Scholarly article on topic 'Six-year longitudinal study of Fasciola hepatica bulk milk antibody ELISA in the dairy dense region of the Republic Ireland'

Six-year longitudinal study of Fasciola hepatica bulk milk antibody ELISA in the dairy dense region of the Republic Ireland Academic research paper on "Animal and dairy science"

CC BY-NC-ND
0
0
Share paper
Academic journal
Preventive Veterinary Medicine
OECD Field of science
Keywords
{" Fasciola hepatica " / Ireland / ELISA / Weather / Dairy / Seasonality / Flukicide}

Abstract of research paper on Animal and dairy science, author of scientific article — M.P. Munita, R. Rea, Y. Bloemhoff, N. Byrne, A.M. Martinez-Ibeas, et al.

Abstract Completion of the F. hepatica lifecycle is dependent on suitable climatic conditions for development of immature stages of the parasite, and its snail intermediate host. Few investigations have been conducted regarding temporal variations in F. hepatica status in Irish dairy herds. The current study aimed to conduct a longitudinal study examining annual and seasonal trends in bulk milk seropositivity over six years, while also investigating associations with soil temperature, rainfall and flukicide treatment. Monthly bulk milk samples (BTM) were submitted by 28 herds between March 2009 and December 2014. In all, 1337 samples were analysed using a Cathepsin L1 ELISA. Soil temperature, rainfall and management data were obtained for general estimating equation and regression analyses. A general decrease in milk seropositivity was observed over the six year study period and was associated with an increased likelihood of treating for liver fluke (OR range=2.73–6.96). Annual and seasonal analyses of rainfall and F. hepatica BTM status yielded conflicting results. Higher annual rainfall (>1150mm) yielded a lower likelihood of being BTM positive than annual rainfall of <1000mm (OR=0.47; P =0.036). This was most likely due to farmers being more proactive in treating for F. hepatica in wetter years, although a ‘wash effect’ by high rainfall of the free living stages and snails cannot be ruled out. Higher seasonal rainfall (>120mm), however, was associated with increased ELISA S/P% values (Coefficient=9.63S/P%; P =0.001). Soil temperature was not found to influence F. hepatica to the same extent as rainfall and may reflect the lack of severe temperature fluctuations in Ireland. Flukicides active against both immature and mature F. hepatica were approximately half as likely to record a positive F. hepatica herd BTM status than a flukicide active against only the mature stage of the parasite (OR≅0.45; P <0.01). This study highlights the importance of examining both annual and seasonal F. hepatica data, which can vary significantly. Additionally, it highlights the progress that can be achieved in fluke control by application of a continuous BTM monitoring program.

Academic research paper on topic "Six-year longitudinal study of Fasciola hepatica bulk milk antibody ELISA in the dairy dense region of the Republic Ireland"

Contents lists available at ScienceDirect

Preventive Veterinary Medicine

journal homepage: www.elsevier.com/locate/prevetmed

Six-year longitudinal study of Fasciola hepatica bulk milk antibody ELISA in the dairy dense region of the Republic Ireland

M.P. Munitaa,b'*, R. Reab, Y. Bloemhoffa, N. Byrnea, A.M. Martinez-Ibeasa, R.G. Sayersa

a Animal and Grassland Research and Innovation Centre (AGRIC), Teagasc, Moorepark, Fermoy, County Cork, Ireland b Biological Sciences, Cork Institute of Technology, Bishopstown, Cork, Ireland

CrossMark

ARTICLE INFO

Article history: Received 22 July 2016 Received in revised form 26 September 2016 Accepted 28 September 2016

Keywords:

Fasciola hepatica

Ireland

Weather

Seasonality Flukicide

ABSTRACT

Completion of the F. hepatica lifecycle is dependent on suitable climatic conditions for development of immature stages of the parasite, and its snail intermediate host. Few investigations have been conducted regarding temporal variations in F. hepatica status in Irish dairy herds. The current study aimed to conduct a longitudinal study examining annual and seasonal trends in bulk milk seropositivity over six years, while also investigating associations with soil temperature, rainfall and flukicide treatment. Monthly bulk milk samples (BTM) were submitted by 28 herds between March 2009 and December 2014. In all, 1337 samples were analysed using a Cathepsin L1 ELISA. Soil temperature, rainfall and management data were obtained for general estimating equation and regression analyses. A general decrease in milk seropos-itivity was observed over the six year study period and was associated with an increased likelihood of treating for liver fluke (OR range = 2.73-6.96). Annual and seasonal analyses of rainfall and F. hepatica BTM status yielded conflicting results. Higher annual rainfall (>1150mm) yielded a lower likelihood of being BTM positive than annual rainfall of <1000mm (0R = 0.47; P=0.036). This was most likely due to farmers being more proactive in treating for F. hepatica in wetter years, although a 'wash effect' by high rainfall of the free living stages and snails cannot be ruled out. Higher seasonal rainfall (>120 mm), however, was associated with increased ELISA S/P% values (Coefficient = 9.63S/P%; P=0.001). Soil temperature was not found to influence F. hepatica to the same extent as rainfall and may reflect the lack of severe temperature fluctuations in Ireland. Flukicides active against both immature and mature F. hepatica were approximately half as likely to record a positive F. hepatica herd BTM status than a flukicide active against only the mature stage of the parasite (OR = 0.45; P<0.01). This study highlights the importance of examining both annual and seasonal F. hepatica data, which can vary significantly. Additionally, it highlights the progress that can be achieved in fluke control by application of a continuous BTM monitoring program.

© 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Fasciola hepatica, commonly known as liver fluke, is an helminth parasite of mammals of the Phylum Platyhelmintes, Class Trema-toda (Urquhart et al., 1996; Borgsteede, 2011). Hosts include cattle, sheep, goat, horse, deer and humans (Taylor et al., 2007). This parasite has a worldwide distribution across these species, although the impact on the human population is highly variable and very much related to socio-economic factors (Mas-Coma et al., 2009). It is considered an important disease of domestic livestock, especially in temperate climatic zones (Bennema et al., 2011), and has been

* Corresponding author at: Animal and Grassland Research and Innovation Centre (AGRIC), Teagasc, Moorepark, Fermoy, County Cork, Ireland.

E-mail address: maria.munitacorbalan@teagasc.ie (M.P. Munita).

estimated to generate annual losses of D 90 million in Irish livestock species and D 2.5 billion worldwide (AHI, 2011).

The predilection site for sexually mature F. hepatica in mammalian hosts is the bile ducts in the liver. From here, mature flukes shed eggs into the gastrointestinal tract which are subsequently released into the environment in faeces. Eggs hatch in the environment and develop into miracidia, a process requiring a minimum temperature of 10 °C and a reported optimum temperature of 22-26 °C (Taylor et al., 2007). Once developed, miracidia penetrate an intermediate molluscan snail host of the family Lymnaeidae, the particular species of which varies by geographical location. Galba truncatula acts as the main intermediate host in Europe, and this species requires moderate climatic conditions and muddy/sodden land to thrive (Thomas, 1883). Further larval stages develop in the snail host until cercariae are formed. On their release from the snail, cercaria encyst on vegetation to form the infective stage i.e.

http://dx.doi.org/10.1016/j.prevetmed.2016.09.024

0167-5877/© 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativec0mm0ns.0rg/licenses/by-nc-nd/4. 0/).

metacercariae. On ingestion by mammalian hosts, such as cattle and sheep, the infectious immature fluke migrate through intestinal and liver tissue until maturation in the bile ducts (Urquhart et al., 1996). The minimal period for the entire F. hepatica lifecycle is 17-18 weeks (Taylor et al., 2007).

Due to the environmental conditions required for fluke larval maturation, and for development of appropriate snail host habitat, temperature and rainfall have been identified as important contributory risk factors for an increased prevalence of F. hepatica. (Mas-Coma et al., 2009). Global warming is a widely accepted phenomenon and unprecedented changes in climate have been recorded since the 1950's. Further warming and long-lasting climate changes are expected (IPCC, 2014) with parallel increases in the risk of fasciolosis predicted (Fox et al., 2011; Van Dijk et al., 2010) due to increased cercarial production (Mas-Coma et al., 2009). The Republic of Ireland (heretofore referred to as Ireland) is located at 53.3442° N, 6.2675° W and is located in a temperate climatic zone. Between 1971 and 2010 the average monthly rainfall was 92.5 mm and average temperature was 10 °C (ECAD, 2016), appropriate conditions for completion of the fluke lifecycle as evidenced by a F. hepatica prevalence of 76.1% in Irish dairy herds in November 2009 (Bloemhoff et al., 2015). Met Éireann, the Irish Meteorological Service, has reported an increase in average air temperature in Ireland of 0.8 °C over the last 100 years and statistical prediction models indicate a rise in global temperature of 2 °C by the end of this century. Met Éireann (2016) also reports an approximate increase in rainfall of 5% over the last 30 years, and an increase in the number of'very wet days' (days with rainfall >20 mm). These ongoing and predicted weather changes in Ireland may present even more favourable conditions for F. hepatica development.

The differences in exposure to F. hepatica in Irish dairy herds have previously been shown to be strongly dependent on the soil type. Soils on farms that did not record fluke exposure were deep and shallow well-drained soils, while poorly drained soils were the main soil types on farms where herds recorded exposure to the parasite (Selemetas et al., 2014). Little evidence has been presented, however, regarding annual variations in exposure to F. hepatica in Irish dairy herds and the impact of annual weather variability on the risk of exposure. In that regard, the current study aimed to conduct a longitudinal study to examine trends in bulk milk seropositivity over time. The study also aimed to investigate associations between F. hepatica bulk milk seropositivity, two weather-related variables (soil temperature and rainfall), and additional risk factors.

2. Materials and methods

2.1. Sample population

A total of 28 herds were included in the study. Of these, 22 were commercial dairy farms, all members of the Dairy Management Information System ('DairyMIS') discussion group coordinated by Teagasc (Irish Agriculture and Food Development Authority). The remaining six were Teagasc research herds. All farms were located in the southernmost Munster region in Ireland. These herds were initially recruited in 2009 to a F. hepatica prevalence study (Bloemhoff et al., 2015). Sampling and data collection were subsequently extended for the purposes of the current research.

2.2. Sampling

Each study farm was requested to submit a monthly bulk tank milk (BTM) sample over the course of each lactation between March 2009 and December 2014. For logistical reasons, monthly samples were not available for spring and summer seasons in 2010. Samples were submitted to Teagasc by post in a standardised sampling

kit, which contained a sample bottle containing broad spectrum Microtab II milk preservative tablets (D&F Control Systems Inc., USA). All samples were received within 48 h of sampling. On receipt at the laboratory, BTM samples were dispensed into 2.5 mL micro tubes (Sarstedt, Germany), centrifuged at 20,000g for one minute, de-fatted, and the supernatant transferred to 1.5 mL microtubes (Sarstedt, Germany). De-fatted samples were frozen at -80° C until analysed.

2.3. Sample analysis

All samples were analysed using a commercially available ELISA kit as previously described by Selemetas et al. (2014). Briefly, the kit is based on the recombinant mutant Cathepsin L1 antigen (rmCLl) (Ildana Biotech, Dublin, Ireland; Collins et al., 2004) and has a documented sensitivity (Se) and specificity (Sp) of 98% (Bloemhoff et al., 2015). It has been previously shown that F. hepatica antibodies detected by this test decrease three to four months following treatment with a flukicide (Bloemhoff et al., 2015).

As per kit manufacturer's instructions, the level of antibodies present in each BTM sample was expressed as S/P%, calculated using the formula S/P% = (sample optical density (OD)/mean OD of positive control) x 100. Samples yielding S/P% values of >15 were classified as positive for exposure to F. hepatica.

2.4. Weather data

Daily rainfall (mm) and soil temperature at depths of 5 cm, 10 cm, and 15 cm below ground level (°C) were recorded by an automated weather measurement station (Campbell Scientific Ltd. Loughborough, UK) based at Teagasc Moorepark, Fermoy, Co Cork, Ireland. This weather measurement station was located within 30 miles of study farms. Daily rainfall measurements were summated to yield monthly values (mm), these values were subsequently averaged to yield seasonal (spring (January, February, March), summer (April, May, June), autumn (July, August, September), winter (October, November, December)) and annual (2009-2014) mean monthly rainfall. Soil temperatures were recorded every five minutes and subsequently averaged to yield seasonal and annual soil temperature values (°C). The 5 cm depth soil temperature reading was chosen for inclusion in the current data analysis as soil temperature is highly correlated with air temperature (Charlier et al., 2014) and this soil depth more closely reflects both fluke larvae and mollusc environments.

2.5. Farm and fluke management data

Farm and fluke management data were collected using a web-based survey tool (https://www.surveymonkey.com). Data collected included predominant dairy cow breed in each herd, farmer self-reported land classification, and whether or not an annual flukicide treatment was administered. Additionally, fluki-cide product(s) used in each year of the study, and how many times the product was administered in a particular year, was recorded. Participation in the study was non-incentivised other than reporting of results of monthly BTM testing in October each year to inform each farmer's annual F. hepatica control programme.

2.6. Herd classification

Each study herd was categorised by herd size (small (71-130 cows), medium (131-190 cows, large (>190 cows)), predominant breed in the herd (Holstein-Friesian (HF), Jersey cross (JEX), Norwegian Red cross (NRX)), and farmer self-reported land classification (dry (predominantly dry paddocks on the farm), occasionally waterlogged (some paddocks prone to becoming wet following

rainfall), wet (predominantly wet paddocks on the farm)). With regard to flukicide treatments, herds were classified as 'treated' or 'not treated' on the basis of whether at least a single treatment was administered in a particular year. The flukicide products administered were classified into five categories depending on the spectrum of activity of the product used and the number of times a product was administered in a particular year. Category 1 farms administered a single dose of a product active against mature F. hepatica alone i.e. albendazole, oxyclozanide, clorsulon. Category 2 farms employed a two dose programme using only products active against mature F. hepatica. Category 3 farms administered a single dose of a product active against both mature and immature F. hepatica i.e. triclabendazole, nitroxynil. Category 4 farms used a combination of flukicide products with differing activity spectra in a multi-dose programme. Finally, category 5 farms included those that did not administer a flukicide product in a particular year. Herd categorisations on the basis of flukicide treatment programme were assigned annually. Finally, each farm was classified as F. hepatica BTM antibody 'positive' or 'negative' on the basis of monthly BTM ELISA S/P% values.

2.7. Data analysis

Data collation and descriptive analyses were completed in Excel (MS Office 2010). Normality of the data was assessed using ladder of powers histogram, which showed visually different power transformations promoting symmetry of the data, normality of residuals was assessed using normal probability plots both constructed in Stata version 11 (StataCorp, USA). Graphical representations, Mann-Whitney test, and t-test were completed in GraphPad Prism 6 (GraphPad Software Inc., 2016). Additional statistical analyses, including Pearson's chi-squared, generalised estimating equations (GEE) and regression analyses, were completed using Stata version 11 (StataCorp, USA).

In order to complete the most comprehensive study of the variability in exposure to F. hepatica over time, both annual (2009-2014) and seasonal (spring, summer, autumn, and winter of each study year) datasets were compiled. Additional independent variables investigated included categorical seasonal soil temperature (low (<5 °C), borderline low (>5-8°C), moderate (>8-12°C), boderline high (>12-15 °C), high (>15-18°C)), annual soil temperature (<10°C, >10 °C), seasonal rainfall (low (41-80 mm), moderate (>80-120 mm), high (>120 mm)), and annual rainfall (low (850-1000 mm), moderate (>1000-1150 mm), high (>1150 mm)). Categorisation for each variable was defined to best represent the spread of data recorded over the duration of the study.

Study herd demographics, number of samples per farm, and monthly S/P% values per farm were tabulated, as well as annual and seasonal S/P%, rainfall and soil temperature summary statistics. The Mann-Whitney test used to determine significant differences between monthly S/P% across each year of the study using 2009 as the reference year, as in this year test results were based on farmers' own treatment program. Similarly, a t-test was used to examine differences in normally distributed rainfall and soil temperature, again using 2009 as the reference year. Boxplots and line graphs of annual and monthly BTM S/P%, mean monthly rainfall, and mean soil temperature, were constructed to allow clearer visualisation of these data. Finally, the frequency of annual use of anthelmintics (flukicides) was computed and plotted.

2.7.1. Generalised estimating equations (GEE)

A univariable generalised estimating equation (GEE) analysis, was used to investigate seasonal differences in F. hepatica S/P% values across each year of the study, using spring 2009 as the reference variable. Herd was included as a repeated measure, and a Gaus-

sian distribution, identity link, and exchangeable correlation used to generate GEE coefficients. Coefficients were subsequently plotted to allow visual assessment of seasonal and annual trends in F. hepatica BTM antibody over the entire duration of the study.

For the purposes of constructing the multivariable GEE model, two-way associations between categorical independent variables were initially identified by Person's chi-squared. Variables recording P values < 0.15 in this analysis were investigated as potential interacting variables in the multivariable GEE model. BTM results were not included in this analysis due to the repeated nature of the data. Herd F. hepatica BTM antibody status, both categorical (positive vs. negative) and continuous (ELISA readings), was the dependent variable in all multivariable GEE analyses. Independent variables included time (year or season), weather (rainfall, soil temperature) flukicide treatment and herd management data. In final models, herd was included as a repeated measure, a Gaussian distribution and identity link function was used for continuous data, and a binominal distribution and logit link function assumed for categorical data. An exchangeable correlation was used for both continuous and categorical GEE analyses. A manual backwards elimination with a forward step was used to build final models with variables recording P values of <0.05 maintained. Second level interactions between independent variables were examined for significance and retained if significant at the 5% level.

2.7.2. Logistic regression

Multivariable logistic regression was performed to examine associations between flukicide treatment (dependent variable) and independent variables identified as significant in the Pearson chi-squared analysis conducted for GEE. Again, a manual backwards elimination with a forward step was used to build the final model with variables recording P values of <0.05 maintained. Second level interactions between independent variables were investigated and included in the final model at a P value <0.05.

3. Results

3.1. Descriptive analysis

A total of 1337 samples from 28 herds were collected and analysed over the entire study period (2009-2014). The mean number of samples per farm per year was 8 (range 0-11 samples). The numbers of samples submitted by each study farm and their mean annual S/P% value each year over the study period are included in Table 1. The majority of herds (n = 27) reported having a spring calving system i.e. the majority of cows calved between February and May of any year. A single herd operated an autumn-calving system during the first four years of study, and subsequently reduced to 60% of the herd calving in autumn for the last two years of the study. All study herds were specialist dairy enterprises with no additional livestock species (e.g. fattening cattle or sheep) present on the farm. All cows had full access to pasture from at least March until November and all were housed in December and January of each year. Housing periods extended to varying degrees into November and February of any particular year depending on grass availability and weather conditions.

The mean herd size across the 28 study herds was 161 cows (range 71-310 cows). Of these, 8 herds contained <100 cows and 13 contained >150 cows. The predominant breed in the majority of herds was Holstein-Friesian (78.6%). The majority of farmers (60.7%) reported having dry land across their farms, the remainder stating their farms were either 'wet' or occasionally waterlogged. More detailed herd demographic data are included in Table 2.

Number of samples collected from study herds in each year of the study and their mean S/P% values.

Herd 2009 2010 2011 2012 2013 2014 Total per farm n Mean S/ 2009 to 2014

n Mean S/P% n Mean S/P% n Mean S/P% n Mean S/P% n Mean S/P% n Mean S/P%

1 10 0.8 5 4.6 9 0 9 0.1 8 0.1 8 -0.9 49 0.54

2 10 126.8 6 134.5 10 101.2 11 58.2 9 54 10 17.9 56 78.46

3 10 36.9 4 93 10 45.3 9 19.7 8 25.9 9 7 50 32.84

4 10 1.3 5 9.4 9 0.6 9 3.6 9 1 8 -0.3 50 2.16

5 10 74.8 4 27.8 8 2.8 9 21.6 9 4.2 8 0.6 48 23.36

6 10 135.8 5 156.4 9 98.6 9 94.4 10 93 8 53.8 51 102.66

7 10 70.2 6 138.7 10 95.2 9 49.7 9 46.8 9 9.4 53 64.91

8 10 1.4 4 3.8 8 6 9 16.4 9 33 9 25.8 49 15.31

9 9 2.8 5 8.2 9 1.3 9 -0.8 9 0.8 7 -0.1 48 1.63

10 10 118.3 5 143.4 10 38.7 10 27.8 9 9.9 8 0.3 52 51.05

11 10 75.7 4 123.8 9 82.1 9 91.1 10 71.7 9 40.7 51 76.26

12 10 13.5 6 9.3 10 3.2 10 14.8 9 9.2 8 -0.8 53 8.45

13 10 91.8 5 133 6 71.7 6 49 3 13.3 0 30 78.24

14 10 103.7 5 129.2 9 54.2 10 23.8 9 6.1 8 -0.6 51 48.21

15 9 9.8 5 4.2 9 3.7 9 17.3 8 17.1 8 20 48 12.45

16 10 2.8 5 0.8 9 0.2 10 0.2 10 1.3 9 0.8 53 1.01

17 10 28.5 4 35.5 9 31.4 9 18 7 47.4 7 15.6 46 28.59

18 9 123.8 5 135 11 95.2 9 78 10 59.9 8 27.6 52 83.78

19 10 134.3 6 94 10 31.1 11 20.5 10 7.5 9 7.9 56 46.19

20 10 9.5 6 25.8 11 0.7 9 15 8 11.1 8 4.6 52 9.97

21 9 14.4 5 38.2 9 34.1 9 22.4 7 36 4 13.8 43 26.45

22 10 101.1 4 128.3 9 76.7 8 36.1 7 55.6 6 12.2 44 67.33

25 10 93.5 6 89.8 11 86.8 11 53.1 9 75.3 10 38.4 57 71.5

26 10 5.1 4 8.8 10 31.4 8 0.3 4 31.3 1 28 37 14.91

27 10 26.4 3 57 2 74 6 0.3 10 48.7 8 28.6 39 33.36

28 10 24.8 6 73.7 10 92.7 11 75.9 9 70.4 10 10.1 56 56.96

29 10 29.5 5 69.4 7 79 8 64.8 8 84.3 6 65.3 44 63.26

32 0 0 0 0 10 36.7 9 0.8 19 19.73

Mean S/P% (all farms) 54.2 70.7 45.5 32.8 34.7 15.2

Total samples (n) 266 133 243 246 237 212 1337 39.98

Table 2

Study herd demographics including herd size and predominant breed.

Variable

Category Herd size (cows) 71-130 131-190 >190

Predominant Breed HFa JEXb NRXc

Calving period Spring-calving Non-spring-calving

Enterprise Dairy only

Farmer land classification"1 Dry

Occasionally waterlogged Wet

Proportion (%)

35.7% 32.1% 32.1%

78.0% 14.3% 7.1%

96.4% 3.6%

100.0%

60.7% 14.0% 25.0%

a Holstein-Friesian. b Jersey crossbreed. c Norwegian Red crossbreed.

d Farmerself-reported classification of predominant land type on each study farm.

from 2010 (Table 3). The highest levels of exposure to F. hepatica were recorded in winter. Examination of annual trends highlighted that 2010 recorded the highest median F. hepatica S/P% values and 2014, the lowest. S/P% values in 2010, 2012, and 2014 were significantly different from 2009 (Table 3), being higher in 2010 and lower in 2012 and 2014. There was no visible trend in annual rainfall and soil temperature, although 2013 tended to be colder than other years. A boxplot of annual and monthly S/P% ELISA values in addition to mean soil temperature and rainfall are shown in Fig. 1.

3.3. Flukicide treatment of milking herd

Flukicide treatment protocols for all six years of the study were available for 25 herds. A single farm did not provide any treatment data and two farms did not have records of 2009 treatment protocols. In all, 134 herd treatments were administered across 27 farms over the six years of study. The majority of farmers administered at least a single treatment in every year of the study (n = 18; 66.7%). The number of farmers using each flukicide product is outlined in Fig. 2a with the frequency of herds in each product treatment category (categories 1-5) by year included in Fig. 2b. The median S/P% of farms that used or did not use treatment across the entire duration of the study was 45 and 28 S/P%, respectively, and the difference was significantly different (P =0.002).

3.2. F. hepatica S/P%, rainfall and soil temperature values

Ladder of powers histograms revealed a normal distribution for weather variables, S/P% values presented a positive skewness and normal probability plot ensured the normality of residuals.

Annual and seasonal BTM seropositivity interquartile range (IQR) and median S/P% values are included in Table 3 and highlighted a general decrease in exposure to F. hepatica in study herds

3.4. GEE and logistic regression analyses

Univariable GEE analysis of BTM ELISA status results over time are outlined in Supplementary Table S1 and highlighted significant differences between seasons. A graphical representation of the difference in S/P% values in each season compared to spring 2009 is outlined in Fig. 3 and exhibits an overall downward trend in BTM result over the course of the study. Univariable associations

Significant differences between S/P%, mean monthly rainfall, and soil temperature across study year (2009 used as reference year) and season (winter used as reference season).

Variable S/P ratio Mean monthly rainfall (mm) Total annual Soil Temperature (°C)

Median IQR Range Mean SD CI (95%) Mean SD CI (95%)

2009 25.50 92.75 0,194 107.80 66.32 101.00,114.50 1293.6 11.03 5.26 10.50, 11.57

2010 66.00a 112.00 0,179 72.40 34.89 68.84, 75.96 868.8 10.23 6.14 9.50, 10.85

2011 30.50 84.00 0, 135 71.30 34.99 67.73, 74.87 855.6 11.2 4.53 10.74, 11.66

2012 13.00a 64.00 -8,125 91.45 57.36 85.60, 97.30 1097.4 10.83 4.49 10.37, 11.28

2013 29.00 52.50 0,123 78.83 42.85 74.46, 83.19 945.9 6.65b 5.16 6.12, 7.18

2014 6.00a 21.00 -2,93 103.20 57.35 97.39,109.10 1238.8 11.38 5.48 10.82, 11.93

Season

Spring 15.50 70.25 -3,159 85.73 58.97 80.82, 90.63 n/a 4.97 2.44 4.76, 5.17

Summer 10.00 50.00 -2,147 73.55 39.56 70.26, 76.84 n/a 13.25 4.18 12.90, 13.6

Autumn 14.00 59.00 -8,176 75.53 46.48 71.66, 79.39 n/a 15.37 3.51 15.07, 15.66

Winter 48.00 92.25 -1,194 115.20 51.54 110.90,119.50 n/a 7.29 3.39 7.01, 7.57

a Significantly different at P< 0.01 using Mann-Whitney test. b Tendancy toward significance at P >0.05 <0.10 using t-test.

Fig. 1. Boxplot of annual (a) and monthly (b) mean S/P% across all study farms. Superimposed line graphs represent mean soil temperature and mean monthly rainfall in each year and month of the study.

Fig. 2. Frequency of use of different flukicide active ingredients during each year of the study (a) and frequency of use based on activity of each active ingredient (categories 1-5) (b).

OXY = oxyclozanide, TCBZ = triclabendazole, ALB = albendazole, NIT = nitroxynil, CLOR=clorsulon, CLOS = closantel

Category 1: a single dose of a product active against mature liver fluke only

Category 2: a two-dose programme using only products active against mature liver fluke

Category 3: a single dose of a product active against both mature and immature liver fluke

Category 4: a combination of flukicide products with differing activity spectra in a multi-dose programme

Category 5: no flukicide product administered.

between additional independent variables are outlined in Supplementary Table S2. Multivariable GEE analyses of both annual and seasonal data are outlined in Tables 4 and 5 (categorical data) and Supplementary Tables S3 and S4 (continuous data). Annual and seasonal analysis of the association between rainfall and F. hepatica BTM yielded contradictory results. Annual rainfall >1000 mm was associated with a lower risk of F. hepatica (OR<0.5; P<0.05), while seasonal rainfall over 120 mm (winter 2009, winter 2013 and

spring 2014) increased the risk (OR= 1.37; P =0.059). Mean annual soil temperature was not significantly associated with F. hepatica status while, in general, seasonal temperatures less than 8 °C and greater than 12 °C reduced the risk of being BTM antibody positive (Table 5). More specifically, a soil temperature between 8 °C and 12 ° C increased the odds of being F. hepatica BTM positive by a factor of two compared to lower soil temperatures (P =0.001). An exception to this was the increased tendency towards being pos-

Fig. 3. Line graph representing the difference inS/P% across seasons in each study year using a univariate GEE. Spring 2009 was used as the reference season for GEE analysis. Atrend line (---) has been included for ease of interpretation. Sp = spring; Su = summer; A = autumn; W=winter.

Table 4

Multivariable GEE of annual F. hepatica BTM antibody status (categorical data).

Dependent variable Independent variable

Odds ratio

Confidence interval

P- value

Model (P-value)

F. hepatica status (Positive vs. Negative) Annual rainfall

High vs. moderate 0.33 0.12, 0.91 0.003

High vs. low 0.47 0.24, 0.95 0.036

Herd size

Medium vs. large 0.19 0.04, 0.88 0.033

Farmer land classification

Wet vs. dry 3.56 0.78,16.19 0.100

Year 0.79 0.62,1.01 0.059

Annual rainfall, herd size, farmer land classification, year (P=0.032)

Annual rainfall: low: 850-1000 mm; moderate: >1000-1150mm; high: >1150mm.

Herd size: small = 71-130 cows; medium = 131-190 cows; large = >190cows.

Only significant associations (P < 0.05) are reported. Tendencies are reported at P > 0.05 < 0.10.

itive (OR= 1.47; P = 0.067) at soil temperatures between 15 °C and 18 °C compared to a soil temperature ofbetween 12 °C and 15 °C.

Additional findings in multivariable GEE included an increased S/P% value (20.20 S/P%, P =0.027 (Supplementary Table S3)) in herds that treated with a flukicide, and a decreased likelihood (OR = 0.50) of being BTM positive in herds that included a flukicide active against mature and immature fluke (categories 3 and 4) in their dosing regimen (Table 4). Finally, depending on the model and time variable used, wetter farms were either more likely (OR= 3.88; P =0.045 (Table 5)) or tended to be more likely (3.56; P =0.10 (Table 4)) to record a BTM antibody positive result.

Compared to 2009, multivariable logistic regression of whether a flukicide treatment was administered or not, indicated that increased levels of dosing were practiced over subsequent years of the study. Larger herds (>190 cows) were most likely to administer a treatment compared to other herd sizes (Table 6) and a predominant breed in a given herd was also an influencing factor. PredominantlyJEX herds were less likely (OR = 0.37; P = 0.004) to administer a flukicide than HF herds. Finally, herds administering a treatment were almost twice as likely to be F. hepatica BTM antibody positive (OR =1.74; P = 0.040).

4. Discussion

The aim of this study was to examine trends in exposure to F. hepatica over a six year period in a sub-set of Irish dairy herds, while also investigating the relationship between soil temperature, rainfall, flukicide treatment and F. hepatica status. Annual

weather variations were found to have relatively minor impact on exposure to F. hepatica, with seasonal weather patterns proving to be more influential. Additionally, the use of a flukicide treatment active against both mature and immature stages of F. hepatica was very much associated with reduction in bulk milk antibody status of herds compared with treatments active against mature F. hepatica only.

Recently, the application of ELISA methodologies to detection of F. hepatica infections has been shown to improve the sensitivity of diagnosis over more conventional corprological techniques (Charlier et al., 2007, 2008). In turn, serum ELISA has been shown to yield greater sensitivity than bulk milk antibody ELISA (Duscher et al., 2011). The majority of F. hepatica infected herds will nonetheless be detected using BTM samples (Bloemhoff et al., 2015), and the practicality of collecting BTM and determination of whole herd status on the basis of a single pooled sample, makes BTM antibody ELISA an attractive alternative to serum ELISA (Salimi-Bejestani et al., 2007; Charlier et al., 2008). In the case of bovine leukaemia virus, it has been shown that repeated measurements improve the diagnostic sensitivity of a bovine leukaemia virus BTM antibody ELISA (Nekouei et al., 2015). Use of repeated monthly samples in the current study, similarly, should provide an effective method of determining whole herd F. hepatica status, and the noteworthy compliance rates achieved in sample submission over a six year period highlights the marked benefits of using BTM samples as a research tool. It should be noted that although compliance rates for sample submission appear lower in 2010, this was due to a suspension of the project for logistical reasons. The majority of

Multivariable GEE of seasonal F. hepatica BTM antibody status (categorical data).

Dependent variable Odds ratio Confidence interval P-value Model

Independent variable (95%) (P-value)

F. hepatica status (Positive vs. Negative)

Seasonal soil temperature

Moderate vs. borderline low 1.95 1.32,2.88 0.001

Moderate vs. borderline high 2.86 1.79,4.55 0.000

Moderate vs. high 1.92 1.24,2.94 0.003

Borderline high vs. low 0.45 0.26,0.80 0.006

Borderline high vs. borderline low 0.69 0.47,1.02 0.062

High vs. borderline high Seasonal rainfall High vs. low Flukicide Type Category 1 vs. category 5 Category 2 vs. category 5 1.47 1.37 0.97,2.23 0.99,1.91 0.067 0.059 Soil temperature, rainfall, farmer land classification, flukicide

2.11 2.19 1.31,3.41 1.26,3.79 0.002 0.005 treatment, product type, season (P< 0.001)

Category 3 vs. category 1 0.43 0.26,0.74 0.002

Category 4 vs. category 1 0.47 0.29,0.76 0.002

Category 3 vs. category 2 0.42 0.22,0.79 0.007

Category 4 vs. category 2 0.46 0.26,0.79 0.005

Farmer land classification

Wet vs. dry 3.88 1.03,14.58 0.045

Season 0.94 0.92,0.97 <0.001

Seasonal soil temperature: Low = <5 0C; borderline low = >5-8 0C; moderate = >8-12 0C; boderline high = >12-15 0C; high = >15-18 0C.

Flukicide categories: Category 1: a single dose of a product active against mature liver fluke only: Category 2: a two-dose programme using only products active against mature liver fluke: Category 3 a single dose of a product active against both mature and immature liver fluke: Category 4: a combination of flukicide products with differing activity spectra in a multi-dose programme: Category 5: no flukicide product administered. Only significant associations (P < 0.05) are reported. Tendencies are reported at P > 0.05 < 0.10.

Seasonal rainfall: low = 41-80mm; moderate = >80-120mm; high = >120mm.

Table 6

Multivariable logistic regression of flukicide treatment (dependent variable) across year, herd size, breed and F. hepatica status (independent variables).

Dependent variable Odds ratio Confidence interval P- value Model

Independent variable (95%) (P-value)

Flukicide treatment vs. no treatment

2010 vs. 2009 2.73 1.27,5.90 0.011

2011 vs. 2009 2.86 1.41,5.80 0.004

2012 vs. 2009 6.96 2.94,16.43 <0.001

2013 vs. 2009 4.29 2.01,9.17 <0.001

2014 vs. 2009 2012 vs. 2010 2012 vs. 2011 Herd size Large vs. medium 3.47 2.43 2.38 6.68 1.67,7.20 0.91,6.54 0.16,1.05 3.18,14.03 0.001 0.078 0.064 <0.001 Year, herd size, breed, F. hepatica bulk milk antibody status (P <0.001)

Large vs. small 2.90 1.36,6.20 0.006

Medium vs. small 0.43 0.25,0.75 0.003

Jersey cross vs. Holstein Friesian 0.37 0.19,0.72 0.004

F. hepatica status

Positive vs. negative 1.74 1.02,2.94 0.040

Herd size: small (71-130 cows); medium (131-190 cows); large (>190 cows).

Only significant associations (P < 0.05) are reported. Tendencies are reported at P > 0.05 < 0.10.

samples submitted in 2010 were submitted in the latter half of the year, which accounts for the significantly higher S/P% median value recorded in that year (Bloemhoff et al., 2015).

A disadvantage of using a serological technique to monitor fasciolosis is that it indicates only the presence (or absence) of antibodies. This does not allow differentiation between current/active and historical (e.g. post-treatment) exposure (Sekiya et al., 2013). The vast majority (>80%) of Irish dairy farmers have previously been shown to treat cows for F. hepatica in winter only, however, when cows in spring-calving herds are housed and not lactating (Bloemhoff et al., 2014). The BTM ELISA used in the current study can detect seasonal changes in exposure to F. hepatica (Bloemhoff et al., 2015) when applied to dairy herds that are predominantly spring-calving and this minimises the disadvantages of using a serological technique in Irish dairy herds.

The seasonal variation in F. hepatica infection highlighted in the current study, indicating higher F. hepatica exposure in autumn

and winter, is similar to that previously reported (Ollerenshaw and Smith, 1969; Relf et al., 2011; Bloemhoff et al., 2015). Conflicting evidence is presented in the literature, however, with regard to the impact of weather conditions on the prevalence of F. hepatica, and its intermediate snail host. Ollerenshaw and Smith (1969), Yilma and Malone (1998), Morley and Lewis (2008), Cruz-Mendoza et al. (2011), Relf et al. (2011), Charlier et al. (2014), Selemetas et al. (2014) an Ducheyne et al. (2015) have all reported significant impacts of weather on the prevalence of F. hepatica. Others, however, report only very minor or weak associations (Bossaert et al., 1999; McCann et al., 2010; Bennema et al., 2011; Kuerpick et al., 2013; Novobilsky et al., 2015). This lack of consistency across studies may be as a result of differences in the time variables used (e.g. seasonal, annual), the statistical models applied, and the jurisdictions in which studies were undertaken. Some studies included weather patterns alone (Ollerenshaw and Smith, 1969; Yilma and Malone, 1998; Bossaert et al., 1999; Morley and Lewis,

2008; McCann et al., 2010; Bennema et al., 2011; Relf et al., 2011; Kuerpick et al., 2013; Selemetas et al., 2014; Ducheyne et al., 2015; Novobilsky et al., 2015), while others also include on-farm management practices such as treatment regimens (Cruz-Mendoza et al., 2011). In the current study, we aimed to provide a comprehensive analysis of six years of monthly data, which, as far as the authors are aware, has not been previously been reported. This afforded a unique opportunity to study both annual and seasonal variations in F. hepatica exposure, their relationship with variations in weather, and finally, the relationship with flukicide treatment practices.

Annual and seasonal analyses of rainfall and exposure to F. hepatica yielded conflicting results. Unexpectedly, years recording annual rainfall of over 1150 mm were associated with an apparent lower risk of F. hepatica (0R=0.47) compared to years recording <1000 mm of rain. Analysis of seasonal data yielded a more typical outcome, i.e. higher seasonal rainfall associated with higher levels ofF. hepatica (OR = 1.37). As mentioned previously, Irish dairy farmers typically treat adult cows for F. hepatica at housing. This allows detection of seasonal variations in F. hepatica largely independent of interference from flukicide treatments. In the case of annual data, however, farmer treatment interventions become important. Anec-dotally, farmers often assume a higher risk of F hepatica infection in a year that they perceive as 'wet'. Levels of perceived risk are known to influence behaviour (Sayers et al., 2013) and it is likely that higher annual rainfall may lead to increased levels of treatment in certain years, manifesting as an apparent reduced risk of fluke in wetter years. This was not highlighted in the logistic regression analysis, however, which examined influences of rainfall on the administration of flukicides. Further investigations are required to examine the relationship between perceived farmer risk and actual weather patterns to further elucidate our findings.

The possibility that an additional factor reduces the risk of exposure to F. hepatica in wetter years cannot be ruled out. In that regard, Rapsch et al. (2008) have reported an important decrease in relative risk of infection at a monthly rainfall threshold of 210 mm whereby G. truncatula and the free-living stages of F. hepatica ('wash effect'). The mean monthly rainfall in the current study was 87.5 mm which may support a lower threshold for the spread of F. hepatica free living stages and the intermediate host, also suggested previously by Bennema et al. (2011), although our seasonal analysis would not support this hypothesis. These data highlight the difficulties in interpreting the parasite risk related to variations in weather, especially when external interventions (treatments) are also potentially influenced by annual weather conditions. Although significant interactions between weather variables and treatment variables were not detected in GEE models, it is both possible and plausible that farmer interventions present a greater influence on annual F. hepatica herd status than annual weather variations. This is supported by the fact that, although the mean annual temperature, and mean monthly rainfall in 2014, were marginally but not significantly higher (11.38 °C vs. 11.03 °C; P>0.10) and marginally but not significantly lower (103.2 mm vs. 107.8 mm; P>0.1) than 2009, respectively, F. hepatica BTM antibody S/P ratio was significantly lower in 2014 (P< 0.01) and the likelihood of administering a flukicide greater (OR = 3.47). It has been reported that the inclusion of management factors can substantially improve spatial distribution models for F. hepatica in temperate climatic zones (Bennema et al., 2011). Certainly the current study would support the addition of flukicide treatments, at a minimum, in F. hepatica modelling studies.

Soil temperature did not appear to influence F. hepatica BTM to the same degree as rainfall, certainly based on annual variations, where no significant association was detected. This is most likely contributed to by the minimal variability of mean annual temperatures across the study compared to seasonal temperature variations. The seasonal temperature influences on F. hepatica

BTM antibody levels would initially appear broadly in line with the development requirements of the intermediate stages of the parasite i.e. temperatures >10°C are optimal (Taylor et al., 2007; Borgsteede, 2011). Unusually, however, a temperature category of less than 12 °C to 15 °C was associated with a significant reduction in BTM results. This may relate to the time taken for infection to build on pastures once cows are turned out. The lifecycle of F. hepatica is relatively long compared to nematode lifecycles, which can be as short as 18 days (Urquhart et al., 1996). This leads to encysted metacercariae increasing on pastures later in the summer season and into autumn which may explain a mean soil temperature of between 8 °C and 12°C appearing most favourable for F. hepatica infection. Rondelaud et al. (2013) and Claxton et al. (1999) found that cercarial production is increases with increasing variability in daily temperature, although many of these cercariae are not viable. Based on the results of the current study, it is unlikely that the variability in temperature in Ireland will dramatically impact on the levels of F. hepatica in Irish dairy cows, with seasonal rainfall being a better predictor of the risk of infection.

Based on logistic regression results for flukicide treatment, farmers were more likely to administer a treatment in the latter years of the study. In that time period, comprehensive parasite awareness programs, including F. hepatica, were initiated by both Teagasc and Animal Health Ireland (AHI, 2011). Additionally, in March 2010, restrictions on the use of certain flukicides in dairy herds were introduced (Bloemhoff et al., 2014), which are also likely to have made farmers more aware of flukicide treatments and their correct usage. This may also explain why larger herds in this study were more likely to administer a flukicide treatment compared to medium and small herds. Larger dairy herds in Ireland (Sayers et al., 2013), the United States (Hoe and Ruegg, 2006), and Canada (Young et al., 2010) have previously been shown to be more likely to engage in health schemes and good farm management. They could be considered the 'first responders' to awareness campaigns and may indicate a trend towards more effective fluke control in Ireland. Our study also reported that small herd categories were more likely than medium-sized herds to administer a flukicide. The majority (80%) of herds classified as 'small' in the current study contained fewer than 100 cows. Small herds lend themselves to closer day-to-day monitoring of animals and may explain the greater likelihood of smaller herds dosing compared to medium-sized herds. This highlights the importance of targeting educational messages to farmers based on farm demographics, with the owners/managers of medium-sized herds in this study lagging behind their counterparts in best practice.

As regards type of flukicide treatment administered, protocols incorporating an ingredient active against all stages of F. hepatica (category 3 or 4) increased throughout the years of study. Since 2012, triclabendazole is the only flukicide licensed in dairy cows in Ireland (HPRA, 2013) active against all stages of F. hepatica (category 3) within the final host (Fairweather and Boray, 1999). Use of a 'two-dose' regimen with a product active against mature F. hepatica only (category 2) likewise did not prove equivalent to treatment with triclabendazole, which was more effective. This is most likely due to the fact that Irish dairy herds operate lengthy grazing seasons incorporating a 60 to 90 day dry period (Dillon et al., 1995). The majority of flukicide treatments are administered during this housed dry period (Bloemhoff et al., 2014). If using an ingredient active against mature F. hepatica only, treatment is administered at the beginning and end of the housing period (AHI, 2013). As the dry period may be as short of 60 days, particularly in the south of Ireland where the current study was conducted, it is unlikely that all encysted metacercariae ingested just prior to housing will have matured to such a degree that a 'two-dose' treatment with an adulticide will be optimally effective (Fairweather and Boray, 1999).

It is interesting to note that herds which did not administer a flukicide treatment also recorded a superior F. hepatica status to those administering category 1 or 2 flukicide treatments. Herds that did not administer flukicides recorded the lowest S/P% values amongst study herds and can essentially be regarded as unin-fected herds. Unlike category 1 and 2 herds, category 3 and 4 (i.e. triclabendazole-treated) herds did not record significantly different results to untreated herds. Category 3 treatment regimens could therefore be considered effective in reducing the burden of F. hepatica to the level of uninfected herds. There is a case, therefore, for using an ingredient active against all stages of F. hepatica alone in a single dose regimen to control F. hepatica in dairy herds. Generation of flukicide resistance, however, must always be a concern when promoting the use of a single active ingredient. The incidence of triclabendazole resistance in livestock is a major threat to global livestock production (Kelley et al., 2016) although reported resistance to triclabendazole in Ireland appears to be restricted to sheep flocks (Fairweather and Boray, 1999; Mooney et al., 2009; Hanna et al., 2015). Internationally, however, cases of triclaben-dazole resistance have been reported in cattle (Moll et al., 2000; Olaechea et al., 2011; Ortiz et al., 2013; Brockwell et al., 2014; Elliot et al., 2015) and care must be taken in promoting use of a single active ingredient.

This study highlighted that a proportion of study herds would not appear to require treatment, once continuous monitoring using an appropriate diagnostic tool is implemented for surveillance purposes. Additionally, regardless of the flukicide used, almost all study herds recorded year-on-year improvements in F. hepatica S/P% values. We would suggest, therefore, that continuous monitoring of F. hepatica status using an ELISA method capable of differentiating seasonal changes in herd status, should be promoted as a decision-support tool for designing a flukicide treatment regimen. It may be possible to use a category 3 or 4 flukicide treatment protocol in the initial years of a control programme. Once levels of F. hepatica have significantly reduced, a category 2 protocol could be implemented on a rotational basis, complimented with a continuous monitoring programme. Access to monthly F. hepatica results was instrumental in changing the behaviour of farmers in the current study and highlights the benefits of a holistic approach (monitoring and treatment) to yield improved and appropriate F. hepatica control.

An unusual finding was the association between a lower likelihood of flukicide treatment in herds where JEX cows were the predominant breed. This may be a function of the low number of such herds in the study (n = 4), although six years of data were available for these herds which should improve the power of the findings. This finding may relate to the overall efficiency of JEX compared to HF cows, which are a larger and less efficient animal (Prendiville et al., 2009). If farmers base a treatment programme on animal performance or body condition score alone, the lower maintenance JEX cows (Kristensen et al., 2014) may not have been treated. This finding requires further investigation to establish whether some additional protective factor exists in these crossbred cows.

Weaknesses of the current study include the fact that weather data were obtained from a single centralised weather station. Local weather variations were, therefore, not accounted for. Having said that, all farms were located within a 30 mile radius of the weather station used and should be sufficient for the analysis undertaken. Additionally, the use of farm self-reported land classification has the potential to introduce a degree of variability, as farmer perceptions of 'wet' and 'dry' land will be subjective. Detailed soil maps have been used by Bloemhoff et al. (2014) and Selemetas et al. (2014) for the purposes of assessing the association between soil type and F. hepatica infection. These maps, however, are available at a resolution of a 10 km grid which does not account for individual paddock variation within farms. This, coupled to the fact that farm-

ers' own perceptions can be important determinants of the action they take in regard to their farms (Van Duinen et al., 2015), justifies the use of self-reported land-classification in the current study. Finally, due to the longitudinal nature of the study, only a subset of dairy farms in the Munster region of Ireland was incorporated into the study which may introduce bias. Irish dairy farmers are encouraged to engage in discussion groups (Teagasc, 2016), however, and the results presented here highlight the progress that can be made within a farmer group once awareness of a disease and appropriate control measures are improved.

5. Conclusion

This study highlights the importance of examining and comparing both seasonal and annual data with regard to environmental and management practices, as F. hepatica outcomes can vary significantly. International models are predicting increases in F. hepatica exposure due to global climate change, and the current study highlights the progress that can be made through awareness campaigns and continuous monitoring programmes. Evidence-based control of F. hepatica in study herds achieved lower BTM milk seropositivity regardless of weather patterns or changes. Further research is now required to design sustainable flukicide treatment programmes that will minimise the possibility of promoting anthelmintic resistance.

Acknowledgements

Funding: This work was supported by the Irish Dairy Levy [grant number not applicable]. The authors wish to thank all participating farmers for timely submission of samples and also are very grateful to AGRIC, Teagasc, Moorepark for supply of weather data.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016Zj.prevetmed.2016. 09.024.

References

Animal Health Ireland (AHI), 2013. A Guide to Parasite Control at Housing, Parasite Control Leaflet, https://online.flippingbook.com/view/941166/7/z Last access: 12/07/2016.

Animal Health Ireland (AHI), 2011. Liver Fluke-the Facts, https://online.

flippingbook.com/view/392838/ Last access: 11/07/2016. Bennema, S.C., Ducheyne, E., Vercruysse, J., Claerebout, E., Hendrickx, G., Charlier, J., 2011. Relative importance of management, meteorological and environmental factors in the spatial distribution of Fasciola hepatica in dairy cattle in a temperate climate zone. Int. J. Parasitol. 41, 225-233. Bloemhoff, Y., Danaher, M., Forbes, A., Morgan, E., Mulcahy, G., Power, C., Sayers, R., 2014. Parasite control practices on pasture-based dairy farms in the Republic of Ireland. Vet. Parasitol. 204,352-363. Bloemhoff, Y., Forbes, A., Danaher, M., Good, B., Morgan, E., Mulcahy, G., Sekiya, M., Sayers, R., 2015. Determining the prevalence and seasonality of Fasciola hepatica in pasture-based dairy herds in Ireland using a bulk tank milk ELISA. Irish Vet. J. 68,1-10.

Borgsteede, F.H.M., 2011. Diseases of dairy animals, parasites, internal: liver flukes. In: Fuquay, J.W. (Ed.), Encyclopedia of Dairy Sciences., 2nd ed. Academic, San Diego, 264 pp.

Bossaert, K., Lonneux, J.F., Godeau, J.M., Peeters, J., Losson, B., 1999. Serological and biochemical follow-ip in cattle naturally infected with Fasciola hepatica, and comparison with a climate model for predicting risk of fasciolosis. Vet. Res. 30, 615-628.

Brockwell, Y.M., Elliott, T.P., Anderson, G.R., Stanton, R., Spithill, T.W., Sangster, N.C., 2014. Confirmation of Fasciola hepatica resistant to triclabendazole in naturally infected Australian beef and dairy cattle. Int. J. Parasitol Drugs Drug Resist. 4, 48-54.

Charlier, J., Duchateau, L., Claerebout, E., Williams, D., Vercruysse, J., 2007.

Associations between anti-Fasciola hepatica antibody levels in bulk-tank milk samples and production parameters in dairy herds. Prev. Vet. Med. 78, 57-66.

Charlier, J., Meulemeester, L., Claerebout, E., Williams, D., Vercruysse, J., 2008. Qualitative and quantitative evaluation of coprological and serological techniques forthe diagnosis of fasciolosis in cattle. Vet. Parasitol. 153, 44-51.

Charlier, J., Vercruysse, J., Morgan, E., van Dijk, J., Williams, D.J.L., 2014. Recent advances in the diagnosis, impact on production and prediction of Fasciola hepatica in cattle. Parasitology 141,326-335.

Claxton, J.R., Sutherst, J., Ortiz, P., Clarkson, M.J., 1999. The effect of cyclic

temperatures on the growth of Fasciola hepatica and Lymnaea viatrix. Vet. J. 157,166-171.

Collins, P.R., Stack, C.M., O'Neill, S.M., Doyle, S., Ryan, T., Brennan, G.P., 2004. Cathepsin L1, the major protease involved in liver fluke (Fasciola hepatica) virulence: propeptide cleavage sites and autoactivation of the zymogen secreted from gastrodermal cells. J. Biol. Chem. 279,17038-17046.

Cruz-Mendoza, I., Quiroz-Romero, H., Correa, D., Gomez-Espinoza, G., 2011. Transmission dynamics of Fasciola hepatica in the Plateau Region of Mexico. Effect of weather and treatment of mammals under current farm management. Vet. Parasitol. 175, 73-79.

Dillon, P., Crosse, S., Stakelum, G., Flynn, F., 1995. The effect of calving date and stocking rate on the performance of spring-calving dairy cows. Grass Forage Sci. 50 (3), 286-299.

Ducheyne, E., Charlier, C., Vercruysse, J., Rinaldi, L., Biggeri, A., Demeler, J., Brandt,

C., de Waal, T., Selemetas, N., Hoglund,J., Kaba, J., Kowalczyk, S.J., Hendrickx, G., 2015. Modelling the spatial distribution of Fasciola hepatica in dairy cattle in Europe. Geospat. Health 9 (2), 261-270.

Duscher, R., Duscher, G., Hofer, J., Tichy, A., Prosl, H., Joachim, A., 2011. Fasciola hepatica - monitoring the milky way? The use of tank milk for liver fluke monitoring in dairy herds as base for treatment strategies. Vet. Parasitol. 178, 273-278.

Elliot, T.P., Kelley, J.M., Rawlin, G., Spithill, T.W., 2015. High prevalence of fasciolosis and evaluation of drug efficacy against Fasciola hepatica in dairy cattle in the Maffra and Bairnsdale districts of Gippsland, Victoria, Australia. Vet. Parasitol. 209,117-124.

European Climate Assessment & Dataset, 2016. (ECAD). Country Met Data, http:// eca.knmi.nl/countries/country_info.php?coun_id=ie Last access: 12/07/2016.

Fairweather, I., Boray, J.C., 1999. Fasciolicides: efficacy, actions, resistance and its management. Vet. J. 158, 81-112.

Fox, N.J., White, P.C.L., McClean, C.J., Marion, G., Evans, A., Hutchings, M.R., 2011. Predicting impacts of climate change on Fasciola hepatica risk. PLoS One 6, e16126.

Hanna, R.E.B., McMahon, C., Ellison, S., Edgar, H.W., Kajugu, P.-E., Gordon, A., Irwin,

D., Barley, J.P., Malone, F.E., Brennan, G.P., Fairweather, I., 2015. Fasciola hepatica: a comparative survey of adult fluke resistance to triclabendazole, nitroxynil and closantel on selected upland and lowland sheep farms in Northern Ireland using faecal egg counting, coproantigen ELISA testing and fluke histology. Vet. Parasitol. 207, 34-43.

Hoe, F.G.H., Ruegg, P.L., 2006. Opinions and practices of Wisconsin dairy producers about biosecurity and animal well-being. J. Dairy Sci. 89, 2297-2308.

Intergovernmental Panel on Climate Change (IPCC), 2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. In: Core writing team, Pachauri, R.K., Meyer, L.A. (Eds.). IPCC, Geneva, Switzerland, 151 pp. http://www.ipcc.ch/pdf/assessment-report/ar5/syr/SYR_AR5_FINAL_full_ wcover.pdfLast access: 27/04/2016.

Kelly, J.M., Elliot, T.P., Beddoe, T., Anderson, G., Skuce, P., Spithill, T.W., 2016. Current threat of triclabendazole resistance in Fasciola hepatica. Trends Parasitol. 32 (6).

Kristensen, T., Jensen, C., Ostergaard, S., Weisbjerg, M.R., Oaes, O., Nielsen, N.I., 2014. Feeding, production, and efficiency of Holstein-Friesian, Jersey, and mixed-breed lactating dairy cows in commercial Danish herds. J. Dairy Sci. 98, 263-274.

Kuerpick, B., Conraths, F., Staubach, C., Frolich, A., Schneider, T., Strube, C., 2013. Seroprevalence and GIS-supported risk factor analysis of Fasciola hepatica infections in dairy herds in Germany. Parasitology 140,1051-1060.

Mas-Coma, S., Valero, M.A., Bargues, M.D., 2009. Climate change effects on

trematodiases, with emphasis on zoonotic fascioliasis and schistosomiasis. Vet. Parasitol. 163, 264-280.

McCann, C.M., Baylis, M., Williams, D.J.L., 2010. The development of linear regression models using environmental variables to explain the spatial distribution of Fasciola hepatica infection in dairy herds in England and Wales. Int. J. Parasitol. 40,1021-1028.

Moll, L., Gaasenbeek, C.P.H., Vellema, P., Borgsteede, F.H.M., 2000. Resistance of Fasciola hepatica against triclabendazole in cattle and sheep in The Netherlands. Vet. Parasitol. 91,153-158.

Mooney, L., Good, B., Hanrahan, J.P., Mulcahy, G., de Waal, T., 2009. The comparative efficacy of four anthelmintics against a natural acquired Fasciola hepatica infection in hill sheep flock in the west of Ireland. Vet. Parasitol. 164, 201-205.

Morley, N.J., Lewis, J.W., 2008. The influence of climatic conditions on long-term changes in the helminth fauna of terrestrial mollusc and the implications for parasite transmission in southern England. J. Helminthol. 82, 325-335.

Nekouei, O., Stryhn, H., VanLeeuwen, J., Kelton, D., Hanna, P., Keefe, G., 2015. Predicting within-herd prevalence of infection with bovine leukemia virus using bulk-tank milk antibody levels. Prev. Vet. Med. 122,53-60.

Novobilsky, A., Novak, J., Bjorkman, C., Hoglund, J., 2015. Impact of meteorological and environmental factors on the spatial distribution of Fasciola hepatica in beef cattle herds in Sweden. BMC Vet. Res. 11,128.

Olaechea, F., Lovera, V., Larroza, M., Raffo, F., Cabrera, R., 2011. Resistance of

Fasciola hepatica against triclabendazole in cattle in Patagonia (Argentina). Vet. Parasitol. 178,364-366.

Ollerenshaw, C.B., Smith, L.P., 1969. Meteorological factors and forecasts of helminthic disease. Adv. Parasitol. 7, 283-323.

Ortiz, P., Scarcella, S., Cerna, C., Rosales, C., Cabrera, M., Guzman, M., Lamenza, P., Solana, H., 2013. Resistance of Fasciola hepatica against triclabendazole in cattle in Cajamarca (Peru): a clinical trial and an in vivo efficacy test in sheep. Vet. Parasitol. 195,118-121.

Prendiville, R., Pierce, K.M., Buckley, F., 2009. An evaluation of production efficiencies among lactating Holstein-Friesian, Jersey, and Jersey x Holstein-Friesian cows at pasture. J. Dairy Sci. 92,6176-6185.

Rapsch, C., Dahinden, T., Heinzmann, D., Torgenson, P.R., Braun, U., Deplazes, P., Hurni, L., Bar, H., Knubben-Schweizer, G., 2008. An interactive map to asses the potential spread of Lymnaea truncatula and the free-living stages of Fasciola hepatica in Switzerland. Vet. Parasitol. 154, 242-249.

Relf, V., Good, B., Hanrahan, J.P., McCarthy, E., Forbes, A.B., deWaal, T., 2011.

Temporal studies on Fasciola hepatica in Galba truncatula in the west of Ireland. Vet. Parasitol. 175, 287-292.

Rondelaud, D., Titi, A., Vignoles, P., Mekroud, A., Dreyfuss, G., 2013. Consequence of temperature changes on cercarial shedding from Galba truncatula infected with Fasciola hepatica or Paramphistomum daubneyi. Parasite 20,10.

Salimi-Bejestani, M.R., Daniel, R., Cripps, P., Felstead, S., Williams, D.J.L., 2007. Evaluation of an enzyme-linked immunosorbent assay for detection of antibodies to Fasciola hepatica in milk. Vet. Parasitol. 149, 290-293.

Sayers, R.G., Sayers, G.P., Mee, J.F., Good, M., Bermingham, M.L., Grant, J., Dillon, P.G., 2013. Implementing biosecurity measures on dairy farms in Ireland. Vet. J. 197, 259-267.

Sekiya, M., Zintl, A., Doherty, M., 2013. Bulk milk ELISA and the diagnosis of parasite infections in dairy herds: a review. Irish Vet. J. 66,14.

Selemetas, N., Phelan, P., O'Kiely, P., Waal, T.D., 2014. Weather and soil type affect incidence of fasciolosis in dairy cow herds. Vet. Rec. 175,371.

Taylor, M.A., Coop, R.L., Wall, R.L., 2007. Fasciola hepatica. In: B.S. Ltd (Ed.), Veterinary Parasitology. , third edition. Blackwell publishing, 600 pp.

Teagasc, 2016. Knowledge Transfer Groups Program, https://www.teagasc.ie/ about/farm-advisory/advisory-services/discussion-groups/ Last access: 12/07/2016.

The Irish National Metereological Service (Met Lireann), 2016. Climate of Ireland, http://www.met.ie/climate-ireland/climate-of-ireland.asp Last access: 12/07/2016.

Thomas, A.P., 1883. Memoirs: the life history of the liver-fluke (Fasciola hepatica). J. Cell Sci. s2-23, 99-133.

Urquhart, G.M., Armour, J., Duncan, J.L., Dunn, A.M., Jennings, F.W., 1996. Fasciola hepatica. Veterinary Parasitology. In: Blackwell science Ltd (Ed.), second ed. Blackwell publishing, pp. 103-112.

Van Dijk, J., Sargison, N.D., Kenyon, F., Skuce, P.J., 2010. Climate change and infectious disease: helminthological challenges to farmed ruminants in temperate regions. Animal 4, 377-392.

Van Duinen, R., Filatova, T., Geurts, P., Van der Veen, A., 2015. Empirical analysis of farmers' drought risk perception: objective factors, personal circumstances, and social influence. Risk Anal. 35, 741-755.

Yilma, J.M., Malone, J.B., 1998. A geographic information system forecast model for strategic control of fasciolosis in Ethiopia. Vet. Parasitol. 78,103-127.

Young, I., Rajic, A., Hendrick, S., Parker, S., Sanchez, J., McClure, J.T., McEwen, S.A., 2010. Attitudes toward the Canadian milk quality program and use of good production practices among Canadian dairy producers. Prev. Vet. Med. 94, 43-53.