars and/ or environmental insecticides and vaccines against CVL, and animal population control through castration were also investigated.Only a small number of owners used these measures on their animals (6.3%, repellent collars; 0.2%, vaccines against CVL; and 1.1%, castration).
The prevalence of CVL in Brazil varies greatly by region 8 .The reported prevalence within the State of Mato Grosso is 3.4% in Cuiabá 5 and 40% in Jaciara 2 .In our study area, the prevalence was 54.7%, thus exceeding the 40% found previously 2 using the same methodology.Different frequencies of infection in humans and dogs are to be expected because of the use of different diagnostic methods and the temporal variability and differences in the dynamics of transmission between the canine and human popula tions 9 , and because of differences resulting from the means used to obtain samples since some samples were obtained by request.
With regard to the epidemiology of VL, occurrences were commonly reported in rural areas in the past.However, an increasing urbanization of the disease has been observed 8,10 .This was also observed in the present study: the risk of infection was three times higher among dogs living in urban areas.
Urbanization is associated with ecotypes that are favorable to the development and adaptation of the vector.This characteristic was also present in Jaciara because many of its dwellings are close to natural environments.Associations with other risk factors for infection were also found such as the presence of plantations (e.g., fruit trees), the proximity of homes with dogs to forests, wastelands, lush vegetation, and areas that had recently undergone environmental change, all of which are conditions suitable to the development of the vector 11 .
The Brazilian Ministry of Heath has recommended using two serological techniques -ELISA as the screening test and IFAT as the confi rmatory method -because this approach has higher sensitivity and reduces the number of false positives 12 .This had previously been recommended by the Ministry of Health, but the techniques now used are ELISA and Dual Path Platform technology (DPP® CVL rapid test/Bio Manguinhos/ FIOCRUZ, Rio de Janeiro, Brazil).The latter employs the immunochromatographic principle, which is fast and easy to use 13 .The agreement between the two techniques was weak, but the prevalence of high titers of antibodies in IFAT showed that infection was present in these animals, thus decreasing the chances of cross-reactions 7 .This justifi es the use of this technique for analyzing risk factors.Isolation in a culture medium for the identifi cation of the species has been recommended; however, the technique was not used in this study.
The technique that was used was more sensitive for detecting the symptomatic dogs among the reactive dogs (p < 0.05).Specifi c antibodies can often be detected before the disease appears and their concentrations may be positively associated with the presence and severity of clinical signs, although some of the dogs may remain asymptomatic for more than fi ve years 14 .Among the symptomatic dogs, the most frequent signs have commonly been cited by other authors 5 .
In the evaluated samples, there were no statistically signifi cant differences regarding sex or breed (p > 0.05); this was also observed in Cuiabá 5 .However, an age greater than 1 year was associated with a higher infection rate.This is corroborated by other epidemiological studies that have shown that the peak incidence of CVL seropositivity occurs at the age of 2-5 years, which is possibly associated with a greater time of exposure to the vector 14 .
The dogs in the municipality of Jaciara (i.e., an urban area) presented a higher risk of being seropositive because this region is endemic for the disease 3 .This also confi rms the autochthonous nature of CVL in this municipality, as does the fact that the dogs had been living in this locality for more than 1 year.Situations in which household animals remain outdoors in areas surrounding the house have been considered a risk factor by other authors 8,14 .This was also observed in the present study.
Short-coated dogs have twice the risk of infection as longhaired dogs because the protection afforded by long hair hinders biting by phlebotomines 4 ; In fact, all dogs in the rural areas were had short coats.
The absence of henhouses or pigsties was a protective factor against infection and may have been associated with a feeding preference among the vectors for poultry or other animals that can act as a source of blood for female sandfl ies 15 , and thereby favor continuation of the vector in this environment.
Issues relating to health education are important in the fi ght against this disease.Although, the knowledge of the owners about the sandfl ies be related to increased risk of infection, a result of exposure to the vector, as reported in the State of Minas Gerais 8 or of the ineffectiveness of control measures in the study area considered endemic.Furthermore, few owners implemented control measures such as using repellent collars and/or environmental insecticides 4 , or having their dogs vaccinated against CVL or castrated as a means of population control.These measures were also investigated in this population.
In conclusion, Jaciara had a high prevalence of risk factors favorable to CVL.These were particularly associated with the environmental conditions, which are important in maintaining the epidemiological cycle of the disease.

FIGURE 1 -
FIGURE 1 -Neighborhoods analyzed for canine visceral leishmaniasis in the municipality of Jaciara, State of Mato Grosso, based on the distribution of cases and the presence of visceral leishmaniasis and Lutzomyia cruzi.

TABLE 1 -Risk factors for infection by Leishmania chagasi in dogs in urban and rural areas in the municipality of Jaciara, State of Mato Grosso.
a urban area; b rural area; c total.