bs_bs_banner
PRODUCTION ANIMALS
PRODUCTION ANIMALS
Breech-strike on mulesed, clipped and unmulesed Merino ewes and hoggets in south-eastern Australia LD Tyrell, JWA Larsen* and N Anderson
Objective Compare breech-strike on Merino ewes and hoggets that were mulesed, had breech and tail clips applied (‘clipped’) or left unmulesed and treated strategically with long-acting insecticide. Design
A cohort study on two farms in southern Victoria.
Methods Three treatment groups were established at lamb marking in 2008 and 2009 on each farm. The unmulesed group was treated with a long-acting insecticide in early spring. The prevalence of breech-strike and key risk factors, including presence of dag, urine stain and breech wrinkle, were compared between groups. Results Breech-strike was detected from October to December on 1.9% and 7.2% of mulesed, and on 14.8% and 12.5% of clipped ewes, respectively, on each farm. Thus, clipped ewes had a relative risk of breech-strike 7.8- and 1.7-fold that of mulesed ewes. Unmulesed ewes treated with insecticide had similar or less breechstrike compared with mulesed ewes (3.4% and 1.4%), but significantly more dag, stain and breech wrinkle. From January, breechstrike on unprotected unmulesed ewes was 8.5% and 2.8%, compared with 3.5% and 0% on mulesed ewes. Conclusion Early-season treatment of unmulesed sheep prevented most breech-strikes during spring and early summer on both hoggets and breeding ewes, confirming it as a short- to medium-term option for the control of breech-strike. Some benefits were associated with the use of clips but, to control breech-strike, clipped sheep should be treated the same as unmulesed sheep. Genetic selection to reduce the prevalence of dag will be required for future strategies to control breech-strike on Merino sheep in south-eastern Australia. Keywords breech-strike; clips; cyromazine; dag; dicyclanil; flystrike; mulesing Abbreviation
FEC, faecal egg count
Aust Vet J 2014;92:348–356
doi: 10.1111/avj.12228
T
he Australian sheep blowfly, Lucilia cuprina, causes flystrike, a major health and welfare problem for sheep production in Australia. Breech-strike is the main form of flystrike in areas with high winter rainfall, although a high prevalence of body-strike can occur in seasons with warm conditions and above-average rainfall. The mules operation, in combination with docking tails at the
*Corresponding author. The Mackinnon Project, University of Melbourne, 250 Princes Hwy, Werribee, Victoria 3030, Australia; [email protected]
348
Australian Veterinary Journal Volume 92, No 9, September 2014
correct length, has been effective in reducing breech-strike, although mulesed sheep with persistent diarrhoea and dag still have a high risk of strike.1,2 As the organisation responsible for coordinating research in the wool industry, Australian Wool Innovation has sponsored efforts to investigate alternatives to mulesing since 2004, with the objective of phasing-out mulesing by December 2010. This objective was not achieved but the research has continued.1 As part of this program, a comparison was made on three farms in Victoria of the prevalence of breech-strike on 12–20-month-old Merino hoggets that were mulesed or had breech and tail clips applied (‘clipped’) or were left unmulesed but treated with a long-acting insecticide early in spring.1 Despite the presence of severe faecal soiling of the breech wool (dag), unmulesed hoggets treated strategically with a long-acting insecticide had a similar prevalence of breech-strike to that on mulesed sheep (overall 0.9%, range 0–2.5%). In contrast, hoggets that had breech and tail clips applied at weaning had a relative risk of breech-strike 18.2-, 16.8- and 3.0-fold that of mulesed sheep on the three farms. There was less dag and urine stain and a shorter time was needed to crutch clipped hoggets, compared with unmulesed ones, but it was concluded that for effective control of breech-strike, the clipped hoggets should be treated in the same way as unmulesed hoggets.1 The relative merits of these treatments were assessed further in a second year, when the initial cohorts of hoggets were maiden ewes and new cohorts of hoggets were available for comparison. The measurements and observations made on these sheep are reported here. Materials and methods Experimental design and preparation of groups A cohort study was undertaken from September 2009 to May 2011 to compare the prevalence and indicators of susceptibility to breechstrike in three groups of maiden ewes and three groups of hoggets. At lamb marking in September 2009, new groups of 4–6-week-old lambs were sequentially allocated to one of three groups: (1) mulesed at lamb marking; (2) clipped (breech and tail clips applied at weaning); (3) left unmulesed. In early spring, at or soon after the expected emergence of L. cuprina flies from over-wintered larvae,3 a long-acting insecticide was applied to the breech and tail of the unmulesed sheep. Subsequently, the prevalence of breech-strike in the three groups was recorded. Indicators of susceptibility to breech-strike were measured at lamb marking, when the sheep were crutched and shorn, and at other times as shown in Figures 1 and 2.
© 2014 Australian Veterinary Association
PRODUCTION ANIMALS
Farm 1 PRODUCTION ANIMALS
5
Dag score
4 3 2 1 0 Oct-09
Jan-10
Apr-10
Jul-10
Oct-10
Jan-11
Oct-10
Jan-11
Farm 3 5
Figure 1. Average dag scores of 2009-born ewe hoggets in three treatment groups at each observation time between October 2009 and March 2011. Black arrows indicate when sheep were crutched, red arrows when sheep were shorn and blue arrows when the insecticide treatment was applied to sheep in the unmulesed group. Bars denote standard errors.
Dag score
4 3 2 1 0 Oct-09
Jan-10
This study was conducted under the University of Melbourne Animal Ethics Committee approval number 08-10949. Farms and sheep All the sheep on farm 2 in the previous study had been sold, so the study was conducted on the other two farms, one located at Nareen in western Victoria (farm 1) and the other at Woodside in East Gippsland (farm 3). The characteristics of the farms and details of the management of sheep on these farms have been published previously.1 The maiden ewes were born in 2008 and the hoggets in 2009. The latter cohorts were obtained from 1186 ewe lambs on farm 1 and 1054 ewe lambs on farm 3, which were marked at 4–6 weeks of age and weaned approximately 13 weeks after the start of lambing. The same operator as in the previous study1 applied the clips and mulesed the lambs, and a combined topical local anaesthetic and antiseptic treatment was applied after mulesing (Trisolfen™; Bayer Australia Ltd, NSW, Australia). Treatments A spray-on formulation of cyromazine or dicyclanil (Novartis Animal Health Australasia Pty Ltd, NSW, Australia) was used for the earlyspring treatment of the unmulesed sheep because of ease of application and the long period of protection (11 weeks for cyromazine, 18–24 weeks for dicyclanil). Cyromazine was used when a product with a shorter withholding period for wool harvesting was required, as it had a withholding period of 2 months versus 3 months for
© 2014 Australian Veterinary Association
Apr-10 Mulesed
Jul-10 Clipped
Unmulesed
dicyclanil; 17–28 mL of a 60 g/L solution of cyromazine spray-on solution was applied in three equal bands, two each side of the breech and overlapping at the top of the tail, and a third band directly over the tail. The dose used was determined by the time elapsed since last shearing, as per the manufacturer’s instructions. For dicyclanil, at least 12 mL of a 50 g/L solution was applied as two bands of 4 mL on each side of the breech, overlapping at the top of the tail, and a third band of 4 mL over the tail. Although not included in the directions provided by the manufacturer, an additional band of equal volume of either cyromazine or dicyclanil was applied over the tail and crutch of sheep with severe dag or stain. These treatments were applied to the unmulesed maiden ewes in August (farm 1) and September 2010 (farm 3), and to the unmulesed hoggets in September 2010 (farm 3) and October (farm 1). Mulesed and clipped groups were treated with insecticide only if the incidence of flystrike exceeded 2% of new cases per week. Measurements
Risk factors for breech-strike. The amount of faecal soiling of breech wool (dag), urine soiling (stain), degree of breech wrinkle and amount of bare area on the breech were scored using scales of 0–5 for dag and 1–5 for the other measures, as previously described.1 The bare area score is the reverse of the breech cover score; a bare area score of 1 indicates no bare area and corresponds to a breech cover score of 5, indicating heavy cover.4
Australian Veterinary Journal Volume 92, No 9, September 2014
349
PRODUCTION ANIMALS
4 Dag score
PRODUCTION ANIMALS
Farm 1 5
3 2 1 0 Oct-08
Jan-09
Apr-09
Jul-09
Oct-09
Jan-10
Apr-10
Jul-10
Oct-10
Jan-11
Farm 3 5
Dag score
4 3 2 1 0 Oct-08
Jan-09
Apr-09
Jul-09
Oct-09
Mulesed
Jan-10
Clipped
Apr-10
Jul-10
Unmulesed
Flystrike.Breech- and body-strike were recorded by the farmer when struck sheep were treated in the paddock and also by the authors when measurements were made. Statistical analysis For ordinal data (dag, stain, wrinkle and bare area), the nonparametric Kruskal-Wallis test was used to compare the data from the three treatment groups at each observation. Dunn’s test was used for pairwise comparisons.5 The prevalence of breech-strike in the three treatment groups was compared using the mid-P value. To estimate multivariable associations between risk factors and breech-strike, the odds ratio of breech-strike for each major risk factor was assessed by univariable exact logistic regression,6 carried out after first categorising the most recent dag, urine stain, breech wrinkle and bare area scores of each sheep into either (a) none to mild (score ≤1) or (b) moderate to severe (score ≥2). Exact logistic regression, rather than logistic regression, was conducted because some categories did not have any struck sheep. Initially, all risk factors were analysed individually by univariable analysis. Subsequently, those with a P-value 0.05).7 The final multivariable model took account of possible confounding between risk factors when there was more than one risk factor included. This analysis used data from the clipped and unmulesed groups, but the mulesed group was not analysed because of the low number of struck sheep in this group.
350
Australian Veterinary Journal Volume 92, No 9, September 2014
Oct-10
Jan-11
Figure 2. Average dag scores of 2008-born maiden ewes in three treatment groups at each observation time between October 2008 and March 2011. Black arrows indicate when sheep were crutched, red arrows when sheep were shorn and blue arrows when the insecticide treatment was applied to sheep in the unmulesed group. Bars denote standard errors.
The correlation between the individual scores at two observations for dag or breech wrinkle or bare area was estimated using Spearman’s rank correlation coefficient. IBM SPSS v20.0.0.0 was used for the parametric and non-parametric analyses, OpenEpi v2.3.1 for calculating the mid-P value and risk ratios, and Stata v11.2 was used for the exact logistic regression and to estimate correlations. Results Management The times of shearing, crutching and strategic application of the long-acting insecticide for the 2009-born hoggets and 2008-born maiden ewes are shown in Figures 1 and 2, respectively. No additional insecticide treatments were applied on either farm after crutching or shearing. Risk factors for breech-strike
Breech wrinkle and bare area. At marking, in October 2009, approximately 60% of the lambs had a breech wrinkle score ≥3, with average scores of 2.8 and 2.9 on farms 1 and 3, respectively. Similarly, these lambs had negligible bare areas on their breeches, with only 7% having a bare area score ≥3 (average 1.3 on both farms). Later, in March 2011, the unmulesed hoggets on farm 1 had significantly more breech wrinkle than either the clipped or mulesed ones, with the scores of the clipped group intermediate (3.0, 2.5 and 1.2 for
© 2014 Australian Veterinary Association
unmulesed, clipped and mulesed, respectively). At this time, unmulesed hoggets on this farm had significantly smaller bare areas on the breeches, with the clipped group again intermediate between, and significantly different from, the mulesed and unmulesed ones (2.5, 2.1 and 1.8 for unmulesed, clipped and mulesed, respectively). For the maiden ewes, the ranking of wrinkle score was the same as when they were hoggets a year earlier, although the average breech wrinkle score of the unmulesed group tended to increase as the sheep matured. Unmulesed maiden ewes had the greatest breech wrinkle score, followed by the clipped and mulesed ewes (average scores 3.0, 2.4 and 1.2, and 3.2, 2.5 and 1.7, on farms 1 and 3, respectively). Unmulesed maiden ewes had the smallest bare area of the breech, followed by the clipped and mulesed ones (average scores 2.0, 2.5 and 3.1, and 2.6, 2.8 and 2.8, on farms 1 and 3, respectively). As sheep matured from hoggets to maiden ewes, the proportion of ewes with a bare area score ≥3 increased by a factor of 2–3 for all groups on both farms.
Dag. The ranking of average dag scores of sheep in the clipped, unmulesed and mulesed groups was consistent at most observations throughout the study (Figures 1, 2). Treated unmulesed sheep usually had the highest dag score, followed by clipped sheep, and mulesed sheep had the least dag. The differences between the groups were generally significant from approximately 5 to 10 months of age onwards, which corresponded to the autumn or winter following their year of birth. For both the hoggets and maiden ewes, on both farms, the amount of dag increased between October and December (Figures 1, 2). At hogget age, sheep on farm 3 consistently had more dag and a higher proportion of sheep with severe dag (dag score ≥4) compared with farm 1. For example, in the treated unmulesed group on farm 3, the
proportion with severe dag was 34.7% when the insecticide was applied in early spring and 76.8% in early summer, whereas on farm 1 the proportions were 7.0% and 35.5%, respectively.
Urine stain. The treated unmulesed hoggets and maiden ewes had significantly more urine stain than the mulesed sheep at all visits on both farms. On farm 1, the clipped group also had significantly more stain than the mulesed sheep, whereas on farm 3, there was generally no difference between these two groups. On both farms, the clipped ewes had significantly less stain than unmulesed ewes. There was a noticeable increase in the amount of stain from early spring to early summer, corresponding to an increase in fleece length. For example, in the treated unmulesed group, the proportion of hoggets with a stain score ≥3 increased from 21% to 69% and from 27% to 74% on farms 1 and 3, respectively.
PRODUCTION ANIMALS
PRODUCTION ANIMALS
Breech-strike The prevalence of breech-strike and estimates of relative risk for breech-strike on hoggets and maiden ewes at both farms, for two risk periods, are shown in Table 1.
Period 1 (September to December). The prevalence of breechstrike was significantly higher on the clipped hoggets and maiden ewes, compared with both the mulesed and unmulesed sheep, on both farms. However, the prevalence of breech-strike on unmulesed sheep after insecticide application, in both age groups, was either similar to or less than that on mulesed sheep. These differences in prevalence led to a significantly greater risk of breech-strike on the clipped hoggets and maiden ewes, relative to the mulesed group. Compared with mulesed sheep, unmulesed ewes and hoggets had no greater risk of breech-strike during this period, with the relative risk being lower for unmulesed maiden ewes on farm 3.
Table 1. Prevalence and relative risk of breech-strike on hoggets and maiden ewes in three treatment groups (mulesed, clipped and unmulesed) on farms 1 and 3 for two risk periodsa Class
Hoggets (2009-born)
Farm
1
3
Maiden ewes (2008-born)
1
3
Groupb
Mulesed Clipped UnmulesedCyr Mulesed Clipped UnmulesedCyr Mulesed Clipped UnmulesedDic Mulesed Clipped UnmulesedCyr
Risk period 2 (Jan–May 2011)c
Risk period 1 (Sept–Dec 2010) No. of sheep
Prevalence (% struck)
RR compared with mulesed (95% CI)
No. of sheep
Prevalence (% struck)
RR compared with mulesed (95% CI)
323 343 327 241 243 250 265 264 263 292 320 292
2.2A 18.1B 0.0C 2.5A 9.9B 0.8A 1.9A 14.8B 3.4A 7.2A 12.5B 1.4C
– 8.3 (3.9–18.0)*** NCP – 4.0 (1.7–9.5)*** 0.3 (0.1–1.6)NS – 7.8 (3.1–19.6)*** 1.8 (0.6–5.3)NS – 1.7 (1.1–2.9)* 0.2 (0.1–0.6)**
273 297 267 223 229 235 234 218 212 286 308 293
0.0A 0.3A 0.7A 1.3A 5.2B 14.9C 0.0A 0.0A 2.8B 3.5A 3.2A 8.5B
– NCP NCP – 3.9 (1.1–13.6)* 11.1 (3.5–35.5)*** – NCP NCP – 0.9 (0.4–2.2)* 2.4 (1.2–5.0)**
Different superscripts within a column for each farm and period denote a significant difference between groups (P < 0.05). Early-spring treatment with insecticide in Period 1 was either a spray-on formulation of cyromazine (Cyr), applied in late Sept or early Oct, or dicyclanil (Dic) applied in Aug. c When the protection period of the insecticide had expired or the breech wool and dags had been removed after crutching. NS, not significant; *P < 0.05, **P < 0.01, ***P < 0.001. 95% CI, 95% confidence interval; NCP, no calculation possible because the reference (mulesed) group had no cases of flystrike; RR, relative risk. a
b
© 2014 Australian Veterinary Association
Australian Veterinary Journal Volume 92, No 9, September 2014
351
PRODUCTION ANIMALS
PRODUCTION ANIMALS Period 2 (January to May). In contrast, during period 2 there was a lower prevalence of breech-strike on both the clipped and mulesed sheep in both age groups, with the only difference being a significantly higher prevalence on clipped hoggets on farm 3 (Table 1). However, unmulesed hoggets on farm 3, and unmulesed maiden ewes on both farms, had a significantly higher prevalence of breech-strike when not protected by insecticide. This led to a significantly higher relative risk of breech-strike for the unmulesed and clipped hoggets, and for the unmulesed maiden ewes, on farm 3. Correlations between risk factors and breech-strike The relationships between risk factors and breech-strike for the clipped hoggets and maiden ewes, from September to December 2010 (Period 1), are summarised in Table 2. Increased dag (score ≥2) was a significant risk factor in both univariable and multivariable analyses, on three of four occasions. Urine stain was a significant risk factor in two of four multivariable analyses, bare area in one of three multivariable analyses and breech wrinkle in one of three univariable but none of the multivariable analyses. No risk factors were significant for cases of breech-strike on treated unmulesed hoggets or maiden ewes during Period 1 on either farm (all P > 0.37), so no multivariable analyses were undertaken. No analysis was undertaken for the mulesed hoggets or maiden ewes for this period because there were too few cases of breech-strike in these groups. The odds ratios for cases of breech-strike in Period 2 (December 2010 to April 2011) on farm 3, when unmulesed sheep were not protected by insecticide, are shown in Table 3. Again, the presence of increased dag (score ≥2) significantly increased the odds of breech-strike on
both unmulesed and clipped hoggets and maiden ewes, being significant in all four analyses. Stain was a significant risk factor in one multivariable analysis (2009-born hoggets) and increased bare area approached significance as a protective trait in one multivariable analysis (clipped maiden ewes). Correlations between risk factors at different ages The correlation for each risk factor for breech-strike at different ages is shown in Table 4. Correlations were only calculated for the unmulesed group for dag score, breech wrinkle and bare area. The dag score at marking and weaning correlated poorly with dag score at hogget and maiden ewe ages, whereas the score at hogget age (15 months) strongly correlated with the dag score of maiden ewes at 27 months of age (r = 0.55 and 0.42 for farms 1 and 3, respectively; both P < 0.0001). Breech wrinkle was assessed at marking and again after crutching or shearing as hoggets and maiden ewes, so no observations were made at weaning. A reasonably strong and significant correlation existed between the breech wrinkle scores at marking and hogget age (r = 0.29–0.36), with less correlation between the wrinkle score at marking and maiden ewe age, but still significant. There was a strong and significant correlation between the wrinkle scores measured at hogget and maiden ewe ages (r = 0.35 and 0.43, respectively). There was a reasonably strong and significant correlation between bare area score of the breech at marking and the score at hogget and maiden ewe age on both farms, between the score at weaning and
Table 2. Odds ratios and 95% confidence interval (CI) of risk factors for cases of breech-strike on hoggets and maiden ewes, calculated by exact logistic regression, during the period that unmulesed sheep were protected from breech-strike by a long-acting insecticide (Period 1: Sept–Dec 2010)a
Class
Hoggets (2009-born)
Farm
1
3 Maiden ewes (2008-born)
1
3
Risk factorb
Dag Stain Bare area Breech wrinkle Dag Stain Dag Stain Bare area Breech wrinkle Dag Stain Bare area Br wrinkle
Univariable analysisc
Multivariable analysis
Odds ratio (95% CI)
P value
Odds ratio (95% CI)
P value
9.7 (2.5–84.3) 1.3 (0.7–2.4) 0.4 (0.2–0.9) 5.6 (0.9–236.8) 3.7 (0.6–∞)* 1.4 (0.5–3.5) 17.4 (3.0–∞)* 3.1 (1.4–6.9) 0.3 (0.1–1.5) 5.4 (0.9–∞)* 33.6 (6.0–∞)* 2.9 (1.3–6.4) 0.8 (0.2–7.2) 7.1 (1.1–293.4)
PRODUCTION ANIMALS
PRODUCTION ANIMALS
Breech-strike on mulesed, clipped and unmulesed Merino ewes and hoggets in south-eastern Australia LD Tyrell, JWA Larsen* and N Anderson
Objective Compare breech-strike on Merino ewes and hoggets that were mulesed, had breech and tail clips applied (‘clipped’) or left unmulesed and treated strategically with long-acting insecticide. Design
A cohort study on two farms in southern Victoria.
Methods Three treatment groups were established at lamb marking in 2008 and 2009 on each farm. The unmulesed group was treated with a long-acting insecticide in early spring. The prevalence of breech-strike and key risk factors, including presence of dag, urine stain and breech wrinkle, were compared between groups. Results Breech-strike was detected from October to December on 1.9% and 7.2% of mulesed, and on 14.8% and 12.5% of clipped ewes, respectively, on each farm. Thus, clipped ewes had a relative risk of breech-strike 7.8- and 1.7-fold that of mulesed ewes. Unmulesed ewes treated with insecticide had similar or less breechstrike compared with mulesed ewes (3.4% and 1.4%), but significantly more dag, stain and breech wrinkle. From January, breechstrike on unprotected unmulesed ewes was 8.5% and 2.8%, compared with 3.5% and 0% on mulesed ewes. Conclusion Early-season treatment of unmulesed sheep prevented most breech-strikes during spring and early summer on both hoggets and breeding ewes, confirming it as a short- to medium-term option for the control of breech-strike. Some benefits were associated with the use of clips but, to control breech-strike, clipped sheep should be treated the same as unmulesed sheep. Genetic selection to reduce the prevalence of dag will be required for future strategies to control breech-strike on Merino sheep in south-eastern Australia. Keywords breech-strike; clips; cyromazine; dag; dicyclanil; flystrike; mulesing Abbreviation
FEC, faecal egg count
Aust Vet J 2014;92:348–356
doi: 10.1111/avj.12228
T
he Australian sheep blowfly, Lucilia cuprina, causes flystrike, a major health and welfare problem for sheep production in Australia. Breech-strike is the main form of flystrike in areas with high winter rainfall, although a high prevalence of body-strike can occur in seasons with warm conditions and above-average rainfall. The mules operation, in combination with docking tails at the
*Corresponding author. The Mackinnon Project, University of Melbourne, 250 Princes Hwy, Werribee, Victoria 3030, Australia; [email protected]
348
Australian Veterinary Journal Volume 92, No 9, September 2014
correct length, has been effective in reducing breech-strike, although mulesed sheep with persistent diarrhoea and dag still have a high risk of strike.1,2 As the organisation responsible for coordinating research in the wool industry, Australian Wool Innovation has sponsored efforts to investigate alternatives to mulesing since 2004, with the objective of phasing-out mulesing by December 2010. This objective was not achieved but the research has continued.1 As part of this program, a comparison was made on three farms in Victoria of the prevalence of breech-strike on 12–20-month-old Merino hoggets that were mulesed or had breech and tail clips applied (‘clipped’) or were left unmulesed but treated with a long-acting insecticide early in spring.1 Despite the presence of severe faecal soiling of the breech wool (dag), unmulesed hoggets treated strategically with a long-acting insecticide had a similar prevalence of breech-strike to that on mulesed sheep (overall 0.9%, range 0–2.5%). In contrast, hoggets that had breech and tail clips applied at weaning had a relative risk of breech-strike 18.2-, 16.8- and 3.0-fold that of mulesed sheep on the three farms. There was less dag and urine stain and a shorter time was needed to crutch clipped hoggets, compared with unmulesed ones, but it was concluded that for effective control of breech-strike, the clipped hoggets should be treated in the same way as unmulesed hoggets.1 The relative merits of these treatments were assessed further in a second year, when the initial cohorts of hoggets were maiden ewes and new cohorts of hoggets were available for comparison. The measurements and observations made on these sheep are reported here. Materials and methods Experimental design and preparation of groups A cohort study was undertaken from September 2009 to May 2011 to compare the prevalence and indicators of susceptibility to breechstrike in three groups of maiden ewes and three groups of hoggets. At lamb marking in September 2009, new groups of 4–6-week-old lambs were sequentially allocated to one of three groups: (1) mulesed at lamb marking; (2) clipped (breech and tail clips applied at weaning); (3) left unmulesed. In early spring, at or soon after the expected emergence of L. cuprina flies from over-wintered larvae,3 a long-acting insecticide was applied to the breech and tail of the unmulesed sheep. Subsequently, the prevalence of breech-strike in the three groups was recorded. Indicators of susceptibility to breech-strike were measured at lamb marking, when the sheep were crutched and shorn, and at other times as shown in Figures 1 and 2.
© 2014 Australian Veterinary Association
PRODUCTION ANIMALS
Farm 1 PRODUCTION ANIMALS
5
Dag score
4 3 2 1 0 Oct-09
Jan-10
Apr-10
Jul-10
Oct-10
Jan-11
Oct-10
Jan-11
Farm 3 5
Figure 1. Average dag scores of 2009-born ewe hoggets in three treatment groups at each observation time between October 2009 and March 2011. Black arrows indicate when sheep were crutched, red arrows when sheep were shorn and blue arrows when the insecticide treatment was applied to sheep in the unmulesed group. Bars denote standard errors.
Dag score
4 3 2 1 0 Oct-09
Jan-10
This study was conducted under the University of Melbourne Animal Ethics Committee approval number 08-10949. Farms and sheep All the sheep on farm 2 in the previous study had been sold, so the study was conducted on the other two farms, one located at Nareen in western Victoria (farm 1) and the other at Woodside in East Gippsland (farm 3). The characteristics of the farms and details of the management of sheep on these farms have been published previously.1 The maiden ewes were born in 2008 and the hoggets in 2009. The latter cohorts were obtained from 1186 ewe lambs on farm 1 and 1054 ewe lambs on farm 3, which were marked at 4–6 weeks of age and weaned approximately 13 weeks after the start of lambing. The same operator as in the previous study1 applied the clips and mulesed the lambs, and a combined topical local anaesthetic and antiseptic treatment was applied after mulesing (Trisolfen™; Bayer Australia Ltd, NSW, Australia). Treatments A spray-on formulation of cyromazine or dicyclanil (Novartis Animal Health Australasia Pty Ltd, NSW, Australia) was used for the earlyspring treatment of the unmulesed sheep because of ease of application and the long period of protection (11 weeks for cyromazine, 18–24 weeks for dicyclanil). Cyromazine was used when a product with a shorter withholding period for wool harvesting was required, as it had a withholding period of 2 months versus 3 months for
© 2014 Australian Veterinary Association
Apr-10 Mulesed
Jul-10 Clipped
Unmulesed
dicyclanil; 17–28 mL of a 60 g/L solution of cyromazine spray-on solution was applied in three equal bands, two each side of the breech and overlapping at the top of the tail, and a third band directly over the tail. The dose used was determined by the time elapsed since last shearing, as per the manufacturer’s instructions. For dicyclanil, at least 12 mL of a 50 g/L solution was applied as two bands of 4 mL on each side of the breech, overlapping at the top of the tail, and a third band of 4 mL over the tail. Although not included in the directions provided by the manufacturer, an additional band of equal volume of either cyromazine or dicyclanil was applied over the tail and crutch of sheep with severe dag or stain. These treatments were applied to the unmulesed maiden ewes in August (farm 1) and September 2010 (farm 3), and to the unmulesed hoggets in September 2010 (farm 3) and October (farm 1). Mulesed and clipped groups were treated with insecticide only if the incidence of flystrike exceeded 2% of new cases per week. Measurements
Risk factors for breech-strike. The amount of faecal soiling of breech wool (dag), urine soiling (stain), degree of breech wrinkle and amount of bare area on the breech were scored using scales of 0–5 for dag and 1–5 for the other measures, as previously described.1 The bare area score is the reverse of the breech cover score; a bare area score of 1 indicates no bare area and corresponds to a breech cover score of 5, indicating heavy cover.4
Australian Veterinary Journal Volume 92, No 9, September 2014
349
PRODUCTION ANIMALS
4 Dag score
PRODUCTION ANIMALS
Farm 1 5
3 2 1 0 Oct-08
Jan-09
Apr-09
Jul-09
Oct-09
Jan-10
Apr-10
Jul-10
Oct-10
Jan-11
Farm 3 5
Dag score
4 3 2 1 0 Oct-08
Jan-09
Apr-09
Jul-09
Oct-09
Mulesed
Jan-10
Clipped
Apr-10
Jul-10
Unmulesed
Flystrike.Breech- and body-strike were recorded by the farmer when struck sheep were treated in the paddock and also by the authors when measurements were made. Statistical analysis For ordinal data (dag, stain, wrinkle and bare area), the nonparametric Kruskal-Wallis test was used to compare the data from the three treatment groups at each observation. Dunn’s test was used for pairwise comparisons.5 The prevalence of breech-strike in the three treatment groups was compared using the mid-P value. To estimate multivariable associations between risk factors and breech-strike, the odds ratio of breech-strike for each major risk factor was assessed by univariable exact logistic regression,6 carried out after first categorising the most recent dag, urine stain, breech wrinkle and bare area scores of each sheep into either (a) none to mild (score ≤1) or (b) moderate to severe (score ≥2). Exact logistic regression, rather than logistic regression, was conducted because some categories did not have any struck sheep. Initially, all risk factors were analysed individually by univariable analysis. Subsequently, those with a P-value 0.05).7 The final multivariable model took account of possible confounding between risk factors when there was more than one risk factor included. This analysis used data from the clipped and unmulesed groups, but the mulesed group was not analysed because of the low number of struck sheep in this group.
350
Australian Veterinary Journal Volume 92, No 9, September 2014
Oct-10
Jan-11
Figure 2. Average dag scores of 2008-born maiden ewes in three treatment groups at each observation time between October 2008 and March 2011. Black arrows indicate when sheep were crutched, red arrows when sheep were shorn and blue arrows when the insecticide treatment was applied to sheep in the unmulesed group. Bars denote standard errors.
The correlation between the individual scores at two observations for dag or breech wrinkle or bare area was estimated using Spearman’s rank correlation coefficient. IBM SPSS v20.0.0.0 was used for the parametric and non-parametric analyses, OpenEpi v2.3.1 for calculating the mid-P value and risk ratios, and Stata v11.2 was used for the exact logistic regression and to estimate correlations. Results Management The times of shearing, crutching and strategic application of the long-acting insecticide for the 2009-born hoggets and 2008-born maiden ewes are shown in Figures 1 and 2, respectively. No additional insecticide treatments were applied on either farm after crutching or shearing. Risk factors for breech-strike
Breech wrinkle and bare area. At marking, in October 2009, approximately 60% of the lambs had a breech wrinkle score ≥3, with average scores of 2.8 and 2.9 on farms 1 and 3, respectively. Similarly, these lambs had negligible bare areas on their breeches, with only 7% having a bare area score ≥3 (average 1.3 on both farms). Later, in March 2011, the unmulesed hoggets on farm 1 had significantly more breech wrinkle than either the clipped or mulesed ones, with the scores of the clipped group intermediate (3.0, 2.5 and 1.2 for
© 2014 Australian Veterinary Association
unmulesed, clipped and mulesed, respectively). At this time, unmulesed hoggets on this farm had significantly smaller bare areas on the breeches, with the clipped group again intermediate between, and significantly different from, the mulesed and unmulesed ones (2.5, 2.1 and 1.8 for unmulesed, clipped and mulesed, respectively). For the maiden ewes, the ranking of wrinkle score was the same as when they were hoggets a year earlier, although the average breech wrinkle score of the unmulesed group tended to increase as the sheep matured. Unmulesed maiden ewes had the greatest breech wrinkle score, followed by the clipped and mulesed ewes (average scores 3.0, 2.4 and 1.2, and 3.2, 2.5 and 1.7, on farms 1 and 3, respectively). Unmulesed maiden ewes had the smallest bare area of the breech, followed by the clipped and mulesed ones (average scores 2.0, 2.5 and 3.1, and 2.6, 2.8 and 2.8, on farms 1 and 3, respectively). As sheep matured from hoggets to maiden ewes, the proportion of ewes with a bare area score ≥3 increased by a factor of 2–3 for all groups on both farms.
Dag. The ranking of average dag scores of sheep in the clipped, unmulesed and mulesed groups was consistent at most observations throughout the study (Figures 1, 2). Treated unmulesed sheep usually had the highest dag score, followed by clipped sheep, and mulesed sheep had the least dag. The differences between the groups were generally significant from approximately 5 to 10 months of age onwards, which corresponded to the autumn or winter following their year of birth. For both the hoggets and maiden ewes, on both farms, the amount of dag increased between October and December (Figures 1, 2). At hogget age, sheep on farm 3 consistently had more dag and a higher proportion of sheep with severe dag (dag score ≥4) compared with farm 1. For example, in the treated unmulesed group on farm 3, the
proportion with severe dag was 34.7% when the insecticide was applied in early spring and 76.8% in early summer, whereas on farm 1 the proportions were 7.0% and 35.5%, respectively.
Urine stain. The treated unmulesed hoggets and maiden ewes had significantly more urine stain than the mulesed sheep at all visits on both farms. On farm 1, the clipped group also had significantly more stain than the mulesed sheep, whereas on farm 3, there was generally no difference between these two groups. On both farms, the clipped ewes had significantly less stain than unmulesed ewes. There was a noticeable increase in the amount of stain from early spring to early summer, corresponding to an increase in fleece length. For example, in the treated unmulesed group, the proportion of hoggets with a stain score ≥3 increased from 21% to 69% and from 27% to 74% on farms 1 and 3, respectively.
PRODUCTION ANIMALS
PRODUCTION ANIMALS
Breech-strike The prevalence of breech-strike and estimates of relative risk for breech-strike on hoggets and maiden ewes at both farms, for two risk periods, are shown in Table 1.
Period 1 (September to December). The prevalence of breechstrike was significantly higher on the clipped hoggets and maiden ewes, compared with both the mulesed and unmulesed sheep, on both farms. However, the prevalence of breech-strike on unmulesed sheep after insecticide application, in both age groups, was either similar to or less than that on mulesed sheep. These differences in prevalence led to a significantly greater risk of breech-strike on the clipped hoggets and maiden ewes, relative to the mulesed group. Compared with mulesed sheep, unmulesed ewes and hoggets had no greater risk of breech-strike during this period, with the relative risk being lower for unmulesed maiden ewes on farm 3.
Table 1. Prevalence and relative risk of breech-strike on hoggets and maiden ewes in three treatment groups (mulesed, clipped and unmulesed) on farms 1 and 3 for two risk periodsa Class
Hoggets (2009-born)
Farm
1
3
Maiden ewes (2008-born)
1
3
Groupb
Mulesed Clipped UnmulesedCyr Mulesed Clipped UnmulesedCyr Mulesed Clipped UnmulesedDic Mulesed Clipped UnmulesedCyr
Risk period 2 (Jan–May 2011)c
Risk period 1 (Sept–Dec 2010) No. of sheep
Prevalence (% struck)
RR compared with mulesed (95% CI)
No. of sheep
Prevalence (% struck)
RR compared with mulesed (95% CI)
323 343 327 241 243 250 265 264 263 292 320 292
2.2A 18.1B 0.0C 2.5A 9.9B 0.8A 1.9A 14.8B 3.4A 7.2A 12.5B 1.4C
– 8.3 (3.9–18.0)*** NCP – 4.0 (1.7–9.5)*** 0.3 (0.1–1.6)NS – 7.8 (3.1–19.6)*** 1.8 (0.6–5.3)NS – 1.7 (1.1–2.9)* 0.2 (0.1–0.6)**
273 297 267 223 229 235 234 218 212 286 308 293
0.0A 0.3A 0.7A 1.3A 5.2B 14.9C 0.0A 0.0A 2.8B 3.5A 3.2A 8.5B
– NCP NCP – 3.9 (1.1–13.6)* 11.1 (3.5–35.5)*** – NCP NCP – 0.9 (0.4–2.2)* 2.4 (1.2–5.0)**
Different superscripts within a column for each farm and period denote a significant difference between groups (P < 0.05). Early-spring treatment with insecticide in Period 1 was either a spray-on formulation of cyromazine (Cyr), applied in late Sept or early Oct, or dicyclanil (Dic) applied in Aug. c When the protection period of the insecticide had expired or the breech wool and dags had been removed after crutching. NS, not significant; *P < 0.05, **P < 0.01, ***P < 0.001. 95% CI, 95% confidence interval; NCP, no calculation possible because the reference (mulesed) group had no cases of flystrike; RR, relative risk. a
b
© 2014 Australian Veterinary Association
Australian Veterinary Journal Volume 92, No 9, September 2014
351
PRODUCTION ANIMALS
PRODUCTION ANIMALS Period 2 (January to May). In contrast, during period 2 there was a lower prevalence of breech-strike on both the clipped and mulesed sheep in both age groups, with the only difference being a significantly higher prevalence on clipped hoggets on farm 3 (Table 1). However, unmulesed hoggets on farm 3, and unmulesed maiden ewes on both farms, had a significantly higher prevalence of breech-strike when not protected by insecticide. This led to a significantly higher relative risk of breech-strike for the unmulesed and clipped hoggets, and for the unmulesed maiden ewes, on farm 3. Correlations between risk factors and breech-strike The relationships between risk factors and breech-strike for the clipped hoggets and maiden ewes, from September to December 2010 (Period 1), are summarised in Table 2. Increased dag (score ≥2) was a significant risk factor in both univariable and multivariable analyses, on three of four occasions. Urine stain was a significant risk factor in two of four multivariable analyses, bare area in one of three multivariable analyses and breech wrinkle in one of three univariable but none of the multivariable analyses. No risk factors were significant for cases of breech-strike on treated unmulesed hoggets or maiden ewes during Period 1 on either farm (all P > 0.37), so no multivariable analyses were undertaken. No analysis was undertaken for the mulesed hoggets or maiden ewes for this period because there were too few cases of breech-strike in these groups. The odds ratios for cases of breech-strike in Period 2 (December 2010 to April 2011) on farm 3, when unmulesed sheep were not protected by insecticide, are shown in Table 3. Again, the presence of increased dag (score ≥2) significantly increased the odds of breech-strike on
both unmulesed and clipped hoggets and maiden ewes, being significant in all four analyses. Stain was a significant risk factor in one multivariable analysis (2009-born hoggets) and increased bare area approached significance as a protective trait in one multivariable analysis (clipped maiden ewes). Correlations between risk factors at different ages The correlation for each risk factor for breech-strike at different ages is shown in Table 4. Correlations were only calculated for the unmulesed group for dag score, breech wrinkle and bare area. The dag score at marking and weaning correlated poorly with dag score at hogget and maiden ewe ages, whereas the score at hogget age (15 months) strongly correlated with the dag score of maiden ewes at 27 months of age (r = 0.55 and 0.42 for farms 1 and 3, respectively; both P < 0.0001). Breech wrinkle was assessed at marking and again after crutching or shearing as hoggets and maiden ewes, so no observations were made at weaning. A reasonably strong and significant correlation existed between the breech wrinkle scores at marking and hogget age (r = 0.29–0.36), with less correlation between the wrinkle score at marking and maiden ewe age, but still significant. There was a strong and significant correlation between the wrinkle scores measured at hogget and maiden ewe ages (r = 0.35 and 0.43, respectively). There was a reasonably strong and significant correlation between bare area score of the breech at marking and the score at hogget and maiden ewe age on both farms, between the score at weaning and
Table 2. Odds ratios and 95% confidence interval (CI) of risk factors for cases of breech-strike on hoggets and maiden ewes, calculated by exact logistic regression, during the period that unmulesed sheep were protected from breech-strike by a long-acting insecticide (Period 1: Sept–Dec 2010)a
Class
Hoggets (2009-born)
Farm
1
3 Maiden ewes (2008-born)
1
3
Risk factorb
Dag Stain Bare area Breech wrinkle Dag Stain Dag Stain Bare area Breech wrinkle Dag Stain Bare area Br wrinkle
Univariable analysisc
Multivariable analysis
Odds ratio (95% CI)
P value
Odds ratio (95% CI)
P value
9.7 (2.5–84.3) 1.3 (0.7–2.4) 0.4 (0.2–0.9) 5.6 (0.9–236.8) 3.7 (0.6–∞)* 1.4 (0.5–3.5) 17.4 (3.0–∞)* 3.1 (1.4–6.9) 0.3 (0.1–1.5) 5.4 (0.9–∞)* 33.6 (6.0–∞)* 2.9 (1.3–6.4) 0.8 (0.2–7.2) 7.1 (1.1–293.4)
