- Email: [email protected]
Epidemiological patterns of bovine besnoitiosis in an endemic beef cattle herd reared under extensive conditions
Accepted Manuscript Title: Epidemiological patterns of bovine besnoitiosis in an endemic beef cattle herd reared under extensive conditions Authors: A. Esteban-Gil, C. Calvete, I. Casas´us, A. Sanz, J. Ferrer, M.P. Peris, J.M. Marc´en-Seral, J.A. Castillo PII: DOI: Reference:
S0304-4017(16)30530-1 http://dx.doi.org/doi:10.1016/j.vetpar.2016.12.018 VETPAR 8218
To appear in:
Veterinary Parasitology
Received date: Revised date: Accepted date:
15-11-2016 22-12-2016 23-12-2016
Please cite this article as: Esteban-Gil, A., Calvete, C., Casas´us, I., Sanz, A., Ferrer, J., Peris, M.P., Marc´en-Seral, J.M., Castillo, J.A., Epidemiological patterns of bovine besnoitiosis in an endemic beef cattle herd reared under extensive conditions.Veterinary Parasitology http://dx.doi.org/10.1016/j.vetpar.2016.12.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Epidemiological patterns of bovine besnoitiosis in an endemic beef cattle herd reared under extensive conditions
A. Esteban-Gila,*, C. Calveteb, I. Casasúsb, A. Sanzb, J. Ferrerb, M.P. Perisa, J.M. Marcén-Serala, J.A. Castilloa
a
Parasitology and Parasitic Disease Area, Animal Pathology Department, Faculty of Veterinary Sciences,
Agrifood Institute of Aragon (IA2), University of Zaragoza-CITA, Miguel Servet 177, 50013, Zaragoza, Spain. b
Animal Health and Production Department, Agrifood Research and Technology Centre of Aragon
(CITA), Agrifood Institute of Aragon (IA2), University of Zaragoza-CITA, Avenida Montañana 930, 50059, Zaragoza, Spain.
*
Corresponding author: A. Esteban-Gil
Tlf.: +34 976844280; fax: +34 976761612 Email addresses: [email protected]; [email protected]
1
Highlights
Epidemiological pattern of bovine besnoitiosis is explored in an endemically affected herd.
A continuous parasite transmission, even among subclinically infected animals, is observed.
Transmission is favoured by housing, when cattle cohabitate in close contact for long periods.
Clinical disease is mainly observed in older animals and is characterized by high immune responses.
Infection is reported in calves younger than 6 months, likely due to contact during suckling.
Abstract Bovine besnoitiosis is a parasitic disease caused by the protozoan Besnoitia besnoiti. Described many decades ago, recent epidemiological studies reveal its important spread within Europe in the last years. To date, many epidemiological aspects related to life cycle, routes of transmission, incidence rates and associated risk factors are lacking; hence, the establishment of appropriate disease control programmes poses an important challenge. Thus, the aim of the present study was to determine the epidemiological pattern of the disease in an endemic herd reared under extensive conditions (Spanish Pyrenees) by identifying main factors associated with infection and clinical disease dynamics. The study population consisted of 276 Brown Swiss and Pirenaica adult animals and 145 calves born and weaned at the farm during the study. Three sampling time frames were used: January 2010, September 2010 and February 2011, which allowed us to differentiate two periods designated as mountain and valley periods. The data related to animals (breed, sex and age) and herd management (animal grouping and time in housing) were recorded. The data collection methodology was mainly based on clinical examinations and defining the serological status against bovine besnoitiosis by the immunofluorescent antibody testing of blood samples. The total prevalence among adult animals was 38.34% (CI95%: 34.53-42.07), with 18.54% of seropositive animals showing clinical signs. In regard to the cumulative incidence, 34.57% of new infections were detected during the mountain period, in contrast to the 24.59% observed in the valley period. The incidence density was 0.058 and 0.061 new infections per animal-month for the mountain and
2
valley periods, respectively. According to the seroepidemiological study, the seroconversion probability of B. besnoiti infection was directly associated with the number of seropositive cows with whom an animal had been stabled as well as the housing period duration, supporting horizontal transmission by close contact as one of the most important methods of disease spread. In addition, the risk of developing the clinical course increased with age, and the presence of clinical signs was related to higher antibody responses. Among calves (from 3.1 to 7.1 months old) sampled once at weaning, the total seroprevalence was 15.17% (CI95%: 9.36-21.04), and the chronic stage was observed in three animals, supporting the ability of B. besnoiti to infect and even cause disease in animals less than 6 months old. Finally, the risk of calf seroconversion was positively related to the serological status of the cows, suggesting postnatal transmission between dams and offspring by contact during the suckling period.
Keywords Besnoitia besnoiti, beef cattle, epidemiological pattern, risk factors, Spanish Pyrenees
1. Introduction: Bovine besnoitiosis is a parasitic disease caused by the protozoa Besnoitia besnoiti (Besnoit & Robin, 1912). Until the beginning of the 21st century, the disease appeared to be restricted to areas in SubSaharan Africa (Hofmeyr, 1945), Asia (Neuman & Nobel, 1960) and southern Europe (Spain, Portugal and France) (Juste et al., 1990; Cortes et al., 2005; Alzieu et al., 2007). Currently, new outbreaks of bovine besnoitiosis are reported each year in countries such as Switzerland (Basso et al., 2013), Hungary (Hornok et al., 2014), Croatia (Beck et al., 2013), Greece (Papadopoulos et al., 2014), Belgium (Vanhoudt et al., 2015) and Ireland (Ryan et al., 2016), which seems to demonstrate a current disease reemergence and spread within European countries (EFSA, 2010). In this sense, cattle trading with subclinically infected animals for the purpose of genetic improvement could present an important risk and contribute to disease transmission across herds and even between countries.
3
Many epidemiological and biological aspects related to the parasite life cycle, routes of disease transmission, prevalence and incidence infection rates and associated risk factors remain unclear; hence, the establishment of appropriate programmes to stop the spread of bovine besnoitiosis is difficult. In addition, control alternatives are lacking in prophylactic measures since there is no treatment or vaccine. Therefore, veterinary surveillance at local and national levels is highly recommended. The aim of the present study was to determine the epidemiological pattern of bovine besnoitiosis in an endemic herd reared under extensive conditions by identifying main factors associated with the infection and clinical disease dynamics.
2. Materials and methods: 2.1. Background, study area and study period: The study was performed from January 2010 to February 2011 at La Garcipollera Research Station, a beef cattle farm located in an endemic area of central Pyrenees (Aragon, Spain). The experimental site ranged in altitude from 950 to 2200 m above sea level, with an average total annual rainfall approximately 1000 l/m3 and a mean annual temperature of 10.9 °C, characteristic of an alpine climate. In September 2009, a few days after the summer grazing season on high mountain pastures, the veterinary practitioner in charge of the farm observed five clinical cases consistent with B. besnoiti infection showing poor performance and the thickening, hardening and folding of the skin with hyperkeratosis and alopecia. Due to the suspicion of a new besnoitiosis outbreak, a herd health investigation was conducted. The studied herd production is based on a mountain system farmed under extensive conditions. Traditional management in the area consists of a winter housing period and a grazing season in which cattle graze on high mountain pastures during summer and on meadows and forest pastures located in the valleys and intermediate areas during spring and autumn. Reproductive management uses natural services, and the cows are divided into two calving groups: spring (from February to May) and autumn (from September to December) (Casasús et al., 2002). Usually, cows are housed in groups of 20 animals, classified by breed, together with one bull during a 2-3 month breeding period. Generally, new heifers
4
included in the studied herd came from self-replacement, except for eight cows introduced into the herd in the spring of 2008 for the purpose of genetic improvement. 2.2. Sampling and data collection: The adult herd (nulliparous heifers and young bulls included) consisted of 276 animals of Brown Swiss (60.7%) and Pirenaica (39.3%) breeds, both managed as suckler cattle, and was composed of 91.4% cows and 8.6% bulls. Their ages varied from 1 to 19 years, with over half of the population younger than 5 years. Based on the farm calving management, sampling was conducted three times to characterize infection and clinical disease dynamics in January 2010, September 2010 and February 2011, which made it possible to differentiate the two periods observed in the study, designated as the mountain and valley periods. The mountain period, which ranged from January to September 2010 (7.3 months), comprised an initial stage of housing (with differences in length between cows depending on the calving group), followed by the grazing season on intermediate areas and high mountain pastures. Concerning the valley period, the duration was 5 months, from September 2010 to February 2011, and involved a short-term grazing season on meadows and forest pastures located in the valley followed by a long-term housing period. Calves born and weaned in the farm during the study were included in the investigation and comprised 145 animals, 83 from autumn-calving and 62 from spring-calving. Both breeds were similarly represented, 58.6% of them were females, and 41.4% were males, with an age ranging from 3.1 to 7.1 months old. Calves were monitored once at weaning, in March 2010 or September 2010 according to the time of the calving, autumn or spring calving, respectively. Data related to the breed, sex and age of the animals were recorded. Regarding herd management measures, time in housing (days) and all animal groupings established on the farm were monitored for each studied period. The control of animal grouping allowed for an estimation of the variable number of seropositive cows with whom an animal had been stabled (designated as Np). When an animal was moved from one group to another, the Np variable was determined by calculating the mean number of seropositive cows. Additionally, the relationships between the dams and offspring were also registered. 2.3. Serology and the clinical examinations:
5
The methodology to assess the herd status was based on clinical examinations and defined the serological condition against bovine besnoitiosis during three sampling time points for the adult herd (January 2010, September 2010 and February 2011) and at weaning for the calves (March 2010 or September 2010 for autumn or spring-calving, respectively). Blood samples were obtained by caudal vein puncture, and the sera were separated by natural precipitation in the sampling tube and stored at −20 °C until use. The presence of typical chronic clinical signs as parasitic cysts in the scleral conjunctivae or vestibula vaginae and alterations of the skin (mainly located in neck, scrotal and leg regions) were carefully checked. Specific antibodies against B. besnoiti infection were detected by the immunofluorescent antibody test (IFAT), one of the most appropriate tools for confirming infection at the individual level (García-Lunar et al., 2013; Cortes et al., 2014). Following the instructions previously described by Fernández-García et al. (2009), we established that an IFAT cut-off equal to or more than 1:100 indicated that an animal was seropositive, and this titre was in perfect agreement with the western blot test (WB) (Fernández-García et al., 2009). 2.4. Data analysis: Serological and clinical prevalence rates were estimated for adult and calf herds for each sampling time point. The serological incidence of B. besnoiti infection was assessed by calculating the cumulative incidence (CI) and the incidence density (ID) for both periods established in the investigation, the mountain and valley periods, using the formulas described by Thrusfield (2007). To solve for the lack of awareness of the accurate moment of infection in most of animals, an approach to calculate the enumerator of the ID formula “population time at risk” was estimated using the assumption that infections had happened around the middle of the period (3.7 and 2.5 months from the beginning of the mountain and valley periods, respectively). Multivariable logistic regression models were fit to explore the epidemiological pattern of bovine besnoitiosis. Firstly, to examine the risk of infection in cows, a model was fit with the probability of seroconversion as the dependent variable (as surrogate of infection), and breed and age, as well as management related variables such as the period, time in housing (days), Np and the interaction between time in housing and Np as independent variables or predictors. Next, a logistic model was used to analyse factors related to the probability of developing clinical disease,
6
including breed, age, period, antibody IFAT titres, the seroconversion code and the interaction between the period and the seroconversion code as independent variables. Similarly, factors associated with antibody IFAT levels were explored using a linear model with the same predictor variables except for the IFAT titre variable, which was replaced by the presence of clinical signs. Males were excluded from the epidemiological model analysis because of the differences between cows and bulls in relation to management conditions. For the calves, potential factors associated with seroconversion probability were assessed by running a logistic model, comprising individual (breed, sex and time of calving) and herd variables (time in housing, Np, time in housing and Np interaction, serological status of dams and the interaction between the latter and the time of calving) as predictors. Otherwise, models related to the probability of developing clinical signs and associated with antibody levels in the calf herd could not be fit. When a model included a numeric variable, a transformation into its square root or base-10 logarithm was performed to reduce the heteroscedasticity. A backward stepwise method was applied to refine initial models and obtain a final predictive model. Once the final model was fit, the exponents of the coefficients resulting in the model equation were considered as a risk marker since this value is comparable to the odds ratio. Additionally, the quantitative variable results were expressed as the mean and standard deviation. 2.5 Ethical considerations: The experimental procedures were in compliance with the guidelines of the European Union (Directive 2010/63/EU) on the protection of animals used for scientific purposes. All methods and sampling in the study were carried out by veterinarians under the supervision of the personnel in charge of the farm. Blood samples were collected using a non-invasive procedure commonly used in the diagnosis of infectious and parasitic diseases.
3. Results: 3.1. Adult herd:
7
The proportion of seropositive adult animals observed in the study was 38.34% (CI95%: 34.53-42.07), with the highest seroprevalence rate detected in September 2010 just after the grazing season (χ²=97.20; df=4; p<0.001 and χ²=37.400; df=4; p<0.001) (Table 1). A total of 85 seropositive animals presented with bovine besnoitiosis clinical signs, and the clinical prevalence rate revealed a statistically significant decrease at the end of the study (February 2011) (LR χ²=50.95; df=4; p<0.001) (Table 1). IFAT antibody titres in seropositive animals ranged from 1/100 to 1/6400 and showed a significant decrease in intensity from 1/400 to 1/200 during the second period of the study (from September 2010 to February 2011) (LR χ²=52.17; df=14; p<0.001). In regard to the CI, 34.57% of new infections were detected within the susceptible population during the mountain period compared to 24.59% observed in the valley period (χ²=128.90; df=2; p<0.001) (Table 2). The IDs found in the study were 0.058 and 0.061 new infections per animal-month for the mountain and valley periods, respectively (Table 2). Concerning factors associated with the probability of seroconversion, the final model accounted for a 6.9% of the deviance. Overall, the probability of becoming seropositive in the herd was associated with a mathematical combination of Np and its interaction with time in housing (Table 3), whose global result suggests a direct relationship between Np and time in housing with seroconversion probability, as illustrated in Figure 1. Moreover, according to the predictive model, the risk associated with seroconversion probability was higher during the mountain period than during the valley period (Table 3). Variables related to clinical disease dynamics explained 32.5% of the deviance in the final model. In agreement with this test, animals that seroconverted during the mountain period had a higher risk of developing bovine besnoitiosis clinical signs (Table 4). Additionally, the risk of suffering clinical disease increased with the age of the animals and was directly related to high immune responses (Table 4). Regarding the average age of animals that seroconverted during the mountain period, cows showing clinical besnoitiosis were notably older (7.7±4.1 years old) than those subclinically infected animals (4.3±2.7 years old) (Student´s t-test=-3.36; df=44; p=0.0020) (Figure 2). A final model was fit for factors associated with antibody levels, which explained 11.71% of the variation, suggesting that the immune response intensity was directly related to the presence of clinical signs (Table 5). Likewise, antibody levels appeared to also be associated with, although not significantly,
8
the breed of the cows (Table 5). According to these results, clinically infected animals revealed stronger antibody responses (1/800) than subclinically infected cows (1/100) (LR χ²=17.58; df=6; p=0.007) (Figure 3). 3.2. Calf herd: The total seroprevalence observed in the calf herd was 15.17% (CI95%: 9.36-21.04), with the seroprevalence rate higher in spring-calving animals (χ²=9.43; gl=1; p=0.002) (Table 6). Furthermore, 50% of seropositive calves (11 out of 22) were born to seropositive dams (χ²=9.47; gl=1; p=0.002) (Table 7). After monitoring the clinical status of the calves, parasitic cysts in the scleral conjunctivae compatible with the chronic stage of bovine besnoitiosis could be noted in 3 out of 16 seropositive spring-calving animals. It is important to highlight that these three clinically affected calves were descendants of seropositive infected cows showing clinical disease. In regard to the immune response of seropositive calves, the antibody titre intensity was higher in the spring-calving animals than in the autumn-calving animals, with antibody mean titres of 1/200 and 1/100, respectively (LR χ²=10.35; df=4; p=0.035). The model was fit to explore factors associated with the probability of seroconversion in calves accounted for 14% of the deviance and was directly related to the time of calving and the serological status of the dams (Table 8). Animals born in the spring-calving season and that were descendants of seropositive dams had a higher risk of seroconversion.
4. Discussion: Despite the wide range of seroprevalence values previously reported, the total seroprevalence rate observed in the adult herd (38.34%) was in agreement with research performed in other endemic areas from France, Italy, Portugal and Spain (Fouquet et al., 2009; Rinaldi et al., 2013; Waap et al., 2014; Gutiérrez-Expósito et al., 2014). A similar trend was observed in a recent study conducted in three infected beef cattle herds located in the Urbasa-Andía Mountains (Navarra, Spain), where an average serological incidence rate of 22% was reported (Gutiérrez-Expósito et al., 2015). Concerning clinical
9
disease, signs compatible with bovine besnoitiosis were observed in 13.30% of the herd, which indicates that the majority of infections were subclinical. However, the seroprevalence detected at the end of the study suggests effective B. besnoiti transmission by subclinically infected cattle. In this context, several authors have hypothesized that seropositive animals without visible clinical signs (commonly referred to as subclinical cases) could occasionally present with B. besnoiti DNA in their skin and may contribute to parasite spread (Frey et al., 2013; Gollnick et al., 2015). In January 2010, all clinically infected animals were seropositive, while in September 2010 and February 2011, seven clinically affected animals did not show an antibody response. This finding could be mainly attributed to an IFAT test failure to detect the antibody response, although it could be also due to an immune tolerance to the parasite or a serological recovery of these animals; however, there is a lack of well-controlled studies on this subject. It should be noted that the high seroprevalence and CI values detected during the mountain period could be explained by the cumulative effect associated with the difference in time between periods (7.3 months for the mountain period versus 5 months for the valley period) and not the concurrence of differentiating epidemiological aspects such as the supraforestal-grazing season or the activity of vector insects during the mountain period. Nevertheless, it was crucial to compensate for the variation in time between the periods by calculating a monthly ID rate. Taking into account all of these circumstances, the similar ID obtained for the mountain and valley periods seems to indicate a continuous parasite transmission throughout the study, which could be in disagreement with the seasonal variation of the disease previously suggested by some authors (Fernandez-García et al., 2010; Jacquiet et al., 2010; Alzieu et al., 2011). As a result of the seroepidemiological study performed by regression models, the seroconversion probability of B. besnoiti infection appeared to be directly associated with the number of seropositive cows with whom an animal had been stabled with, as well as the housing period duration, supporting the role of horizontal transmission by close contact as one of the most important ways of disease spread. The final model fit for factors associated with the probability of seroconversion indicated that the more seropositive cows in cohabitation and the longer the housing period duration, the greater the probability an animal would become infected. These findings suggest that disease transmission in herds is mainly driven by housing, when management measures are more intensive and when cattle cohabitate in close
10
contact for long periods. Similarly, a cohabitation experiment conducted in Germany, where six healthy animals were housed on a pasture together with three besnoitiosis clinically affected cows, showed that after 12 weeks, 3 out of the 6 healthy animals become infected, highlighting the important role of close contact between animals in disease transmission (Gollnick et al., 2015). Furthermore, Bigalke already demonstrated the mechanical transmission of the disease through direct contact by inoculating parasite cystic stages (bradyzoites) into the nostrils of a cow, which led to the hypothesis of the ability of parasite cysts to pass through mucosal membranes (Bigalke, 1968). Another important aspect to consider, as a consequence of this analysis, was the apparent lack of relationship between the breed and the age of the animals and their seroconversion probability. These results confirm previous observations describing the susceptibility to become infected among a wide variety of breeds (Álvarez-García et al., 2014a; ÁlvarezGarcía et al., 2014b). Our results seem to suggest that age is not a risk factor, contrary to the increase in prevalence with age reported by other authors (Álvarez-García et al., 2014a; Álvarez-García et al., 2014b; Gutiérrez-Expósito et al., 2014; Waap et al., 2014). However, it is important to explain that we studied the probability of seroconversion among healthy animals in a particular period of time depending on their age, while other investigations described the prevalence by age, and the increase observed may be associated with a cumulative effect over time. In fact, most authors agree that the increase in seroprevalence found in older animals could be mainly due to successive parasite exposures accumulated with age. In contrast, this investigation shows that the risk of developing bovine besnoitiosis clinical course increased with animal age. During the mountain period, seropositive clinically infected animals were on average three year older than seropositive subclinically infected animals. As previously described, certain authors suggest an increase in seroprevalence with age together with the hypothesis that older animals are more often clinically affected (Fernández- García et al., 2010; Álvarez-García et al., 2014b). Supporting this theory, Liénard et al. observed, in a longitudinal study performed in French dairy cattle, that only animals four years old and older showed clinical signs of besnoitiosis (cysts in the scleral conjunctivae) (Liénard et al., 2011). Potential causes that could lead an infected animal to develop chronic clinical disease are largely unknown. In this study, we demonstrated that age seems to play an important role in that regard. This could be explained by the existence of biological differences due to animal age, such as aspects related to the immune system, or by certain characteristics associated with animal behaviour depending on age, which could predispose animals towards one or another way of transmission, and consequently, animals would receive different doses or exposures to parasites.
11
Furthermore, the development of clinical signs in this study implied higher antibody responses, with mean dilutions of 1/800 and 1/100 for antibody titres from clinically and subclinically infected cows, respectively. This association between the intensity of the immune response and clinical signs have been previously described by other authors, who suggest that this correspondence may be due to a continuous stimulation of the immune system proportional to the presence of tissue cysts (Liénard et al., 2011; Esteban-Gil et al., 2014.). With the aim of decreasing disease transmission in herds, these results highlight the possibility of using antibody titres as a marker of disease severity and more specifically, as a tool of choice for the selective screening of highly infectious animals (Alzieu et al., 2011; Gollnick et al., 2015). The prevalence of antibodies and the clinical signs recorded in calves demonstrates the ability of B. besnoiti to infect and even cause disease in animals younger than 6 months old, supporting the susceptibility of cattle to the disease regardless of age, as Alzieu et al. demonstrated before detecting tissue cysts in young calves from 4-5 months old (Alzieu et al., 2011). Supposing that the antibodies observed in our study were due to a transfer of maternal immunity (Shkap et al., 1994), we would expect that all calves born to B. besnoiti seropositive dams presented with antibodies. However, only half of the infected calves descended from seropositive cows, suggesting that the antibodies detected were consistent with recent infections. Consequently, these results indicate that maternal antibodies were not transferred to calves for a protective immune response against subsequent infections. Regarding putative routes of transmission to calves, Hornok et al. noted an unlikely transmission pathway via the colostrum, given the lack of B. besnoiti DNA excretion in the colostrums of seropositive cows (Hornok et al., 2015). Similarly, Frey et al. described the improbability of the vertical transmission of the disease, as previously reported by Nobel et al. (Nobel et al., 1981), due to the failure to detect the parasite in the upper tract of the reproductive organs of infected cows (Frey et al., 2013). Based on similar studies of bovine neosporosis (Schares et al., 1998; Davison et al., 1999), in which congenital transmission was observed in 94% of seropositive cows, considering transplacental or congenital disease transmission in our study seems unlikely since approximately 70% of seropositive infected dams had healthy offspring, and contrary, 10% of healthy females had infected offspring, which suggests postnatal transmission between dams and offspring through close contact during the suckling period. Evidence for this pathway could be observed in the regression models of the present study, which revealed that calf descendants of seropositive dams had double the probability of seroconversion than those born to seronegative cows. According to a recent
12
study, for seropositive pregnant cattle, the immediate separation of the calves after the ingestion of colostrum and subsequent artificial feeding could be a successful strategy for preventing parasite transmission (Hornok et al., 2015). Additionally, the seroprevalence was higher in the spring than in the autumn-calving animals, which could be due to the slight difference of age between the calving time points (spring-calving animals were one month and a half older than autumn-calving animals, being 6.5 and 5.2 months old, respectively). Therefore, among calves born in the spring, the supraforestal-grazing season, the increased activity of vector insects during the summer months or just the difference in age could increase the risk of infection, which would explain the increased seroprevalence detected in calves born in the autumn. Additionally, the antibody titres of animals born in the spring were slightly higher, consistent with the presence of calves with clinical disease. As in the adult herd, clinically affected calves had higher antibody titres (≥ 1/400).
Regarding the explanatory capacity of the models and, specifically, the models of seroconversion probabilities, their explanatory power was low, suggesting the existence of other factors that modulated the epidemiology of the disease and could not be controlled in this study. However, it is important to consider that the explanatory capacity of the models may be reduced by not controlling for important factors and also a lack of means for recording such factors. Thus, the reduced explanatory power observed in this study could be mainly because complex epidemiological mechanisms of the disease, such as direct parasite transmission rates between cows, were measured with rough estimates of the time in housing and the number of seropositive cows with whom an animal had been stabled, with the added difficulty of not knowing how many of these seropositive cows were infectious. Finally, a hypothesis about the origin of the outbreak on the farm was suggested. Based on the evidence observed in the research, the beginning of the outbreak could be attributed to cattle trading, specifically to the introduction of eight new cows into the herd in spring 2008 with a genetic improvement purpose. In fact, all of them showed clinical signs in the course of this research. Consequently, and considering all the results drawn from the study, we recommend a set of measures to avoid infection diffusion in herds. Disease control strategies in farms should be based on two main approaches. Farms should avoid the introduction of parasites into herds by rigorous controls as new animals arrive. Additionally, affected farms should avoid parasite transmission by focusing on visual observation of parasitic cysts in the scleral conjunctivae and serological analyses coupled with the progressive culling of infected animals and
13
following management measures including the physical separation of infected and healthy cattle (even infected dams and their offspring) until the total eradication of seropositive animals is achieved. To conclude, the present work identified the main factors associated with infection and clinical disease dynamics in a bovine besnoitiosis outbreak, which may act as predictive markers of infection and disease transmission under field conditions. The seroepidemiological investigation suggests continuous parasite transmission throughout the study, including among subclinically infected animals, likely due to long periods of close contact occurring when animals are housed in the farm and when management measures become more intensive. Clinical disease was mainly observed in older animals and was characterized by immune responses of higher intensity. Moreover, the results demonstrate that calves can become infected and can even develop clinical disease when they are less than 6 months old. Besnoitia besnoiti transmission between dams and offspring in this herd may be attributed to postnatal transmission via close contact during the suckling period.
Conflict of interest statement: All authors declare that they do not have any financial, personal or other relationships with other people or organizations that could inappropriately influence their work.
Acknowledgments: Part of this research was financially supported by the General Administration of Food and Agrifood Development (Husbandry Department of Aragon Government) (project number OTRI 2011/1023). The authors wish to thank the staff of La Garcipollera Research Farm and the Agrifood Research and Technology Centre of Aragon (CITA) for supporting the acquisition and collection of the samples and data. Likewise, we acknowledge the SALUVET group (Animal Health Department, Faculty of Veterinary Sciences, Complutense University of Madrid) for their excellent technical assistance. Finally, Adriana Esteban-Gil would to thank the professors and colleagues from the Parasitology and Parasitic Disease Department of the Veterinary Science Faculty of Zaragoza for all these years of collaboration.
14
References: Álvarez-García, G., Fernández-García, A., Gutiérrez-Expósito, D., Ruíz-Santa Quiteria, J.A., Aguado-Martínez, A. & Ortega-Mora, L.M., 2014a. Seroprevalence of Besnoitia besnoiti infection and associated risk factors in cattle from an endemic region in Europe. Vet. J. 200, 328331. Álvarez-García, G., García-Lunar, P., Gutiérrez-Expósito, D., Shkap, V. & Ortega-Mora, L.M., 2014b. Dynamics of Besnoitia besnoiti infection in cattle. Parasitology 141, 1419-1435. Alzieu, J.P., Dorchies, P., Schelcher, F. & Gottstein, B., 2007. L´extension de la besnoitiose bovine en France. Le Point Vétérinaire 38, 37-43. Alzieu, J.P., Jacquiet, P., Liénard, E., Grisez, C., Prevot, F., Malavieille, R., Desclaux, X., Hussbaum, S., Bergeaud, J.P., Franc, M., Shelcher, F., Corboz, N., Bastien, F., GuerrierChätelet, M.C., Gaver, L., Schneider, V. & Boulon, C., 2011. Réémergence de la besnoitiose bovine: démarche diagnostique et possibilités de contrôle. Bulletin des GTV 58, 71-86. Basso, W., Lesser, M., Grimm, F., Hilbe, M., Syder, T., Trösch, L., Hansueli, O., Braun, U. & Deplazes, P., 2013. Bovine besnoitiosis in Switzerland: Imported cases and local transmission. Vet. Parasitol. 198, 265-273. Beck, R., Stokovic, I., Pleadin, J. & Beck, A., 2013. Bovine besnoitiosis in Croatia. In: Proceedings of the 2nd International Meeting on Apicomplexan Parasites in Farm Animals (ApiCOWplexa: Kusadasi, Turkey), p. 64. Besnoit, C. & Robin, V., 1912. Sarcosporidiose cutanée chez une vache. Recueil Vétérinaire 37, 649663. Bigalke, R.D., 1968. New concepts on the epidemiological features of bovine besnoitiosis as determined by laboratory and field investigations. Onderstepoort J. Vet. Res. 35, 3-137. Casasús, I., Sanz, A., Villalba, D., Ferrer, R. & Revilla, R., 2002. Factors affecting animal performance during the grazing season in a mountain cattle production system. J. Anim. Sci. 80, 1638–51. Cortes, H., Leitao, A., Vidal, R., Vila-Vicosa, M.J., Ferreira, M.L., Caeiro, V. & Hjerpe, C.A., 2005.
15
Besnoitiosis in bulls in Portugal. Vet. Rec. 157, 262-264. Cortes, H., Leitão, A., Gottstein, B. & Hemphill, A., 2014. A review on bovine besnoitiosis: a disease with economic impact in herd health management, caused by Besnoitia besnoiti (Franco and Borges, 1916). Parasitology 141, 1406–17. Davison, H.C., Otter, A. & Trees, A.J., 1999. Estimation of vertical and horizontal transmission parameters of Neospora caninum infections in dairy cattle. Int. J. Parasitol. 29, 1683–9. EFSA European Food Safety Authority, 2010. Bovine besnoitiosis: an emerging disease in Europe. Scientific statement on Bovine Besnoitiosis, Question No EFSA-Q-2009-00879, EFSA Journal 8, 1499. http://www.efsa.europa.eu/en/efsajournal/pub/1499.htm. Adopted 28 January 2010. Esteban-Gil, A., Grisez, C., Prevot, F., Florentin, S., Decaudin, A., Picard-Hagen, N., Berthelot, X., Ronsin, P., Alzieu, J.P., Marois, M., Corboz, N., Peglion, M., Vilardell, C., Liénard, E., Bouhsira, E., Castillo, J.A., Franc, M. & Jacquiet, P., 2014. No detection of Besnoitia besnoiti DNA in the semen of chronically infected bulls. Parasitol. Res. 113, 355-362. Fernandez-Garcia, A., Alvarez-Garcia, G., Risco-Castillo, V., Aguado-Martinez, A., MaruganHernandez, V. & Ortega-Mora, L.M., 2009. Pattern of recognition of Besnoitia besnoiti tachyzoite and bradyzoite antigens by naturally infected cattle. Vet. Parasitol. 164, 104-110. Fernandez-García, A., Alvarez-Garcia, G., Risco-Castillo, V., Aguado-Martinez, A., Marcen, J.M., Rojo-Montejo, S., Castillo, J.A. & Ortega-Mora, L.M., 2010. Development and use of an indirect ELISA in an outbreak of bovine besnoitiosis in Spain. Vet. Rec. 166, 818-822. Fouquet, C.M., 2009. La besnoitiose bovine: suivi épidémiologique de l´epizootie de la región PACA. Thèse Vétérinaire Médecine, Lyon. Frey, C.F., Gutiérrez-Expósito, D., Ortega-Mora, L.M., Benavides, J., Marcén, J.M., Castillo, J.A., Casasús, I., Sanz, A., García-Lunar, P., Esteban-Gil, A. & Alvarez-García, G., 2013. Chronic bovine besnoitiosis: intra-organ parasite distribution, parasite loads and parasiteassociated lesions in subclinical cases. Vet. Parasitol. 197, 95-103. García-Lunar, P., Ortega-Mora, L.M., Schares, G., Gollnick, N.S., Jacquiet, P., Grisez, C., Prevot, F., Frey, C.F., Gottstein, B. & Álvarez-García, G., 2013. An Inter-Laboratory Comparative
16
Study of Serological Tools Employed in the Diagnosis of Besnoitia besnoiti Infection in Bovines. Transbound. Emerg. Dis. 60, 59–68. Gollnick, N.S., Scharr, J.C., Schares, G. & Langenmayer, M.C., 2015. Natural Besnoitia besnoiti infections in cattle: chronology of disease progression. BMC Vet. Res. 11, 35. Gutiérrez-Expósito, D., Esteban-Gil, A., Ortega-Mora, L.M., García-Lunar, P., Castillo, J.A., Marcén, J.M. & Álvarez-García, G., 2014. Prevalence of Besnoitia besnoiti infection in beef cattle from the Spanish Pyrenees. Vet. J. 200, 468-470. Gutiérrez-Expósito, D., Ortega-Mora, L.M., García-Lunar, P., Rojo-Montejo, S., Zabala, J., Serrano, M. & Álvarez-García, G., 2015. Clinical and serological dynamics of Besnoitia besnoiti infection in three endemically infected beef cattle herds. Transbound. Emerg. Dis. DOI: 10.1111/tbed.12402. Hofmeyr, C.F.B., 1945. Globidiosis in cattle. S. Afr. Vet. Med. Assoc. 6, 102-109. Hornok, S., Fedák, A., Baska, F., Hofmann-Lehmann, R. & Basso, W., 2014. Bovine besnoitiosis emerging in Central-Esatern Europe, Hungary. Parasit. Vectors. 7, 20. Hornok, S., Fedák, A., Baska, F., Basso, W., Dencso, L., Tóth, G., Szeredi, L., Abonyi, T. & Dénes, B., 2015. Vector-borne transmission of Besnoitia besnoiti by blood-sucking and secretophagous flies: epidemiological and clinicopathological implications. Parasit. Vectors. 8, 450. Jacquiet, P., Liénard, E. & Franc, M., 2010. Bovine besnoitiosis: Epidemiological and clinical aspects. Vet. Parasitol. 174, 30-60. Juste, R.A., Cuervo, L.A., Marco, J.C. & Oregui, L.M., 1990. La besnoitiosis bovina: ¿desconocida en España?. Medicina Veterinaria 7, 613-618. Liénard, E., Salem, A., Grisez, C., Prévot, F., Bergeaud, J.P., Franc, M., Gottstein, B., Alzieu, J.P., Lagalisse, Y. & Jacquiet, P., 2011. A longitudinal study of Besnoitia besnoiti infections and seasonal abundance of Stomoxys calcitrans in a dairy cattle farm of southwest France. Vet. Parasitol. 177, 20-27.
17
Neuman, M. & Nobel, T.A. 1960. Globidiosis of cattle and sheep in Israel. Refuah Veterinarith 17, 101103. Nobel, T.A., Klopfer, U., Perl, S., Nyska, A., Neumann, M. & Brenner, G., 1981. Histopathology of genital besnoitiosis of cows in Israel. Vet. Parasitol. 8, 271-276. Papadopoulos, E., Arsenos, G., Ptochos, S., Katsoulos, P., Oikonomou, G., Karatzia, M.A. & Karatzias, H., 2014. First report of Besnoitia besnoiti seropositive cattle in Greece. J. Hellenic Vet. Med. Soc. 65, 115-120. Rinaldi, L., Maurelli, M.P., Musella, V., Bosco, A., Cortes, H. & Cringoli, G., 2013. First crosssectional serological survey on Besnoitia besnoiti in cattle in Italy. Parasitol. Res. 112, 18051807. Ryan, E.G., Lee, A., Carty, C., O´ Shaughnessy, J., Kelly, P., Cassidy, J.P., Sheehan, M., Johnson, A. & de Waal, T., 2016. Paper Bovine besnoitiosis (Besnoitia besnoiti) in an Irish dairy herd. Vet. Rec.178. Schares, G., Peters, M., Wurm, R., Bärwald, A. & Conraths, F.J., 1998. The efficiency of vertical transmission of Neospora caninum in dairy cattle analyzed by serological techniques. Vet. Parasitol. 80, 87–98. Shkap, V., Pipano, E., Marcus, S. & Krigel, Y., 1994. Bovine besnoitiosis: transfer of colostral antibodies with observations possibly relating to natural transmission of the infection. Onderstepoort J. Vet. Res. 61, 273-275. Thrusfield, M., 2007. Veterinary Epidemiology, 3rd Edition. Blackwell publishing. Vanhoudt, A., Pardon, B., De Schutter, P., Bosseler, L., Sarre, C., Vercruysse, J. & Deprez, P., 2015. First confirmed case of bovine besnoitiosis in an imported bull in Belgium. Vlaams Diergeneeskd Tijdschr 84, 205–11. Waap, H., Nunes, T., Cortes, H., Leitao, A. & Vaz, Y., 2014. Prevalence and geographic distribution of Besnoitia besnoiti infection in cattle herds in Portugal. Parasitol. Res. 113, 3703-3711.
18
Figure legend: Figure 1. 3D surface plot illustrating the seroconversion probability (Y) dependent on the number of seropositive cows with whom an animal had been stabled with (X) and the time in housing (Z) in the mountain (left-hand) and valley (right hand) periods.
19
Figure 2. Age distribution (years old) of clinically (black bar) and subclinically (grey bar) infected cows that seroconverted during the mountain period.
20
Figure 3. IFAT antibody titres of clinically (black bar) and subclinically (grey bar) infected cows.
21
Tables: Table 1. Within-herd seroprevalence and clinical prevalence rates in the adult herd at three times of sampling.
Seroprevalence
Clinical prevalence
n
Seropositive animals
Serological rate (%) CI95%
Animals showing clinical signs
Clinical (%)
219
65
29.68 (23.65-35.75)
38
17.35
223
115
51.57(44.94-58.06)a
44
19.73
February 2011
197
65
32.99 (26.43-39.57)
3
1.52b
Total
639
245
38.34 (34.53-42.07)
85
13.30
Time sampling
of
January 2010 September 2010
rate
Statistical significance according to Pearson´s chi-squared testa (χ²=97.20; df=4; p<0.001 and χ²=37.400; df=4; p<0.001) and Likelihood ratiob (LR χ²=50.95; df=4; p<0.001)
22
Table 2. Cumulative incidence (CI) (% of new infected cases among disease-free population) and incidence density (ID) (new infected cases per animal-month) for the mountain and valley periods.
Study period
CI
ID
Mountain
34.57
0.058
Valley
24.59
0.061
p
<0.001*
*Statistical significance according to Pearson´s chi-squared test (χ²=128.90; df=2; p<0.001)
23
Table 3. Final model fit for factors associated with the seroconversion probability (explained 6.9% of final model deviance).
Variable
Coefficient
SE
Wald
p-value
Constant
-1.411
0.588
5.75
0.016
Period (mountain)
0.530
0.260
4.16
0.041
√Number of seropositive cows (Np)
-1.125
0.492
5.23
0.022
√Time in housing (days) *√Np
0.229
0.104
8.22
0.004
24
Table 4. Final model fit for factors related to clinical disease dynamics (explained 32.5% of final model deviance).
Variable
Coefficient
SE
Wald
p-value
Constant
-8.799
1.835
22.99
<0.001
Period (mountain)
1.159
0.533
4.72
0.030
Age
1.259
0.351
12.89
<0.001
IFAT titre log
1.478
0.504
8.61
0.003
Period*seroconversion code
0.672
0.243
7.64
0.006
25
Table 5. Final model fit for factors associated with antibody levels detected by IFAT (explained 11.71% of final model variation).
Variable
Coefficient
SE
t
p-value
Constant
2.684
0.044
60.69
<0.001
Clinical signs (positive)
0.171
0.043
3.97
<0.001
Breed (Brown Swiss)
-0.068
0.038
-1.77
0.078
26
Table 6. Within-herd seroprevalence and clinical prevalence rates in the calf herd according to the time of calving.
Seroprevalence Time calving
of
Clinical prevalence
n
Seropositive animals
Serological rate (%) CI95%
Animals showing clinical signs
Clinical rate (%)
Autumn 2009
83
6
7.23 (1.64-12.76)
0
0
Spring 2010
62
16
25.81 (14.91-36.69)*
3
4.84
Total
145
22
15.17 (9.36-21.04)
3
4.84
*Statistical significance according to Pearson´s chi-squared test (χ²=9.43; df=1; p=0.002)
27
Table 7. Frequency distribution of serological calf results according to the serological status of their dams.
Calf results
Serological status of dams
Seropositive
Seronegative
Total
Seropositive
11*
24
35
Seronegative
11
99
110
Total
22
123
145
*Statistical significance according to Pearson´s chi-squared test (χ²=9.47; df=1; p=0.002)
28
Table 8. Final model fit for factors related to the seroconversion probability in the calf herd (explained 14% of final model deviance).
Variable
Coefficient
SE
Wald
p-value
Constant
-1.572
0.266
34.77
<0.001
0.708
0.253
7.83
0.005
0.747
0.265
7.97
0.005
Serological (positive)
status
of
dams
Time of calving (spring 2010)
29
S0304-4017(16)30530-1 http://dx.doi.org/doi:10.1016/j.vetpar.2016.12.018 VETPAR 8218
To appear in:
Veterinary Parasitology
Received date: Revised date: Accepted date:
15-11-2016 22-12-2016 23-12-2016
Please cite this article as: Esteban-Gil, A., Calvete, C., Casas´us, I., Sanz, A., Ferrer, J., Peris, M.P., Marc´en-Seral, J.M., Castillo, J.A., Epidemiological patterns of bovine besnoitiosis in an endemic beef cattle herd reared under extensive conditions.Veterinary Parasitology http://dx.doi.org/10.1016/j.vetpar.2016.12.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Epidemiological patterns of bovine besnoitiosis in an endemic beef cattle herd reared under extensive conditions
A. Esteban-Gila,*, C. Calveteb, I. Casasúsb, A. Sanzb, J. Ferrerb, M.P. Perisa, J.M. Marcén-Serala, J.A. Castilloa
a
Parasitology and Parasitic Disease Area, Animal Pathology Department, Faculty of Veterinary Sciences,
Agrifood Institute of Aragon (IA2), University of Zaragoza-CITA, Miguel Servet 177, 50013, Zaragoza, Spain. b
Animal Health and Production Department, Agrifood Research and Technology Centre of Aragon
(CITA), Agrifood Institute of Aragon (IA2), University of Zaragoza-CITA, Avenida Montañana 930, 50059, Zaragoza, Spain.
*
Corresponding author: A. Esteban-Gil
Tlf.: +34 976844280; fax: +34 976761612 Email addresses: [email protected]; [email protected]
1
Highlights
Epidemiological pattern of bovine besnoitiosis is explored in an endemically affected herd.
A continuous parasite transmission, even among subclinically infected animals, is observed.
Transmission is favoured by housing, when cattle cohabitate in close contact for long periods.
Clinical disease is mainly observed in older animals and is characterized by high immune responses.
Infection is reported in calves younger than 6 months, likely due to contact during suckling.
Abstract Bovine besnoitiosis is a parasitic disease caused by the protozoan Besnoitia besnoiti. Described many decades ago, recent epidemiological studies reveal its important spread within Europe in the last years. To date, many epidemiological aspects related to life cycle, routes of transmission, incidence rates and associated risk factors are lacking; hence, the establishment of appropriate disease control programmes poses an important challenge. Thus, the aim of the present study was to determine the epidemiological pattern of the disease in an endemic herd reared under extensive conditions (Spanish Pyrenees) by identifying main factors associated with infection and clinical disease dynamics. The study population consisted of 276 Brown Swiss and Pirenaica adult animals and 145 calves born and weaned at the farm during the study. Three sampling time frames were used: January 2010, September 2010 and February 2011, which allowed us to differentiate two periods designated as mountain and valley periods. The data related to animals (breed, sex and age) and herd management (animal grouping and time in housing) were recorded. The data collection methodology was mainly based on clinical examinations and defining the serological status against bovine besnoitiosis by the immunofluorescent antibody testing of blood samples. The total prevalence among adult animals was 38.34% (CI95%: 34.53-42.07), with 18.54% of seropositive animals showing clinical signs. In regard to the cumulative incidence, 34.57% of new infections were detected during the mountain period, in contrast to the 24.59% observed in the valley period. The incidence density was 0.058 and 0.061 new infections per animal-month for the mountain and
2
valley periods, respectively. According to the seroepidemiological study, the seroconversion probability of B. besnoiti infection was directly associated with the number of seropositive cows with whom an animal had been stabled as well as the housing period duration, supporting horizontal transmission by close contact as one of the most important methods of disease spread. In addition, the risk of developing the clinical course increased with age, and the presence of clinical signs was related to higher antibody responses. Among calves (from 3.1 to 7.1 months old) sampled once at weaning, the total seroprevalence was 15.17% (CI95%: 9.36-21.04), and the chronic stage was observed in three animals, supporting the ability of B. besnoiti to infect and even cause disease in animals less than 6 months old. Finally, the risk of calf seroconversion was positively related to the serological status of the cows, suggesting postnatal transmission between dams and offspring by contact during the suckling period.
Keywords Besnoitia besnoiti, beef cattle, epidemiological pattern, risk factors, Spanish Pyrenees
1. Introduction: Bovine besnoitiosis is a parasitic disease caused by the protozoa Besnoitia besnoiti (Besnoit & Robin, 1912). Until the beginning of the 21st century, the disease appeared to be restricted to areas in SubSaharan Africa (Hofmeyr, 1945), Asia (Neuman & Nobel, 1960) and southern Europe (Spain, Portugal and France) (Juste et al., 1990; Cortes et al., 2005; Alzieu et al., 2007). Currently, new outbreaks of bovine besnoitiosis are reported each year in countries such as Switzerland (Basso et al., 2013), Hungary (Hornok et al., 2014), Croatia (Beck et al., 2013), Greece (Papadopoulos et al., 2014), Belgium (Vanhoudt et al., 2015) and Ireland (Ryan et al., 2016), which seems to demonstrate a current disease reemergence and spread within European countries (EFSA, 2010). In this sense, cattle trading with subclinically infected animals for the purpose of genetic improvement could present an important risk and contribute to disease transmission across herds and even between countries.
3
Many epidemiological and biological aspects related to the parasite life cycle, routes of disease transmission, prevalence and incidence infection rates and associated risk factors remain unclear; hence, the establishment of appropriate programmes to stop the spread of bovine besnoitiosis is difficult. In addition, control alternatives are lacking in prophylactic measures since there is no treatment or vaccine. Therefore, veterinary surveillance at local and national levels is highly recommended. The aim of the present study was to determine the epidemiological pattern of bovine besnoitiosis in an endemic herd reared under extensive conditions by identifying main factors associated with the infection and clinical disease dynamics.
2. Materials and methods: 2.1. Background, study area and study period: The study was performed from January 2010 to February 2011 at La Garcipollera Research Station, a beef cattle farm located in an endemic area of central Pyrenees (Aragon, Spain). The experimental site ranged in altitude from 950 to 2200 m above sea level, with an average total annual rainfall approximately 1000 l/m3 and a mean annual temperature of 10.9 °C, characteristic of an alpine climate. In September 2009, a few days after the summer grazing season on high mountain pastures, the veterinary practitioner in charge of the farm observed five clinical cases consistent with B. besnoiti infection showing poor performance and the thickening, hardening and folding of the skin with hyperkeratosis and alopecia. Due to the suspicion of a new besnoitiosis outbreak, a herd health investigation was conducted. The studied herd production is based on a mountain system farmed under extensive conditions. Traditional management in the area consists of a winter housing period and a grazing season in which cattle graze on high mountain pastures during summer and on meadows and forest pastures located in the valleys and intermediate areas during spring and autumn. Reproductive management uses natural services, and the cows are divided into two calving groups: spring (from February to May) and autumn (from September to December) (Casasús et al., 2002). Usually, cows are housed in groups of 20 animals, classified by breed, together with one bull during a 2-3 month breeding period. Generally, new heifers
4
included in the studied herd came from self-replacement, except for eight cows introduced into the herd in the spring of 2008 for the purpose of genetic improvement. 2.2. Sampling and data collection: The adult herd (nulliparous heifers and young bulls included) consisted of 276 animals of Brown Swiss (60.7%) and Pirenaica (39.3%) breeds, both managed as suckler cattle, and was composed of 91.4% cows and 8.6% bulls. Their ages varied from 1 to 19 years, with over half of the population younger than 5 years. Based on the farm calving management, sampling was conducted three times to characterize infection and clinical disease dynamics in January 2010, September 2010 and February 2011, which made it possible to differentiate the two periods observed in the study, designated as the mountain and valley periods. The mountain period, which ranged from January to September 2010 (7.3 months), comprised an initial stage of housing (with differences in length between cows depending on the calving group), followed by the grazing season on intermediate areas and high mountain pastures. Concerning the valley period, the duration was 5 months, from September 2010 to February 2011, and involved a short-term grazing season on meadows and forest pastures located in the valley followed by a long-term housing period. Calves born and weaned in the farm during the study were included in the investigation and comprised 145 animals, 83 from autumn-calving and 62 from spring-calving. Both breeds were similarly represented, 58.6% of them were females, and 41.4% were males, with an age ranging from 3.1 to 7.1 months old. Calves were monitored once at weaning, in March 2010 or September 2010 according to the time of the calving, autumn or spring calving, respectively. Data related to the breed, sex and age of the animals were recorded. Regarding herd management measures, time in housing (days) and all animal groupings established on the farm were monitored for each studied period. The control of animal grouping allowed for an estimation of the variable number of seropositive cows with whom an animal had been stabled (designated as Np). When an animal was moved from one group to another, the Np variable was determined by calculating the mean number of seropositive cows. Additionally, the relationships between the dams and offspring were also registered. 2.3. Serology and the clinical examinations:
5
The methodology to assess the herd status was based on clinical examinations and defined the serological condition against bovine besnoitiosis during three sampling time points for the adult herd (January 2010, September 2010 and February 2011) and at weaning for the calves (March 2010 or September 2010 for autumn or spring-calving, respectively). Blood samples were obtained by caudal vein puncture, and the sera were separated by natural precipitation in the sampling tube and stored at −20 °C until use. The presence of typical chronic clinical signs as parasitic cysts in the scleral conjunctivae or vestibula vaginae and alterations of the skin (mainly located in neck, scrotal and leg regions) were carefully checked. Specific antibodies against B. besnoiti infection were detected by the immunofluorescent antibody test (IFAT), one of the most appropriate tools for confirming infection at the individual level (García-Lunar et al., 2013; Cortes et al., 2014). Following the instructions previously described by Fernández-García et al. (2009), we established that an IFAT cut-off equal to or more than 1:100 indicated that an animal was seropositive, and this titre was in perfect agreement with the western blot test (WB) (Fernández-García et al., 2009). 2.4. Data analysis: Serological and clinical prevalence rates were estimated for adult and calf herds for each sampling time point. The serological incidence of B. besnoiti infection was assessed by calculating the cumulative incidence (CI) and the incidence density (ID) for both periods established in the investigation, the mountain and valley periods, using the formulas described by Thrusfield (2007). To solve for the lack of awareness of the accurate moment of infection in most of animals, an approach to calculate the enumerator of the ID formula “population time at risk” was estimated using the assumption that infections had happened around the middle of the period (3.7 and 2.5 months from the beginning of the mountain and valley periods, respectively). Multivariable logistic regression models were fit to explore the epidemiological pattern of bovine besnoitiosis. Firstly, to examine the risk of infection in cows, a model was fit with the probability of seroconversion as the dependent variable (as surrogate of infection), and breed and age, as well as management related variables such as the period, time in housing (days), Np and the interaction between time in housing and Np as independent variables or predictors. Next, a logistic model was used to analyse factors related to the probability of developing clinical disease,
6
including breed, age, period, antibody IFAT titres, the seroconversion code and the interaction between the period and the seroconversion code as independent variables. Similarly, factors associated with antibody IFAT levels were explored using a linear model with the same predictor variables except for the IFAT titre variable, which was replaced by the presence of clinical signs. Males were excluded from the epidemiological model analysis because of the differences between cows and bulls in relation to management conditions. For the calves, potential factors associated with seroconversion probability were assessed by running a logistic model, comprising individual (breed, sex and time of calving) and herd variables (time in housing, Np, time in housing and Np interaction, serological status of dams and the interaction between the latter and the time of calving) as predictors. Otherwise, models related to the probability of developing clinical signs and associated with antibody levels in the calf herd could not be fit. When a model included a numeric variable, a transformation into its square root or base-10 logarithm was performed to reduce the heteroscedasticity. A backward stepwise method was applied to refine initial models and obtain a final predictive model. Once the final model was fit, the exponents of the coefficients resulting in the model equation were considered as a risk marker since this value is comparable to the odds ratio. Additionally, the quantitative variable results were expressed as the mean and standard deviation. 2.5 Ethical considerations: The experimental procedures were in compliance with the guidelines of the European Union (Directive 2010/63/EU) on the protection of animals used for scientific purposes. All methods and sampling in the study were carried out by veterinarians under the supervision of the personnel in charge of the farm. Blood samples were collected using a non-invasive procedure commonly used in the diagnosis of infectious and parasitic diseases.
3. Results: 3.1. Adult herd:
7
The proportion of seropositive adult animals observed in the study was 38.34% (CI95%: 34.53-42.07), with the highest seroprevalence rate detected in September 2010 just after the grazing season (χ²=97.20; df=4; p<0.001 and χ²=37.400; df=4; p<0.001) (Table 1). A total of 85 seropositive animals presented with bovine besnoitiosis clinical signs, and the clinical prevalence rate revealed a statistically significant decrease at the end of the study (February 2011) (LR χ²=50.95; df=4; p<0.001) (Table 1). IFAT antibody titres in seropositive animals ranged from 1/100 to 1/6400 and showed a significant decrease in intensity from 1/400 to 1/200 during the second period of the study (from September 2010 to February 2011) (LR χ²=52.17; df=14; p<0.001). In regard to the CI, 34.57% of new infections were detected within the susceptible population during the mountain period compared to 24.59% observed in the valley period (χ²=128.90; df=2; p<0.001) (Table 2). The IDs found in the study were 0.058 and 0.061 new infections per animal-month for the mountain and valley periods, respectively (Table 2). Concerning factors associated with the probability of seroconversion, the final model accounted for a 6.9% of the deviance. Overall, the probability of becoming seropositive in the herd was associated with a mathematical combination of Np and its interaction with time in housing (Table 3), whose global result suggests a direct relationship between Np and time in housing with seroconversion probability, as illustrated in Figure 1. Moreover, according to the predictive model, the risk associated with seroconversion probability was higher during the mountain period than during the valley period (Table 3). Variables related to clinical disease dynamics explained 32.5% of the deviance in the final model. In agreement with this test, animals that seroconverted during the mountain period had a higher risk of developing bovine besnoitiosis clinical signs (Table 4). Additionally, the risk of suffering clinical disease increased with the age of the animals and was directly related to high immune responses (Table 4). Regarding the average age of animals that seroconverted during the mountain period, cows showing clinical besnoitiosis were notably older (7.7±4.1 years old) than those subclinically infected animals (4.3±2.7 years old) (Student´s t-test=-3.36; df=44; p=0.0020) (Figure 2). A final model was fit for factors associated with antibody levels, which explained 11.71% of the variation, suggesting that the immune response intensity was directly related to the presence of clinical signs (Table 5). Likewise, antibody levels appeared to also be associated with, although not significantly,
8
the breed of the cows (Table 5). According to these results, clinically infected animals revealed stronger antibody responses (1/800) than subclinically infected cows (1/100) (LR χ²=17.58; df=6; p=0.007) (Figure 3). 3.2. Calf herd: The total seroprevalence observed in the calf herd was 15.17% (CI95%: 9.36-21.04), with the seroprevalence rate higher in spring-calving animals (χ²=9.43; gl=1; p=0.002) (Table 6). Furthermore, 50% of seropositive calves (11 out of 22) were born to seropositive dams (χ²=9.47; gl=1; p=0.002) (Table 7). After monitoring the clinical status of the calves, parasitic cysts in the scleral conjunctivae compatible with the chronic stage of bovine besnoitiosis could be noted in 3 out of 16 seropositive spring-calving animals. It is important to highlight that these three clinically affected calves were descendants of seropositive infected cows showing clinical disease. In regard to the immune response of seropositive calves, the antibody titre intensity was higher in the spring-calving animals than in the autumn-calving animals, with antibody mean titres of 1/200 and 1/100, respectively (LR χ²=10.35; df=4; p=0.035). The model was fit to explore factors associated with the probability of seroconversion in calves accounted for 14% of the deviance and was directly related to the time of calving and the serological status of the dams (Table 8). Animals born in the spring-calving season and that were descendants of seropositive dams had a higher risk of seroconversion.
4. Discussion: Despite the wide range of seroprevalence values previously reported, the total seroprevalence rate observed in the adult herd (38.34%) was in agreement with research performed in other endemic areas from France, Italy, Portugal and Spain (Fouquet et al., 2009; Rinaldi et al., 2013; Waap et al., 2014; Gutiérrez-Expósito et al., 2014). A similar trend was observed in a recent study conducted in three infected beef cattle herds located in the Urbasa-Andía Mountains (Navarra, Spain), where an average serological incidence rate of 22% was reported (Gutiérrez-Expósito et al., 2015). Concerning clinical
9
disease, signs compatible with bovine besnoitiosis were observed in 13.30% of the herd, which indicates that the majority of infections were subclinical. However, the seroprevalence detected at the end of the study suggests effective B. besnoiti transmission by subclinically infected cattle. In this context, several authors have hypothesized that seropositive animals without visible clinical signs (commonly referred to as subclinical cases) could occasionally present with B. besnoiti DNA in their skin and may contribute to parasite spread (Frey et al., 2013; Gollnick et al., 2015). In January 2010, all clinically infected animals were seropositive, while in September 2010 and February 2011, seven clinically affected animals did not show an antibody response. This finding could be mainly attributed to an IFAT test failure to detect the antibody response, although it could be also due to an immune tolerance to the parasite or a serological recovery of these animals; however, there is a lack of well-controlled studies on this subject. It should be noted that the high seroprevalence and CI values detected during the mountain period could be explained by the cumulative effect associated with the difference in time between periods (7.3 months for the mountain period versus 5 months for the valley period) and not the concurrence of differentiating epidemiological aspects such as the supraforestal-grazing season or the activity of vector insects during the mountain period. Nevertheless, it was crucial to compensate for the variation in time between the periods by calculating a monthly ID rate. Taking into account all of these circumstances, the similar ID obtained for the mountain and valley periods seems to indicate a continuous parasite transmission throughout the study, which could be in disagreement with the seasonal variation of the disease previously suggested by some authors (Fernandez-García et al., 2010; Jacquiet et al., 2010; Alzieu et al., 2011). As a result of the seroepidemiological study performed by regression models, the seroconversion probability of B. besnoiti infection appeared to be directly associated with the number of seropositive cows with whom an animal had been stabled with, as well as the housing period duration, supporting the role of horizontal transmission by close contact as one of the most important ways of disease spread. The final model fit for factors associated with the probability of seroconversion indicated that the more seropositive cows in cohabitation and the longer the housing period duration, the greater the probability an animal would become infected. These findings suggest that disease transmission in herds is mainly driven by housing, when management measures are more intensive and when cattle cohabitate in close
10
contact for long periods. Similarly, a cohabitation experiment conducted in Germany, where six healthy animals were housed on a pasture together with three besnoitiosis clinically affected cows, showed that after 12 weeks, 3 out of the 6 healthy animals become infected, highlighting the important role of close contact between animals in disease transmission (Gollnick et al., 2015). Furthermore, Bigalke already demonstrated the mechanical transmission of the disease through direct contact by inoculating parasite cystic stages (bradyzoites) into the nostrils of a cow, which led to the hypothesis of the ability of parasite cysts to pass through mucosal membranes (Bigalke, 1968). Another important aspect to consider, as a consequence of this analysis, was the apparent lack of relationship between the breed and the age of the animals and their seroconversion probability. These results confirm previous observations describing the susceptibility to become infected among a wide variety of breeds (Álvarez-García et al., 2014a; ÁlvarezGarcía et al., 2014b). Our results seem to suggest that age is not a risk factor, contrary to the increase in prevalence with age reported by other authors (Álvarez-García et al., 2014a; Álvarez-García et al., 2014b; Gutiérrez-Expósito et al., 2014; Waap et al., 2014). However, it is important to explain that we studied the probability of seroconversion among healthy animals in a particular period of time depending on their age, while other investigations described the prevalence by age, and the increase observed may be associated with a cumulative effect over time. In fact, most authors agree that the increase in seroprevalence found in older animals could be mainly due to successive parasite exposures accumulated with age. In contrast, this investigation shows that the risk of developing bovine besnoitiosis clinical course increased with animal age. During the mountain period, seropositive clinically infected animals were on average three year older than seropositive subclinically infected animals. As previously described, certain authors suggest an increase in seroprevalence with age together with the hypothesis that older animals are more often clinically affected (Fernández- García et al., 2010; Álvarez-García et al., 2014b). Supporting this theory, Liénard et al. observed, in a longitudinal study performed in French dairy cattle, that only animals four years old and older showed clinical signs of besnoitiosis (cysts in the scleral conjunctivae) (Liénard et al., 2011). Potential causes that could lead an infected animal to develop chronic clinical disease are largely unknown. In this study, we demonstrated that age seems to play an important role in that regard. This could be explained by the existence of biological differences due to animal age, such as aspects related to the immune system, or by certain characteristics associated with animal behaviour depending on age, which could predispose animals towards one or another way of transmission, and consequently, animals would receive different doses or exposures to parasites.
11
Furthermore, the development of clinical signs in this study implied higher antibody responses, with mean dilutions of 1/800 and 1/100 for antibody titres from clinically and subclinically infected cows, respectively. This association between the intensity of the immune response and clinical signs have been previously described by other authors, who suggest that this correspondence may be due to a continuous stimulation of the immune system proportional to the presence of tissue cysts (Liénard et al., 2011; Esteban-Gil et al., 2014.). With the aim of decreasing disease transmission in herds, these results highlight the possibility of using antibody titres as a marker of disease severity and more specifically, as a tool of choice for the selective screening of highly infectious animals (Alzieu et al., 2011; Gollnick et al., 2015). The prevalence of antibodies and the clinical signs recorded in calves demonstrates the ability of B. besnoiti to infect and even cause disease in animals younger than 6 months old, supporting the susceptibility of cattle to the disease regardless of age, as Alzieu et al. demonstrated before detecting tissue cysts in young calves from 4-5 months old (Alzieu et al., 2011). Supposing that the antibodies observed in our study were due to a transfer of maternal immunity (Shkap et al., 1994), we would expect that all calves born to B. besnoiti seropositive dams presented with antibodies. However, only half of the infected calves descended from seropositive cows, suggesting that the antibodies detected were consistent with recent infections. Consequently, these results indicate that maternal antibodies were not transferred to calves for a protective immune response against subsequent infections. Regarding putative routes of transmission to calves, Hornok et al. noted an unlikely transmission pathway via the colostrum, given the lack of B. besnoiti DNA excretion in the colostrums of seropositive cows (Hornok et al., 2015). Similarly, Frey et al. described the improbability of the vertical transmission of the disease, as previously reported by Nobel et al. (Nobel et al., 1981), due to the failure to detect the parasite in the upper tract of the reproductive organs of infected cows (Frey et al., 2013). Based on similar studies of bovine neosporosis (Schares et al., 1998; Davison et al., 1999), in which congenital transmission was observed in 94% of seropositive cows, considering transplacental or congenital disease transmission in our study seems unlikely since approximately 70% of seropositive infected dams had healthy offspring, and contrary, 10% of healthy females had infected offspring, which suggests postnatal transmission between dams and offspring through close contact during the suckling period. Evidence for this pathway could be observed in the regression models of the present study, which revealed that calf descendants of seropositive dams had double the probability of seroconversion than those born to seronegative cows. According to a recent
12
study, for seropositive pregnant cattle, the immediate separation of the calves after the ingestion of colostrum and subsequent artificial feeding could be a successful strategy for preventing parasite transmission (Hornok et al., 2015). Additionally, the seroprevalence was higher in the spring than in the autumn-calving animals, which could be due to the slight difference of age between the calving time points (spring-calving animals were one month and a half older than autumn-calving animals, being 6.5 and 5.2 months old, respectively). Therefore, among calves born in the spring, the supraforestal-grazing season, the increased activity of vector insects during the summer months or just the difference in age could increase the risk of infection, which would explain the increased seroprevalence detected in calves born in the autumn. Additionally, the antibody titres of animals born in the spring were slightly higher, consistent with the presence of calves with clinical disease. As in the adult herd, clinically affected calves had higher antibody titres (≥ 1/400).
Regarding the explanatory capacity of the models and, specifically, the models of seroconversion probabilities, their explanatory power was low, suggesting the existence of other factors that modulated the epidemiology of the disease and could not be controlled in this study. However, it is important to consider that the explanatory capacity of the models may be reduced by not controlling for important factors and also a lack of means for recording such factors. Thus, the reduced explanatory power observed in this study could be mainly because complex epidemiological mechanisms of the disease, such as direct parasite transmission rates between cows, were measured with rough estimates of the time in housing and the number of seropositive cows with whom an animal had been stabled, with the added difficulty of not knowing how many of these seropositive cows were infectious. Finally, a hypothesis about the origin of the outbreak on the farm was suggested. Based on the evidence observed in the research, the beginning of the outbreak could be attributed to cattle trading, specifically to the introduction of eight new cows into the herd in spring 2008 with a genetic improvement purpose. In fact, all of them showed clinical signs in the course of this research. Consequently, and considering all the results drawn from the study, we recommend a set of measures to avoid infection diffusion in herds. Disease control strategies in farms should be based on two main approaches. Farms should avoid the introduction of parasites into herds by rigorous controls as new animals arrive. Additionally, affected farms should avoid parasite transmission by focusing on visual observation of parasitic cysts in the scleral conjunctivae and serological analyses coupled with the progressive culling of infected animals and
13
following management measures including the physical separation of infected and healthy cattle (even infected dams and their offspring) until the total eradication of seropositive animals is achieved. To conclude, the present work identified the main factors associated with infection and clinical disease dynamics in a bovine besnoitiosis outbreak, which may act as predictive markers of infection and disease transmission under field conditions. The seroepidemiological investigation suggests continuous parasite transmission throughout the study, including among subclinically infected animals, likely due to long periods of close contact occurring when animals are housed in the farm and when management measures become more intensive. Clinical disease was mainly observed in older animals and was characterized by immune responses of higher intensity. Moreover, the results demonstrate that calves can become infected and can even develop clinical disease when they are less than 6 months old. Besnoitia besnoiti transmission between dams and offspring in this herd may be attributed to postnatal transmission via close contact during the suckling period.
Conflict of interest statement: All authors declare that they do not have any financial, personal or other relationships with other people or organizations that could inappropriately influence their work.
Acknowledgments: Part of this research was financially supported by the General Administration of Food and Agrifood Development (Husbandry Department of Aragon Government) (project number OTRI 2011/1023). The authors wish to thank the staff of La Garcipollera Research Farm and the Agrifood Research and Technology Centre of Aragon (CITA) for supporting the acquisition and collection of the samples and data. Likewise, we acknowledge the SALUVET group (Animal Health Department, Faculty of Veterinary Sciences, Complutense University of Madrid) for their excellent technical assistance. Finally, Adriana Esteban-Gil would to thank the professors and colleagues from the Parasitology and Parasitic Disease Department of the Veterinary Science Faculty of Zaragoza for all these years of collaboration.
14
References: Álvarez-García, G., Fernández-García, A., Gutiérrez-Expósito, D., Ruíz-Santa Quiteria, J.A., Aguado-Martínez, A. & Ortega-Mora, L.M., 2014a. Seroprevalence of Besnoitia besnoiti infection and associated risk factors in cattle from an endemic region in Europe. Vet. J. 200, 328331. Álvarez-García, G., García-Lunar, P., Gutiérrez-Expósito, D., Shkap, V. & Ortega-Mora, L.M., 2014b. Dynamics of Besnoitia besnoiti infection in cattle. Parasitology 141, 1419-1435. Alzieu, J.P., Dorchies, P., Schelcher, F. & Gottstein, B., 2007. L´extension de la besnoitiose bovine en France. Le Point Vétérinaire 38, 37-43. Alzieu, J.P., Jacquiet, P., Liénard, E., Grisez, C., Prevot, F., Malavieille, R., Desclaux, X., Hussbaum, S., Bergeaud, J.P., Franc, M., Shelcher, F., Corboz, N., Bastien, F., GuerrierChätelet, M.C., Gaver, L., Schneider, V. & Boulon, C., 2011. Réémergence de la besnoitiose bovine: démarche diagnostique et possibilités de contrôle. Bulletin des GTV 58, 71-86. Basso, W., Lesser, M., Grimm, F., Hilbe, M., Syder, T., Trösch, L., Hansueli, O., Braun, U. & Deplazes, P., 2013. Bovine besnoitiosis in Switzerland: Imported cases and local transmission. Vet. Parasitol. 198, 265-273. Beck, R., Stokovic, I., Pleadin, J. & Beck, A., 2013. Bovine besnoitiosis in Croatia. In: Proceedings of the 2nd International Meeting on Apicomplexan Parasites in Farm Animals (ApiCOWplexa: Kusadasi, Turkey), p. 64. Besnoit, C. & Robin, V., 1912. Sarcosporidiose cutanée chez une vache. Recueil Vétérinaire 37, 649663. Bigalke, R.D., 1968. New concepts on the epidemiological features of bovine besnoitiosis as determined by laboratory and field investigations. Onderstepoort J. Vet. Res. 35, 3-137. Casasús, I., Sanz, A., Villalba, D., Ferrer, R. & Revilla, R., 2002. Factors affecting animal performance during the grazing season in a mountain cattle production system. J. Anim. Sci. 80, 1638–51. Cortes, H., Leitao, A., Vidal, R., Vila-Vicosa, M.J., Ferreira, M.L., Caeiro, V. & Hjerpe, C.A., 2005.
15
Besnoitiosis in bulls in Portugal. Vet. Rec. 157, 262-264. Cortes, H., Leitão, A., Gottstein, B. & Hemphill, A., 2014. A review on bovine besnoitiosis: a disease with economic impact in herd health management, caused by Besnoitia besnoiti (Franco and Borges, 1916). Parasitology 141, 1406–17. Davison, H.C., Otter, A. & Trees, A.J., 1999. Estimation of vertical and horizontal transmission parameters of Neospora caninum infections in dairy cattle. Int. J. Parasitol. 29, 1683–9. EFSA European Food Safety Authority, 2010. Bovine besnoitiosis: an emerging disease in Europe. Scientific statement on Bovine Besnoitiosis, Question No EFSA-Q-2009-00879, EFSA Journal 8, 1499. http://www.efsa.europa.eu/en/efsajournal/pub/1499.htm. Adopted 28 January 2010. Esteban-Gil, A., Grisez, C., Prevot, F., Florentin, S., Decaudin, A., Picard-Hagen, N., Berthelot, X., Ronsin, P., Alzieu, J.P., Marois, M., Corboz, N., Peglion, M., Vilardell, C., Liénard, E., Bouhsira, E., Castillo, J.A., Franc, M. & Jacquiet, P., 2014. No detection of Besnoitia besnoiti DNA in the semen of chronically infected bulls. Parasitol. Res. 113, 355-362. Fernandez-Garcia, A., Alvarez-Garcia, G., Risco-Castillo, V., Aguado-Martinez, A., MaruganHernandez, V. & Ortega-Mora, L.M., 2009. Pattern of recognition of Besnoitia besnoiti tachyzoite and bradyzoite antigens by naturally infected cattle. Vet. Parasitol. 164, 104-110. Fernandez-García, A., Alvarez-Garcia, G., Risco-Castillo, V., Aguado-Martinez, A., Marcen, J.M., Rojo-Montejo, S., Castillo, J.A. & Ortega-Mora, L.M., 2010. Development and use of an indirect ELISA in an outbreak of bovine besnoitiosis in Spain. Vet. Rec. 166, 818-822. Fouquet, C.M., 2009. La besnoitiose bovine: suivi épidémiologique de l´epizootie de la región PACA. Thèse Vétérinaire Médecine, Lyon. Frey, C.F., Gutiérrez-Expósito, D., Ortega-Mora, L.M., Benavides, J., Marcén, J.M., Castillo, J.A., Casasús, I., Sanz, A., García-Lunar, P., Esteban-Gil, A. & Alvarez-García, G., 2013. Chronic bovine besnoitiosis: intra-organ parasite distribution, parasite loads and parasiteassociated lesions in subclinical cases. Vet. Parasitol. 197, 95-103. García-Lunar, P., Ortega-Mora, L.M., Schares, G., Gollnick, N.S., Jacquiet, P., Grisez, C., Prevot, F., Frey, C.F., Gottstein, B. & Álvarez-García, G., 2013. An Inter-Laboratory Comparative
16
Study of Serological Tools Employed in the Diagnosis of Besnoitia besnoiti Infection in Bovines. Transbound. Emerg. Dis. 60, 59–68. Gollnick, N.S., Scharr, J.C., Schares, G. & Langenmayer, M.C., 2015. Natural Besnoitia besnoiti infections in cattle: chronology of disease progression. BMC Vet. Res. 11, 35. Gutiérrez-Expósito, D., Esteban-Gil, A., Ortega-Mora, L.M., García-Lunar, P., Castillo, J.A., Marcén, J.M. & Álvarez-García, G., 2014. Prevalence of Besnoitia besnoiti infection in beef cattle from the Spanish Pyrenees. Vet. J. 200, 468-470. Gutiérrez-Expósito, D., Ortega-Mora, L.M., García-Lunar, P., Rojo-Montejo, S., Zabala, J., Serrano, M. & Álvarez-García, G., 2015. Clinical and serological dynamics of Besnoitia besnoiti infection in three endemically infected beef cattle herds. Transbound. Emerg. Dis. DOI: 10.1111/tbed.12402. Hofmeyr, C.F.B., 1945. Globidiosis in cattle. S. Afr. Vet. Med. Assoc. 6, 102-109. Hornok, S., Fedák, A., Baska, F., Hofmann-Lehmann, R. & Basso, W., 2014. Bovine besnoitiosis emerging in Central-Esatern Europe, Hungary. Parasit. Vectors. 7, 20. Hornok, S., Fedák, A., Baska, F., Basso, W., Dencso, L., Tóth, G., Szeredi, L., Abonyi, T. & Dénes, B., 2015. Vector-borne transmission of Besnoitia besnoiti by blood-sucking and secretophagous flies: epidemiological and clinicopathological implications. Parasit. Vectors. 8, 450. Jacquiet, P., Liénard, E. & Franc, M., 2010. Bovine besnoitiosis: Epidemiological and clinical aspects. Vet. Parasitol. 174, 30-60. Juste, R.A., Cuervo, L.A., Marco, J.C. & Oregui, L.M., 1990. La besnoitiosis bovina: ¿desconocida en España?. Medicina Veterinaria 7, 613-618. Liénard, E., Salem, A., Grisez, C., Prévot, F., Bergeaud, J.P., Franc, M., Gottstein, B., Alzieu, J.P., Lagalisse, Y. & Jacquiet, P., 2011. A longitudinal study of Besnoitia besnoiti infections and seasonal abundance of Stomoxys calcitrans in a dairy cattle farm of southwest France. Vet. Parasitol. 177, 20-27.
17
Neuman, M. & Nobel, T.A. 1960. Globidiosis of cattle and sheep in Israel. Refuah Veterinarith 17, 101103. Nobel, T.A., Klopfer, U., Perl, S., Nyska, A., Neumann, M. & Brenner, G., 1981. Histopathology of genital besnoitiosis of cows in Israel. Vet. Parasitol. 8, 271-276. Papadopoulos, E., Arsenos, G., Ptochos, S., Katsoulos, P., Oikonomou, G., Karatzia, M.A. & Karatzias, H., 2014. First report of Besnoitia besnoiti seropositive cattle in Greece. J. Hellenic Vet. Med. Soc. 65, 115-120. Rinaldi, L., Maurelli, M.P., Musella, V., Bosco, A., Cortes, H. & Cringoli, G., 2013. First crosssectional serological survey on Besnoitia besnoiti in cattle in Italy. Parasitol. Res. 112, 18051807. Ryan, E.G., Lee, A., Carty, C., O´ Shaughnessy, J., Kelly, P., Cassidy, J.P., Sheehan, M., Johnson, A. & de Waal, T., 2016. Paper Bovine besnoitiosis (Besnoitia besnoiti) in an Irish dairy herd. Vet. Rec.178. Schares, G., Peters, M., Wurm, R., Bärwald, A. & Conraths, F.J., 1998. The efficiency of vertical transmission of Neospora caninum in dairy cattle analyzed by serological techniques. Vet. Parasitol. 80, 87–98. Shkap, V., Pipano, E., Marcus, S. & Krigel, Y., 1994. Bovine besnoitiosis: transfer of colostral antibodies with observations possibly relating to natural transmission of the infection. Onderstepoort J. Vet. Res. 61, 273-275. Thrusfield, M., 2007. Veterinary Epidemiology, 3rd Edition. Blackwell publishing. Vanhoudt, A., Pardon, B., De Schutter, P., Bosseler, L., Sarre, C., Vercruysse, J. & Deprez, P., 2015. First confirmed case of bovine besnoitiosis in an imported bull in Belgium. Vlaams Diergeneeskd Tijdschr 84, 205–11. Waap, H., Nunes, T., Cortes, H., Leitao, A. & Vaz, Y., 2014. Prevalence and geographic distribution of Besnoitia besnoiti infection in cattle herds in Portugal. Parasitol. Res. 113, 3703-3711.
18
Figure legend: Figure 1. 3D surface plot illustrating the seroconversion probability (Y) dependent on the number of seropositive cows with whom an animal had been stabled with (X) and the time in housing (Z) in the mountain (left-hand) and valley (right hand) periods.
19
Figure 2. Age distribution (years old) of clinically (black bar) and subclinically (grey bar) infected cows that seroconverted during the mountain period.
20
Figure 3. IFAT antibody titres of clinically (black bar) and subclinically (grey bar) infected cows.
21
Tables: Table 1. Within-herd seroprevalence and clinical prevalence rates in the adult herd at three times of sampling.
Seroprevalence
Clinical prevalence
n
Seropositive animals
Serological rate (%) CI95%
Animals showing clinical signs
Clinical (%)
219
65
29.68 (23.65-35.75)
38
17.35
223
115
51.57(44.94-58.06)a
44
19.73
February 2011
197
65
32.99 (26.43-39.57)
3
1.52b
Total
639
245
38.34 (34.53-42.07)
85
13.30
Time sampling
of
January 2010 September 2010
rate
Statistical significance according to Pearson´s chi-squared testa (χ²=97.20; df=4; p<0.001 and χ²=37.400; df=4; p<0.001) and Likelihood ratiob (LR χ²=50.95; df=4; p<0.001)
22
Table 2. Cumulative incidence (CI) (% of new infected cases among disease-free population) and incidence density (ID) (new infected cases per animal-month) for the mountain and valley periods.
Study period
CI
ID
Mountain
34.57
0.058
Valley
24.59
0.061
p
<0.001*
*Statistical significance according to Pearson´s chi-squared test (χ²=128.90; df=2; p<0.001)
23
Table 3. Final model fit for factors associated with the seroconversion probability (explained 6.9% of final model deviance).
Variable
Coefficient
SE
Wald
p-value
Constant
-1.411
0.588
5.75
0.016
Period (mountain)
0.530
0.260
4.16
0.041
√Number of seropositive cows (Np)
-1.125
0.492
5.23
0.022
√Time in housing (days) *√Np
0.229
0.104
8.22
0.004
24
Table 4. Final model fit for factors related to clinical disease dynamics (explained 32.5% of final model deviance).
Variable
Coefficient
SE
Wald
p-value
Constant
-8.799
1.835
22.99
<0.001
Period (mountain)
1.159
0.533
4.72
0.030
Age
1.259
0.351
12.89
<0.001
IFAT titre log
1.478
0.504
8.61
0.003
Period*seroconversion code
0.672
0.243
7.64
0.006
25
Table 5. Final model fit for factors associated with antibody levels detected by IFAT (explained 11.71% of final model variation).
Variable
Coefficient
SE
t
p-value
Constant
2.684
0.044
60.69
<0.001
Clinical signs (positive)
0.171
0.043
3.97
<0.001
Breed (Brown Swiss)
-0.068
0.038
-1.77
0.078
26
Table 6. Within-herd seroprevalence and clinical prevalence rates in the calf herd according to the time of calving.
Seroprevalence Time calving
of
Clinical prevalence
n
Seropositive animals
Serological rate (%) CI95%
Animals showing clinical signs
Clinical rate (%)
Autumn 2009
83
6
7.23 (1.64-12.76)
0
0
Spring 2010
62
16
25.81 (14.91-36.69)*
3
4.84
Total
145
22
15.17 (9.36-21.04)
3
4.84
*Statistical significance according to Pearson´s chi-squared test (χ²=9.43; df=1; p=0.002)
27
Table 7. Frequency distribution of serological calf results according to the serological status of their dams.
Calf results
Serological status of dams
Seropositive
Seronegative
Total
Seropositive
11*
24
35
Seronegative
11
99
110
Total
22
123
145
*Statistical significance according to Pearson´s chi-squared test (χ²=9.47; df=1; p=0.002)
28
Table 8. Final model fit for factors related to the seroconversion probability in the calf herd (explained 14% of final model deviance).
Variable
Coefficient
SE
Wald
p-value
Constant
-1.572
0.266
34.77
<0.001
0.708
0.253
7.83
0.005
0.747
0.265
7.97
0.005
Serological (positive)
status
of
dams
Time of calving (spring 2010)
29