Intensive rotational grazing assists control of gastrointestinal nematodosis of sheep in a cool temperate environment with summer-dominant rainfall

Available online at www.sciencedirect.com

Veterinary Parasitology 153 (2008) 108–120 www.elsevier.com/locate/vetpar

Intensive rotational grazing assists control of gastrointestinal nematodosis of sheep in a cool temperate environment with summer-dominant rainfall A.F. Colvin a,*, S.W. Walkden-Brown a, M.R. Knox b, J.M. Scott c a

Centre for Animal Health and Welfare, School of Rural Science and Agriculture, University of New England, NSW 2351, Australia b CSIRO Livestock Industries, F D McMaster Laboratory-Chiswick, Locked Bag 1, Armidale, NSW 2350, Australia c Centre for Sustainable Farming Systems, University of New England, NSW 2351, Australia Received 29 August 2006; received in revised form 10 January 2008; accepted 11 January 2008

Abstract While rotational grazing methods have an accepted role in the management of gastrointestinal nematodosis (GIN) of small ruminants in humid tropical regions, their efficacy and application in cool temperate regions is more controversial. This study evaluated GIN over 2 years in three classes of fine wool Merino sheep (lambs, hoggets and ewes) under three different sheep management systems in a cool tableland environment near Armidale NSW Australia (950 m altitude, 308310 S, 1518390 E). The management systems were High input (HI) with high fertiliser inputs, a target of 100% sown pasture, high stocking rate (13.4 dry sheep equivalents/ha) and relatively long grazing periods; Typical (TYP) New England management system with moderate fertiliser inputs and stocking rate (9.3 DSE/ha) and relatively long grazing periods; and Intensive rotational grazing (IRG) with moderate fertiliser inputs and stocking rate (8.8 DSE/ha) but very short (mean 5 days) grazing periods and long (mean 103 days) rest periods. Twenty sheep of each class in each management treatment were sampled monthly for faecal worm egg counts (WEC, followed by larval differentiation), and body weights with a blood sample taken for haematology every second month. The proportion of sheep with WEC above zero did not differ between management systems but the magnitude of WEC did, with sheep under IRG displaying lower mean WEC than those on the other treatments (IRG: 326, HI: 594, TYP: 536, eggs/g P < 0.0001). This was despite a significantly longer mean interval between anthelmintic treatments (IRG: 144 days, HI: 77 days, TYP: 78 days, P < 0.0001). The IRG management system also influenced the composition of the infections with sheep on this treatment having a significantly lower proportion of Haemonchus contortus in their faecal cultures (IRG: 59.7%, HI: 79.4%, TYP: 80.9%, P < 0.05) and a significantly higher proportion of Trichostrongylus spp. Sheep on the IRG treatment also had a significantly higher haematocrit (HCT) than those on the other management systems. Despite the lower WEC and higher HCT, sheep under IRG also had significantly lower bodyweights and fleece weights overall, although this was only evident in sheep raised to adulthood prior to the experiment, not those raised during the 2-year experimental period. The results demonstrate that IRG systems with short grazing periods and long rest periods between grazing events can assist with control of GIN in cool temperate climates where H. contortus is the dominant parasite. # 2008 Elsevier B.V. All rights reserved. Keywords: Sheep-nematoda; Rotational grazing; Grazing management; Haemonchus contortus; Trichostrongylus

* Corresponding author at: Centre for Animal Health and Welfare, School of Rural Science and Agriculture, University of New England, NSW 2351, Australia. Tel.: +61 2 67733239. E-mail address: [email protected] (A.F. Colvin). 0304-4017/$ – see front matter # 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.vetpar.2008.01.014

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

1. Introduction Our understanding of the ecology of the free-living stages of parasitic nematodes suggests that rotational grazing systems could assist in the control of gastrointestinal nematodosis (GIN) of sheep by interrupting the nematode lifecycle (Donald, 1967). However, up to the late 1980s, there was little success in developing and implementing practical rotational grazing systems that reduced GIN. Early studies on rotational grazing in cool temperate environments involved grazing periods of 7 days and rest periods between grazing events ranging from 3 to 7 weeks (Morgan, 1933; Morgan and Oldham, 1934; Roe et al., 1959; Gibson and Everett, 1968). These rotations, however, were ideal for the proliferation of parasitic nematodes allowing both autoinfection during the first grazing period, and timing the next grazing at the peak of L3 availability. Thus, time of development from egg to L3 is an important consideration for the length of the grazing period. Haemonchus contortus will develop from egg to L3 in 3–5 days at 25–26 8C but will take 15– 30 days at 10–11 8C (Rose, 1963). Season therefore determines the length of safe grazing periods that prevent autoinfection. The time of peak L3 on pasture in the Sydney Basin, NSW is generally around 35 days after deposition with smaller peaks at days 14 and 28 (Donald, 1967). This author concluded that the spelling period for a paddock should be no less than 8 weeks to enable a significant reduction in pasture infectivity. This may also vary with season as L3 on pasture survive longer in cooler conditions than warm or hot conditions (Ransom, 1906; Monnig, 1930; Dinaburg, 1944a,b; Silverman and Campbell, 1959; Thomas and Boag, 1972; Levine et al., 1974; Southcott et al., 1976; Besier and Dunsmore, 1993). As recognised by Donald (1967) such long rest periods may be inefficient in terms of optimal pasture utilisation. Such inefficiency was demonstrated in the study of Robertson and Fraser (1933) in which ‘progressional grazing’ comprising 10 days grazing followed by 100 days rest significantly reduced the level of infection with H. contortus. However, over-mature grass, resulting from the long rest period, undermined the success of the grazing system with sheep failing to maintain body weight despite lower parasite burdens. More recently Eysker et al. (2005) working in the Netherlands found that a rotational grazing system based on a grazing period of 4 weeks was not sufficient for worm control, with graze periods of 3 or 2 weeks being more successful. The authors proposed grazing for 2 weeks coupled with 3 months rest for sufficient

109

control of H. contortus in that climate. However, there were still significant summer infections in lambs grazed on this regime (mean WEC > 3000 eggs/g). It is likely that such long rest periods would also result in major reductions in pasture quality. In recent times more success has been had in the tropics, based on the findings of Banks et al. (1990) in Fiji on the rates of larval development and survival in hot, humid environments. An effective rapid rotational grazing system based on a grazing period of 3.5 days and a rest period of 31.5 days was developed and has been used with success throughout the tropics in both sheep and goats (Barger et al., 1994; Chandrawathani et al., 1995; Sani et al., 1996; Gray et al., 2000). However, Banks et al. (1990) and Barger et al. (1994) both suggested that such a rapid rotational grazing system would not be economically viable in cooler climates, presumably because the rigid application of timing of graze and rest periods would not be suitable given the seasonal variability of temperature and rainfall in these regions. In the early 1990s in the temperate regions of Australia, intensive rotational grazing (IRG) systems such as ‘cell grazing’ and ‘holistic grazing’ were introduced based on claims of improved pasture production, quality and sustainability (Earl and Jones, 1996; McCosker, 2000; Sparke, 2000). They involve the use of large groups of animals at high stock densities moving through a series of 20–40 paddocks at a rate dependant on the amount of feed on offer and pasture growth rate rather than rigid time periods. The grazing period generally ranges from 1 to 3 days with rest periods of 40–90 days, resulting in paddocks being rested for 90–95% of the year (Earl and Jones, 1996). This type of grazing management has become increasingly used throughout Australia with its highest prevalence being in the New England region on the Northern Tablelands of NSW (Reeve and Thompson, 2005). The consequences of such intensive grazing systems on GIN in sheep in cool temperate climates have not been documented despite considerable anecdotal evidence of marked reductions in faecal worm egg counts (WEC) and the number of anthelmintic treatments required. We investigated this issue on the Cicerone Project, a long-term project comparing three different sheep management systems (including an Intensive rotational grazing treatment) in the New England region of Northern NSW. An analysis of historical data from the project had indicated that significant differences in WEC between management systems did occur and were worthy of more detailed investigation (Healey et al., 2004).

110

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

Table 1 Summary of management system treatments on the Cicerone Project Inc. Management system

Grazing/Pasture management Rest periods Graze periods Number of paddocks Percentage of sown pastures (%) Target soil phosphorus (ppm) Target soil sulphur (ppm) Average stocking rate (DSE*) Average number of sheep (2003–2005) *

High input (HI)

Typical (TYP)

Intensive rotational grazing (IRG)

Flexible using Prograze1 principles Moderate Long 8 50–80 60 10 13.4 581

Flexible using Prograze1 principles Moderate Long 8 30–50 20 6.5 9.2 444

Intensive rotational Long Short 64 30–50 20 6.5 8.8 395

Dry sheep equivalents/ha based on sheep numbers in each treatment.

The aims of this experiment were to (a) determine whether management system influences the incidence and severity of GIN as determined by WEC, blood parameters and animal performance, (b) determine whether the effects of management system differ for the different major gastrointestinal nematode species and (c) gain insight into possible underlying reasons for observed management system effects on WEC. 2. Materials and methods 2.1. Experimental design and location This experiment was a longitudinal study of GIN under three farm management systems over a 2 year period from November 2003 to October 2005. The key factors in the design were:  Three management systems: High input (HI), Typical (TYP) and Intensive rotational grazing (IRG) (Table 1).

 Three classes of sheep: lambs, hoggets and ewes.  Time: the study covered 2 years, year 1 (November 2003 to October 2005) and year 2 (November 2004 to October 2005). The experiment took place on the Cicerone Project farm located 18 km south of Armidale, NSW, Australia (950 m altitude 308310 S, 1518390 E). Climatic data for Armidale during the experimental period are presented in Fig. 1. It has a cool temperate climate with the majority of rain falling in the summer months. Winters are generally dry, cold and frosty with occasional light snowfall. Summer days are warm but nights are generally cool. The average annual rainfall is 790 mm. 2.2. Management system treatments The Cicerone Project Inc. is a producer-led group aiming to increase the profitability and sustainability of grazing based agriculture in the New England region. It

Fig. 1. Mean monthly rainfall (black bar), evaporation (grey bar), minimum (solid line) and maximum (dashed line) temperatures (8C) at Armidale for the experimental period November 2003 to October 2005.

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

111

Table 2 Total days of supplementary feed per sheep class under each management system over the experimental period from November 2003 to October 2005 Management system

Class

Total days fed

Average rate of feed (kg/sheep/week)

Year 1

Year 2

Year 1

Year 2

High input (HI)

Ewes Hoggets Lambs

118 133 0

97 0 271

1.3 1.0 1.0

2.5 0.0 3.5

Typical (TYP)

Ewes Hoggets Lambs

111 133 0

77 0 272

0.7 1.0 0.0

2.0 0.0 1.7

Intensive rotational grazing (IRG)

Ewes

118

91

0.7

1.2

Hoggets Lambs

133 0

0 272

1.0 0.0

0.0 1.7

commenced in 1998 with development of three farmlets, each approximately 50 ha, which were managed under three different management systems. The sub-division of land was a complex process of apportioning paddocks equally to each system on the basis of hydrology, slope, soil type and fertiliser history (Scott et al., 2004). Thus, the management systems comprise interspersed allocations of paddocks rather than being geographically discrete sections. Stocking of the farm and implementation of the management systems (Table 1) commenced in July 2000. Use of supplementary feeding was limited and is summarised in Table 2. Pasture dry matter availability under each of the management systems throughout the experimental period is shown in Fig. 2. The HI treatment had lower total dry matter with a higher proportion of green dry matter than treatments CON and IRG, which were similar. 2.3. Experimental sheep and their husbandry The experimental sheep were fine wool merinos. The classes of sheep monitored during the experiment were all females and defined as ewes (females over 2 years of age) hoggets (12–24 months of age) or lambs (<12

months of age). In November 2003, 20 ewes, 20 hoggets (September 2002 born) and 20 lambs (September 2003 born) from each management treatment were randomly selected and individually ear-tagged. These 180 experimental animals were sampled each month throughout the experiment, with lambs becoming hoggets and hoggets becoming ewes in November 2004. In that month an additional 20 lambs (September 2004 born) from each management treatment were included thus bringing the total number sampled in year 2 up to 240. Mating occurred over a 5-week period in April/May each year with ewes and hoggets from all management treatments moved to a common periphery paddock for joining with common sires to ensure similar genetics across management systems. Lambing started in mid-September and was finished by the end of October with lamb marking in mid-November and weaning in late January. Shearing of all sheep took place at the end of July each year in an off-site shearing shed. Anthelmintic treatments comprised two fixed treatments across all management treatments and tactical treatments applied within management groups on the basis of monitored WEC, larval differentiation and advice from the project’s consulting veterinarian. The

Fig. 2. Total pasture green dry matter (black) and dead dry matter (white) by management treatment over the experimental period (Mulchay, C. unpublished).

112

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

fixed treatments comprised a quarantine treatment given to all sheep as they moved off the property for shearing in July and a pre-mating treatment for all ewes in March–April prior to grouping in a paddock external to the management treatment areas. All anthelmintic treatments for the duration of the experiment are detailed in Table 3. The anthelmintics used were moxidectin (Cydectin1, Fort Dodge Australia Pty Ltd.), albendazole oxide with levamisole hydrochloride (Combi1, Novartis Animal Health Australasia Pty Ltd., Australia), levamisole hydrochloride (Levamisole Gold1, Virbac, Australia) and naphthalophos (Rametin1, Bayer Healthcare, Australia) at manufacturers’ recommended dose rates unless otherwise specified.

2.4. Measurements The date of all stock movements into and out of paddocks was recorded. Each month all of the experimental sheep were faecally sampled per rectum for determination of WEC, and bodyweight recorded. Every second month a blood sample in K3EDTA tubes was collected for haematological analysis. WEC were performed in random order using a modification of the McMaster technique. Faeces left over from the individual counts were bulked within management treatment and class, mixed with vermiculite, moistened and incubated at 27 8C for 7 days. Larvae were then recovered and 100 L3 differentiated into species

Table 3 Anthelmintic treatments during the experimental period by sheep class and management treatment; moxidectin (MOX), albendazole (ABZ), levamisole (LEV), napthalophos (NAP) Class

Ewes

Hoggets

Lambs

Total treatments

Date of treatment

Anthelmintic treatment High input (HI)

Typical management (TYP)

Intensive rotational grazing (IRG)

23/12/2003 19/04/2004 * 30/07/2004 * 17/11/2004 14/12/2004 24/01/2005 24/02/2005 23/04/2005 * 23/07/2005 *

MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV LEV double dose NAP/ABZ NAP/ABZ NAP/ABZ/LEV

MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV LEV double dose NAP/ABZ NAP/ABZ NAP/ABZ/LEV

MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV

23/12/2003 19/04/2004 * 30/07/2004 * 28/9/2004 15/11/2004 5/01/2005 24/01/2005 24/02/2005 23/04/2005 * 23/07/2005 *

MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV LEV double dose

MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV

MOX + ABZ + LEV LEV double dose MOX + ABZ + LEV

23/12/2003 10/03/2004 8/05/2004 2/07/2004 30/07/2004 * 28/09/2004 14/12/2004 24/01/2005 24/02/2005 23/04/2005 23/07/2005 *

MOX LEV double dose LEV double dose LEV double dose MOX + ABZ + LEV LEV double dose MOX LEV double dose NAP/ABZ NAP/ABZ NAP/ABZ/LEV

MOX LEV double dose LEV double dose LEV double dose MOX + ABZ + LEV

LEV double dose

MOX LEV double dose NAP/ABZ NAP/ABZ NAP/ABZ/LEV

NAP/ABZ

28

27

15

LEV double dose NAP/ABZ NAP/ABZ NAP/ABZ/LEV

LEV double dose MOX LEV double dose NAP/ABZ NAP/ABZ NAP/ABZ/LEV

NAP/ABZ NAP/ABZ/LEV

NAP/ABZ NAP/ABZ/LEV MOX

MOX + ABZ + LEV

NAP/ABZ/LEV

Where LEV was given without other anthelmintics it was administered at double the recommended dose rate under veterinary advice (Dr. E. Hall, personal communication). * Fixed treatments imposed for reasons other than perceived risk of GIN.

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

(H. contortus, Teladorsagia circumcincta) or genus (Trichostrongylus, Oesophogostomum or Cooperia) after iodine staining. Blood cell parameters were measured using a Cell Dyn 3500 automated haematology analyser (Abbott Diagnostics, USA) specifically calibrated for sheep. The majority of samples were analysed on the day of collection with a maximum delay of 24 h. 2.5. Statistical analysis Data for several variables were transformed prior to analysis: WEC, log10(x +p1); eosinophil data, cubedffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi root; HCT (%), ArcSineð proportionÞ. WEC data for individual species or genera were derived by multiplying total WEC by the proportion of larvae of the relevant species in the pooled culture. WEC data for all strongyle species (except Nematodirus spp.) was is referred to as total WEC, that attributable to H. contortus is referred to as HcWEC and that attributable to Trichostrongylus spp. is referred to as TWEC. There were insufficient data for meaningful analysis of the other individual species. Data were analysed by fitting appropriate linear mixed models followed by analysis of variance using the statistical package JMP IN version 5.1 (SAS Institute Inc., NC, USA). The effects tested in the models were: management treatment (HI, TYP, or IRG), class (Ewe, Hogget or Lamb), month, season or year and tag number (fitted as a random variable for repeated measures data). Seasons were spring/summer (September to February), autumn/winter (March to August). Significant two- and three-way interactions were retained in the model. Appropriate covariates were included in the models to either account for their effects (retained in the model) or to explore the effect on the variable under analysis (not retained in the model). The interval between anthelmintic treatments was analysed using Group (original Ewes, 2002-born, 2003-born and 2004-born sheep) in place of class, as the changeover of animals between classes would otherwise confound the data. Tukey’s HSD post hoc test or specific linear contrasts were used to test for significant differences between means. As no transformation made the entire WEC dataset amenable to analysis with a linear mixed model due to a high proportion of zero values, it was subjected to two analyses. Analysis 1 used a nominal logistic model to analyse the occurrence of GIN (i.e. WEC zero or >0). Analysis 2 analysed the magnitude of WEC in those animals where WEC > 0 using linear mixed models as described above. A level of significance of P < 0.05 is used throughout and data are presented as least square

113

means  standard error (L.S.M.  S.E.M.) or in the case of transformed data, back transformed LSM with 95% confidence intervals. 3. Results 3.1. Implementation of treatments 3.1.1. Paddock rotations Overall, raw mean graze period on IRG was shorter (5  0.3 days) and raw mean rest period longer (103  5 days) than HI (Graze: 63  9 days; Rest: 65  8 days) and TYP (Graze: 80  12 days; Rest: 89  14 days). Graze periods were shorter on IRG in spring/summer compared with autumn/winter and the converse was true for rest periods in the seasons, however this was not the case on the other management treatments (Fig. 3). 3.2. Parasitological variables Fig. 4 shows total WEC, HcWEC and TWEC for each management treatment and class throughout the experiment and the timing and type of anthelmintic treatment given. 3.2.1. Interval between anthelmintic treatments There was a significant overall effect of management system on anthelmintic treatment interval (P < 0.0001) with IRG having the longest interval at 144 days, nearly double that of management treatments HI (77 days) and TYP (78 days). There was a significant interaction between management treatment and year with the interval between treatments declining for HI and TYP from year 1 to year 2 (HI: 96  11 and 58  7 days, TYP: 97  11 and 59  7 days) but increasing for IRG (114  14 and 173  1 2 days). There was no significant effect of year (P  0.48), group (P  0.11), nor any interaction between management treatment and group (P  0.97).

Fig. 3. Raw mean rest period (grey) and raw mean graze period (white) with standard errors by season and management treatment.

114

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

Fig. 4. Arithmetic mean ewe (a) hogget (b) and lamb (c) total worm egg counts (white), Haemonchus contortus worm egg counts (grey) and Trichostrongylus spp. worm egg counts (black) over the experimental period with anthelmintic treatments indicated by arrows moxidectin (white), short acting (black), weaning (W), mating (J) and quarantine (Q). Management treatments – HI: High input, TYP: Typical and IRG: Intensive rotational grazing.

3.2.2. Faecal worm egg count by species or genus The overall raw mean proportion of nematode species and genera determined by larval differentiation of WEC were H. contortus (70%), Trichostrongylus spp. (22%), T. circumcincta (4%), Oesophagostomum spp. (3%) and Cooperia spp. (1%). There were higher proportions of H. contortus on the HI and TYP treatments than the IRG (76.3, 79.7 and 59.5%, respectively). The IRG treatment had the highest proportion of Trichostrongylus spp. (HI: 20.8; TYP: 14.3; IRG: 27.9) and T. circumcincta (HI: 2.2; TYP: 2.1; IRG: 8.6). 3.2.2.1. Analysis 1––Incidence of positive faecal worm egg count. Nominal logistic analysis revealed no overall effect of management treatment on the

incidence of positive (WEC > 0) total WEC (HI: 69.1%; TYP: 65.1% and IRG: 64.1%, P  0.99), HcWEC (HI: 66.7%; TYP: 63.0% and IRG: 60.9%, P  0.99) or TWEC (HI: 56.9%; TYP: 48.6% and IRG: 62.9%, P  0.99). However, in each case there was a significant interaction between the effects of management treatment and class (P < 0.0001) as illustrated in Fig. 5. Month was also significant as would be expected with seasonal fluctuations in the incidence of WEC (P < 0.0001). 3.2.2.2. Analysis 2––Magnitude of WEC in WECpositive sheep. HI and TYP management treatments had a higher total WEC than IRG (546, CI 389–713; 582, CI 414–763 and 304, CI 200–414 eggs/g,

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

115

Fig. 5. Mean (SEM) incidence of positive faecal worm egg counts (%) showing the management treatment by class interaction for all strongyle species, H. contortus and Trichostrongylus spp. Means within management treatments not sharing a common letter differ significantly (P < 0.05). Management treatments – HI: High input, TYP: Typical and IRG: Intensive rotational grazing. Class – E: ewes, H: hoggets, L: lambs.

respectively, P < 0.0001). All treatments were significantly different from each other for HcWEC, HI sheep having the highest mean followed by TYP then IRG (555, CI 469–647; 438, CI 372–508 and 209 CI 173– 249 eggs/g, respectively, P < 0.0001). The situation was slightly different for TWEC with the HI management treatment having a significantly higher mean (51, CI 43–60 eggs/g) than IRG (39, CI 32–46 eggs/g) and TYP (37, CI 31–44 eggs/g, P < 0.001). A significant interaction between management treatment, class and year was also revealed for total WEC (P < 0.0001, Fig. 6). Bodyweight, fitted as a covariate, had a significant effect on HcWEC (P < 0.0001) with a weak, but significant negative association (R2 = 0.03, P < 0.0001). However, it did not significantly affect total WEC (P  0.11) or TWEC (P  0.61). Anthelmintic treatment interval (days since last treatment) also fitted as a regressor had a significant positive association with WEC (P < 0.0001), as would be expected. Neither bodyweight nor treatment interval were retained in the final model. The effect of year was significant (P < 0.0001) with higher mean WEC in year 2 (635, CI 541– 737 eggs/g) than year 1 (445, CI 378–518 eggs/g).

3.3. Bodyweight Bodyweights were higher on HI and TYP than IRG (39.3  0.3, 39.1  0.3 and 38.1  0.3 kg, respectively, P < 0.05). Initial bodyweight was a significant covariate (P < 0.0001) and was retained in the model. The overall effects of month (P < 0.0001) and year (P < 0.0001) were also significant with higher bodyweights in year 1 (40.4  0.2 kg) than year 2 (37.3  0.2 kg). Fitting initial bodyweight as a covariate did nothing to change the lamb bodyweights, but it did correct some initial differences in the hoggets and in the ewes in year 2 (Fig. 7). These overall effects masked significant interactions between management treatment and year (P < 0.0001), management treatment and month (P < 0.0001) and management treatment and class (P < 0.0001). The management treatment by year interaction was due to large declines in bodyweight between years 1 and 2 on HI and TYP treatments while bodyweights on the IRG treatment declined far less. The management treatment by class interaction was due to ewe and hogget bodyweights on the IRG treatment being significantly

Fig. 6. Total WEC (Back transformed LSM with 95% confidence intervals) showing significant interaction between the effects of year, management treatment and class. Means within years not sharing a common letter differ significantly (P < 0.05). Management treatments – HI: High input, TYP: Typical and IRG: Intensive rotational grazing. Class – E: ewes, H: hoggets, L: lambs.

116

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

Fig. 7. Least squares means for monthly bodyweight by management treatment (HI-solid line, TYP-dashed line, IRG-dotted line) and class over the experimental period. Vertical dotted lines indicate the transition between classes. Management treatments – HI: High input, TYP: Typical and IRG: Intensive rotational grazing.

lower than for HI or TYP, while the lamb bodyweights did not differ between management treatments (Fig. 7). 3.4. Haematocrit Sheep in the IRG treatment had higher HCT values than those on the TYP treatment with those on the HI treatment being intermediate (34.5, CI 33.5–35.5%; 33.3, CI 32.3–34.3%; and 34.1, 33.1–35.1%, respectively, P < 0.05). Overall, lambs had significantly higher HCT (36.9, CI 35.5–38.4%) than hoggets (33.3, CI 32.5–64.1%) which in turn had higher HCT than ewes (31.7, CI 30.7–32.6%) (P < 0.0001). The effects of month (P < 0.0001) and year (P < 0.0001) were also significant with significant interactions between management treatment and class (P < 0.05), management treatment and month (P < 0.05), management treatment and year (P < 0.05), and class and year (P < 0.0001). Bodyweight had a significant positive relationship with HCT and was retained in the model (P < 0.0001). HCT values decreased from year 1 to year 2 for HI (Year 1: 35.5, CI 35.0–35.0%; Year 2: 32.7, CI 32.2–33.2%) and TYP (Year 1: 34.4, CI 34.0– 34.8%; Year 2: 32.2, CI 31.7–32.6%) treatments but not the IRG treatment (Year 1: 35.1, CI 34.6–35.5%; Year 2: 33.9, CI 33.1–34.4%).

year 2 than year 1 (167, CI (141–193)  106/ml and 127, CI (109–145)  106/ml, respectively). The effect of month was significant (P < 0.0001) and there was also a significant positive effect of bodyweight when fitted as a covariate (P < 0.0001). There were significant interactions between management treatment and class (P < 0.001), management treatment and month (P < 0.001), class and year (P < 0.0001) and management treatment, class and year (P < 0.05) which can be visualised in Fig. 8. Eosinophil counts for ewes and lambs did not differ between years on all management treatments while in each case hoggets had higher eosinophil counts in year 2 than year 1. The eosinophil counts were not particularly high and remained within the normal range of 0–1000 million cells/l (Radostits et al., 2000).

3.5. Eosinophil count There was no significant effect of management treatment on total eosinophil count (P  0.99) but there was a significant effect of class (P < 0.0001) with higher eosinophil counts in hoggets (180, CI (152– 209)  106/ml) than ewes and lambs which did not differ (147, CI (117–176)  106/ml and 116, CI (78–153)  106/ml, respectively). Eosinophil counts differed between years (P < 0.0001) being higher in

Fig. 8. Eosinophil count (Back transformed LSM with 95% confidence intervals) showing interaction between management treatment (HI: white, TYP: grey, IRG: striped), class and year. Means within management treatments not sharing a common letter differ significantly (P < 0.05). Management treatments – HI: High input, TYP: Typical and IRG: Intensive rotational grazing. Class – E: ewes, H: hoggets, L: lambs.

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

4. Discussion The main finding of this experiment was that sheep on the IRG treatment had significantly lower total WEC (46% lower) and HcWEC (53% lower) than those under the HI and TYP management systems. This was despite receiving approximately half the number of anthelmintic treatments (15) across the three classes of sheep over two years, than either the HI (28) or TYP treatments (27). Trichostrongylus WEC was not reduced by IRG, suggesting a specific action on H. contortus. Despite the difference in WEC, there was no advantage in bodyweights to sheep on the IRG system, and eosinophil counts did not differ, suggesting that the observed management system effects are less likely to be hostmediated than mediated by effects on the environmental phase of the nematode lifecycle. Support for this is found in the striking differences in grazing and rest periods between the IRG and the other treatments; differences which on the basis of existing epidemiological knowledge would predict large effects on infections with H. contortus, but less so for Trichostrongylus species. These findings are discussed in more detail below in light of the original objectives. The incidence of sheep with positive WEC did not differ much between management systems being approximately 65% overall. Similarly, there were no differences in the incidence of those with positive HcWEC or TWEC. However, the number of anthelmintic treatments given on the management treatments needs to be considered. With management treatments HI and TYP receiving almost double the number of anthelmintic treatments given relative to IRG, these management treatments were sampled for WEC soon after an anthelmintic treatment on more occasions than sheep on the IRG management treatment, which had much longer anthelmintic treatment intervals. Thus, although management treatments were statistically similar in the proportion of sheep with positive WEC, the IRG management treatment may actually have had a lower incidence of GIN had the number of anthelmintic treatments been standardised across treatments. In contrast to the lack of effect on the incidence of worm infection, the magnitude of infection (as determined by analysis of WEC values above zero) was markedly affected by management system, with significantly lower values for total WEC on the IRG treatment than the other two treatments, which did not differ. The pattern of worm infestation over both years of the experiment followed established trends for this region (Gordon, 1948; Roe et al., 1959; Barger et al.,

117

1972; Southcott et al., 1976). High spring and early summer rainfall resulted in heavy infections of H. contortus in those seasons, along with relatively high Trichostrongylus spp. infections. Faecal worm egg count was relatively low for all species in autumn and winter. Sheep in the IRG treatment completely avoided the very large peaks of WEC in late spring and summer observed in all classes of sheep in the other treatments. In November and December 2004, there were losses of lambs and ewes with acute haemonchosis in the HI and TYP treatments (10–15 ewes and lambs on each of these management systems). Not only was WEC lower on the IRG treatment throughout the year with no deaths occurring, but this was achieved with a much lower level of anthelmintic treatment. There was a higher mean WEC in year 2 than year 1 associated with acute haemonchosis in HI and TYP sheep in the spring and summer of year 2. There was clear evidence of a differential effect of IRG on nematode species with the reduction of WEC observed on the IRG treatment mostly attributable to reductions in H. contortus WEC. The effect of IRG on Trichostrongylus spp. was less clear with TWEC on the IRG treatment being equivalent to that of the TYP treatment and significantly lower than that of the HI treatment. This indicates that both the IRG and TYP treatments effectively reduced TWEC relative to HI. There are two lines of evidence suggesting an effect of IRG on Trichostrongylus spp., albeit a smaller effect than on Haemonchus. Firstly, there is the significant reduction in TWEC relative to the HI treatment and, secondly, there is the fact that TWEC on the IRG treatment was equivalent to that on the TYP treatment, despite the IRG treatment receiving 12 fewer anthelmintic treatments over the experimental period. This strongly suggests a suppressive effect of IRG on TWEC had the anthelmintic treatments been the same across treatments. The reasons for the higher TWEC on the HI treatment relative to the TYP treatment are not clear. They can possibly be attributed to the higher stocking rate, however, there is conflicting evidence of the effect of stocking rate on worm burdens, with some studies reporting no effect, while others report a significant effect related to worm species (Beveridge et al., 1985; Brown et al., 1985; Waller et al., 1987; Thamsborg et al., 1996). A limitation on our interpretation of the Trichostrongylus WEC is the association between WEC and worm counts in this species due to densitydependant effects on fecundity (Dobson et al., 1990). Infection levels with the other nematode species found were insufficient for a meaningful analysis of their burdens, but the considerably greater contribution of T.

118

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

circumcincta to infections on the IRG treatment suggests that it is more resistant to the effects of IRG than either Haemonchus or Trichostrongylus. This is not surprising given the ability of T. circumcincta to overwinter, develop at low temperatures (below 4 8C), and to survive desiccation in the pre-hatch stage (Kates, 1950; Gibson and Everett, 1972; Rossanigo and Gruner, 1995). It is not clear whether the highly effective rapid rotational grazing system of Barger et al. (1994) in the tropics was effective against both H. contortus and Trichostrongylus spp., but the authors did note that all genera appeared to be remarkably similar in their ability to hatch and develop to L3 and survive under Tongan conditions. Banks et al. (1990) reported similar results in Fiji. This is unlikely to be the case in the cool temperate climate of Armidale which is more conducive to the development and survival of Trichostrongylus spp. eggs and larvae between grazing events (Anderson et al., 1966; Waller and Donald, 1970; Levine and Anderson, 1973; Levine et al., 1974). On the other hand, the susceptibility of H. contortus eggs to desiccation and low temperatures is likely to make them just as susceptible to rapid rotational grazing as they would be under tropical conditions, but probably for different reasons. In the humid tropics, a higher proportion of eggs deposited on pasture would hatch and develop into infective larvae than in this cool temperate climate, thus control of H. contortus would hinge more on rate of larval decay than it would in cooler climates. On the other hand, temperature and moisture are often limiting in the New England for development of H. contortus eggs to L3 (O’Connor et al., 2006, 2007), and failure of development is likely to be a major mechanism operating in this environment. The superior ability of Trichostrongylus species to survive in the embryonated egg stage during cool dry conditions relative to Haemonchus could account for the species difference in response to IRG observed in this experiment. Of great importance in the IRG system, both in the tropics and in cool temperate regions, is the short grazing period which precludes autoinfection from the current grazing. This ensures that the nematodes must run the gauntlet of high L3 death rates in the tropics, or intermittent development and/or high death rates in cool temperate regions, prior to being presented to host animals at the next grazing event. It is unlikely that the effects of IRG were mediated by improved host immunity either through better host nutrition or greater exposure to infective larvae. The older sheep on the IRG treatment were 2–3 kg lighter than those on the other treatments, suggestive of an

inferior nutritional status, yet these animals had consistently lower HcWEC and higher HCT values. Lambs on the other hand did not differ in bodyweight between management treatments, yet the effects of IRG were seen consistently across each class of sheep irrespective of differences in productivity. Mean circulating eosinophil count also did not vary between management treatments. The lower overall levels of sheep productivity seen on the IRG treatment appear to be due mainly to effects preceding the present experimental period. This is best illustrated in the bodyweights, with major differences in ewe bodyweight present at the start of the experiment, largely maintained thereafter. Similarly, bodyweight in hoggets differed widely prior to the start of the experiment, with marked convergence between treatments by the end of the first year. In the second year there are no treatment differences between hoggets. For lambs, there are no major treatment effects in either year, suggesting that, for animals born and reared during the experimental period, differences in productivity were minimal. Nevertheless, there are important challenges with implementation of IRG treatments and it must be remembered that they are used primarily to improve animal production and the sustainability of grazing systems rather than to limit the effects of GIN. While IRG systems are implemented for animal and pasture management, they may become an important tool for integrated parasite management in the face of increased anthelmintic resistance and the demand for more sustainable animal production systems. We conclude that IRG systems such as that used in this experiment are able to markedly reduce WEC in all major classes of stock. This action appears to be preferential for H. contortus and we propose that it is mediated by effects on the free-living stages of the lifecycle rather than on the host. The mechanisms behind the observed effects are currently under investigation. We believe that these findings demonstrate significant potential for the use of IRG for integrated parasite management for the control of H. contortus in temperate environments. Further studies on the impact of such systems in regions where Trichostrongylus spp. or T. circumcincta predominate is warranted. Acknowledgements The Cicerone Project Inc. is supported by the Australian Wool Innovation and member subscriptions. Dr. A. Colvin held an Australian Sheep Industry CRC Ph.D. scholarship and received separate project funding

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

from the Australian Wool Innovation for which we are grateful. We thank Dr. Betty Hall and Ian Barger for their helpful discussions in the planning stages of this work. We also thank Mr. Justin Hoad (Cicerone Project Farm Manager) and Mrs. Caroline Gaden (Past Cicerone Project Inc. co-ordinator), our technical assistants Mr. Michael Raue, Mr. Neill Baillie and Mr. John Gorham and the many others who helped with the sampling. We acknowledge the use of pasture data collected by Mr. Colin Mulchay on behalf of the Cicerone Project Inc. and weather data collected by Mr. P. Burr. References Anderson, F.L., Wang, G., Levine, N.D., 1966. Effect of temperature on survival of the free-living stages of Trichostrongylus colubriformis. J. Parasitol. 52, 712–721. Banks, D.J.D., Singh, R., Barger, I.A., Pratap, B., Le Jambre, L.F., 1990. Development and survival of infective larvae of Haemonchus contortus and Trichostrongylus colubriformis on pasture in a tropical environment. Int. J. Parasitol. 20, 155–160. Barger, I.A., Beynon, P.R., Southcott, W.H., 1972. Simulation of pasture larval populations of Haemonchus contortus. In: Proceedings of the Australian Society of Animal Production, vol. 9. pp. 38–42. Barger, I.A., Siale, K., Banks, D.J.D., Le Jambre, L.F., 1994. Rotational grazing for control of gastrointestinal nematodes of goats in a wet tropical environment. Vet. Parasitol. 53, 109–116. Besier, R.B., Dunsmore, J.D., 1993. The ecology of Haemonchus contortus in a winter rainfall climate in Australia: the survival of infective larvae on pasture. Vet. Parasitol. 45, 293–306. Beveridge, I., Brown, T.H., Fitzsimons, S.M., Ford, G.E., Judson, G.J., Martin, R.R., Miller, D.W., 1985. Mortality in weaner sheep in South Australia under different regimes of anthelmintic treatment. Aust. J. Agric. Res. 36, 857–865. Brown, T.H., Ford, G.E., Miller, D.W., Beveridge, I., 1985. Effect of anthelmintic dosing and stocking rate on the productivity of weaner sheep in a mediterranean climate environment. Aust. J. Agric. Res. 36, 845–855. Chandrawathani, P., Jamnah, O., Rajamanickham, C., 1995. Testing potential control programmes for helminthosis in sheep on open pasture. In: 17th Malaysian Society of Animal Production Annual Conference, vol. 17, Penang, pp. 114–115. Dinaburg, A.G., 1944a. Development and survival under outdoor conditions of eggs and larvae of common ruminant stomach worm, Haemonchus contortus. J. Agric. Res. 69, 421–433. Dinaburg, A.G., 1944b. The survival of the infective larvae of the common ruminant stomach worm, Haemonchus contortus, on outdoor grass plots. Am. J. Vet. Res. 5, 32–37. Dobson, R.J., Waller, P.J., Donald, A.D., 1990. Population dynamics of Trichostrongylus colubriformis in sheep: the effect of infection rate on the establishment of infective larvae and parasite fecundity. Int. J. Parasitol. 20, 347–352. Donald, A.D., 1967. Populations of strongyloid infective larvae in pastures after sheep are removed from grazing. Aust. Vet. J. 43, 122–128. Earl, J.M., Jones, C.E., 1996. The need for a new approach to grazing management––is cell grazing the answer? Rangeland J. 18, 327–350.

119

Eysker, M., Bakker, N., Kooyman, F.N.J., Ploeger, H.W., 2005. The possibilities and limitations of evasive grazing as a control measure for parasitic gastroenteritis on small ruminants in temperate environments. Vet. Parasitol. 129, 95–104. Gibson, T.E., Everett, G., 1968. A comparison of set stocking and rotational grazing for the control of Trichostrongylosis in sheep. Br. Vet. J. 124, 287. Gibson, T.E., Everett, G., 1972. The ecology of the free-living stages of Ostertagia circumcincta. Parasitology 64, 451–460. Gordon, H.M., 1948. The epidemiology of parasitic diseases, with special reference to studies with nematode parasites of sheep. Aust. Vet. J. 26. Gray, G.D., Anchetta, P.A., Arguzon, J.A., Barcelo, P.M., Casis, C., Cerbito, W.A., Dobson, R.J., Dumilon, R., Cruz, E.M., Knox, M.R., Le Jambre, L.F., Mateo, K.S., Suba, M.S., Villar, E.C., 2000. Better worm control for goats and sheep in the Philippines. AsianAust. J. Anim. Sci. 13, 548. Healey, A.F., Hall, E., Gaden, C.A., Scott, J.M., Walkden-Brown, S.W., 2004. Intensive Rotational Grazing Reduces Nematode Faecal Egg Counts in Sheep on the Cicerone Project. Animal Production in Australia, vol. 25. CSIRO Publishing, Melbourne, pp. 85–88. Kates, K.C., 1950. Survival on pasture of free-living stages of some common gastrointestinal nematodes of sheep. In: Proceedings of the Helminthological Society of Washington, 17. pp. 39–58. Levine, N.D., Anderson, F.L., 1973. Development and survival of Trichostrongylus colubriformis on pasture. J. Parasitol. 59, 147– 165. Levine, N.D., Todd, K.S.J., Boatman, P.A., 1974. Development and Survival of Haemonchus contortus on pasture. Am. J. Vet. Res. 35, 1413–1422. McCosker, T., 2000. Cell grazing-the first 10 years in Australia. Trop. Grasslands 34, 207–218. Monnig, H.O., 1930. Studies on the bionomics of the free-living stages of Trichostrongylus spp. and other parasitic nematodes. Report of the Director of Veterinary Services and Animal Industry, Union of South Africa 16, pp. 175–198. Morgan, D.O., 1933. The effect of heavy stocking on the worm burden under a system of rotational grazing. J. Helminthol. 11, 169–180. Morgan, D.O., Oldham, J.N., 1934. Further observations on the effect of heavy stocking on the worm burden under a system of rotational grazing. J. Helminthol. 12, 177–182. O’Connor, L.J., Walkden-Brown, S.W., Kahn, L.P., 2006. Ecology of the free-living stages of major trichostrongylid parasites of sheep. Vet. Parasitol. 142, 1–15. O’Connor, L.J., Kahn, L.P., Walkden-Brown, S.W., 2007. The effects of amount, timing and distribution of simulated rainfall on the development of Haemonchus contortus to the infective larval stage. Vet. Parasitol. 146, 90–101. Radostits, O.M., Gay, C.C., Blood, D.C., Hinchcliff, K.W., 2000. Veterinary Medicine: A textbook of the Diseases of Cattle, Sheep, Pigs, Goats and Horses. W.B. Saunders Company Ltd., London. Ransom, B.H., 1906. The life history of the twisted wireworm (Haemonchus contortus) of sheep and other ruminants. Circ. US Bureaux Anim. Ind. 93, 7. Reeve, I., Thompson, L., 2005. 2004 Benchmark Survey for the Integrated Parasite Management in Sheep (IPM-s) project. Australian Wool Innovation Ltd., Sydney, Australia, 140 pp. Robertson, D., Fraser, A.H.H., 1933. On the incidence of stomach worms in lambs in the North of Scotland and their control by progressive sectional grazing. J. Helminthol. 11, 187–194.

120

A.F. Colvin et al. / Veterinary Parasitology 153 (2008) 108–120

Roe, R., Southcott, W.H., Turner, H.N., 1959. Grazing management of native pastures in the New England region of New South Wales. 1. Pasture and sheep production with special reference to systems of grazing and internal parasites. Aust. J. Agric. Res. 10, 530–554. Rose, J.H., 1963. Observations on the free-living stages of the stomach worm Haemonchus contortus. Parasitology 53, 469–481. Rossanigo, C.E., Gruner, L., 1995. Moisture and temperature requirements in faeces for the development of free-living stages of gastrointestinal nematodes of sheep, cattle and deer. J. Helminthol. 69, 363–366. Sani, R.A., Mazila, M., Zaki, M., Chong, D.T., Tajuddin, I., 1996. Grazing management for worm control in a sheep-rubber integrated system. In: Silver-Jubilee Malaysian Society of Animal Production Conference, Sarawak, pp. 301–302. Scott, J.M., Gaden, C.A., Shakhane, L., Healey, A.F., 2004. Holding a measuring stick up to the Cicerone farmlets: how are they shaping up? In: Proceedings of the 19th Grasslands Society of NSW Conference, Gunnedah, pp. 63–68.

Silverman, P.H., Campbell, J.A., 1959. Studies on parasitic worms of sheep in Scotland.I. Embryonic and larval development of Haemonchus contortus at constant conditions. Parasitology 49, 23–38. Southcott, W.H., Major, G.W., Barger, I.A., 1976. Seasonal pasture contamination and availability of nematodes for grazing sheep. Aus. J. Agric. Res. 27, 277–286. Sparke, R., 2000. Cell grazing––a producer’s perspective. Trop. Grasslands 34, 219–222. Thamsborg, S.M., Jorgenson, R.J., Waller, P.J., Nanson, P., 1996. The influence of stocking rate on gastrointestinal nematode infections of sheep over a 2-year grazing period. Vet. Parasitol. 67, 207–224. Thomas, R.J., Boag, B., 1972. Epidemiological studies on gastrointestinal nematode parasites of sheep. Res. Vet. Sci. 13, 61–69. Waller, P.J., Donald, A.D., 1970. The response to desiccation of eggs of Trichostrongylus colubriformis and Haemonchus contortus (Nematoda: Trichostrongylidae). Parasitology 61, 195–204. Waller, P.J., Axelsen, A., Donald, A.D., Morley, F.H.W., Dobson, R.J., Donnelly, J.R., 1987. Effects of helminth infection on the preweaning production of ewes and lambs: comparison between safe and contaminated pasture. Aust. Vet. J. 64, 357–362.