Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Hospital-acquired infection wikipedia , lookup
Henipavirus wikipedia , lookup
Meningococcal disease wikipedia , lookup
Brucellosis wikipedia , lookup
Marburg virus disease wikipedia , lookup
Oesophagostomum wikipedia , lookup
Bovine spongiform encephalopathy wikipedia , lookup
Eradication of infectious diseases wikipedia , lookup
SRBESSENTIALS 05 Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. Kerli Raaperi*, Toomas Orro, Arvo Viltrop Institute of Veterinary Medicine and Animal Sciences, Estonian University of Life Sciences, Kreutzwaldi 62, Tartu 51014, Estonia THE PLACE FOR THE EXPERTS IN CATTLE DISEASES G Model ARTICLE IN PRESS PREVET-3732; No. of Pages 10 Preventive Veterinary Medicine xxx (2015) xxx–xxx Contents lists available at ScienceDirect Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds Kerli Raaperi ∗ , Toomas Orro, Arvo Viltrop Institute of Veterinary Medicine and Animal Sciences, Estonian University of Life Sciences, Kreutzwaldi 62, Tartu 51014, Estonia a r t i c l e i n f o Article history: Received 23 March 2014 Received in revised form 13 January 2015 Accepted 17 January 2015 Keywords: Bovine herpesvirus 1 Vaccination Herd health Reproduction Respiratory disease a b s t r a c t The aim of this study was to estimate the effect of the bovine herpesvirus 1 (BHV-1) vaccination on herd health and production in BHV-1 infected Estonian dairy cattle herds. Seven herds vaccinated with inactivated gE-negative BHV-1 marker vaccines and seven matched non-vaccinated herds were selected. In vaccinated herds the calving interval was on average 7.01 days shorter compared to that in the non-vaccinated herds (coef = −7.01, 95% CI = −11.98, −2.03, p = 0.008) during the study period (2007–2012). In non-vaccinated herds the insemination index had an increasing trend (coef(log scale) = 0.03, 95% CI = −0.003, 0.06, p = 0.054) and the first service conception rate decreased (coef = −2.19, 95% CI = −3.91, −0.47, p = 0.015), whereas no significant changes occurred in vaccinated herds. Average yearly milk yield per cow increased (coef = 145.30, 95% CI = −6.11, 296.71, p = 0.065) and length of the dry period decreased in BHV-1 vaccinated herds (coef(log scale) = −0.02, 95% CI = −0.04, 0.004, p = 0.056) compared to non-vaccinated herds during the study years. Youngstock and the cow culling rate due to respiratory disease was significantly lower in vaccinated herds compared to non-vaccinated herds (coef = −0.29, 95% CI = −0.47, −0.11, p = 0.003 and coef = −0.15, 95% CI = −0.29, −0.007, p = 0.043, respectively). These results suggest that vaccination against BHV-1 is associated with herd health and productivity. © 2015 Elsevier B.V. All rights reserved. 1. Introduction Bovine herpesvirus 1 (BHV-1) is an important cattle pathogen causing infectious bovine rhinotracheitis (IBR), infectious pustulous vulvovaginitis/balanoposthitis (IPV/IPB), abortions as well as systemic illness in young calves. BHV-1 establishes life-long latency following acute infection and can reactivate under unfavourable conditions for the virus-carrier animal (Kaashoek et al., 1996; Jones and Chowdhury, 2007). ∗ Corresponding author. Tel.: +372 56219397; fax: +372 7313706. E-mail address: [email protected] (K. Raaperi). Several countries and larger districts of a few countries are to date recognised as IBR-free. European Commissionapproved eradication programmes are ongoing in some countries and regions to which the additional guarantees for IBR apply (2011/674/EU). In addition to negative influences on animal health, IBR has become a limiting factor in livestock trade (64/432/EEC Article 9). Eradication programmes relying entirely or partly on BHV-1 DIVA vaccines (syn. marker vaccines) are currently running in several countries. A DIVA vaccine carries at least one antigenic protein less than the corresponding wild-type virus. After infection, but not after vaccination, it is possible to detect an antibody response for the specific protein with a companion diagnostic test (van Oirschot, 1999). To date, only glycoprotein E-deleted virus http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 0167-5877/© 2015 Elsevier B.V. All rights reserved. Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 G Model PREVET-3732; No. of Pages 10 ARTICLE IN PRESS K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx 2 vaccines are in use in the European Union for IBR eradication (Sanco/C3/AH/R20/2000). Continuous vaccination with live or inactivated gE-negative vaccines is reducing the circulation of the virus within a herd (Bosch et al., 1998; Mars et al., 2001; Makoschey et al., 2007; Vilmos et al., 2007; Jacevičius et al., 2008; Ampe et al., 2012; Raaperi et al., 2012a). By limiting the risk of having new BHV-1 gE-positive animals within the herd and by the gradual culling of field virus-infected animals, the herd may achieve IBR-free status (Makoschey and Bielsa, 2007). Published reports on the impact of BHV-1 infections and outbreaks of BHV-1 on animal health under field conditions have described reduced milk production, an increase of respiratory disease, higher mortality of calves as well as reproductive problems (Allan et al., 1980; Janzen et al., 1980; Wiseman et al., 1980; Greig et al., 1981; Cook, 1998; van Schaik et al., 1999; Holzhauer et al., 2003; Rissi et al., 2008). On the other hand, subclinical spread of BHV-1 in a naive herd has been demonstrated (van Oirschot et al., 1993; Hage et al., 1998; Pritchard et al., 2003). Clinical and pathological effects of BHV-1 infection have been shown in experimental challenge studies (Bitsch, 1973; Miller and Van der Maaten, 1987; Belknap et al., 1994). Several studies showed associations between the prevalence of BHV-1 serum antibodies and impaired health and performance in cattle herds (Biuk-Rudan et al., 1999; Waldner, 2005; Mineo et al., 2006; Raaperi et al., 2012b,c; Roshtkhari et al., 2012). Most of the experimental and observational studies have followed an epidemic of IBR in a previously naive herd in order to estimate losses. In countries where BHV-1 infection is endemic, a number of herds are permanently infected and clinical manifestations of the BHV-1 infection may not be as obvious as that following the first introduction of the virus. Due to overlapping clinical manifestation of several infections causing respiratory and reproductive disorders, contribution of the IBR virus to the overall performance of the animals within the herd is obscure. Concerning BHV-1, implementation of a vaccination protocol may reduce or eliminate the impact of the infection on herd health. Comparing vaccinated herds with those not implementing any control measures against BHV-1, enables evaluating the actual role of IBR infection in herd health in real farm conditions. Asking the farmer’s opinion about the change in herd health during vaccinating years would allow evaluation of whether the changes are recognisable to the farmer. The objective of this study was to estimate the impact of BHV-1 infection on health and performance of dairy cattle herds via the evaluation of the effect of vaccination with inactivated gE-negative marker vaccines. 2. Materials and methods 2.1. Study design 2.1.1. Herd selection After the prevalence study of Raaperi et al. (2010), control programmes were composed for seven dairy cattle herds aiming to eradicate the virus from the herds. Characteristics of those herds are given in Table 1 (V+ herds). In those herds, inactivated gE-negative marker vaccines were used for all animals over three months of age twice a year. Rispoval IBR Marker inactivatum (Pfizer Animal Health) (V+ I, II, III, IV, V and VII) and/or Bovilis IBR marker inac. (Intervet International) (V+ V and VI) were used (Raaperi et al., 2012a). Herds started with the vaccination between April 2007 and November 2008. Also, seven non-vaccinated dairy herds were selected and matched individually with vaccinated herds based on the information available from BHV-1 prevalence study conducted from 2006 to 2008 (Raaperi et al., 2010). Matching criteria were: similar herd size, milk productivity, breed, animal keeping system, BHV-1 prevalence before vaccination and herd bovine viral diarrhoea virus (BVDV) status. None of the participating herds had vaccinated against BHV-1 or BVDV before (Raaperi et al., 2010). Description of the herds included in this study is given in Table 1. 2.1.2. Data collection Herd health data for the years 2007–2012 were received from the Estonian Animal Recording Centre (EARC). For the two herds that started IBR vaccination in 2008, data of five years (2008–2012) were included and consequently also for the two non-vaccinated herds. In five herds the vaccination programme was initiated in 2007 and therefore six study years (2007–2012) were included for those herds as well as for the five non-vaccinated herds. The dataset contained 80 records in total. An internet questionnaire was composed for all vaccinating farms and completed by the farmer or farm manager. A pre-testing of the questionnaire was accomplished by sending the questionnaire to one farmer not participating in the study. After completion the author had a discussion with the test farmer to discover if any of the questions were confusing. All the comments of the test farmer were taken into account to improve the questionnaire. The questionnaire was composed to collect information about the continuation and regularity of the vaccination programme in order to confirm the suitability of the herd to be included. Specific questions were asked about the change in respiratory disease incidence in calves, heifers and cows, calf mortality, abortion incidence in heifers and cows, reproduction performance of heifers and cows, and milk production and treatment costs (antimicrobial drugs) compared to the pre-vaccination period. For that, multiple-choice questions with pre-defined answer categories (‘increased’, ‘decreased’, ‘remained the same’, ‘don’t know’) were used. We also questioned farmers from non-vaccinating farms about vaccinations against BHV-1 in the last ten years in order to confirm the suitability of the farm as a ‘nonvaccinated herd’ in our study. 2.1.3. Data analysis Linear mixed-effects models were constructed to evaluate the effect of vaccination to herd health and productivity parameters. Outcome variables analysed were ‘calving interval’, ‘average cow milk production’, ‘average somatic cell count’, ‘first service conception rate’, ‘insemination index’ (number of inseminations per pregnancy), ‘herd average number of lactations’, ‘days from Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 V− I V+ II V− II V+ III V− III V+ IV V− IV V+ V V− V V+ VI V− VI V+ VII V− VII 402 7523 EH, ER 415 6200 EH, ER 1000 8369 EH, ER 1000 7238 EH 530 7000 EH, ER 613 7400 EH, ER 1820 8416 EH 524 8700 EH, ER 2187 8325 EH 832 6301 EH, ER 674 8300 EH 1791 6400 EH 619 8085 EH, ER 608 8284 EH Farm management Number of barns Veterinarian is the employee of the farm Inseminator is the employee of the farm Using bull 1 Yes >1 Yes >1 Yes >1 Yes >1 Yes >1 No >1 Yes >1 Yes >1 Yes >1 No >1 Yes >1 Yes >1 No >1 Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No No H C H H H No H No H No No No No No No No No Yes Yes Yes No Yes Yes Yes Yes ≥2 ≥2 ≥2 ≥2 ≥2 ≥2 0 ≥2 ≥2 ≥2 ≥2 1 ≥2 0 Farm biosecurity Inseminator gives service to other farms Number of animals purchased in last 3 years Cow management Barn type Keeping system Grazing CB L Yes CB L No CB L No CB L Yes CB L No CB L Yes CB L No CB L No CB L No CB L Yes WB F Yes WB F Yes WB F Yes WB F Yes Youngstock management Way of keeping Keeping system Relocating animals between the barns (number of times per year) Grazing TC M 0 S L >2 S M >2 S M >2 S M 2 S L >2 S M >2 S L >2 S M >2 S L >2 S L >2 S M >2 S F >2 S M >2 Yes No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Infection status Herd BVD status IBR prevalence in cows IBR prevalence in youngstock Pos 90 38 Neg 81 4 Pos 84 24 Neg 51 9 Pos 89 91 Pos 95 63 Pos 95 89 Neg 100 16 Pos 97 60 Neg 21 7 Pos 98 44 Pos 92 89 Pos 67 2 Neg 48 0 V+ – vaccinated, V− – non-vaccinated, EH – Estonian Holstein, ER – Estonian Red, H – heifers, C – cows, CB – cold barn, WB – warm barn, F – fixed, L – loose, M – mixed, TC – together with cows, S – separately from cows, Pos – positive, Neg – negative. ARTICLE IN PRESS V+ I Number of cows Year milk production (kg/cow) Breed G Model PREVET-3732; No. of Pages 10 Farm Farm characteristics K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 Table 1 Description of herds in the longitudinal bovine herpesvirus 1 vaccination study during 2007–2012 in Estonia. 3 G Model ARTICLE IN PRESS PREVET-3732; No. of Pages 10 4 K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx calving to first pregnancy’, ‘age of first calving’, ‘length of dry period’, ‘youngstock (from first days of life to calving) culling rate (number of youngstock culled/total number of cows in 31st December*100)’, ‘youngstock culling rate due to: respiratory disease/reproduction problems/ abortions/appearance fault/selling/digestive disorders/ metabolic disorders/foot disorders/infectious diseases/ traumas/loosing/accident/meat production/fattening/ other reasons’ (number of youngstock culled due to given reason/total number of cows*100), ‘cow culling rate’ (number of cows culled/total number of cows*100), ‘cow culling rate due to: age/low productivity/udder diseases and mastitis/reproduction problems/abortion/difficult calving/foot disorders/digestive disorders/respiratory diseases/infectious diseases/traumas/loosing/accident/other reasons (number of cows culled due to given reason/total number of cows*100)’, ‘abortion % (number of abortions/total number of cows*100)’ and ‘stillbirth % (number of stillbirths/total number of cows*100)’. Fixed effects for the model were ‘year’ and ‘vaccination status’ (vaccinated/non-vaccinated) with interaction term between those two. The herd itself was considered a random effect. As our main interest was to determine the trend over time of the outcome variable with our models, ‘year’ was included as a continuous variable. General equation of the linear mixed models was as follows: Yi = ˇ0 + ˇ1 X1(i) ∗ X2(i) + uherd(i) + εi in which Y is the outcome variable, ˇ0 is the intercept, ˇ1 is the size of the effect of independent variables X1 (year) and X2 (herd category: vaccinated or not-vaccinated), uherd(i) is the random effect (herd) with the first-order autoregressive (AR1) covariance structure for repeated measures and εi is an error term. Linearity of the relationship was controlled graphically and statistically. In order to check if the relationship was curvilinear, square, cubic and quadratic terms for ‘year’ were inspected for significance. In order to confirm constant variance and residual normality, a residual plot of the model was composed to detect for homoscedasticity and q–q plots to check residuals for normality. Residuals were plotted against independent variables to discover hidden patterns. Significance of the random effect was evaluated by comparing models using likelihood ratio test (Table 2). Four models showed positive log likelihood values, however the model with a random effect gave AIC and BIC values closer to negative infinity and was therefore considered better in terms of statistical fitness. In order to complete the linearity assumption of the linear models, values of the variables ‘insemination index’ and ‘length of dry period’ were log-transformed and square root was taken from ‘youngstock culling rate’. R version 2.13.0 (The R Foundation for Statistical Computing) was used for linear mixed model building via using the lme function. The variables ‘youngstock culling rate due to respiratory disease’ and ‘cow culling rate due to respiratory disease’ had a negative binomial distribution; therefore mixed-effects negative binomial models were composed for those parameters. Statistical software SAS version 9.2 (SAS Institute Inc., USA) with GLIMMIX procedure was used for these models. Descriptive statistics were used to summarise the responses to questionnaires. 3. Results 3.1. Effect of BHV-1 vaccination on health and productivity of dairy cattle Results of the outcome variables significantly associated with vaccination status are presented in Table 2. Figs. 1 and 2 present the raw data together with results of the statistical models in which the significance of the main effect year (Y) in vaccinated (V+) and non-vaccinated herds (V−) as well as the interaction between the year and vaccination status of the farm (Y*V) is presented. The trend of change in the two groups of herds is given with lines of average fitted values obtained from the model. For graphical presentation the model results were back-transformed to the original scale (Figs. 1 and 2). In V+ herds the calving interval showed a decreasing trend over six years (coef = −7.01 days, 95% CI = −11.98, −2.03, p = 0.008) compared to the non-vaccinated herds. The insemination index demonstrated an increasing trend (coef(log scale) = 0.03, 95% CI = −0.003, 0.06, p = 0.054) and the first service conception rate decreased (coef = −2.19, 95% CI = −3.91, −0.47, p = 0.015) in V− herds, whereas insignificant change was ascertained in vaccinated herds during study years (Fig. 1, Table 2). Average yearly milk yield per cow increased (coef = 145.30 kg, 95% CI = −6.11, 296.71, p = 0.065) and length of the dry period decreased (coef(log scale) = −0.02, 95% CI = −0.04, 0.004, p = 0.056) in V+ herds compared to non-vaccinated herds (Fig. 1, Table 2). The youngstock culling rate was higher in V+ herds (coef(square root) = 0.20, 95% CI = −0.002, 0.40, p = 0.054) compared to V− herds. The youngstock culling rate due to respiratory disease increased in V− herds (coef = 0.24, 95% CI = 0.12, 0.36, p < 0.001), whereas no statistically significant change was ascertained in V+ herds (p = 0.461). The same trend was observed among cows where culling rate due to respiratory disease was lower in V+ herds (coef = −0.15, 95% CI = −0.29, −0.007, p = 0.043) than in V− herds (Fig. 2, Table 2). 3.2. Results of questionnaire data Six out of seven vaccinating farmers completed the internet questionnaire. Five farmers declared that they have been vaccinating continuously and no vaccination round was skipped. ‘V+ V’ farm stated that due to financial problems vaccination was halted from spring 2012 to spring 2013. All farmers of the non-vaccinated herds confirmed that they had not been vaccinating against BHV-1. According to the farmers’ opinions, calf respiratory disease incidence decreased in five out of six vaccinating farms and increased in one herd (V+ II) compared to the pre-vaccination period. Farmers claimed that after starting with vaccination, respiratory disease incidence among Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 G Model ARTICLE IN PRESS PREVET-3732; No. of Pages 10 K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx 5 Table 2 Results of the linear and negative binomial mixed-effect models for longitudinal bovine herpesvirus 1 vaccination study during 2007–2012 in Estonia. p-Value 95% CI Random effect p-value LogLik (MM) LogLik (FEM) Calving interval (days) 410.15 Intercept 3.44 Y 23.44 V+ Y*V+ −7.01 Coefficient 0.000 0.06 0.071 0.008 393.73; 426.57 −0.08; 6.96 0.22; 46.66 −11.98; −2.03 <0.0001 −325.06 −342.18 Milk production (kg/year) 8148.17 Intercept 65.39 Y −561.29 V+ 145.30 Y*V+ 0.000 0.236 0.289 0.065 7446.62; 8849.72 −41.68; 172.46 −1553.43; 430.85 −6.11; 296.71 <0.0001 −589.41 −628.86 Insemination index (log scale) 0.61 Intercept 0.03 Y V+ 0.30 −0.05 Y*V+ 0.000 0.054 0.047 0.033 0.42; 0.80 −0.003; 0.06 0.04; 0.56 −0.10; −0.003 <0.0001 44.21 1.72 Length of dry period (days) (log scale) 4.25 0.000 Intercept −0.004 0.643 Y 0.05 0.407 V+ Y*V+ −0.02 0.056 4.17; 4.33 −0.02; 0.01 −0.06; 0.16 −0.04; 0.004 <0.0001 70.91 57.20 First service conception rate (%) 53.18 Intercept Y −2.19 −16.44 V+ 3.14 Y*V+ 43.08; 63.28 −3.91; −0.47 −30.73; −2.15 0.71; 5.57 <0.0001 −262.88 −303.56 2.98; 4.26 −0.19; 0.09 −2.00; −0.18 −0.002; 0.40 <0.0001 −74.16 −92.59 <0.0001 206.87 253.32 <0.0001 167.73 137.61 0.000 0.015 0.044 0.014 Youngstock culling rate (%) (square root) Intercept 3.62 0.000 −0.05 0.535 Y −1.09 0.036 V+ Y*V+ 0.20 0.054 Youngstock culling rate (%) due to respiratory diseasea Intercept −0.51 0.204 −1.29; 0.26 0.24 0.0004 0.12; 0.36 Y 0.86 0.128 −0.21; 1.93 V+ Y*V+ −0.29 0.003 −0.47; −0.11 Cow culling rate (%) due to respiratory diseasea Intercept −0.74 0.122 0.11 0.054 Y V+ 0.03 0.969 −0.15 0.043 Y*V+ −1.62; 0.15 0.0001; 0.21 −1.20; 1.25 −0.29; −0.007 a Negative binomial mixed-effects model. MM – mixed effect model, FEM – fixed effect model, Y – year, V – vaccination status, V+ – vaccinated herds. heifers and cows has decreased in six and four herds out of six, respectively. Farmers indicated that calf mortality also decreased in five out of six herds (V+ I, II, III, V, VII) during the vaccination period. Heifer and cow abortion incidence was ascertained to decrease in one (V+ II) and two herds (V+ III and IV) respectively, whereas increase in abortion incidence was not indicated by any of the vaccinating farms. Two farmers (V+ II and III) stated that overall reproduction status improved after starting IBR vaccination, three of them (V+ I, V, VII) stated that it remained the same and one (V+ IV) could not answer the question. Milk productivity was estimated to increase in three farms (V+ II, III, VII) out of six and treatment costs were stated to be increased in two farms (V+ II, III) and decreased in two vaccinated herds (V+ I, IV) (Table 3). An open question asked to specify any other change occurring during the vaccination period that was not asked in the questionnaire. The farmer from herd ‘V+ IV’ reported that in addition to reduced incidence of pneumonia, there was a decrease in conjunctivitis cases and therefore less blindness in animals. 4. Discussion The aim of this study was to estimate the impact of BHV-1 infection on health and performance of dairy cattle herds via the evaluation of the effect of vaccination with inactivated gE-negative marker vaccines. 4.1. Change in herd health related to vaccination against BHV-1 4.1.1. Respiratory disease Most of the farmers claimed they experienced a decline in calf, heifer and cow respiratory disease as well as calf mortality in the vaccination period compared with the Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 G Model ARTICLE IN PRESS PREVET-3732; No. of Pages 10 A Calving inte rval (days) 500 V+ V- Y(V-) Y(V+) Y*V 475 p = 0.060 p = 0.051 p = 0.008 450 425 400 375 2007 2008 2009 2010 2011 2012 Year Average cow milk production (kg/year) K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx 6 V+ V- 4.0 Y(V-) Y(V+) Y*V p = 0.054 p = 0.269 p = 0.033 Length of dry pe riod (days) Insemination index Y(V-) Y(V+) Y*V V+ V- 11000 p = 0.236 p < 0.001 p = 0.065 10000 9000 8000 7000 6000 2007 2008 2009 2010 2011 2012 Ye ar D C 4.5 3.5 3.0 2.5 2.0 1.5 2007 2008 2009 2010 2011 2012 Year First service conception rate (%) B 12000 100 V+ Y(V-) Y(V+) Y*V V- 90 p = 0.643 p = 0.002 p = 0.056 80 70 60 50 2007 2008 2009 2010 2011 2012 Ye ar E 90 Y(V-) Y(V+) Y*V V+ V- 80 p = 0.015 p = 0.285 p = 0.014 70 60 50 40 30 20 2007 2008 2009 2010 2011 2012 Ye ar Fig. 1. Reproduction and productivity parameters in herds of a longitudinal bovine herpesvirus 1 vaccination study during years 2007–2012 in Estonia. Graph points represent the raw data in vaccinated () and non-vaccinated herds (♦). Results of the linear mixed model are presented with p-values for the main effect year (Y) in vaccinated (V+) and non-vaccinated herds (V−) as well as the interaction term between the year and vaccination status of the farm (Y*V). The trend of change in the two groups of herds is given with lines (solid line for vaccinated herds and dashed line for non-vaccinated herds) of average fitted values obtained from the model. pre-vaccination period. From our previous studies we reported a significant association between high occurrence of respiratory disease in unweaned calves and the presence of BHV-1 among cows (Raaperi et al., 2012b). However, BHV-1 was not significantly associated with high respiratory disease occurrence among heifers (Raaperi et al., 2012b) or cows (Raaperi et al., 2012c). Unfortunately there is no obligation to insert disease incidence data into the EARC database, so we could not test the change of respiratory disease incidence statistically. In vaccinated herds the overall culling rate of youngstock increased, whereas it remained unchanged in non-vaccinated herds. Still, in vaccinated herds the average culling rate was lower than in non-vaccinated herds in the beginning of the study period and reached the same level by the end of it (year 2012) (Fig. 2A). The culling Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 G Model PREVET-3732; No. of Pages 10 ARTICLE IN PRESS K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx 7 Table 3 Results of farmer’s opinion about the change in herd health during BHV-1 vaccination period compared to pre-vaccination period. Question Answer Replies (n/n total) Farm V+ Change in calf respiratory disease incidence Increased Decreased Remained the same Don’t know 1/6 5/6 0 0 II I, III, IV, V, VII Change in heifer (6 months to calving) respiratory disease incidence Increased Decreased Remained the same Don’t know 0 6/6 0 0 Change in respiratory disease incidence among cows Increased Decreased Remained the same Don’t know 0 4/6 1/6 1/6 Change in calf mortality Increased Decreased Remained the same Don’t know 0 5/6 0 1/6 Change in heifer abortion incidence Increased Decreased Remained the same Don’t know 0 1/6 1/6 4/6 II VII I, III, IV, V Change in cow abortion incidence Increased Decreased Remained the same Don’t know 0 2/6 2/6 2/6 III, IV II, VII I, V Change in heifer/cow reproduction performance Improved Worsened Remained the same Don’t know 2/6 0 3/6 1/6 II, III Change in milk production Increased Decreased Remained the same Don’t know 3/6 0 0 3/6 II, III, VII Change in treatment costs (antibacterial drugs) Increased Decreased Remained the same Don’t know 2/6 2/6 1/6 1/6 II, III I, IV V VII I, II, III, IV, V, VII I, II, IV, VII V III I, II, III, V, VII IV I, V, VII IV I, IV, V V+ – vaccinated. rate of youngstock due to respiratory disease remained statistically unchanged in vaccinated herds but increased in non-vaccinated herds during six study years (Fig. 2B). This indicates that in vaccinated herds youngstock was culled more often for other reasons than respiratory disease. We analysed culling data for all other culling reasons and only found that the culling rate due to metabolic disorders increased in the second, third and fourth year of the vaccination period followed by a decline. After that it remained at the same level as it was among non-vaccinated herds at the end of the study period (data not shown). EARC discriminates between metabolic disorders and digestive disorders in their youngstock culling database. Unfortunately there are no instructions available for the farmer by how to interpret different culling reasons. The average culling age due to metabolic and digestive disorders among youngstock was six months and one month, respectively, in the EARC Annual Synopsis 2012. Based on that, we can assume that culling due to metabolic disorders is related with conditions leading to inadequate growth, whereas mortality or culling due to calf diarrhoea can be excluded. As most of the farmers claimed that calf mortality decreased after vaccination we can assume that more heifers stayed in the herd and the farmer could probably do more voluntary culling among youngstock, excluding the inadequately growing heifers or animals with phenotype imperfection. Culling cows due to respiratory disease showed an increasing trend in non-vaccinated herds but no changes were ascertained among vaccinated herds (Fig. 2C). It is difficult to state whether or not this is linked to BHV-1. We did not carry out any investigations in the non-vaccinated herds and are unable to explain what is actually behind that trend. We also asked about the change in treatment costs (antimicrobial agents) in vaccinated herds during the vaccination period. Decrease in treatment costs was declared by two farmers (V+ I and IV); two farms stated they experienced an increase in treatment costs (V+ II and III). We did not ask the cause of treatment; therefore changes in treatment costs cannot be attributed to a specific cause. Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 G Model ARTICLE IN PRESS PREVET-3732; No. of Pages 10 K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx 8 Youngstock culling rate (%) A 40 V+ V- 30 Y(V-) Y(V+) Y*V p = 0.535 p = 0.035 p = 0.054 2009 2010 20 10 0 2007 2008 2011 2012 2011 2012 Year Youngstock culling rate (%) due to respiratory disease B 7 V+ V- 6 Y(V-) Y(V+) Y*V p < 0.001 p = 0.461 p = 0.003 2009 2010 5 4 3 2 1 0 2007 2008 Year C Cow culling rate (%) due to respiratory disease 4 V+ V- 3 Y(V-) Y(V+) Y*V p = 0.054 p = 0.374 p = 0.043 2 1 0 2007 2008 2009 2010 2011 2012 Year Fig. 2. Culling rates in herds of a longitudinal bovine herpesvirus 1 vaccination study during years 2007–2012 in Estonia. Graph points represent the raw data in vaccinated () and non-vaccinated herds (♦). Results of the linear mixed model (A) and mixed effect negative binomial model (B and C) are presented with p-values for the main effect year (Y) in vaccinated (V+) and non-vaccinated herds (V−) as well as the interaction term between the year and vaccination status of the farm (Y*V). The trend of change in the two groups of herds is given with lines (solid line for vaccinated herds and dashed line for non-vaccinated herds) of average fitted values obtained from the model. 4.1.2. Reproduction and milk production According to the farmers’ opinions, the overall reproduction status among heifers and cows improved or remained the same after the beginning of the vaccination programme. Farmers evaluated the change in abortion incidence differently, but none of them claimed that the abortion incidence had increased. The insemination index showed an increasing trend and the first service conception rate decreased significantly in non-vaccinated herds but remained unchanged in vaccinated herds (Fig. 1C and E). The length of the dry period shortened considerably in vaccinated herds, whereas it remained statistically unchanged and longer in non-vaccinated herds (Fig. 1D). Studies have shown improvement in reproductive performance with a shorter dry period in multiparous animals. Due to shorter dry periods, the postpartum negative energy balance is improved, resulting in fewer days to first ovulation and a greater success of artificial insemination (Watters et al., 2009; Grummer et al., 2010; Mansfeld et al., 2012). This may have an impact on the overall calving interval, which showed a decreasing trend in the vaccinated herds over the years, whereas an adverse trend was observed in nonvaccinated herds (Fig. 1A). A negative effect of BHV-1 infection on milk production has been reported from earlier outbreak investigations (Hage et al., 1998; van Schaik et al., 1999). In the simulation model by Vonk Noordegraaf et al. (1998) the average decrease in milk production of gE-positive cows was considered to be about 150 kg per year, which is about 2%. Our results show that inhibiting virus circulation within the herd and keeping the young replacement cows free of BHV-1 infection results in an increase in milk productivity on average 145.30 kg/year (95% CI = −6.11, 296.71, p = 0.065) compared to that in non-vaccinated herds. This may also be the consequence of shorter dry periods and calving intervals, leading to higher number of days in milk with concurrent increase in milk production (Kuhn et al., 2006; Bello et al., 2012). 4.1.3. Validity and bias of the study The limited sample size has lowered the power of the study; therefore some relationships might have been insignificant. Herds participating the study were large with more than 400 dairy cows. The total number of herds in this herd size category (herds with >400 cows) was 59 in year 2010 in Estonia (about 1% of all dairy cow herds) (Estonian Animal Register, 2010). Thus the study sample (24% of largest herds) may be considered to represent well this subpopulation of herds. Also, 38% of all dairy cows of Estonia were kept in these 59 herds (Estonian Animal Register, 2010) meaning the study covered substantial part of all dairy cows in the country. Consequently, the results of this study can be extrapolated mainly to larger herds with intensive production systems and relatively high productivity. All herds originated from the same source population which consist of herds randomly selected from the list of all herds with more than 400 cows and positive BHV-1 antibody test result in 2004 which participated the prevalence study in 2006–2008 (Raaperi et al., 2010). The study included all herds that volunteered to start vaccinating Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 G Model PREVET-3732; No. of Pages 10 ARTICLE IN PRESS K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx after the prevalence study (Raaperi et al., 2010) and nonvaccinated herds were matched with vaccinated herds according to herd and management characteristics. Therefore the study herds should well reflect the source and target population and selection bias is reduced to minimum. Time gap of up to two years occurred between measurement of the initial prevalence and start of the vaccination programme, therefore the initial BHV-1 prevalence estimates may be somewhat biased. Still, the process of endemic infection was continuous in all study herds during that intermediary time and the probability of considerable changes in within-herd prevalence of BHV-1 is the same for all fourteen herds. However, herd BVDV status was one of the matching criterions and five herds were BVDV negative at the time the first farm inspection and sampling took place. Some herds might have introduced the infection during that time gap causing misclassification bias of the herd BVDV status turning to concurrent selection bias in the matching process. Vaccinated herds may differ from non-vaccinated herds as well as being more prone to make improvements also in other animal husbandry practices, e.g. feeding, keeping management, controlling other infectious diseases; as well as attitude, education and awareness of the farm owner or manager. Therefore some confounding bias may still remain even after matching on herd variables. Also, due to the limited number of possible non-vaccinated herds available, we could not match herds perfectly in all aspects (milk production level, BVDV status, etc.) (Table 1). Vaccination was interrupted in the ‘V+ V’ farm due to financial problems from spring 2012 to spring 2013. We don’t expect the effect of vaccination to wane sharply after a five-year vaccination period and no dramatic changes in herd health data were noticed; therefore we still included the herd in our study. Due to communication difficulties, questionnaire data were not received from the ‘V+ VI’ herd, limiting our conclusions about the farmer’s opinion in the change of herd health. Despite that, confirmation about the regular continuity of the vaccination programme was gained after communication with the veterinarian employed by that farm, allowing us to include the farm data in the statistical analysis. 5. Conclusions Our findings indicate that the BHV-1 infection may have clinical significance for Estonian dairy cattle herds and is most probably a limiting factor for dairy production. Although sample size of this study was relatively small, significant relationships were detected between BHV-1 vaccination and productivity parameters. Larger number of herds of different size participating in a study would allow us to evaluate more broadly the impacts of the BHV-1 control on herd health and productivity in different production systems. This information is especially valuable for farmers and veterinary advisors from countries with voluntary BHV-1 control programmes, as improvement of herd health and productivity are the major incentives to invest in a rather costly, laborious and long-term disease control programme. 9 Conflict of interest statement The authors declare that they have no financial or personal relationships with other people or organisations that could inappropriately influence the results of this study. Acknowledgements This research was supported financially by the Estonian Ministry of Agriculture (Research contract 34-23 2006–2008) and by Institutional Research Funding (IUT81) of the Estonian Ministry of Education and Research. The authors thank the farmers and veterinarians participating in the study and the Estonian Animal Recording Centre for providing the data. References Allan, E.M., Pirie, H.M., Msolla, P.M., Selman, I.E., Wiseman, A., 1980. The pathological features of severe cases of infectious bovine rhinotracheitis. Vet. Rec. 107, 441–445. Ampe, B., Duchateau, L., Speybroeck, N., Berkvens, D., Dupont, A., Kerkhofs, P., Thiry, E., Dispas, M., 2012. Assessment of the long-term effect of vaccination on transmission of infectious bovine rhinotracheitis virus in cattle herds hyperimmunized with glycoprotein E-deleted marker vaccine. Am. J. Vet. Res. 73, 1787–1793. Belknap, E.B., Collins, J.K., Ayers, V.K., Schultheiss, P.C., 1994. Experimental infection of neonatal calves with neurovirulent bovine herpesvirus type 1.3. Vet. Pathol. 31, 358–365. Bello, N.M., Stevenson, J.S., Tempelman, R.J., 2012. Invited review: milk production and reproductive performance: modern interdisciplinary insights into an enduring axiom. J. Dairy Sci. 95, 5461–5475. Bitsch, V., 1973. Infectious bovine rhinotracheitis virus infection in bulls, with special reference to preputial infection. Appl. Microbiol. 26, 337–343. Biuk-Rudan, N., Cvetníc, S., Madić, J., Rudan, D., 1999. Prevalence of antibodies to IBR and BVD viruses in dairy cows with reproductive disorders. Theriogenology 51, 875–881. Bosch, J.C., De Jong, M.C., Franken, P., Frankena, K., Hage, J.J., Kaashoek, M.J., Maris-Veldhuis, M.A., Noordhuizen, J.P., Van der Poel, W.H., Verhoeff, J., et al., 1998. An inactivated gE-negative marker vaccine and an experimental gD-subunit vaccine reduce the incidence of bovine herpesvirus 1 infections in the field. Vaccine 16, 265–271. Cook, N., 1998. Combined outbreak of the genital and conjunctival forms of bovine herpesvirus 1 infection in a UK dairy herd. Vet. Rec. 143, 561–562. Greig, A., Gibson, I.R., Nettleton, P.F., Herring, J.A., 1981. Disease outbreak in calves caused by a mixed infection with infectious bovine rhinotracheitis virus and bovine virus diarrhoea virus. Vet. Rec. 108, 480. Grummer, R.R., Wiltbank, M.C., Fricke, P.M., Watters, R.D., Silva-Del-Rio, N., 2010. Management of dry and transition cows to improve energy balance and reproduction. J. Reprod. Dev. 56 (Suppl.), S22–S28. Hage, J.J., Schukken, Y.H., Dijkstra, T., Barkema, H.W., van Valkengoed, P.H., Wentink, G.H., 1998. Milk production and reproduction during a subclinical bovine herpesvirus 1 infection on a dairy farm. Prev. Vet. Med. 34, 97–106. Holzhauer, M., Dijk, R., Mars, J., 2003. Infectious bovine rhinotracheitis outbreak on a mostly BHV-1 free farm can result in great damage. Tijdschr. Diergeneeskd. 128, 593–595. Jacevičius, E., Šalomskas, A., Milius, J., Petkevičius, S., Mockeliūnas, R., Jacevičienē, I., Lelešius, R., Pridotkas, G., 2008. Prevalence and control measures of infectious bovine rhinotracheitis in Lithuania. In: Animals. Health. Food Hygiene, Jelgava., pp. 49–53. Janzen, E.D., Smart, J.N., Nicholson, H.H., 1980. Observations on an outbreak of infectious bovine rhinotracheitis in a bull test station. Can. Vet. J. 21, 24–27. Jones, C., Chowdhury, S., 2007. A review of the biology of bovine herpesvirus type 1 (BHV-1), its role as a cofactor in the bovine respiratory disease complex and development of improved vaccines. Anim. Health Res. Rev. 8, 187–205. Kaashoek, M.J., Straver, P.H., Van Rooij, E.M., Quak, J., Van Oirschot, J.T., 1996. Virulence, immunogenicity and reactivation of seven bovine herpesvirus 1.1 strains: clinical and virological aspects. Vet. Rec. 139, 416–421. Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014 G Model PREVET-3732; No. of Pages 10 10 ARTICLE IN PRESS K. Raaperi et al. / Preventive Veterinary Medicine xxx (2015) xxx–xxx Kuhn, M.T., Hutchison, J.L., Norman, H.D., 2006. Dry period length to maximize production across adjacent lactations and lifetime production. J. Dairy Sci. 89, 1713–1722. Makoschey, B., Bielsa, J.M., 2007. Europe’s progress in IBR virus eradication. Int. Dairy Top. 6, 13–14. Makoschey, B., Zehle, H.H., Bussacchini, M., Valla, G., Pálfi, V., Földi, J., 2007. Efficacy of a live bovine herpesvirus type 1 marker vaccine under field conditions in three countries. Vet. Rec. 161, 295–298. Mansfeld, R., Sauter-Louis, C., Martin, R., 2012. Effects of dry period length on milk production, health, fertility, and quality of colostrum in dairy cows. Invited review. Tierarztl. Prax. Ausg. G Grosstiere Nutztiere 40, 239–250. Mars, M.H., de Jong, M.C., Franken, P., van Oirschot, J.T., 2001. Efficacy of a live glycoprotein E-negative bovine herpesvirus 1 vaccine in cattle in the field. Vaccine 19, 1924–1930. Miller, J.M., Van der Maaten, M.J., 1987. Early embryonic death in heifers after inoculation with bovine herpesvirus-1 and reactivation of latent virus in reproductive tissues. Am. J. Vet. Res. 48, 1555–1558. Mineo, T.W., Alenius, S., Näslund, K., Montassier, H.J., Björkman, C., 2006. Distribution of antibodies against Neospora caninum, BVDV and BHV1 among cows in Brazilian dairy herds with reproductive disorders. Rev. Bras. Parasitol. Vet. 15, 188–192. Pritchard, G.C., Banks, M., Vernon, R.E., 2003. Subclinical breakdown with infectious bovine rhinotracheitis virus infection in dairy herd of high health status. Vet. Rec. 153, 113–117. Raaperi, K., Aleksejev, A., Orro, T., Viltrop, A., 2012a. Dynamics of bovine herpesvirus 1 infection in Estonian dairy herds with and without a control programme. Vet. Rec. 171, 99. Raaperi, K., Bougeard, S., Aleksejev, A., Orro, T., Viltrop, A., 2012b. Association of herd BHV-1 seroprevalence with respiratory disease in youngstock in Estonian dairy cattle. Res. Vet. Sci. 93, 641–648. Raaperi, K., Bougeard, S., Aleksejev, A., Orro, T., Viltrop, A., 2012c. Association of herd BRSV and BHV-1 seroprevalence with respiratory disease and reproductive performance in adult dairy cattle. Acta Vet. Scand. 54, 4. Raaperi, K., Nurmoja, I., Orro, T., Viltrop, A., 2010. Seroepidemiology of bovine herpesvirus 1 (BHV1) infection among Estonian dairy herds and risk factors for the spread within herds. Prev. Vet. Med. 96, 74–81. Rissi, D.R., Pierezan, F., Sá e Silva, M., Flores, E.F., de Barros, C.S., 2008. Neurological disease in cattle in southern Brazil associated with Bovine herpesvirus infection. J. Vet. Diagn. Invest. 20, 346–349. Roshtkhari, F., Mohammadi, G., Mayameei, A., 2012. Serological evaluation of relationship between viral pathogens (BHV-1, BVDV, BRSV, PI-3V, and Adeno3) and dairy calf pneumonia by indirect ELISA. Trop. Anim. Health Prod. 44, 1105–1110. Sanco/C3/AH/R20/2000. Report on Bovine Herpesvirus 1 (BHV1) marker vaccines and the accompanying diagnostic tests. http://ec.europa.eu/ food/fs/sc/scah/out49 en.pdf (accessed 01.10.13). van Oirschot, J.T., 1999. Diva vaccines that reduce virus transmission. J. Biotechnol. 73, 195–205. van Oirschot, J.T., Straver, P.J., van Lieshout, J.A., Quak, J., Westenbrink, F., van Exsel, A.C., 1993. A subclinical infection of bulls with bovine herpesvirus type 1 at an artificial insemination centre. Vet. Rec. 132, 32–35. van Schaik, G., Shoukri, M., Martin, S.W., Schukken, Y.H., Nielen, M., Hage, J.J., Dijkhuizen, A.A., 1999. Modeling the effect of an outbreak of bovine herpesvirus type 1 on herd-level milk production of Dutch dairy farms. J. Dairy Sci. 82, 944–952. Vilmos, P., Laszlo, O., Jozsef, F., Emese, H.F., 2007. Field experiences on the eradication with live IBR marker vaccine in large-scale herds in Hungary. Magy. Allatorvosok. Lapja 129, 323–331. Vonk Noordegraaf, A., Buijtels, J.A., Dijkhuizen, A.A., Franken, P., Stegeman, J.A., Verhoeff, J., 1998. An epidemiological and economic simulation model to evaluate the spread and control of infectious bovine rhinotracheitis in The Netherlands. Prev. Vet. Med. 36, 219–238. Waldner, C.L., 2005. Serological status for N. caninum, bovine viral diarrhea virus, and infectious bovine rhinotracheitis virus at pregnancy testing and reproductive performance in beef herds. Anim. Reprod. Sci. 90, 219–242. Watters, R.D., Wiltbank, M.C., Guenther, J.N., Brickner, A.E., Rastani, R.R., Fricke, P.M., Grummer, R.R., 2009. Effect of dry period length on reproduction during the subsequent lactation. J. Dairy Sci. 92, 3081–3090. Wiseman, A., Msolla, P.M., Selman, I.E., Allan, E.M., Pirie, H.M., 1980. Clinical and epidemiological features of 15 incidents of severe infectious bovine rhinotracheitis. Vet. Rec. 107, 436–441. Please cite this article in press as: Raaperi, K., et al., Effect of vaccination against bovine herpesvirus 1 with inactivated gE-negative marker vaccines on the health of dairy cattle herds. PREVET (2015), http://dx.doi.org/10.1016/j.prevetmed.2015.01.014