Parasitism by gastrointestinal nematodes is responsible for marked economic losses in sheep and goat industries around the world because of the direct costs of livestock loss and reduced weight gain and indirect costs of prevention and treatment measures.1 Nematode infections of goats in the southeastern United States are exacerbated by mild winters that limit nematode death from ground freeze and by high precipitation, particularly in the summer, that limits nematode death from desiccation. Of chief concern in this region are infections by Haemonchus contortus, a nematode of the Strongylidae family that causes mild to severe anemia in sheep and goats.2
Anthelmintic drugs are traditionally used for the treatment of infections by H contortus.3,4 Although the use of these products has often been effective at controlling parasitism by H contortus in the short term, marked anthelmintic resistance has been documented1,5 in recent decades because overuse or underdosing has contributed to nematode anthelmintic resistance on farms4,6 and dispersion of resistant nematode eggs on pastures.7 Nematode parasites in goats appear to be particularly successful at achieving resistance, compared with nematode parasites in sheep, possibly owing to the more rapid drug metabolism of goats versus sheep, which results in higher dosage requirements for effective anthelmintic treatment and, therefore, greater opportunity for underdosing.3,8
In recent years, accounts of anthelmintic resistance and concerns about environmental toxicants and chemical residues in food have encouraged serious investigation into alternative approaches to control nematode infections.9 Alternative, nonpharmaceutical approaches to gastrointestinal nematode control have had varying success rates. Attempts to mitigate nematode parasitism are modeled on 3 mechanisms: eliminating the worms already in the host, bolstering host immunity, and reducing contact between the host and infective-stage larvae.1
To eliminate worms already in the host, possible alternatives to the use of anthelmintic drugs include addition of material suspected to have inherent antiparasitic qualities to the host's diet. For instance, researchers have demonstrated marked success by administering copper oxide wire particles for the treatment of small ruminants with abomasal H contortus infections.10–12 However, unlike copper oxide wire particles, dietary copper sulfate has not been shown to effectively reduce FECs in goats.13 Similarly, the use of herbal dewormers in goats and the inclusion of diatomaceous earth in their diet have also been evaluated with no documented benefit.14,15 Dietary supplementation with bioactive, high-tannin plants provides some benefit in that the addition of sericea lespedeza (Lespedeza cuneate) to caprine diets has been shown to reduce FEC,16,17 as has administration of condensed-tannin quebracho (Schinopsis sp) extract.18 Similarly, access (vs no access) to heather has been associated with a lower FEC and higher resilience of goats to infection,19 and inclusion of Desmodium intortum in their diet has been associated with a reduction in worm burden as well as a decrease in egg production by female worms.20 Oral administration of nematophagous fungi has also been shown21 to be successful in reducing parasitic impact through the nematocidal or oviductal effects of fungal enzyme activity.
As for bolstering host immunity, genetic differences linked to parasite tolerance have been identified in goats, suggesting that breeding management may play a useful role in parasite control on farms.1,22,23 Approaches bolstering individual goat tolerance and immunity through various nutritional strategies have also been investigated.24 One of the most successful strategies has been to increase goats' dietary protein intake in an attempt to boost their resilience to infection, particularly during times of increased metabolic demand (eg, during pregnancy or lactation).25–27 Dietary supplementation with urea as a nonprotein nitrogen source has been associated with a decrease in FEC and improvement in weight gain in sheep,28 as has supplementation with supra-nutritional selenium yeast.29 In Brazil, the combined treatment with IM-administered sodium selenite and SC-administered copper lactobionate, copper gluconate, or copper octadecanoate solution has also been shown30 to successfully reduce H contortus FEC in sheep. Additionally, studies in Spain and China involving vaccines against H contortus infection, such as those involving recombinant cysteine protease31 and H contortus actin-encoding DNA,32 have yielded promising results. Whether these strategies for improving host immunity to gastrointestinal nematodes have the same mechanisms of action is unclear; therefore, it is unknown whether additive or synergistic effects exist when such strategies are combined.
Attempts to mitigate parasitism by reducing contact between the host and infective-stage larvae primarily focus on pasture and grazing management.1 Cograzing of cattle with goats is reportedly a successful strategy,33,34 although multispecies grazing has also been associated with an increase in the prevalence of anthelmintic resistance.35 Rotational grazing, with pasture use carefully planned around nematode life cycles, has also been successful.36 In addition, exposure to H contortus may be limited by increasing goats' access to above ground forages,2 such as plants typically found in wooded areas, instead of or in addition to ground-level pasture forages.
Despite a large body of research on alternative strategies for parasite control, information regarding interactions between various strategies is limited. We hypothesized that combinations of strategies, particularly those with different mechanisms of action, would have a more pronounced effect than single strategies in reducing the prevalence of gastrointestinal nematode infection (specifically, strongyle infection) in goats. The purpose of the study reported here was to identify factors associated with strongyle infection and parasite reduction strategies associated with a low strongyle FEC in goats on farms in North Carolina.
Materials and Methods
Enrollment and participation
Based on an estimated population of 5,000 goat farms in North Carolina at the time of the study and an 85% to 90% prevalence of strongyle infections in goats, a sample size calculation revealed that participation from 35 to 49 goat farms would be necessary to estimate the population prevalence with 95% confidence and a 10% margin of error. To be 95% certain of detecting 10% prevalence of goats with ≥ 1,000 EPG of feces within herds, and assuming 70% test sensitivity, we determined that fecal samples should be collected from all goats on a given farm, to a maximum of 28 goats. On farms with > 28 goats, fecal samples would be collected from a randomly selected subset of 28 goats.
A single recruitment mailing was sent to 411 goat owners identified from records of the North Carolina Department of Agriculture, goat breed organizations, and veterinarians. The mailing invited voluntary participation by owners or managers at any location that housed at least 1 goat (defined for study purposes as a farm). At each so-defined farm, fecal samples were collected from all adult goats (up to a maximum of 28). Goats were excluded from the study if they were < 6 months of age, could not be caught, or had no fecal sample available. The study protocol was approved by the North Carolina State University Institutional Animal Care and Use Committee and Institutional Review Board for the Protection of Human Subjects in Research.
Fecal sample and goat data collection
From June through August 2013, farms were visited to collect fecal samples manually and directly from the rectum of each goat. Collected samples were placed in individual bags,a stored on ice in a cooler, transported directly to the laboratory, and refrigerated overnight for next-day analysis. At the time of fecal sample collection, participating farm owners or managers were asked to provide individual goat information, including age, breed, pregnancy status, lactation status, and date of last kidding.
Strongyle FECs
The McMaster egg counting technique was used to perform an FEC for each collected sample. Briefly, approximately 2 g of each fecal sample, measured by 2 mL of fluid displacement, was added to 13 mL of sodium nitrate fecal flotation solution.b Fecal matter was homogeneously mixed into the solution by use of a wooden tongue depressor. The mixture was strained through a 4 × 4-inch gauze pad into a secondary container. The filtrate was swirled, and a 7-mL transfer pipette was used to transfer the mixture into both 0.15-mL chambers of a double-chambered microscope slide.c Slides were incubated for 10 minutes at 25°C, and then light microscopy was used to count the total number of eggs seen within the 2 chambers. This total number was multiplied by 25 to estimate the total number of strongyle-type EPG of feces. Although the FEC was performed for strongyle-type eggs only, other identified types of parasite eggs or oocysts were quantified in the same manner, except for coccidia, which were subjectively quantified by use of a 4-point scoring system as follows: 1, ≤ 10 EPG; 2, 11 to 30 EPG; 3, 31 to 50 EPG; and 4, > 50 EPG.
Farm data collection
At the time of fecal sample collection, an initial questionnaire was administered to participating farm owners or managers to obtain general farm information, including acreage, feed type, and parasite control techniques. Afterward, it was determined that more specific information was required to achieve more in-depth comparisons of farm environments and management practices. Therefore, a second questionnaire was administered via email or mail approximately 1 to 3 months after initial sample collection to collect information regarding potential risk factors for strongyle infection (Supplementary Appendix S1, available at avmajournals.avma.org/doi/suppl/10.2460/javma.253.7.907). This questionnaire included multiple-choice and fill-in-the-blank questions to elicit more precise information and therefore provide more objective data for analyses.
Statistical analysis
Questionnaire responses considered too highly variable to quantify (eg, responses reflecting inconsistent treatment or that were too specific to be categorized) were eliminated from the response dataset. The remaining responses were included in the dataset and grouped as pertaining to farm- or individual goat–level variables. The primary outcome of interest for all analyses was strongyle FEC.
Farm-level univariate analyses—For analyses regarding farm-level variables, 2 farm-level outcome measures were used: farm mean FEC and the percentage of tested goats on the farm with an FEC > 800 EPG (a value associated with production loss37). As an exploratory analysis, a series of 1-way ANOVAs was performed with the aid of statistical softwared to assess associations between each outcome measure and the following categorical variables: owner-defined purpose of goats (dairy, meat, or companion), breed type (dairy, meat, companion, or mixed), anthelmintic drug administration strategy (no anthelmintic drugs, scheduled dosing, dosed as needed, or both scheduled and dosed as needed), height of grazed pasture grass (kept on wooded pasture, < 3 inches, 3 to 6 inches, or > 6 inches), means of lespedeza access (none, available in pasture, or fed), pasture fertilization (none [existing goat manure only], soil additives, or additional manure), clostridial disease vaccination (twice annually, annually, never; analyzed because this was believed to be a potential indicator of management intensiveness), presence of poultry (yes or no; analyzed because of a common belief that the presence of poultry could reduce parasite populations), and owner perception of difficulty in controlling parasitism (a lot of difficulty, a little difficulty, or no difficulty). In addition, a series of nonparametric Mann-Whitney rank sum tests was performed to assess associations between each outcome measure and the categorical variables dew exposure (yes or no), lespedeza access (yes or no), off-pasture housing provided at night (yes or no), hay or straw bedding provided (yes or no), natural water source access (yes or no), and use of raised feeding troughs (yes or no). Pearson correlations (r) were calculated to quantify linear associations between each outcome measure (both natural and square root–transformed values) and the continuous variables farm acreage (number of acres grazed by goats), goat herd size, goat density (number of goats/acre), time (months) since last anthelmintic treatment, number of anthelmintic drug classes used (benzimidazoles, macrocyclic lactones, or nicotinic receptor agonists), number of nonchemical interventions used for parasite control (selected from a list provided in questionnaire), percentage of grazing area that is wooded, and duration (years) of closed herd status. Assumptions were checked and met for each analysis.
Individual goat–level univariate analyses—At the individual goat level, a series of 1-way ANOVAs was performed to assess associations between individual goat strongyle FEC and the following categorical variables: percentage of grazing area that is wooded (< 20, 20 to 49, 50 to 79, or ≥ 80), height of grazed pasture grass (kept on wooded pasture, < 3 inches, 3 to 6 inches, or > 6 inches), duration (years) of closed herd status, pasture shared with other species (no [goats only], poultry, cattle, horses or donkeys, llamas or alpacas, and sheep), goat sex (female, sexually intact male, or castrated male), breed, breed type (meat, dairy, companion, or mixed), owner-defined purpose (meat, dairy, or companion), time since last kidding (never [female]; < 6 weeks, 6 weeks to < 6 months, or > 1 year; or never [male]), lactation status (lactating female, nonlactating female, or male), pregnancy status (not pregnant, early pregnancy, late pregnancy, recently kidded, or lactating), and number of high-tannin plant species in pasture (0 to 1, 2 to 4, or 5 to 7 [list of qualifying plant species provided in questionnaire]). A series of 1-way ANOVAs was also performed to assess associations between individual goat fecal coccidia scores and the categorical variables sex, breed, and lactation status. Mann-Whitney rank sum tests were performed to assess associations between individual goat strongyle FEC and the following categorical variables: pasture shared with poultry (yes or no; analyzed separately because of a common belief that the presence of poultry could reduce parasite populations), pasture shared with species other than poultry (yes or no), and provision of dietary diatomaceous earth (yes or no). Pearson correlations were calculated to quantify linear associations between individual goat strongyle FEC (both natural and square root–transformed values) and age (years), daily protein intake (males and nonlactating females and lactating females only), and prevalence of other fecal parasite species (coccidia score, Strongyloides FEC, Trichuris FEC, and Moniezia FEC). Assumptions were checked and met for each analysis.
Multivariate analyses—Factors identified as nonsignificant on univariate analysis were eliminated from further multivariate analysis, except for those with previously established associations with gastrointestinal nematode infection in small ruminants. Multivariate analysis was performede to fit a generalized linear model in which the response variable was transformed to square-root values to address a violation of the assumption of homogeneous variances. Farm-level models were created to identify associations between the response variable (farm mean FEC) and the following farm-level predictor variables: anthelmintic drug administration strategy, time since last anthelmintic treatment, lespedeza access (yes or no), dew exposure (yes or no), clostridial disease vaccination (yes or no), and percentage of grazing area that is wooded. Individual goat-level models compared the response variable (square root of individual goat FEC) with the individual goat-level predictor variables goat age (logarithmically transformed), breed, breed type, owner-defined purpose, time since last kidding, lactation status, pregnancy status, daily protein intake, and fecal coccidia score.
All analyses—Because several variables were presumed to be highly related to each other, and because this multicollinearity could have adversely affected the fit of multivariate models and complicated interpretation of multivariate model results, results of both univariate and multivariate modeling are reported. During univariate analyses, the Dunn multiple comparisons test was performed to identify specific significant differences between categories (levels) of any categorical variable that was identified as significantly associated with outcome. Bonferroni correction was performed for all analyses to adjust for the high number of comparisons, and values of P < 0.0038 were consequently considered significant.
Results
Goats and farms
A total of 631 goats on 52 farms were included in the study. Farm herd size, including goats of all ages, ranged from 1 to 240 goats (mean ± SD, 24 ± 37 goats). Twenty-three (44%) farms had ≤ 10 goats, 16 (31%) had 11 to 25 goats, 7 (13%) had 26 to 50 goats, and 6 (12%) had ≥ 51 goats.
Goats included 499 (79.1%) females and 132 (20.9%) males, ranging in age from 6 months to 14 years (mean ± SD, 3.2 ± 2.4 years). Owner-defined purposes of the goats included meat production (n = 262 [41.5%]), dairy production (209 [33.1%]), and companionship (160 [25.4%]). Breeds (characterized by predominant bloodline) included Boer (n = 224 [35.5%]), Alpine (112 [17.7%]), Nigerian Dwarf (70 [11.1%]), Nubian (31 [4.9%]), Spanish (19 [3.0%]), Kiko (12 [1.9%]), LaMancha (11 [1.7%]), Oberhasli (10 [1.6%]), Pygmy (10 [1.6%]), Saanen (10 [1.6%]), Savanna (7 [1.1%]), and Myotonic (6 [1.0%]). The remainder (109 [17.3%]) were of various mixed breeds.
Parasite prevalence
Overall prevalence of strongyle infection in goats as determined through detection of strongyle eggs in fecal samples was 88.9% (561/631). Mean ± SD FEC for all included goats was 660.8 ± 1,114.5 EPG (range, 0.0 to 8,825.0 EPG), for strongyle-positive goats was 743.3 ± 1,155.9 EPG (range, 20 to 8,825 EPG), and for farms in general was 557 ± 474.1 EPG (range, 33 to 1,760 EPG). One hundred thirty-nine (22.0%) goats had an FEC > 800 EPG, and 54 (8.6%) had an FEC > 2,000 EPG.
In addition to strongyle-type eggs, coccidia oocysts were identified in the fecal samples of 57.8% (365/631) of goats on 96% (50/52) of farms, and the coccidia score was positively correlated (r = 0.23; P < 0.001) with individual goat strongyle FEC. This correlation was further supported by the high prevalence (74.8%) of coccidia in fecal samples from goats with a strongyle FEC > 800 EPG and the even higher prevalence (75.2%) in goats with a strongyle FEC > 2,000 EPG. Other identified types of parasite eggs were less common and were classified as Strongyloides spp for 31 of the 631 (4.9%) goats and 8 of the 52 (15.4%) farms, Trichuris spp for 9 (1.4%) goats and 3 (5.8%) farms, and Moniezia spp for 2 (0.3%) goats and 1 (1.9%) farm.
Factors associated with strongyle infection at the farm level
Results of univariate analyses indicated several factors were significantly associated with farm mean FEC or the proportion of the herd with an FEC > 800 EPG when a value of P < 0.05 was used to indicate significance (Tables 1 and 2); however, few factors remained significant (P < 0.0038) following Bonferroni correction. The only factor retaining a significant (P = 0.002) association with mean FEC was owner or manager perception of difficulty in controlling parasitism, with those reporting a lot of difficulty representing farms with the highest mean FECs. The only factor retaining a significant (P = 0.002) association with percentage of the herd that had an FEC > 800 EPG was the number of anthelmintic classes (ie, benzimidazoles, macrocyclic lactones, or nicotinic receptor agonists) used on the farm, for which a positive correlation (r = 0.43) was identified.
Results of univariate analyses to identify categorical variables associated with strongyle infection at the farm level (farm mean strongyle FEC and percentage of tested goats with an FEC > 800 EPG) in 631 goats on 52 farms in North Carolina.
| Â | Â | Mean FEC | Percentage of tested goats with an FEC > 800 | ||
|---|---|---|---|---|---|
| Variable | No. of farms in category | Median (IQR) | P value | Median (IQR) | P value |
| Owner-defined purpose of goats | 52 | Â | 0.02 | Â | 0.04* |
|  Dairy | 14 | 476 (238–947)a,b |  | 15 (0–49) |  |
|  Meat | 15 | 754 (292–1,030)a |  | 26 (0–36) |  |
|  Companion | 23 | 229 (105–518)b |  | 0 (0–19) |  |
| Breed type | 52 | Â | 0.01 | Â | 0.20 |
|  Dairy | 21 | 244 (109–720)a,b |  | 0 (0–27) |  |
|  Meat | 19 | 588 (250–1,030)a |  | 21 (0–30) |  |
|  Companion | 3 | 100 (33–165)b |  | 0 (0–0) |  |
|  Mixed | 9 | 518 (254–1,060)a,b |  | 14 (0–44) |  |
| Anthelmintic drug administration strategy | 52 | Â | 0.19 | Â | 0.05 |
|  As needed | 32 | 554 (229–993) |  | 21 (0–39) |  |
|  Scheduled | 8 | 236 (176–378) |  | 0 (0–9) |  |
|  Both scheduled and as needed | 8 | 357 (208–880) |  | 11 (0–23) |  |
|  Never | 4 | 224 (130–386) |  | 0 (0–0) |  |
| Height (in) of grazed pasture grass | 52 | Â | 0.06 | Â | 0.07 |
|  < 3 | 7 | 239 (38–999) |  | 9 (0–48) |  |
|  3–6 | 26 | 616 (213–1,167) |  | 20 (0–37) |  |
|  > 6 | 13 | 350 (238–700) |  | 11 (4–28) |  |
|  No pasture grass | 6 | 140 (89–311) |  | 0 (0–0) |  |
| Dew exposure | 52 | Â | 0.09 | Â | 0.48 |
|  Yes | 31 | 438 (239–925) |  | 14 (0–33) |  |
|  No | 21 | 228 (100–914) |  | 0 (0–31) |  |
| Lespedeza access | 52 | Â | 0.06 | Â | 0.06 |
|  Yes | 14 | 506 (326–1,007) |  | 18 (10–47) |  |
|  No | 38 | 260 (172–769) |  | 0 (0–30) |  |
| Means of lespedeza access | 52 | Â | 0.17 | Â | 0.16 |
|  Fed | 6 | 476 (298–947) |  | 15 (8–49) |  |
|  Available in pasture | 8 | 733 (304–1,412) |  | 24 (8–42) |  |
|  None | 38 | 269 (184–856) |  | 0 (0–30) |  |
| Off-pasture housing provided at night | 50 | Â | 0.98 | Â | 0.45 |
|  Yes | 28 | 344 (222–922) |  | 4 (0–29) |  |
|  No | 22 | 460 (145–841) |  | 15 (0–34) |  |
| Hay or straw bedding provided | 52 | Â | 0.58 | Â | 0.23 |
|  Yes | 33 | 420 (224–912) |  | 11 (0–41) |  |
|  No | 19 | 269 (165–975) |  | 0 (0–29) |  |
| Natural water source access | 52 | Â | 0.04 | Â | 0.04 |
|  Yes | 15 | 714 (269–930)a |  | 26 (0–50)a |  |
|  No | 37 | 310 (160–740)b |  | 0 (0–27)b |  |
| Fertilizers used on pasture | 52 | Â | 0.11 | Â | 0.06 |
|  Soil additives | 13 | 754 (262–1,183) |  | 21 (9–47) |  |
|  Additional manure | 3 | 754 (430–930) |  | 29 (15–50) |  |
|  None (existing goat manure only) | 38 | 290 (172–747) |  | 0 (0–30) |  |
| Use of raised feeding troughs | 52 | Â | 0.009 | Â | 0.08 |
|  Yes | 32 | 504 (250–964)a |  | 20 (0–39) |  |
|  No | 20 | 230 (107–475)b |  | 0 (0–14) |  |
| Clostridial disease vaccination | 50 | Â | 0.046* | Â | 0.44 |
|  Twice annually | 10 | 651 (340–1,073) |  | 20 (0–35) |  |
|  Annually | 32 | 242 (124–747) |  | 8 (0–32) |  |
|  Never | 8 | 667 (234–1,237) |  | 0 (0–50) |  |
| Owner or manager perception of difficulty controlling parasitism | 52 | Â | 0.002 | Â | 0.02* |
|  A lot of difficulty | 6 | 1,060 (295–1,304)a |  | 42 (0–53) |  |
|  A little difficulty | 20 | 553 (234–1,016)a |  | 20 (7–32) |  |
|  No difficulty | 26 | 236 (104–499)b |  | 0 (0–16) |  |
The ANOVA indicated an uncorrected significant (P < 0.05) difference among categories, but subsequent pairwise comparisons revealed no significant differences between pairs of categories.
Within a variable, values with different superscript letters differ significantly (1-way ANOVA, Dunn method pairwise multiple comparisons procedure, P < 0.05).
Comparisons of mean FEC among ≥ 3 categories within a variable were performed via 1-way ANOVA; comparisons between 2 (yes or no) categories within a variable were performed with the Mann-Whitney rank sum test. Bonferroni correction for multiple comparisons was applied when making conclusions regarding significant findings, and values of P < 0.0038 were considered significant.
Results of univariate analyses to identify continuous variables associated with strongyle infection at the farm level (farm mean strongyle FEC and percentage of tested goats with an FEC > 800 EPG) in the goats of Table 1.
| Â | Â | Mean FEC | Percentage of goats with an FEC > 800 | ||
|---|---|---|---|---|---|
| Variable | Mean ± SD | Pearson correlation | P value | Pearson correlation | P value |
| Farm acreage | 12.35 ± 17.79 | 0.22 | 0.12 | 0.20 | 0.16 |
| Goat herd size (No. of goats) | 24.31 ± 37.17 | 0.18 | 0.21 | 0.19 | 0.18 |
| No. of goats/acre | 6.11 ± 10.86 | –0.23 | 0.10 | –0.22 | 0.12 |
| Time (mo) since last anthelmintic treatment | 3.75 ± 3.72 | –0.08 | 0.60 | 0.06 | 0.69 |
| No. of anthelmintic drug classes* used | 1.63 ± 0.88 | 0.37 | 0.009 | 0.43 | 0.002 |
| No. of nonchemical interventions used for parasite control | 1.60 ± 1.21 | 0.25 | 0.07 | 0.16 | 0.26 |
| Percentage of grazing area that is wooded | 11.85 ± 32.15 | –0.28 | 0.04 | –0.31 | 0.02 |
| Duration (y) of closed herd status | 2.39 ± 3.38 | –0.05 | 0.70 | –0.001 | 1.00 |
Counted drug classes included benzimidazoles, macrocyclic lactones, and nicotinic receptor agonists.
See Table 1 for remainder of key.
Most variables included in the final multivariate model for associations with farm mean strongyle FEC were nonsignificant following Bonferroni correction, including time since last anthelmintic treatment (P = 0.18), lespedeza access (P = 0.09), dew exposure (P = 0.02 [higher FEC with exposure when P < 0.05 indicated significance]), clostridial disease vaccination (P = 0.03 [higher FEC without vaccination when P < 0.05 indicated significance]), and percentage of grazing area that is wooded (P = 0.02). Only anthelmintic drug administration strategy was significant (P < 0.001), whereby farms on which no anthelmintic drugs had ever been used had the lowest mean FECs, farms that administered the drugs on a schedule had higher mean FECs, and farms on which FAMACHA38 scoring or FEC monitoring was used to determine when to administer anthelmintic drugs had the highest mean FECs.
Factors associated with strongyle infection at the individual goat level
Univariate analyses revealed several factors significantly associated with individual goat strongyle FECs when a value of P < 0.05 was used to indicate significance (Tables 3 and 4); most of these factors remained significant (P < 0.0038) following Bonferroni correction. Goats kept as companion animals had significantly (P < 0.001) lower FECs than goats kept for dairy or meat production. Similarly, companion-type goats had significantly (P < 0.001) lower FECs than dairy- or meat-type goats, according to traditional breed usage. Pygmies and Nigerian Dwarfs had significantly (P < 0.001) lower FECs than LaManchas, Kikos, Alpines, and Boers. Goats that had access to grazing areas that were ≥ 80% wooded had significantly (P < 0.001) lower FECs than goats in less wooded grazing areas, whereas grazing pasture grass with a height of 3 to 6 inches was associated with significantly (P = 0.003) higher FECs than grazing pasture grass with a height < 3 or > 6 inches. Furthermore, goats on farms with a closed herd status duration of < 0.5 (6 months) or > 3 years at the time of sample collection had significantly (P < 0.001) higher FECs, compared with goats on farms with a closed herd status duration of 0.5 to < 1 year or 1 to 3 years. Females that had kidded < 6 weeks prior to fecal sample collection had significantly (P = 0.001) higher FECs than other females or males. Individual goat FEC was positively correlated (r = 0.13; P = 0.003) with daily protein intake, which was attributed to lactating females (r = 0.30; P = 0.001) as opposed to nonlactating goats (r = 0.07; P = 0.13) on further analysis. Individual goat FEC was also positively correlated (r = 0.23; P < 0.001) with fecal coccidia score.
Results of univariate analyses to identify categorical variables associated with strongyle infection at the individual level (individual goat strongyle FECs) for the goats in Table 1.
| Variable | No. of goats in category | Median (IQR) FEC | P value |
|---|---|---|---|
| Percentage of grazing area that is wooded | 423 | Â | < 0.001 |
|  < 20 | 119 | 275 (50–675)a |  |
|  20–49 | 140 | 250 (75–650)a |  |
|  50–79 | 106 | 338 (75–888)a |  |
|  ≥ 80 | 58 | 100 (25–306)b |  |
| Height (in) of grazed pasture grass | 390 | Â | 0.003 |
|  < 3 | 43 | 125 (25–1,000)a |  |
|  3–6 | 195 | 350 (100–800)b |  |
|   > 6 | 152 | 250 (75–488)a |  |
| Duration (y) of closed herd status | 424 | Â | < 0.001 |
|  > 0.5 | 177 | 300 (100–175)b |  |
|  0.5 to < 1.0 | 104 | 100 (25–469)a |  |
|  1.0 to < 3.0 | 60 | 162 (5–419)a |  |
|  ≥ 3.0 | 83 | 375 (150–725)b |  |
| Pasture shared with other species | 423 | Â | 0.04* |
|  No | 156 | 200 (50–595) |  |
|  Poultry | 142 | 275 (75–756) |  |
|  Cattle | 77 | 300 (175–762) |  |
|  Horses or donkeys | 114 | 300 (175–762) |  |
|  Llamas or alpacas | 13 | 125 (12–238) |  |
|  Sheep | 7 | 200 (25–650) |  |
| Pasture shared with poultry | 423 | Â | 0.15 |
|  Yes | 142 | 275 (75–756) |  |
|  No | 281 | 250 (50–590) |  |
| Pasture shared with species other than poultry | 423 | Â | 0.08 |
|  Yes | 267 | 275 (75–650) |  |
|  No | 156 | 200 (50–595) |  |
| Sex | 607 | Â | 0.14 |
|  Female | 498 | 262 (75–700) |  |
|  Sexually intact male | 67 | 350 (75–1,775) |  |
|  Castrated male | 42 | 188 (44–506) |  |
| Breed classification | 631 | Â | < 0.001 |
|  LaMancha | 11 | 725 (240–1,700)a |  |
|  Kiko | 12 | 475 (300–2,494)a |  |
|  Alpine | 112 | 422 (180–1,256)a |  |
|  Boer | 224 | 362 (100–825)a |  |
|  Nubian | 31 | 325 (175–580)a,b |  |
|  Spanish | 19 | 300 (175–800)a,b |  |
|  Savanna | 7 | 250 (175–350)a,b |  |
|  Saanen | 10 | 238 (144–520)a,b |  |
|  Mixed breed | 109 | 225 (45–550)a,b |  |
|  Myotonic | 6 | 140 (50–655)a,b |  |
|  Oberhasli | 10 | 138 (75–269)a,b |  |
|  Nigerian Dwarf | 70 | 50 (0–131)b |  |
|  Pygmy | 10 | 25 (0–86)b |  |
| Breed type | 630 | Â | < 0.001 |
|  Meat | 295 | 245 (70–649)a |  |
|  Dairy | 276 | 325 (100–825)a |  |
|  Companion | 23 | 75 (12–250)b |  |
|  Mixed | 36 | 175 (40–650)a,b |  |
| Owner-defined purpose | 631 | Â | < 0.001 |
|  Meat | 262 | 350 (106–825)a |  |
|  Dairy | 209 | 325 (75–900)a |  |
|  Companion | 160 | 150 (106–825)b |  |
| Time since last kidding | 568 | Â | 0.001 |
|  Never (female) | 153 | 300 (62–712)a |  |
|  < 6 wk | 8 | 2,325 (375–3,519)b |  |
|  6 wk to < 6 mo | 173 | 275 (100–778)a |  |
|   > 1 y | 101 | 175 (50–375)a |  |
|  Never (male) | 133 | 250 (75–850)a |  |
| Lactation status | 615 | Â | 0.03 |
|  Lactating female | 121 | 375 (100–955)a |  |
|  Nonlactating female | 361 | 250 (50–650)b |  |
|  Male | 133 | 250 (75–850)a,b |  |
| Pregnancy status | 455 | Â | 0.007 |
|  Not pregnant | 253 | 250 (62–700)a,b |  |
|  Early pregnancy | 16 | 62 (0–950)a |  |
|  Late pregnancy | 71 | 200 (50–400)a |  |
|  Recently kidded | 8 | 2,325 (375–3,519)b |  |
|  Lactating | 107 | 400 (100–950)a,b |  |
| Dietary diatomaceous earth provided | 423 | Â | 0.25 |
|  Yes | 11 | 80 (40–400) |  |
|  No | 412 | 250 (75–650) |  |
| No. of high-tannin plant species in pasture | 422 | Â | 0.02 |
|  0–1 | 10 | 38 (0–225)a |  |
|  2–4 | 191 | 250 (75–650)b |  |
|  5–7 | 221 | 250 (75–712)b |  |
See Table 1 for key.
Results of univariate analyses to identify continuous variables associated with individual strongyle FEC for the goats in Table 1.
| Variable | No. of goats | Mean ± SD | Pearson correlation | P value |
|---|---|---|---|---|
| Age (y) at time of sample collection | 443 | 3.0 ± 2.6 | –0.12 | 0.02 |
| Daily protein intake (oz) |  | 2.8 ± 3.0 | 0.13 | 0.003 |
|  Males and nonlactating females | 423 | 2.0 ± 1.7 | 0.07 | 0.13 |
|  Lactating females | 121 | 6.0 ± 4.3 | 0.30 | 0.001 |
| Other fecal parasite species | Â | Â | Â | Â |
|  Coccidia score* | 631 | 0.94 ± 1.02 | 0.23 | < 0.001 |
|  Strongyloides FEC | 631 | 0.06 ± 0.29 | –0.002 | 0.96 |
|  Trichuris FEC | 631 | 0.02 ± 0.15 | –0.03 | 0.51 |
|  Moniezia FEC | 631 | 0.00 ± 0.06 | –0.02 | 0.70 |
Scored on a 4-point scale, where 1 = ≤ 10 EPG, 2 = 11 to 30 EPG, 3 = 31 to 50 EPG, and 4 = > 50 EPG.
See Table 1 for key.
Several variables in the final multivariable model for associations with individual goat strongyle FEC were insignificant following Bonferroni correction, including goat age (P = 0.02 [lower FEC with increased age with P value cutoff of < 0.05]), time since last kidding (P = 0.01 [higher FEC with recent kidding with P value cutoff of < 0.05]), and lactation status (P = 0.005 [higher FEC with lactation with P value cutoff of < 0.05]). The variables deemed significant following Bonferroni correction were breed (P < 0.001), breed type (P < 0.001), owner-defined purpose (P < 0.001), daily dietary protein intake (P < 0.001), coccidia presence (P < 0.001), and pregnancy status (P = 0.002). Breed comparisons revealed that Pygmies had the lowest FECs, followed by Nigerian Dwarfs, whereas Kikos had the highest FECs, followed by LaManchas. Similarly, breeds traditionally used as companion animals (eg, Pygmy) had the lowest FECs, breeds traditionally used for dairy (eg, Nigerian Dwarfs and LaManchas) had higher FECs, and breeds traditionally used for meat (eg, Kikos) had the highest FECs. Additionally, comparisons of goats grouped by owner-defined purpose (regardless of breed or breed type) indicated that goats used as companion animals had lower FECs, compared with goats used for dairy or meat production. Other comparisons revealed that goats with a larger amount of protein in their daily diet had higher FECs, and goats with higher strongyle loads also had higher burdens of coccidia. Regarding pregnancy status, there was a slight decline in FEC with increased gestation, a sharp rise in FEC in the periparturient period, and a decline in FEC with lactation to a level above those maintained by nonlactating does; does that had recently kidded (in the past 6 weeks) had the highest FECs, whereas does in the early stages of pregnancy had the lowest FECs.
Discussion
Results of the present study confirmed that strongyle infection, as determined by FEC, was common in the included goats. All 52 North Carolina farms had ≥ 1 infected goat, and 22.0% of goats had a strongyle FEC associated with production loss (> 800 EPG). Comparisons of individual goat FECs and farm mean FECs revealed that strongyle burden was higher in some goats and some farms than others. The unequal impact also became apparent during our conversations with farm owners and managers and was most likely attributable to differences among farms in terms of management practices and among goats in response to gastrointestinal parasites.
The high prevalence of coccidia infection combined with the positive correlation identified between fecal coccidia score and strongyle FEC suggested that the same factors that influence the likelihood or degree of strongyle infection may also influence the likelihood or degree of coccidia infection. Such factors may include young age and stressful events.39,40 In the present study, the youngest goats had the highest strongyle FECs as indicated by correlation analysis, a finding similar to results of a study41 on goats in India.
Markedly high strongyle FECs were associated with kidding and lactation in the present study, demonstrating that the reported periparturient rise in FEC that occurs in sheep25,42,43 also occurred in goats. Despite the intrinsic relationship between kidding and lactation, lactation status was more important than the time since last kidding in the prediction of FEC in goats, consistent with reported findings for Creole goats.44
Comparisons involving breed-related variables, such as breed in terms of predominant bloodline or breed in terms of traditional purpose (ie, breed type [companion, meat, or dairy] in the present study), suggested that inherent genetic differences or traditional production use predisposed specific breeds to high strongyle burdens. However, it was challenging to separate the influence of each of these variables on outcome because most goats of given breeds were also used for their traditional purpose. Our findings were consistent with a report45 that suggested parasite resistance could be improved through breed selection; however, selection for parasite resistance must be balanced with production goals, including those for meat and dairy production, which may select against the use of small-breed goats.
Our finding that Kikos had higher strongyle FECs than other breeds contradicted the widely held belief that this breed is less susceptible to intestinal parasitism. The resistance of Kikos may have been overestimated in the past, or owners of farms with a high parasite burden in the present study may have been more likely to choose Kikos than other breeds. Because we did not assess goats for clinical signs of disease or measure anemia, it remains unclear whether Kikos are less likely to become infected with strongyles or are simply more tolerant of high parasite burdens and have fewer clinical effects than other breeds.
To better understand the influence of goat breeding versus the influence of breed type on strongyle FECs, another breed-related variable, owner-defined purpose (ie, dairy, meat, or companion), was analyzed because not all goats served in a traditional purpose, and actual purpose, regardless of its breed, may have outweighed genetic predispositions to high or low FECs. Results of multivariate analysis indicated that goats of any breed used as companion animals had significantly lower FECs than goats used by owners for other purposes, suggesting that owner-defined purpose was also a predictor of FEC along with breed and breed type.
The lower strongyle FECs of companion goats versus meat or dairy goats may have been attributable to various factors, including more individual attention from the owner (perhaps combined with greater financial investment in veterinary care), a longer life span, and absence of the many physiologic stressors associated with kidding and lactation of production animals. Interestingly, owner-defined purpose was not related to goat density because the number of goats per acre was not associated with the farm median strongyle FEC.
In addition to identifying factors associated with strongyle infection on goat farms, an aim of the present study was to determine which parasite control strategies implemented on farms were associated with low median strongyle FECs. One such strategy, which was not significant, was prevention of grazing during early mornings (inferred through no reported exposure to dew). A high abundance of water droplets on pasture increases the opportunity for parasite larvae to reside on forage and thus be consumed by a host.46 Furthermore, univariate analyses revealed that farms where goats had access to a natural water source had higher median strongyle FECs than farms without such access; however, survey responses were often inadequate to determine whether a standing natural water source originated simply from wet weather patterns (known to increase larval migration from feces onto grass47) or from a more constant natural water source that might have affected FECs. Because no weather data were collected at the time of fecal sample collection, this potentially important confounding variable was excluded from multivariate analysis involving farm median FECs.
Height of grazed pasture grass was also hypothesized to influence goat exposure to infective-stage nematode larvae because upward migration of larvae on blades of grass is reportedly limited to a maximum height of 5 cm,48 although this can increase to heights up to 22.5 cm in rainy conditions of experimental potted grass models.42 Univariate analyses at the individual goat level revealed that medium-height grass (3 to 6 inches [7.6 to 15.2 cm] tall) was associated with significantly higher FECs than taller or shorter grass. The lack of significance for shorter grass could have been attributable to limited grass growth caused by drier conditions, which would also limit survival of infective stage larvae, or regular mowing of grass. This variable was not significant on multivariate analysis.
Neither wooded pasture nor lespedeza exposure was associated with farm median strongyle FECs after other variables were controlled for in the present study. This was surprising given that these factors have been reported to impact FECs of small ruminants.2,18,19 A benefit may not have been evident because lespedeza access and the opportunity for goats to graze in a mostly wooded pasture were limited to a minority of farms. Further, it was possible that such conditions existed only on farms with an established parasite problem, therefore confounding any ability to have shown a positive impact. These conditions may have also explained the finding that use of raised feeding troughs was associated with higher farm median FECs on univariate analyses before Bonferroni correction. Additionally, goat exposure to tannin-containing plants (believed to have antiparasitic properties) and the percentage of grazing area that was wooded (believed to reduce contact with infective-stage larvae) were not independently associated with outcome on multivariate analysis. Grazing of goats on tannin-containing plants has been shown to have had a greater effect on the reduction of egg production by female H contortus worms than on other nematodes,49 and this could have contributed to the lack of effect identified in the present study because no attempt was made to identify the species of the counted strongyle-type eggs. Prospective, controlled studies would be helpful in determining the direct effect of these variables on individual goat and farm mean FECs.
In the study reported here, dietary protein intake was positively correlated with individual goat FEC. Dietary protein supplementation reportedly improves host resilience,50 particularly in kids and periparturient females.27 The positive correlation observed in the present study was unexpected and was likely attributable to goat purpose, given that most goats receiving a large amount of grain were lactating females that were more inclined to have high FECs, regardless of the amount of protein in their diet. The higher amount of dietary protein could have helped boost host immunity, although not enough to overcome the higher FEC subsequent to kidding and lactation. We were unable to assess the nutritional quality of the feed supplement or overall diet, which could also have influenced our findings.
Farms on which no anthelmintic drugs had ever been used had the lowest mean FECs, and those on which such drugs were administered on a schedule (eg, yearly or monthly) had higher mean FECs, whereas farms on which FAMACHA38 scoring or FEC monitoring was used to determine when to administer the drugs had the highest mean FECs. Use of FAMACHA scoring or FEC monitoring was expected to be associated with high farm mean FECs because individual treatments of goats (instead of whole-herd treatments) reduces FECs in smaller percentages of the involved herds, compared with larger percentages in whole-herd treatments. The finding that no history of anthelmintic drug administration was associated with the lowest farm mean FECs suggested that alternative methods of parasite control and prevention, rather than treatment, could be successful and, by extension, may be viable options for controlling anthelmintic resistance.
Although we believed strongyle FEC represented a reasonable estimate of H contortus burden within the study population, the strongyle eggs were not specifically identified as H contortus eggs. Straining of fecal samples before the FEC was performed could have resulted in lower absolute counts than in unstrained samples.51 Also, the impact of season on strongyle FEC could not have been assessed because all goat fecal samples were collected during the summer. Another limitation of the present study was the lack of controlling for farm within the individual goat analysis; goats on the same farm could be expected to be most similar owing to shared management practices, or the presence of a goat with a strongyle infection on a farm would put other goats on that farm at risk of strongyle exposure. It is also important to note that the use of a Bonferroni-corrected P value for indicating significant differences may have increased the chance of type II error along with the intended increased confidence; therefore, the potential influence on outcome variables that did not meet this significance cutoff should not be ruled out. Although the cross-sectional nature of the present study precluded establishment of causal relationships between any specific farm or individual goat factors and degree of parasitism, the findings obtained would be useful to explore further in future prospective studies.
Acknowledgments
Supported by the North Carolina State College of Veterinary Medicine.
The authors declare that there were no conflicts of interest.
ABBREVIATIONS
| EPG | Eggs per gram |
| FAMACHA | Faffa Malan chart |
| FEC | Fecal egg count |
| IQR | Interquartile (25th to 75th percentile) range |
Footnotes
Whirl-Pak, Nasco, Fort Atkinson, Wis.
Feca-Med, Vedco Inc, St Joseph, Mo.
Two-Chamber Green Grid McMaster Slide, CHALEX LLC, Park City, Utah.
SigmaPlot, version 12, Systat Software Inc, San Jose, Calif.
PROC GENMOD, SAS, version 9.3, SAS Institute Inc, Cary, NC.
References
1. Torres-Acosta JFJ, Hoste H. Alternative or improved methods to limit gastro-intestinal parasitism in grazing sheep and goats. Small Rumin Res 2008;77:159–173.
2. Miller JE, Kaplan RM, Pugh DG. Internal parasites. In: Pugh DG, Baird AN, eds. Sheep and goat medicine. 2nd ed. Maryland Heights, Mo: Elsevier Saunders, 2012;106–125.
3. Lespine A, Chartier C, Hoste H, et al. Endectocides in goats: pharmacology, efficacy and use conditions in the context of anthelmintic resistance. Small Rumin Res 2012;103:10–17.
4. Kaplan RM. Drug resistance in nematodes of veterinary importance: a status report. Trends Parasitol 2004;20:477–481.
5. Barton NJ, Trainor BL, Urie JS, et al. Anthelmintic resistance in nematode parasites of goats. Aust Vet J 1985;62:224–227.
6. Craig TM. Anthelmintic resistance and alternative control measures. Vet Clin North Am Food Anim Pract 2006;22:567–581.
7. Charles TP, Pompeu J, Miranda DB. Efficacy of three broad-spectrum anthelmintics against gastrointestinal nematode infections of goats. Vet Parasitol 1989;34:71–75.
8. Manfredi MT, Di Cerbo AR, Zanzani S, et al. Breeding management in goat farms of Lombardy, northern Italy: risk factors connected to gastrointestinal parasites. Small Rumin Res 2010;88:113–118.
9. Williams JC. Anthelmintic treatment strategies: current status and future. Vet Parasitol 1997;72:461–477.
10. Vatta AF, Waller PJ, Githiori JB, et al. Persistence of the efficacy of copper oxide wire particles against Haemonchus contortus in grazing South African goats. Vet Parasitol 2012;190:159–166.
11. Soli F, Terrill TH, Shaik SA, et al. Efficacy of copper oxide wire particles against gastrointestinal nematodes in sheep and goats. Vet Parasitol 2010;168:93–96.
12. Chartier C, Etter E, Hoste H, et al. Efficacy of copper oxide needles for the control of nematode parasites in dairy goats. Vet Res Commun 2000;24:389–399.
13. Burke JM, Miller JE. Dietary copper sulfate for control of gastrointestinal nematodes in goats. Vet Parasitol 2008;154:289–293.
14. Burke JM, Wells A, Casey P, et al. Herbal dewormer fails to control gastrointestinal nematodes in goats. Vet Parasitol 2009;160:168–170.
15. Bernard G, Worku M, Ahmedna M. The effects of diatomaceous earth on parasite infected goats. Bull Georgian Nat Acad Sci 2009;3:129–135.
16. Burke JM, Orlik S, Miller JE, et al. Using copper oxide wire particles or Sericea lespedeza to prevent peri-parturient gastrointestinal nematode infection in sheep and goats. Livest Sci 2010;132:13–18.
17. Min BR, Hart SP, Miller D, et al. The effect of grazing forage containing condensed tannins of gastrointestinal parasite infection and milk composition in Angora does. Vet Parasitol 2005;130:105–113.
18. Paolini V, Bergeaud JP, Grisez C, et al. Effects of condensed tannin of goats experimentally infected with Haemonchus contortus. Vet Parasitol 2003;113:253–261.
19. Moreno-Gonzalo J, Ferre I, Celaya R, et al. Potential use of heather to control gastrointestinal nematodes in goats. Small Rumin Res 2012;103:60–68.
20. Debela E, Tolera A, Eik LO, et al. Condensed tannins from Sesbania sesban and Desmodium intortum as a means of Haemonchus contortus control in goats. Trop Anim Health Prod 2012;44:1939–1944.
21. Braga FR, Araújo JV. Nematophagous fungi for biological control of gastrointestinal nematodes in domestic animals. Appl Microbiol Biotechnol 2014;98:71–82.
22. Hoste H, Sotiraki S, Torres-Acosta JFJ. Control of endoparasitic nematode infections in goats. Vet Clin North Am Food Anim Pract 2011; 27:163–173.
23. Bishop SC, Morris CA. Genetics of disease resistance in sheep and goats. Small Rumin Res 2007;70:48–59.
24. Houdjik JGM, Kyriazakis I, Kidane A, et al. Manipulating small ruminant parasite epidemiology through the combination of nutritional strategies. Vet Parasitol 2012;186:38–50.
25. Houdijk JGM. Influence of periparturient nutritional demand on resistance to parasites in livestock. Parasite Immunol 2008;30:113–121.
26. MartÃnez Ortiz de Montellano C, Vargas-Magana JJ, Aguilar-Caballero AJ, et al. Combining the effects of supplementary feeding and copper oxide needles for the control of gastrointestinal nematodes in browsing goats. Vet Parasitol 2007;146:66–76.
27. Knox MR, Torres-Acosta JFJ, Aguilar-Caballero AJ. Exploiting the effect of dietary supplementation of small ruminants on resilience and resistance against gastrointestinal nematodes. Vet Parasitol 2006;139:385–393.
28. Knox MR, Steel JW. The effects of urea supplementation on production and parasitological responses of sheep infected with Haemonchus contortus and Trichostrongylus colubriformis. Vet Parasitol 1999;83:123–135.
29. Hooper KJ, Bobe G, Voracheck WR, et al. Effect of selenium yeast supplementation on naturally acquired parasitic infection in ewes. Biol Trace Elem Res 2014;161:308–317.
30. Fausto GC, Pivoto FL, Costa MM, et al. Protein profile of lambs experimentally infected with Haemonchus contortus and supplemented with selenium and copper. Parasit Vectors 2014;7:355–360.
31. Muleke CI, Yan R, Sun Y, et al. Vaccination of goats against Haemonchus contortus with a recombinant cysteine protease. Small Rumin Res 2007;73:95–102.
32. Yan R, Wang J, Xu L, et al. DNA vaccine encoding Haemonchus contortus actin induces partial protection in goats. Acta Parasitol 2014;59:698–709.
33. d'Alexis S, Periacarpin F, Jackson F, et al. Mixed grazing systems of goats with cattle in tropical conditions: an alternative to improving animal production in the pasture. Animal 2014;8:1282–1289.
34. Marshall R, Gebrelul S, Gray L, et al. Mixed species grazing of cattle and goats on gastrointestinal infections of Haemonchus contortus. Am J Anim Vet Sci 2012;7:61–66.
35. Falzon LC, O'Neill TJ, Menzies PI, et al. A systematic review and meta-analysis of factors associated with anthelmintic resistance in sheep. Prev Vet Med 2014;117:388–402.
36. Barger IA, Siale K, Banks DJD, et al. Rotational grazing for control of gastrointestinal nematodes of goats in a wet tropical environment. Vet Parasitol 1994;53:109–116.
37. Blackburn HD, Rocha JL, Figueiredo EP, et al. Interaction of parasitism and nutrition and their effects on production and clinical parameters in goats. Vet Parasitol 1991;40:99–112.
38. van Wyk JA, Bath GF. The FAMACHA system for managing haemonchosis in sheep and goats by clinically identifying individual animals for treatment. Vet Res 2002;33:509–529.
39. Kanyari PW. The relationship between coccidial and helminth infections in sheep and goats in Kenya. Vet Parasitol 1993;51:137–141.
40. Kanyari PW. Coccidiosis in goats and aspects of epidemiology. Aust Vet J 1988;65:257–258.
41. Tariq KA, Chishti MZ, Ahmad F. Gastrointestinal nematode infections in goats relative to season, host sex and age from Kashmir Valley, India. J Helminthol 2010;84:93–97.
42. Fleming MW. Selection for a strain of Haemonchus contortus that exhibits periparturient egg rise in sheep. J Parasitol 1993;79:399–402.
43. Beasley AM, Kahn LP, Windon RG. The periparturient relaxation of immunity in Merino ewes infected with Trichostrongylus colubriformis: parasitological and immunological responses. Vet Parasitol 2010;168:60–70.
44. Mandonnet N, Bachand M, Mahieu M, et al. Impact on productivity of peri-parturient rise in fecal egg counts in Creole goats in the humid tropics. Vet Parasitol 2005;134:249–259.
45. Bishop SC. Possibilities to breed for resistance to nematode parasite infections in small ruminants in tropical production systems. Animal 2012;6:741–747.
46. Machen R, Craddock F, Craig T, et al. A Haemonchus contortus management plan for sheep and goats in Texas. Pamphlet L-5095. College Station, Tex: AgriLife Communications, Texas A&M System, 1998.
47. Amaradasa BS, Lane RA, Manage A. Vertical migration of Haemonchus contortus infective larvae on Cynodon dactylon and Paspalum notatum pastures in response to climatic conditions. Vet Parasitol 2010;170:78–87.
48. Hoste H, Sotiraki S, Landau SY, et al. Goat-nematode interactions: think differently. Trends Parasitol 2010;26:376–381.
49. Shaik SA, Terrill TH, Miller JE, et al. Sericea lespedeza hay as a natural deworming agent against gastrointestinal nematode infection in goats. Vet Parasitol 2006;139:150–157.
50. Nnadi PA, Kamalu TN, Onah DN. The effect of dietary protein on the productivity of West African Dwarf (WAD) goats infected with Haemonchus contortus. Vet Parasitol 2009;161:232–238.
51. Bowman DD. Diagnostic parasitology. In: Bowman DD, ed. Georgi's parasitology for veterinarians. 9th ed. Maryland Heights, Mo: Elsevier Saunders, 2009;300.
