- Email: [email protected]
Sero-prevalence and risk factors of infectious bovine rhinotracheitis virus (type 1) in Meru County, Kenya
Journal Pre-proof Sero-prevalence and risk factors of infectious bovine rhinotracheitis virus (type 1) in Meru County, Kenya Essau Serem Kipyego
PII:
S0167-5877(19)30456-8
DOI:
https://doi.org/10.1016/j.prevetmed.2019.104863
Reference:
PREVET 104863
To appear in:
Preventive Veterinary Medicine
Received Date:
7 July 2019
Revised Date:
16 November 2019
Accepted Date:
29 November 2019
Please cite this article as: Kipyego ES, Sero-prevalence and risk factors of infectious bovine rhinotracheitis virus (type 1) in Meru County, Kenya, Preventive Veterinary Medicine (2019), doi: https://doi.org/10.1016/j.prevetmed.2019.104863
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Published by Elsevier.
Sero-prevalence and risk factors of infectious bovine rhinotracheitis virus (type 1) in Meru County, Kenya
Essau Serem Kipyego
University of Nairobi
ro of
[email protected]
Abstract
The aim of the study was to determine the antibody sero-prevalence of Bovine Herpesvirus-1
-p
which cause Infectious Bovine Rhinotracheitis (IBR) and to identify risk factors associated with
re
BHV-1 antibody seropositivity among smallholder dairy farms in Meru County, Kenya. A cross-sectional study was conducted in the Naari area of Meru County, Kenya between
lP
September-October 2016 and March-April 2017. The 149 farmers were randomly selected from members of the Naari Dairy Farmers Cooperative Society who were actively delivering milk to
ur na
the society at the time of the study. Serum samples were obtained from 403 female dairy cattle. Farm level management and animal factors were collected through direct interviews with the owner or someone who was knowledgeable about the animals. All serum samples were processed
Jo
with an indirect enzyme-linked immunosorbent assay (gB ELISA) to determine the presence of antibodies to BHV-1. The overall farm-level and animal-level sero-prevalences of BHV-1 antibodies were 30.9% (95% CI: 23.6% to 39.0%) and 17.4% (95% CI: 13.8% to 21.4%), respectively. In the final multivariable analysis, the factors significantly associated with BHV-1 antibodies included; age of the dairy cattle (OR=1.200, p=0.001), age of the principal female farmers (OR=0.182,
1
p=0.001) and rearing goats in the farm (OR=26.77, p=0.000). There was a significant interaction between rearing goats on the farm and age of the dairy cattle (p < 0.010); younger cattle seemed to have been exposed to BHV or a cross-reacting caprine herpesvirus when goats were on the farm. The results showed that BHV-1 was circulating among the cattle population in the Naari area of Meru County. Given that there is not BHV-1 vaccination use in this study population, training on the importance of biosecurity and vaccination for BHV-1 are recommended to reduce the transmission and impacts of BHV-1.
ro of
Keywords: Dairy Cattle, Infectious Bovine Rhinotracheitis (IBR), Bovine Herpesvirus type 1
Jo
ur na
lP
re
-p
(BHV-1), Sero-Prevalence, Risk factors.
2
Introduction Infectious Bovine Rhinotracheitis (IBR) is a multi-organ disease of significant economic importance worldwide caused by Bovine Herpes Virus-1 (BHV-1), and it affects both domestic and wild ruminants (Bowland et al., 2000; Muylkens et al., 2007). Bovine Herpes Virus-1 is a virus of the genus Varicellovirus, subfamily Alphaherpesvirinae and family Herpesviridae, and is a highly contagious and infectious virus (King et al., 2012; Biswas et al., 2013; Newcomer and Givens, 2016). Infectious bovine rhinotracheitis in bovine animals is caused by BHV-1.1, the respiratory subtypes, while strain BHV-1.2a and BHV-1.2b are the genital subtypes, and BHV-
ro of
1.3 is the encephalitic subtype (Muylkens et al., 2007). Contact with Infected cattle is the main source of BHV-1 infection to susceptible animals and herds. These cattle shed the virus through various body secretions and excretions (Takiuchi et al.,
-p
2005; Constable et al., 2017). For the respiratory subtype, the virus is transmitted through direct transmission of nasal discharges between animals or by indirect transmission of aerosol droplets,
re
and the infectivity duration of these aerosols is dependent on environmental factors such as
lP
humidity and temperature. Direct contact with non-nasal mucosal discharges, contaminated semen, fetal tissues, and genital fluid can also lead to transmission (Mars et al., 1999; Mars et al.,
ur na
2000; Kahrs, 2001).
Cattle of all breeds and ages are equally susceptible, and the disease is common in cattle above 6 months of age due to waning of maternal antibody protection and increased mixing of cattle populations (Majumder et al., 2015; Seyfi Abad Shapouri et al., 2016; Constable et al., 2017).
Jo
The IBR diseases have no inherent seasonal variability, although in temperate countries, during the months of fall and winter, disease occurrence is high due to feedlot cattle assembling (Majumder et al., 2015). The managerial and environmental risk factors contributing to the spread of BHV–1 include; purchasing of infected cattle, participation in agricultural shows, increased herd size and type of production system (Gay and Barnouin, 2009; Constable et al., 2017). Uncontrolled movement of visitors and cattle into the farm, and unreliable records of vaccination 3
dates have also been associated with the spread of the disease (Boelaert et al., 2005; GonzálezGarcia et al., 2009). The prevalence and risk factors of BHV-1 infection in Kenya have not been recently researched in predominantly zero-grazing farming regions of Kenya. In this cross-sectional study, a random sample of dairy cattle was examined to determine the BHV-1 antibody sero-prevalence, and to identify risk factors associated with BHV-1 antibody seropositivity on smallholder dairy farms using primarily zero-grazing in Naari area, Meru County, Kenya.
ro of
Materials and Methods Ethical approval
The study was approved by the Biosafety, Animal use and Ethics Committee, Faculty of
-p
Veterinary Medicine, University of Nairobi (FVM/BAUEC/2018/146).
re
Selection of study area and farms
The study area of Naari in Meru County, Kenya (Figure 1), was purposively selected since this
lP
research formed part of a larger study involving smallholder dairy farmers. A non-governmental organization called Farmers Helping Farmers, the University of Prince Edward Island (Canada)
ur na
and the University of Nairobi had an existing developmental partnership with the Naari Dairy Farmers Cooperative Society, and this formed the basis for the entry point to the community. Meru lies at latitude and longitude 0°6'0" N and 37°34'60" E, respectively, and is located in the Mount Kenya region of Kenya, approximately 270 km North of Nairobi, the capital city of
Jo
Kenya. The Naari area is situated at an altitude of approximately 2000 m above sea level. The climate in Meru is warm and temperate with volcanic ash soils, and therefore is considered to have high agricultural potential. The average annual temperature in the Meru area ranges from 14oC to 17oC in the highlands (where Naari is), while in the lowlands it ranges from 22oC to 27oC. Precipitation in the high altitude Meru areas average 2200 mm, while low altitude areas
4
average 500 mm. The main agricultural activities in Naari include; lumbering, horticulture, crop production and dairy production. [Insert Figure 1 near here] The sampling frame for the study consisted of 568 farmers who were active members of the Naari Dairy Farmers Cooperative Society (NDFCS) and delivering milk to the cooperative. The sample size of dairy cattle sampled in the study was estimated based on a prevalence of BHV-1 infection of 50% (for a maximum variance), a precision of 5% and confidence level of 95%, giving a
ro of
required sample size of 385 (Dohoo et al., 2009). The average number of cattle above 6 months of age per farm had been established to be approximately 3, thus 149 farms were randomly selected from the 568 smallholder dairy farms from the registry of active members using software-based random number generation. Therefore,
re
-p
the study population of farms represented 26% of the farmers who were members of the NDFCS.
Data and sample collection
lP
The selected farms were visited between September-October 2016 and March-April 2017, and a questionnaire was administered to capture farm- and animal-level factors of BHV-1. The data
ur na
collected included collection of animal-level information about milking cows and heifers on the farm. A systematic scrutiny of written farm records was also conducted to obtain the age and calving history of the cattle, along with history of respiratory and reproductive diseases, periparturient conditions, and mastitis cases. Other farm-level management information collected
Jo
included: feed and mineral supplementation, vaccination status (including use of IBR vaccine), cattle owner attendances to any dairy husbandry training, herd size, awareness and monitoring of heat signs, and source of animals with respect to the last 12 months. In addition, 5 ml of whole blood was collected in plain redtop vacutainer tubes via the tail vein of each dairy animal greater than 6 months of age, and stored in an ice box during the day. Each evening, the blood tubes were taken from the ice box and allowed to clot, and thereafter, the 5
serum was transferred to Eppendorf® vials. The serum was then frozen at -200C and transported on ice to the Hematology and Biochemistry Laboratory, Department of Clinical Studies, Faculty of Veterinary Medicine, University of Nairobi, and then stored at -200C until all samples were collected for testing.
Laboratory analysis The frozen sera were thawed to room temperature (18-260C). The IBR test kit used was the IDEXX IBR gB X3 Ab Test (IDEXX, Liebefeld-Bern, Switzerland), comprising of a BHV-1 Antigen Coated Plate and all the reagents, which was also warmed from its storage temperature of
ro of
40C to room temperature. The IDEXX IBR gB X3 Ab Test is an Indirect Enzyme-Linked Immunosorbent Assay (ELISA) which has been developed to detect presence of antibodies against IBR (type 1) in individual bovine plasma, milk and serum samples. Antibody responses
-p
induced by vaccines which contain the glycoprotein B (gB) of BHV-1 are detected as well. This
re
kit has a reported sensitivity of 100% and specificity of 95%. The testing procedure was done following the protocol described by the manufacturer in Liebefeld-Bern, Switzerland.
lP
The ELISA results were read from a microplate photometer, Mindray Microplate Reader (MR96A, Shenzhen Mindray Bio-Medical Electronics Company Limited), where optical density (OD)
ur na
was measured either at a single wavelength of 450 nm [A (450)] or dual wavelength of 450 nm and 650 nm [A (450/650) for enhanced sensitivity]. The blocking percentage was calculated by using the absorbance obtained with the test sample and subtracting the absorbance of the negative
Jo
control containing no BHV-1-specific antibodies and dividing the result with the negative control containing no BHV-1-specific antibodies. Interpretation of the results was determined via comparing the sample blocking percentages to the expected percentages in accordance with the manufacturer’s test instructions: blocking % < 45 = negative; 45 ≤ blocking % < 55 = suspect; and blocking % ≥ 55 = positive. The suspects were considered as positive in order to obtain a dichotomous outcome. A positive control was used on each plate to confirm that the procedures were followed properly. 6
Data entry and statistical analysis Data collected through the questionnaires and the laboratory results were first entered into MS Excel (Microsoft Inc., Sacramento, California, USA) and then imported to Stata 15 (StataCorp LLC, College station, Texas, USA) for analyses. Initially, the data were checked for accuracy, coded and analyzed using descriptive statistics. Proportions were determined for categorical variables and presented as a percentage of the overall number, along with a 95% confidence interval, where applicable.
ro of
Mixed-effect logistic regression analysis was performed, accounting for clustering of cattle among herds, to determine associations between the categorical and continuous variables and the dichotomized seropositivity outcome (presence or absence of BHV-1 antibody). In the first step, univariable multi-level mixed models for all the predictor variables were fitted into separate
-p
logistic regression models, employing the functional logit. In the second step, a multivariable
re
mixed logistic regression analysis was fitted for all the univariable associations with p≤0.30 in the first step. Correlations between predictor variables were identified using pair-wise correlation,
lP
and where two or more variables were highly correlated (correlation coefficient >0.5), statistical significance and biological plausibility were used to identify which variable would be offered to
ur na
the modeling process. The final models were built using backward stepwise elimination, leaving those variables which had a p-value ≤0.05. Explanatory variables were considered confounders if their removal from the multivariable model modified the coefficients of other significant variables by 30%. Plausible biological interactions between significant explanatory variables in final model
Jo
were also tested and the significant interaction terms were included in the final models (Dohoo et al, 2009). The area under the curve (AUC) of the receiver operating characteristic was used to evaluate the overall model performance.
Results Farm and animal demographics 7
A total of 403 cattle and 149 farms were involved in the study. The principal farmer was mostly comprised of men (56.4%), while women as principal farmers in this sample population were fewer (28.2%), and 15.4% of farms had both the male and female considered jointly as the principal farmer. Most of the principal farmers were married (83.9%), some had lost their spouse, and a few of them were young people who were still single but had established themselves as dairy farmers (5.3%). Among the principal farmers, approximately half (50.3%) were below 45 years. A large majority of the female farmers had completed primary school (91.2%) but only 39.4% had completed secondary school, while the proportions of male farmers having completed
ro of
primary and secondary school were slightly different at 81.5% and 42.3%, respectively. The majority (81.2%) of principal farmers had training on dairy production.
The mean household size recorded in this study was 3.7 ± 1.54, with a minimum of 1 person and
-p
a maximum of 11. The mean total land holdings owned by the respondents was 2.11 ± 2.04 acres. In addition, some of the farmers also had access to other pieces of land (0.51±0.84 acres) through
re
renting, borrowing and government-owned lands leased to them.
lP
Holstein-Friesians formed nearly half of the dairy cattle reared in the region, with small proportions of other exotic and indigenous breeds (Table 1). The mean herd size among the
ur na
sampled farms was 5.7 with a range of 1 to 16 cattle. The mean number of milking cows was 1.8 with a range of 0 and 8. The mean number of dry cows was 0.3 with a range of 0 to 3. Mature cows comprised the majority of animals at 79.7% (321/403), while unbred heifers comprised 12.2% (49/403). The mean age of the sampled animals was 5.5 years with a range of between 0.5
Jo
and 17.0 years. The parities of the cows ranged between 0 and 8, with multiparous cows forming nearly 60% of sampled cattle. It was reported that 9.0% (32/354) of the cows and bred heifers had experienced an abortion in the last one year. Farm management practices at the farms The managerial practices found among the 149 smallholder dairy farms recruited in the study are summarized in Table 1. The zero-grazing and semi-zero-grazing systems formed the majority 8
(nearly 80%) of smallholder dairy farms. Other than dairy production, many of the recruited farmers also practiced sheep and goat production systems. The majority of smallholder dairy farms used artificial insemination exclusively at 57.7% (86/149). More than half of the farmers grazed their cows in community pastures where they could contact other cattle, while less than a quarter of farms disallowed fence-line contact with other cattle. However, there was insignificant movement of the cattle between farms across the region, with less than 10% of farms borrowing cows, lending cows, and purchasing cows. None of the study farmers used a BHV-1 vaccine in the past.
ro of
Sero-prevalence of Bovine Herpesvirus-1 Infection The overall apparent animal-level sero-prevalence to BHV-1 was 17.4% (70/403), with a 95% CI of 13.8% to 21.4%. The overall estimated true animal-level sero-prevalence to BHV-1, adjusting
-p
for the imperfect specificity of the test, was calculated to be 13.0%, with a 95% CI of 9.5% to
re
17.2%. The herd-level sero-prevalence to BHV-1 was 30.9% (46/149), with a 95% CI of 23.6% to 39.0%. The sero-prevalences of the antibodies to the IBR virus in relation to a number of cow
lP
and farm variables are summarized in Table 1.
The sero-prevalence for BHV-1 infection was highest among cattle that were older and
ur na
multiparous, with slightly elevated sero-prevalences among Ayrshire crosses and animals with a history of abortion. At the farm level, farms with goats had a 3 times higher sero-prevalence for BHV-1 among the cattle versus farms without goats, while farms with sheep only had a slightly
Jo
higher BHV-1 sero-prevalence in cattle. The cattle sero-prevalence for BHV-1 was also substantially higher on farms that lent and/or borrowed cattle to/from other farms, and on farms where fence-line contact with neighbour’s cattle was allowed, [Insert Table 1 near here] Factors associated with Bovine Herpesvirus-1 sero-positivity
9
The following animal-level variables met the P–value cut-off criterion (P-value < 0.3) for univariable regression analysis with BHV-1 antibody status: parity category, age category, dry cow status, and BVDV antibody status. Farm-level variables meeting the cut-off criterion included: the farm allowed fence-line contact with other cattle, grazed on community pasture, lent cows to other farms, borrowed cows from other farms, reared goats on the farm, reared sheep on the farm, used natural breeding, age of the male principal farmer, and age of the female principal farmer (Table 2). Pair-wise correlation was carried out and age, parity and age category of the cattle were found to
ro of
be highly correlated (r>0.5). Fence-line contact with other cattle and grazing on community pasture were also found to be highly correlated. Therefore, age of the animal and fence-line contact were offered to the final multivariable model because they had lower P-Values than their
-p
correlates.
re
[Insert Table 2 near here]
The factors associated with sero-positivity to BHV-1 infection in the final multivariable analysis
lP
included (Table 3): dichotomized age of the principal female farmers, age of the dairy cattle reared on the farm, and rearing goats on the farm. In this final model, controlling for confounding
ur na
of variables in the model, the odds of older dairy cattle to have BHV-1 antibody positivity were 1.200 times higher with each additional year of age. Cattle on farms that were rearing goats had 26.77 times the odds to have BHV-1 antibodies compared to cattle on farms not rearing goats. In addition, when the age of the female principal farmer was over 45 years, the odds of cattle on the
Jo
farm to have BHV-1 antibodies were lower by 0.182 times compared with cattle reared by younger women principal farmers under or equal to 45 years (Table 3). The area under the ROC curve for this final model was 0.86, indicating a good overall goodness-of–fit of the observed data (Figure 2). [Insert Table 3 near here]
10
[Insert figure 2 near here] The 95% CI for 1` year and 6 years do not overlap, showing that the interaction is significant (p<0. 0.010). The relationship between cattle age and BHV antibodies depends on whether there are goats or not, with there being a significant positive association when the cattle are young (< 6 years) but no association when the cattle are older than 6 years. Similarly, the relationship between goats on the farm and BHV antibodies depends on the cattle age, with there being a positive association between BHV antibodies and the presence of goats, but only when cattle age is < 6 years (Figure 3).
ro of
[Insert figure 3 near here] Discussion
-p
The results of this study confirmed the presence of IBR infection through sero-prevalence of BHV-1 antibodies among the smallholder dairy cattle reared in primarily zero-grazing units in
re
Naari area of Meru County. The observed sero-prevalence for BHV-1 antibodies from this study was estimated at 17.4%, which was slightly lower compared to studies in other parts of Kenya of
lP
20.9% in the western Kenya (Callaby et al., 2016) and 28.0% in the former Malindi District of the Coastal Region (Kenyanjui et al, 2007). This slightly lower sero-prevalence could have been due
ur na
to the different livestock production systems studied, for example, 42% of our study farms used only zero-grazing, and open grazing is a known risk factor for BHV–1 infection (Snowder et al., 2006; Callaby et al., 2016).
Jo
The observed sero-prevalence to BHV-1 of 17.4% from this study was within the range of 16% 54% that was observed by McDermott et al., (1997) in former Districts across Kenya in 1991– 1992. However, in this study, the suspects were considered as positive in order to obtain a dichotomous outcome, and this interpretation might have led to higher magnitudes and direction of estimates of the sero-prevalence (Friesen et al., 2007 and Dohoo et al., 2009). Conversely, the sero-prevalence observed in Naari area of Meru County was clearly lower than that estimated in
11
traditionally managed herds (i.e. open grazing) in Zambia, which ranged from 42%-76% (Ghirotti et al., 1991), and in Egypt which ranged from 63%-86% (Mahmoud et al., 2009). The observed differences in antibody sero-prevalence in different regions and countries could be explained by factors such as production systems, differences in herd sizes, type of breeding methods, differences in disease-control measures, and age of the cattle (Mainar-Jaime et al., 2001 and Ackermann and Engels, 2006). It has been observed that higher BHV-1 prevalences recorded in larger herd sizes and intensively farmed cattle could be associated with a high level of contact between individual animals within a herd (Snowder et al., 2006). For extensively managed farms
ro of
with median herd size of 5 animals, the risk of contact between a susceptible individual with an infected or persistently infected animal is lower (Callaby et al., 2016). The studies reported above involved animals of various ages, such as 51 weeks old (Callaby et al., 2016), 3 months to adults
-p
(Ghirotti et al., 1991), and zebu adults (Kenyanjui et al., 2007), while our study involved heifer and adult dairy cows. Therefore, the age differences among the animals studied may also have
re
explained some of the variations in the sero-prevalence.
lP
Ghirotti et al., (1991) and Kenyanjui et al., (2007) employed virus neutralization tests (VNT) compared to ELISA tests used in our study, and the differences in test specificity and sensitivity
ur na
may also have also contributed to observed differences in prevalence (Graham et al., 1998). In the studies done by Saravanajayam et al., (2001) it was reported that the ELISA had a higher specificity and sensitivity compared to VNT, thus ELISA is considered a rapid, reliable and technically superior test for detection of BHV-1 antibodies. Callaby et al., (2016) also employed
Jo
an indirect ELISA test, and thus may have contributed a similar prevalence to our study. The results of the final model of this study revealed that a number of risk factors were associated with sero-prevalence of BHV-1 antibody, such as age of the cattle, rearing goats in the farms, and cows that had antibodies against BVDV. Several serological studies carried out worldwide identified similar risk factors associated with BHV–1 sero-positivity (Muylkens et al., 2007; Callaby et al., 2016; Constable et al., 2017). The current study also agrees with other studies 12
which reported a number of risk factors in their studies, such as large herd size, intensive production systems, housing, and management practices, like animal age and sex (males are more frequently positive than females), and direct animal contact through cattle shows and purchasing of cattle (van Schaik, 2001; van Schaik et al., 2002; Solis-Calderon et al., 2003; VonkNoordegraaf et al., 2004; Boelaert et al., 2005). Age of the cattle is a frequent risk factor reported in BHV-1 seropositive cattle, where findings from Solis-Calderon et al., (2003), Carbonero et al., (2011), Romero-Salas et al., (2013), Saravanajayam et al., (2015) and Segura-Correa et al., (2016) showed that older animals had a
ro of
higher sero-prevalence compared to young animals. Since infected cattle remain infected for life, it is not surprising that proportions of infected cattle increase with age.
Cattle on farms with older female principal farmers had a lower odds of infection compared to
-p
cattle on farms with younger female principal farmers. It is unclear why this variable remained significant in our final model, but it may be that older farmers are more knowledgeable of the
re
importance of biosecurity and disease control measures. Further research would help to clarify
lP
this finding.
Based on past epidemiological studies, this study showed that IBR and BVD viruses are closely
ur na
associated in the univariable analyses, and infection for one disease has predisposed to the other disease elsewhere (Callaby et al., 2016). The study also showed that BVDV sero-prevalence was significantly associated with BHV-1 at the farm level. This finding agrees with those reported earlier (Fulton et al., 2000; Callaby et al., 2016). A positive correlation has also been
Jo
demonstrated between infections with the viruses causing IBR and PI (Callaby et al., 2016). The present study revealed that rearing of goats in the farm was associated with BHV-1 seroprevalence in the cattle. Mixed rearing of sheep, goats and cattle was not uncommon in our study population, allowing this variable to become part of the risk factors in the final model. Infectious Bovine Rhinotracheitis has the potential for cross-reaction with four other herpesviruses from other animals including goats and buffaloes (Handel et al., 2011; Callaby et al., 2016). Therefore, 13
this risk factor may be attributed to the potential ability of viral cross-infection with related herpesvirus (Biswas et al., 2013). Further research is needed to clarify this potential crossreaction. The present study showed that introduction of new animals into the herd was significantly associated sero-positivity of BHV-1 in univariable analyses but did not remain significant in the final model. Other past studies by Solis-Calderon et al., (2003) and Segura-Correa et al., (2016) reported that introduction of an animal into the herd was not significantly associated with seroprevalence of BHV-1. However, van Schaik et al., (1998) and Gay and Barnouin, (2009) reported
ro of
increased animal movement into a herd and close proximity neighbouring farms increased the risk of IBR spread through contact between naïve herd and infected animals. While introduction of new animals would biologically make sense as a risk factor for BHV-1, the relatively low sero-
-p
prevalence of BHV-1 in our study population (13.0% estimated true prevalence) and infrequent movement of cattle between herds (6.7%, 8.1%, and 8.7% of farmers borrowed, lent, or bought
re
cattle, respectively – Table 1) would mean that the chances of an infected animal moving between
lP
farms is low.
For the observed significant interaction between cattle age and goats on the farm (p < 0.01),
ur na
younger cattle seemed to have been exposed to BHV or a cross-reacting caprine herpesvirus when goats were on the farm. This could be because farms with cattle < 6 years old just brought in goats to the farm in the last 6 years. It could also be that farms bringing in goats into the farm in the last 6 years is correlated with farms bringing in other cattle in the last 6 years, and so those
Jo
younger cattle are more likely to be exposed to BHV with these new cattle introductions than older cattle.
This study was carried out in smallholder farms with approximate herd size of 3 animals and little variation in herd size, as a result, the study did not have enough variation and power to identify herd size as a risk factor for seropositivity. Other studies suggested that at herd level, herd size is the main important risk factor for BHV-1 infection (Snowder et al., 2006; Segura-Correa et al., 14
2016). Our herd sizes were small, making it less likely for our herds to be aggregates of animals from other farms. A limitation to this study is that it is a cross-sectional design, which has limited ability to confirm temporality of the risk factors for BHV-1 infection. Furthermore, the study used a BHV-1specific antibody ELISA testing kit, thus this test cannot establish whether the test-positive animal is due to the virus retention, or due to adaptive immunity from previous exposure to the infection. In the future, a study is needed to establish if clinical IBR occurs in smallholder dairy farmers in the Meru area of Kenya in order to determine the long-term effects of BHV-1 infection
ro of
in cattle production that would justify formulation of prevention and control measures, such as vaccination. Conclusions and recommendations
-p
Overall, BHV-1 was found to be naturally circulating among the cattle population in Meru
re
County. There was a positive association between BHV-1 infection and cattle age, rearing of goats in the farm together with cattle, and younger women principal farmers. Given that there is
lP
not BHV-1 vaccination use in this study population, training on the importance of biosecurity and vaccination for BHV-1 are recommended to reduce the transmission and impacts of BHV-1.
None.
ur na
Declaration of interest
Acknowledgement
Jo
We would like to acknowledge the developmental partnership between Farmers Helping Farmers, the University of Prince Edward Island and the University of Nairobi for their full support towards the research project. As well, the support of the Naari Dairy Farmers Cooperative Society made it all possible. References
15
Ackermann M., Engels M., 2006. Pro and contra IBR-eradication. Veterinary Microbiology, 113 (3-4), 293-302. https://doi.org/10.1016/j.vetmic.2005.11.043. Biswas S., Bandyopadhyay S., Dimri U., Patra H. P., 2013. Bovine herpesvirus-1 (BHV-1) – a reemerging concern in livestock: a revisit to its biology, epidemiology, diagnosis, and prophylaxis.
Veterinary
Quarterly,
33
(2),
68-81.
https://doi.org/10.1080/01652176.2013.799301. Boelaert F., Speybroeck N., De Kruif A., Aerts M., Burzykowski T., Molenberghs G., Berkvens D. L., 2005. Risk factors for bovine herpesvirus-1 seropositivity. Preventive Veterinary
ro of
Medicine, 69, 285–295. https://doi.org/10.1016/j.prevetmed.2005.02.010. Bowland S. L., Shewen P. E., 2000. Bovine respiratory disease: commercial vaccines currently available
in
Canada.
Canadian
Veterinary
Journal,
41,
33–48.
-p
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1476343/.
Callaby R., Toye P. G., Jennings A., Thumbi S. M., Coveter J. A. W, Van Wyk I. C., Hanotte O.,
re
Mbole-Kariuki M. N., Bronsvoort M. D. C., Kruuk L. E. B., Woolhouse M. E. J., Kiara
Kenya.
Research
lP
H., 2016. Sero-prevalence of respiratory viral pathogens of indigenous calves in Western in
Veterinary
Science,
108,
120-124.
http://dx.doi.org/10.1016/j.rvsc.2016.08.010./.
ur na
Carbonero A., Sa a L. R., Jara D. V., García-Bocanegra I., Arenas A., Borge C., Perea A., 2011. Sero-prevalence and risk factors associated to Bovine Herpesvirus 1 (BHV-1) infection in non-vaccinated dairy and dual purpose cattle herds in Ecuador. Preventive Veterinary
Jo
Medicine, 100 (1), 84-88. https://doi.org/10.1016/j.prevetmed.2011.03.006. Constable P. D., Hinchcliff K. W., Done S. H., Grünberg W., 2017. Infectious Bovine Rhinotracheitis. Veterinary Medicine: A Textbook of the Diseases of Cattle, Horses, Sheep, Pigs, and Goats, 11th edition. Elsevier, St. Louis, Missouri pp. 952–961. Dohoo I. R., Martin W., Stryhn H., 2009. Veterinary Epidemiologic Research. 2nd ed. AVC Inc., Charlottetown, PEI, Canada. 16
Friesen M. C., Davies H. W., Teschke K., Ostry A. S., Hertzman C. and Demers P. A., 2007. Impact of the specificity of the exposure metric on exposure-response relationships. Epidemiology, 18, 88-94. https://doi.org/10.1097/01.ede.0000249558.18960.6b. Fulton R. W., Purdy C. W., Confer A. W., Saliki J. T., Loan R. W., Briggs R. E., Burge L. J., 2000. Bovine viral diarrhea viral infections in feeder calves with respiratory disease: interactions with Pasteurella spp., parainfluenza-3 virus, and bovine respiratory syncytial virus.
Canadian
Journal
of
Veterinary
Research,
64,
151–159.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1189606/.
disease
in
France.
Preventive
Veterinary
https://doi.org/10.1016/j.prevetmed.2009.02.013
ro of
Gay, E., Barnouin, J., 2009. A nation-wide epidemiological study of acute bovine respiratory Medicine,
89,
265–271.
-p
Ghirotti M., Semproni G., Meneghi D., Mungaba F. N., Nannini D., Calzetta G., Paganico G., 1991. Sero-prevalence’s of selected cattle diseases in the Kafue flats of Zambia. Research
Communication,
15,
25–36.
re
Veterinary
https://link.springer.com/article/10.1007/BF00497787.
lP
Graham D. A., 2013. Bovine herpes virus – 1 (BHV – 1) in cattle – a review with emphasis on reproductive impacts and emergence of infection in Ireland and United Kingdom. Irish
ur na
Veterinary Journal, 66, 15. https://doi.org/10.1186/2046-0481-66-15. Handel I. G., Willoughby K., Land F., Koterwas B., Morgan K. L., Tanya V. N., Barend M., 2011. Seroepidemiology of bovine viral diarrhea virus (BVDV) in the Adamawa region of Cameroon and use of the SPOT test to identify herds with PI calves. PloS one, 6(7),
Jo
p.e21620. https://doi.org/10.1371/journal.pone.0021620.
Kahrs R. F., 2001. Infectious bovine rhinotracheitis and infections pustular vulvovaginitis. In: Viral Diseases of Cattle. 2nd ed. Ames, IA: Iowa State University Press, 159–170. https://doi.org/10.1080/01652176.2013.799301 Kenyanjui M., Steiger Y., Thorpe W., 2007. Virus neutralizing Anitbodies to bovine herpes virus type 1 (BHV-1), bovine viral diarrhoea (BVD) and rinderpest (RPV) viruses in 17
smallholder east African zebu cattle in coastal Kenya. Kenya Veterinarians, 18, 21–24. https://www.ajol.info/index.php/kenvet/article/view/39494 Keuser V., Schynts F., Detry B., Collard A., Robert B., Vanderplasschen A., Pastoret P., Thirty E., 2004. Improved antigenic methods for differential of bovine diagnosis of bovine, caprine, and corvine Alpha-herpesviruses related to bovine herpesvirus-1. Clinical Microbiology, 42, 1228–1235. https://doi.org/10.1128/JCM.42.3.1228-1235.2004. Mahmoud M. A., Mahmoud N. A., Allam A. M., 2009. Investigation on infectious bovine rhinotracheitis in Egyptian cattle and buffaloes. Global Veterinaria, 3, 335–340.
ro of
Mainar-Jaime R. C., Berzal-Herranz B., Arias P., Rojo-Vázquez F. A., 2001. Epidemiological pattern and risk factors associated with bovine viral-diarrhoea virus (BVDV) infection in a non-vaccinated dairy-cattle population from the Asturias region of Spain. Preventive
-p
Veterinary Medicine, 52 (1), 63-73. https://doi.org/10.1016/S0167-5877(01)00239-2. Majumder S., Ramakrishnan M. A., Nandi S., 2015. Infectious Bovine Rhinotracheitis: An Indian
re
Perspective. Current Microbiology and Applied Sciences, 4 (10), 844-858.
lP
Mars M. H., de Jong M., Franken P., van Oirschot J. T., 2001. Efficacy of a live glycoprotein Enegative bovine herpesvirus 1 vaccine in cattle in the field. Vaccine, 19, 1924–1930.
ur na
https://doi.org/10.1016/S0264-410X(00)00435-7. Mars M. H., De Jong M. C., Van-Maanen C., Hage J. J., Van Oirschot J. T., 2000. Airborne transmission of bovine herpesvirus 1 infections in calves under field conditions. Veterinary Microbiology, 76, 1–13. https://doi.org/10.1016/S0378-1135(00)00218-2.
Jo
Mars M. H., Bruschke C. J., van Oirschot J. T., 1999. Airborne transmission of BHV1, BRSV, and BVDV among cattle is possible under experimental conditions. Veterinary Microbiology, 66 (3), 197-207. https://doi.org/10.1016/S0378-1135(99)00009-7.
McDermott J. J., Kadohira M., O'Callaghan C. J., Shoukri M. M. 1997. A comparison of different models for assessing variations in the sero-prevalence of infectious bovine rhinotracheitis
18
by farm, area and district in Kenya. Preventive Veterinary Medicine, 32 (3-4): 219-234. https://doi.org/10.1016/S0167-5877(97)00025-1. Muylkens B., Thiry J., Kirten P., Schynts F., Thiry E., 2007. Bovine herpesvirus 1 infection and infectious bovine rhinotracheitis. Veterinary Record, 38, 181-209. https://doi.org. 10.1051/vetres:2006059.2007.181.209. https://doi.org/10.1051/vetres:2006059. Newcomer B. W., Givens D., 2016. Diagnosis and control of viral diseases of reproductive importance: Infectious Bovine Rhinotracheitis and Bovine Viral Diarrhea. Veterinary Clinics
of
North
America
-
Food
Animal
34,
425-441.
ro of
https://doi.org/10.1016/j.cvfa.2016.01.011.
Practice,
Radostits O. M., Gay C. C., Blood D. C., Hinchcliff K. W., 2000. Infectious bovine Rhinotracheitis (IBR, red nose), Bovine herpesvirus -1 (BHV-1) infection, 9th ed., W.B.
-p
Saunders, London. UK, 1173-1184.
Saravanajayam M., Kumanan K., Balasubramaniam A., 2015. Seroepidemiology of infectious
re
bovine rhinotracheitis infection in unvaccinated cattle. Veterinary World, 8 (12), 1416-
lP
1419. https://doi.org/10.14202/vetworld.2015.1416-1419. Segura-Correa J. C., Zapata-Campos C. C., Jasso-Obregón J. O., Martinez-Burnes J., López-
ur na
Zavala R., 2016. Sero-prevalence and risk factors associated with bovine herpesvirus 1 and bovine viral diarrhea virus in North-Eastern Mexico. Open Veterinary Journal, 6 (2), 143-149. http://dx.doi.org/10.4314/ovj.v6i2.12. Seyfi Abad Shapouri M., Ghiami R. M., Haji Hajikolaei M. R., Mahmoodi P., Karami A.,
Jo
Daghari M., 2016. Isolation of Bovine Herpesvirus-1 (BoHV-1) From Latently Infected/Carrier Cattle in Ahvaz. Iranian Journal of Ruminants Health Research, 1 (1), 1120.
Solis-Calderon J. J., Segura-Correa V. M., Segura-Correa J. C., Alvarado-Islas A., 2003. Seroprevalence of and risk factors for infectious bovine rhinotracheitis in beef cattle herds of
19
Yucatan,
Mexico.
Preventive
Veterinary
Medicine,
57,
199–208.
https://doi.org/10.1016/S0167-5877(02)00230-1. Snowder, G.D., Van Vleck, L.D., Cundiff, L.V., Bennett, G.L., 2006. Bovine respiratory disease in feedlot cattle: environmental, genetic, and economic factors. Journal of Animal Science, 84, 1999–2008. https://doi.org/10.2527/jas.2006-046 Straub O. C., 1990. Infectious bovine rhinotracheitis virus. In: Dinter, Z. and Morein, B., editors. Virus infections of ruminants. Oxford: Elsevier Science Publishers BV, 71–108. Takiuchi E., Medici K., Alfieri A., Alfieri A., 2005. Bovine herpesvirus type 1 abortions detected
ro of
by a semi-nested PCR in Brazilian cattle herds. Research Veterinary Sciences 79, 85 88. https://doi.org/10.1016/j.rvsc.2004.11.005.
van Schaik G., 2001. Risk and economics of disease introduction to dairy farms. Tijdschr.
-p
Diergeneeskd, 126, 414–418 (in Dutch). https://www.ncbi.nlm.nih.gov/pubmed/11436606
re
van Schaik G., Schukken Y. H., Nielen M., Dijkhuizen A. A., Barkema H. W., Benedictus G., 2002. Probability of and risk factors for introduction of infectious diseases into Dutch SPF
lP
dairy farms: a cohort study. Preventive Veterinary Medicine, 54, 279–289. https://doi.org/10.1016/S0167-5877(02)00004-1.
ur na
van Schaik, G., Dijkhuizen, A.A., Huirne, R.B.M., Schukken, Y.H., Nielen, M., Hage, H.J., 1998. Risk factors for existence of bovine herpes virus 1 antibodies on nonvaccinating Dutch dairy farms. Preventive Veterinary Medicine, 34, 125–136. https://doi.org/10.1016/S0167-
Jo
5877(97)00085-8
Vonk-Noordegraaf A., Labrovic A., Frankena K., Pfeiffer D. U., Nielen M., 2004. Simulated hazards of losing infection-free status in a Dutch BHV1 model. Preventive Veterinary Medicine, 62, 51–58. https://doi.org/10.1016/j.prevetmed.2003.09.001. Yesilbag K., Bilge-dagalap S., Okur-gumusova S., Gunged B., 2003. Studies on herpesvirus infections of goats in Turkey: Prevalence of antibodies to bovine herpesvirus 1. Revue de 20
Médecine
Vétérinaire,
154,
772–774.
Jo
ur na
lP
re
-p
ro of
https://www.revmedvet.com/2003/RMV154_772_774.pdf.
21
ro of
Jo
ur na
lP
re
-p
Figure 1: A map showing Naari Sub-location (middle of county) in Meru County of Kenya.
22
Table 1: Description of categorical variables for animal- and farm-level factors for 403 cattle on 149 smallholder dairy farms in Meru County, Kenya in 2017, along with bovine herpesvirus-1 infection percentages by category Variable Category Category BHV-1 # Frequency (Percent) + (Percent) by Category Animal level factors Breed
Ayrshire Friesian Guernsey Zebu Cow Heifer Parity 0 Parity1 Parity >1 No history of abortion History of abortion Positive Negative Positive Negative
Age category Parity category
History of an abortion
Showed BVDV signsb Herd level factors Type of feeding system
Jo
ur na
lP
re
-p
Zero– grazing Semi-zero-grazing Grazing Yes Rearing goats on the farm No Yes Rearing sheep on the farm No Yes Use of natural mating No Yes Fence-line contact with other cattle No Yes Grazing on community pasture No Yes Borrowed cows from other farms No Yes Lent cows to other farms No Yes Cattle bought into the farm No Dichotomized age of the female principal ≤ 45 years farmers > 45 years Dichotomized age of the male principal farmers ≤ 45 years > 45 years a Bovine Viral Diarrhea Disease Virus antibodies-positive cows b
14 (20.0) 31 (16.3) 20 (17.9) 5 (16.1) 60 (18.7) 10 (12.2) 12 (15.0) 11 (12.8) 47 (19.8) 55 (17.1) 15 (18.5) 32 (23.4) 20 (12.7) 41 (16.1) 29 (19.6)
63 (42.2) 53 (35.6) 33 (22.2) 23 (15.4) 126 (84.6) 57 (38.3) 92 (61.7) 63 (42.3) 86 (57.7) 120 (80.5) 29 (19.5) 98 (65.8) 51 (34.2) 10 (6.7) 139 (93.3) 12 (8.1) 137 (91.9) 13 (8.7) 136 (91.3) 74 (49.7) 75 (50.3) 76 (51.0) 73 (49.0)
11 (17.5) 10 (18.9) 4 (12.1) 9 (39.1) 17 (13.5) 11 (19.3) 15 (16.3) 12 (19.0) 14 (16.3) 23 (19.2) 3 (10.3) 19 (19.4) 7 (13.7) 5 (50.0) 21 (15.1) 4 (33.3) 22 (16.1) 3 (23.1) 23 (16.9) 14 (23.9) 4 (5.3) 12 (15.8) 6 (8.2)
ro of
BVDV ab positivea
70 (17.4) 190 (47.2) 112 (27.8) 31 (7.7) 321 (79.7) 82 (20.3) 80 (19.9) 86 (21.3) 237 (58.8) 322 (79.9) 32 (9.0) 137 (46.6) 157 (53.4) 254 (63.2) 148 (36.8)
Bovine Viral Diarrhea disease
23
Table 2: Univariable logistic mixed models of the outcome variable bovine herpesvirus-1 antibody seropositivity, while accounting for clustering of 403 cows among 149 smallholder dairy farms in Meru County, Kenya in 2017, for variables of interest with P–Value ≤ 0.3 Categories
OR
95% CIOR
Parity 0 Parity 1 Parity 2-8 Heifer Cows n/a No Yes Negative Positive
baseline 1.059 1.414 baseline 2.181 1.112 baseline 1.446 baseline 2.262
0.206! 0.361 – 3.103 0.917 0.803 – 4.428 0.145
No Yes No Yes No Yes No Yes No Yes No Yes No Yes < 45 years
baseline 2.850 baseline 2.150 baseline 4.893 baseline 2.486 baseline 3.024 baseline 1.076 baseline 1.706 baseline
> 45 years < 45 years > 45 years ! Overall P-values for categorical variables with >2 categories,
0.230 baseline 0.394
Animal level variables Parity category
Age category Age of the dairy cattle Dry cow status BVDV ab status positivea Herd level variables Fence-line contact with other cattle Grazing on community pasture Borrowed cows from other farms Lent cows to other farms
-p
Rearing goats on the farm
Use of natural mating
lP
Dichotomized age of the female principal farmer
re
Rearing sheep on the farm
Dichotomized age of the male principal farmer
0.913 – 5.211 0.079 1.012 – 1.222 0.027 0.685 – 3.049 0.249 1.129 – 4.533 0.021 0.924 – 8.790 0.068 0.918 – 5.033 0.078 1.328–16.031 0.017 0.697 – 8.859 0.014 1.694 – 5.396 0.001 0.969 – 1.192 0.173 0.769 – 3.786 0.150 0.108 – 0.498 0.001 0.181 – 0.898 0.019
Bovine Viral Diarrhea Disease Virus antibody positive cows
ur na
a
P–value
ro of
Variable
OR: Odds Ratio
95% CIOR: 95% Confidence Interval of OR
Jo
Table 3: Final multivariable logistic mixed model for variables associated with bovine herpesvirus-1 infection antibody sero-positivity for 403 cows on 149 smallholder dairy farms in Meru County, Kenya in 2017 Variable OR 95% CIOR P–value 26.77 5.328 – 134.5 <0.001 Rearing goats on the farm 1.200 1.079 – 1.335 0.001 Age of the dairy cattle 0.744 0.593 – 0.933 0.010 Rearing goats on the farm * Age of the dairy cattle 0.182 0.087 – 0.382 0.001 Dichotomized age of female principal farmer OR: Odd Ratio 95% CIOR: 95% Confidence Interval of OR 24
ro of
Jo
ur na
lP
re
-p
Figure 2: Area-under-the-curve graph showing goodness-of-fit of the final model in a study on risk factors for infection with Bovine Herpesvirus-1 for 403 cows on 149 smallholder dairy farms in Meru County, Kenya in 2017.
25
Jo
ur na
lP
re
-p
ro of
Figure 3: Interaction plot of the marginal predicted probability and 95% confidence interval bars of BHV test positivity for age of dairy cattle by goats on (circle) and off (square) the farms, based on the final model on 403 cows on 149 smallholder dairy farms in Meru County, Kenya in 2017.
26
PII:
S0167-5877(19)30456-8
DOI:
https://doi.org/10.1016/j.prevetmed.2019.104863
Reference:
PREVET 104863
To appear in:
Preventive Veterinary Medicine
Received Date:
7 July 2019
Revised Date:
16 November 2019
Accepted Date:
29 November 2019
Please cite this article as: Kipyego ES, Sero-prevalence and risk factors of infectious bovine rhinotracheitis virus (type 1) in Meru County, Kenya, Preventive Veterinary Medicine (2019), doi: https://doi.org/10.1016/j.prevetmed.2019.104863
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Published by Elsevier.
Sero-prevalence and risk factors of infectious bovine rhinotracheitis virus (type 1) in Meru County, Kenya
Essau Serem Kipyego
University of Nairobi
ro of
[email protected]
Abstract
The aim of the study was to determine the antibody sero-prevalence of Bovine Herpesvirus-1
-p
which cause Infectious Bovine Rhinotracheitis (IBR) and to identify risk factors associated with
re
BHV-1 antibody seropositivity among smallholder dairy farms in Meru County, Kenya. A cross-sectional study was conducted in the Naari area of Meru County, Kenya between
lP
September-October 2016 and March-April 2017. The 149 farmers were randomly selected from members of the Naari Dairy Farmers Cooperative Society who were actively delivering milk to
ur na
the society at the time of the study. Serum samples were obtained from 403 female dairy cattle. Farm level management and animal factors were collected through direct interviews with the owner or someone who was knowledgeable about the animals. All serum samples were processed
Jo
with an indirect enzyme-linked immunosorbent assay (gB ELISA) to determine the presence of antibodies to BHV-1. The overall farm-level and animal-level sero-prevalences of BHV-1 antibodies were 30.9% (95% CI: 23.6% to 39.0%) and 17.4% (95% CI: 13.8% to 21.4%), respectively. In the final multivariable analysis, the factors significantly associated with BHV-1 antibodies included; age of the dairy cattle (OR=1.200, p=0.001), age of the principal female farmers (OR=0.182,
1
p=0.001) and rearing goats in the farm (OR=26.77, p=0.000). There was a significant interaction between rearing goats on the farm and age of the dairy cattle (p < 0.010); younger cattle seemed to have been exposed to BHV or a cross-reacting caprine herpesvirus when goats were on the farm. The results showed that BHV-1 was circulating among the cattle population in the Naari area of Meru County. Given that there is not BHV-1 vaccination use in this study population, training on the importance of biosecurity and vaccination for BHV-1 are recommended to reduce the transmission and impacts of BHV-1.
ro of
Keywords: Dairy Cattle, Infectious Bovine Rhinotracheitis (IBR), Bovine Herpesvirus type 1
Jo
ur na
lP
re
-p
(BHV-1), Sero-Prevalence, Risk factors.
2
Introduction Infectious Bovine Rhinotracheitis (IBR) is a multi-organ disease of significant economic importance worldwide caused by Bovine Herpes Virus-1 (BHV-1), and it affects both domestic and wild ruminants (Bowland et al., 2000; Muylkens et al., 2007). Bovine Herpes Virus-1 is a virus of the genus Varicellovirus, subfamily Alphaherpesvirinae and family Herpesviridae, and is a highly contagious and infectious virus (King et al., 2012; Biswas et al., 2013; Newcomer and Givens, 2016). Infectious bovine rhinotracheitis in bovine animals is caused by BHV-1.1, the respiratory subtypes, while strain BHV-1.2a and BHV-1.2b are the genital subtypes, and BHV-
ro of
1.3 is the encephalitic subtype (Muylkens et al., 2007). Contact with Infected cattle is the main source of BHV-1 infection to susceptible animals and herds. These cattle shed the virus through various body secretions and excretions (Takiuchi et al.,
-p
2005; Constable et al., 2017). For the respiratory subtype, the virus is transmitted through direct transmission of nasal discharges between animals or by indirect transmission of aerosol droplets,
re
and the infectivity duration of these aerosols is dependent on environmental factors such as
lP
humidity and temperature. Direct contact with non-nasal mucosal discharges, contaminated semen, fetal tissues, and genital fluid can also lead to transmission (Mars et al., 1999; Mars et al.,
ur na
2000; Kahrs, 2001).
Cattle of all breeds and ages are equally susceptible, and the disease is common in cattle above 6 months of age due to waning of maternal antibody protection and increased mixing of cattle populations (Majumder et al., 2015; Seyfi Abad Shapouri et al., 2016; Constable et al., 2017).
Jo
The IBR diseases have no inherent seasonal variability, although in temperate countries, during the months of fall and winter, disease occurrence is high due to feedlot cattle assembling (Majumder et al., 2015). The managerial and environmental risk factors contributing to the spread of BHV–1 include; purchasing of infected cattle, participation in agricultural shows, increased herd size and type of production system (Gay and Barnouin, 2009; Constable et al., 2017). Uncontrolled movement of visitors and cattle into the farm, and unreliable records of vaccination 3
dates have also been associated with the spread of the disease (Boelaert et al., 2005; GonzálezGarcia et al., 2009). The prevalence and risk factors of BHV-1 infection in Kenya have not been recently researched in predominantly zero-grazing farming regions of Kenya. In this cross-sectional study, a random sample of dairy cattle was examined to determine the BHV-1 antibody sero-prevalence, and to identify risk factors associated with BHV-1 antibody seropositivity on smallholder dairy farms using primarily zero-grazing in Naari area, Meru County, Kenya.
ro of
Materials and Methods Ethical approval
The study was approved by the Biosafety, Animal use and Ethics Committee, Faculty of
-p
Veterinary Medicine, University of Nairobi (FVM/BAUEC/2018/146).
re
Selection of study area and farms
The study area of Naari in Meru County, Kenya (Figure 1), was purposively selected since this
lP
research formed part of a larger study involving smallholder dairy farmers. A non-governmental organization called Farmers Helping Farmers, the University of Prince Edward Island (Canada)
ur na
and the University of Nairobi had an existing developmental partnership with the Naari Dairy Farmers Cooperative Society, and this formed the basis for the entry point to the community. Meru lies at latitude and longitude 0°6'0" N and 37°34'60" E, respectively, and is located in the Mount Kenya region of Kenya, approximately 270 km North of Nairobi, the capital city of
Jo
Kenya. The Naari area is situated at an altitude of approximately 2000 m above sea level. The climate in Meru is warm and temperate with volcanic ash soils, and therefore is considered to have high agricultural potential. The average annual temperature in the Meru area ranges from 14oC to 17oC in the highlands (where Naari is), while in the lowlands it ranges from 22oC to 27oC. Precipitation in the high altitude Meru areas average 2200 mm, while low altitude areas
4
average 500 mm. The main agricultural activities in Naari include; lumbering, horticulture, crop production and dairy production. [Insert Figure 1 near here] The sampling frame for the study consisted of 568 farmers who were active members of the Naari Dairy Farmers Cooperative Society (NDFCS) and delivering milk to the cooperative. The sample size of dairy cattle sampled in the study was estimated based on a prevalence of BHV-1 infection of 50% (for a maximum variance), a precision of 5% and confidence level of 95%, giving a
ro of
required sample size of 385 (Dohoo et al., 2009). The average number of cattle above 6 months of age per farm had been established to be approximately 3, thus 149 farms were randomly selected from the 568 smallholder dairy farms from the registry of active members using software-based random number generation. Therefore,
re
-p
the study population of farms represented 26% of the farmers who were members of the NDFCS.
Data and sample collection
lP
The selected farms were visited between September-October 2016 and March-April 2017, and a questionnaire was administered to capture farm- and animal-level factors of BHV-1. The data
ur na
collected included collection of animal-level information about milking cows and heifers on the farm. A systematic scrutiny of written farm records was also conducted to obtain the age and calving history of the cattle, along with history of respiratory and reproductive diseases, periparturient conditions, and mastitis cases. Other farm-level management information collected
Jo
included: feed and mineral supplementation, vaccination status (including use of IBR vaccine), cattle owner attendances to any dairy husbandry training, herd size, awareness and monitoring of heat signs, and source of animals with respect to the last 12 months. In addition, 5 ml of whole blood was collected in plain redtop vacutainer tubes via the tail vein of each dairy animal greater than 6 months of age, and stored in an ice box during the day. Each evening, the blood tubes were taken from the ice box and allowed to clot, and thereafter, the 5
serum was transferred to Eppendorf® vials. The serum was then frozen at -200C and transported on ice to the Hematology and Biochemistry Laboratory, Department of Clinical Studies, Faculty of Veterinary Medicine, University of Nairobi, and then stored at -200C until all samples were collected for testing.
Laboratory analysis The frozen sera were thawed to room temperature (18-260C). The IBR test kit used was the IDEXX IBR gB X3 Ab Test (IDEXX, Liebefeld-Bern, Switzerland), comprising of a BHV-1 Antigen Coated Plate and all the reagents, which was also warmed from its storage temperature of
ro of
40C to room temperature. The IDEXX IBR gB X3 Ab Test is an Indirect Enzyme-Linked Immunosorbent Assay (ELISA) which has been developed to detect presence of antibodies against IBR (type 1) in individual bovine plasma, milk and serum samples. Antibody responses
-p
induced by vaccines which contain the glycoprotein B (gB) of BHV-1 are detected as well. This
re
kit has a reported sensitivity of 100% and specificity of 95%. The testing procedure was done following the protocol described by the manufacturer in Liebefeld-Bern, Switzerland.
lP
The ELISA results were read from a microplate photometer, Mindray Microplate Reader (MR96A, Shenzhen Mindray Bio-Medical Electronics Company Limited), where optical density (OD)
ur na
was measured either at a single wavelength of 450 nm [A (450)] or dual wavelength of 450 nm and 650 nm [A (450/650) for enhanced sensitivity]. The blocking percentage was calculated by using the absorbance obtained with the test sample and subtracting the absorbance of the negative
Jo
control containing no BHV-1-specific antibodies and dividing the result with the negative control containing no BHV-1-specific antibodies. Interpretation of the results was determined via comparing the sample blocking percentages to the expected percentages in accordance with the manufacturer’s test instructions: blocking % < 45 = negative; 45 ≤ blocking % < 55 = suspect; and blocking % ≥ 55 = positive. The suspects were considered as positive in order to obtain a dichotomous outcome. A positive control was used on each plate to confirm that the procedures were followed properly. 6
Data entry and statistical analysis Data collected through the questionnaires and the laboratory results were first entered into MS Excel (Microsoft Inc., Sacramento, California, USA) and then imported to Stata 15 (StataCorp LLC, College station, Texas, USA) for analyses. Initially, the data were checked for accuracy, coded and analyzed using descriptive statistics. Proportions were determined for categorical variables and presented as a percentage of the overall number, along with a 95% confidence interval, where applicable.
ro of
Mixed-effect logistic regression analysis was performed, accounting for clustering of cattle among herds, to determine associations between the categorical and continuous variables and the dichotomized seropositivity outcome (presence or absence of BHV-1 antibody). In the first step, univariable multi-level mixed models for all the predictor variables were fitted into separate
-p
logistic regression models, employing the functional logit. In the second step, a multivariable
re
mixed logistic regression analysis was fitted for all the univariable associations with p≤0.30 in the first step. Correlations between predictor variables were identified using pair-wise correlation,
lP
and where two or more variables were highly correlated (correlation coefficient >0.5), statistical significance and biological plausibility were used to identify which variable would be offered to
ur na
the modeling process. The final models were built using backward stepwise elimination, leaving those variables which had a p-value ≤0.05. Explanatory variables were considered confounders if their removal from the multivariable model modified the coefficients of other significant variables by 30%. Plausible biological interactions between significant explanatory variables in final model
Jo
were also tested and the significant interaction terms were included in the final models (Dohoo et al, 2009). The area under the curve (AUC) of the receiver operating characteristic was used to evaluate the overall model performance.
Results Farm and animal demographics 7
A total of 403 cattle and 149 farms were involved in the study. The principal farmer was mostly comprised of men (56.4%), while women as principal farmers in this sample population were fewer (28.2%), and 15.4% of farms had both the male and female considered jointly as the principal farmer. Most of the principal farmers were married (83.9%), some had lost their spouse, and a few of them were young people who were still single but had established themselves as dairy farmers (5.3%). Among the principal farmers, approximately half (50.3%) were below 45 years. A large majority of the female farmers had completed primary school (91.2%) but only 39.4% had completed secondary school, while the proportions of male farmers having completed
ro of
primary and secondary school were slightly different at 81.5% and 42.3%, respectively. The majority (81.2%) of principal farmers had training on dairy production.
The mean household size recorded in this study was 3.7 ± 1.54, with a minimum of 1 person and
-p
a maximum of 11. The mean total land holdings owned by the respondents was 2.11 ± 2.04 acres. In addition, some of the farmers also had access to other pieces of land (0.51±0.84 acres) through
re
renting, borrowing and government-owned lands leased to them.
lP
Holstein-Friesians formed nearly half of the dairy cattle reared in the region, with small proportions of other exotic and indigenous breeds (Table 1). The mean herd size among the
ur na
sampled farms was 5.7 with a range of 1 to 16 cattle. The mean number of milking cows was 1.8 with a range of 0 and 8. The mean number of dry cows was 0.3 with a range of 0 to 3. Mature cows comprised the majority of animals at 79.7% (321/403), while unbred heifers comprised 12.2% (49/403). The mean age of the sampled animals was 5.5 years with a range of between 0.5
Jo
and 17.0 years. The parities of the cows ranged between 0 and 8, with multiparous cows forming nearly 60% of sampled cattle. It was reported that 9.0% (32/354) of the cows and bred heifers had experienced an abortion in the last one year. Farm management practices at the farms The managerial practices found among the 149 smallholder dairy farms recruited in the study are summarized in Table 1. The zero-grazing and semi-zero-grazing systems formed the majority 8
(nearly 80%) of smallholder dairy farms. Other than dairy production, many of the recruited farmers also practiced sheep and goat production systems. The majority of smallholder dairy farms used artificial insemination exclusively at 57.7% (86/149). More than half of the farmers grazed their cows in community pastures where they could contact other cattle, while less than a quarter of farms disallowed fence-line contact with other cattle. However, there was insignificant movement of the cattle between farms across the region, with less than 10% of farms borrowing cows, lending cows, and purchasing cows. None of the study farmers used a BHV-1 vaccine in the past.
ro of
Sero-prevalence of Bovine Herpesvirus-1 Infection The overall apparent animal-level sero-prevalence to BHV-1 was 17.4% (70/403), with a 95% CI of 13.8% to 21.4%. The overall estimated true animal-level sero-prevalence to BHV-1, adjusting
-p
for the imperfect specificity of the test, was calculated to be 13.0%, with a 95% CI of 9.5% to
re
17.2%. The herd-level sero-prevalence to BHV-1 was 30.9% (46/149), with a 95% CI of 23.6% to 39.0%. The sero-prevalences of the antibodies to the IBR virus in relation to a number of cow
lP
and farm variables are summarized in Table 1.
The sero-prevalence for BHV-1 infection was highest among cattle that were older and
ur na
multiparous, with slightly elevated sero-prevalences among Ayrshire crosses and animals with a history of abortion. At the farm level, farms with goats had a 3 times higher sero-prevalence for BHV-1 among the cattle versus farms without goats, while farms with sheep only had a slightly
Jo
higher BHV-1 sero-prevalence in cattle. The cattle sero-prevalence for BHV-1 was also substantially higher on farms that lent and/or borrowed cattle to/from other farms, and on farms where fence-line contact with neighbour’s cattle was allowed, [Insert Table 1 near here] Factors associated with Bovine Herpesvirus-1 sero-positivity
9
The following animal-level variables met the P–value cut-off criterion (P-value < 0.3) for univariable regression analysis with BHV-1 antibody status: parity category, age category, dry cow status, and BVDV antibody status. Farm-level variables meeting the cut-off criterion included: the farm allowed fence-line contact with other cattle, grazed on community pasture, lent cows to other farms, borrowed cows from other farms, reared goats on the farm, reared sheep on the farm, used natural breeding, age of the male principal farmer, and age of the female principal farmer (Table 2). Pair-wise correlation was carried out and age, parity and age category of the cattle were found to
ro of
be highly correlated (r>0.5). Fence-line contact with other cattle and grazing on community pasture were also found to be highly correlated. Therefore, age of the animal and fence-line contact were offered to the final multivariable model because they had lower P-Values than their
-p
correlates.
re
[Insert Table 2 near here]
The factors associated with sero-positivity to BHV-1 infection in the final multivariable analysis
lP
included (Table 3): dichotomized age of the principal female farmers, age of the dairy cattle reared on the farm, and rearing goats on the farm. In this final model, controlling for confounding
ur na
of variables in the model, the odds of older dairy cattle to have BHV-1 antibody positivity were 1.200 times higher with each additional year of age. Cattle on farms that were rearing goats had 26.77 times the odds to have BHV-1 antibodies compared to cattle on farms not rearing goats. In addition, when the age of the female principal farmer was over 45 years, the odds of cattle on the
Jo
farm to have BHV-1 antibodies were lower by 0.182 times compared with cattle reared by younger women principal farmers under or equal to 45 years (Table 3). The area under the ROC curve for this final model was 0.86, indicating a good overall goodness-of–fit of the observed data (Figure 2). [Insert Table 3 near here]
10
[Insert figure 2 near here] The 95% CI for 1` year and 6 years do not overlap, showing that the interaction is significant (p<0. 0.010). The relationship between cattle age and BHV antibodies depends on whether there are goats or not, with there being a significant positive association when the cattle are young (< 6 years) but no association when the cattle are older than 6 years. Similarly, the relationship between goats on the farm and BHV antibodies depends on the cattle age, with there being a positive association between BHV antibodies and the presence of goats, but only when cattle age is < 6 years (Figure 3).
ro of
[Insert figure 3 near here] Discussion
-p
The results of this study confirmed the presence of IBR infection through sero-prevalence of BHV-1 antibodies among the smallholder dairy cattle reared in primarily zero-grazing units in
re
Naari area of Meru County. The observed sero-prevalence for BHV-1 antibodies from this study was estimated at 17.4%, which was slightly lower compared to studies in other parts of Kenya of
lP
20.9% in the western Kenya (Callaby et al., 2016) and 28.0% in the former Malindi District of the Coastal Region (Kenyanjui et al, 2007). This slightly lower sero-prevalence could have been due
ur na
to the different livestock production systems studied, for example, 42% of our study farms used only zero-grazing, and open grazing is a known risk factor for BHV–1 infection (Snowder et al., 2006; Callaby et al., 2016).
Jo
The observed sero-prevalence to BHV-1 of 17.4% from this study was within the range of 16% 54% that was observed by McDermott et al., (1997) in former Districts across Kenya in 1991– 1992. However, in this study, the suspects were considered as positive in order to obtain a dichotomous outcome, and this interpretation might have led to higher magnitudes and direction of estimates of the sero-prevalence (Friesen et al., 2007 and Dohoo et al., 2009). Conversely, the sero-prevalence observed in Naari area of Meru County was clearly lower than that estimated in
11
traditionally managed herds (i.e. open grazing) in Zambia, which ranged from 42%-76% (Ghirotti et al., 1991), and in Egypt which ranged from 63%-86% (Mahmoud et al., 2009). The observed differences in antibody sero-prevalence in different regions and countries could be explained by factors such as production systems, differences in herd sizes, type of breeding methods, differences in disease-control measures, and age of the cattle (Mainar-Jaime et al., 2001 and Ackermann and Engels, 2006). It has been observed that higher BHV-1 prevalences recorded in larger herd sizes and intensively farmed cattle could be associated with a high level of contact between individual animals within a herd (Snowder et al., 2006). For extensively managed farms
ro of
with median herd size of 5 animals, the risk of contact between a susceptible individual with an infected or persistently infected animal is lower (Callaby et al., 2016). The studies reported above involved animals of various ages, such as 51 weeks old (Callaby et al., 2016), 3 months to adults
-p
(Ghirotti et al., 1991), and zebu adults (Kenyanjui et al., 2007), while our study involved heifer and adult dairy cows. Therefore, the age differences among the animals studied may also have
re
explained some of the variations in the sero-prevalence.
lP
Ghirotti et al., (1991) and Kenyanjui et al., (2007) employed virus neutralization tests (VNT) compared to ELISA tests used in our study, and the differences in test specificity and sensitivity
ur na
may also have also contributed to observed differences in prevalence (Graham et al., 1998). In the studies done by Saravanajayam et al., (2001) it was reported that the ELISA had a higher specificity and sensitivity compared to VNT, thus ELISA is considered a rapid, reliable and technically superior test for detection of BHV-1 antibodies. Callaby et al., (2016) also employed
Jo
an indirect ELISA test, and thus may have contributed a similar prevalence to our study. The results of the final model of this study revealed that a number of risk factors were associated with sero-prevalence of BHV-1 antibody, such as age of the cattle, rearing goats in the farms, and cows that had antibodies against BVDV. Several serological studies carried out worldwide identified similar risk factors associated with BHV–1 sero-positivity (Muylkens et al., 2007; Callaby et al., 2016; Constable et al., 2017). The current study also agrees with other studies 12
which reported a number of risk factors in their studies, such as large herd size, intensive production systems, housing, and management practices, like animal age and sex (males are more frequently positive than females), and direct animal contact through cattle shows and purchasing of cattle (van Schaik, 2001; van Schaik et al., 2002; Solis-Calderon et al., 2003; VonkNoordegraaf et al., 2004; Boelaert et al., 2005). Age of the cattle is a frequent risk factor reported in BHV-1 seropositive cattle, where findings from Solis-Calderon et al., (2003), Carbonero et al., (2011), Romero-Salas et al., (2013), Saravanajayam et al., (2015) and Segura-Correa et al., (2016) showed that older animals had a
ro of
higher sero-prevalence compared to young animals. Since infected cattle remain infected for life, it is not surprising that proportions of infected cattle increase with age.
Cattle on farms with older female principal farmers had a lower odds of infection compared to
-p
cattle on farms with younger female principal farmers. It is unclear why this variable remained significant in our final model, but it may be that older farmers are more knowledgeable of the
re
importance of biosecurity and disease control measures. Further research would help to clarify
lP
this finding.
Based on past epidemiological studies, this study showed that IBR and BVD viruses are closely
ur na
associated in the univariable analyses, and infection for one disease has predisposed to the other disease elsewhere (Callaby et al., 2016). The study also showed that BVDV sero-prevalence was significantly associated with BHV-1 at the farm level. This finding agrees with those reported earlier (Fulton et al., 2000; Callaby et al., 2016). A positive correlation has also been
Jo
demonstrated between infections with the viruses causing IBR and PI (Callaby et al., 2016). The present study revealed that rearing of goats in the farm was associated with BHV-1 seroprevalence in the cattle. Mixed rearing of sheep, goats and cattle was not uncommon in our study population, allowing this variable to become part of the risk factors in the final model. Infectious Bovine Rhinotracheitis has the potential for cross-reaction with four other herpesviruses from other animals including goats and buffaloes (Handel et al., 2011; Callaby et al., 2016). Therefore, 13
this risk factor may be attributed to the potential ability of viral cross-infection with related herpesvirus (Biswas et al., 2013). Further research is needed to clarify this potential crossreaction. The present study showed that introduction of new animals into the herd was significantly associated sero-positivity of BHV-1 in univariable analyses but did not remain significant in the final model. Other past studies by Solis-Calderon et al., (2003) and Segura-Correa et al., (2016) reported that introduction of an animal into the herd was not significantly associated with seroprevalence of BHV-1. However, van Schaik et al., (1998) and Gay and Barnouin, (2009) reported
ro of
increased animal movement into a herd and close proximity neighbouring farms increased the risk of IBR spread through contact between naïve herd and infected animals. While introduction of new animals would biologically make sense as a risk factor for BHV-1, the relatively low sero-
-p
prevalence of BHV-1 in our study population (13.0% estimated true prevalence) and infrequent movement of cattle between herds (6.7%, 8.1%, and 8.7% of farmers borrowed, lent, or bought
re
cattle, respectively – Table 1) would mean that the chances of an infected animal moving between
lP
farms is low.
For the observed significant interaction between cattle age and goats on the farm (p < 0.01),
ur na
younger cattle seemed to have been exposed to BHV or a cross-reacting caprine herpesvirus when goats were on the farm. This could be because farms with cattle < 6 years old just brought in goats to the farm in the last 6 years. It could also be that farms bringing in goats into the farm in the last 6 years is correlated with farms bringing in other cattle in the last 6 years, and so those
Jo
younger cattle are more likely to be exposed to BHV with these new cattle introductions than older cattle.
This study was carried out in smallholder farms with approximate herd size of 3 animals and little variation in herd size, as a result, the study did not have enough variation and power to identify herd size as a risk factor for seropositivity. Other studies suggested that at herd level, herd size is the main important risk factor for BHV-1 infection (Snowder et al., 2006; Segura-Correa et al., 14
2016). Our herd sizes were small, making it less likely for our herds to be aggregates of animals from other farms. A limitation to this study is that it is a cross-sectional design, which has limited ability to confirm temporality of the risk factors for BHV-1 infection. Furthermore, the study used a BHV-1specific antibody ELISA testing kit, thus this test cannot establish whether the test-positive animal is due to the virus retention, or due to adaptive immunity from previous exposure to the infection. In the future, a study is needed to establish if clinical IBR occurs in smallholder dairy farmers in the Meru area of Kenya in order to determine the long-term effects of BHV-1 infection
ro of
in cattle production that would justify formulation of prevention and control measures, such as vaccination. Conclusions and recommendations
-p
Overall, BHV-1 was found to be naturally circulating among the cattle population in Meru
re
County. There was a positive association between BHV-1 infection and cattle age, rearing of goats in the farm together with cattle, and younger women principal farmers. Given that there is
lP
not BHV-1 vaccination use in this study population, training on the importance of biosecurity and vaccination for BHV-1 are recommended to reduce the transmission and impacts of BHV-1.
None.
ur na
Declaration of interest
Acknowledgement
Jo
We would like to acknowledge the developmental partnership between Farmers Helping Farmers, the University of Prince Edward Island and the University of Nairobi for their full support towards the research project. As well, the support of the Naari Dairy Farmers Cooperative Society made it all possible. References
15
Ackermann M., Engels M., 2006. Pro and contra IBR-eradication. Veterinary Microbiology, 113 (3-4), 293-302. https://doi.org/10.1016/j.vetmic.2005.11.043. Biswas S., Bandyopadhyay S., Dimri U., Patra H. P., 2013. Bovine herpesvirus-1 (BHV-1) – a reemerging concern in livestock: a revisit to its biology, epidemiology, diagnosis, and prophylaxis.
Veterinary
Quarterly,
33
(2),
68-81.
https://doi.org/10.1080/01652176.2013.799301. Boelaert F., Speybroeck N., De Kruif A., Aerts M., Burzykowski T., Molenberghs G., Berkvens D. L., 2005. Risk factors for bovine herpesvirus-1 seropositivity. Preventive Veterinary
ro of
Medicine, 69, 285–295. https://doi.org/10.1016/j.prevetmed.2005.02.010. Bowland S. L., Shewen P. E., 2000. Bovine respiratory disease: commercial vaccines currently available
in
Canada.
Canadian
Veterinary
Journal,
41,
33–48.
-p
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1476343/.
Callaby R., Toye P. G., Jennings A., Thumbi S. M., Coveter J. A. W, Van Wyk I. C., Hanotte O.,
re
Mbole-Kariuki M. N., Bronsvoort M. D. C., Kruuk L. E. B., Woolhouse M. E. J., Kiara
Kenya.
Research
lP
H., 2016. Sero-prevalence of respiratory viral pathogens of indigenous calves in Western in
Veterinary
Science,
108,
120-124.
http://dx.doi.org/10.1016/j.rvsc.2016.08.010./.
ur na
Carbonero A., Sa a L. R., Jara D. V., García-Bocanegra I., Arenas A., Borge C., Perea A., 2011. Sero-prevalence and risk factors associated to Bovine Herpesvirus 1 (BHV-1) infection in non-vaccinated dairy and dual purpose cattle herds in Ecuador. Preventive Veterinary
Jo
Medicine, 100 (1), 84-88. https://doi.org/10.1016/j.prevetmed.2011.03.006. Constable P. D., Hinchcliff K. W., Done S. H., Grünberg W., 2017. Infectious Bovine Rhinotracheitis. Veterinary Medicine: A Textbook of the Diseases of Cattle, Horses, Sheep, Pigs, and Goats, 11th edition. Elsevier, St. Louis, Missouri pp. 952–961. Dohoo I. R., Martin W., Stryhn H., 2009. Veterinary Epidemiologic Research. 2nd ed. AVC Inc., Charlottetown, PEI, Canada. 16
Friesen M. C., Davies H. W., Teschke K., Ostry A. S., Hertzman C. and Demers P. A., 2007. Impact of the specificity of the exposure metric on exposure-response relationships. Epidemiology, 18, 88-94. https://doi.org/10.1097/01.ede.0000249558.18960.6b. Fulton R. W., Purdy C. W., Confer A. W., Saliki J. T., Loan R. W., Briggs R. E., Burge L. J., 2000. Bovine viral diarrhea viral infections in feeder calves with respiratory disease: interactions with Pasteurella spp., parainfluenza-3 virus, and bovine respiratory syncytial virus.
Canadian
Journal
of
Veterinary
Research,
64,
151–159.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1189606/.
disease
in
France.
Preventive
Veterinary
https://doi.org/10.1016/j.prevetmed.2009.02.013
ro of
Gay, E., Barnouin, J., 2009. A nation-wide epidemiological study of acute bovine respiratory Medicine,
89,
265–271.
-p
Ghirotti M., Semproni G., Meneghi D., Mungaba F. N., Nannini D., Calzetta G., Paganico G., 1991. Sero-prevalence’s of selected cattle diseases in the Kafue flats of Zambia. Research
Communication,
15,
25–36.
re
Veterinary
https://link.springer.com/article/10.1007/BF00497787.
lP
Graham D. A., 2013. Bovine herpes virus – 1 (BHV – 1) in cattle – a review with emphasis on reproductive impacts and emergence of infection in Ireland and United Kingdom. Irish
ur na
Veterinary Journal, 66, 15. https://doi.org/10.1186/2046-0481-66-15. Handel I. G., Willoughby K., Land F., Koterwas B., Morgan K. L., Tanya V. N., Barend M., 2011. Seroepidemiology of bovine viral diarrhea virus (BVDV) in the Adamawa region of Cameroon and use of the SPOT test to identify herds with PI calves. PloS one, 6(7),
Jo
p.e21620. https://doi.org/10.1371/journal.pone.0021620.
Kahrs R. F., 2001. Infectious bovine rhinotracheitis and infections pustular vulvovaginitis. In: Viral Diseases of Cattle. 2nd ed. Ames, IA: Iowa State University Press, 159–170. https://doi.org/10.1080/01652176.2013.799301 Kenyanjui M., Steiger Y., Thorpe W., 2007. Virus neutralizing Anitbodies to bovine herpes virus type 1 (BHV-1), bovine viral diarrhoea (BVD) and rinderpest (RPV) viruses in 17
smallholder east African zebu cattle in coastal Kenya. Kenya Veterinarians, 18, 21–24. https://www.ajol.info/index.php/kenvet/article/view/39494 Keuser V., Schynts F., Detry B., Collard A., Robert B., Vanderplasschen A., Pastoret P., Thirty E., 2004. Improved antigenic methods for differential of bovine diagnosis of bovine, caprine, and corvine Alpha-herpesviruses related to bovine herpesvirus-1. Clinical Microbiology, 42, 1228–1235. https://doi.org/10.1128/JCM.42.3.1228-1235.2004. Mahmoud M. A., Mahmoud N. A., Allam A. M., 2009. Investigation on infectious bovine rhinotracheitis in Egyptian cattle and buffaloes. Global Veterinaria, 3, 335–340.
ro of
Mainar-Jaime R. C., Berzal-Herranz B., Arias P., Rojo-Vázquez F. A., 2001. Epidemiological pattern and risk factors associated with bovine viral-diarrhoea virus (BVDV) infection in a non-vaccinated dairy-cattle population from the Asturias region of Spain. Preventive
-p
Veterinary Medicine, 52 (1), 63-73. https://doi.org/10.1016/S0167-5877(01)00239-2. Majumder S., Ramakrishnan M. A., Nandi S., 2015. Infectious Bovine Rhinotracheitis: An Indian
re
Perspective. Current Microbiology and Applied Sciences, 4 (10), 844-858.
lP
Mars M. H., de Jong M., Franken P., van Oirschot J. T., 2001. Efficacy of a live glycoprotein Enegative bovine herpesvirus 1 vaccine in cattle in the field. Vaccine, 19, 1924–1930.
ur na
https://doi.org/10.1016/S0264-410X(00)00435-7. Mars M. H., De Jong M. C., Van-Maanen C., Hage J. J., Van Oirschot J. T., 2000. Airborne transmission of bovine herpesvirus 1 infections in calves under field conditions. Veterinary Microbiology, 76, 1–13. https://doi.org/10.1016/S0378-1135(00)00218-2.
Jo
Mars M. H., Bruschke C. J., van Oirschot J. T., 1999. Airborne transmission of BHV1, BRSV, and BVDV among cattle is possible under experimental conditions. Veterinary Microbiology, 66 (3), 197-207. https://doi.org/10.1016/S0378-1135(99)00009-7.
McDermott J. J., Kadohira M., O'Callaghan C. J., Shoukri M. M. 1997. A comparison of different models for assessing variations in the sero-prevalence of infectious bovine rhinotracheitis
18
by farm, area and district in Kenya. Preventive Veterinary Medicine, 32 (3-4): 219-234. https://doi.org/10.1016/S0167-5877(97)00025-1. Muylkens B., Thiry J., Kirten P., Schynts F., Thiry E., 2007. Bovine herpesvirus 1 infection and infectious bovine rhinotracheitis. Veterinary Record, 38, 181-209. https://doi.org. 10.1051/vetres:2006059.2007.181.209. https://doi.org/10.1051/vetres:2006059. Newcomer B. W., Givens D., 2016. Diagnosis and control of viral diseases of reproductive importance: Infectious Bovine Rhinotracheitis and Bovine Viral Diarrhea. Veterinary Clinics
of
North
America
-
Food
Animal
34,
425-441.
ro of
https://doi.org/10.1016/j.cvfa.2016.01.011.
Practice,
Radostits O. M., Gay C. C., Blood D. C., Hinchcliff K. W., 2000. Infectious bovine Rhinotracheitis (IBR, red nose), Bovine herpesvirus -1 (BHV-1) infection, 9th ed., W.B.
-p
Saunders, London. UK, 1173-1184.
Saravanajayam M., Kumanan K., Balasubramaniam A., 2015. Seroepidemiology of infectious
re
bovine rhinotracheitis infection in unvaccinated cattle. Veterinary World, 8 (12), 1416-
lP
1419. https://doi.org/10.14202/vetworld.2015.1416-1419. Segura-Correa J. C., Zapata-Campos C. C., Jasso-Obregón J. O., Martinez-Burnes J., López-
ur na
Zavala R., 2016. Sero-prevalence and risk factors associated with bovine herpesvirus 1 and bovine viral diarrhea virus in North-Eastern Mexico. Open Veterinary Journal, 6 (2), 143-149. http://dx.doi.org/10.4314/ovj.v6i2.12. Seyfi Abad Shapouri M., Ghiami R. M., Haji Hajikolaei M. R., Mahmoodi P., Karami A.,
Jo
Daghari M., 2016. Isolation of Bovine Herpesvirus-1 (BoHV-1) From Latently Infected/Carrier Cattle in Ahvaz. Iranian Journal of Ruminants Health Research, 1 (1), 1120.
Solis-Calderon J. J., Segura-Correa V. M., Segura-Correa J. C., Alvarado-Islas A., 2003. Seroprevalence of and risk factors for infectious bovine rhinotracheitis in beef cattle herds of
19
Yucatan,
Mexico.
Preventive
Veterinary
Medicine,
57,
199–208.
https://doi.org/10.1016/S0167-5877(02)00230-1. Snowder, G.D., Van Vleck, L.D., Cundiff, L.V., Bennett, G.L., 2006. Bovine respiratory disease in feedlot cattle: environmental, genetic, and economic factors. Journal of Animal Science, 84, 1999–2008. https://doi.org/10.2527/jas.2006-046 Straub O. C., 1990. Infectious bovine rhinotracheitis virus. In: Dinter, Z. and Morein, B., editors. Virus infections of ruminants. Oxford: Elsevier Science Publishers BV, 71–108. Takiuchi E., Medici K., Alfieri A., Alfieri A., 2005. Bovine herpesvirus type 1 abortions detected
ro of
by a semi-nested PCR in Brazilian cattle herds. Research Veterinary Sciences 79, 85 88. https://doi.org/10.1016/j.rvsc.2004.11.005.
van Schaik G., 2001. Risk and economics of disease introduction to dairy farms. Tijdschr.
-p
Diergeneeskd, 126, 414–418 (in Dutch). https://www.ncbi.nlm.nih.gov/pubmed/11436606
re
van Schaik G., Schukken Y. H., Nielen M., Dijkhuizen A. A., Barkema H. W., Benedictus G., 2002. Probability of and risk factors for introduction of infectious diseases into Dutch SPF
lP
dairy farms: a cohort study. Preventive Veterinary Medicine, 54, 279–289. https://doi.org/10.1016/S0167-5877(02)00004-1.
ur na
van Schaik, G., Dijkhuizen, A.A., Huirne, R.B.M., Schukken, Y.H., Nielen, M., Hage, H.J., 1998. Risk factors for existence of bovine herpes virus 1 antibodies on nonvaccinating Dutch dairy farms. Preventive Veterinary Medicine, 34, 125–136. https://doi.org/10.1016/S0167-
Jo
5877(97)00085-8
Vonk-Noordegraaf A., Labrovic A., Frankena K., Pfeiffer D. U., Nielen M., 2004. Simulated hazards of losing infection-free status in a Dutch BHV1 model. Preventive Veterinary Medicine, 62, 51–58. https://doi.org/10.1016/j.prevetmed.2003.09.001. Yesilbag K., Bilge-dagalap S., Okur-gumusova S., Gunged B., 2003. Studies on herpesvirus infections of goats in Turkey: Prevalence of antibodies to bovine herpesvirus 1. Revue de 20
Médecine
Vétérinaire,
154,
772–774.
Jo
ur na
lP
re
-p
ro of
https://www.revmedvet.com/2003/RMV154_772_774.pdf.
21
ro of
Jo
ur na
lP
re
-p
Figure 1: A map showing Naari Sub-location (middle of county) in Meru County of Kenya.
22
Table 1: Description of categorical variables for animal- and farm-level factors for 403 cattle on 149 smallholder dairy farms in Meru County, Kenya in 2017, along with bovine herpesvirus-1 infection percentages by category Variable Category Category BHV-1 # Frequency (Percent) + (Percent) by Category Animal level factors Breed
Ayrshire Friesian Guernsey Zebu Cow Heifer Parity 0 Parity1 Parity >1 No history of abortion History of abortion Positive Negative Positive Negative
Age category Parity category
History of an abortion
Showed BVDV signsb Herd level factors Type of feeding system
Jo
ur na
lP
re
-p
Zero– grazing Semi-zero-grazing Grazing Yes Rearing goats on the farm No Yes Rearing sheep on the farm No Yes Use of natural mating No Yes Fence-line contact with other cattle No Yes Grazing on community pasture No Yes Borrowed cows from other farms No Yes Lent cows to other farms No Yes Cattle bought into the farm No Dichotomized age of the female principal ≤ 45 years farmers > 45 years Dichotomized age of the male principal farmers ≤ 45 years > 45 years a Bovine Viral Diarrhea Disease Virus antibodies-positive cows b
14 (20.0) 31 (16.3) 20 (17.9) 5 (16.1) 60 (18.7) 10 (12.2) 12 (15.0) 11 (12.8) 47 (19.8) 55 (17.1) 15 (18.5) 32 (23.4) 20 (12.7) 41 (16.1) 29 (19.6)
63 (42.2) 53 (35.6) 33 (22.2) 23 (15.4) 126 (84.6) 57 (38.3) 92 (61.7) 63 (42.3) 86 (57.7) 120 (80.5) 29 (19.5) 98 (65.8) 51 (34.2) 10 (6.7) 139 (93.3) 12 (8.1) 137 (91.9) 13 (8.7) 136 (91.3) 74 (49.7) 75 (50.3) 76 (51.0) 73 (49.0)
11 (17.5) 10 (18.9) 4 (12.1) 9 (39.1) 17 (13.5) 11 (19.3) 15 (16.3) 12 (19.0) 14 (16.3) 23 (19.2) 3 (10.3) 19 (19.4) 7 (13.7) 5 (50.0) 21 (15.1) 4 (33.3) 22 (16.1) 3 (23.1) 23 (16.9) 14 (23.9) 4 (5.3) 12 (15.8) 6 (8.2)
ro of
BVDV ab positivea
70 (17.4) 190 (47.2) 112 (27.8) 31 (7.7) 321 (79.7) 82 (20.3) 80 (19.9) 86 (21.3) 237 (58.8) 322 (79.9) 32 (9.0) 137 (46.6) 157 (53.4) 254 (63.2) 148 (36.8)
Bovine Viral Diarrhea disease
23
Table 2: Univariable logistic mixed models of the outcome variable bovine herpesvirus-1 antibody seropositivity, while accounting for clustering of 403 cows among 149 smallholder dairy farms in Meru County, Kenya in 2017, for variables of interest with P–Value ≤ 0.3 Categories
OR
95% CIOR
Parity 0 Parity 1 Parity 2-8 Heifer Cows n/a No Yes Negative Positive
baseline 1.059 1.414 baseline 2.181 1.112 baseline 1.446 baseline 2.262
0.206! 0.361 – 3.103 0.917 0.803 – 4.428 0.145
No Yes No Yes No Yes No Yes No Yes No Yes No Yes < 45 years
baseline 2.850 baseline 2.150 baseline 4.893 baseline 2.486 baseline 3.024 baseline 1.076 baseline 1.706 baseline
> 45 years < 45 years > 45 years ! Overall P-values for categorical variables with >2 categories,
0.230 baseline 0.394
Animal level variables Parity category
Age category Age of the dairy cattle Dry cow status BVDV ab status positivea Herd level variables Fence-line contact with other cattle Grazing on community pasture Borrowed cows from other farms Lent cows to other farms
-p
Rearing goats on the farm
Use of natural mating
lP
Dichotomized age of the female principal farmer
re
Rearing sheep on the farm
Dichotomized age of the male principal farmer
0.913 – 5.211 0.079 1.012 – 1.222 0.027 0.685 – 3.049 0.249 1.129 – 4.533 0.021 0.924 – 8.790 0.068 0.918 – 5.033 0.078 1.328–16.031 0.017 0.697 – 8.859 0.014 1.694 – 5.396 0.001 0.969 – 1.192 0.173 0.769 – 3.786 0.150 0.108 – 0.498 0.001 0.181 – 0.898 0.019
Bovine Viral Diarrhea Disease Virus antibody positive cows
ur na
a
P–value
ro of
Variable
OR: Odds Ratio
95% CIOR: 95% Confidence Interval of OR
Jo
Table 3: Final multivariable logistic mixed model for variables associated with bovine herpesvirus-1 infection antibody sero-positivity for 403 cows on 149 smallholder dairy farms in Meru County, Kenya in 2017 Variable OR 95% CIOR P–value 26.77 5.328 – 134.5 <0.001 Rearing goats on the farm 1.200 1.079 – 1.335 0.001 Age of the dairy cattle 0.744 0.593 – 0.933 0.010 Rearing goats on the farm * Age of the dairy cattle 0.182 0.087 – 0.382 0.001 Dichotomized age of female principal farmer OR: Odd Ratio 95% CIOR: 95% Confidence Interval of OR 24
ro of
Jo
ur na
lP
re
-p
Figure 2: Area-under-the-curve graph showing goodness-of-fit of the final model in a study on risk factors for infection with Bovine Herpesvirus-1 for 403 cows on 149 smallholder dairy farms in Meru County, Kenya in 2017.
25
Jo
ur na
lP
re
-p
ro of
Figure 3: Interaction plot of the marginal predicted probability and 95% confidence interval bars of BHV test positivity for age of dairy cattle by goats on (circle) and off (square) the farms, based on the final model on 403 cows on 149 smallholder dairy farms in Meru County, Kenya in 2017.
26