Prevalence and risk factors associated with Maedi-Visna infection in sheep in the State of Maranhão, Brazil

In order to determine Maedi-Visna virus (MVV) seroprevalence and risk factors associated with infection in sheep, 445 animals of both sexes and different ages were tested using the Agarose Gel Immunodiffusion technique (IDGA). The animals were divided into two groups: group 1 composed of exhibition animals (n=70) and group 2 composed of animals from properties from the north, east and central mesoregions of the state of Maranhão (n=375). The general prevalence of MVV infection was 2.02%; 1.42% in group 1 and 2.13% in group 2. In the north mesoregion the prevalence was 2.20%, while a total of 40% of the animals living in municipalities for MVV. It was observed that 1.15% of males and 2.23% of females were seropositive (p> 0.20). Regarding breeds, the Dorper (1.66%); Santa Inês (1.67%); White Dorper (33.33%) and Texel (4.34%) were the most affected. This was the only variable among all the risk factors studied that had a significant association in multivariate analysis (p <0.05). MVV infection is present regardless of the purpose of breeding, and the animals are exposed to the same risk of infection. This demonstrates the need to implement public policies for the prevention, control, and eradication of this disease.


Introduction
Maedi-Visna (MV) is a chronic and progressive multisystemic disease that mainly affects sheep (Araújo et al., 2004;Lombardi et al., 2009). It is caused by an RNA virus belonging to the genus Lentivirus, family Retroviridae, subfamily Lentivirinae . The infection is characterized by the development of chronic inflammatory lesions in different organs, mainly in the lung, mammary gland and central nervous system (CNS). In the CNS, a histiocytic or lymphocytic inflammatory pattern is predominantly described (Gayo et al., 2018). Animals with a histiocytic pattern are easily recognizable by the ELISA test, whereas sheep with a lymphocytic lesion pattern may go unnoticed as seronegative and are probably a remnant source of infection (Gayo et al., 2017).
The disease is easily spread and handling conditions can favor the spread among animals. However, as the main routes of infection and transmission of the agent, direct contact with secretions rich in cells of the monocyte-phagocytic system, such as blood, milk or colostrum (Callado et al., 2001;Blacklaws et al., 2004;Silva & Lima, 2007). In addition, transmission can occur mainly through respiratory exudates and secretions, most often facilitated between animals kept in confinement periods (Callado et al., 2001;Cortez-Romero et al., 2013).
Among the serological tests available, an Agarose Gel Immunodiffusion (AGID) is recommended by the World Organization for Animal Health (OIE) and is widely used for diagnosis of small Ruminants Lentivirosis the most used around the world (OIE, 2012).
Animal agglomerations represent an important link in the transmission of infectious diseases . In the absence of effective vaccines, a comprehensive understanding of the epidemiology of this diseare is of great importance to limit their spread. Many variables influence the expected costs and benefits of control programs and must be evaluated in order to implement measures that can lead to the control of this disease in sheep herds infections (Minguijón et al., 2015).
The economic damage caused by this disease is considerable, as it can result in severe weight loss in animals, reproductive failures, the need for early replacement of breeding stocks and matrices and death; as well as limiting international trade due to the sanitary barriers it can cause. The paucity of information about Maedi-Visna Virus (MVV), its distribution and dispersion has contributed to a lack of effective control measures. The aim of the present study was to determine the prevalence and identify risk factors associated with Maedi-Visna Virus in sheep from the State of Maranhão.

Methodology
The study was approved by the Ethics Committee in Animal  In total, 445 purebred sheep were included in the survey. Of these, 87 were male and 358 were female, aged from two months to eight years, of the Dorper, Santa Inês, White Dorper and Texel breeds.
The number of animals in Group 1 was determined by the non-probabilistic sampling technique (Thrusfield, 2004). The detection of Maedi-Visna virus antibodies was performed using the agar gel immunodiffusion technique (AGID), with a commercial kit (Biovetech®, Recife, Pernambuco, Brazil) and following the manufacturer's recommendations. Two readings were performed, one after 24h and one after 48h, by three independent observers. The result was considered that taken after the second reading. Samples that tested positive were repeated three times to confirm the result.
Properties were considered positive for the presence of the Maedi-Visna Virus when at least one reagent animal was found. First, the prevalence of seropositive animals was estimated and then a descriptive statistical analysis of the groups by sex, breed and age was performed. The frequencies in the studied strata were calculated based on the serological results by simple percentage.
The variables that were related in the epidemiological survey allowed the study of risk factors. A database was generated, with the animal considered to be an epidemiological unit, to identify which of the studied factors were associated with the Maedi-Visna virus. The logistic regression model was used (Hosmer & Lemeshow, 1989).
Data were analyzed in the STATA 9.0 program. Univariate analysis of each independent variable was performed by selecting those with a p-value <0.20. Subsequently, the combined association of the factors studied with respect to the occurrence of Maedi-Visna was analyzed using the multivariate logistic regression technique, with only the variables that presented a pvalue <0.05 remaining in the model. Prevalence ratios (PR) and 95% confidence intervals were estimated. The level of Research, Society andDevelopment, v. 10, n. 5, e2210514440, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i5.14440 4 significance adopted was 5%. The comparison of the frequencies between groups was performed by the chi-squared test and logistic regression analysis was used to identify risk factors with a significant association with seropositivity.
The general prevalence of the Maedi-Visna Virus (MVV) without clinical signs characteristic of the disease was 2.02%  (Table 2). Variables with a p value less than 0.20 (p <0.20) were included in the multivariate logistic regression model; the "breed" variable was the only risk factor with a statistically significant association (p <0.05). White Dorper animals were 44 times more likely to be infected with MVV than the other breeds tested. Reagent animals were observed in all the evaluated breeds; 1.66% (2/120) for the Dorper breed, 1.67% (5/299) for Santa Inês, 33.33% (1/3) for White Dorper and 4.34% (1/23) for Texel (Table   2).
Of all the positive animals, 2.20% (9/409) were from the north mesoregion of the state of Maranhão. Of all the municipal regions, 40% (4/10) had at least one property with a seropositive animal.
A prevalence of 1.42% (1/70) in group 1 and 2.13% (8/375) in group 2 was observed (Table 3). In terms of the overall prevalence of the disease, Group 1 had 0.22% of seropositive animals and group 2 had 1.80% of seropositive animals. Statistical analysis, however, revealed that the frequencies of MVV did not differ between the groups.

Discussion
The general prevalence of the Maedi-Visna (MV) found in this study is considered low by Reina et al. (2009) (1 to 9%), and is similar to that observed in other states where a semi-extensive breeding system also predominates. This may be related to the recent introduction of the virus in the region and to the fact that this rate varies greatly in purebred animals, according to Batista et al. (2004).
Low prevalences were also found in the state of Pernambuco (1.1%) by Costa et al. (2007), Bahia (0.5% and 0.34%) by Souza et al. (2007) and Martinez et al. (2011) and in Tocantins (0.9%), by Moura Sobrinho et al. (2008). In studies conducted by Guilherme et al. (2017),  et al. (2008) and Lombardi et al. (2009). However, results presented in this work corroborate with data from Sobrinho et al. (2010) findings of the present study, as they also found a significant association between breed and the occurrence of Lentivirosis in small ruminants in the state of Tocantins. Callado et al. (2001) stated that breed susceptibility could not be identified, as the studies are difficult to interpret due to various factors related to handling.
The null prevalence found in the other states surveyed and in the central and east mesoregions can be explained by the reduced number of animals tested. However, the presence of reactive sheep in the north mesoregion may be due to the fact that this region had the highest concentration of sheep breeders in the study area, as well as the greatest number of cities, properties and animals sampled. Teixeira et al. (2016) observed an overall prevalence of MVV of 0.7%, and prevalences of 0.5%, 0.7% and 1% in the central, eastern and northern mesoregions of the state of Maranhão, respectively.
Studies carried out by Moura Sobrinho et al. (2008) observed that there was a numerically greater number of positive cases among animals of the Santa Inês breed (6/511) than among non-defined breed animals (2/324). However, no significant association was found between breed and MVV occurrence. In the results obtained by Teixeira et al. (2016) there was no association between seropositivity and the breed affected. The claim that MVV infection can affect animals of any age was verified in studies by Rowe & East (1997), where it was observed that animals aged less than one year and older than 48 months were seropositive, and that a higher percentage of infection was found in the latter stratum (5/9 -3.28%). According to Snowder et al. (1990) and Cutlip et al. (1992), older animals are exposed to the infectious agent for a longer period while seroprevalence tends to be higher in this age group, as it is a chronic disease with a long, slow course, allowing the confirmation of seroconversion. Teixeira et al. (2016) also found that there was no significant difference in this variable.
There were no animals with clinical signs of MVV, probably due to the long incubation period of the virus, a finding similar to the observations of Araújo et al. (2004), Souza et al. (2007) and Lombardi et al. (2009). This result reinforces the hypothesis that the infections are recent.

Conclusion
Maedi-Visna Virus infection is present in non-defined breed sheep in the state of Maranhão, and all sheep are exposed to the same risk of infection irrespective of the purpose of breeding. The occurrence of Maedi-Visna indicates the need to implement preventive and control measures, especially as these animals are used as herd enhancers, and have the capacity to spread the disease quickly and across a large geographic area.

Ethics Committee in Animal Experimentation
The study was approved by the Ethics Committee in Animal Experimentation of the State University of Maranhão (CEAA/UEMA -012/2013).